diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..262ee2e9fde17fce384c879b2d5495f523a21a21 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +checkpoint-1624/tokenizer.json filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-1218/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-2433/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-2030/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..5cc707f21a2ec05fe10d969c7db62073e1f9a712 --- /dev/null +++ b/README.md @@ -0,0 +1,145 @@ +--- +library_name: transformers +license: llama3.1 +base_model: meta-llama/Llama-3.1-8B-Instruct +tags: +- generated_from_trainer +datasets: +- tokenized_novelty_dataset_4/train_full.parquet +model-index: +- name: outputs_4/llama31-8B-liger-ds-full + results: [] +--- + + + +[Built with Axolotl](https://github.com/axolotl-ai-cloud/axolotl) +
See axolotl config + +axolotl version: `0.12.2` +```yaml +base_model: meta-llama/Llama-3.1-8B-Instruct + +#load_in_4bit: true + +#adapter: lora +#lora_r: 16 +#lora_alpha: 32 +#lora_dropout: 0.05 +#lora_target_modules: +# - q_proj +# - v_proj +# - k_proj +# - o_proj + + +plugins: + - axolotl.integrations.liger.LigerPlugin +liger_rope: true +liger_rms_norm: true +liger_glu_activation: true +liger_fused_linear_cross_entropy: true + +strict: false + +chat_template: llama3 +datasets: + - path: "tokenized_novelty_dataset_4/train_full.parquet" + type: + ds_type: parquet + +dataset_prepared_path: +val_set_size: 0.00 +output_dir: ./outputs_4/llama31-8B-liger-ds-full + +dataset_processes: 16 + +sequence_len: 32120 +sample_packing: false +pad_to_sequence_len: true + +gradient_accumulation_steps: 1 +micro_batch_size: 1 +num_epochs: 3 +optimizer: adamw_torch +lr_scheduler: cosine +learning_rate: 2e-5 + +train_on_inputs: false +group_by_length: false +bf16: auto +fp16: +tf32: true + +gradient_checkpointing: true +gradient_checkpointing_kwargs: + use_reentrant: false +early_stopping_patience: +resume_from_checkpoint: +logging_steps: 1 +flash_attention: true + +warmup_steps: 50 +evals_per_epoch: 0 +eval_table_size: +saves_per_epoch: 2 +save_only_model: true +debug: +deepspeed: deepspeed_configs/zero3_bf16.json +weight_decay: 0.0 +fsdp: +fsdp_config: +special_tokens: + pad_token: <|finetune_right_pad_id|> + eos_token: <|eot_id|> + + +``` + +

+ +# outputs_4/llama31-8B-liger-ds-full + +This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the tokenized_novelty_dataset_4/train_full.parquet dataset. + +## Model description + +More information needed + +## Intended uses & limitations + +More information needed + +## Training and evaluation data + +More information needed + +## Training procedure + +### Training hyperparameters + +The following hyperparameters were used during training: +- learning_rate: 2e-05 +- train_batch_size: 1 +- eval_batch_size: 1 +- seed: 42 +- distributed_type: multi-GPU +- num_devices: 32 +- total_train_batch_size: 32 +- total_eval_batch_size: 32 +- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: cosine +- lr_scheduler_warmup_steps: 50 +- training_steps: 2433 + +### Training results + + + +### Framework versions + +- Transformers 4.55.2 +- Pytorch 2.6.0+cu124 +- Datasets 4.0.0 +- Tokenizers 0.21.4 diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..870322b8f52314199bfd74bf68b8b8f1126395ae --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,5 @@ +{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} diff --git a/checkpoint-1218/chat_template.jinja b/checkpoint-1218/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..870322b8f52314199bfd74bf68b8b8f1126395ae --- /dev/null +++ b/checkpoint-1218/chat_template.jinja @@ -0,0 +1,5 @@ +{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} diff --git a/checkpoint-1218/config.json b/checkpoint-1218/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4c0f53b6c66a0b8a2a4eb621c949412e2a7f353a --- /dev/null +++ b/checkpoint-1218/config.json @@ -0,0 +1,35 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.2", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/checkpoint-1218/generation_config.json b/checkpoint-1218/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b513e54e3195b917260c9a8a04c9f3683f19de35 --- /dev/null +++ b/checkpoint-1218/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.55.2" +} diff --git a/checkpoint-1218/model-00004-of-00004.safetensors b/checkpoint-1218/model-00004-of-00004.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..db3c849aa89c7985d5c7ea3a8b68c8f8086d217b --- /dev/null +++ b/checkpoint-1218/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aac380227388fd5111951a0c7ba0de5ae01a5ca74e33626a4a4a16c95a039987 +size 1168138808 diff --git a/checkpoint-1218/model.safetensors.index.json b/checkpoint-1218/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..58a7ef40d52ab3340cf131cf33f8923f5144aaf9 --- /dev/null +++ b/checkpoint-1218/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-1218/special_tokens_map.json b/checkpoint-1218/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3c1d04911c269b925af977a3151c9704e990e4d0 --- /dev/null +++ b/checkpoint-1218/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1218/tokenizer.json b/checkpoint-1218/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-1218/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-1218/tokenizer_config.json b/checkpoint-1218/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..c68051fe3c4d23234a59316bc52d21f6e3a4182c --- /dev/null +++ b/checkpoint-1218/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|finetune_right_pad_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1218/trainer_state.json b/checkpoint-1218/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..f67d0e58ffb7fa683426577bae42c929bbbe199e --- /dev/null +++ b/checkpoint-1218/trainer_state.json @@ -0,0 +1,12214 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.501849568434032, + "eval_steps": 500, + "global_step": 1218, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0012330456226880395, + "grad_norm": 13.955555906863093, + "learning_rate": 0.0, + "loss": 1.9039, + "memory/device_mem_reserved(gib)": 19.83, + "memory/max_mem_active(gib)": 16.16, + "memory/max_mem_allocated(gib)": 16.16, + "step": 1 + }, + { + "epoch": 0.002466091245376079, + "grad_norm": 14.68412675273428, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.9395, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 2 + }, + { + "epoch": 0.0036991368680641184, + "grad_norm": 15.404314959825989, + "learning_rate": 8.000000000000001e-07, + "loss": 1.9524, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 3 + }, + { + "epoch": 0.004932182490752158, + "grad_norm": 14.099502171385124, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.8527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 4 + }, + { + "epoch": 0.006165228113440197, + "grad_norm": 12.213499070556583, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.8853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 5 + }, + { + "epoch": 0.007398273736128237, + "grad_norm": 7.4228290511653725, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.7686, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 6 + }, + { + "epoch": 0.008631319358816275, + "grad_norm": 6.989334002600767, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.7976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 7 + }, + { + "epoch": 0.009864364981504316, + "grad_norm": 7.330479156896915, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6029, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 8 + }, + { + "epoch": 0.011097410604192354, + "grad_norm": 7.091571204851302, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.6097, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 9 + }, + { + "epoch": 0.012330456226880395, + "grad_norm": 5.900225931668668, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.6512, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 10 + }, + { + "epoch": 0.013563501849568433, + "grad_norm": 12.562784448457464, + "learning_rate": 4.000000000000001e-06, + "loss": 1.4824, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 11 + }, + { + "epoch": 0.014796547472256474, + "grad_norm": 6.056074456731366, + "learning_rate": 4.4e-06, + "loss": 1.5581, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 12 + }, + { + "epoch": 0.016029593094944512, + "grad_norm": 5.051119502724206, + "learning_rate": 4.800000000000001e-06, + "loss": 1.5275, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 13 + }, + { + "epoch": 0.01726263871763255, + "grad_norm": 4.022180485081632, + "learning_rate": 5.2e-06, + "loss": 1.5075, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 14 + }, + { + "epoch": 0.018495684340320593, + "grad_norm": 4.5104551135107815, + "learning_rate": 5.600000000000001e-06, + "loss": 1.4638, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 15 + }, + { + "epoch": 0.01972872996300863, + "grad_norm": 4.3429897237996276, + "learning_rate": 6e-06, + "loss": 1.3996, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 16 + }, + { + "epoch": 0.02096177558569667, + "grad_norm": 3.9791138623880764, + "learning_rate": 6.4000000000000006e-06, + "loss": 1.4059, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 17 + }, + { + "epoch": 0.02219482120838471, + "grad_norm": 4.194505204288471, + "learning_rate": 6.800000000000001e-06, + "loss": 1.3747, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 18 + }, + { + "epoch": 0.02342786683107275, + "grad_norm": 3.6309957760327713, + "learning_rate": 7.2000000000000005e-06, + "loss": 1.3565, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 19 + }, + { + "epoch": 0.02466091245376079, + "grad_norm": 3.431355025722372, + "learning_rate": 7.600000000000001e-06, + "loss": 1.4019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 20 + }, + { + "epoch": 0.025893958076448828, + "grad_norm": 3.258474422309909, + "learning_rate": 8.000000000000001e-06, + "loss": 1.364, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 21 + }, + { + "epoch": 0.027127003699136867, + "grad_norm": 3.7750095693616736, + "learning_rate": 8.400000000000001e-06, + "loss": 1.3798, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 22 + }, + { + "epoch": 0.02836004932182491, + "grad_norm": 3.1192450402705125, + "learning_rate": 8.8e-06, + "loss": 1.2677, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 23 + }, + { + "epoch": 0.029593094944512947, + "grad_norm": 3.34669710403596, + "learning_rate": 9.200000000000002e-06, + "loss": 1.3465, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 24 + }, + { + "epoch": 0.030826140567200986, + "grad_norm": 3.0969841766129638, + "learning_rate": 9.600000000000001e-06, + "loss": 1.3662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 25 + }, + { + "epoch": 0.032059186189889025, + "grad_norm": 2.9521843506112466, + "learning_rate": 1e-05, + "loss": 1.3295, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 26 + }, + { + "epoch": 0.03329223181257707, + "grad_norm": 2.9117498945141085, + "learning_rate": 1.04e-05, + "loss": 1.3066, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 27 + }, + { + "epoch": 0.0345252774352651, + "grad_norm": 2.9215831263409013, + "learning_rate": 1.0800000000000002e-05, + "loss": 1.3221, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 28 + }, + { + "epoch": 0.035758323057953144, + "grad_norm": 2.97321184472987, + "learning_rate": 1.1200000000000001e-05, + "loss": 1.3247, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 29 + }, + { + "epoch": 0.036991368680641186, + "grad_norm": 3.293476484963883, + "learning_rate": 1.16e-05, + "loss": 1.2582, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 30 + }, + { + "epoch": 0.03822441430332922, + "grad_norm": 2.8642108219246745, + "learning_rate": 1.2e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 31 + }, + { + "epoch": 0.03945745992601726, + "grad_norm": 2.8916726488438824, + "learning_rate": 1.2400000000000002e-05, + "loss": 1.2628, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 32 + }, + { + "epoch": 0.040690505548705305, + "grad_norm": 3.17999982306396, + "learning_rate": 1.2800000000000001e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 33 + }, + { + "epoch": 0.04192355117139334, + "grad_norm": 2.6965579635696453, + "learning_rate": 1.3200000000000002e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 34 + }, + { + "epoch": 0.04315659679408138, + "grad_norm": 2.9489592197594896, + "learning_rate": 1.3600000000000002e-05, + "loss": 1.3113, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 35 + }, + { + "epoch": 0.04438964241676942, + "grad_norm": 2.8898479257030734, + "learning_rate": 1.4e-05, + "loss": 1.3387, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 36 + }, + { + "epoch": 0.04562268803945746, + "grad_norm": 2.6370539695770425, + "learning_rate": 1.4400000000000001e-05, + "loss": 1.3314, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 37 + }, + { + "epoch": 0.0468557336621455, + "grad_norm": 2.7915119494225813, + "learning_rate": 1.48e-05, + "loss": 1.2603, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 38 + }, + { + "epoch": 0.04808877928483354, + "grad_norm": 2.602735589719503, + "learning_rate": 1.5200000000000002e-05, + "loss": 1.3492, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 39 + }, + { + "epoch": 0.04932182490752158, + "grad_norm": 2.9123309438390113, + "learning_rate": 1.5600000000000003e-05, + "loss": 1.2291, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 40 + }, + { + "epoch": 0.05055487053020962, + "grad_norm": 2.577807086473754, + "learning_rate": 1.6000000000000003e-05, + "loss": 1.3098, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 41 + }, + { + "epoch": 0.051787916152897656, + "grad_norm": 2.548383978805174, + "learning_rate": 1.64e-05, + "loss": 1.2893, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 42 + }, + { + "epoch": 0.0530209617755857, + "grad_norm": 2.6372216084932814, + "learning_rate": 1.6800000000000002e-05, + "loss": 1.2701, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 43 + }, + { + "epoch": 0.05425400739827373, + "grad_norm": 2.60286437866227, + "learning_rate": 1.72e-05, + "loss": 1.3579, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 44 + }, + { + "epoch": 0.055487053020961775, + "grad_norm": 2.536071820500707, + "learning_rate": 1.76e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 45 + }, + { + "epoch": 0.05672009864364982, + "grad_norm": 2.8143481627767963, + "learning_rate": 1.8e-05, + "loss": 1.1729, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 46 + }, + { + "epoch": 0.05795314426633785, + "grad_norm": 2.8600643284872596, + "learning_rate": 1.8400000000000003e-05, + "loss": 1.3246, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 47 + }, + { + "epoch": 0.059186189889025895, + "grad_norm": 2.618463019216554, + "learning_rate": 1.88e-05, + "loss": 1.2878, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 48 + }, + { + "epoch": 0.06041923551171394, + "grad_norm": 2.757496450593693, + "learning_rate": 1.9200000000000003e-05, + "loss": 1.3454, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 49 + }, + { + "epoch": 0.06165228113440197, + "grad_norm": 2.5356573576135126, + "learning_rate": 1.9600000000000002e-05, + "loss": 1.2719, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 50 + }, + { + "epoch": 0.06288532675709001, + "grad_norm": 2.532419474064049, + "learning_rate": 2e-05, + "loss": 1.332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 51 + }, + { + "epoch": 0.06411837237977805, + "grad_norm": 2.711902763412396, + "learning_rate": 1.999999130996347e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 52 + }, + { + "epoch": 0.06535141800246609, + "grad_norm": 2.654767047589697, + "learning_rate": 1.9999965239868983e-05, + "loss": 1.3092, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 53 + }, + { + "epoch": 0.06658446362515413, + "grad_norm": 3.2372229064932454, + "learning_rate": 1.9999921789761843e-05, + "loss": 1.3428, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 54 + }, + { + "epoch": 0.06781750924784218, + "grad_norm": 2.723103510772695, + "learning_rate": 1.999986095971757e-05, + "loss": 1.3918, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 55 + }, + { + "epoch": 0.0690505548705302, + "grad_norm": 2.849098835881172, + "learning_rate": 1.9999782749841895e-05, + "loss": 1.3304, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 56 + }, + { + "epoch": 0.07028360049321825, + "grad_norm": 2.4115082044348557, + "learning_rate": 1.9999687160270735e-05, + "loss": 1.2818, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 57 + }, + { + "epoch": 0.07151664611590629, + "grad_norm": 2.7720437269939495, + "learning_rate": 1.9999574191170233e-05, + "loss": 1.2799, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 58 + }, + { + "epoch": 0.07274969173859433, + "grad_norm": 2.4891158758553042, + "learning_rate": 1.9999443842736723e-05, + "loss": 1.2408, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 59 + }, + { + "epoch": 0.07398273736128237, + "grad_norm": 2.8066935220153737, + "learning_rate": 1.999929611519676e-05, + "loss": 1.3157, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 60 + }, + { + "epoch": 0.0752157829839704, + "grad_norm": 2.6715709033170705, + "learning_rate": 1.9999131008807094e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 61 + }, + { + "epoch": 0.07644882860665844, + "grad_norm": 2.586596297320694, + "learning_rate": 1.9998948523854676e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 62 + }, + { + "epoch": 0.07768187422934648, + "grad_norm": 2.646887045495792, + "learning_rate": 1.999874866065667e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 63 + }, + { + "epoch": 0.07891491985203453, + "grad_norm": 2.419730330481743, + "learning_rate": 1.999853141956044e-05, + "loss": 1.2829, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 64 + }, + { + "epoch": 0.08014796547472257, + "grad_norm": 2.5201815842521245, + "learning_rate": 1.999829680094355e-05, + "loss": 1.3243, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 65 + }, + { + "epoch": 0.08138101109741061, + "grad_norm": 2.503429794099987, + "learning_rate": 1.9998044805213773e-05, + "loss": 1.2681, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 66 + }, + { + "epoch": 0.08261405672009864, + "grad_norm": 2.348439339795846, + "learning_rate": 1.9997775432809074e-05, + "loss": 1.3359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 67 + }, + { + "epoch": 0.08384710234278668, + "grad_norm": 2.3828095918277357, + "learning_rate": 1.999748868419763e-05, + "loss": 1.3526, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 68 + }, + { + "epoch": 0.08508014796547472, + "grad_norm": 2.5076969489265273, + "learning_rate": 1.9997184559877806e-05, + "loss": 1.3531, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 69 + }, + { + "epoch": 0.08631319358816276, + "grad_norm": 2.388938276200468, + "learning_rate": 1.9996863060378176e-05, + "loss": 1.334, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 70 + }, + { + "epoch": 0.08754623921085081, + "grad_norm": 2.283154945083281, + "learning_rate": 1.999652418625751e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 71 + }, + { + "epoch": 0.08877928483353884, + "grad_norm": 2.2613574311048734, + "learning_rate": 1.9996167938104767e-05, + "loss": 1.2858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 72 + }, + { + "epoch": 0.09001233045622688, + "grad_norm": 2.3931162231369076, + "learning_rate": 1.999579431653912e-05, + "loss": 1.3089, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 73 + }, + { + "epoch": 0.09124537607891492, + "grad_norm": 2.5073647819033194, + "learning_rate": 1.9995403322209913e-05, + "loss": 1.382, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 74 + }, + { + "epoch": 0.09247842170160296, + "grad_norm": 2.5388011932611536, + "learning_rate": 1.999499495579671e-05, + "loss": 1.402, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 75 + }, + { + "epoch": 0.093711467324291, + "grad_norm": 2.3884809904992252, + "learning_rate": 1.9994569218009247e-05, + "loss": 1.3079, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 76 + }, + { + "epoch": 0.09494451294697903, + "grad_norm": 2.3454483578272094, + "learning_rate": 1.999412610958746e-05, + "loss": 1.2653, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 77 + }, + { + "epoch": 0.09617755856966707, + "grad_norm": 2.186902835408518, + "learning_rate": 1.9993665631301474e-05, + "loss": 1.2069, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 78 + }, + { + "epoch": 0.09741060419235512, + "grad_norm": 2.3845058131797368, + "learning_rate": 1.9993187783951603e-05, + "loss": 1.3541, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 79 + }, + { + "epoch": 0.09864364981504316, + "grad_norm": 2.256399889979939, + "learning_rate": 1.9992692568368354e-05, + "loss": 1.2119, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 80 + }, + { + "epoch": 0.0998766954377312, + "grad_norm": 2.3361092398369903, + "learning_rate": 1.9992179985412416e-05, + "loss": 1.2585, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 81 + }, + { + "epoch": 0.10110974106041924, + "grad_norm": 2.289061082197673, + "learning_rate": 1.999165003597465e-05, + "loss": 1.3459, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 82 + }, + { + "epoch": 0.10234278668310727, + "grad_norm": 2.228187448538031, + "learning_rate": 1.999110272097613e-05, + "loss": 1.3739, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 83 + }, + { + "epoch": 0.10357583230579531, + "grad_norm": 2.157361593613978, + "learning_rate": 1.9990538041368078e-05, + "loss": 1.2575, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 84 + }, + { + "epoch": 0.10480887792848335, + "grad_norm": 2.294809937254642, + "learning_rate": 1.998995599813192e-05, + "loss": 1.2834, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 85 + }, + { + "epoch": 0.1060419235511714, + "grad_norm": 2.1831743941808925, + "learning_rate": 1.998935659227925e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 86 + }, + { + "epoch": 0.10727496917385944, + "grad_norm": 2.247228611971362, + "learning_rate": 1.998873982485183e-05, + "loss": 1.2654, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 87 + }, + { + "epoch": 0.10850801479654747, + "grad_norm": 2.2844560193856416, + "learning_rate": 1.9988105696921622e-05, + "loss": 1.2927, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 88 + }, + { + "epoch": 0.10974106041923551, + "grad_norm": 2.3780157649218157, + "learning_rate": 1.9987454209590736e-05, + "loss": 1.2621, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 89 + }, + { + "epoch": 0.11097410604192355, + "grad_norm": 2.3017268441414003, + "learning_rate": 1.9986785363991466e-05, + "loss": 1.2676, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 90 + }, + { + "epoch": 0.11220715166461159, + "grad_norm": 2.1906552279173503, + "learning_rate": 1.9986099161286266e-05, + "loss": 1.2281, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 91 + }, + { + "epoch": 0.11344019728729964, + "grad_norm": 2.4009419002081276, + "learning_rate": 1.9985395602667764e-05, + "loss": 1.2267, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 92 + }, + { + "epoch": 0.11467324290998766, + "grad_norm": 2.190217063330867, + "learning_rate": 1.9984674689358744e-05, + "loss": 1.2778, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 93 + }, + { + "epoch": 0.1159062885326757, + "grad_norm": 2.305624793209019, + "learning_rate": 1.998393642261217e-05, + "loss": 1.2821, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 94 + }, + { + "epoch": 0.11713933415536375, + "grad_norm": 2.162537568697834, + "learning_rate": 1.9983180803711146e-05, + "loss": 1.3775, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 95 + }, + { + "epoch": 0.11837237977805179, + "grad_norm": 2.2632096292316706, + "learning_rate": 1.9982407833968947e-05, + "loss": 1.3358, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 96 + }, + { + "epoch": 0.11960542540073983, + "grad_norm": 2.268508818243908, + "learning_rate": 1.9981617514728998e-05, + "loss": 1.2368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 97 + }, + { + "epoch": 0.12083847102342787, + "grad_norm": 2.2721363895167377, + "learning_rate": 1.9980809847364884e-05, + "loss": 1.2774, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 98 + }, + { + "epoch": 0.1220715166461159, + "grad_norm": 2.1661250921439974, + "learning_rate": 1.9979984833280332e-05, + "loss": 1.359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 99 + }, + { + "epoch": 0.12330456226880394, + "grad_norm": 2.2521676172131873, + "learning_rate": 1.997914247390922e-05, + "loss": 1.216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 100 + }, + { + "epoch": 0.12453760789149199, + "grad_norm": 2.3907655242108095, + "learning_rate": 1.9978282770715586e-05, + "loss": 1.2468, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 101 + }, + { + "epoch": 0.12577065351418001, + "grad_norm": 2.2007275768445074, + "learning_rate": 1.9977405725193587e-05, + "loss": 1.2415, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 102 + }, + { + "epoch": 0.12700369913686807, + "grad_norm": 2.3304230330598346, + "learning_rate": 1.9976511338867542e-05, + "loss": 1.2438, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 103 + }, + { + "epoch": 0.1282367447595561, + "grad_norm": 2.346171964705753, + "learning_rate": 1.99755996132919e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 104 + }, + { + "epoch": 0.12946979038224415, + "grad_norm": 2.286623534979458, + "learning_rate": 1.9974670550051248e-05, + "loss": 1.2897, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 105 + }, + { + "epoch": 0.13070283600493218, + "grad_norm": 2.245774575899364, + "learning_rate": 1.99737241507603e-05, + "loss": 1.2923, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 106 + }, + { + "epoch": 0.1319358816276202, + "grad_norm": 2.2352297973588953, + "learning_rate": 1.9972760417063912e-05, + "loss": 1.1434, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 107 + }, + { + "epoch": 0.13316892725030827, + "grad_norm": 2.169802916370051, + "learning_rate": 1.9971779350637054e-05, + "loss": 1.2353, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 108 + }, + { + "epoch": 0.1344019728729963, + "grad_norm": 2.333533512175979, + "learning_rate": 1.9970780953184828e-05, + "loss": 1.2641, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 109 + }, + { + "epoch": 0.13563501849568435, + "grad_norm": 2.2985803581198674, + "learning_rate": 1.996976522644246e-05, + "loss": 1.3036, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 110 + }, + { + "epoch": 0.13686806411837238, + "grad_norm": 2.0992498120010272, + "learning_rate": 1.996873217217528e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 111 + }, + { + "epoch": 0.1381011097410604, + "grad_norm": 2.396895991067544, + "learning_rate": 1.9967681792178758e-05, + "loss": 1.2533, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 112 + }, + { + "epoch": 0.13933415536374846, + "grad_norm": 2.0738932505082768, + "learning_rate": 1.9966614088278456e-05, + "loss": 1.3332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 113 + }, + { + "epoch": 0.1405672009864365, + "grad_norm": 2.1049156004628395, + "learning_rate": 1.9965529062330048e-05, + "loss": 1.2793, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 114 + }, + { + "epoch": 0.14180024660912455, + "grad_norm": 2.2137820282577847, + "learning_rate": 1.9964426716219316e-05, + "loss": 1.1932, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 115 + }, + { + "epoch": 0.14303329223181258, + "grad_norm": 2.1545610182661976, + "learning_rate": 1.9963307051862152e-05, + "loss": 1.2662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 116 + }, + { + "epoch": 0.1442663378545006, + "grad_norm": 2.151286882718536, + "learning_rate": 1.9962170071204542e-05, + "loss": 1.205, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 117 + }, + { + "epoch": 0.14549938347718866, + "grad_norm": 2.026608825482131, + "learning_rate": 1.996101577622256e-05, + "loss": 1.195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 118 + }, + { + "epoch": 0.1467324290998767, + "grad_norm": 2.156975981742732, + "learning_rate": 1.9959844168922377e-05, + "loss": 1.3284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 119 + }, + { + "epoch": 0.14796547472256474, + "grad_norm": 2.2311865782919025, + "learning_rate": 1.9958655251340265e-05, + "loss": 1.2183, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 120 + }, + { + "epoch": 0.14919852034525277, + "grad_norm": 2.2940740150536376, + "learning_rate": 1.9957449025542567e-05, + "loss": 1.2625, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 121 + }, + { + "epoch": 0.1504315659679408, + "grad_norm": 2.214218654766027, + "learning_rate": 1.9956225493625708e-05, + "loss": 1.2177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 122 + }, + { + "epoch": 0.15166461159062886, + "grad_norm": 2.1690434056776775, + "learning_rate": 1.9954984657716202e-05, + "loss": 1.2049, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 123 + }, + { + "epoch": 0.15289765721331688, + "grad_norm": 2.1450511034984046, + "learning_rate": 1.9953726519970625e-05, + "loss": 1.232, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 124 + }, + { + "epoch": 0.15413070283600494, + "grad_norm": 2.0475086343318543, + "learning_rate": 1.995245108257563e-05, + "loss": 1.2665, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 125 + }, + { + "epoch": 0.15536374845869297, + "grad_norm": 2.1232594967339318, + "learning_rate": 1.9951158347747945e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 126 + }, + { + "epoch": 0.15659679408138102, + "grad_norm": 2.1915516478247308, + "learning_rate": 1.9949848317734345e-05, + "loss": 1.177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 127 + }, + { + "epoch": 0.15782983970406905, + "grad_norm": 2.2848682922250205, + "learning_rate": 1.9948520994811674e-05, + "loss": 1.2406, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 128 + }, + { + "epoch": 0.15906288532675708, + "grad_norm": 2.2223107161107505, + "learning_rate": 1.9947176381286825e-05, + "loss": 1.327, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 129 + }, + { + "epoch": 0.16029593094944514, + "grad_norm": 2.1247307489434837, + "learning_rate": 1.994581447949675e-05, + "loss": 1.1634, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 130 + }, + { + "epoch": 0.16152897657213316, + "grad_norm": 1.926245615401859, + "learning_rate": 1.9944435291808445e-05, + "loss": 1.1655, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 131 + }, + { + "epoch": 0.16276202219482122, + "grad_norm": 2.2531819769934858, + "learning_rate": 1.9943038820618946e-05, + "loss": 1.2216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 132 + }, + { + "epoch": 0.16399506781750925, + "grad_norm": 2.2183949614858305, + "learning_rate": 1.994162506835533e-05, + "loss": 1.2202, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 133 + }, + { + "epoch": 0.16522811344019728, + "grad_norm": 2.164554433266706, + "learning_rate": 1.994019403747471e-05, + "loss": 1.2245, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 134 + }, + { + "epoch": 0.16646115906288533, + "grad_norm": 2.17653642461108, + "learning_rate": 1.9938745730464228e-05, + "loss": 1.1899, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 135 + }, + { + "epoch": 0.16769420468557336, + "grad_norm": 1.9936998341266101, + "learning_rate": 1.9937280149841055e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 136 + }, + { + "epoch": 0.16892725030826142, + "grad_norm": 2.27069520590907, + "learning_rate": 1.9935797298152374e-05, + "loss": 1.3023, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 137 + }, + { + "epoch": 0.17016029593094945, + "grad_norm": 2.244074718609361, + "learning_rate": 1.9934297177975397e-05, + "loss": 1.3292, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 138 + }, + { + "epoch": 0.17139334155363747, + "grad_norm": 2.2121305803685094, + "learning_rate": 1.9932779791917343e-05, + "loss": 1.3028, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 139 + }, + { + "epoch": 0.17262638717632553, + "grad_norm": 2.7262731821688697, + "learning_rate": 1.9931245142615442e-05, + "loss": 1.2861, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 140 + }, + { + "epoch": 0.17385943279901356, + "grad_norm": 2.2051993935523497, + "learning_rate": 1.992969323273692e-05, + "loss": 1.3134, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 141 + }, + { + "epoch": 0.17509247842170161, + "grad_norm": 2.088688387254833, + "learning_rate": 1.992812406497901e-05, + "loss": 1.262, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 142 + }, + { + "epoch": 0.17632552404438964, + "grad_norm": 2.19652156002121, + "learning_rate": 1.9926537642068942e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 143 + }, + { + "epoch": 0.17755856966707767, + "grad_norm": 1.887489027726817, + "learning_rate": 1.9924933966763924e-05, + "loss": 1.2847, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 144 + }, + { + "epoch": 0.17879161528976573, + "grad_norm": 2.088506357330406, + "learning_rate": 1.992331304185116e-05, + "loss": 1.2218, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 145 + }, + { + "epoch": 0.18002466091245375, + "grad_norm": 2.102792679186299, + "learning_rate": 1.9921674870147827e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 146 + }, + { + "epoch": 0.1812577065351418, + "grad_norm": 2.0847095844816015, + "learning_rate": 1.992001945450108e-05, + "loss": 1.2986, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 147 + }, + { + "epoch": 0.18249075215782984, + "grad_norm": 2.344867998715345, + "learning_rate": 1.9918346797788045e-05, + "loss": 1.1347, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 148 + }, + { + "epoch": 0.18372379778051787, + "grad_norm": 2.1013292606380602, + "learning_rate": 1.991665690291581e-05, + "loss": 1.1319, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 149 + }, + { + "epoch": 0.18495684340320592, + "grad_norm": 2.2475258887436538, + "learning_rate": 1.9914949772821418e-05, + "loss": 1.1648, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 150 + }, + { + "epoch": 0.18618988902589395, + "grad_norm": 2.223126056771945, + "learning_rate": 1.9913225410471888e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 151 + }, + { + "epoch": 0.187422934648582, + "grad_norm": 2.102849036750559, + "learning_rate": 1.9911483818864163e-05, + "loss": 1.2062, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 152 + }, + { + "epoch": 0.18865598027127004, + "grad_norm": 2.2107888348018805, + "learning_rate": 1.9909725001025146e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 153 + }, + { + "epoch": 0.18988902589395806, + "grad_norm": 2.1730720194510456, + "learning_rate": 1.9907948960011677e-05, + "loss": 1.2133, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 154 + }, + { + "epoch": 0.19112207151664612, + "grad_norm": 2.0874944224993115, + "learning_rate": 1.9906155698910523e-05, + "loss": 1.1966, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 155 + }, + { + "epoch": 0.19235511713933415, + "grad_norm": 2.047788318815714, + "learning_rate": 1.990434522083839e-05, + "loss": 1.26, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 156 + }, + { + "epoch": 0.1935881627620222, + "grad_norm": 2.0992674413368415, + "learning_rate": 1.99025175289419e-05, + "loss": 1.2544, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 157 + }, + { + "epoch": 0.19482120838471023, + "grad_norm": 2.279539056045867, + "learning_rate": 1.9900672626397593e-05, + "loss": 1.3123, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 158 + }, + { + "epoch": 0.1960542540073983, + "grad_norm": 1.9940111001673004, + "learning_rate": 1.9898810516411926e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 159 + }, + { + "epoch": 0.19728729963008632, + "grad_norm": 2.278268154556979, + "learning_rate": 1.989693120222126e-05, + "loss": 1.236, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 160 + }, + { + "epoch": 0.19852034525277434, + "grad_norm": 2.3184527515323796, + "learning_rate": 1.9895034687091854e-05, + "loss": 1.1691, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 161 + }, + { + "epoch": 0.1997533908754624, + "grad_norm": 2.094943513595831, + "learning_rate": 1.9893120974319867e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 162 + }, + { + "epoch": 0.20098643649815043, + "grad_norm": 2.296308593300257, + "learning_rate": 1.9891190067231346e-05, + "loss": 1.2976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 163 + }, + { + "epoch": 0.20221948212083848, + "grad_norm": 2.131546792333376, + "learning_rate": 1.988924196918222e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 164 + }, + { + "epoch": 0.2034525277435265, + "grad_norm": 1.981669864872828, + "learning_rate": 1.9887276683558302e-05, + "loss": 1.2934, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 165 + }, + { + "epoch": 0.20468557336621454, + "grad_norm": 2.0382654138005423, + "learning_rate": 1.9885294213775267e-05, + "loss": 1.3207, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 166 + }, + { + "epoch": 0.2059186189889026, + "grad_norm": 2.074242050982157, + "learning_rate": 1.9883294563278665e-05, + "loss": 1.1832, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 167 + }, + { + "epoch": 0.20715166461159062, + "grad_norm": 2.1994142392378317, + "learning_rate": 1.9881277735543902e-05, + "loss": 1.2284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 168 + }, + { + "epoch": 0.20838471023427868, + "grad_norm": 2.1893568284451987, + "learning_rate": 1.987924373407624e-05, + "loss": 1.2572, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 169 + }, + { + "epoch": 0.2096177558569667, + "grad_norm": 2.0490903431710352, + "learning_rate": 1.987719256241079e-05, + "loss": 1.2808, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 170 + }, + { + "epoch": 0.21085080147965474, + "grad_norm": 2.3424605103097176, + "learning_rate": 1.9875124224112498e-05, + "loss": 1.2924, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 171 + }, + { + "epoch": 0.2120838471023428, + "grad_norm": 2.100563128603325, + "learning_rate": 1.987303872277616e-05, + "loss": 1.1368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 172 + }, + { + "epoch": 0.21331689272503082, + "grad_norm": 2.303295425870437, + "learning_rate": 1.9870936062026385e-05, + "loss": 1.3425, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 173 + }, + { + "epoch": 0.21454993834771888, + "grad_norm": 1.9108233071934615, + "learning_rate": 1.9868816245517615e-05, + "loss": 1.1926, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 174 + }, + { + "epoch": 0.2157829839704069, + "grad_norm": 2.2539584509521173, + "learning_rate": 1.986667927693411e-05, + "loss": 1.222, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 175 + }, + { + "epoch": 0.21701602959309493, + "grad_norm": 2.006165392274905, + "learning_rate": 1.986452515998993e-05, + "loss": 1.2853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 176 + }, + { + "epoch": 0.218249075215783, + "grad_norm": 2.1337815004505454, + "learning_rate": 1.9862353898428948e-05, + "loss": 1.2671, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 177 + }, + { + "epoch": 0.21948212083847102, + "grad_norm": 2.062911126777883, + "learning_rate": 1.986016549602484e-05, + "loss": 1.2397, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 178 + }, + { + "epoch": 0.22071516646115907, + "grad_norm": 2.115740291239205, + "learning_rate": 1.9857959956581053e-05, + "loss": 1.2534, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 179 + }, + { + "epoch": 0.2219482120838471, + "grad_norm": 2.3733626044353837, + "learning_rate": 1.9855737283930843e-05, + "loss": 1.2082, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 180 + }, + { + "epoch": 0.22318125770653513, + "grad_norm": 2.0972006372187373, + "learning_rate": 1.985349748193722e-05, + "loss": 1.1091, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 181 + }, + { + "epoch": 0.22441430332922319, + "grad_norm": 2.0727561431265635, + "learning_rate": 1.9851240554492985e-05, + "loss": 1.2153, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 182 + }, + { + "epoch": 0.22564734895191121, + "grad_norm": 2.094605431898615, + "learning_rate": 1.9848966505520694e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 183 + }, + { + "epoch": 0.22688039457459927, + "grad_norm": 2.1524188959905866, + "learning_rate": 1.9846675338972654e-05, + "loss": 1.219, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 184 + }, + { + "epoch": 0.2281134401972873, + "grad_norm": 2.008037870627797, + "learning_rate": 1.9844367058830933e-05, + "loss": 1.2635, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 185 + }, + { + "epoch": 0.22934648581997533, + "grad_norm": 2.050448462235658, + "learning_rate": 1.984204166910734e-05, + "loss": 1.187, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 186 + }, + { + "epoch": 0.23057953144266338, + "grad_norm": 2.2727997887406066, + "learning_rate": 1.983969917384342e-05, + "loss": 1.2709, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 187 + }, + { + "epoch": 0.2318125770653514, + "grad_norm": 2.0295960774209414, + "learning_rate": 1.9837339577110443e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 188 + }, + { + "epoch": 0.23304562268803947, + "grad_norm": 2.250775161795276, + "learning_rate": 1.9834962883009406e-05, + "loss": 1.2566, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 189 + }, + { + "epoch": 0.2342786683107275, + "grad_norm": 2.221930554057359, + "learning_rate": 1.9832569095671025e-05, + "loss": 1.2547, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 190 + }, + { + "epoch": 0.23551171393341552, + "grad_norm": 2.071699318441252, + "learning_rate": 1.9830158219255715e-05, + "loss": 1.1712, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 191 + }, + { + "epoch": 0.23674475955610358, + "grad_norm": 1.9444133453118777, + "learning_rate": 1.98277302579536e-05, + "loss": 1.1858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 192 + }, + { + "epoch": 0.2379778051787916, + "grad_norm": 2.134049888070869, + "learning_rate": 1.9825285215984495e-05, + "loss": 1.2212, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 193 + }, + { + "epoch": 0.23921085080147966, + "grad_norm": 2.1418697560189988, + "learning_rate": 1.9822823097597893e-05, + "loss": 1.2602, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 194 + }, + { + "epoch": 0.2404438964241677, + "grad_norm": 2.049709663111986, + "learning_rate": 1.9820343907072982e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 195 + }, + { + "epoch": 0.24167694204685575, + "grad_norm": 2.1074579631320605, + "learning_rate": 1.9817847648718615e-05, + "loss": 1.1863, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 196 + }, + { + "epoch": 0.24290998766954378, + "grad_norm": 2.279482709995341, + "learning_rate": 1.9815334326873304e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 197 + }, + { + "epoch": 0.2441430332922318, + "grad_norm": 2.055596994912923, + "learning_rate": 1.9812803945905215e-05, + "loss": 1.2952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 198 + }, + { + "epoch": 0.24537607891491986, + "grad_norm": 2.132145299143716, + "learning_rate": 1.9810256510212178e-05, + "loss": 1.2276, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 199 + }, + { + "epoch": 0.2466091245376079, + "grad_norm": 2.077623874214478, + "learning_rate": 1.9807692024221647e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 200 + }, + { + "epoch": 0.24784217016029594, + "grad_norm": 2.0797643511906356, + "learning_rate": 1.9805110492390723e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 201 + }, + { + "epoch": 0.24907521578298397, + "grad_norm": 2.0930244790953854, + "learning_rate": 1.9802511919206123e-05, + "loss": 1.1984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 202 + }, + { + "epoch": 0.25030826140567203, + "grad_norm": 1.9804465011478614, + "learning_rate": 1.9799896309184192e-05, + "loss": 1.1436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 203 + }, + { + "epoch": 0.25154130702836003, + "grad_norm": 2.054252771935451, + "learning_rate": 1.9797263666870876e-05, + "loss": 1.1913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 204 + }, + { + "epoch": 0.2527743526510481, + "grad_norm": 1.9941485664509049, + "learning_rate": 1.9794613996841723e-05, + "loss": 1.1524, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 205 + }, + { + "epoch": 0.25400739827373614, + "grad_norm": 2.051706932133756, + "learning_rate": 1.9791947303701882e-05, + "loss": 1.2003, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 206 + }, + { + "epoch": 0.25524044389642414, + "grad_norm": 1.9992360549131547, + "learning_rate": 1.9789263592086085e-05, + "loss": 1.2392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 207 + }, + { + "epoch": 0.2564734895191122, + "grad_norm": 2.0558176906965375, + "learning_rate": 1.9786562866658644e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 208 + }, + { + "epoch": 0.25770653514180025, + "grad_norm": 2.2225484776914426, + "learning_rate": 1.9783845132113438e-05, + "loss": 1.1771, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 209 + }, + { + "epoch": 0.2589395807644883, + "grad_norm": 2.2832873135247516, + "learning_rate": 1.978111039317391e-05, + "loss": 1.2226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 210 + }, + { + "epoch": 0.2601726263871763, + "grad_norm": 1.9887162353358598, + "learning_rate": 1.9778358654593054e-05, + "loss": 1.1936, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 211 + }, + { + "epoch": 0.26140567200986436, + "grad_norm": 1.9738209615627333, + "learning_rate": 1.9775589921153417e-05, + "loss": 1.2666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 212 + }, + { + "epoch": 0.2626387176325524, + "grad_norm": 1.8817044406816048, + "learning_rate": 1.9772804197667074e-05, + "loss": 1.1842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 213 + }, + { + "epoch": 0.2638717632552404, + "grad_norm": 2.013799438378123, + "learning_rate": 1.9770001488975632e-05, + "loss": 1.2242, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 214 + }, + { + "epoch": 0.2651048088779285, + "grad_norm": 1.9119015873636653, + "learning_rate": 1.9767181799950222e-05, + "loss": 1.0898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 215 + }, + { + "epoch": 0.26633785450061653, + "grad_norm": 1.9353703261833344, + "learning_rate": 1.976434513549148e-05, + "loss": 1.2241, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 216 + }, + { + "epoch": 0.2675709001233046, + "grad_norm": 1.8684279347954125, + "learning_rate": 1.9761491500529555e-05, + "loss": 1.209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 217 + }, + { + "epoch": 0.2688039457459926, + "grad_norm": 2.1880109676776556, + "learning_rate": 1.9758620900024083e-05, + "loss": 1.1956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 218 + }, + { + "epoch": 0.27003699136868065, + "grad_norm": 2.083271319194177, + "learning_rate": 1.9755733338964186e-05, + "loss": 1.3062, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 219 + }, + { + "epoch": 0.2712700369913687, + "grad_norm": 2.093319969744569, + "learning_rate": 1.9752828822368468e-05, + "loss": 1.2083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 220 + }, + { + "epoch": 0.2725030826140567, + "grad_norm": 1.8811834241518346, + "learning_rate": 1.9749907355285e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 221 + }, + { + "epoch": 0.27373612823674476, + "grad_norm": 2.012899236904161, + "learning_rate": 1.9746968942791314e-05, + "loss": 1.1476, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 222 + }, + { + "epoch": 0.2749691738594328, + "grad_norm": 1.9990139044183095, + "learning_rate": 1.9744013589994397e-05, + "loss": 1.1257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 223 + }, + { + "epoch": 0.2762022194821208, + "grad_norm": 1.9924345221184356, + "learning_rate": 1.9741041302030666e-05, + "loss": 1.224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 224 + }, + { + "epoch": 0.27743526510480887, + "grad_norm": 2.2180474451395344, + "learning_rate": 1.9738052084065982e-05, + "loss": 1.1713, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 225 + }, + { + "epoch": 0.2786683107274969, + "grad_norm": 1.983690334551248, + "learning_rate": 1.973504594129563e-05, + "loss": 1.179, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 226 + }, + { + "epoch": 0.279901356350185, + "grad_norm": 2.283777738854751, + "learning_rate": 1.97320228789443e-05, + "loss": 1.2869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 227 + }, + { + "epoch": 0.281134401972873, + "grad_norm": 1.943369966527272, + "learning_rate": 1.9728982902266107e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 228 + }, + { + "epoch": 0.28236744759556104, + "grad_norm": 2.0711638809016333, + "learning_rate": 1.9725926016544544e-05, + "loss": 1.16, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 229 + }, + { + "epoch": 0.2836004932182491, + "grad_norm": 1.9786842049852629, + "learning_rate": 1.972285222709251e-05, + "loss": 1.2338, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 230 + }, + { + "epoch": 0.2848335388409371, + "grad_norm": 1.9447068528661613, + "learning_rate": 1.9719761539252268e-05, + "loss": 1.1827, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 231 + }, + { + "epoch": 0.28606658446362515, + "grad_norm": 2.0161169975225475, + "learning_rate": 1.971665395839545e-05, + "loss": 1.117, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 232 + }, + { + "epoch": 0.2872996300863132, + "grad_norm": 2.181216178694371, + "learning_rate": 1.9713529489923067e-05, + "loss": 1.1807, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 233 + }, + { + "epoch": 0.2885326757090012, + "grad_norm": 2.254001309068264, + "learning_rate": 1.9710388139265458e-05, + "loss": 1.0036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 234 + }, + { + "epoch": 0.28976572133168926, + "grad_norm": 2.028042430685245, + "learning_rate": 1.9707229911882318e-05, + "loss": 1.1175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 235 + }, + { + "epoch": 0.2909987669543773, + "grad_norm": 2.0601619279255177, + "learning_rate": 1.9704054813262666e-05, + "loss": 1.2758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 236 + }, + { + "epoch": 0.2922318125770654, + "grad_norm": 2.125107661538684, + "learning_rate": 1.9700862848924852e-05, + "loss": 1.2611, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 237 + }, + { + "epoch": 0.2934648581997534, + "grad_norm": 2.0416663193733746, + "learning_rate": 1.9697654024416528e-05, + "loss": 1.1986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 238 + }, + { + "epoch": 0.29469790382244143, + "grad_norm": 2.0193702602947554, + "learning_rate": 1.9694428345314657e-05, + "loss": 1.181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 239 + }, + { + "epoch": 0.2959309494451295, + "grad_norm": 2.0150819278141148, + "learning_rate": 1.9691185817225493e-05, + "loss": 1.1118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 240 + }, + { + "epoch": 0.2971639950678175, + "grad_norm": 2.355208602911491, + "learning_rate": 1.9687926445784576e-05, + "loss": 1.1013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 241 + }, + { + "epoch": 0.29839704069050554, + "grad_norm": 1.9770821951357245, + "learning_rate": 1.968465023665671e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 242 + }, + { + "epoch": 0.2996300863131936, + "grad_norm": 1.992866162682536, + "learning_rate": 1.9681357195535974e-05, + "loss": 1.2264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 243 + }, + { + "epoch": 0.3008631319358816, + "grad_norm": 1.9560049490402185, + "learning_rate": 1.9678047328145702e-05, + "loss": 1.1903, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 244 + }, + { + "epoch": 0.30209617755856966, + "grad_norm": 2.0070855957858114, + "learning_rate": 1.967472064023846e-05, + "loss": 1.1793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 245 + }, + { + "epoch": 0.3033292231812577, + "grad_norm": 2.0661988646186313, + "learning_rate": 1.9671377137596064e-05, + "loss": 1.2222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 246 + }, + { + "epoch": 0.30456226880394577, + "grad_norm": 2.1056352825098816, + "learning_rate": 1.9668016826029538e-05, + "loss": 1.2107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 247 + }, + { + "epoch": 0.30579531442663377, + "grad_norm": 1.98815122638957, + "learning_rate": 1.9664639711379135e-05, + "loss": 1.2816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 248 + }, + { + "epoch": 0.3070283600493218, + "grad_norm": 2.1440514280412737, + "learning_rate": 1.9661245799514298e-05, + "loss": 1.2021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 249 + }, + { + "epoch": 0.3082614056720099, + "grad_norm": 2.0231899704902805, + "learning_rate": 1.9657835096333676e-05, + "loss": 1.2358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 250 + }, + { + "epoch": 0.3094944512946979, + "grad_norm": 1.9883606430802652, + "learning_rate": 1.9654407607765098e-05, + "loss": 1.1847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 251 + }, + { + "epoch": 0.31072749691738594, + "grad_norm": 2.1163621969666786, + "learning_rate": 1.965096333976556e-05, + "loss": 1.2595, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 252 + }, + { + "epoch": 0.311960542540074, + "grad_norm": 1.9898069180734743, + "learning_rate": 1.9647502298321223e-05, + "loss": 1.2286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 253 + }, + { + "epoch": 0.31319358816276205, + "grad_norm": 1.929003891101706, + "learning_rate": 1.9644024489447403e-05, + "loss": 1.2394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 254 + }, + { + "epoch": 0.31442663378545005, + "grad_norm": 1.98661541473719, + "learning_rate": 1.9640529919188562e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 255 + }, + { + "epoch": 0.3156596794081381, + "grad_norm": 1.9483621179175639, + "learning_rate": 1.9637018593618285e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 256 + }, + { + "epoch": 0.31689272503082616, + "grad_norm": 1.8370569270185535, + "learning_rate": 1.9633490518839285e-05, + "loss": 1.1627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 257 + }, + { + "epoch": 0.31812577065351416, + "grad_norm": 2.2350844635149376, + "learning_rate": 1.9629945700983376e-05, + "loss": 1.1717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 258 + }, + { + "epoch": 0.3193588162762022, + "grad_norm": 2.3938085963940847, + "learning_rate": 1.9626384146211482e-05, + "loss": 1.2982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 259 + }, + { + "epoch": 0.3205918618988903, + "grad_norm": 1.9382250145807791, + "learning_rate": 1.962280586071361e-05, + "loss": 1.1414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 260 + }, + { + "epoch": 0.3218249075215783, + "grad_norm": 2.048034832679974, + "learning_rate": 1.9619210850708844e-05, + "loss": 1.2367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 261 + }, + { + "epoch": 0.32305795314426633, + "grad_norm": 1.868035630300947, + "learning_rate": 1.9615599122445338e-05, + "loss": 1.1582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 262 + }, + { + "epoch": 0.3242909987669544, + "grad_norm": 1.9842993509323321, + "learning_rate": 1.9611970682200308e-05, + "loss": 1.1214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 263 + }, + { + "epoch": 0.32552404438964244, + "grad_norm": 2.009754214636891, + "learning_rate": 1.9608325536280003e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 264 + }, + { + "epoch": 0.32675709001233044, + "grad_norm": 2.0322676023608097, + "learning_rate": 1.9604663691019715e-05, + "loss": 1.1514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 265 + }, + { + "epoch": 0.3279901356350185, + "grad_norm": 1.9898336979024067, + "learning_rate": 1.960098515278376e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 266 + }, + { + "epoch": 0.32922318125770655, + "grad_norm": 2.1120578466218007, + "learning_rate": 1.959728992796546e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 267 + }, + { + "epoch": 0.33045622688039455, + "grad_norm": 2.230681018136028, + "learning_rate": 1.959357802298715e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 268 + }, + { + "epoch": 0.3316892725030826, + "grad_norm": 2.121990574572694, + "learning_rate": 1.9589849444300142e-05, + "loss": 1.1686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 269 + }, + { + "epoch": 0.33292231812577067, + "grad_norm": 1.9793276124371497, + "learning_rate": 1.958610419838473e-05, + "loss": 1.2605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 270 + }, + { + "epoch": 0.33415536374845867, + "grad_norm": 2.160304448339165, + "learning_rate": 1.958234229175019e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 271 + }, + { + "epoch": 0.3353884093711467, + "grad_norm": 2.0375075579110646, + "learning_rate": 1.9578563730934733e-05, + "loss": 1.1282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 272 + }, + { + "epoch": 0.3366214549938348, + "grad_norm": 2.0496730732316313, + "learning_rate": 1.9574768522505528e-05, + "loss": 1.2336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 273 + }, + { + "epoch": 0.33785450061652283, + "grad_norm": 2.0096053009394987, + "learning_rate": 1.9570956673058678e-05, + "loss": 1.1737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 274 + }, + { + "epoch": 0.33908754623921084, + "grad_norm": 2.064991805627586, + "learning_rate": 1.9567128189219203e-05, + "loss": 1.2067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 275 + }, + { + "epoch": 0.3403205918618989, + "grad_norm": 2.016629291677699, + "learning_rate": 1.9563283077641032e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 276 + }, + { + "epoch": 0.34155363748458695, + "grad_norm": 2.026571915003702, + "learning_rate": 1.9559421345007004e-05, + "loss": 0.9722, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 277 + }, + { + "epoch": 0.34278668310727495, + "grad_norm": 2.0408528020000896, + "learning_rate": 1.9555542998028836e-05, + "loss": 1.0951, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 278 + }, + { + "epoch": 0.344019728729963, + "grad_norm": 2.4490550716870336, + "learning_rate": 1.955164804344712e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 279 + }, + { + "epoch": 0.34525277435265106, + "grad_norm": 2.0298228033571517, + "learning_rate": 1.9547736488031322e-05, + "loss": 1.2091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 280 + }, + { + "epoch": 0.34648581997533906, + "grad_norm": 1.928172339548102, + "learning_rate": 1.9543808338579746e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 281 + }, + { + "epoch": 0.3477188655980271, + "grad_norm": 1.9880687290891643, + "learning_rate": 1.953986360191955e-05, + "loss": 1.0397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 282 + }, + { + "epoch": 0.34895191122071517, + "grad_norm": 2.3895656693876823, + "learning_rate": 1.9535902284906717e-05, + "loss": 1.2004, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 283 + }, + { + "epoch": 0.35018495684340323, + "grad_norm": 1.9995303874089259, + "learning_rate": 1.9531924394426036e-05, + "loss": 1.2001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 284 + }, + { + "epoch": 0.35141800246609123, + "grad_norm": 1.9751518073159213, + "learning_rate": 1.9527929937391118e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 285 + }, + { + "epoch": 0.3526510480887793, + "grad_norm": 2.002111448708798, + "learning_rate": 1.9523918920744357e-05, + "loss": 1.0845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 286 + }, + { + "epoch": 0.35388409371146734, + "grad_norm": 2.0131766303020036, + "learning_rate": 1.9519891351456926e-05, + "loss": 1.1764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 287 + }, + { + "epoch": 0.35511713933415534, + "grad_norm": 2.2331356025573785, + "learning_rate": 1.9515847236528772e-05, + "loss": 1.225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 288 + }, + { + "epoch": 0.3563501849568434, + "grad_norm": 2.0628294266012204, + "learning_rate": 1.9511786582988597e-05, + "loss": 1.2259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 289 + }, + { + "epoch": 0.35758323057953145, + "grad_norm": 1.9151442433897523, + "learning_rate": 1.9507709397893846e-05, + "loss": 1.2103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 290 + }, + { + "epoch": 0.3588162762022195, + "grad_norm": 1.958231535128413, + "learning_rate": 1.9503615688330698e-05, + "loss": 1.105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 291 + }, + { + "epoch": 0.3600493218249075, + "grad_norm": 2.149244076290205, + "learning_rate": 1.9499505461414045e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 292 + }, + { + "epoch": 0.36128236744759556, + "grad_norm": 2.017117422238215, + "learning_rate": 1.9495378724287498e-05, + "loss": 1.1007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 293 + }, + { + "epoch": 0.3625154130702836, + "grad_norm": 1.9210343338897897, + "learning_rate": 1.949123548412335e-05, + "loss": 1.0755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 294 + }, + { + "epoch": 0.3637484586929716, + "grad_norm": 1.9955568267380113, + "learning_rate": 1.9487075748122586e-05, + "loss": 1.1702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 295 + }, + { + "epoch": 0.3649815043156597, + "grad_norm": 2.014663690647869, + "learning_rate": 1.948289952351486e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 296 + }, + { + "epoch": 0.36621454993834773, + "grad_norm": 2.0560814736523922, + "learning_rate": 1.9478706817558474e-05, + "loss": 1.1981, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 297 + }, + { + "epoch": 0.36744759556103573, + "grad_norm": 2.072244571774935, + "learning_rate": 1.947449763754039e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 298 + }, + { + "epoch": 0.3686806411837238, + "grad_norm": 2.3134145282189724, + "learning_rate": 1.947027199077619e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 299 + }, + { + "epoch": 0.36991368680641185, + "grad_norm": 1.9505648569157141, + "learning_rate": 1.9466029884610072e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 300 + }, + { + "epoch": 0.3711467324290999, + "grad_norm": 1.9850708859825466, + "learning_rate": 1.946177132641486e-05, + "loss": 1.0863, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 301 + }, + { + "epoch": 0.3723797780517879, + "grad_norm": 2.3267730903818755, + "learning_rate": 1.945749632359195e-05, + "loss": 1.1405, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 302 + }, + { + "epoch": 0.37361282367447596, + "grad_norm": 2.403438763479749, + "learning_rate": 1.9453204883571333e-05, + "loss": 1.1977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 303 + }, + { + "epoch": 0.374845869297164, + "grad_norm": 2.028737664163498, + "learning_rate": 1.9448897013811558e-05, + "loss": 1.2029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 304 + }, + { + "epoch": 0.376078914919852, + "grad_norm": 1.9981856925053472, + "learning_rate": 1.9444572721799743e-05, + "loss": 1.1873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 305 + }, + { + "epoch": 0.37731196054254007, + "grad_norm": 1.968701546642257, + "learning_rate": 1.9440232015051526e-05, + "loss": 1.1721, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 306 + }, + { + "epoch": 0.3785450061652281, + "grad_norm": 1.864390281708884, + "learning_rate": 1.9435874901111102e-05, + "loss": 1.1797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 307 + }, + { + "epoch": 0.3797780517879161, + "grad_norm": 2.0705929640726577, + "learning_rate": 1.9431501387551158e-05, + "loss": 1.1453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 308 + }, + { + "epoch": 0.3810110974106042, + "grad_norm": 1.884359293663656, + "learning_rate": 1.942711148197289e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 309 + }, + { + "epoch": 0.38224414303329224, + "grad_norm": 1.936787839699547, + "learning_rate": 1.9422705192005994e-05, + "loss": 1.2022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 310 + }, + { + "epoch": 0.3834771886559803, + "grad_norm": 1.9636401987324612, + "learning_rate": 1.941828252530863e-05, + "loss": 1.2015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 311 + }, + { + "epoch": 0.3847102342786683, + "grad_norm": 2.0693459412185495, + "learning_rate": 1.9413843489567423e-05, + "loss": 1.054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 312 + }, + { + "epoch": 0.38594327990135635, + "grad_norm": 2.0182958064572256, + "learning_rate": 1.9409388092497453e-05, + "loss": 1.1086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 313 + }, + { + "epoch": 0.3871763255240444, + "grad_norm": 2.093433911405551, + "learning_rate": 1.940491634184223e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 314 + }, + { + "epoch": 0.3884093711467324, + "grad_norm": 2.018372038783884, + "learning_rate": 1.9400428245373693e-05, + "loss": 1.0813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 315 + }, + { + "epoch": 0.38964241676942046, + "grad_norm": 1.9703329339612685, + "learning_rate": 1.9395923810892184e-05, + "loss": 1.0768, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 316 + }, + { + "epoch": 0.3908754623921085, + "grad_norm": 1.9327973664586777, + "learning_rate": 1.939140304622644e-05, + "loss": 1.1157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 317 + }, + { + "epoch": 0.3921085080147966, + "grad_norm": 2.019497002973341, + "learning_rate": 1.938686595923359e-05, + "loss": 0.9835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 318 + }, + { + "epoch": 0.3933415536374846, + "grad_norm": 2.089968778066188, + "learning_rate": 1.9382312557799116e-05, + "loss": 1.0213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 319 + }, + { + "epoch": 0.39457459926017263, + "grad_norm": 2.0510099518796006, + "learning_rate": 1.937774284983687e-05, + "loss": 1.099, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 320 + }, + { + "epoch": 0.3958076448828607, + "grad_norm": 2.0175024562414063, + "learning_rate": 1.9373156843289037e-05, + "loss": 1.0764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 321 + }, + { + "epoch": 0.3970406905055487, + "grad_norm": 2.020977606778504, + "learning_rate": 1.9368554546126126e-05, + "loss": 1.0562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 322 + }, + { + "epoch": 0.39827373612823674, + "grad_norm": 2.2138761252967125, + "learning_rate": 1.936393596634697e-05, + "loss": 1.196, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 323 + }, + { + "epoch": 0.3995067817509248, + "grad_norm": 1.855560397827274, + "learning_rate": 1.9359301111978683e-05, + "loss": 1.0725, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 324 + }, + { + "epoch": 0.4007398273736128, + "grad_norm": 2.0805790476522636, + "learning_rate": 1.9354649991076684e-05, + "loss": 1.1665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 325 + }, + { + "epoch": 0.40197287299630086, + "grad_norm": 1.8807037569919354, + "learning_rate": 1.9349982611724654e-05, + "loss": 1.1366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 326 + }, + { + "epoch": 0.4032059186189889, + "grad_norm": 1.9286740986289053, + "learning_rate": 1.9345298982034536e-05, + "loss": 1.132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 327 + }, + { + "epoch": 0.40443896424167697, + "grad_norm": 2.0846402787959786, + "learning_rate": 1.9340599110146504e-05, + "loss": 1.1055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 328 + }, + { + "epoch": 0.40567200986436497, + "grad_norm": 2.2183587793468993, + "learning_rate": 1.9335883004228974e-05, + "loss": 1.1544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 329 + }, + { + "epoch": 0.406905055487053, + "grad_norm": 2.2213015226303643, + "learning_rate": 1.9331150672478575e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 330 + }, + { + "epoch": 0.4081381011097411, + "grad_norm": 2.2277113238319726, + "learning_rate": 1.9326402123120132e-05, + "loss": 1.0451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 331 + }, + { + "epoch": 0.4093711467324291, + "grad_norm": 2.259911506493339, + "learning_rate": 1.9321637364406658e-05, + "loss": 1.0236, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 332 + }, + { + "epoch": 0.41060419235511714, + "grad_norm": 2.3573090703399417, + "learning_rate": 1.931685640461934e-05, + "loss": 1.1412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 333 + }, + { + "epoch": 0.4118372379778052, + "grad_norm": 2.1097456297814468, + "learning_rate": 1.9312059252067517e-05, + "loss": 1.1901, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 334 + }, + { + "epoch": 0.4130702836004932, + "grad_norm": 2.121952527251297, + "learning_rate": 1.930724591508868e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 335 + }, + { + "epoch": 0.41430332922318125, + "grad_norm": 2.091327214317948, + "learning_rate": 1.930241640204844e-05, + "loss": 1.0135, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 336 + }, + { + "epoch": 0.4155363748458693, + "grad_norm": 2.1272216173587255, + "learning_rate": 1.929757072134053e-05, + "loss": 1.1754, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 337 + }, + { + "epoch": 0.41676942046855736, + "grad_norm": 1.9973226929736474, + "learning_rate": 1.929270888138678e-05, + "loss": 1.1392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 338 + }, + { + "epoch": 0.41800246609124536, + "grad_norm": 1.9318995412408386, + "learning_rate": 1.9287830890637093e-05, + "loss": 1.0762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 339 + }, + { + "epoch": 0.4192355117139334, + "grad_norm": 2.785840304521055, + "learning_rate": 1.9282936757569463e-05, + "loss": 1.1699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 340 + }, + { + "epoch": 0.4204685573366215, + "grad_norm": 2.171910512782072, + "learning_rate": 1.9278026490689925e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 341 + }, + { + "epoch": 0.4217016029593095, + "grad_norm": 2.052852932277261, + "learning_rate": 1.927310009853256e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 342 + }, + { + "epoch": 0.42293464858199753, + "grad_norm": 1.9538275366336955, + "learning_rate": 1.926815758965947e-05, + "loss": 1.1934, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 343 + }, + { + "epoch": 0.4241676942046856, + "grad_norm": 1.9772612426660126, + "learning_rate": 1.9263198972660774e-05, + "loss": 1.1061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 344 + }, + { + "epoch": 0.4254007398273736, + "grad_norm": 1.9522364508331842, + "learning_rate": 1.9258224256154588e-05, + "loss": 1.101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 345 + }, + { + "epoch": 0.42663378545006164, + "grad_norm": 2.1060276682972714, + "learning_rate": 1.9253233448787e-05, + "loss": 1.2094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 346 + }, + { + "epoch": 0.4278668310727497, + "grad_norm": 2.020183264992464, + "learning_rate": 1.9248226559232074e-05, + "loss": 1.2012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 347 + }, + { + "epoch": 0.42909987669543775, + "grad_norm": 2.1361451948859256, + "learning_rate": 1.9243203596191815e-05, + "loss": 1.2714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 348 + }, + { + "epoch": 0.43033292231812575, + "grad_norm": 1.890346669461953, + "learning_rate": 1.9238164568396177e-05, + "loss": 1.067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 349 + }, + { + "epoch": 0.4315659679408138, + "grad_norm": 1.908858157385325, + "learning_rate": 1.9233109484603025e-05, + "loss": 1.1123, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 350 + }, + { + "epoch": 0.43279901356350187, + "grad_norm": 1.9095667461482675, + "learning_rate": 1.9228038353598122e-05, + "loss": 1.0834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 351 + }, + { + "epoch": 0.43403205918618987, + "grad_norm": 1.8950946192756821, + "learning_rate": 1.9222951184195146e-05, + "loss": 1.0686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 352 + }, + { + "epoch": 0.4352651048088779, + "grad_norm": 2.5409486063614066, + "learning_rate": 1.921784798523562e-05, + "loss": 1.1168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 353 + }, + { + "epoch": 0.436498150431566, + "grad_norm": 1.9251760877216109, + "learning_rate": 1.9212728765588954e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 354 + }, + { + "epoch": 0.43773119605425403, + "grad_norm": 2.12327469473301, + "learning_rate": 1.9207593534152383e-05, + "loss": 1.1871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 355 + }, + { + "epoch": 0.43896424167694204, + "grad_norm": 2.0404259513933214, + "learning_rate": 1.9202442299850977e-05, + "loss": 1.0985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 356 + }, + { + "epoch": 0.4401972872996301, + "grad_norm": 2.10787306308974, + "learning_rate": 1.9197275071637617e-05, + "loss": 1.0063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 357 + }, + { + "epoch": 0.44143033292231815, + "grad_norm": 2.2402685683097996, + "learning_rate": 1.919209185849299e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 358 + }, + { + "epoch": 0.44266337854500615, + "grad_norm": 2.148922705064427, + "learning_rate": 1.9186892669425546e-05, + "loss": 1.0889, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 359 + }, + { + "epoch": 0.4438964241676942, + "grad_norm": 1.8789248361497928, + "learning_rate": 1.9181677513471525e-05, + "loss": 1.1796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 360 + }, + { + "epoch": 0.44512946979038226, + "grad_norm": 1.8886466190263609, + "learning_rate": 1.9176446399694903e-05, + "loss": 1.1494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 361 + }, + { + "epoch": 0.44636251541307026, + "grad_norm": 1.9432580511904547, + "learning_rate": 1.9171199337187395e-05, + "loss": 1.2599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 362 + }, + { + "epoch": 0.4475955610357583, + "grad_norm": 1.8820277177030855, + "learning_rate": 1.916593633506843e-05, + "loss": 1.1884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 363 + }, + { + "epoch": 0.44882860665844637, + "grad_norm": 2.249700768154195, + "learning_rate": 1.9160657402485147e-05, + "loss": 1.0503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 364 + }, + { + "epoch": 0.45006165228113443, + "grad_norm": 2.082324666968816, + "learning_rate": 1.9155362548612366e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 365 + }, + { + "epoch": 0.45129469790382243, + "grad_norm": 2.2090950658394193, + "learning_rate": 1.915005178265259e-05, + "loss": 1.1742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 366 + }, + { + "epoch": 0.4525277435265105, + "grad_norm": 2.2007793765172683, + "learning_rate": 1.914472511383596e-05, + "loss": 1.2617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 367 + }, + { + "epoch": 0.45376078914919854, + "grad_norm": 1.9166813253050292, + "learning_rate": 1.913938255142027e-05, + "loss": 1.1444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 368 + }, + { + "epoch": 0.45499383477188654, + "grad_norm": 1.908891419449054, + "learning_rate": 1.913402410469093e-05, + "loss": 1.1139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 369 + }, + { + "epoch": 0.4562268803945746, + "grad_norm": 2.0434795360025566, + "learning_rate": 1.9128649782960965e-05, + "loss": 1.1761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 370 + }, + { + "epoch": 0.45745992601726265, + "grad_norm": 2.2319795236629165, + "learning_rate": 1.9123259595570978e-05, + "loss": 0.9997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 371 + }, + { + "epoch": 0.45869297163995065, + "grad_norm": 1.9325225410679332, + "learning_rate": 1.911785355188916e-05, + "loss": 1.1191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 372 + }, + { + "epoch": 0.4599260172626387, + "grad_norm": 2.3718583427313384, + "learning_rate": 1.9112431661311248e-05, + "loss": 1.1801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 373 + }, + { + "epoch": 0.46115906288532676, + "grad_norm": 2.2778458302261853, + "learning_rate": 1.9106993933260535e-05, + "loss": 1.1973, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 374 + }, + { + "epoch": 0.4623921085080148, + "grad_norm": 2.104485888382934, + "learning_rate": 1.910154037718783e-05, + "loss": 1.206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 375 + }, + { + "epoch": 0.4636251541307028, + "grad_norm": 2.149849667802547, + "learning_rate": 1.909607100257145e-05, + "loss": 1.0479, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 376 + }, + { + "epoch": 0.4648581997533909, + "grad_norm": 2.242058517922103, + "learning_rate": 1.9090585818917205e-05, + "loss": 1.0902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 377 + }, + { + "epoch": 0.46609124537607893, + "grad_norm": 2.0943438214587307, + "learning_rate": 1.9085084835758394e-05, + "loss": 1.2428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 378 + }, + { + "epoch": 0.46732429099876693, + "grad_norm": 2.1534146036573647, + "learning_rate": 1.907956806265576e-05, + "loss": 1.0392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 379 + }, + { + "epoch": 0.468557336621455, + "grad_norm": 2.2361240083173324, + "learning_rate": 1.9074035509197495e-05, + "loss": 1.1337, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 380 + }, + { + "epoch": 0.46979038224414305, + "grad_norm": 2.0666798517773795, + "learning_rate": 1.906848718499922e-05, + "loss": 1.0574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 381 + }, + { + "epoch": 0.47102342786683105, + "grad_norm": 2.0056597359218094, + "learning_rate": 1.906292309970396e-05, + "loss": 1.1982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 382 + }, + { + "epoch": 0.4722564734895191, + "grad_norm": 1.931784648165126, + "learning_rate": 1.9057343262982136e-05, + "loss": 1.0671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 383 + }, + { + "epoch": 0.47348951911220716, + "grad_norm": 2.2650382784813408, + "learning_rate": 1.905174768453155e-05, + "loss": 1.1384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 384 + }, + { + "epoch": 0.4747225647348952, + "grad_norm": 2.2386337800038594, + "learning_rate": 1.904613637407735e-05, + "loss": 1.169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 385 + }, + { + "epoch": 0.4759556103575832, + "grad_norm": 1.985439502238856, + "learning_rate": 1.9040509341372045e-05, + "loss": 1.0544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 386 + }, + { + "epoch": 0.47718865598027127, + "grad_norm": 1.9991995362972141, + "learning_rate": 1.903486659619545e-05, + "loss": 1.1048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 387 + }, + { + "epoch": 0.4784217016029593, + "grad_norm": 1.938313304762426, + "learning_rate": 1.90292081483547e-05, + "loss": 1.0217, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 388 + }, + { + "epoch": 0.4796547472256473, + "grad_norm": 2.0593580639988605, + "learning_rate": 1.9023534007684217e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 389 + }, + { + "epoch": 0.4808877928483354, + "grad_norm": 2.0113392094384723, + "learning_rate": 1.90178441840457e-05, + "loss": 1.2114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 390 + }, + { + "epoch": 0.48212083847102344, + "grad_norm": 1.9865461362024583, + "learning_rate": 1.901213868732811e-05, + "loss": 1.0818, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 391 + }, + { + "epoch": 0.4833538840937115, + "grad_norm": 1.9044551829036014, + "learning_rate": 1.9006417527447634e-05, + "loss": 1.1046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 392 + }, + { + "epoch": 0.4845869297163995, + "grad_norm": 1.8508018232087575, + "learning_rate": 1.9000680714347694e-05, + "loss": 1.1411, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 393 + }, + { + "epoch": 0.48581997533908755, + "grad_norm": 1.98419367965867, + "learning_rate": 1.8994928257998907e-05, + "loss": 0.9804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 394 + }, + { + "epoch": 0.4870530209617756, + "grad_norm": 2.1671038349199567, + "learning_rate": 1.8989160168399093e-05, + "loss": 1.1656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 395 + }, + { + "epoch": 0.4882860665844636, + "grad_norm": 2.0570349889938284, + "learning_rate": 1.898337645557323e-05, + "loss": 1.2111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 396 + }, + { + "epoch": 0.48951911220715166, + "grad_norm": 1.9619760913338724, + "learning_rate": 1.8977577129573452e-05, + "loss": 1.0932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 397 + }, + { + "epoch": 0.4907521578298397, + "grad_norm": 1.9608046473112009, + "learning_rate": 1.897176220047903e-05, + "loss": 1.1389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 398 + }, + { + "epoch": 0.4919852034525277, + "grad_norm": 1.9645035824539876, + "learning_rate": 1.8965931678396357e-05, + "loss": 1.1056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 399 + }, + { + "epoch": 0.4932182490752158, + "grad_norm": 1.961734329156117, + "learning_rate": 1.896008557345892e-05, + "loss": 1.1322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 400 + }, + { + "epoch": 0.49445129469790383, + "grad_norm": 1.844560254648902, + "learning_rate": 1.8954223895827288e-05, + "loss": 1.0712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 401 + }, + { + "epoch": 0.4956843403205919, + "grad_norm": 2.1465454734369933, + "learning_rate": 1.8948346655689108e-05, + "loss": 1.1294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 402 + }, + { + "epoch": 0.4969173859432799, + "grad_norm": 2.078244569426143, + "learning_rate": 1.8942453863259058e-05, + "loss": 1.0546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 403 + }, + { + "epoch": 0.49815043156596794, + "grad_norm": 2.264508341916039, + "learning_rate": 1.893654552877886e-05, + "loss": 1.0568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 404 + }, + { + "epoch": 0.499383477188656, + "grad_norm": 2.0051680479216696, + "learning_rate": 1.893062166251724e-05, + "loss": 1.0156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 405 + }, + { + "epoch": 0.5006165228113441, + "grad_norm": 2.1569911537953725, + "learning_rate": 1.8924682274769925e-05, + "loss": 1.173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 406 + }, + { + "epoch": 0.5018495684340321, + "grad_norm": 2.002983438517559, + "learning_rate": 1.8918727375859608e-05, + "loss": 1.0073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 407 + }, + { + "epoch": 0.5030826140567201, + "grad_norm": 1.9516175334775212, + "learning_rate": 1.891275697613595e-05, + "loss": 1.1404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 408 + }, + { + "epoch": 0.5043156596794082, + "grad_norm": 2.1494580236712832, + "learning_rate": 1.890677108597555e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 409 + }, + { + "epoch": 0.5055487053020962, + "grad_norm": 2.025395779691303, + "learning_rate": 1.8900769715781928e-05, + "loss": 0.9756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 410 + }, + { + "epoch": 0.5067817509247842, + "grad_norm": 1.9927648271180456, + "learning_rate": 1.889475287598551e-05, + "loss": 1.1394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 411 + }, + { + "epoch": 0.5080147965474723, + "grad_norm": 1.9818065701766985, + "learning_rate": 1.8888720577043603e-05, + "loss": 1.1683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 412 + }, + { + "epoch": 0.5092478421701603, + "grad_norm": 1.9410101081952555, + "learning_rate": 1.8882672829440393e-05, + "loss": 1.0839, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 413 + }, + { + "epoch": 0.5104808877928483, + "grad_norm": 1.8396805813100958, + "learning_rate": 1.8876609643686906e-05, + "loss": 1.114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 414 + }, + { + "epoch": 0.5117139334155364, + "grad_norm": 2.0105399328568905, + "learning_rate": 1.8870531030321004e-05, + "loss": 1.1439, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 415 + }, + { + "epoch": 0.5129469790382244, + "grad_norm": 2.090098149989323, + "learning_rate": 1.886443699990736e-05, + "loss": 1.1153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 416 + }, + { + "epoch": 0.5141800246609125, + "grad_norm": 2.0517637089595246, + "learning_rate": 1.885832756303745e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 417 + }, + { + "epoch": 0.5154130702836005, + "grad_norm": 3.0738713157895674, + "learning_rate": 1.885220273032951e-05, + "loss": 1.0222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 418 + }, + { + "epoch": 0.5166461159062885, + "grad_norm": 2.6677283836123005, + "learning_rate": 1.884606251242855e-05, + "loss": 1.1038, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 419 + }, + { + "epoch": 0.5178791615289766, + "grad_norm": 2.125547466007061, + "learning_rate": 1.8839906920006315e-05, + "loss": 1.2128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 420 + }, + { + "epoch": 0.5191122071516646, + "grad_norm": 2.1599202425686013, + "learning_rate": 1.883373596376126e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 421 + }, + { + "epoch": 0.5203452527743526, + "grad_norm": 2.1229927533556436, + "learning_rate": 1.8827549654418566e-05, + "loss": 1.1353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 422 + }, + { + "epoch": 0.5215782983970407, + "grad_norm": 1.9567023805342967, + "learning_rate": 1.8821348002730072e-05, + "loss": 1.0584, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 423 + }, + { + "epoch": 0.5228113440197287, + "grad_norm": 1.8989916861837828, + "learning_rate": 1.88151310194743e-05, + "loss": 0.9837, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 424 + }, + { + "epoch": 0.5240443896424167, + "grad_norm": 2.1067680957156014, + "learning_rate": 1.8808898715456415e-05, + "loss": 1.0673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 425 + }, + { + "epoch": 0.5252774352651048, + "grad_norm": 2.6013242993173926, + "learning_rate": 1.88026511015082e-05, + "loss": 1.112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 426 + }, + { + "epoch": 0.5265104808877928, + "grad_norm": 2.0539914273851907, + "learning_rate": 1.8796388188488055e-05, + "loss": 1.1031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 427 + }, + { + "epoch": 0.5277435265104808, + "grad_norm": 2.0386084337690584, + "learning_rate": 1.8790109987280972e-05, + "loss": 1.1556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 428 + }, + { + "epoch": 0.528976572133169, + "grad_norm": 2.093419206457373, + "learning_rate": 1.878381650879851e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 429 + }, + { + "epoch": 0.530209617755857, + "grad_norm": 2.2549735586348985, + "learning_rate": 1.877750776397878e-05, + "loss": 1.0663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 430 + }, + { + "epoch": 0.531442663378545, + "grad_norm": 2.1062036310291266, + "learning_rate": 1.877118376378642e-05, + "loss": 1.0455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 431 + }, + { + "epoch": 0.5326757090012331, + "grad_norm": 1.995564286989001, + "learning_rate": 1.87648445192126e-05, + "loss": 0.982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 432 + }, + { + "epoch": 0.5339087546239211, + "grad_norm": 2.41307031725252, + "learning_rate": 1.8758490041274963e-05, + "loss": 1.0778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 433 + }, + { + "epoch": 0.5351418002466092, + "grad_norm": 2.831604603209417, + "learning_rate": 1.875212034101765e-05, + "loss": 1.1345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 434 + }, + { + "epoch": 0.5363748458692972, + "grad_norm": 2.2244728383006604, + "learning_rate": 1.8745735429511236e-05, + "loss": 0.938, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 435 + }, + { + "epoch": 0.5376078914919852, + "grad_norm": 2.0679863467245823, + "learning_rate": 1.8739335317852747e-05, + "loss": 1.0763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 436 + }, + { + "epoch": 0.5388409371146733, + "grad_norm": 2.2454972633071915, + "learning_rate": 1.8732920017165624e-05, + "loss": 1.1207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 437 + }, + { + "epoch": 0.5400739827373613, + "grad_norm": 1.952229569146223, + "learning_rate": 1.8726489538599707e-05, + "loss": 1.0255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 438 + }, + { + "epoch": 0.5413070283600493, + "grad_norm": 1.9236876015939406, + "learning_rate": 1.8720043893331214e-05, + "loss": 0.9933, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 439 + }, + { + "epoch": 0.5425400739827374, + "grad_norm": 1.9736270928366435, + "learning_rate": 1.8713583092562727e-05, + "loss": 1.107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 440 + }, + { + "epoch": 0.5437731196054254, + "grad_norm": 2.1162408965440633, + "learning_rate": 1.870710714752316e-05, + "loss": 1.0441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 441 + }, + { + "epoch": 0.5450061652281134, + "grad_norm": 1.9515336502135756, + "learning_rate": 1.8700616069467753e-05, + "loss": 0.9855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 442 + }, + { + "epoch": 0.5462392108508015, + "grad_norm": 2.566164154732447, + "learning_rate": 1.8694109869678047e-05, + "loss": 1.1289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 443 + }, + { + "epoch": 0.5474722564734895, + "grad_norm": 2.3448298186890026, + "learning_rate": 1.868758855946187e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 444 + }, + { + "epoch": 0.5487053020961775, + "grad_norm": 2.076679946731619, + "learning_rate": 1.8681052150153305e-05, + "loss": 1.0494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 445 + }, + { + "epoch": 0.5499383477188656, + "grad_norm": 2.281980101143931, + "learning_rate": 1.8674500653112674e-05, + "loss": 1.0456, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 446 + }, + { + "epoch": 0.5511713933415536, + "grad_norm": 2.147469472559495, + "learning_rate": 1.8667934079726537e-05, + "loss": 1.0682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 447 + }, + { + "epoch": 0.5524044389642416, + "grad_norm": 2.0621898279926527, + "learning_rate": 1.866135244140763e-05, + "loss": 1.0527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 448 + }, + { + "epoch": 0.5536374845869297, + "grad_norm": 2.0688727890116785, + "learning_rate": 1.8654755749594905e-05, + "loss": 1.0549, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 449 + }, + { + "epoch": 0.5548705302096177, + "grad_norm": 2.0101374651739325, + "learning_rate": 1.864814401575345e-05, + "loss": 1.002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 450 + }, + { + "epoch": 0.5561035758323057, + "grad_norm": 2.2861388762596344, + "learning_rate": 1.8641517251374514e-05, + "loss": 1.0666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 451 + }, + { + "epoch": 0.5573366214549939, + "grad_norm": 2.006862831841472, + "learning_rate": 1.8634875467975456e-05, + "loss": 1.0061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 452 + }, + { + "epoch": 0.5585696670776819, + "grad_norm": 2.144270279582644, + "learning_rate": 1.862821867709975e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 453 + }, + { + "epoch": 0.55980271270037, + "grad_norm": 1.9667742618855084, + "learning_rate": 1.862154689031694e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 454 + }, + { + "epoch": 0.561035758323058, + "grad_norm": 1.8857968215229095, + "learning_rate": 1.8614860119222643e-05, + "loss": 1.0861, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 455 + }, + { + "epoch": 0.562268803945746, + "grad_norm": 1.9374803828274445, + "learning_rate": 1.860815837543852e-05, + "loss": 0.9784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 456 + }, + { + "epoch": 0.5635018495684341, + "grad_norm": 2.0764596642416433, + "learning_rate": 1.8601441670612247e-05, + "loss": 1.1005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 457 + }, + { + "epoch": 0.5647348951911221, + "grad_norm": 2.123062949997989, + "learning_rate": 1.8594710016417505e-05, + "loss": 1.1953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 458 + }, + { + "epoch": 0.5659679408138101, + "grad_norm": 2.311252498934928, + "learning_rate": 1.858796342455396e-05, + "loss": 1.1049, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 459 + }, + { + "epoch": 0.5672009864364982, + "grad_norm": 1.9780001048890374, + "learning_rate": 1.8581201906747242e-05, + "loss": 1.1393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 460 + }, + { + "epoch": 0.5684340320591862, + "grad_norm": 1.968874552657262, + "learning_rate": 1.857442547474891e-05, + "loss": 1.1578, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 461 + }, + { + "epoch": 0.5696670776818742, + "grad_norm": 1.9395225862380807, + "learning_rate": 1.8567634140336457e-05, + "loss": 1.0823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 462 + }, + { + "epoch": 0.5709001233045623, + "grad_norm": 1.9422213147988507, + "learning_rate": 1.856082791531327e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 463 + }, + { + "epoch": 0.5721331689272503, + "grad_norm": 1.9095099328124767, + "learning_rate": 1.855400681150862e-05, + "loss": 1.1041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 464 + }, + { + "epoch": 0.5733662145499383, + "grad_norm": 1.9651570734782207, + "learning_rate": 1.8547170840777635e-05, + "loss": 0.9924, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 465 + }, + { + "epoch": 0.5745992601726264, + "grad_norm": 1.9083237960346455, + "learning_rate": 1.8540320015001283e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 466 + }, + { + "epoch": 0.5758323057953144, + "grad_norm": 2.0442809672147484, + "learning_rate": 1.8533454346086345e-05, + "loss": 0.9449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 467 + }, + { + "epoch": 0.5770653514180024, + "grad_norm": 2.839273126939154, + "learning_rate": 1.8526573845965407e-05, + "loss": 1.1446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 468 + }, + { + "epoch": 0.5782983970406905, + "grad_norm": 2.8401028851750083, + "learning_rate": 1.8519678526596827e-05, + "loss": 1.0793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 469 + }, + { + "epoch": 0.5795314426633785, + "grad_norm": 2.538177485476921, + "learning_rate": 1.8512768399964725e-05, + "loss": 1.1066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 470 + }, + { + "epoch": 0.5807644882860666, + "grad_norm": 2.1457149698206526, + "learning_rate": 1.8505843478078944e-05, + "loss": 1.1138, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 471 + }, + { + "epoch": 0.5819975339087546, + "grad_norm": 1.8879825929112735, + "learning_rate": 1.8498903772975056e-05, + "loss": 1.0181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 472 + }, + { + "epoch": 0.5832305795314426, + "grad_norm": 2.083069072716887, + "learning_rate": 1.8491949296714312e-05, + "loss": 1.0307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 473 + }, + { + "epoch": 0.5844636251541308, + "grad_norm": 2.135777387695568, + "learning_rate": 1.848498006138365e-05, + "loss": 1.0346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 474 + }, + { + "epoch": 0.5856966707768188, + "grad_norm": 2.092600182986245, + "learning_rate": 1.8477996079095643e-05, + "loss": 1.041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 475 + }, + { + "epoch": 0.5869297163995068, + "grad_norm": 2.1246789769186414, + "learning_rate": 1.8470997361988514e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 476 + }, + { + "epoch": 0.5881627620221949, + "grad_norm": 2.062971539274987, + "learning_rate": 1.8463983922226076e-05, + "loss": 1.111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 477 + }, + { + "epoch": 0.5893958076448829, + "grad_norm": 2.092774168134711, + "learning_rate": 1.8456955771997744e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 478 + }, + { + "epoch": 0.5906288532675709, + "grad_norm": 2.23500231174451, + "learning_rate": 1.844991292351849e-05, + "loss": 1.1008, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 479 + }, + { + "epoch": 0.591861898890259, + "grad_norm": 2.04828338908309, + "learning_rate": 1.844285538902884e-05, + "loss": 1.0183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 480 + }, + { + "epoch": 0.593094944512947, + "grad_norm": 2.1680070152194064, + "learning_rate": 1.8435783180794836e-05, + "loss": 1.092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 481 + }, + { + "epoch": 0.594327990135635, + "grad_norm": 2.0706136769588603, + "learning_rate": 1.8428696311108033e-05, + "loss": 1.1438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 482 + }, + { + "epoch": 0.5955610357583231, + "grad_norm": 1.9925622155649279, + "learning_rate": 1.842159479228546e-05, + "loss": 1.0359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 483 + }, + { + "epoch": 0.5967940813810111, + "grad_norm": 1.9425841064456546, + "learning_rate": 1.841447863666961e-05, + "loss": 1.2084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 484 + }, + { + "epoch": 0.5980271270036991, + "grad_norm": 1.929626615569851, + "learning_rate": 1.8407347856628405e-05, + "loss": 1.0615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 485 + }, + { + "epoch": 0.5992601726263872, + "grad_norm": 2.049215260436237, + "learning_rate": 1.8400202464555202e-05, + "loss": 1.0429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 486 + }, + { + "epoch": 0.6004932182490752, + "grad_norm": 2.0464723350262624, + "learning_rate": 1.8393042472868746e-05, + "loss": 1.1504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 487 + }, + { + "epoch": 0.6017262638717632, + "grad_norm": 1.9060167485555373, + "learning_rate": 1.8385867894013144e-05, + "loss": 0.9809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 488 + }, + { + "epoch": 0.6029593094944513, + "grad_norm": 1.9897115419719036, + "learning_rate": 1.837867874045788e-05, + "loss": 1.0573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 489 + }, + { + "epoch": 0.6041923551171393, + "grad_norm": 2.241807622458982, + "learning_rate": 1.8371475024697746e-05, + "loss": 1.1149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 490 + }, + { + "epoch": 0.6054254007398274, + "grad_norm": 2.141493902645165, + "learning_rate": 1.8364256759252855e-05, + "loss": 1.0396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 491 + }, + { + "epoch": 0.6066584463625154, + "grad_norm": 2.168929995419785, + "learning_rate": 1.8357023956668606e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 492 + }, + { + "epoch": 0.6078914919852034, + "grad_norm": 2.095414950529066, + "learning_rate": 1.834977662951566e-05, + "loss": 1.0313, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 493 + }, + { + "epoch": 0.6091245376078915, + "grad_norm": 2.2287918098101516, + "learning_rate": 1.834251479038993e-05, + "loss": 1.0118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 494 + }, + { + "epoch": 0.6103575832305795, + "grad_norm": 2.0458360619004154, + "learning_rate": 1.833523845191254e-05, + "loss": 1.0644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 495 + }, + { + "epoch": 0.6115906288532675, + "grad_norm": 2.1216602719929245, + "learning_rate": 1.8327947626729825e-05, + "loss": 1.1734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 496 + }, + { + "epoch": 0.6128236744759556, + "grad_norm": 2.1445297291236947, + "learning_rate": 1.8320642327513285e-05, + "loss": 0.9661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 497 + }, + { + "epoch": 0.6140567200986436, + "grad_norm": 2.061159959033469, + "learning_rate": 1.831332256695959e-05, + "loss": 1.189, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 498 + }, + { + "epoch": 0.6152897657213316, + "grad_norm": 1.966885029141375, + "learning_rate": 1.8305988357790533e-05, + "loss": 1.0711, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 499 + }, + { + "epoch": 0.6165228113440198, + "grad_norm": 2.127061899116888, + "learning_rate": 1.8298639712753025e-05, + "loss": 0.9351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 500 + }, + { + "epoch": 0.6177558569667078, + "grad_norm": 2.2239600877101835, + "learning_rate": 1.8291276644619066e-05, + "loss": 1.0569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 501 + }, + { + "epoch": 0.6189889025893958, + "grad_norm": 2.0281028990006913, + "learning_rate": 1.8283899166185723e-05, + "loss": 1.0698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 502 + }, + { + "epoch": 0.6202219482120839, + "grad_norm": 2.1224950625771677, + "learning_rate": 1.8276507290275104e-05, + "loss": 0.9905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 503 + }, + { + "epoch": 0.6214549938347719, + "grad_norm": 2.0799215178987, + "learning_rate": 1.8269101029734343e-05, + "loss": 1.0631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 504 + }, + { + "epoch": 0.6226880394574599, + "grad_norm": 2.2495223027498388, + "learning_rate": 1.8261680397435577e-05, + "loss": 0.9869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 505 + }, + { + "epoch": 0.623921085080148, + "grad_norm": 2.036885073926771, + "learning_rate": 1.825424540627592e-05, + "loss": 1.0939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 506 + }, + { + "epoch": 0.625154130702836, + "grad_norm": 2.0423951754968064, + "learning_rate": 1.824679606917744e-05, + "loss": 0.9793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 507 + }, + { + "epoch": 0.6263871763255241, + "grad_norm": 2.118412450801216, + "learning_rate": 1.8239332399087136e-05, + "loss": 1.0214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 508 + }, + { + "epoch": 0.6276202219482121, + "grad_norm": 2.273320358602361, + "learning_rate": 1.8231854408976924e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 509 + }, + { + "epoch": 0.6288532675709001, + "grad_norm": 1.9352848815012758, + "learning_rate": 1.822436211184361e-05, + "loss": 1.0194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 510 + }, + { + "epoch": 0.6300863131935882, + "grad_norm": 2.1086688743882713, + "learning_rate": 1.8216855520708854e-05, + "loss": 0.9404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 511 + }, + { + "epoch": 0.6313193588162762, + "grad_norm": 2.268255476351323, + "learning_rate": 1.820933464861917e-05, + "loss": 0.9734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 512 + }, + { + "epoch": 0.6325524044389642, + "grad_norm": 2.3533715256185705, + "learning_rate": 1.8201799508645887e-05, + "loss": 1.1794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 513 + }, + { + "epoch": 0.6337854500616523, + "grad_norm": 2.005798886598211, + "learning_rate": 1.8194250113885134e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 514 + }, + { + "epoch": 0.6350184956843403, + "grad_norm": 1.9276850525809597, + "learning_rate": 1.8186686477457817e-05, + "loss": 1.0122, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 515 + }, + { + "epoch": 0.6362515413070283, + "grad_norm": 2.089756823449061, + "learning_rate": 1.8179108612509587e-05, + "loss": 1.1076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 516 + }, + { + "epoch": 0.6374845869297164, + "grad_norm": 2.047489618950505, + "learning_rate": 1.8171516532210825e-05, + "loss": 1.1261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 517 + }, + { + "epoch": 0.6387176325524044, + "grad_norm": 1.9819427784604287, + "learning_rate": 1.8163910249756634e-05, + "loss": 0.9829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 518 + }, + { + "epoch": 0.6399506781750924, + "grad_norm": 2.1061977657317574, + "learning_rate": 1.815628977836678e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 519 + }, + { + "epoch": 0.6411837237977805, + "grad_norm": 2.0221352739914624, + "learning_rate": 1.8148655131285697e-05, + "loss": 1.0413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 520 + }, + { + "epoch": 0.6424167694204685, + "grad_norm": 2.0126443495430033, + "learning_rate": 1.8141006321782464e-05, + "loss": 1.0737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 521 + }, + { + "epoch": 0.6436498150431565, + "grad_norm": 1.9687053123524947, + "learning_rate": 1.8133343363150762e-05, + "loss": 1.0534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 522 + }, + { + "epoch": 0.6448828606658447, + "grad_norm": 2.1885490221346036, + "learning_rate": 1.812566626870887e-05, + "loss": 1.0746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 523 + }, + { + "epoch": 0.6461159062885327, + "grad_norm": 2.279950249415201, + "learning_rate": 1.8117975051799636e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 524 + }, + { + "epoch": 0.6473489519112207, + "grad_norm": 2.2376288751244986, + "learning_rate": 1.811026972579045e-05, + "loss": 1.0655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 525 + }, + { + "epoch": 0.6485819975339088, + "grad_norm": 2.0687763759549704, + "learning_rate": 1.810255030407323e-05, + "loss": 0.9515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 526 + }, + { + "epoch": 0.6498150431565968, + "grad_norm": 1.9230312514706345, + "learning_rate": 1.8094816800064376e-05, + "loss": 1.0144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 527 + }, + { + "epoch": 0.6510480887792849, + "grad_norm": 1.9036609363684858, + "learning_rate": 1.8087069227204784e-05, + "loss": 1.1092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 528 + }, + { + "epoch": 0.6522811344019729, + "grad_norm": 2.0395351493662104, + "learning_rate": 1.807930759895979e-05, + "loss": 1.0345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 529 + }, + { + "epoch": 0.6535141800246609, + "grad_norm": 2.011352927400422, + "learning_rate": 1.8071531928819158e-05, + "loss": 0.9913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 530 + }, + { + "epoch": 0.654747225647349, + "grad_norm": 1.9222212916878523, + "learning_rate": 1.8063742230297063e-05, + "loss": 1.0643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 531 + }, + { + "epoch": 0.655980271270037, + "grad_norm": 2.203621374645629, + "learning_rate": 1.8055938516932056e-05, + "loss": 1.0912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 532 + }, + { + "epoch": 0.657213316892725, + "grad_norm": 1.9492742413222397, + "learning_rate": 1.804812080228705e-05, + "loss": 0.9613, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 533 + }, + { + "epoch": 0.6584463625154131, + "grad_norm": 2.1694867453328683, + "learning_rate": 1.8040289099949285e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 534 + }, + { + "epoch": 0.6596794081381011, + "grad_norm": 2.0272612869084474, + "learning_rate": 1.8032443423530322e-05, + "loss": 1.0473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 535 + }, + { + "epoch": 0.6609124537607891, + "grad_norm": 2.059356328787384, + "learning_rate": 1.8024583786666007e-05, + "loss": 1.1633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 536 + }, + { + "epoch": 0.6621454993834772, + "grad_norm": 1.9976930500073966, + "learning_rate": 1.8016710203016434e-05, + "loss": 1.0907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 537 + }, + { + "epoch": 0.6633785450061652, + "grad_norm": 2.2046172414513747, + "learning_rate": 1.8008822686265958e-05, + "loss": 1.0435, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 538 + }, + { + "epoch": 0.6646115906288532, + "grad_norm": 1.9172442142747625, + "learning_rate": 1.8000921250123144e-05, + "loss": 1.0905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 539 + }, + { + "epoch": 0.6658446362515413, + "grad_norm": 1.9118211475087676, + "learning_rate": 1.7993005908320737e-05, + "loss": 0.9362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 540 + }, + { + "epoch": 0.6670776818742293, + "grad_norm": 2.2793615054405096, + "learning_rate": 1.7985076674615663e-05, + "loss": 1.1407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 541 + }, + { + "epoch": 0.6683107274969173, + "grad_norm": 2.0968780040667636, + "learning_rate": 1.797713356278899e-05, + "loss": 1.0263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 542 + }, + { + "epoch": 0.6695437731196054, + "grad_norm": 2.130295025454763, + "learning_rate": 1.79691765866459e-05, + "loss": 0.99, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 543 + }, + { + "epoch": 0.6707768187422934, + "grad_norm": 2.050843680123929, + "learning_rate": 1.7961205760015678e-05, + "loss": 0.937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 544 + }, + { + "epoch": 0.6720098643649816, + "grad_norm": 2.00556261298427, + "learning_rate": 1.7953221096751678e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 545 + }, + { + "epoch": 0.6732429099876696, + "grad_norm": 2.104573565750377, + "learning_rate": 1.7945222610731306e-05, + "loss": 0.9851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 546 + }, + { + "epoch": 0.6744759556103576, + "grad_norm": 2.009986716494675, + "learning_rate": 1.793721031585599e-05, + "loss": 0.9842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 547 + }, + { + "epoch": 0.6757090012330457, + "grad_norm": 2.053047844057653, + "learning_rate": 1.792918422605115e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 548 + }, + { + "epoch": 0.6769420468557337, + "grad_norm": 2.0361666532452714, + "learning_rate": 1.7921144355266195e-05, + "loss": 1.0002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 549 + }, + { + "epoch": 0.6781750924784217, + "grad_norm": 2.02324182186846, + "learning_rate": 1.7913090717474476e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 550 + }, + { + "epoch": 0.6794081381011098, + "grad_norm": 1.9162969465242339, + "learning_rate": 1.7905023326673277e-05, + "loss": 0.9256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 551 + }, + { + "epoch": 0.6806411837237978, + "grad_norm": 2.055814853732051, + "learning_rate": 1.789694219688378e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 552 + }, + { + "epoch": 0.6818742293464858, + "grad_norm": 2.0596956840908276, + "learning_rate": 1.7888847342151047e-05, + "loss": 1.096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 553 + }, + { + "epoch": 0.6831072749691739, + "grad_norm": 1.9518592781827393, + "learning_rate": 1.7880738776543994e-05, + "loss": 0.9907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 554 + }, + { + "epoch": 0.6843403205918619, + "grad_norm": 2.0542440716859485, + "learning_rate": 1.787261651415537e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 555 + }, + { + "epoch": 0.6855733662145499, + "grad_norm": 1.9676290362610278, + "learning_rate": 1.786448056910173e-05, + "loss": 0.96, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 556 + }, + { + "epoch": 0.686806411837238, + "grad_norm": 2.0236847394224418, + "learning_rate": 1.7856330955523396e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 557 + }, + { + "epoch": 0.688039457459926, + "grad_norm": 2.166651989793048, + "learning_rate": 1.7848167687584466e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 558 + }, + { + "epoch": 0.689272503082614, + "grad_norm": 2.037401342146827, + "learning_rate": 1.783999077947275e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 559 + }, + { + "epoch": 0.6905055487053021, + "grad_norm": 2.0931620663370465, + "learning_rate": 1.783180024539978e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 560 + }, + { + "epoch": 0.6917385943279901, + "grad_norm": 2.5512794237438032, + "learning_rate": 1.7823596099600767e-05, + "loss": 1.056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 561 + }, + { + "epoch": 0.6929716399506781, + "grad_norm": 2.2145616452367305, + "learning_rate": 1.781537835633457e-05, + "loss": 1.0677, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 562 + }, + { + "epoch": 0.6942046855733662, + "grad_norm": 2.304945412618975, + "learning_rate": 1.7807147029883688e-05, + "loss": 1.0025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 563 + }, + { + "epoch": 0.6954377311960542, + "grad_norm": 2.2714171401510272, + "learning_rate": 1.7798902134554225e-05, + "loss": 0.927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 564 + }, + { + "epoch": 0.6966707768187423, + "grad_norm": 1.9934114654014257, + "learning_rate": 1.7790643684675874e-05, + "loss": 0.9226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 565 + }, + { + "epoch": 0.6979038224414303, + "grad_norm": 2.4001795539004425, + "learning_rate": 1.7782371694601882e-05, + "loss": 0.9898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 566 + }, + { + "epoch": 0.6991368680641183, + "grad_norm": 2.1543466787664944, + "learning_rate": 1.777408617870902e-05, + "loss": 0.8765, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 567 + }, + { + "epoch": 0.7003699136868065, + "grad_norm": 2.125530896627932, + "learning_rate": 1.7765787151397584e-05, + "loss": 0.9383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 568 + }, + { + "epoch": 0.7016029593094945, + "grad_norm": 2.399579117157628, + "learning_rate": 1.775747462709134e-05, + "loss": 1.116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 569 + }, + { + "epoch": 0.7028360049321825, + "grad_norm": 2.3911388414036843, + "learning_rate": 1.7749148620237515e-05, + "loss": 0.983, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 570 + }, + { + "epoch": 0.7040690505548706, + "grad_norm": 2.037432480432792, + "learning_rate": 1.7740809145306773e-05, + "loss": 1.0094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 571 + }, + { + "epoch": 0.7053020961775586, + "grad_norm": 2.0999588564507685, + "learning_rate": 1.773245621679318e-05, + "loss": 1.046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 572 + }, + { + "epoch": 0.7065351418002466, + "grad_norm": 1.9822662619027398, + "learning_rate": 1.7724089849214184e-05, + "loss": 1.0177, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 573 + }, + { + "epoch": 0.7077681874229347, + "grad_norm": 2.296463261709746, + "learning_rate": 1.77157100571106e-05, + "loss": 0.9207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 574 + }, + { + "epoch": 0.7090012330456227, + "grad_norm": 2.0434250630443045, + "learning_rate": 1.7707316855046563e-05, + "loss": 1.0088, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 575 + }, + { + "epoch": 0.7102342786683107, + "grad_norm": 2.067794231415559, + "learning_rate": 1.7698910257609524e-05, + "loss": 1.0033, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 576 + }, + { + "epoch": 0.7114673242909988, + "grad_norm": 1.9884025311346816, + "learning_rate": 1.7690490279410205e-05, + "loss": 1.0395, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 577 + }, + { + "epoch": 0.7127003699136868, + "grad_norm": 2.0663488956854676, + "learning_rate": 1.768205693508259e-05, + "loss": 1.0609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 578 + }, + { + "epoch": 0.7139334155363748, + "grad_norm": 1.9661847340747427, + "learning_rate": 1.7673610239283898e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 579 + }, + { + "epoch": 0.7151664611590629, + "grad_norm": 2.4048347386453135, + "learning_rate": 1.7665150206694543e-05, + "loss": 1.0294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 580 + }, + { + "epoch": 0.7163995067817509, + "grad_norm": 2.049926176641203, + "learning_rate": 1.765667685201813e-05, + "loss": 0.9953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 581 + }, + { + "epoch": 0.717632552404439, + "grad_norm": 2.0425901676211398, + "learning_rate": 1.7648190189981403e-05, + "loss": 1.0155, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 582 + }, + { + "epoch": 0.718865598027127, + "grad_norm": 2.111319944982802, + "learning_rate": 1.7639690235334246e-05, + "loss": 1.012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 583 + }, + { + "epoch": 0.720098643649815, + "grad_norm": 1.9625598394287143, + "learning_rate": 1.7631177002849648e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 584 + }, + { + "epoch": 0.7213316892725031, + "grad_norm": 2.0737975490297025, + "learning_rate": 1.7622650507323658e-05, + "loss": 0.9954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 585 + }, + { + "epoch": 0.7225647348951911, + "grad_norm": 2.2220755560999175, + "learning_rate": 1.7614110763575393e-05, + "loss": 1.093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 586 + }, + { + "epoch": 0.7237977805178791, + "grad_norm": 2.10369867254611, + "learning_rate": 1.7605557786446997e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 587 + }, + { + "epoch": 0.7250308261405672, + "grad_norm": 2.023199550080708, + "learning_rate": 1.7596991590803597e-05, + "loss": 1.0909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 588 + }, + { + "epoch": 0.7262638717632552, + "grad_norm": 2.1139350266949197, + "learning_rate": 1.7588412191533305e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 589 + }, + { + "epoch": 0.7274969173859432, + "grad_norm": 2.1032070896543926, + "learning_rate": 1.7579819603547187e-05, + "loss": 0.9305, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 590 + }, + { + "epoch": 0.7287299630086314, + "grad_norm": 2.1111448680444798, + "learning_rate": 1.7571213841779213e-05, + "loss": 1.0821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 591 + }, + { + "epoch": 0.7299630086313194, + "grad_norm": 2.0849758733816692, + "learning_rate": 1.7562594921186268e-05, + "loss": 0.9747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 592 + }, + { + "epoch": 0.7311960542540074, + "grad_norm": 2.7468709410129573, + "learning_rate": 1.7553962856748098e-05, + "loss": 1.0598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 593 + }, + { + "epoch": 0.7324290998766955, + "grad_norm": 2.297654831376881, + "learning_rate": 1.754531766346729e-05, + "loss": 1.0414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 594 + }, + { + "epoch": 0.7336621454993835, + "grad_norm": 2.0094458450808874, + "learning_rate": 1.7536659356369254e-05, + "loss": 0.977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 595 + }, + { + "epoch": 0.7348951911220715, + "grad_norm": 1.9967204791438438, + "learning_rate": 1.7527987950502196e-05, + "loss": 0.9274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 596 + }, + { + "epoch": 0.7361282367447596, + "grad_norm": 1.9493969416772718, + "learning_rate": 1.7519303460937073e-05, + "loss": 1.0292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 597 + }, + { + "epoch": 0.7373612823674476, + "grad_norm": 2.0023257821799247, + "learning_rate": 1.7510605902767602e-05, + "loss": 0.9574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 598 + }, + { + "epoch": 0.7385943279901356, + "grad_norm": 2.076672200643684, + "learning_rate": 1.7501895291110197e-05, + "loss": 0.9971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 599 + }, + { + "epoch": 0.7398273736128237, + "grad_norm": 2.0612367813173584, + "learning_rate": 1.7493171641103965e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 600 + }, + { + "epoch": 0.7410604192355117, + "grad_norm": 2.16071395185368, + "learning_rate": 1.7484434967910676e-05, + "loss": 1.0001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 601 + }, + { + "epoch": 0.7422934648581998, + "grad_norm": 2.0810754791890664, + "learning_rate": 1.747568528671473e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 602 + }, + { + "epoch": 0.7435265104808878, + "grad_norm": 2.0015256143479525, + "learning_rate": 1.7466922612723133e-05, + "loss": 0.9288, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 603 + }, + { + "epoch": 0.7447595561035758, + "grad_norm": 1.9574522371127243, + "learning_rate": 1.7458146961165484e-05, + "loss": 1.0365, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 604 + }, + { + "epoch": 0.7459926017262639, + "grad_norm": 1.999399085309589, + "learning_rate": 1.7449358347293925e-05, + "loss": 1.0457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 605 + }, + { + "epoch": 0.7472256473489519, + "grad_norm": 2.016778757176555, + "learning_rate": 1.7440556786383132e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 606 + }, + { + "epoch": 0.7484586929716399, + "grad_norm": 2.011184042360808, + "learning_rate": 1.743174229373028e-05, + "loss": 0.9568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 607 + }, + { + "epoch": 0.749691738594328, + "grad_norm": 1.9994468999326396, + "learning_rate": 1.7422914884655027e-05, + "loss": 0.9822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 608 + }, + { + "epoch": 0.750924784217016, + "grad_norm": 2.1373417665891163, + "learning_rate": 1.7414074574499465e-05, + "loss": 0.9128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 609 + }, + { + "epoch": 0.752157829839704, + "grad_norm": 2.414511732797334, + "learning_rate": 1.7405221378628133e-05, + "loss": 1.0866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 610 + }, + { + "epoch": 0.7533908754623921, + "grad_norm": 2.1820009401893605, + "learning_rate": 1.7396355312427935e-05, + "loss": 1.0538, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 611 + }, + { + "epoch": 0.7546239210850801, + "grad_norm": 2.0281507644839163, + "learning_rate": 1.7387476391308168e-05, + "loss": 0.9224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 612 + }, + { + "epoch": 0.7558569667077681, + "grad_norm": 2.387292229834452, + "learning_rate": 1.7378584630700454e-05, + "loss": 0.7561, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 613 + }, + { + "epoch": 0.7570900123304563, + "grad_norm": 2.0036252294718975, + "learning_rate": 1.7369680046058744e-05, + "loss": 0.9816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 614 + }, + { + "epoch": 0.7583230579531443, + "grad_norm": 2.1474265419679566, + "learning_rate": 1.7360762652859275e-05, + "loss": 0.9312, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 615 + }, + { + "epoch": 0.7595561035758323, + "grad_norm": 2.3913311043981387, + "learning_rate": 1.735183246660053e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 616 + }, + { + "epoch": 0.7607891491985204, + "grad_norm": 2.070836177310932, + "learning_rate": 1.7342889502803246e-05, + "loss": 0.8846, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 617 + }, + { + "epoch": 0.7620221948212084, + "grad_norm": 1.9909512311630846, + "learning_rate": 1.7333933777010357e-05, + "loss": 0.9683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 618 + }, + { + "epoch": 0.7632552404438965, + "grad_norm": 2.048373917294825, + "learning_rate": 1.732496530478698e-05, + "loss": 1.021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 619 + }, + { + "epoch": 0.7644882860665845, + "grad_norm": 2.046848859346405, + "learning_rate": 1.7315984101720388e-05, + "loss": 0.9949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 620 + }, + { + "epoch": 0.7657213316892725, + "grad_norm": 2.199270192761742, + "learning_rate": 1.7306990183419972e-05, + "loss": 0.9319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 621 + }, + { + "epoch": 0.7669543773119606, + "grad_norm": 2.2362149652218877, + "learning_rate": 1.729798356551723e-05, + "loss": 1.0579, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 622 + }, + { + "epoch": 0.7681874229346486, + "grad_norm": 2.072113656676678, + "learning_rate": 1.7288964263665735e-05, + "loss": 0.984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 623 + }, + { + "epoch": 0.7694204685573366, + "grad_norm": 2.2058183913371647, + "learning_rate": 1.7279932293541092e-05, + "loss": 1.0727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 624 + }, + { + "epoch": 0.7706535141800247, + "grad_norm": 2.1099223160270104, + "learning_rate": 1.7270887670840935e-05, + "loss": 0.852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 625 + }, + { + "epoch": 0.7718865598027127, + "grad_norm": 2.014275874425781, + "learning_rate": 1.7261830411284884e-05, + "loss": 1.0219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 626 + }, + { + "epoch": 0.7731196054254007, + "grad_norm": 2.1002718881715605, + "learning_rate": 1.7252760530614522e-05, + "loss": 0.9339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 627 + }, + { + "epoch": 0.7743526510480888, + "grad_norm": 2.530650182566485, + "learning_rate": 1.724367804459337e-05, + "loss": 1.1248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 628 + }, + { + "epoch": 0.7755856966707768, + "grad_norm": 2.072394234232429, + "learning_rate": 1.7234582969006847e-05, + "loss": 0.963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 629 + }, + { + "epoch": 0.7768187422934648, + "grad_norm": 2.077401892279152, + "learning_rate": 1.722547531966227e-05, + "loss": 0.9589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 630 + }, + { + "epoch": 0.7780517879161529, + "grad_norm": 2.0461500607118612, + "learning_rate": 1.7216355112388797e-05, + "loss": 0.9892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 631 + }, + { + "epoch": 0.7792848335388409, + "grad_norm": 1.9481531627295827, + "learning_rate": 1.7207222363037414e-05, + "loss": 0.9963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 632 + }, + { + "epoch": 0.7805178791615289, + "grad_norm": 2.0363659895001067, + "learning_rate": 1.7198077087480903e-05, + "loss": 0.9206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 633 + }, + { + "epoch": 0.781750924784217, + "grad_norm": 2.367528975566535, + "learning_rate": 1.718891930161383e-05, + "loss": 1.1255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 634 + }, + { + "epoch": 0.782983970406905, + "grad_norm": 2.007050047111477, + "learning_rate": 1.7179749021352482e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 635 + }, + { + "epoch": 0.7842170160295932, + "grad_norm": 1.9196848202462296, + "learning_rate": 1.7170566262634883e-05, + "loss": 0.9448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 636 + }, + { + "epoch": 0.7854500616522812, + "grad_norm": 2.145941740629715, + "learning_rate": 1.7161371041420723e-05, + "loss": 0.9428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 637 + }, + { + "epoch": 0.7866831072749692, + "grad_norm": 2.170240354212872, + "learning_rate": 1.7152163373691377e-05, + "loss": 1.0323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 638 + }, + { + "epoch": 0.7879161528976573, + "grad_norm": 2.0553179179576015, + "learning_rate": 1.714294327544983e-05, + "loss": 0.8661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 639 + }, + { + "epoch": 0.7891491985203453, + "grad_norm": 2.1113466348102397, + "learning_rate": 1.7133710762720683e-05, + "loss": 1.034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 640 + }, + { + "epoch": 0.7903822441430333, + "grad_norm": 2.146414650333374, + "learning_rate": 1.712446585155011e-05, + "loss": 1.0681, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 641 + }, + { + "epoch": 0.7916152897657214, + "grad_norm": 2.0440039273987227, + "learning_rate": 1.7115208558005834e-05, + "loss": 0.9085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 642 + }, + { + "epoch": 0.7928483353884094, + "grad_norm": 2.3653004518970535, + "learning_rate": 1.7105938898177097e-05, + "loss": 0.8514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 643 + }, + { + "epoch": 0.7940813810110974, + "grad_norm": 2.058806308472887, + "learning_rate": 1.709665688817464e-05, + "loss": 0.9258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 644 + }, + { + "epoch": 0.7953144266337855, + "grad_norm": 2.3406153888995798, + "learning_rate": 1.708736254413066e-05, + "loss": 0.9866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 645 + }, + { + "epoch": 0.7965474722564735, + "grad_norm": 3.2012831686915293, + "learning_rate": 1.70780558821988e-05, + "loss": 1.0916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 646 + }, + { + "epoch": 0.7977805178791615, + "grad_norm": 2.23457477201596, + "learning_rate": 1.7068736918554102e-05, + "loss": 0.9076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 647 + }, + { + "epoch": 0.7990135635018496, + "grad_norm": 2.255512398782474, + "learning_rate": 1.7059405669392993e-05, + "loss": 0.9113, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 648 + }, + { + "epoch": 0.8002466091245376, + "grad_norm": 2.114904513093256, + "learning_rate": 1.7050062150933252e-05, + "loss": 0.9587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 649 + }, + { + "epoch": 0.8014796547472256, + "grad_norm": 1.931643331052564, + "learning_rate": 1.7040706379413983e-05, + "loss": 0.9888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 650 + }, + { + "epoch": 0.8027127003699137, + "grad_norm": 2.0094212900142816, + "learning_rate": 1.7031338371095588e-05, + "loss": 0.9961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 651 + }, + { + "epoch": 0.8039457459926017, + "grad_norm": 2.1928733306625188, + "learning_rate": 1.702195814225973e-05, + "loss": 1.0158, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 652 + }, + { + "epoch": 0.8051787916152897, + "grad_norm": 2.2216734919656544, + "learning_rate": 1.7012565709209317e-05, + "loss": 1.0201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 653 + }, + { + "epoch": 0.8064118372379778, + "grad_norm": 2.0613969675724686, + "learning_rate": 1.700316108826847e-05, + "loss": 0.9941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 654 + }, + { + "epoch": 0.8076448828606658, + "grad_norm": 1.9372031821836655, + "learning_rate": 1.6993744295782482e-05, + "loss": 1.0026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 655 + }, + { + "epoch": 0.8088779284833539, + "grad_norm": 2.077194491232311, + "learning_rate": 1.698431534811781e-05, + "loss": 0.8491, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 656 + }, + { + "epoch": 0.8101109741060419, + "grad_norm": 2.1425857340105754, + "learning_rate": 1.6974874261662035e-05, + "loss": 0.9719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 657 + }, + { + "epoch": 0.8113440197287299, + "grad_norm": 1.8547512115782296, + "learning_rate": 1.6965421052823834e-05, + "loss": 1.0107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 658 + }, + { + "epoch": 0.812577065351418, + "grad_norm": 1.9098479812610942, + "learning_rate": 1.695595573803295e-05, + "loss": 0.95, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 659 + }, + { + "epoch": 0.813810110974106, + "grad_norm": 1.892767796833539, + "learning_rate": 1.6946478333740174e-05, + "loss": 0.86, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 660 + }, + { + "epoch": 0.815043156596794, + "grad_norm": 2.0228841723951043, + "learning_rate": 1.69369888564173e-05, + "loss": 0.9881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 661 + }, + { + "epoch": 0.8162762022194822, + "grad_norm": 2.144845399605076, + "learning_rate": 1.692748732255711e-05, + "loss": 0.8493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 662 + }, + { + "epoch": 0.8175092478421702, + "grad_norm": 2.1710135611234027, + "learning_rate": 1.6917973748673343e-05, + "loss": 0.9356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 663 + }, + { + "epoch": 0.8187422934648582, + "grad_norm": 2.200157070237925, + "learning_rate": 1.6908448151300654e-05, + "loss": 0.9922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 664 + }, + { + "epoch": 0.8199753390875463, + "grad_norm": 2.1046294162556385, + "learning_rate": 1.6898910546994605e-05, + "loss": 0.9714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 665 + }, + { + "epoch": 0.8212083847102343, + "grad_norm": 2.1921813927002622, + "learning_rate": 1.688936095233162e-05, + "loss": 1.0186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 666 + }, + { + "epoch": 0.8224414303329223, + "grad_norm": 2.5290727417605767, + "learning_rate": 1.6879799383908967e-05, + "loss": 0.9147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 667 + }, + { + "epoch": 0.8236744759556104, + "grad_norm": 2.0992091012861605, + "learning_rate": 1.6870225858344715e-05, + "loss": 1.0331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 668 + }, + { + "epoch": 0.8249075215782984, + "grad_norm": 2.228115711337027, + "learning_rate": 1.6860640392277728e-05, + "loss": 1.0372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 669 + }, + { + "epoch": 0.8261405672009864, + "grad_norm": 2.104862934581671, + "learning_rate": 1.6851043002367612e-05, + "loss": 0.9556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 670 + }, + { + "epoch": 0.8273736128236745, + "grad_norm": 2.1153449603435575, + "learning_rate": 1.6841433705294704e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 671 + }, + { + "epoch": 0.8286066584463625, + "grad_norm": 1.9113210816564241, + "learning_rate": 1.6831812517760036e-05, + "loss": 0.9553, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 672 + }, + { + "epoch": 0.8298397040690506, + "grad_norm": 2.146278639653085, + "learning_rate": 1.6822179456485292e-05, + "loss": 0.8621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 673 + }, + { + "epoch": 0.8310727496917386, + "grad_norm": 2.092521119953653, + "learning_rate": 1.681253453821281e-05, + "loss": 0.9576, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 674 + }, + { + "epoch": 0.8323057953144266, + "grad_norm": 2.108757495841555, + "learning_rate": 1.6802877779705526e-05, + "loss": 0.9349, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 675 + }, + { + "epoch": 0.8335388409371147, + "grad_norm": 2.165541838354097, + "learning_rate": 1.679320919774696e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 676 + }, + { + "epoch": 0.8347718865598027, + "grad_norm": 2.0901280484241322, + "learning_rate": 1.6783528809141175e-05, + "loss": 0.9577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 677 + }, + { + "epoch": 0.8360049321824907, + "grad_norm": 2.049770632568499, + "learning_rate": 1.6773836630712757e-05, + "loss": 0.8912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 678 + }, + { + "epoch": 0.8372379778051788, + "grad_norm": 1.9415439366071066, + "learning_rate": 1.6764132679306785e-05, + "loss": 0.9678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 679 + }, + { + "epoch": 0.8384710234278668, + "grad_norm": 2.159517279560615, + "learning_rate": 1.6754416971788795e-05, + "loss": 0.9739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 680 + }, + { + "epoch": 0.8397040690505548, + "grad_norm": 2.0578899051147745, + "learning_rate": 1.674468952504476e-05, + "loss": 0.9187, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 681 + }, + { + "epoch": 0.840937114673243, + "grad_norm": 2.1905094099573135, + "learning_rate": 1.673495035598105e-05, + "loss": 0.9753, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 682 + }, + { + "epoch": 0.842170160295931, + "grad_norm": 2.194631531971183, + "learning_rate": 1.6725199481524418e-05, + "loss": 1.0267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 683 + }, + { + "epoch": 0.843403205918619, + "grad_norm": 2.040276435918069, + "learning_rate": 1.671543691862195e-05, + "loss": 0.9703, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 684 + }, + { + "epoch": 0.8446362515413071, + "grad_norm": 1.9627382699239366, + "learning_rate": 1.6705662684241053e-05, + "loss": 0.8955, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 685 + }, + { + "epoch": 0.8458692971639951, + "grad_norm": 2.288828851032713, + "learning_rate": 1.6695876795369418e-05, + "loss": 0.946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 686 + }, + { + "epoch": 0.8471023427866831, + "grad_norm": 2.1152782339808884, + "learning_rate": 1.668607926901499e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 687 + }, + { + "epoch": 0.8483353884093712, + "grad_norm": 2.798938790414006, + "learning_rate": 1.6676270122205946e-05, + "loss": 0.8988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 688 + }, + { + "epoch": 0.8495684340320592, + "grad_norm": 2.5307103300223766, + "learning_rate": 1.6666449371990644e-05, + "loss": 0.8814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 689 + }, + { + "epoch": 0.8508014796547472, + "grad_norm": 2.2986978033861294, + "learning_rate": 1.6656617035437635e-05, + "loss": 0.9402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 690 + }, + { + "epoch": 0.8520345252774353, + "grad_norm": 2.1645514508732546, + "learning_rate": 1.664677312963558e-05, + "loss": 0.9551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 691 + }, + { + "epoch": 0.8532675709001233, + "grad_norm": 2.144781636504613, + "learning_rate": 1.6636917671693264e-05, + "loss": 1.0505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 692 + }, + { + "epoch": 0.8545006165228114, + "grad_norm": 2.068308383986026, + "learning_rate": 1.6627050678739544e-05, + "loss": 0.9237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 693 + }, + { + "epoch": 0.8557336621454994, + "grad_norm": 2.4201265499871325, + "learning_rate": 1.6617172167923327e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 694 + }, + { + "epoch": 0.8569667077681874, + "grad_norm": 1.9743124171175193, + "learning_rate": 1.660728215641353e-05, + "loss": 0.9649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 695 + }, + { + "epoch": 0.8581997533908755, + "grad_norm": 2.4976283410874847, + "learning_rate": 1.6597380661399077e-05, + "loss": 0.9445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 696 + }, + { + "epoch": 0.8594327990135635, + "grad_norm": 2.2344847337230855, + "learning_rate": 1.6587467700088834e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 697 + }, + { + "epoch": 0.8606658446362515, + "grad_norm": 2.1366592393550343, + "learning_rate": 1.6577543289711597e-05, + "loss": 0.8845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 698 + }, + { + "epoch": 0.8618988902589396, + "grad_norm": 2.0879117181580753, + "learning_rate": 1.6567607447516064e-05, + "loss": 0.7972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 699 + }, + { + "epoch": 0.8631319358816276, + "grad_norm": 2.117890235500837, + "learning_rate": 1.6557660190770803e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 700 + }, + { + "epoch": 0.8643649815043156, + "grad_norm": 2.227804015025162, + "learning_rate": 1.654770153676422e-05, + "loss": 0.9105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 701 + }, + { + "epoch": 0.8655980271270037, + "grad_norm": 2.262676552209609, + "learning_rate": 1.6537731502804525e-05, + "loss": 0.9706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 702 + }, + { + "epoch": 0.8668310727496917, + "grad_norm": 2.146446661141531, + "learning_rate": 1.6527750106219715e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 703 + }, + { + "epoch": 0.8680641183723797, + "grad_norm": 2.08167189099969, + "learning_rate": 1.6517757364357526e-05, + "loss": 0.9381, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 704 + }, + { + "epoch": 0.8692971639950678, + "grad_norm": 1.9310579355796305, + "learning_rate": 1.6507753294585417e-05, + "loss": 0.9047, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 705 + }, + { + "epoch": 0.8705302096177558, + "grad_norm": 2.0325358606748978, + "learning_rate": 1.6497737914290535e-05, + "loss": 0.9094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 706 + }, + { + "epoch": 0.8717632552404438, + "grad_norm": 2.0543982954124966, + "learning_rate": 1.648771124087969e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 707 + }, + { + "epoch": 0.872996300863132, + "grad_norm": 2.295806181212224, + "learning_rate": 1.64776732917793e-05, + "loss": 0.9444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 708 + }, + { + "epoch": 0.87422934648582, + "grad_norm": 2.1899960866925303, + "learning_rate": 1.646762408443541e-05, + "loss": 1.0083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 709 + }, + { + "epoch": 0.8754623921085081, + "grad_norm": 2.040526145894979, + "learning_rate": 1.6457563636313598e-05, + "loss": 0.8215, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 710 + }, + { + "epoch": 0.8766954377311961, + "grad_norm": 1.9967218569140772, + "learning_rate": 1.6447491964899008e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 711 + }, + { + "epoch": 0.8779284833538841, + "grad_norm": 2.131774886953309, + "learning_rate": 1.6437409087696283e-05, + "loss": 0.8786, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 712 + }, + { + "epoch": 0.8791615289765722, + "grad_norm": 2.0642410512824414, + "learning_rate": 1.6427315022229525e-05, + "loss": 0.8831, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 713 + }, + { + "epoch": 0.8803945745992602, + "grad_norm": 2.2793436000912255, + "learning_rate": 1.64172097860423e-05, + "loss": 0.8413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 714 + }, + { + "epoch": 0.8816276202219482, + "grad_norm": 1.9501864497105927, + "learning_rate": 1.640709339669758e-05, + "loss": 0.8856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 715 + }, + { + "epoch": 0.8828606658446363, + "grad_norm": 2.0279122102292977, + "learning_rate": 1.6396965871777727e-05, + "loss": 0.8835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 716 + }, + { + "epoch": 0.8840937114673243, + "grad_norm": 2.181455569803073, + "learning_rate": 1.638682722888445e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 717 + }, + { + "epoch": 0.8853267570900123, + "grad_norm": 2.1877848027614313, + "learning_rate": 1.637667748563879e-05, + "loss": 0.8314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 718 + }, + { + "epoch": 0.8865598027127004, + "grad_norm": 2.0901252578597713, + "learning_rate": 1.6366516659681067e-05, + "loss": 0.9333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 719 + }, + { + "epoch": 0.8877928483353884, + "grad_norm": 2.1846108211938673, + "learning_rate": 1.6356344768670875e-05, + "loss": 0.9309, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 720 + }, + { + "epoch": 0.8890258939580764, + "grad_norm": 2.600109718803851, + "learning_rate": 1.6346161830287034e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 721 + }, + { + "epoch": 0.8902589395807645, + "grad_norm": 2.2057915960500276, + "learning_rate": 1.6335967862227567e-05, + "loss": 0.9764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 722 + }, + { + "epoch": 0.8914919852034525, + "grad_norm": 2.1009909758008267, + "learning_rate": 1.632576288220966e-05, + "loss": 0.8741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 723 + }, + { + "epoch": 0.8927250308261405, + "grad_norm": 2.1617838023881, + "learning_rate": 1.6315546907969652e-05, + "loss": 0.9573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 724 + }, + { + "epoch": 0.8939580764488286, + "grad_norm": 2.091522410165622, + "learning_rate": 1.6305319957262972e-05, + "loss": 0.9534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 725 + }, + { + "epoch": 0.8951911220715166, + "grad_norm": 2.2845231322682418, + "learning_rate": 1.6295082047864138e-05, + "loss": 0.9156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 726 + }, + { + "epoch": 0.8964241676942046, + "grad_norm": 2.062614641816067, + "learning_rate": 1.6284833197566712e-05, + "loss": 0.9539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 727 + }, + { + "epoch": 0.8976572133168927, + "grad_norm": 2.157165674006361, + "learning_rate": 1.627457342418327e-05, + "loss": 0.9926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 728 + }, + { + "epoch": 0.8988902589395807, + "grad_norm": 2.039737473184125, + "learning_rate": 1.6264302745545374e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 729 + }, + { + "epoch": 0.9001233045622689, + "grad_norm": 2.057800120641984, + "learning_rate": 1.625402117950354e-05, + "loss": 0.9139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 730 + }, + { + "epoch": 0.9013563501849569, + "grad_norm": 2.240321306869684, + "learning_rate": 1.62437287439272e-05, + "loss": 1.0758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 731 + }, + { + "epoch": 0.9025893958076449, + "grad_norm": 2.0947466285552436, + "learning_rate": 1.6233425456704686e-05, + "loss": 0.8467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 732 + }, + { + "epoch": 0.903822441430333, + "grad_norm": 2.061121182714553, + "learning_rate": 1.6223111335743187e-05, + "loss": 0.9093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 733 + }, + { + "epoch": 0.905055487053021, + "grad_norm": 2.209914581799182, + "learning_rate": 1.6212786398968717e-05, + "loss": 0.8805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 734 + }, + { + "epoch": 0.906288532675709, + "grad_norm": 2.468473842682747, + "learning_rate": 1.6202450664326093e-05, + "loss": 0.9013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 735 + }, + { + "epoch": 0.9075215782983971, + "grad_norm": 2.3365951628381056, + "learning_rate": 1.61921041497789e-05, + "loss": 0.9112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 736 + }, + { + "epoch": 0.9087546239210851, + "grad_norm": 2.474277588242769, + "learning_rate": 1.6181746873309455e-05, + "loss": 0.9186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 737 + }, + { + "epoch": 0.9099876695437731, + "grad_norm": 2.0716027082353228, + "learning_rate": 1.6171378852918778e-05, + "loss": 0.9486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 738 + }, + { + "epoch": 0.9112207151664612, + "grad_norm": 2.165213004434134, + "learning_rate": 1.6161000106626564e-05, + "loss": 0.9726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 739 + }, + { + "epoch": 0.9124537607891492, + "grad_norm": 2.4218033902151475, + "learning_rate": 1.615061065247115e-05, + "loss": 0.8259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 740 + }, + { + "epoch": 0.9136868064118372, + "grad_norm": 1.970928985580259, + "learning_rate": 1.614021050850949e-05, + "loss": 0.9788, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 741 + }, + { + "epoch": 0.9149198520345253, + "grad_norm": 2.2639971055328934, + "learning_rate": 1.6129799692817097e-05, + "loss": 0.8745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 742 + }, + { + "epoch": 0.9161528976572133, + "grad_norm": 2.5602631751385028, + "learning_rate": 1.611937822348806e-05, + "loss": 0.9512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 743 + }, + { + "epoch": 0.9173859432799013, + "grad_norm": 2.22841865866248, + "learning_rate": 1.6108946118634957e-05, + "loss": 0.805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 744 + }, + { + "epoch": 0.9186189889025894, + "grad_norm": 2.4182686550221626, + "learning_rate": 1.6098503396388868e-05, + "loss": 1.0202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 745 + }, + { + "epoch": 0.9198520345252774, + "grad_norm": 2.2298122692949645, + "learning_rate": 1.608805007489932e-05, + "loss": 0.8972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 746 + }, + { + "epoch": 0.9210850801479655, + "grad_norm": 2.362160156268979, + "learning_rate": 1.607758617233426e-05, + "loss": 0.9039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 747 + }, + { + "epoch": 0.9223181257706535, + "grad_norm": 2.1106519602619533, + "learning_rate": 1.6067111706880027e-05, + "loss": 0.8318, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 748 + }, + { + "epoch": 0.9235511713933415, + "grad_norm": 2.0825477479243424, + "learning_rate": 1.605662669674132e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 749 + }, + { + "epoch": 0.9247842170160296, + "grad_norm": 2.036239449814601, + "learning_rate": 1.6046131160141167e-05, + "loss": 0.8727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 750 + }, + { + "epoch": 0.9260172626387176, + "grad_norm": 2.1703406452205787, + "learning_rate": 1.603562511532088e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 751 + }, + { + "epoch": 0.9272503082614056, + "grad_norm": 2.1398345899160143, + "learning_rate": 1.6025108580540046e-05, + "loss": 0.8691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 752 + }, + { + "epoch": 0.9284833538840938, + "grad_norm": 2.0888823283975375, + "learning_rate": 1.601458157407648e-05, + "loss": 0.8547, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 753 + }, + { + "epoch": 0.9297163995067818, + "grad_norm": 2.269523743217552, + "learning_rate": 1.6004044114226193e-05, + "loss": 1.0141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 754 + }, + { + "epoch": 0.9309494451294698, + "grad_norm": 2.0091635625820063, + "learning_rate": 1.5993496219303367e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 755 + }, + { + "epoch": 0.9321824907521579, + "grad_norm": 2.1386095902203612, + "learning_rate": 1.5982937907640327e-05, + "loss": 0.9388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 756 + }, + { + "epoch": 0.9334155363748459, + "grad_norm": 2.1260034823775467, + "learning_rate": 1.5972369197587486e-05, + "loss": 0.9153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 757 + }, + { + "epoch": 0.9346485819975339, + "grad_norm": 2.1225283234506156, + "learning_rate": 1.5961790107513343e-05, + "loss": 1.0178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 758 + }, + { + "epoch": 0.935881627620222, + "grad_norm": 2.0422192666840795, + "learning_rate": 1.595120065580444e-05, + "loss": 0.9856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 759 + }, + { + "epoch": 0.93711467324291, + "grad_norm": 2.051963293225036, + "learning_rate": 1.5940600860865315e-05, + "loss": 0.8559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 760 + }, + { + "epoch": 0.938347718865598, + "grad_norm": 2.127612803214104, + "learning_rate": 1.592999074111849e-05, + "loss": 0.8298, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 761 + }, + { + "epoch": 0.9395807644882861, + "grad_norm": 2.286554941831041, + "learning_rate": 1.591937031500443e-05, + "loss": 0.8796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 762 + }, + { + "epoch": 0.9408138101109741, + "grad_norm": 2.282412038316439, + "learning_rate": 1.5908739600981517e-05, + "loss": 1.022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 763 + }, + { + "epoch": 0.9420468557336621, + "grad_norm": 2.197857160398704, + "learning_rate": 1.5898098617526006e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 764 + }, + { + "epoch": 0.9432799013563502, + "grad_norm": 2.1702846778760607, + "learning_rate": 1.5887447383132008e-05, + "loss": 0.9507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 765 + }, + { + "epoch": 0.9445129469790382, + "grad_norm": 2.045700294662405, + "learning_rate": 1.587678591631144e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 766 + }, + { + "epoch": 0.9457459926017263, + "grad_norm": 2.426595174168101, + "learning_rate": 1.5866114235594014e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 767 + }, + { + "epoch": 0.9469790382244143, + "grad_norm": 2.085172001225346, + "learning_rate": 1.585543235952718e-05, + "loss": 0.9166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 768 + }, + { + "epoch": 0.9482120838471023, + "grad_norm": 2.1181076484114665, + "learning_rate": 1.5844740306676133e-05, + "loss": 0.8829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 769 + }, + { + "epoch": 0.9494451294697904, + "grad_norm": 2.64845194344853, + "learning_rate": 1.583403809562373e-05, + "loss": 0.8927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 770 + }, + { + "epoch": 0.9506781750924784, + "grad_norm": 2.4929599774250186, + "learning_rate": 1.5823325744970488e-05, + "loss": 0.9407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 771 + }, + { + "epoch": 0.9519112207151664, + "grad_norm": 2.368031870072283, + "learning_rate": 1.581260327333456e-05, + "loss": 0.9463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 772 + }, + { + "epoch": 0.9531442663378545, + "grad_norm": 2.136177985699207, + "learning_rate": 1.580187069935167e-05, + "loss": 0.8571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 773 + }, + { + "epoch": 0.9543773119605425, + "grad_norm": 2.128181510143405, + "learning_rate": 1.5791128041675114e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 774 + }, + { + "epoch": 0.9556103575832305, + "grad_norm": 2.220458022640665, + "learning_rate": 1.5780375318975715e-05, + "loss": 0.8445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 775 + }, + { + "epoch": 0.9568434032059187, + "grad_norm": 2.098330673951412, + "learning_rate": 1.5769612549941777e-05, + "loss": 0.964, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 776 + }, + { + "epoch": 0.9580764488286067, + "grad_norm": 2.094864541454194, + "learning_rate": 1.575883975327907e-05, + "loss": 0.8254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 777 + }, + { + "epoch": 0.9593094944512947, + "grad_norm": 2.3159416415455047, + "learning_rate": 1.5748056947710804e-05, + "loss": 0.9472, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 778 + }, + { + "epoch": 0.9605425400739828, + "grad_norm": 1.9950192842009455, + "learning_rate": 1.573726415197756e-05, + "loss": 0.7965, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 779 + }, + { + "epoch": 0.9617755856966708, + "grad_norm": 2.006344851533844, + "learning_rate": 1.5726461384837306e-05, + "loss": 0.8203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 780 + }, + { + "epoch": 0.9630086313193588, + "grad_norm": 2.2545503988149007, + "learning_rate": 1.5715648665065324e-05, + "loss": 0.9282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 781 + }, + { + "epoch": 0.9642416769420469, + "grad_norm": 2.2372695819528294, + "learning_rate": 1.5704826011454203e-05, + "loss": 0.9608, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 782 + }, + { + "epoch": 0.9654747225647349, + "grad_norm": 2.1710522433750357, + "learning_rate": 1.5693993442813792e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 783 + }, + { + "epoch": 0.966707768187423, + "grad_norm": 2.400731212803864, + "learning_rate": 1.568315097797118e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 784 + }, + { + "epoch": 0.967940813810111, + "grad_norm": 2.096674414566688, + "learning_rate": 1.5672298635770645e-05, + "loss": 0.6728, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 785 + }, + { + "epoch": 0.969173859432799, + "grad_norm": 2.2214704493931205, + "learning_rate": 1.566143643507364e-05, + "loss": 1.0617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 786 + }, + { + "epoch": 0.9704069050554871, + "grad_norm": 2.21278029237075, + "learning_rate": 1.5650564394758743e-05, + "loss": 0.8487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 787 + }, + { + "epoch": 0.9716399506781751, + "grad_norm": 2.3400569095138763, + "learning_rate": 1.5639682533721643e-05, + "loss": 0.9532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 788 + }, + { + "epoch": 0.9728729963008631, + "grad_norm": 2.1937723298850784, + "learning_rate": 1.5628790870875093e-05, + "loss": 0.9272, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 789 + }, + { + "epoch": 0.9741060419235512, + "grad_norm": 2.117801922125244, + "learning_rate": 1.561788942514889e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 790 + }, + { + "epoch": 0.9753390875462392, + "grad_norm": 2.436307402269952, + "learning_rate": 1.560697821548982e-05, + "loss": 0.8599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 791 + }, + { + "epoch": 0.9765721331689272, + "grad_norm": 2.2722775653462848, + "learning_rate": 1.5596057260861645e-05, + "loss": 0.8503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 792 + }, + { + "epoch": 0.9778051787916153, + "grad_norm": 2.22960384352197, + "learning_rate": 1.5585126580245066e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 793 + }, + { + "epoch": 0.9790382244143033, + "grad_norm": 2.1141630678799275, + "learning_rate": 1.557418619263768e-05, + "loss": 0.8418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 794 + }, + { + "epoch": 0.9802712700369913, + "grad_norm": 2.4590856272408224, + "learning_rate": 1.5563236117053965e-05, + "loss": 0.9197, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 795 + }, + { + "epoch": 0.9815043156596794, + "grad_norm": 2.678264319346431, + "learning_rate": 1.5552276372525237e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 796 + }, + { + "epoch": 0.9827373612823674, + "grad_norm": 2.262445020658096, + "learning_rate": 1.554130697809961e-05, + "loss": 0.9101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 797 + }, + { + "epoch": 0.9839704069050554, + "grad_norm": 2.327233920544754, + "learning_rate": 1.5530327952841965e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 798 + }, + { + "epoch": 0.9852034525277436, + "grad_norm": 2.1249691788358334, + "learning_rate": 1.551933931583393e-05, + "loss": 0.8824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 799 + }, + { + "epoch": 0.9864364981504316, + "grad_norm": 2.0648396343207325, + "learning_rate": 1.5508341086173843e-05, + "loss": 0.8895, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 800 + }, + { + "epoch": 0.9876695437731196, + "grad_norm": 2.109919443242373, + "learning_rate": 1.5497333282976698e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 801 + }, + { + "epoch": 0.9889025893958077, + "grad_norm": 2.3306692057820797, + "learning_rate": 1.5486315925374146e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 802 + }, + { + "epoch": 0.9901356350184957, + "grad_norm": 2.365009122779881, + "learning_rate": 1.547528903251443e-05, + "loss": 1.0604, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 803 + }, + { + "epoch": 0.9913686806411838, + "grad_norm": 2.2289048394713262, + "learning_rate": 1.5464252623562375e-05, + "loss": 0.8141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 804 + }, + { + "epoch": 0.9926017262638718, + "grad_norm": 2.457422539818384, + "learning_rate": 1.5453206717699334e-05, + "loss": 1.0228, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 805 + }, + { + "epoch": 0.9938347718865598, + "grad_norm": 2.0659430518171558, + "learning_rate": 1.544215133412318e-05, + "loss": 0.8641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 806 + }, + { + "epoch": 0.9950678175092479, + "grad_norm": 2.154728199437817, + "learning_rate": 1.543108649204824e-05, + "loss": 0.8029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 807 + }, + { + "epoch": 0.9963008631319359, + "grad_norm": 2.3679631548982196, + "learning_rate": 1.54200122107053e-05, + "loss": 0.866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 808 + }, + { + "epoch": 0.9975339087546239, + "grad_norm": 2.218685707693942, + "learning_rate": 1.540892850934154e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 809 + }, + { + "epoch": 0.998766954377312, + "grad_norm": 2.3274870898977187, + "learning_rate": 1.5397835407220508e-05, + "loss": 0.9724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 810 + }, + { + "epoch": 1.0, + "grad_norm": 2.159965188324255, + "learning_rate": 1.53867329236221e-05, + "loss": 0.762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 811 + }, + { + "epoch": 1.0012330456226881, + "grad_norm": 2.8883947578149782, + "learning_rate": 1.5375621077842517e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 812 + }, + { + "epoch": 1.002466091245376, + "grad_norm": 2.1490979216413963, + "learning_rate": 1.5364499889194225e-05, + "loss": 0.8653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 813 + }, + { + "epoch": 1.0036991368680641, + "grad_norm": 2.2100133362019787, + "learning_rate": 1.5353369377005928e-05, + "loss": 0.8871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 814 + }, + { + "epoch": 1.0049321824907522, + "grad_norm": 2.1549827645085444, + "learning_rate": 1.5342229560622545e-05, + "loss": 0.9146, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 815 + }, + { + "epoch": 1.0061652281134401, + "grad_norm": 2.0929823364178155, + "learning_rate": 1.5331080459405153e-05, + "loss": 0.8513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 816 + }, + { + "epoch": 1.0073982737361282, + "grad_norm": 2.220368656816853, + "learning_rate": 1.5319922092730974e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 817 + }, + { + "epoch": 1.0086313193588163, + "grad_norm": 2.1262847612815197, + "learning_rate": 1.5308754479993326e-05, + "loss": 0.8457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 818 + }, + { + "epoch": 1.0098643649815042, + "grad_norm": 2.1791530750144443, + "learning_rate": 1.5297577640601607e-05, + "loss": 0.7645, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 819 + }, + { + "epoch": 1.0110974106041923, + "grad_norm": 2.1726539474102426, + "learning_rate": 1.5286391593981243e-05, + "loss": 0.8684, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 820 + }, + { + "epoch": 1.0123304562268804, + "grad_norm": 2.193333692016209, + "learning_rate": 1.5275196359573667e-05, + "loss": 0.7566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 821 + }, + { + "epoch": 1.0135635018495683, + "grad_norm": 2.9212088946802623, + "learning_rate": 1.526399195683627e-05, + "loss": 1.0019, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 822 + }, + { + "epoch": 1.0147965474722564, + "grad_norm": 2.0580823212102417, + "learning_rate": 1.52527784052424e-05, + "loss": 0.9646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 823 + }, + { + "epoch": 1.0160295930949446, + "grad_norm": 2.19343597586527, + "learning_rate": 1.5241555724281283e-05, + "loss": 0.9453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 824 + }, + { + "epoch": 1.0172626387176325, + "grad_norm": 2.294578049001136, + "learning_rate": 1.523032393345802e-05, + "loss": 1.0273, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 825 + }, + { + "epoch": 1.0184956843403206, + "grad_norm": 1.9511593069959263, + "learning_rate": 1.5219083052293546e-05, + "loss": 0.8161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 826 + }, + { + "epoch": 1.0197287299630087, + "grad_norm": 2.421213793478604, + "learning_rate": 1.5207833100324595e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 827 + }, + { + "epoch": 1.0209617755856966, + "grad_norm": 2.1254569627132422, + "learning_rate": 1.5196574097103669e-05, + "loss": 1.0024, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 828 + }, + { + "epoch": 1.0221948212083847, + "grad_norm": 2.1741883221339213, + "learning_rate": 1.5185306062198993e-05, + "loss": 0.8694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 829 + }, + { + "epoch": 1.0234278668310728, + "grad_norm": 2.340571355784493, + "learning_rate": 1.5174029015194496e-05, + "loss": 0.9235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 830 + }, + { + "epoch": 1.024660912453761, + "grad_norm": 2.173793908678777, + "learning_rate": 1.5162742975689769e-05, + "loss": 0.8448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 831 + }, + { + "epoch": 1.0258939580764488, + "grad_norm": 2.149214329172728, + "learning_rate": 1.5151447963300027e-05, + "loss": 0.8388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 832 + }, + { + "epoch": 1.027127003699137, + "grad_norm": 2.059108722897467, + "learning_rate": 1.5140143997656092e-05, + "loss": 0.7303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 833 + }, + { + "epoch": 1.028360049321825, + "grad_norm": 2.2165337335607536, + "learning_rate": 1.512883109840433e-05, + "loss": 0.873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 834 + }, + { + "epoch": 1.029593094944513, + "grad_norm": 2.0093629352145923, + "learning_rate": 1.511750928520665e-05, + "loss": 0.7804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 835 + }, + { + "epoch": 1.030826140567201, + "grad_norm": 1.9963120627280608, + "learning_rate": 1.5106178577740442e-05, + "loss": 0.8642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 836 + }, + { + "epoch": 1.0320591861898891, + "grad_norm": 2.1963002199807384, + "learning_rate": 1.5094838995698555e-05, + "loss": 0.8688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 837 + }, + { + "epoch": 1.033292231812577, + "grad_norm": 2.1874228944498473, + "learning_rate": 1.5083490558789275e-05, + "loss": 0.9912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 838 + }, + { + "epoch": 1.0345252774352651, + "grad_norm": 2.12533056093305, + "learning_rate": 1.5072133286736261e-05, + "loss": 0.7925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 839 + }, + { + "epoch": 1.0357583230579532, + "grad_norm": 2.5000691541559745, + "learning_rate": 1.5060767199278536e-05, + "loss": 0.9544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 840 + }, + { + "epoch": 1.0369913686806411, + "grad_norm": 2.0622785615015125, + "learning_rate": 1.5049392316170443e-05, + "loss": 0.8433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 841 + }, + { + "epoch": 1.0382244143033292, + "grad_norm": 2.2009668761322563, + "learning_rate": 1.5038008657181616e-05, + "loss": 0.8157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 842 + }, + { + "epoch": 1.0394574599260173, + "grad_norm": 2.0959248263408385, + "learning_rate": 1.5026616242096932e-05, + "loss": 0.9167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 843 + }, + { + "epoch": 1.0406905055487052, + "grad_norm": 2.1893631450779893, + "learning_rate": 1.5015215090716493e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 844 + }, + { + "epoch": 1.0419235511713933, + "grad_norm": 2.077397878791412, + "learning_rate": 1.5003805222855584e-05, + "loss": 0.8383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 845 + }, + { + "epoch": 1.0431565967940815, + "grad_norm": 2.4055084927944623, + "learning_rate": 1.4992386658344643e-05, + "loss": 0.8404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 846 + }, + { + "epoch": 1.0443896424167693, + "grad_norm": 2.171777480987558, + "learning_rate": 1.4980959417029208e-05, + "loss": 0.7997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 847 + }, + { + "epoch": 1.0456226880394575, + "grad_norm": 2.157645250920109, + "learning_rate": 1.4969523518769917e-05, + "loss": 0.8738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 848 + }, + { + "epoch": 1.0468557336621456, + "grad_norm": 2.396639754777913, + "learning_rate": 1.4958078983442441e-05, + "loss": 0.6992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 849 + }, + { + "epoch": 1.0480887792848335, + "grad_norm": 2.4186334460214765, + "learning_rate": 1.4946625830937472e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 850 + }, + { + "epoch": 1.0493218249075216, + "grad_norm": 2.2977540698502996, + "learning_rate": 1.4935164081160663e-05, + "loss": 0.8683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 851 + }, + { + "epoch": 1.0505548705302097, + "grad_norm": 2.1379865919168, + "learning_rate": 1.4923693754032626e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 852 + }, + { + "epoch": 1.0517879161528976, + "grad_norm": 2.259076969791956, + "learning_rate": 1.491221486948887e-05, + "loss": 0.7691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 853 + }, + { + "epoch": 1.0530209617755857, + "grad_norm": 2.1377806626548357, + "learning_rate": 1.490072744747978e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 854 + }, + { + "epoch": 1.0542540073982738, + "grad_norm": 2.1044084023956167, + "learning_rate": 1.4889231507970581e-05, + "loss": 0.8401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 855 + }, + { + "epoch": 1.0554870530209617, + "grad_norm": 2.1271140055745286, + "learning_rate": 1.4877727070941302e-05, + "loss": 0.8084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 856 + }, + { + "epoch": 1.0567200986436498, + "grad_norm": 2.1718066261306817, + "learning_rate": 1.4866214156386734e-05, + "loss": 0.9014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 857 + }, + { + "epoch": 1.057953144266338, + "grad_norm": 2.1176596897595714, + "learning_rate": 1.4854692784316409e-05, + "loss": 0.7752, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 858 + }, + { + "epoch": 1.0591861898890258, + "grad_norm": 2.171891486084014, + "learning_rate": 1.4843162974754555e-05, + "loss": 0.6988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 859 + }, + { + "epoch": 1.060419235511714, + "grad_norm": 2.0055850881304718, + "learning_rate": 1.4831624747740065e-05, + "loss": 0.7928, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 860 + }, + { + "epoch": 1.061652281134402, + "grad_norm": 2.1272055457979624, + "learning_rate": 1.4820078123326464e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 861 + }, + { + "epoch": 1.06288532675709, + "grad_norm": 2.2409166254632416, + "learning_rate": 1.4808523121581871e-05, + "loss": 0.8428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 862 + }, + { + "epoch": 1.064118372379778, + "grad_norm": 2.167571295890406, + "learning_rate": 1.4796959762588957e-05, + "loss": 0.9326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 863 + }, + { + "epoch": 1.0653514180024661, + "grad_norm": 1.9896497336935064, + "learning_rate": 1.4785388066444928e-05, + "loss": 0.7178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 864 + }, + { + "epoch": 1.066584463625154, + "grad_norm": 1.9782830220589522, + "learning_rate": 1.4773808053261478e-05, + "loss": 0.8734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 865 + }, + { + "epoch": 1.0678175092478421, + "grad_norm": 2.1094953563881447, + "learning_rate": 1.4762219743164752e-05, + "loss": 0.72, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 866 + }, + { + "epoch": 1.0690505548705302, + "grad_norm": 2.172378157117765, + "learning_rate": 1.4750623156295316e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 867 + }, + { + "epoch": 1.0702836004932181, + "grad_norm": 2.777083857064727, + "learning_rate": 1.473901831280813e-05, + "loss": 0.887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 868 + }, + { + "epoch": 1.0715166461159062, + "grad_norm": 2.773482686862429, + "learning_rate": 1.4727405232872489e-05, + "loss": 0.9897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 869 + }, + { + "epoch": 1.0727496917385944, + "grad_norm": 2.303987298346897, + "learning_rate": 1.4715783936672017e-05, + "loss": 0.8267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 870 + }, + { + "epoch": 1.0739827373612825, + "grad_norm": 2.5090103249366718, + "learning_rate": 1.4704154444404606e-05, + "loss": 0.8297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 871 + }, + { + "epoch": 1.0752157829839704, + "grad_norm": 2.1804014393140334, + "learning_rate": 1.4692516776282404e-05, + "loss": 0.8673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 872 + }, + { + "epoch": 1.0764488286066585, + "grad_norm": 2.114741425345962, + "learning_rate": 1.4680870952531758e-05, + "loss": 0.7878, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 873 + }, + { + "epoch": 1.0776818742293466, + "grad_norm": 2.1487592065744043, + "learning_rate": 1.4669216993393195e-05, + "loss": 0.702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 874 + }, + { + "epoch": 1.0789149198520345, + "grad_norm": 2.250699104710453, + "learning_rate": 1.4657554919121388e-05, + "loss": 0.7805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 875 + }, + { + "epoch": 1.0801479654747226, + "grad_norm": 2.652130922890576, + "learning_rate": 1.46458847499851e-05, + "loss": 0.8594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 876 + }, + { + "epoch": 1.0813810110974107, + "grad_norm": 2.618911622711661, + "learning_rate": 1.4634206506267172e-05, + "loss": 0.974, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 877 + }, + { + "epoch": 1.0826140567200986, + "grad_norm": 2.104711420819804, + "learning_rate": 1.462252020826448e-05, + "loss": 0.7511, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 878 + }, + { + "epoch": 1.0838471023427867, + "grad_norm": 2.181334503374274, + "learning_rate": 1.4610825876287892e-05, + "loss": 0.8632, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 879 + }, + { + "epoch": 1.0850801479654748, + "grad_norm": 2.2281952334339126, + "learning_rate": 1.4599123530662242e-05, + "loss": 0.7623, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 880 + }, + { + "epoch": 1.0863131935881627, + "grad_norm": 2.2779707650922254, + "learning_rate": 1.4587413191726296e-05, + "loss": 0.7533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 881 + }, + { + "epoch": 1.0875462392108508, + "grad_norm": 2.227912879258747, + "learning_rate": 1.4575694879832704e-05, + "loss": 0.9742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 882 + }, + { + "epoch": 1.088779284833539, + "grad_norm": 2.200751986221906, + "learning_rate": 1.4563968615347983e-05, + "loss": 0.9859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 883 + }, + { + "epoch": 1.0900123304562268, + "grad_norm": 2.1474005643679503, + "learning_rate": 1.455223441865246e-05, + "loss": 0.885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 884 + }, + { + "epoch": 1.091245376078915, + "grad_norm": 2.5635895574555967, + "learning_rate": 1.4540492310140262e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 885 + }, + { + "epoch": 1.092478421701603, + "grad_norm": 2.086551890894098, + "learning_rate": 1.4528742310219251e-05, + "loss": 0.7838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 886 + }, + { + "epoch": 1.093711467324291, + "grad_norm": 2.100577896933077, + "learning_rate": 1.4516984439311017e-05, + "loss": 0.8245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 887 + }, + { + "epoch": 1.094944512946979, + "grad_norm": 2.0824253868885427, + "learning_rate": 1.4505218717850826e-05, + "loss": 0.8121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 888 + }, + { + "epoch": 1.0961775585696671, + "grad_norm": 2.424807773941655, + "learning_rate": 1.449344516628759e-05, + "loss": 0.6852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 889 + }, + { + "epoch": 1.097410604192355, + "grad_norm": 2.2203614431330587, + "learning_rate": 1.4481663805083823e-05, + "loss": 0.9022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 890 + }, + { + "epoch": 1.0986436498150431, + "grad_norm": 2.194684728317858, + "learning_rate": 1.4469874654715618e-05, + "loss": 0.8018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 891 + }, + { + "epoch": 1.0998766954377313, + "grad_norm": 2.2557943511918066, + "learning_rate": 1.4458077735672606e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 892 + }, + { + "epoch": 1.1011097410604191, + "grad_norm": 2.350846713741873, + "learning_rate": 1.4446273068457916e-05, + "loss": 0.8373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 893 + }, + { + "epoch": 1.1023427866831073, + "grad_norm": 2.3206912584026584, + "learning_rate": 1.4434460673588151e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 894 + }, + { + "epoch": 1.1035758323057954, + "grad_norm": 2.083251461091603, + "learning_rate": 1.4422640571593336e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 895 + }, + { + "epoch": 1.1048088779284833, + "grad_norm": 2.2429047820431456, + "learning_rate": 1.4410812783016893e-05, + "loss": 0.8066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 896 + }, + { + "epoch": 1.1060419235511714, + "grad_norm": 2.39736831788261, + "learning_rate": 1.4398977328415608e-05, + "loss": 0.8961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 897 + }, + { + "epoch": 1.1072749691738595, + "grad_norm": 2.4569397246397995, + "learning_rate": 1.4387134228359587e-05, + "loss": 0.9745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 898 + }, + { + "epoch": 1.1085080147965474, + "grad_norm": 2.194341713414294, + "learning_rate": 1.4375283503432223e-05, + "loss": 0.7642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 899 + }, + { + "epoch": 1.1097410604192355, + "grad_norm": 2.108094195937167, + "learning_rate": 1.4363425174230161e-05, + "loss": 0.7856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 900 + }, + { + "epoch": 1.1109741060419236, + "grad_norm": 2.140652113297743, + "learning_rate": 1.4351559261363267e-05, + "loss": 0.6648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 901 + }, + { + "epoch": 1.1122071516646117, + "grad_norm": 2.1936452316718063, + "learning_rate": 1.4339685785454587e-05, + "loss": 0.7336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 902 + }, + { + "epoch": 1.1134401972872996, + "grad_norm": 2.2269402261375038, + "learning_rate": 1.4327804767140303e-05, + "loss": 0.7708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 903 + }, + { + "epoch": 1.1146732429099877, + "grad_norm": 2.2989998022422453, + "learning_rate": 1.4315916227069713e-05, + "loss": 0.743, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 904 + }, + { + "epoch": 1.1159062885326758, + "grad_norm": 2.2917968867963427, + "learning_rate": 1.4304020185905189e-05, + "loss": 0.8126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 905 + }, + { + "epoch": 1.1171393341553637, + "grad_norm": 2.217875844794138, + "learning_rate": 1.4292116664322134e-05, + "loss": 0.6544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 906 + }, + { + "epoch": 1.1183723797780518, + "grad_norm": 2.4162119819935803, + "learning_rate": 1.4280205683008959e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 907 + }, + { + "epoch": 1.11960542540074, + "grad_norm": 2.3603757259614047, + "learning_rate": 1.4268287262667033e-05, + "loss": 0.7766, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 908 + }, + { + "epoch": 1.1208384710234278, + "grad_norm": 2.323141591914079, + "learning_rate": 1.4256361424010663e-05, + "loss": 0.8892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 909 + }, + { + "epoch": 1.122071516646116, + "grad_norm": 2.196302554500726, + "learning_rate": 1.4244428187767039e-05, + "loss": 0.9378, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 910 + }, + { + "epoch": 1.123304562268804, + "grad_norm": 2.2395403393552393, + "learning_rate": 1.4232487574676212e-05, + "loss": 1.0055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 911 + }, + { + "epoch": 1.124537607891492, + "grad_norm": 2.15769661287864, + "learning_rate": 1.4220539605491057e-05, + "loss": 0.748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 912 + }, + { + "epoch": 1.12577065351418, + "grad_norm": 2.0601339442187605, + "learning_rate": 1.4208584300977233e-05, + "loss": 0.7894, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 913 + }, + { + "epoch": 1.1270036991368682, + "grad_norm": 2.10447957980717, + "learning_rate": 1.4196621681913145e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 914 + }, + { + "epoch": 1.128236744759556, + "grad_norm": 2.3961229528506265, + "learning_rate": 1.4184651769089911e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 915 + }, + { + "epoch": 1.1294697903822442, + "grad_norm": 2.457287266967065, + "learning_rate": 1.417267458331133e-05, + "loss": 0.8519, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 916 + }, + { + "epoch": 1.1307028360049323, + "grad_norm": 2.023004753946474, + "learning_rate": 1.4160690145393838e-05, + "loss": 0.8449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 917 + }, + { + "epoch": 1.1319358816276202, + "grad_norm": 2.2533720475013634, + "learning_rate": 1.4148698476166471e-05, + "loss": 0.8221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 918 + }, + { + "epoch": 1.1331689272503083, + "grad_norm": 2.105957236229041, + "learning_rate": 1.4136699596470846e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 919 + }, + { + "epoch": 1.1344019728729964, + "grad_norm": 2.1652513956532946, + "learning_rate": 1.4124693527161095e-05, + "loss": 0.8322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 920 + }, + { + "epoch": 1.1356350184956843, + "grad_norm": 2.2463359048305693, + "learning_rate": 1.4112680289103858e-05, + "loss": 0.8458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 921 + }, + { + "epoch": 1.1368680641183724, + "grad_norm": 2.1946517808282784, + "learning_rate": 1.4100659903178231e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 922 + }, + { + "epoch": 1.1381011097410605, + "grad_norm": 2.116652661158196, + "learning_rate": 1.4088632390275731e-05, + "loss": 0.8496, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 923 + }, + { + "epoch": 1.1393341553637484, + "grad_norm": 2.1260376630680677, + "learning_rate": 1.4076597771300267e-05, + "loss": 0.7646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 924 + }, + { + "epoch": 1.1405672009864365, + "grad_norm": 2.1008385205849684, + "learning_rate": 1.4064556067168092e-05, + "loss": 0.8303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 925 + }, + { + "epoch": 1.1418002466091246, + "grad_norm": 2.1995475134175573, + "learning_rate": 1.4052507298807774e-05, + "loss": 0.8636, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 926 + }, + { + "epoch": 1.1430332922318125, + "grad_norm": 2.238025377523696, + "learning_rate": 1.404045148716016e-05, + "loss": 0.9095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 927 + }, + { + "epoch": 1.1442663378545006, + "grad_norm": 2.1584647692083636, + "learning_rate": 1.4028388653178346e-05, + "loss": 0.9552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 928 + }, + { + "epoch": 1.1454993834771887, + "grad_norm": 2.277913607137681, + "learning_rate": 1.4016318817827618e-05, + "loss": 0.6826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 929 + }, + { + "epoch": 1.1467324290998766, + "grad_norm": 2.1210301206048494, + "learning_rate": 1.4004242002085441e-05, + "loss": 0.7648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 930 + }, + { + "epoch": 1.1479654747225647, + "grad_norm": 2.1730216571228143, + "learning_rate": 1.3992158226941409e-05, + "loss": 0.7348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 931 + }, + { + "epoch": 1.1491985203452528, + "grad_norm": 2.2903920964212197, + "learning_rate": 1.3980067513397208e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 932 + }, + { + "epoch": 1.1504315659679407, + "grad_norm": 2.207582230014581, + "learning_rate": 1.3967969882466595e-05, + "loss": 0.8393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 933 + }, + { + "epoch": 1.1516646115906288, + "grad_norm": 2.1487993957640605, + "learning_rate": 1.3955865355175334e-05, + "loss": 0.9618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 934 + }, + { + "epoch": 1.152897657213317, + "grad_norm": 2.1100013581157673, + "learning_rate": 1.3943753952561179e-05, + "loss": 0.7185, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 935 + }, + { + "epoch": 1.1541307028360048, + "grad_norm": 2.1580493494759505, + "learning_rate": 1.3931635695673845e-05, + "loss": 0.8656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 936 + }, + { + "epoch": 1.155363748458693, + "grad_norm": 2.449600660057695, + "learning_rate": 1.3919510605574944e-05, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 937 + }, + { + "epoch": 1.156596794081381, + "grad_norm": 2.2204766170828782, + "learning_rate": 1.3907378703337977e-05, + "loss": 0.823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 938 + }, + { + "epoch": 1.157829839704069, + "grad_norm": 2.563974651134322, + "learning_rate": 1.389524001004827e-05, + "loss": 0.7685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 939 + }, + { + "epoch": 1.159062885326757, + "grad_norm": 2.258821529498267, + "learning_rate": 1.388309454680297e-05, + "loss": 0.673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 940 + }, + { + "epoch": 1.1602959309494452, + "grad_norm": 2.292191724384705, + "learning_rate": 1.3870942334710978e-05, + "loss": 0.8204, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 941 + }, + { + "epoch": 1.161528976572133, + "grad_norm": 2.283429457608797, + "learning_rate": 1.3858783394892925e-05, + "loss": 0.7777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 942 + }, + { + "epoch": 1.1627620221948212, + "grad_norm": 2.186748888333124, + "learning_rate": 1.3846617748481137e-05, + "loss": 0.7862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 943 + }, + { + "epoch": 1.1639950678175093, + "grad_norm": 2.262303399886297, + "learning_rate": 1.3834445416619602e-05, + "loss": 0.7559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 944 + }, + { + "epoch": 1.1652281134401972, + "grad_norm": 2.262832253985301, + "learning_rate": 1.3822266420463914e-05, + "loss": 0.9429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 945 + }, + { + "epoch": 1.1664611590628853, + "grad_norm": 2.150186199729938, + "learning_rate": 1.3810080781181264e-05, + "loss": 0.814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 946 + }, + { + "epoch": 1.1676942046855734, + "grad_norm": 2.343694448653775, + "learning_rate": 1.3797888519950373e-05, + "loss": 0.93, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 947 + }, + { + "epoch": 1.1689272503082615, + "grad_norm": 2.1813728957413385, + "learning_rate": 1.3785689657961494e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 948 + }, + { + "epoch": 1.1701602959309494, + "grad_norm": 2.119521364690522, + "learning_rate": 1.3773484216416326e-05, + "loss": 0.6718, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 949 + }, + { + "epoch": 1.1713933415536375, + "grad_norm": 2.2831289714651537, + "learning_rate": 1.3761272216528023e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 950 + }, + { + "epoch": 1.1726263871763256, + "grad_norm": 2.431755167028705, + "learning_rate": 1.3749053679521123e-05, + "loss": 0.8664, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 951 + }, + { + "epoch": 1.1738594327990135, + "grad_norm": 2.148435358078549, + "learning_rate": 1.3736828626631539e-05, + "loss": 0.6746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 952 + }, + { + "epoch": 1.1750924784217016, + "grad_norm": 2.124965667169116, + "learning_rate": 1.3724597079106498e-05, + "loss": 0.9202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 953 + }, + { + "epoch": 1.1763255240443897, + "grad_norm": 2.2382797156893397, + "learning_rate": 1.3712359058204525e-05, + "loss": 0.7696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 954 + }, + { + "epoch": 1.1775585696670776, + "grad_norm": 2.2207967591300632, + "learning_rate": 1.3700114585195381e-05, + "loss": 0.8654, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 955 + }, + { + "epoch": 1.1787916152897657, + "grad_norm": 2.3010582856815347, + "learning_rate": 1.3687863681360059e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 956 + }, + { + "epoch": 1.1800246609124538, + "grad_norm": 2.146527954418168, + "learning_rate": 1.367560636799071e-05, + "loss": 0.6291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 957 + }, + { + "epoch": 1.1812577065351417, + "grad_norm": 2.203217169625211, + "learning_rate": 1.3663342666390642e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 958 + }, + { + "epoch": 1.1824907521578298, + "grad_norm": 2.380280804061753, + "learning_rate": 1.365107259787425e-05, + "loss": 0.8015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 959 + }, + { + "epoch": 1.183723797780518, + "grad_norm": 2.7057879375717766, + "learning_rate": 1.3638796183767013e-05, + "loss": 0.8469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 960 + }, + { + "epoch": 1.1849568434032058, + "grad_norm": 2.0549864237362856, + "learning_rate": 1.362651344540542e-05, + "loss": 0.7724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 961 + }, + { + "epoch": 1.186189889025894, + "grad_norm": 2.165523344913771, + "learning_rate": 1.3614224404136964e-05, + "loss": 0.7119, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 962 + }, + { + "epoch": 1.187422934648582, + "grad_norm": 2.193503404484943, + "learning_rate": 1.3601929081320085e-05, + "loss": 0.769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 963 + }, + { + "epoch": 1.18865598027127, + "grad_norm": 2.30296466364579, + "learning_rate": 1.358962749832415e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 964 + }, + { + "epoch": 1.189889025893958, + "grad_norm": 2.3370321886887226, + "learning_rate": 1.3577319676529394e-05, + "loss": 0.8251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 965 + }, + { + "epoch": 1.1911220715166462, + "grad_norm": 2.1447616659612727, + "learning_rate": 1.3565005637326902e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 966 + }, + { + "epoch": 1.192355117139334, + "grad_norm": 2.1126610192998263, + "learning_rate": 1.3552685402118569e-05, + "loss": 0.893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 967 + }, + { + "epoch": 1.1935881627620222, + "grad_norm": 2.145481581859885, + "learning_rate": 1.3540358992317048e-05, + "loss": 0.8777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 968 + }, + { + "epoch": 1.1948212083847103, + "grad_norm": 2.0436843289234714, + "learning_rate": 1.3528026429345733e-05, + "loss": 0.7041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 969 + }, + { + "epoch": 1.1960542540073984, + "grad_norm": 2.3729052917065863, + "learning_rate": 1.351568773463871e-05, + "loss": 0.7822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 970 + }, + { + "epoch": 1.1972872996300863, + "grad_norm": 2.2771218973970977, + "learning_rate": 1.3503342929640713e-05, + "loss": 0.6887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 971 + }, + { + "epoch": 1.1985203452527744, + "grad_norm": 2.114743129136917, + "learning_rate": 1.3490992035807108e-05, + "loss": 0.6587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 972 + }, + { + "epoch": 1.1997533908754625, + "grad_norm": 2.210055964932895, + "learning_rate": 1.3478635074603844e-05, + "loss": 0.6986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 973 + }, + { + "epoch": 1.2009864364981504, + "grad_norm": 2.376467969342028, + "learning_rate": 1.34662720675074e-05, + "loss": 0.8992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 974 + }, + { + "epoch": 1.2022194821208385, + "grad_norm": 2.3169180290916893, + "learning_rate": 1.3453903036004784e-05, + "loss": 0.8468, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 975 + }, + { + "epoch": 1.2034525277435266, + "grad_norm": 2.4929171744348095, + "learning_rate": 1.3441528001593448e-05, + "loss": 0.8746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 976 + }, + { + "epoch": 1.2046855733662145, + "grad_norm": 2.3498012552061223, + "learning_rate": 1.3429146985781305e-05, + "loss": 0.848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 977 + }, + { + "epoch": 1.2059186189889026, + "grad_norm": 2.3519351668545223, + "learning_rate": 1.3416760010086647e-05, + "loss": 0.8045, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 978 + }, + { + "epoch": 1.2071516646115907, + "grad_norm": 2.152128272403449, + "learning_rate": 1.3404367096038127e-05, + "loss": 0.6621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 979 + }, + { + "epoch": 1.2083847102342786, + "grad_norm": 2.2310865684580126, + "learning_rate": 1.339196826517472e-05, + "loss": 0.8074, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 980 + }, + { + "epoch": 1.2096177558569667, + "grad_norm": 2.4218206259337336, + "learning_rate": 1.3379563539045688e-05, + "loss": 0.7489, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 981 + }, + { + "epoch": 1.2108508014796548, + "grad_norm": 2.3171020050526754, + "learning_rate": 1.3367152939210532e-05, + "loss": 0.6967, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 982 + }, + { + "epoch": 1.2120838471023427, + "grad_norm": 2.253666664504199, + "learning_rate": 1.3354736487238968e-05, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 983 + }, + { + "epoch": 1.2133168927250308, + "grad_norm": 2.202923113820963, + "learning_rate": 1.3342314204710878e-05, + "loss": 0.6643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 984 + }, + { + "epoch": 1.214549938347719, + "grad_norm": 2.3907506714199216, + "learning_rate": 1.332988611321628e-05, + "loss": 0.7483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 985 + }, + { + "epoch": 1.2157829839704068, + "grad_norm": 2.3288855585799726, + "learning_rate": 1.3317452234355286e-05, + "loss": 0.8775, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 986 + }, + { + "epoch": 1.217016029593095, + "grad_norm": 2.4270404362139124, + "learning_rate": 1.3305012589738075e-05, + "loss": 0.9152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 987 + }, + { + "epoch": 1.218249075215783, + "grad_norm": 2.3906498632865825, + "learning_rate": 1.3292567200984833e-05, + "loss": 0.6825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 988 + }, + { + "epoch": 1.219482120838471, + "grad_norm": 2.1776282578685495, + "learning_rate": 1.3280116089725743e-05, + "loss": 0.7098, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 989 + }, + { + "epoch": 1.220715166461159, + "grad_norm": 2.3972197284848265, + "learning_rate": 1.3267659277600921e-05, + "loss": 0.8804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 990 + }, + { + "epoch": 1.2219482120838472, + "grad_norm": 2.272454158882342, + "learning_rate": 1.3255196786260402e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 991 + }, + { + "epoch": 1.223181257706535, + "grad_norm": 2.0898886573973985, + "learning_rate": 1.3242728637364084e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 992 + }, + { + "epoch": 1.2244143033292232, + "grad_norm": 2.1283771526768422, + "learning_rate": 1.3230254852581703e-05, + "loss": 0.7541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 993 + }, + { + "epoch": 1.2256473489519113, + "grad_norm": 2.5947641083937785, + "learning_rate": 1.3217775453592786e-05, + "loss": 0.6516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 994 + }, + { + "epoch": 1.2268803945745992, + "grad_norm": 2.2994678179446155, + "learning_rate": 1.3205290462086623e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 995 + }, + { + "epoch": 1.2281134401972873, + "grad_norm": 2.419174005886699, + "learning_rate": 1.3192799899762216e-05, + "loss": 0.6653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 996 + }, + { + "epoch": 1.2293464858199754, + "grad_norm": 2.7052778802902724, + "learning_rate": 1.318030378832826e-05, + "loss": 0.8398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 997 + }, + { + "epoch": 1.2305795314426633, + "grad_norm": 2.182898548205837, + "learning_rate": 1.3167802149503079e-05, + "loss": 0.7902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 998 + }, + { + "epoch": 1.2318125770653514, + "grad_norm": 2.319601966909909, + "learning_rate": 1.315529500501462e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 999 + }, + { + "epoch": 1.2330456226880395, + "grad_norm": 2.3070852285534875, + "learning_rate": 1.3142782376600389e-05, + "loss": 0.7369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1000 + }, + { + "epoch": 1.2342786683107274, + "grad_norm": 2.3823064106232956, + "learning_rate": 1.3130264286007427e-05, + "loss": 0.8759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1001 + }, + { + "epoch": 1.2355117139334155, + "grad_norm": 2.2502915191290405, + "learning_rate": 1.3117740754992263e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1002 + }, + { + "epoch": 1.2367447595561036, + "grad_norm": 2.329022774742351, + "learning_rate": 1.310521180532089e-05, + "loss": 0.6222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1003 + }, + { + "epoch": 1.2379778051787915, + "grad_norm": 2.451176371025935, + "learning_rate": 1.3092677458768712e-05, + "loss": 0.9302, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1004 + }, + { + "epoch": 1.2392108508014796, + "grad_norm": 2.266087500621217, + "learning_rate": 1.3080137737120514e-05, + "loss": 0.7132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1005 + }, + { + "epoch": 1.2404438964241677, + "grad_norm": 2.2897262982963205, + "learning_rate": 1.3067592662170424e-05, + "loss": 0.7692, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1006 + }, + { + "epoch": 1.2416769420468556, + "grad_norm": 2.321897633330746, + "learning_rate": 1.3055042255721877e-05, + "loss": 0.8054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1007 + }, + { + "epoch": 1.2429099876695437, + "grad_norm": 2.327507620912996, + "learning_rate": 1.3042486539587566e-05, + "loss": 0.8495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1008 + }, + { + "epoch": 1.2441430332922319, + "grad_norm": 2.144625913300158, + "learning_rate": 1.3029925535589423e-05, + "loss": 0.7546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1009 + }, + { + "epoch": 1.2453760789149197, + "grad_norm": 2.1664603999352217, + "learning_rate": 1.301735926555856e-05, + "loss": 0.758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1010 + }, + { + "epoch": 1.2466091245376079, + "grad_norm": 2.2424395261695422, + "learning_rate": 1.300478775133525e-05, + "loss": 0.6794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1011 + }, + { + "epoch": 1.247842170160296, + "grad_norm": 2.135987556369877, + "learning_rate": 1.2992211014768872e-05, + "loss": 0.7814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1012 + }, + { + "epoch": 1.2490752157829839, + "grad_norm": 2.3429443823198812, + "learning_rate": 1.297962907771789e-05, + "loss": 0.756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1013 + }, + { + "epoch": 1.250308261405672, + "grad_norm": 2.3806466769436083, + "learning_rate": 1.29670419620498e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1014 + }, + { + "epoch": 1.25154130702836, + "grad_norm": 2.2614648564029536, + "learning_rate": 1.2954449689641105e-05, + "loss": 0.6227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1015 + }, + { + "epoch": 1.252774352651048, + "grad_norm": 2.690263466615422, + "learning_rate": 1.294185228237726e-05, + "loss": 0.8233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1016 + }, + { + "epoch": 1.254007398273736, + "grad_norm": 2.473463220519423, + "learning_rate": 1.2929249762152659e-05, + "loss": 0.7693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1017 + }, + { + "epoch": 1.2552404438964242, + "grad_norm": 2.203357353825905, + "learning_rate": 1.2916642150870567e-05, + "loss": 0.7486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1018 + }, + { + "epoch": 1.256473489519112, + "grad_norm": 2.407886169080358, + "learning_rate": 1.2904029470443105e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1019 + }, + { + "epoch": 1.2577065351418002, + "grad_norm": 2.416338321530697, + "learning_rate": 1.289141174279121e-05, + "loss": 0.7849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1020 + }, + { + "epoch": 1.2589395807644883, + "grad_norm": 2.1578914081370617, + "learning_rate": 1.2878788989844577e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1021 + }, + { + "epoch": 1.2601726263871762, + "grad_norm": 2.111289324604086, + "learning_rate": 1.2866161233541648e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1022 + }, + { + "epoch": 1.2614056720098643, + "grad_norm": 2.463350736417066, + "learning_rate": 1.2853528495829557e-05, + "loss": 0.7408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1023 + }, + { + "epoch": 1.2626387176325524, + "grad_norm": 2.5227574157477424, + "learning_rate": 1.284089079866409e-05, + "loss": 0.6943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1024 + }, + { + "epoch": 1.2638717632552403, + "grad_norm": 2.550177757961712, + "learning_rate": 1.2828248164009658e-05, + "loss": 0.8063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1025 + }, + { + "epoch": 1.2651048088779284, + "grad_norm": 2.279399386552855, + "learning_rate": 1.2815600613839254e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1026 + }, + { + "epoch": 1.2663378545006165, + "grad_norm": 2.403266684764464, + "learning_rate": 1.2802948170134411e-05, + "loss": 0.8159, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1027 + }, + { + "epoch": 1.2675709001233046, + "grad_norm": 2.3974372248557088, + "learning_rate": 1.2790290854885173e-05, + "loss": 0.7733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1028 + }, + { + "epoch": 1.2688039457459925, + "grad_norm": 2.75325380275428, + "learning_rate": 1.2777628690090035e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1029 + }, + { + "epoch": 1.2700369913686806, + "grad_norm": 2.253046145862575, + "learning_rate": 1.2764961697755947e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1030 + }, + { + "epoch": 1.2712700369913688, + "grad_norm": 2.0946644139817967, + "learning_rate": 1.2752289899898225e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1031 + }, + { + "epoch": 1.2725030826140566, + "grad_norm": 2.2771512156869576, + "learning_rate": 1.2739613318540546e-05, + "loss": 0.8351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1032 + }, + { + "epoch": 1.2737361282367448, + "grad_norm": 2.360325908676525, + "learning_rate": 1.2726931975714904e-05, + "loss": 0.8367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1033 + }, + { + "epoch": 1.2749691738594329, + "grad_norm": 2.310569034824901, + "learning_rate": 1.2714245893461568e-05, + "loss": 0.7346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1034 + }, + { + "epoch": 1.2762022194821208, + "grad_norm": 2.1364800896900413, + "learning_rate": 1.2701555093829034e-05, + "loss": 0.6637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1035 + }, + { + "epoch": 1.2774352651048089, + "grad_norm": 2.2320085770485734, + "learning_rate": 1.2688859598874013e-05, + "loss": 0.6849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1036 + }, + { + "epoch": 1.278668310727497, + "grad_norm": 2.2073805269530067, + "learning_rate": 1.2676159430661361e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1037 + }, + { + "epoch": 1.279901356350185, + "grad_norm": 2.2795714091180694, + "learning_rate": 1.2663454611264065e-05, + "loss": 0.8542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1038 + }, + { + "epoch": 1.281134401972873, + "grad_norm": 2.4086359220078535, + "learning_rate": 1.2650745162763195e-05, + "loss": 0.6699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1039 + }, + { + "epoch": 1.282367447595561, + "grad_norm": 2.3833290498603787, + "learning_rate": 1.2638031107247871e-05, + "loss": 0.7976, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1040 + }, + { + "epoch": 1.2836004932182492, + "grad_norm": 2.4078757121437873, + "learning_rate": 1.26253124668152e-05, + "loss": 0.7284, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1041 + }, + { + "epoch": 1.284833538840937, + "grad_norm": 2.3345693841609045, + "learning_rate": 1.2612589263570287e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1042 + }, + { + "epoch": 1.2860665844636252, + "grad_norm": 2.2804486360942633, + "learning_rate": 1.2599861519626142e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1043 + }, + { + "epoch": 1.2872996300863133, + "grad_norm": 2.219596427341039, + "learning_rate": 1.2587129257103684e-05, + "loss": 0.723, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1044 + }, + { + "epoch": 1.2885326757090012, + "grad_norm": 2.318632178358617, + "learning_rate": 1.2574392498131671e-05, + "loss": 0.7334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1045 + }, + { + "epoch": 1.2897657213316893, + "grad_norm": 2.1919076842831933, + "learning_rate": 1.2561651264846692e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1046 + }, + { + "epoch": 1.2909987669543774, + "grad_norm": 2.4787933074669186, + "learning_rate": 1.2548905579393096e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1047 + }, + { + "epoch": 1.2922318125770653, + "grad_norm": 2.4023622763250883, + "learning_rate": 1.2536155463922982e-05, + "loss": 0.826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1048 + }, + { + "epoch": 1.2934648581997534, + "grad_norm": 2.117125770604436, + "learning_rate": 1.252340094059614e-05, + "loss": 0.8052, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1049 + }, + { + "epoch": 1.2946979038224415, + "grad_norm": 2.2921041661226926, + "learning_rate": 1.251064203158003e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1050 + }, + { + "epoch": 1.2959309494451294, + "grad_norm": 2.5171575376979334, + "learning_rate": 1.2497878759049722e-05, + "loss": 0.6485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1051 + }, + { + "epoch": 1.2971639950678175, + "grad_norm": 2.1128556413546806, + "learning_rate": 1.248511114518788e-05, + "loss": 0.6076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1052 + }, + { + "epoch": 1.2983970406905057, + "grad_norm": 2.1493124843340907, + "learning_rate": 1.2472339212184712e-05, + "loss": 0.7343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1053 + }, + { + "epoch": 1.2996300863131935, + "grad_norm": 2.889368089302167, + "learning_rate": 1.2459562982237933e-05, + "loss": 0.7978, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1054 + }, + { + "epoch": 1.3008631319358817, + "grad_norm": 2.495383798802243, + "learning_rate": 1.2446782477552718e-05, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1055 + }, + { + "epoch": 1.3020961775585698, + "grad_norm": 2.517167273202911, + "learning_rate": 1.2433997720341682e-05, + "loss": 0.8663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1056 + }, + { + "epoch": 1.3033292231812577, + "grad_norm": 2.33725722509719, + "learning_rate": 1.2421208732824823e-05, + "loss": 0.7587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1057 + }, + { + "epoch": 1.3045622688039458, + "grad_norm": 2.29088652068904, + "learning_rate": 1.2408415537229497e-05, + "loss": 0.8488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1058 + }, + { + "epoch": 1.3057953144266339, + "grad_norm": 2.283233783970098, + "learning_rate": 1.239561815579037e-05, + "loss": 0.6389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1059 + }, + { + "epoch": 1.3070283600493218, + "grad_norm": 2.2928393078851568, + "learning_rate": 1.2382816610749388e-05, + "loss": 0.7261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1060 + }, + { + "epoch": 1.3082614056720099, + "grad_norm": 2.163749641191809, + "learning_rate": 1.2370010924355726e-05, + "loss": 0.6366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1061 + }, + { + "epoch": 1.309494451294698, + "grad_norm": 2.1229525047726963, + "learning_rate": 1.2357201118865762e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1062 + }, + { + "epoch": 1.3107274969173859, + "grad_norm": 2.272499973037141, + "learning_rate": 1.234438721654303e-05, + "loss": 0.6777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1063 + }, + { + "epoch": 1.311960542540074, + "grad_norm": 2.2172413085789207, + "learning_rate": 1.233156923965819e-05, + "loss": 0.8492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1064 + }, + { + "epoch": 1.313193588162762, + "grad_norm": 2.215217499299492, + "learning_rate": 1.2318747210488973e-05, + "loss": 0.816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1065 + }, + { + "epoch": 1.31442663378545, + "grad_norm": 2.2046121405476384, + "learning_rate": 1.2305921151320164e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1066 + }, + { + "epoch": 1.315659679408138, + "grad_norm": 2.410535974252942, + "learning_rate": 1.2293091084443547e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1067 + }, + { + "epoch": 1.3168927250308262, + "grad_norm": 2.2346919759512427, + "learning_rate": 1.2280257032157873e-05, + "loss": 0.5882, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1068 + }, + { + "epoch": 1.318125770653514, + "grad_norm": 2.217353695685963, + "learning_rate": 1.2267419016768814e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1069 + }, + { + "epoch": 1.3193588162762022, + "grad_norm": 2.3310973124804892, + "learning_rate": 1.225457706058894e-05, + "loss": 0.8387, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1070 + }, + { + "epoch": 1.3205918618988903, + "grad_norm": 2.363091956760975, + "learning_rate": 1.224173118593766e-05, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1071 + }, + { + "epoch": 1.3218249075215782, + "grad_norm": 2.3530303319299835, + "learning_rate": 1.22288814151412e-05, + "loss": 0.7271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1072 + }, + { + "epoch": 1.3230579531442663, + "grad_norm": 2.3412040138713137, + "learning_rate": 1.2216027770532559e-05, + "loss": 0.7482, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1073 + }, + { + "epoch": 1.3242909987669544, + "grad_norm": 2.388832200908418, + "learning_rate": 1.2203170274451462e-05, + "loss": 0.6927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1074 + }, + { + "epoch": 1.3255240443896423, + "grad_norm": 2.107265565119065, + "learning_rate": 1.219030894924433e-05, + "loss": 0.6069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1075 + }, + { + "epoch": 1.3267570900123304, + "grad_norm": 2.249447889766729, + "learning_rate": 1.2177443817264241e-05, + "loss": 0.7429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1076 + }, + { + "epoch": 1.3279901356350186, + "grad_norm": 2.337103531405435, + "learning_rate": 1.2164574900870887e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1077 + }, + { + "epoch": 1.3292231812577064, + "grad_norm": 2.4954282627925806, + "learning_rate": 1.2151702222430542e-05, + "loss": 0.781, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1078 + }, + { + "epoch": 1.3304562268803946, + "grad_norm": 2.4023586193420714, + "learning_rate": 1.2138825804316015e-05, + "loss": 0.7126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1079 + }, + { + "epoch": 1.3316892725030827, + "grad_norm": 2.293592183809771, + "learning_rate": 1.2125945668906611e-05, + "loss": 0.859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1080 + }, + { + "epoch": 1.3329223181257706, + "grad_norm": 2.280827683105619, + "learning_rate": 1.2113061838588104e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1081 + }, + { + "epoch": 1.3341553637484587, + "grad_norm": 2.2618423177043647, + "learning_rate": 1.2100174335752683e-05, + "loss": 0.8226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1082 + }, + { + "epoch": 1.3353884093711468, + "grad_norm": 2.263174374532883, + "learning_rate": 1.2087283182798923e-05, + "loss": 0.8671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1083 + }, + { + "epoch": 1.3366214549938347, + "grad_norm": 2.3088875209533875, + "learning_rate": 1.2074388402131737e-05, + "loss": 0.7575, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1084 + }, + { + "epoch": 1.3378545006165228, + "grad_norm": 2.217074039041576, + "learning_rate": 1.2061490016162357e-05, + "loss": 0.8498, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1085 + }, + { + "epoch": 1.339087546239211, + "grad_norm": 2.558767932572183, + "learning_rate": 1.2048588047308263e-05, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1086 + }, + { + "epoch": 1.3403205918618988, + "grad_norm": 2.1426590192119392, + "learning_rate": 1.203568251799318e-05, + "loss": 0.8473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1087 + }, + { + "epoch": 1.341553637484587, + "grad_norm": 2.3130116023826903, + "learning_rate": 1.2022773450647002e-05, + "loss": 0.6533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1088 + }, + { + "epoch": 1.342786683107275, + "grad_norm": 2.2474452770176505, + "learning_rate": 1.2009860867705793e-05, + "loss": 0.7731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1089 + }, + { + "epoch": 1.344019728729963, + "grad_norm": 2.553351511741815, + "learning_rate": 1.1996944791611711e-05, + "loss": 0.8881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1090 + }, + { + "epoch": 1.345252774352651, + "grad_norm": 2.4417233034036747, + "learning_rate": 1.1984025244812987e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1091 + }, + { + "epoch": 1.3464858199753391, + "grad_norm": 2.02108560427268, + "learning_rate": 1.1971102249763895e-05, + "loss": 0.5307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1092 + }, + { + "epoch": 1.347718865598027, + "grad_norm": 2.4627196027809655, + "learning_rate": 1.1958175828924692e-05, + "loss": 0.6095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1093 + }, + { + "epoch": 1.3489519112207151, + "grad_norm": 2.2087226765380703, + "learning_rate": 1.194524600476159e-05, + "loss": 0.7225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1094 + }, + { + "epoch": 1.3501849568434032, + "grad_norm": 2.268031072263066, + "learning_rate": 1.1932312799746722e-05, + "loss": 0.7492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1095 + }, + { + "epoch": 1.3514180024660911, + "grad_norm": 2.187415973980175, + "learning_rate": 1.1919376236358087e-05, + "loss": 0.6372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1096 + }, + { + "epoch": 1.3526510480887792, + "grad_norm": 2.3776853142465755, + "learning_rate": 1.1906436337079531e-05, + "loss": 0.757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1097 + }, + { + "epoch": 1.3538840937114673, + "grad_norm": 2.633083891326042, + "learning_rate": 1.189349312440069e-05, + "loss": 0.741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1098 + }, + { + "epoch": 1.3551171393341552, + "grad_norm": 2.3175112181742703, + "learning_rate": 1.1880546620816971e-05, + "loss": 0.577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1099 + }, + { + "epoch": 1.3563501849568433, + "grad_norm": 2.57847955905109, + "learning_rate": 1.1867596848829479e-05, + "loss": 0.7485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1100 + }, + { + "epoch": 1.3575832305795315, + "grad_norm": 2.2807181061723223, + "learning_rate": 1.1854643830945023e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1101 + }, + { + "epoch": 1.3588162762022196, + "grad_norm": 2.150298874669539, + "learning_rate": 1.1841687589676037e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1102 + }, + { + "epoch": 1.3600493218249075, + "grad_norm": 2.239857482190429, + "learning_rate": 1.1828728147540565e-05, + "loss": 0.5845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1103 + }, + { + "epoch": 1.3612823674475956, + "grad_norm": 2.391169760631781, + "learning_rate": 1.1815765527062207e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1104 + }, + { + "epoch": 1.3625154130702837, + "grad_norm": 2.233510553215683, + "learning_rate": 1.18027997507701e-05, + "loss": 0.687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1105 + }, + { + "epoch": 1.3637484586929716, + "grad_norm": 2.3406994321803376, + "learning_rate": 1.1789830841198854e-05, + "loss": 0.541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1106 + }, + { + "epoch": 1.3649815043156597, + "grad_norm": 3.0813520964620884, + "learning_rate": 1.1776858820888532e-05, + "loss": 0.8939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1107 + }, + { + "epoch": 1.3662145499383478, + "grad_norm": 2.289968441014776, + "learning_rate": 1.1763883712384593e-05, + "loss": 0.6178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1108 + }, + { + "epoch": 1.3674475955610357, + "grad_norm": 2.753373823273893, + "learning_rate": 1.1750905538237877e-05, + "loss": 0.8218, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1109 + }, + { + "epoch": 1.3686806411837238, + "grad_norm": 2.582931593682227, + "learning_rate": 1.1737924321004544e-05, + "loss": 0.5923, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1110 + }, + { + "epoch": 1.369913686806412, + "grad_norm": 2.577422739044973, + "learning_rate": 1.1724940083246042e-05, + "loss": 0.7151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1111 + }, + { + "epoch": 1.3711467324291, + "grad_norm": 2.4016601831106867, + "learning_rate": 1.171195284752907e-05, + "loss": 0.6719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1112 + }, + { + "epoch": 1.372379778051788, + "grad_norm": 2.2381623856871187, + "learning_rate": 1.1698962636425546e-05, + "loss": 0.7466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1113 + }, + { + "epoch": 1.373612823674476, + "grad_norm": 2.3077954380459498, + "learning_rate": 1.1685969472512544e-05, + "loss": 0.6609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1114 + }, + { + "epoch": 1.3748458692971641, + "grad_norm": 2.528220870127657, + "learning_rate": 1.1672973378372283e-05, + "loss": 0.7914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1115 + }, + { + "epoch": 1.376078914919852, + "grad_norm": 2.2640778971544946, + "learning_rate": 1.1659974376592065e-05, + "loss": 0.5914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1116 + }, + { + "epoch": 1.3773119605425401, + "grad_norm": 2.3357561361414994, + "learning_rate": 1.1646972489764254e-05, + "loss": 0.7194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1117 + }, + { + "epoch": 1.3785450061652282, + "grad_norm": 2.578566043676433, + "learning_rate": 1.163396774048622e-05, + "loss": 0.8069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1118 + }, + { + "epoch": 1.3797780517879161, + "grad_norm": 2.6394995201547853, + "learning_rate": 1.1620960151360317e-05, + "loss": 0.797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1119 + }, + { + "epoch": 1.3810110974106042, + "grad_norm": 2.2935912265432963, + "learning_rate": 1.1607949744993829e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1120 + }, + { + "epoch": 1.3822441430332923, + "grad_norm": 2.3848813510148394, + "learning_rate": 1.1594936543998933e-05, + "loss": 0.5816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1121 + }, + { + "epoch": 1.3834771886559802, + "grad_norm": 2.266951138614489, + "learning_rate": 1.1581920570992673e-05, + "loss": 0.8167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1122 + }, + { + "epoch": 1.3847102342786684, + "grad_norm": 2.0749563395142867, + "learning_rate": 1.1568901848596905e-05, + "loss": 0.6801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1123 + }, + { + "epoch": 1.3859432799013565, + "grad_norm": 2.259958476393895, + "learning_rate": 1.1555880399438257e-05, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1124 + }, + { + "epoch": 1.3871763255240444, + "grad_norm": 2.329878469976296, + "learning_rate": 1.1542856246148109e-05, + "loss": 0.7625, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1125 + }, + { + "epoch": 1.3884093711467325, + "grad_norm": 2.626476895279836, + "learning_rate": 1.1529829411362532e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1126 + }, + { + "epoch": 1.3896424167694206, + "grad_norm": 2.2360683032625297, + "learning_rate": 1.1516799917722262e-05, + "loss": 0.6588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1127 + }, + { + "epoch": 1.3908754623921085, + "grad_norm": 2.2606389128750664, + "learning_rate": 1.1503767787872652e-05, + "loss": 0.612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1128 + }, + { + "epoch": 1.3921085080147966, + "grad_norm": 2.3488413613095043, + "learning_rate": 1.149073304446364e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1129 + }, + { + "epoch": 1.3933415536374847, + "grad_norm": 2.187014482346159, + "learning_rate": 1.1477695710149706e-05, + "loss": 0.6897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1130 + }, + { + "epoch": 1.3945745992601726, + "grad_norm": 2.232874524639436, + "learning_rate": 1.1464655807589829e-05, + "loss": 0.7495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1131 + }, + { + "epoch": 1.3958076448828607, + "grad_norm": 2.2422283956051277, + "learning_rate": 1.1451613359447462e-05, + "loss": 0.7748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1132 + }, + { + "epoch": 1.3970406905055488, + "grad_norm": 2.2170691512303327, + "learning_rate": 1.1438568388390465e-05, + "loss": 0.7026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1133 + }, + { + "epoch": 1.3982737361282367, + "grad_norm": 2.3511388510197495, + "learning_rate": 1.1425520917091104e-05, + "loss": 0.7438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1134 + }, + { + "epoch": 1.3995067817509248, + "grad_norm": 2.302626316717242, + "learning_rate": 1.141247096822597e-05, + "loss": 0.6808, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1135 + }, + { + "epoch": 1.400739827373613, + "grad_norm": 2.358682446508914, + "learning_rate": 1.1399418564475977e-05, + "loss": 0.7462, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1136 + }, + { + "epoch": 1.4019728729963008, + "grad_norm": 2.4921870940266864, + "learning_rate": 1.138636372852629e-05, + "loss": 0.7565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1137 + }, + { + "epoch": 1.403205918618989, + "grad_norm": 2.881251766211438, + "learning_rate": 1.1373306483066315e-05, + "loss": 0.8531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1138 + }, + { + "epoch": 1.404438964241677, + "grad_norm": 2.374914568635526, + "learning_rate": 1.1360246850789638e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1139 + }, + { + "epoch": 1.405672009864365, + "grad_norm": 2.4892197803325478, + "learning_rate": 1.1347184854394e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1140 + }, + { + "epoch": 1.406905055487053, + "grad_norm": 2.335470361345138, + "learning_rate": 1.1334120516581239e-05, + "loss": 0.6239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1141 + }, + { + "epoch": 1.4081381011097411, + "grad_norm": 2.295703427576132, + "learning_rate": 1.1321053860057275e-05, + "loss": 0.7663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1142 + }, + { + "epoch": 1.409371146732429, + "grad_norm": 2.266519214119513, + "learning_rate": 1.1307984907532047e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1143 + }, + { + "epoch": 1.4106041923551171, + "grad_norm": 2.3854379731705917, + "learning_rate": 1.1294913681719497e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1144 + }, + { + "epoch": 1.4118372379778052, + "grad_norm": 2.3244362810232473, + "learning_rate": 1.1281840205337504e-05, + "loss": 0.694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1145 + }, + { + "epoch": 1.4130702836004931, + "grad_norm": 2.3269738125045727, + "learning_rate": 1.1268764501107872e-05, + "loss": 0.6867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1146 + }, + { + "epoch": 1.4143033292231812, + "grad_norm": 2.1038829157154533, + "learning_rate": 1.1255686591756265e-05, + "loss": 0.4939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1147 + }, + { + "epoch": 1.4155363748458694, + "grad_norm": 2.3057935606552356, + "learning_rate": 1.124260650001219e-05, + "loss": 0.697, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1148 + }, + { + "epoch": 1.4167694204685573, + "grad_norm": 2.4457435727408448, + "learning_rate": 1.1229524248608936e-05, + "loss": 0.7129, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1149 + }, + { + "epoch": 1.4180024660912454, + "grad_norm": 2.5516324512905793, + "learning_rate": 1.1216439860283556e-05, + "loss": 0.7116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1150 + }, + { + "epoch": 1.4192355117139335, + "grad_norm": 2.439191648697924, + "learning_rate": 1.120335335777681e-05, + "loss": 0.7269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1151 + }, + { + "epoch": 1.4204685573366214, + "grad_norm": 2.381736438404098, + "learning_rate": 1.1190264763833138e-05, + "loss": 0.7226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1152 + }, + { + "epoch": 1.4217016029593095, + "grad_norm": 2.326191969965805, + "learning_rate": 1.1177174101200609e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1153 + }, + { + "epoch": 1.4229346485819976, + "grad_norm": 2.4443478986341964, + "learning_rate": 1.1164081392630895e-05, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1154 + }, + { + "epoch": 1.4241676942046855, + "grad_norm": 2.4191171109083274, + "learning_rate": 1.115098666087921e-05, + "loss": 0.671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1155 + }, + { + "epoch": 1.4254007398273736, + "grad_norm": 2.331429993762429, + "learning_rate": 1.1137889928704303e-05, + "loss": 0.7875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1156 + }, + { + "epoch": 1.4266337854500617, + "grad_norm": 2.6182853389449905, + "learning_rate": 1.1124791218868383e-05, + "loss": 0.7465, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1157 + }, + { + "epoch": 1.4278668310727496, + "grad_norm": 2.3276838263801474, + "learning_rate": 1.111169055413711e-05, + "loss": 0.6731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1158 + }, + { + "epoch": 1.4290998766954377, + "grad_norm": 2.3555978375187783, + "learning_rate": 1.109858795727953e-05, + "loss": 0.7079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1159 + }, + { + "epoch": 1.4303329223181258, + "grad_norm": 2.243198455884724, + "learning_rate": 1.1085483451068056e-05, + "loss": 0.6767, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1160 + }, + { + "epoch": 1.4315659679408137, + "grad_norm": 2.426755129308782, + "learning_rate": 1.107237705827841e-05, + "loss": 0.6942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1161 + }, + { + "epoch": 1.4327990135635018, + "grad_norm": 2.310530088918712, + "learning_rate": 1.1059268801689605e-05, + "loss": 0.7414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1162 + }, + { + "epoch": 1.43403205918619, + "grad_norm": 2.3870562170497984, + "learning_rate": 1.104615870408388e-05, + "loss": 0.3904, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1163 + }, + { + "epoch": 1.4352651048088778, + "grad_norm": 2.185670591654591, + "learning_rate": 1.1033046788246684e-05, + "loss": 0.5667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1164 + }, + { + "epoch": 1.436498150431566, + "grad_norm": 2.2363567918319904, + "learning_rate": 1.1019933076966624e-05, + "loss": 0.6582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1165 + }, + { + "epoch": 1.437731196054254, + "grad_norm": 2.4989532650808917, + "learning_rate": 1.1006817593035423e-05, + "loss": 0.8567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1166 + }, + { + "epoch": 1.438964241676942, + "grad_norm": 2.3045828034357614, + "learning_rate": 1.0993700359247889e-05, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1167 + }, + { + "epoch": 1.44019728729963, + "grad_norm": 2.4408566344557263, + "learning_rate": 1.0980581398401875e-05, + "loss": 0.867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1168 + }, + { + "epoch": 1.4414303329223181, + "grad_norm": 2.433134793461258, + "learning_rate": 1.0967460733298224e-05, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1169 + }, + { + "epoch": 1.442663378545006, + "grad_norm": 2.084326240070432, + "learning_rate": 1.0954338386740746e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1170 + }, + { + "epoch": 1.4438964241676941, + "grad_norm": 2.3889216941699165, + "learning_rate": 1.0941214381536185e-05, + "loss": 0.7649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1171 + }, + { + "epoch": 1.4451294697903823, + "grad_norm": 2.3052157462538787, + "learning_rate": 1.092808874049415e-05, + "loss": 0.6548, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1172 + }, + { + "epoch": 1.4463625154130701, + "grad_norm": 2.412258190430439, + "learning_rate": 1.0914961486427103e-05, + "loss": 0.6655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1173 + }, + { + "epoch": 1.4475955610357583, + "grad_norm": 2.3820079102806755, + "learning_rate": 1.0901832642150308e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1174 + }, + { + "epoch": 1.4488286066584464, + "grad_norm": 2.826438039521343, + "learning_rate": 1.0888702230481788e-05, + "loss": 0.6666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1175 + }, + { + "epoch": 1.4500616522811345, + "grad_norm": 2.3005434823783166, + "learning_rate": 1.0875570274242305e-05, + "loss": 0.7333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1176 + }, + { + "epoch": 1.4512946979038224, + "grad_norm": 2.580920616275187, + "learning_rate": 1.0862436796255285e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1177 + }, + { + "epoch": 1.4525277435265105, + "grad_norm": 2.39619982968134, + "learning_rate": 1.0849301819346809e-05, + "loss": 0.6086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1178 + }, + { + "epoch": 1.4537607891491986, + "grad_norm": 2.4469170067316246, + "learning_rate": 1.0836165366345572e-05, + "loss": 0.7415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1179 + }, + { + "epoch": 1.4549938347718865, + "grad_norm": 2.4895765738835327, + "learning_rate": 1.0823027460082815e-05, + "loss": 0.8269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1180 + }, + { + "epoch": 1.4562268803945746, + "grad_norm": 2.1191361440164935, + "learning_rate": 1.0809888123392321e-05, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1181 + }, + { + "epoch": 1.4574599260172627, + "grad_norm": 2.0580916232206263, + "learning_rate": 1.0796747379110351e-05, + "loss": 0.5483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1182 + }, + { + "epoch": 1.4586929716399506, + "grad_norm": 2.358708614850794, + "learning_rate": 1.0783605250075615e-05, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1183 + }, + { + "epoch": 1.4599260172626387, + "grad_norm": 2.3706216835188068, + "learning_rate": 1.0770461759129225e-05, + "loss": 0.6925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1184 + }, + { + "epoch": 1.4611590628853268, + "grad_norm": 2.075953282437584, + "learning_rate": 1.0757316929114673e-05, + "loss": 0.7095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1185 + }, + { + "epoch": 1.462392108508015, + "grad_norm": 2.3425821034074175, + "learning_rate": 1.0744170782877765e-05, + "loss": 0.5685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1186 + }, + { + "epoch": 1.4636251541307028, + "grad_norm": 2.364864320197578, + "learning_rate": 1.07310233432666e-05, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1187 + }, + { + "epoch": 1.464858199753391, + "grad_norm": 2.269198583092931, + "learning_rate": 1.071787463313152e-05, + "loss": 0.5853, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1188 + }, + { + "epoch": 1.466091245376079, + "grad_norm": 2.3353973695396504, + "learning_rate": 1.0704724675325086e-05, + "loss": 0.6851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1189 + }, + { + "epoch": 1.467324290998767, + "grad_norm": 2.297514940312218, + "learning_rate": 1.0691573492702014e-05, + "loss": 0.6437, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1190 + }, + { + "epoch": 1.468557336621455, + "grad_norm": 2.347920635849606, + "learning_rate": 1.0678421108119161e-05, + "loss": 0.5607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1191 + }, + { + "epoch": 1.4697903822441432, + "grad_norm": 2.4998335942015792, + "learning_rate": 1.0665267544435464e-05, + "loss": 0.715, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1192 + }, + { + "epoch": 1.471023427866831, + "grad_norm": 2.6799025570634494, + "learning_rate": 1.065211282451192e-05, + "loss": 0.855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1193 + }, + { + "epoch": 1.4722564734895192, + "grad_norm": 2.3846645442657497, + "learning_rate": 1.0638956971211515e-05, + "loss": 0.8093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1194 + }, + { + "epoch": 1.4734895191122073, + "grad_norm": 2.289987275457904, + "learning_rate": 1.0625800007399232e-05, + "loss": 0.615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1195 + }, + { + "epoch": 1.4747225647348952, + "grad_norm": 2.398231591907997, + "learning_rate": 1.061264195594196e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1196 + }, + { + "epoch": 1.4759556103575833, + "grad_norm": 2.598957432876697, + "learning_rate": 1.0599482839708492e-05, + "loss": 0.5761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1197 + }, + { + "epoch": 1.4771886559802714, + "grad_norm": 2.3008244142116308, + "learning_rate": 1.0586322681569468e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1198 + }, + { + "epoch": 1.4784217016029593, + "grad_norm": 2.62152754882142, + "learning_rate": 1.0573161504397345e-05, + "loss": 0.9059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1199 + }, + { + "epoch": 1.4796547472256474, + "grad_norm": 2.7027129465803323, + "learning_rate": 1.0559999331066334e-05, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1200 + }, + { + "epoch": 1.4808877928483355, + "grad_norm": 2.441005731777059, + "learning_rate": 1.0546836184452397e-05, + "loss": 0.8212, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1201 + }, + { + "epoch": 1.4821208384710234, + "grad_norm": 2.4156909019245516, + "learning_rate": 1.0533672087433176e-05, + "loss": 0.6755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1202 + }, + { + "epoch": 1.4833538840937115, + "grad_norm": 2.373408895848631, + "learning_rate": 1.0520507062887966e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1203 + }, + { + "epoch": 1.4845869297163996, + "grad_norm": 2.1607701722680814, + "learning_rate": 1.050734113369768e-05, + "loss": 0.6858, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1204 + }, + { + "epoch": 1.4858199753390875, + "grad_norm": 2.211217022679158, + "learning_rate": 1.0494174322744797e-05, + "loss": 0.7383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1205 + }, + { + "epoch": 1.4870530209617756, + "grad_norm": 2.180028246293776, + "learning_rate": 1.0481006652913328e-05, + "loss": 0.5521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1206 + }, + { + "epoch": 1.4882860665844637, + "grad_norm": 2.393847834562169, + "learning_rate": 1.0467838147088785e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1207 + }, + { + "epoch": 1.4895191122071516, + "grad_norm": 2.20100213605133, + "learning_rate": 1.0454668828158122e-05, + "loss": 0.7316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1208 + }, + { + "epoch": 1.4907521578298397, + "grad_norm": 2.43405475601361, + "learning_rate": 1.0441498719009715e-05, + "loss": 0.7545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1209 + }, + { + "epoch": 1.4919852034525278, + "grad_norm": 2.3073004776119004, + "learning_rate": 1.0428327842533309e-05, + "loss": 0.5749, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1210 + }, + { + "epoch": 1.4932182490752157, + "grad_norm": 2.5013121128147535, + "learning_rate": 1.041515622161998e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1211 + }, + { + "epoch": 1.4944512946979038, + "grad_norm": 2.7168877702234537, + "learning_rate": 1.0401983879162108e-05, + "loss": 0.7342, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1212 + }, + { + "epoch": 1.495684340320592, + "grad_norm": 2.35281737411902, + "learning_rate": 1.0388810838053316e-05, + "loss": 0.6203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1213 + }, + { + "epoch": 1.4969173859432798, + "grad_norm": 2.400077140272344, + "learning_rate": 1.0375637121188447e-05, + "loss": 0.6201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1214 + }, + { + "epoch": 1.498150431565968, + "grad_norm": 2.5745039938322796, + "learning_rate": 1.0362462751463518e-05, + "loss": 0.6622, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1215 + }, + { + "epoch": 1.499383477188656, + "grad_norm": 2.4221988301394966, + "learning_rate": 1.0349287751775677e-05, + "loss": 0.7025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1216 + }, + { + "epoch": 1.500616522811344, + "grad_norm": 2.386783019524135, + "learning_rate": 1.033611214502317e-05, + "loss": 0.5671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1217 + }, + { + "epoch": 1.501849568434032, + "grad_norm": 2.539552131989461, + "learning_rate": 1.0322935954105303e-05, + "loss": 0.6469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1218 + } + ], + "logging_steps": 1, + "max_steps": 2433, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 406, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2000347799224320.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1218/training_args.bin b/checkpoint-1218/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7b675d785639ed4b99af767b6c87a1227c13ec1b --- /dev/null +++ b/checkpoint-1218/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e80ea310ffc6c44543f32aa82ecc00b1bf9d3e329881647443a895f041e17955 +size 9208 diff --git a/checkpoint-1624/chat_template.jinja b/checkpoint-1624/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..870322b8f52314199bfd74bf68b8b8f1126395ae --- /dev/null +++ b/checkpoint-1624/chat_template.jinja @@ -0,0 +1,5 @@ +{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} diff --git a/checkpoint-1624/config.json b/checkpoint-1624/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4c0f53b6c66a0b8a2a4eb621c949412e2a7f353a --- /dev/null +++ b/checkpoint-1624/config.json @@ -0,0 +1,35 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.2", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/checkpoint-1624/generation_config.json b/checkpoint-1624/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b513e54e3195b917260c9a8a04c9f3683f19de35 --- /dev/null +++ b/checkpoint-1624/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.55.2" +} diff --git a/checkpoint-1624/model-00004-of-00004.safetensors b/checkpoint-1624/model-00004-of-00004.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..407f46c2976b4956345913b950af709a4da35522 --- /dev/null +++ b/checkpoint-1624/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ead28dad6d8868227174436ecfab3a285011f798e03cc48211a8993734d39cf +size 1168138808 diff --git a/checkpoint-1624/model.safetensors.index.json b/checkpoint-1624/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..58a7ef40d52ab3340cf131cf33f8923f5144aaf9 --- /dev/null +++ b/checkpoint-1624/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-1624/special_tokens_map.json b/checkpoint-1624/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3c1d04911c269b925af977a3151c9704e990e4d0 --- /dev/null +++ b/checkpoint-1624/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1624/tokenizer.json b/checkpoint-1624/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-1624/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-1624/tokenizer_config.json b/checkpoint-1624/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..c68051fe3c4d23234a59316bc52d21f6e3a4182c --- /dev/null +++ b/checkpoint-1624/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|finetune_right_pad_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1624/trainer_state.json b/checkpoint-1624/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..1a6d5c9c591cb1b4e21b0bc9f1c23565ecbbeb39 --- /dev/null +++ b/checkpoint-1624/trainer_state.json @@ -0,0 +1,16274 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.0024660912453762, + "eval_steps": 500, + "global_step": 1624, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0012330456226880395, + "grad_norm": 13.955555906863093, + "learning_rate": 0.0, + "loss": 1.9039, + "memory/device_mem_reserved(gib)": 19.83, + "memory/max_mem_active(gib)": 16.16, + "memory/max_mem_allocated(gib)": 16.16, + "step": 1 + }, + { + "epoch": 0.002466091245376079, + "grad_norm": 14.68412675273428, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.9395, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 2 + }, + { + "epoch": 0.0036991368680641184, + "grad_norm": 15.404314959825989, + "learning_rate": 8.000000000000001e-07, + "loss": 1.9524, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 3 + }, + { + "epoch": 0.004932182490752158, + "grad_norm": 14.099502171385124, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.8527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 4 + }, + { + "epoch": 0.006165228113440197, + "grad_norm": 12.213499070556583, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.8853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 5 + }, + { + "epoch": 0.007398273736128237, + "grad_norm": 7.4228290511653725, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.7686, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 6 + }, + { + "epoch": 0.008631319358816275, + "grad_norm": 6.989334002600767, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.7976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 7 + }, + { + "epoch": 0.009864364981504316, + "grad_norm": 7.330479156896915, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6029, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 8 + }, + { + "epoch": 0.011097410604192354, + "grad_norm": 7.091571204851302, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.6097, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 9 + }, + { + "epoch": 0.012330456226880395, + "grad_norm": 5.900225931668668, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.6512, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 10 + }, + { + "epoch": 0.013563501849568433, + "grad_norm": 12.562784448457464, + "learning_rate": 4.000000000000001e-06, + "loss": 1.4824, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 11 + }, + { + "epoch": 0.014796547472256474, + "grad_norm": 6.056074456731366, + "learning_rate": 4.4e-06, + "loss": 1.5581, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 12 + }, + { + "epoch": 0.016029593094944512, + "grad_norm": 5.051119502724206, + "learning_rate": 4.800000000000001e-06, + "loss": 1.5275, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 13 + }, + { + "epoch": 0.01726263871763255, + "grad_norm": 4.022180485081632, + "learning_rate": 5.2e-06, + "loss": 1.5075, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 14 + }, + { + "epoch": 0.018495684340320593, + "grad_norm": 4.5104551135107815, + "learning_rate": 5.600000000000001e-06, + "loss": 1.4638, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 15 + }, + { + "epoch": 0.01972872996300863, + "grad_norm": 4.3429897237996276, + "learning_rate": 6e-06, + "loss": 1.3996, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 16 + }, + { + "epoch": 0.02096177558569667, + "grad_norm": 3.9791138623880764, + "learning_rate": 6.4000000000000006e-06, + "loss": 1.4059, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 17 + }, + { + "epoch": 0.02219482120838471, + "grad_norm": 4.194505204288471, + "learning_rate": 6.800000000000001e-06, + "loss": 1.3747, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 18 + }, + { + "epoch": 0.02342786683107275, + "grad_norm": 3.6309957760327713, + "learning_rate": 7.2000000000000005e-06, + "loss": 1.3565, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 19 + }, + { + "epoch": 0.02466091245376079, + "grad_norm": 3.431355025722372, + "learning_rate": 7.600000000000001e-06, + "loss": 1.4019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 20 + }, + { + "epoch": 0.025893958076448828, + "grad_norm": 3.258474422309909, + "learning_rate": 8.000000000000001e-06, + "loss": 1.364, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 21 + }, + { + "epoch": 0.027127003699136867, + "grad_norm": 3.7750095693616736, + "learning_rate": 8.400000000000001e-06, + "loss": 1.3798, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 22 + }, + { + "epoch": 0.02836004932182491, + "grad_norm": 3.1192450402705125, + "learning_rate": 8.8e-06, + "loss": 1.2677, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 23 + }, + { + "epoch": 0.029593094944512947, + "grad_norm": 3.34669710403596, + "learning_rate": 9.200000000000002e-06, + "loss": 1.3465, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 24 + }, + { + "epoch": 0.030826140567200986, + "grad_norm": 3.0969841766129638, + "learning_rate": 9.600000000000001e-06, + "loss": 1.3662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 25 + }, + { + "epoch": 0.032059186189889025, + "grad_norm": 2.9521843506112466, + "learning_rate": 1e-05, + "loss": 1.3295, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 26 + }, + { + "epoch": 0.03329223181257707, + "grad_norm": 2.9117498945141085, + "learning_rate": 1.04e-05, + "loss": 1.3066, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 27 + }, + { + "epoch": 0.0345252774352651, + "grad_norm": 2.9215831263409013, + "learning_rate": 1.0800000000000002e-05, + "loss": 1.3221, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 28 + }, + { + "epoch": 0.035758323057953144, + "grad_norm": 2.97321184472987, + "learning_rate": 1.1200000000000001e-05, + "loss": 1.3247, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 29 + }, + { + "epoch": 0.036991368680641186, + "grad_norm": 3.293476484963883, + "learning_rate": 1.16e-05, + "loss": 1.2582, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 30 + }, + { + "epoch": 0.03822441430332922, + "grad_norm": 2.8642108219246745, + "learning_rate": 1.2e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 31 + }, + { + "epoch": 0.03945745992601726, + "grad_norm": 2.8916726488438824, + "learning_rate": 1.2400000000000002e-05, + "loss": 1.2628, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 32 + }, + { + "epoch": 0.040690505548705305, + "grad_norm": 3.17999982306396, + "learning_rate": 1.2800000000000001e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 33 + }, + { + "epoch": 0.04192355117139334, + "grad_norm": 2.6965579635696453, + "learning_rate": 1.3200000000000002e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 34 + }, + { + "epoch": 0.04315659679408138, + "grad_norm": 2.9489592197594896, + "learning_rate": 1.3600000000000002e-05, + "loss": 1.3113, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 35 + }, + { + "epoch": 0.04438964241676942, + "grad_norm": 2.8898479257030734, + "learning_rate": 1.4e-05, + "loss": 1.3387, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 36 + }, + { + "epoch": 0.04562268803945746, + "grad_norm": 2.6370539695770425, + "learning_rate": 1.4400000000000001e-05, + "loss": 1.3314, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 37 + }, + { + "epoch": 0.0468557336621455, + "grad_norm": 2.7915119494225813, + "learning_rate": 1.48e-05, + "loss": 1.2603, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 38 + }, + { + "epoch": 0.04808877928483354, + "grad_norm": 2.602735589719503, + "learning_rate": 1.5200000000000002e-05, + "loss": 1.3492, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 39 + }, + { + "epoch": 0.04932182490752158, + "grad_norm": 2.9123309438390113, + "learning_rate": 1.5600000000000003e-05, + "loss": 1.2291, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 40 + }, + { + "epoch": 0.05055487053020962, + "grad_norm": 2.577807086473754, + "learning_rate": 1.6000000000000003e-05, + "loss": 1.3098, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 41 + }, + { + "epoch": 0.051787916152897656, + "grad_norm": 2.548383978805174, + "learning_rate": 1.64e-05, + "loss": 1.2893, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 42 + }, + { + "epoch": 0.0530209617755857, + "grad_norm": 2.6372216084932814, + "learning_rate": 1.6800000000000002e-05, + "loss": 1.2701, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 43 + }, + { + "epoch": 0.05425400739827373, + "grad_norm": 2.60286437866227, + "learning_rate": 1.72e-05, + "loss": 1.3579, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 44 + }, + { + "epoch": 0.055487053020961775, + "grad_norm": 2.536071820500707, + "learning_rate": 1.76e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 45 + }, + { + "epoch": 0.05672009864364982, + "grad_norm": 2.8143481627767963, + "learning_rate": 1.8e-05, + "loss": 1.1729, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 46 + }, + { + "epoch": 0.05795314426633785, + "grad_norm": 2.8600643284872596, + "learning_rate": 1.8400000000000003e-05, + "loss": 1.3246, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 47 + }, + { + "epoch": 0.059186189889025895, + "grad_norm": 2.618463019216554, + "learning_rate": 1.88e-05, + "loss": 1.2878, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 48 + }, + { + "epoch": 0.06041923551171394, + "grad_norm": 2.757496450593693, + "learning_rate": 1.9200000000000003e-05, + "loss": 1.3454, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 49 + }, + { + "epoch": 0.06165228113440197, + "grad_norm": 2.5356573576135126, + "learning_rate": 1.9600000000000002e-05, + "loss": 1.2719, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 50 + }, + { + "epoch": 0.06288532675709001, + "grad_norm": 2.532419474064049, + "learning_rate": 2e-05, + "loss": 1.332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 51 + }, + { + "epoch": 0.06411837237977805, + "grad_norm": 2.711902763412396, + "learning_rate": 1.999999130996347e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 52 + }, + { + "epoch": 0.06535141800246609, + "grad_norm": 2.654767047589697, + "learning_rate": 1.9999965239868983e-05, + "loss": 1.3092, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 53 + }, + { + "epoch": 0.06658446362515413, + "grad_norm": 3.2372229064932454, + "learning_rate": 1.9999921789761843e-05, + "loss": 1.3428, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 54 + }, + { + "epoch": 0.06781750924784218, + "grad_norm": 2.723103510772695, + "learning_rate": 1.999986095971757e-05, + "loss": 1.3918, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 55 + }, + { + "epoch": 0.0690505548705302, + "grad_norm": 2.849098835881172, + "learning_rate": 1.9999782749841895e-05, + "loss": 1.3304, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 56 + }, + { + "epoch": 0.07028360049321825, + "grad_norm": 2.4115082044348557, + "learning_rate": 1.9999687160270735e-05, + "loss": 1.2818, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 57 + }, + { + "epoch": 0.07151664611590629, + "grad_norm": 2.7720437269939495, + "learning_rate": 1.9999574191170233e-05, + "loss": 1.2799, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 58 + }, + { + "epoch": 0.07274969173859433, + "grad_norm": 2.4891158758553042, + "learning_rate": 1.9999443842736723e-05, + "loss": 1.2408, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 59 + }, + { + "epoch": 0.07398273736128237, + "grad_norm": 2.8066935220153737, + "learning_rate": 1.999929611519676e-05, + "loss": 1.3157, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 60 + }, + { + "epoch": 0.0752157829839704, + "grad_norm": 2.6715709033170705, + "learning_rate": 1.9999131008807094e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 61 + }, + { + "epoch": 0.07644882860665844, + "grad_norm": 2.586596297320694, + "learning_rate": 1.9998948523854676e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 62 + }, + { + "epoch": 0.07768187422934648, + "grad_norm": 2.646887045495792, + "learning_rate": 1.999874866065667e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 63 + }, + { + "epoch": 0.07891491985203453, + "grad_norm": 2.419730330481743, + "learning_rate": 1.999853141956044e-05, + "loss": 1.2829, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 64 + }, + { + "epoch": 0.08014796547472257, + "grad_norm": 2.5201815842521245, + "learning_rate": 1.999829680094355e-05, + "loss": 1.3243, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 65 + }, + { + "epoch": 0.08138101109741061, + "grad_norm": 2.503429794099987, + "learning_rate": 1.9998044805213773e-05, + "loss": 1.2681, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 66 + }, + { + "epoch": 0.08261405672009864, + "grad_norm": 2.348439339795846, + "learning_rate": 1.9997775432809074e-05, + "loss": 1.3359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 67 + }, + { + "epoch": 0.08384710234278668, + "grad_norm": 2.3828095918277357, + "learning_rate": 1.999748868419763e-05, + "loss": 1.3526, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 68 + }, + { + "epoch": 0.08508014796547472, + "grad_norm": 2.5076969489265273, + "learning_rate": 1.9997184559877806e-05, + "loss": 1.3531, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 69 + }, + { + "epoch": 0.08631319358816276, + "grad_norm": 2.388938276200468, + "learning_rate": 1.9996863060378176e-05, + "loss": 1.334, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 70 + }, + { + "epoch": 0.08754623921085081, + "grad_norm": 2.283154945083281, + "learning_rate": 1.999652418625751e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 71 + }, + { + "epoch": 0.08877928483353884, + "grad_norm": 2.2613574311048734, + "learning_rate": 1.9996167938104767e-05, + "loss": 1.2858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 72 + }, + { + "epoch": 0.09001233045622688, + "grad_norm": 2.3931162231369076, + "learning_rate": 1.999579431653912e-05, + "loss": 1.3089, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 73 + }, + { + "epoch": 0.09124537607891492, + "grad_norm": 2.5073647819033194, + "learning_rate": 1.9995403322209913e-05, + "loss": 1.382, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 74 + }, + { + "epoch": 0.09247842170160296, + "grad_norm": 2.5388011932611536, + "learning_rate": 1.999499495579671e-05, + "loss": 1.402, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 75 + }, + { + "epoch": 0.093711467324291, + "grad_norm": 2.3884809904992252, + "learning_rate": 1.9994569218009247e-05, + "loss": 1.3079, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 76 + }, + { + "epoch": 0.09494451294697903, + "grad_norm": 2.3454483578272094, + "learning_rate": 1.999412610958746e-05, + "loss": 1.2653, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 77 + }, + { + "epoch": 0.09617755856966707, + "grad_norm": 2.186902835408518, + "learning_rate": 1.9993665631301474e-05, + "loss": 1.2069, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 78 + }, + { + "epoch": 0.09741060419235512, + "grad_norm": 2.3845058131797368, + "learning_rate": 1.9993187783951603e-05, + "loss": 1.3541, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 79 + }, + { + "epoch": 0.09864364981504316, + "grad_norm": 2.256399889979939, + "learning_rate": 1.9992692568368354e-05, + "loss": 1.2119, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 80 + }, + { + "epoch": 0.0998766954377312, + "grad_norm": 2.3361092398369903, + "learning_rate": 1.9992179985412416e-05, + "loss": 1.2585, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 81 + }, + { + "epoch": 0.10110974106041924, + "grad_norm": 2.289061082197673, + "learning_rate": 1.999165003597465e-05, + "loss": 1.3459, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 82 + }, + { + "epoch": 0.10234278668310727, + "grad_norm": 2.228187448538031, + "learning_rate": 1.999110272097613e-05, + "loss": 1.3739, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 83 + }, + { + "epoch": 0.10357583230579531, + "grad_norm": 2.157361593613978, + "learning_rate": 1.9990538041368078e-05, + "loss": 1.2575, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 84 + }, + { + "epoch": 0.10480887792848335, + "grad_norm": 2.294809937254642, + "learning_rate": 1.998995599813192e-05, + "loss": 1.2834, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 85 + }, + { + "epoch": 0.1060419235511714, + "grad_norm": 2.1831743941808925, + "learning_rate": 1.998935659227925e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 86 + }, + { + "epoch": 0.10727496917385944, + "grad_norm": 2.247228611971362, + "learning_rate": 1.998873982485183e-05, + "loss": 1.2654, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 87 + }, + { + "epoch": 0.10850801479654747, + "grad_norm": 2.2844560193856416, + "learning_rate": 1.9988105696921622e-05, + "loss": 1.2927, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 88 + }, + { + "epoch": 0.10974106041923551, + "grad_norm": 2.3780157649218157, + "learning_rate": 1.9987454209590736e-05, + "loss": 1.2621, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 89 + }, + { + "epoch": 0.11097410604192355, + "grad_norm": 2.3017268441414003, + "learning_rate": 1.9986785363991466e-05, + "loss": 1.2676, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 90 + }, + { + "epoch": 0.11220715166461159, + "grad_norm": 2.1906552279173503, + "learning_rate": 1.9986099161286266e-05, + "loss": 1.2281, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 91 + }, + { + "epoch": 0.11344019728729964, + "grad_norm": 2.4009419002081276, + "learning_rate": 1.9985395602667764e-05, + "loss": 1.2267, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 92 + }, + { + "epoch": 0.11467324290998766, + "grad_norm": 2.190217063330867, + "learning_rate": 1.9984674689358744e-05, + "loss": 1.2778, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 93 + }, + { + "epoch": 0.1159062885326757, + "grad_norm": 2.305624793209019, + "learning_rate": 1.998393642261217e-05, + "loss": 1.2821, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 94 + }, + { + "epoch": 0.11713933415536375, + "grad_norm": 2.162537568697834, + "learning_rate": 1.9983180803711146e-05, + "loss": 1.3775, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 95 + }, + { + "epoch": 0.11837237977805179, + "grad_norm": 2.2632096292316706, + "learning_rate": 1.9982407833968947e-05, + "loss": 1.3358, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 96 + }, + { + "epoch": 0.11960542540073983, + "grad_norm": 2.268508818243908, + "learning_rate": 1.9981617514728998e-05, + "loss": 1.2368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 97 + }, + { + "epoch": 0.12083847102342787, + "grad_norm": 2.2721363895167377, + "learning_rate": 1.9980809847364884e-05, + "loss": 1.2774, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 98 + }, + { + "epoch": 0.1220715166461159, + "grad_norm": 2.1661250921439974, + "learning_rate": 1.9979984833280332e-05, + "loss": 1.359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 99 + }, + { + "epoch": 0.12330456226880394, + "grad_norm": 2.2521676172131873, + "learning_rate": 1.997914247390922e-05, + "loss": 1.216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 100 + }, + { + "epoch": 0.12453760789149199, + "grad_norm": 2.3907655242108095, + "learning_rate": 1.9978282770715586e-05, + "loss": 1.2468, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 101 + }, + { + "epoch": 0.12577065351418001, + "grad_norm": 2.2007275768445074, + "learning_rate": 1.9977405725193587e-05, + "loss": 1.2415, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 102 + }, + { + "epoch": 0.12700369913686807, + "grad_norm": 2.3304230330598346, + "learning_rate": 1.9976511338867542e-05, + "loss": 1.2438, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 103 + }, + { + "epoch": 0.1282367447595561, + "grad_norm": 2.346171964705753, + "learning_rate": 1.99755996132919e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 104 + }, + { + "epoch": 0.12946979038224415, + "grad_norm": 2.286623534979458, + "learning_rate": 1.9974670550051248e-05, + "loss": 1.2897, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 105 + }, + { + "epoch": 0.13070283600493218, + "grad_norm": 2.245774575899364, + "learning_rate": 1.99737241507603e-05, + "loss": 1.2923, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 106 + }, + { + "epoch": 0.1319358816276202, + "grad_norm": 2.2352297973588953, + "learning_rate": 1.9972760417063912e-05, + "loss": 1.1434, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 107 + }, + { + "epoch": 0.13316892725030827, + "grad_norm": 2.169802916370051, + "learning_rate": 1.9971779350637054e-05, + "loss": 1.2353, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 108 + }, + { + "epoch": 0.1344019728729963, + "grad_norm": 2.333533512175979, + "learning_rate": 1.9970780953184828e-05, + "loss": 1.2641, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 109 + }, + { + "epoch": 0.13563501849568435, + "grad_norm": 2.2985803581198674, + "learning_rate": 1.996976522644246e-05, + "loss": 1.3036, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 110 + }, + { + "epoch": 0.13686806411837238, + "grad_norm": 2.0992498120010272, + "learning_rate": 1.996873217217528e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 111 + }, + { + "epoch": 0.1381011097410604, + "grad_norm": 2.396895991067544, + "learning_rate": 1.9967681792178758e-05, + "loss": 1.2533, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 112 + }, + { + "epoch": 0.13933415536374846, + "grad_norm": 2.0738932505082768, + "learning_rate": 1.9966614088278456e-05, + "loss": 1.3332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 113 + }, + { + "epoch": 0.1405672009864365, + "grad_norm": 2.1049156004628395, + "learning_rate": 1.9965529062330048e-05, + "loss": 1.2793, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 114 + }, + { + "epoch": 0.14180024660912455, + "grad_norm": 2.2137820282577847, + "learning_rate": 1.9964426716219316e-05, + "loss": 1.1932, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 115 + }, + { + "epoch": 0.14303329223181258, + "grad_norm": 2.1545610182661976, + "learning_rate": 1.9963307051862152e-05, + "loss": 1.2662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 116 + }, + { + "epoch": 0.1442663378545006, + "grad_norm": 2.151286882718536, + "learning_rate": 1.9962170071204542e-05, + "loss": 1.205, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 117 + }, + { + "epoch": 0.14549938347718866, + "grad_norm": 2.026608825482131, + "learning_rate": 1.996101577622256e-05, + "loss": 1.195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 118 + }, + { + "epoch": 0.1467324290998767, + "grad_norm": 2.156975981742732, + "learning_rate": 1.9959844168922377e-05, + "loss": 1.3284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 119 + }, + { + "epoch": 0.14796547472256474, + "grad_norm": 2.2311865782919025, + "learning_rate": 1.9958655251340265e-05, + "loss": 1.2183, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 120 + }, + { + "epoch": 0.14919852034525277, + "grad_norm": 2.2940740150536376, + "learning_rate": 1.9957449025542567e-05, + "loss": 1.2625, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 121 + }, + { + "epoch": 0.1504315659679408, + "grad_norm": 2.214218654766027, + "learning_rate": 1.9956225493625708e-05, + "loss": 1.2177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 122 + }, + { + "epoch": 0.15166461159062886, + "grad_norm": 2.1690434056776775, + "learning_rate": 1.9954984657716202e-05, + "loss": 1.2049, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 123 + }, + { + "epoch": 0.15289765721331688, + "grad_norm": 2.1450511034984046, + "learning_rate": 1.9953726519970625e-05, + "loss": 1.232, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 124 + }, + { + "epoch": 0.15413070283600494, + "grad_norm": 2.0475086343318543, + "learning_rate": 1.995245108257563e-05, + "loss": 1.2665, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 125 + }, + { + "epoch": 0.15536374845869297, + "grad_norm": 2.1232594967339318, + "learning_rate": 1.9951158347747945e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 126 + }, + { + "epoch": 0.15659679408138102, + "grad_norm": 2.1915516478247308, + "learning_rate": 1.9949848317734345e-05, + "loss": 1.177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 127 + }, + { + "epoch": 0.15782983970406905, + "grad_norm": 2.2848682922250205, + "learning_rate": 1.9948520994811674e-05, + "loss": 1.2406, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 128 + }, + { + "epoch": 0.15906288532675708, + "grad_norm": 2.2223107161107505, + "learning_rate": 1.9947176381286825e-05, + "loss": 1.327, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 129 + }, + { + "epoch": 0.16029593094944514, + "grad_norm": 2.1247307489434837, + "learning_rate": 1.994581447949675e-05, + "loss": 1.1634, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 130 + }, + { + "epoch": 0.16152897657213316, + "grad_norm": 1.926245615401859, + "learning_rate": 1.9944435291808445e-05, + "loss": 1.1655, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 131 + }, + { + "epoch": 0.16276202219482122, + "grad_norm": 2.2531819769934858, + "learning_rate": 1.9943038820618946e-05, + "loss": 1.2216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 132 + }, + { + "epoch": 0.16399506781750925, + "grad_norm": 2.2183949614858305, + "learning_rate": 1.994162506835533e-05, + "loss": 1.2202, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 133 + }, + { + "epoch": 0.16522811344019728, + "grad_norm": 2.164554433266706, + "learning_rate": 1.994019403747471e-05, + "loss": 1.2245, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 134 + }, + { + "epoch": 0.16646115906288533, + "grad_norm": 2.17653642461108, + "learning_rate": 1.9938745730464228e-05, + "loss": 1.1899, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 135 + }, + { + "epoch": 0.16769420468557336, + "grad_norm": 1.9936998341266101, + "learning_rate": 1.9937280149841055e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 136 + }, + { + "epoch": 0.16892725030826142, + "grad_norm": 2.27069520590907, + "learning_rate": 1.9935797298152374e-05, + "loss": 1.3023, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 137 + }, + { + "epoch": 0.17016029593094945, + "grad_norm": 2.244074718609361, + "learning_rate": 1.9934297177975397e-05, + "loss": 1.3292, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 138 + }, + { + "epoch": 0.17139334155363747, + "grad_norm": 2.2121305803685094, + "learning_rate": 1.9932779791917343e-05, + "loss": 1.3028, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 139 + }, + { + "epoch": 0.17262638717632553, + "grad_norm": 2.7262731821688697, + "learning_rate": 1.9931245142615442e-05, + "loss": 1.2861, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 140 + }, + { + "epoch": 0.17385943279901356, + "grad_norm": 2.2051993935523497, + "learning_rate": 1.992969323273692e-05, + "loss": 1.3134, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 141 + }, + { + "epoch": 0.17509247842170161, + "grad_norm": 2.088688387254833, + "learning_rate": 1.992812406497901e-05, + "loss": 1.262, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 142 + }, + { + "epoch": 0.17632552404438964, + "grad_norm": 2.19652156002121, + "learning_rate": 1.9926537642068942e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 143 + }, + { + "epoch": 0.17755856966707767, + "grad_norm": 1.887489027726817, + "learning_rate": 1.9924933966763924e-05, + "loss": 1.2847, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 144 + }, + { + "epoch": 0.17879161528976573, + "grad_norm": 2.088506357330406, + "learning_rate": 1.992331304185116e-05, + "loss": 1.2218, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 145 + }, + { + "epoch": 0.18002466091245375, + "grad_norm": 2.102792679186299, + "learning_rate": 1.9921674870147827e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 146 + }, + { + "epoch": 0.1812577065351418, + "grad_norm": 2.0847095844816015, + "learning_rate": 1.992001945450108e-05, + "loss": 1.2986, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 147 + }, + { + "epoch": 0.18249075215782984, + "grad_norm": 2.344867998715345, + "learning_rate": 1.9918346797788045e-05, + "loss": 1.1347, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 148 + }, + { + "epoch": 0.18372379778051787, + "grad_norm": 2.1013292606380602, + "learning_rate": 1.991665690291581e-05, + "loss": 1.1319, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 149 + }, + { + "epoch": 0.18495684340320592, + "grad_norm": 2.2475258887436538, + "learning_rate": 1.9914949772821418e-05, + "loss": 1.1648, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 150 + }, + { + "epoch": 0.18618988902589395, + "grad_norm": 2.223126056771945, + "learning_rate": 1.9913225410471888e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 151 + }, + { + "epoch": 0.187422934648582, + "grad_norm": 2.102849036750559, + "learning_rate": 1.9911483818864163e-05, + "loss": 1.2062, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 152 + }, + { + "epoch": 0.18865598027127004, + "grad_norm": 2.2107888348018805, + "learning_rate": 1.9909725001025146e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 153 + }, + { + "epoch": 0.18988902589395806, + "grad_norm": 2.1730720194510456, + "learning_rate": 1.9907948960011677e-05, + "loss": 1.2133, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 154 + }, + { + "epoch": 0.19112207151664612, + "grad_norm": 2.0874944224993115, + "learning_rate": 1.9906155698910523e-05, + "loss": 1.1966, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 155 + }, + { + "epoch": 0.19235511713933415, + "grad_norm": 2.047788318815714, + "learning_rate": 1.990434522083839e-05, + "loss": 1.26, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 156 + }, + { + "epoch": 0.1935881627620222, + "grad_norm": 2.0992674413368415, + "learning_rate": 1.99025175289419e-05, + "loss": 1.2544, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 157 + }, + { + "epoch": 0.19482120838471023, + "grad_norm": 2.279539056045867, + "learning_rate": 1.9900672626397593e-05, + "loss": 1.3123, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 158 + }, + { + "epoch": 0.1960542540073983, + "grad_norm": 1.9940111001673004, + "learning_rate": 1.9898810516411926e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 159 + }, + { + "epoch": 0.19728729963008632, + "grad_norm": 2.278268154556979, + "learning_rate": 1.989693120222126e-05, + "loss": 1.236, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 160 + }, + { + "epoch": 0.19852034525277434, + "grad_norm": 2.3184527515323796, + "learning_rate": 1.9895034687091854e-05, + "loss": 1.1691, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 161 + }, + { + "epoch": 0.1997533908754624, + "grad_norm": 2.094943513595831, + "learning_rate": 1.9893120974319867e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 162 + }, + { + "epoch": 0.20098643649815043, + "grad_norm": 2.296308593300257, + "learning_rate": 1.9891190067231346e-05, + "loss": 1.2976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 163 + }, + { + "epoch": 0.20221948212083848, + "grad_norm": 2.131546792333376, + "learning_rate": 1.988924196918222e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 164 + }, + { + "epoch": 0.2034525277435265, + "grad_norm": 1.981669864872828, + "learning_rate": 1.9887276683558302e-05, + "loss": 1.2934, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 165 + }, + { + "epoch": 0.20468557336621454, + "grad_norm": 2.0382654138005423, + "learning_rate": 1.9885294213775267e-05, + "loss": 1.3207, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 166 + }, + { + "epoch": 0.2059186189889026, + "grad_norm": 2.074242050982157, + "learning_rate": 1.9883294563278665e-05, + "loss": 1.1832, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 167 + }, + { + "epoch": 0.20715166461159062, + "grad_norm": 2.1994142392378317, + "learning_rate": 1.9881277735543902e-05, + "loss": 1.2284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 168 + }, + { + "epoch": 0.20838471023427868, + "grad_norm": 2.1893568284451987, + "learning_rate": 1.987924373407624e-05, + "loss": 1.2572, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 169 + }, + { + "epoch": 0.2096177558569667, + "grad_norm": 2.0490903431710352, + "learning_rate": 1.987719256241079e-05, + "loss": 1.2808, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 170 + }, + { + "epoch": 0.21085080147965474, + "grad_norm": 2.3424605103097176, + "learning_rate": 1.9875124224112498e-05, + "loss": 1.2924, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 171 + }, + { + "epoch": 0.2120838471023428, + "grad_norm": 2.100563128603325, + "learning_rate": 1.987303872277616e-05, + "loss": 1.1368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 172 + }, + { + "epoch": 0.21331689272503082, + "grad_norm": 2.303295425870437, + "learning_rate": 1.9870936062026385e-05, + "loss": 1.3425, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 173 + }, + { + "epoch": 0.21454993834771888, + "grad_norm": 1.9108233071934615, + "learning_rate": 1.9868816245517615e-05, + "loss": 1.1926, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 174 + }, + { + "epoch": 0.2157829839704069, + "grad_norm": 2.2539584509521173, + "learning_rate": 1.986667927693411e-05, + "loss": 1.222, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 175 + }, + { + "epoch": 0.21701602959309493, + "grad_norm": 2.006165392274905, + "learning_rate": 1.986452515998993e-05, + "loss": 1.2853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 176 + }, + { + "epoch": 0.218249075215783, + "grad_norm": 2.1337815004505454, + "learning_rate": 1.9862353898428948e-05, + "loss": 1.2671, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 177 + }, + { + "epoch": 0.21948212083847102, + "grad_norm": 2.062911126777883, + "learning_rate": 1.986016549602484e-05, + "loss": 1.2397, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 178 + }, + { + "epoch": 0.22071516646115907, + "grad_norm": 2.115740291239205, + "learning_rate": 1.9857959956581053e-05, + "loss": 1.2534, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 179 + }, + { + "epoch": 0.2219482120838471, + "grad_norm": 2.3733626044353837, + "learning_rate": 1.9855737283930843e-05, + "loss": 1.2082, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 180 + }, + { + "epoch": 0.22318125770653513, + "grad_norm": 2.0972006372187373, + "learning_rate": 1.985349748193722e-05, + "loss": 1.1091, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 181 + }, + { + "epoch": 0.22441430332922319, + "grad_norm": 2.0727561431265635, + "learning_rate": 1.9851240554492985e-05, + "loss": 1.2153, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 182 + }, + { + "epoch": 0.22564734895191121, + "grad_norm": 2.094605431898615, + "learning_rate": 1.9848966505520694e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 183 + }, + { + "epoch": 0.22688039457459927, + "grad_norm": 2.1524188959905866, + "learning_rate": 1.9846675338972654e-05, + "loss": 1.219, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 184 + }, + { + "epoch": 0.2281134401972873, + "grad_norm": 2.008037870627797, + "learning_rate": 1.9844367058830933e-05, + "loss": 1.2635, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 185 + }, + { + "epoch": 0.22934648581997533, + "grad_norm": 2.050448462235658, + "learning_rate": 1.984204166910734e-05, + "loss": 1.187, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 186 + }, + { + "epoch": 0.23057953144266338, + "grad_norm": 2.2727997887406066, + "learning_rate": 1.983969917384342e-05, + "loss": 1.2709, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 187 + }, + { + "epoch": 0.2318125770653514, + "grad_norm": 2.0295960774209414, + "learning_rate": 1.9837339577110443e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 188 + }, + { + "epoch": 0.23304562268803947, + "grad_norm": 2.250775161795276, + "learning_rate": 1.9834962883009406e-05, + "loss": 1.2566, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 189 + }, + { + "epoch": 0.2342786683107275, + "grad_norm": 2.221930554057359, + "learning_rate": 1.9832569095671025e-05, + "loss": 1.2547, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 190 + }, + { + "epoch": 0.23551171393341552, + "grad_norm": 2.071699318441252, + "learning_rate": 1.9830158219255715e-05, + "loss": 1.1712, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 191 + }, + { + "epoch": 0.23674475955610358, + "grad_norm": 1.9444133453118777, + "learning_rate": 1.98277302579536e-05, + "loss": 1.1858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 192 + }, + { + "epoch": 0.2379778051787916, + "grad_norm": 2.134049888070869, + "learning_rate": 1.9825285215984495e-05, + "loss": 1.2212, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 193 + }, + { + "epoch": 0.23921085080147966, + "grad_norm": 2.1418697560189988, + "learning_rate": 1.9822823097597893e-05, + "loss": 1.2602, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 194 + }, + { + "epoch": 0.2404438964241677, + "grad_norm": 2.049709663111986, + "learning_rate": 1.9820343907072982e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 195 + }, + { + "epoch": 0.24167694204685575, + "grad_norm": 2.1074579631320605, + "learning_rate": 1.9817847648718615e-05, + "loss": 1.1863, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 196 + }, + { + "epoch": 0.24290998766954378, + "grad_norm": 2.279482709995341, + "learning_rate": 1.9815334326873304e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 197 + }, + { + "epoch": 0.2441430332922318, + "grad_norm": 2.055596994912923, + "learning_rate": 1.9812803945905215e-05, + "loss": 1.2952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 198 + }, + { + "epoch": 0.24537607891491986, + "grad_norm": 2.132145299143716, + "learning_rate": 1.9810256510212178e-05, + "loss": 1.2276, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 199 + }, + { + "epoch": 0.2466091245376079, + "grad_norm": 2.077623874214478, + "learning_rate": 1.9807692024221647e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 200 + }, + { + "epoch": 0.24784217016029594, + "grad_norm": 2.0797643511906356, + "learning_rate": 1.9805110492390723e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 201 + }, + { + "epoch": 0.24907521578298397, + "grad_norm": 2.0930244790953854, + "learning_rate": 1.9802511919206123e-05, + "loss": 1.1984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 202 + }, + { + "epoch": 0.25030826140567203, + "grad_norm": 1.9804465011478614, + "learning_rate": 1.9799896309184192e-05, + "loss": 1.1436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 203 + }, + { + "epoch": 0.25154130702836003, + "grad_norm": 2.054252771935451, + "learning_rate": 1.9797263666870876e-05, + "loss": 1.1913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 204 + }, + { + "epoch": 0.2527743526510481, + "grad_norm": 1.9941485664509049, + "learning_rate": 1.9794613996841723e-05, + "loss": 1.1524, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 205 + }, + { + "epoch": 0.25400739827373614, + "grad_norm": 2.051706932133756, + "learning_rate": 1.9791947303701882e-05, + "loss": 1.2003, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 206 + }, + { + "epoch": 0.25524044389642414, + "grad_norm": 1.9992360549131547, + "learning_rate": 1.9789263592086085e-05, + "loss": 1.2392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 207 + }, + { + "epoch": 0.2564734895191122, + "grad_norm": 2.0558176906965375, + "learning_rate": 1.9786562866658644e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 208 + }, + { + "epoch": 0.25770653514180025, + "grad_norm": 2.2225484776914426, + "learning_rate": 1.9783845132113438e-05, + "loss": 1.1771, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 209 + }, + { + "epoch": 0.2589395807644883, + "grad_norm": 2.2832873135247516, + "learning_rate": 1.978111039317391e-05, + "loss": 1.2226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 210 + }, + { + "epoch": 0.2601726263871763, + "grad_norm": 1.9887162353358598, + "learning_rate": 1.9778358654593054e-05, + "loss": 1.1936, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 211 + }, + { + "epoch": 0.26140567200986436, + "grad_norm": 1.9738209615627333, + "learning_rate": 1.9775589921153417e-05, + "loss": 1.2666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 212 + }, + { + "epoch": 0.2626387176325524, + "grad_norm": 1.8817044406816048, + "learning_rate": 1.9772804197667074e-05, + "loss": 1.1842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 213 + }, + { + "epoch": 0.2638717632552404, + "grad_norm": 2.013799438378123, + "learning_rate": 1.9770001488975632e-05, + "loss": 1.2242, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 214 + }, + { + "epoch": 0.2651048088779285, + "grad_norm": 1.9119015873636653, + "learning_rate": 1.9767181799950222e-05, + "loss": 1.0898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 215 + }, + { + "epoch": 0.26633785450061653, + "grad_norm": 1.9353703261833344, + "learning_rate": 1.976434513549148e-05, + "loss": 1.2241, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 216 + }, + { + "epoch": 0.2675709001233046, + "grad_norm": 1.8684279347954125, + "learning_rate": 1.9761491500529555e-05, + "loss": 1.209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 217 + }, + { + "epoch": 0.2688039457459926, + "grad_norm": 2.1880109676776556, + "learning_rate": 1.9758620900024083e-05, + "loss": 1.1956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 218 + }, + { + "epoch": 0.27003699136868065, + "grad_norm": 2.083271319194177, + "learning_rate": 1.9755733338964186e-05, + "loss": 1.3062, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 219 + }, + { + "epoch": 0.2712700369913687, + "grad_norm": 2.093319969744569, + "learning_rate": 1.9752828822368468e-05, + "loss": 1.2083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 220 + }, + { + "epoch": 0.2725030826140567, + "grad_norm": 1.8811834241518346, + "learning_rate": 1.9749907355285e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 221 + }, + { + "epoch": 0.27373612823674476, + "grad_norm": 2.012899236904161, + "learning_rate": 1.9746968942791314e-05, + "loss": 1.1476, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 222 + }, + { + "epoch": 0.2749691738594328, + "grad_norm": 1.9990139044183095, + "learning_rate": 1.9744013589994397e-05, + "loss": 1.1257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 223 + }, + { + "epoch": 0.2762022194821208, + "grad_norm": 1.9924345221184356, + "learning_rate": 1.9741041302030666e-05, + "loss": 1.224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 224 + }, + { + "epoch": 0.27743526510480887, + "grad_norm": 2.2180474451395344, + "learning_rate": 1.9738052084065982e-05, + "loss": 1.1713, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 225 + }, + { + "epoch": 0.2786683107274969, + "grad_norm": 1.983690334551248, + "learning_rate": 1.973504594129563e-05, + "loss": 1.179, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 226 + }, + { + "epoch": 0.279901356350185, + "grad_norm": 2.283777738854751, + "learning_rate": 1.97320228789443e-05, + "loss": 1.2869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 227 + }, + { + "epoch": 0.281134401972873, + "grad_norm": 1.943369966527272, + "learning_rate": 1.9728982902266107e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 228 + }, + { + "epoch": 0.28236744759556104, + "grad_norm": 2.0711638809016333, + "learning_rate": 1.9725926016544544e-05, + "loss": 1.16, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 229 + }, + { + "epoch": 0.2836004932182491, + "grad_norm": 1.9786842049852629, + "learning_rate": 1.972285222709251e-05, + "loss": 1.2338, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 230 + }, + { + "epoch": 0.2848335388409371, + "grad_norm": 1.9447068528661613, + "learning_rate": 1.9719761539252268e-05, + "loss": 1.1827, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 231 + }, + { + "epoch": 0.28606658446362515, + "grad_norm": 2.0161169975225475, + "learning_rate": 1.971665395839545e-05, + "loss": 1.117, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 232 + }, + { + "epoch": 0.2872996300863132, + "grad_norm": 2.181216178694371, + "learning_rate": 1.9713529489923067e-05, + "loss": 1.1807, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 233 + }, + { + "epoch": 0.2885326757090012, + "grad_norm": 2.254001309068264, + "learning_rate": 1.9710388139265458e-05, + "loss": 1.0036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 234 + }, + { + "epoch": 0.28976572133168926, + "grad_norm": 2.028042430685245, + "learning_rate": 1.9707229911882318e-05, + "loss": 1.1175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 235 + }, + { + "epoch": 0.2909987669543773, + "grad_norm": 2.0601619279255177, + "learning_rate": 1.9704054813262666e-05, + "loss": 1.2758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 236 + }, + { + "epoch": 0.2922318125770654, + "grad_norm": 2.125107661538684, + "learning_rate": 1.9700862848924852e-05, + "loss": 1.2611, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 237 + }, + { + "epoch": 0.2934648581997534, + "grad_norm": 2.0416663193733746, + "learning_rate": 1.9697654024416528e-05, + "loss": 1.1986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 238 + }, + { + "epoch": 0.29469790382244143, + "grad_norm": 2.0193702602947554, + "learning_rate": 1.9694428345314657e-05, + "loss": 1.181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 239 + }, + { + "epoch": 0.2959309494451295, + "grad_norm": 2.0150819278141148, + "learning_rate": 1.9691185817225493e-05, + "loss": 1.1118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 240 + }, + { + "epoch": 0.2971639950678175, + "grad_norm": 2.355208602911491, + "learning_rate": 1.9687926445784576e-05, + "loss": 1.1013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 241 + }, + { + "epoch": 0.29839704069050554, + "grad_norm": 1.9770821951357245, + "learning_rate": 1.968465023665671e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 242 + }, + { + "epoch": 0.2996300863131936, + "grad_norm": 1.992866162682536, + "learning_rate": 1.9681357195535974e-05, + "loss": 1.2264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 243 + }, + { + "epoch": 0.3008631319358816, + "grad_norm": 1.9560049490402185, + "learning_rate": 1.9678047328145702e-05, + "loss": 1.1903, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 244 + }, + { + "epoch": 0.30209617755856966, + "grad_norm": 2.0070855957858114, + "learning_rate": 1.967472064023846e-05, + "loss": 1.1793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 245 + }, + { + "epoch": 0.3033292231812577, + "grad_norm": 2.0661988646186313, + "learning_rate": 1.9671377137596064e-05, + "loss": 1.2222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 246 + }, + { + "epoch": 0.30456226880394577, + "grad_norm": 2.1056352825098816, + "learning_rate": 1.9668016826029538e-05, + "loss": 1.2107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 247 + }, + { + "epoch": 0.30579531442663377, + "grad_norm": 1.98815122638957, + "learning_rate": 1.9664639711379135e-05, + "loss": 1.2816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 248 + }, + { + "epoch": 0.3070283600493218, + "grad_norm": 2.1440514280412737, + "learning_rate": 1.9661245799514298e-05, + "loss": 1.2021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 249 + }, + { + "epoch": 0.3082614056720099, + "grad_norm": 2.0231899704902805, + "learning_rate": 1.9657835096333676e-05, + "loss": 1.2358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 250 + }, + { + "epoch": 0.3094944512946979, + "grad_norm": 1.9883606430802652, + "learning_rate": 1.9654407607765098e-05, + "loss": 1.1847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 251 + }, + { + "epoch": 0.31072749691738594, + "grad_norm": 2.1163621969666786, + "learning_rate": 1.965096333976556e-05, + "loss": 1.2595, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 252 + }, + { + "epoch": 0.311960542540074, + "grad_norm": 1.9898069180734743, + "learning_rate": 1.9647502298321223e-05, + "loss": 1.2286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 253 + }, + { + "epoch": 0.31319358816276205, + "grad_norm": 1.929003891101706, + "learning_rate": 1.9644024489447403e-05, + "loss": 1.2394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 254 + }, + { + "epoch": 0.31442663378545005, + "grad_norm": 1.98661541473719, + "learning_rate": 1.9640529919188562e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 255 + }, + { + "epoch": 0.3156596794081381, + "grad_norm": 1.9483621179175639, + "learning_rate": 1.9637018593618285e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 256 + }, + { + "epoch": 0.31689272503082616, + "grad_norm": 1.8370569270185535, + "learning_rate": 1.9633490518839285e-05, + "loss": 1.1627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 257 + }, + { + "epoch": 0.31812577065351416, + "grad_norm": 2.2350844635149376, + "learning_rate": 1.9629945700983376e-05, + "loss": 1.1717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 258 + }, + { + "epoch": 0.3193588162762022, + "grad_norm": 2.3938085963940847, + "learning_rate": 1.9626384146211482e-05, + "loss": 1.2982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 259 + }, + { + "epoch": 0.3205918618988903, + "grad_norm": 1.9382250145807791, + "learning_rate": 1.962280586071361e-05, + "loss": 1.1414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 260 + }, + { + "epoch": 0.3218249075215783, + "grad_norm": 2.048034832679974, + "learning_rate": 1.9619210850708844e-05, + "loss": 1.2367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 261 + }, + { + "epoch": 0.32305795314426633, + "grad_norm": 1.868035630300947, + "learning_rate": 1.9615599122445338e-05, + "loss": 1.1582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 262 + }, + { + "epoch": 0.3242909987669544, + "grad_norm": 1.9842993509323321, + "learning_rate": 1.9611970682200308e-05, + "loss": 1.1214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 263 + }, + { + "epoch": 0.32552404438964244, + "grad_norm": 2.009754214636891, + "learning_rate": 1.9608325536280003e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 264 + }, + { + "epoch": 0.32675709001233044, + "grad_norm": 2.0322676023608097, + "learning_rate": 1.9604663691019715e-05, + "loss": 1.1514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 265 + }, + { + "epoch": 0.3279901356350185, + "grad_norm": 1.9898336979024067, + "learning_rate": 1.960098515278376e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 266 + }, + { + "epoch": 0.32922318125770655, + "grad_norm": 2.1120578466218007, + "learning_rate": 1.959728992796546e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 267 + }, + { + "epoch": 0.33045622688039455, + "grad_norm": 2.230681018136028, + "learning_rate": 1.959357802298715e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 268 + }, + { + "epoch": 0.3316892725030826, + "grad_norm": 2.121990574572694, + "learning_rate": 1.9589849444300142e-05, + "loss": 1.1686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 269 + }, + { + "epoch": 0.33292231812577067, + "grad_norm": 1.9793276124371497, + "learning_rate": 1.958610419838473e-05, + "loss": 1.2605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 270 + }, + { + "epoch": 0.33415536374845867, + "grad_norm": 2.160304448339165, + "learning_rate": 1.958234229175019e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 271 + }, + { + "epoch": 0.3353884093711467, + "grad_norm": 2.0375075579110646, + "learning_rate": 1.9578563730934733e-05, + "loss": 1.1282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 272 + }, + { + "epoch": 0.3366214549938348, + "grad_norm": 2.0496730732316313, + "learning_rate": 1.9574768522505528e-05, + "loss": 1.2336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 273 + }, + { + "epoch": 0.33785450061652283, + "grad_norm": 2.0096053009394987, + "learning_rate": 1.9570956673058678e-05, + "loss": 1.1737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 274 + }, + { + "epoch": 0.33908754623921084, + "grad_norm": 2.064991805627586, + "learning_rate": 1.9567128189219203e-05, + "loss": 1.2067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 275 + }, + { + "epoch": 0.3403205918618989, + "grad_norm": 2.016629291677699, + "learning_rate": 1.9563283077641032e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 276 + }, + { + "epoch": 0.34155363748458695, + "grad_norm": 2.026571915003702, + "learning_rate": 1.9559421345007004e-05, + "loss": 0.9722, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 277 + }, + { + "epoch": 0.34278668310727495, + "grad_norm": 2.0408528020000896, + "learning_rate": 1.9555542998028836e-05, + "loss": 1.0951, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 278 + }, + { + "epoch": 0.344019728729963, + "grad_norm": 2.4490550716870336, + "learning_rate": 1.955164804344712e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 279 + }, + { + "epoch": 0.34525277435265106, + "grad_norm": 2.0298228033571517, + "learning_rate": 1.9547736488031322e-05, + "loss": 1.2091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 280 + }, + { + "epoch": 0.34648581997533906, + "grad_norm": 1.928172339548102, + "learning_rate": 1.9543808338579746e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 281 + }, + { + "epoch": 0.3477188655980271, + "grad_norm": 1.9880687290891643, + "learning_rate": 1.953986360191955e-05, + "loss": 1.0397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 282 + }, + { + "epoch": 0.34895191122071517, + "grad_norm": 2.3895656693876823, + "learning_rate": 1.9535902284906717e-05, + "loss": 1.2004, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 283 + }, + { + "epoch": 0.35018495684340323, + "grad_norm": 1.9995303874089259, + "learning_rate": 1.9531924394426036e-05, + "loss": 1.2001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 284 + }, + { + "epoch": 0.35141800246609123, + "grad_norm": 1.9751518073159213, + "learning_rate": 1.9527929937391118e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 285 + }, + { + "epoch": 0.3526510480887793, + "grad_norm": 2.002111448708798, + "learning_rate": 1.9523918920744357e-05, + "loss": 1.0845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 286 + }, + { + "epoch": 0.35388409371146734, + "grad_norm": 2.0131766303020036, + "learning_rate": 1.9519891351456926e-05, + "loss": 1.1764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 287 + }, + { + "epoch": 0.35511713933415534, + "grad_norm": 2.2331356025573785, + "learning_rate": 1.9515847236528772e-05, + "loss": 1.225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 288 + }, + { + "epoch": 0.3563501849568434, + "grad_norm": 2.0628294266012204, + "learning_rate": 1.9511786582988597e-05, + "loss": 1.2259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 289 + }, + { + "epoch": 0.35758323057953145, + "grad_norm": 1.9151442433897523, + "learning_rate": 1.9507709397893846e-05, + "loss": 1.2103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 290 + }, + { + "epoch": 0.3588162762022195, + "grad_norm": 1.958231535128413, + "learning_rate": 1.9503615688330698e-05, + "loss": 1.105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 291 + }, + { + "epoch": 0.3600493218249075, + "grad_norm": 2.149244076290205, + "learning_rate": 1.9499505461414045e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 292 + }, + { + "epoch": 0.36128236744759556, + "grad_norm": 2.017117422238215, + "learning_rate": 1.9495378724287498e-05, + "loss": 1.1007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 293 + }, + { + "epoch": 0.3625154130702836, + "grad_norm": 1.9210343338897897, + "learning_rate": 1.949123548412335e-05, + "loss": 1.0755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 294 + }, + { + "epoch": 0.3637484586929716, + "grad_norm": 1.9955568267380113, + "learning_rate": 1.9487075748122586e-05, + "loss": 1.1702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 295 + }, + { + "epoch": 0.3649815043156597, + "grad_norm": 2.014663690647869, + "learning_rate": 1.948289952351486e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 296 + }, + { + "epoch": 0.36621454993834773, + "grad_norm": 2.0560814736523922, + "learning_rate": 1.9478706817558474e-05, + "loss": 1.1981, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 297 + }, + { + "epoch": 0.36744759556103573, + "grad_norm": 2.072244571774935, + "learning_rate": 1.947449763754039e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 298 + }, + { + "epoch": 0.3686806411837238, + "grad_norm": 2.3134145282189724, + "learning_rate": 1.947027199077619e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 299 + }, + { + "epoch": 0.36991368680641185, + "grad_norm": 1.9505648569157141, + "learning_rate": 1.9466029884610072e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 300 + }, + { + "epoch": 0.3711467324290999, + "grad_norm": 1.9850708859825466, + "learning_rate": 1.946177132641486e-05, + "loss": 1.0863, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 301 + }, + { + "epoch": 0.3723797780517879, + "grad_norm": 2.3267730903818755, + "learning_rate": 1.945749632359195e-05, + "loss": 1.1405, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 302 + }, + { + "epoch": 0.37361282367447596, + "grad_norm": 2.403438763479749, + "learning_rate": 1.9453204883571333e-05, + "loss": 1.1977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 303 + }, + { + "epoch": 0.374845869297164, + "grad_norm": 2.028737664163498, + "learning_rate": 1.9448897013811558e-05, + "loss": 1.2029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 304 + }, + { + "epoch": 0.376078914919852, + "grad_norm": 1.9981856925053472, + "learning_rate": 1.9444572721799743e-05, + "loss": 1.1873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 305 + }, + { + "epoch": 0.37731196054254007, + "grad_norm": 1.968701546642257, + "learning_rate": 1.9440232015051526e-05, + "loss": 1.1721, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 306 + }, + { + "epoch": 0.3785450061652281, + "grad_norm": 1.864390281708884, + "learning_rate": 1.9435874901111102e-05, + "loss": 1.1797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 307 + }, + { + "epoch": 0.3797780517879161, + "grad_norm": 2.0705929640726577, + "learning_rate": 1.9431501387551158e-05, + "loss": 1.1453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 308 + }, + { + "epoch": 0.3810110974106042, + "grad_norm": 1.884359293663656, + "learning_rate": 1.942711148197289e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 309 + }, + { + "epoch": 0.38224414303329224, + "grad_norm": 1.936787839699547, + "learning_rate": 1.9422705192005994e-05, + "loss": 1.2022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 310 + }, + { + "epoch": 0.3834771886559803, + "grad_norm": 1.9636401987324612, + "learning_rate": 1.941828252530863e-05, + "loss": 1.2015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 311 + }, + { + "epoch": 0.3847102342786683, + "grad_norm": 2.0693459412185495, + "learning_rate": 1.9413843489567423e-05, + "loss": 1.054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 312 + }, + { + "epoch": 0.38594327990135635, + "grad_norm": 2.0182958064572256, + "learning_rate": 1.9409388092497453e-05, + "loss": 1.1086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 313 + }, + { + "epoch": 0.3871763255240444, + "grad_norm": 2.093433911405551, + "learning_rate": 1.940491634184223e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 314 + }, + { + "epoch": 0.3884093711467324, + "grad_norm": 2.018372038783884, + "learning_rate": 1.9400428245373693e-05, + "loss": 1.0813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 315 + }, + { + "epoch": 0.38964241676942046, + "grad_norm": 1.9703329339612685, + "learning_rate": 1.9395923810892184e-05, + "loss": 1.0768, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 316 + }, + { + "epoch": 0.3908754623921085, + "grad_norm": 1.9327973664586777, + "learning_rate": 1.939140304622644e-05, + "loss": 1.1157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 317 + }, + { + "epoch": 0.3921085080147966, + "grad_norm": 2.019497002973341, + "learning_rate": 1.938686595923359e-05, + "loss": 0.9835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 318 + }, + { + "epoch": 0.3933415536374846, + "grad_norm": 2.089968778066188, + "learning_rate": 1.9382312557799116e-05, + "loss": 1.0213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 319 + }, + { + "epoch": 0.39457459926017263, + "grad_norm": 2.0510099518796006, + "learning_rate": 1.937774284983687e-05, + "loss": 1.099, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 320 + }, + { + "epoch": 0.3958076448828607, + "grad_norm": 2.0175024562414063, + "learning_rate": 1.9373156843289037e-05, + "loss": 1.0764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 321 + }, + { + "epoch": 0.3970406905055487, + "grad_norm": 2.020977606778504, + "learning_rate": 1.9368554546126126e-05, + "loss": 1.0562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 322 + }, + { + "epoch": 0.39827373612823674, + "grad_norm": 2.2138761252967125, + "learning_rate": 1.936393596634697e-05, + "loss": 1.196, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 323 + }, + { + "epoch": 0.3995067817509248, + "grad_norm": 1.855560397827274, + "learning_rate": 1.9359301111978683e-05, + "loss": 1.0725, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 324 + }, + { + "epoch": 0.4007398273736128, + "grad_norm": 2.0805790476522636, + "learning_rate": 1.9354649991076684e-05, + "loss": 1.1665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 325 + }, + { + "epoch": 0.40197287299630086, + "grad_norm": 1.8807037569919354, + "learning_rate": 1.9349982611724654e-05, + "loss": 1.1366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 326 + }, + { + "epoch": 0.4032059186189889, + "grad_norm": 1.9286740986289053, + "learning_rate": 1.9345298982034536e-05, + "loss": 1.132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 327 + }, + { + "epoch": 0.40443896424167697, + "grad_norm": 2.0846402787959786, + "learning_rate": 1.9340599110146504e-05, + "loss": 1.1055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 328 + }, + { + "epoch": 0.40567200986436497, + "grad_norm": 2.2183587793468993, + "learning_rate": 1.9335883004228974e-05, + "loss": 1.1544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 329 + }, + { + "epoch": 0.406905055487053, + "grad_norm": 2.2213015226303643, + "learning_rate": 1.9331150672478575e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 330 + }, + { + "epoch": 0.4081381011097411, + "grad_norm": 2.2277113238319726, + "learning_rate": 1.9326402123120132e-05, + "loss": 1.0451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 331 + }, + { + "epoch": 0.4093711467324291, + "grad_norm": 2.259911506493339, + "learning_rate": 1.9321637364406658e-05, + "loss": 1.0236, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 332 + }, + { + "epoch": 0.41060419235511714, + "grad_norm": 2.3573090703399417, + "learning_rate": 1.931685640461934e-05, + "loss": 1.1412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 333 + }, + { + "epoch": 0.4118372379778052, + "grad_norm": 2.1097456297814468, + "learning_rate": 1.9312059252067517e-05, + "loss": 1.1901, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 334 + }, + { + "epoch": 0.4130702836004932, + "grad_norm": 2.121952527251297, + "learning_rate": 1.930724591508868e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 335 + }, + { + "epoch": 0.41430332922318125, + "grad_norm": 2.091327214317948, + "learning_rate": 1.930241640204844e-05, + "loss": 1.0135, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 336 + }, + { + "epoch": 0.4155363748458693, + "grad_norm": 2.1272216173587255, + "learning_rate": 1.929757072134053e-05, + "loss": 1.1754, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 337 + }, + { + "epoch": 0.41676942046855736, + "grad_norm": 1.9973226929736474, + "learning_rate": 1.929270888138678e-05, + "loss": 1.1392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 338 + }, + { + "epoch": 0.41800246609124536, + "grad_norm": 1.9318995412408386, + "learning_rate": 1.9287830890637093e-05, + "loss": 1.0762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 339 + }, + { + "epoch": 0.4192355117139334, + "grad_norm": 2.785840304521055, + "learning_rate": 1.9282936757569463e-05, + "loss": 1.1699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 340 + }, + { + "epoch": 0.4204685573366215, + "grad_norm": 2.171910512782072, + "learning_rate": 1.9278026490689925e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 341 + }, + { + "epoch": 0.4217016029593095, + "grad_norm": 2.052852932277261, + "learning_rate": 1.927310009853256e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 342 + }, + { + "epoch": 0.42293464858199753, + "grad_norm": 1.9538275366336955, + "learning_rate": 1.926815758965947e-05, + "loss": 1.1934, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 343 + }, + { + "epoch": 0.4241676942046856, + "grad_norm": 1.9772612426660126, + "learning_rate": 1.9263198972660774e-05, + "loss": 1.1061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 344 + }, + { + "epoch": 0.4254007398273736, + "grad_norm": 1.9522364508331842, + "learning_rate": 1.9258224256154588e-05, + "loss": 1.101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 345 + }, + { + "epoch": 0.42663378545006164, + "grad_norm": 2.1060276682972714, + "learning_rate": 1.9253233448787e-05, + "loss": 1.2094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 346 + }, + { + "epoch": 0.4278668310727497, + "grad_norm": 2.020183264992464, + "learning_rate": 1.9248226559232074e-05, + "loss": 1.2012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 347 + }, + { + "epoch": 0.42909987669543775, + "grad_norm": 2.1361451948859256, + "learning_rate": 1.9243203596191815e-05, + "loss": 1.2714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 348 + }, + { + "epoch": 0.43033292231812575, + "grad_norm": 1.890346669461953, + "learning_rate": 1.9238164568396177e-05, + "loss": 1.067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 349 + }, + { + "epoch": 0.4315659679408138, + "grad_norm": 1.908858157385325, + "learning_rate": 1.9233109484603025e-05, + "loss": 1.1123, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 350 + }, + { + "epoch": 0.43279901356350187, + "grad_norm": 1.9095667461482675, + "learning_rate": 1.9228038353598122e-05, + "loss": 1.0834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 351 + }, + { + "epoch": 0.43403205918618987, + "grad_norm": 1.8950946192756821, + "learning_rate": 1.9222951184195146e-05, + "loss": 1.0686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 352 + }, + { + "epoch": 0.4352651048088779, + "grad_norm": 2.5409486063614066, + "learning_rate": 1.921784798523562e-05, + "loss": 1.1168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 353 + }, + { + "epoch": 0.436498150431566, + "grad_norm": 1.9251760877216109, + "learning_rate": 1.9212728765588954e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 354 + }, + { + "epoch": 0.43773119605425403, + "grad_norm": 2.12327469473301, + "learning_rate": 1.9207593534152383e-05, + "loss": 1.1871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 355 + }, + { + "epoch": 0.43896424167694204, + "grad_norm": 2.0404259513933214, + "learning_rate": 1.9202442299850977e-05, + "loss": 1.0985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 356 + }, + { + "epoch": 0.4401972872996301, + "grad_norm": 2.10787306308974, + "learning_rate": 1.9197275071637617e-05, + "loss": 1.0063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 357 + }, + { + "epoch": 0.44143033292231815, + "grad_norm": 2.2402685683097996, + "learning_rate": 1.919209185849299e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 358 + }, + { + "epoch": 0.44266337854500615, + "grad_norm": 2.148922705064427, + "learning_rate": 1.9186892669425546e-05, + "loss": 1.0889, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 359 + }, + { + "epoch": 0.4438964241676942, + "grad_norm": 1.8789248361497928, + "learning_rate": 1.9181677513471525e-05, + "loss": 1.1796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 360 + }, + { + "epoch": 0.44512946979038226, + "grad_norm": 1.8886466190263609, + "learning_rate": 1.9176446399694903e-05, + "loss": 1.1494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 361 + }, + { + "epoch": 0.44636251541307026, + "grad_norm": 1.9432580511904547, + "learning_rate": 1.9171199337187395e-05, + "loss": 1.2599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 362 + }, + { + "epoch": 0.4475955610357583, + "grad_norm": 1.8820277177030855, + "learning_rate": 1.916593633506843e-05, + "loss": 1.1884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 363 + }, + { + "epoch": 0.44882860665844637, + "grad_norm": 2.249700768154195, + "learning_rate": 1.9160657402485147e-05, + "loss": 1.0503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 364 + }, + { + "epoch": 0.45006165228113443, + "grad_norm": 2.082324666968816, + "learning_rate": 1.9155362548612366e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 365 + }, + { + "epoch": 0.45129469790382243, + "grad_norm": 2.2090950658394193, + "learning_rate": 1.915005178265259e-05, + "loss": 1.1742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 366 + }, + { + "epoch": 0.4525277435265105, + "grad_norm": 2.2007793765172683, + "learning_rate": 1.914472511383596e-05, + "loss": 1.2617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 367 + }, + { + "epoch": 0.45376078914919854, + "grad_norm": 1.9166813253050292, + "learning_rate": 1.913938255142027e-05, + "loss": 1.1444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 368 + }, + { + "epoch": 0.45499383477188654, + "grad_norm": 1.908891419449054, + "learning_rate": 1.913402410469093e-05, + "loss": 1.1139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 369 + }, + { + "epoch": 0.4562268803945746, + "grad_norm": 2.0434795360025566, + "learning_rate": 1.9128649782960965e-05, + "loss": 1.1761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 370 + }, + { + "epoch": 0.45745992601726265, + "grad_norm": 2.2319795236629165, + "learning_rate": 1.9123259595570978e-05, + "loss": 0.9997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 371 + }, + { + "epoch": 0.45869297163995065, + "grad_norm": 1.9325225410679332, + "learning_rate": 1.911785355188916e-05, + "loss": 1.1191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 372 + }, + { + "epoch": 0.4599260172626387, + "grad_norm": 2.3718583427313384, + "learning_rate": 1.9112431661311248e-05, + "loss": 1.1801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 373 + }, + { + "epoch": 0.46115906288532676, + "grad_norm": 2.2778458302261853, + "learning_rate": 1.9106993933260535e-05, + "loss": 1.1973, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 374 + }, + { + "epoch": 0.4623921085080148, + "grad_norm": 2.104485888382934, + "learning_rate": 1.910154037718783e-05, + "loss": 1.206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 375 + }, + { + "epoch": 0.4636251541307028, + "grad_norm": 2.149849667802547, + "learning_rate": 1.909607100257145e-05, + "loss": 1.0479, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 376 + }, + { + "epoch": 0.4648581997533909, + "grad_norm": 2.242058517922103, + "learning_rate": 1.9090585818917205e-05, + "loss": 1.0902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 377 + }, + { + "epoch": 0.46609124537607893, + "grad_norm": 2.0943438214587307, + "learning_rate": 1.9085084835758394e-05, + "loss": 1.2428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 378 + }, + { + "epoch": 0.46732429099876693, + "grad_norm": 2.1534146036573647, + "learning_rate": 1.907956806265576e-05, + "loss": 1.0392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 379 + }, + { + "epoch": 0.468557336621455, + "grad_norm": 2.2361240083173324, + "learning_rate": 1.9074035509197495e-05, + "loss": 1.1337, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 380 + }, + { + "epoch": 0.46979038224414305, + "grad_norm": 2.0666798517773795, + "learning_rate": 1.906848718499922e-05, + "loss": 1.0574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 381 + }, + { + "epoch": 0.47102342786683105, + "grad_norm": 2.0056597359218094, + "learning_rate": 1.906292309970396e-05, + "loss": 1.1982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 382 + }, + { + "epoch": 0.4722564734895191, + "grad_norm": 1.931784648165126, + "learning_rate": 1.9057343262982136e-05, + "loss": 1.0671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 383 + }, + { + "epoch": 0.47348951911220716, + "grad_norm": 2.2650382784813408, + "learning_rate": 1.905174768453155e-05, + "loss": 1.1384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 384 + }, + { + "epoch": 0.4747225647348952, + "grad_norm": 2.2386337800038594, + "learning_rate": 1.904613637407735e-05, + "loss": 1.169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 385 + }, + { + "epoch": 0.4759556103575832, + "grad_norm": 1.985439502238856, + "learning_rate": 1.9040509341372045e-05, + "loss": 1.0544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 386 + }, + { + "epoch": 0.47718865598027127, + "grad_norm": 1.9991995362972141, + "learning_rate": 1.903486659619545e-05, + "loss": 1.1048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 387 + }, + { + "epoch": 0.4784217016029593, + "grad_norm": 1.938313304762426, + "learning_rate": 1.90292081483547e-05, + "loss": 1.0217, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 388 + }, + { + "epoch": 0.4796547472256473, + "grad_norm": 2.0593580639988605, + "learning_rate": 1.9023534007684217e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 389 + }, + { + "epoch": 0.4808877928483354, + "grad_norm": 2.0113392094384723, + "learning_rate": 1.90178441840457e-05, + "loss": 1.2114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 390 + }, + { + "epoch": 0.48212083847102344, + "grad_norm": 1.9865461362024583, + "learning_rate": 1.901213868732811e-05, + "loss": 1.0818, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 391 + }, + { + "epoch": 0.4833538840937115, + "grad_norm": 1.9044551829036014, + "learning_rate": 1.9006417527447634e-05, + "loss": 1.1046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 392 + }, + { + "epoch": 0.4845869297163995, + "grad_norm": 1.8508018232087575, + "learning_rate": 1.9000680714347694e-05, + "loss": 1.1411, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 393 + }, + { + "epoch": 0.48581997533908755, + "grad_norm": 1.98419367965867, + "learning_rate": 1.8994928257998907e-05, + "loss": 0.9804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 394 + }, + { + "epoch": 0.4870530209617756, + "grad_norm": 2.1671038349199567, + "learning_rate": 1.8989160168399093e-05, + "loss": 1.1656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 395 + }, + { + "epoch": 0.4882860665844636, + "grad_norm": 2.0570349889938284, + "learning_rate": 1.898337645557323e-05, + "loss": 1.2111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 396 + }, + { + "epoch": 0.48951911220715166, + "grad_norm": 1.9619760913338724, + "learning_rate": 1.8977577129573452e-05, + "loss": 1.0932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 397 + }, + { + "epoch": 0.4907521578298397, + "grad_norm": 1.9608046473112009, + "learning_rate": 1.897176220047903e-05, + "loss": 1.1389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 398 + }, + { + "epoch": 0.4919852034525277, + "grad_norm": 1.9645035824539876, + "learning_rate": 1.8965931678396357e-05, + "loss": 1.1056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 399 + }, + { + "epoch": 0.4932182490752158, + "grad_norm": 1.961734329156117, + "learning_rate": 1.896008557345892e-05, + "loss": 1.1322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 400 + }, + { + "epoch": 0.49445129469790383, + "grad_norm": 1.844560254648902, + "learning_rate": 1.8954223895827288e-05, + "loss": 1.0712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 401 + }, + { + "epoch": 0.4956843403205919, + "grad_norm": 2.1465454734369933, + "learning_rate": 1.8948346655689108e-05, + "loss": 1.1294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 402 + }, + { + "epoch": 0.4969173859432799, + "grad_norm": 2.078244569426143, + "learning_rate": 1.8942453863259058e-05, + "loss": 1.0546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 403 + }, + { + "epoch": 0.49815043156596794, + "grad_norm": 2.264508341916039, + "learning_rate": 1.893654552877886e-05, + "loss": 1.0568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 404 + }, + { + "epoch": 0.499383477188656, + "grad_norm": 2.0051680479216696, + "learning_rate": 1.893062166251724e-05, + "loss": 1.0156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 405 + }, + { + "epoch": 0.5006165228113441, + "grad_norm": 2.1569911537953725, + "learning_rate": 1.8924682274769925e-05, + "loss": 1.173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 406 + }, + { + "epoch": 0.5018495684340321, + "grad_norm": 2.002983438517559, + "learning_rate": 1.8918727375859608e-05, + "loss": 1.0073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 407 + }, + { + "epoch": 0.5030826140567201, + "grad_norm": 1.9516175334775212, + "learning_rate": 1.891275697613595e-05, + "loss": 1.1404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 408 + }, + { + "epoch": 0.5043156596794082, + "grad_norm": 2.1494580236712832, + "learning_rate": 1.890677108597555e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 409 + }, + { + "epoch": 0.5055487053020962, + "grad_norm": 2.025395779691303, + "learning_rate": 1.8900769715781928e-05, + "loss": 0.9756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 410 + }, + { + "epoch": 0.5067817509247842, + "grad_norm": 1.9927648271180456, + "learning_rate": 1.889475287598551e-05, + "loss": 1.1394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 411 + }, + { + "epoch": 0.5080147965474723, + "grad_norm": 1.9818065701766985, + "learning_rate": 1.8888720577043603e-05, + "loss": 1.1683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 412 + }, + { + "epoch": 0.5092478421701603, + "grad_norm": 1.9410101081952555, + "learning_rate": 1.8882672829440393e-05, + "loss": 1.0839, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 413 + }, + { + "epoch": 0.5104808877928483, + "grad_norm": 1.8396805813100958, + "learning_rate": 1.8876609643686906e-05, + "loss": 1.114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 414 + }, + { + "epoch": 0.5117139334155364, + "grad_norm": 2.0105399328568905, + "learning_rate": 1.8870531030321004e-05, + "loss": 1.1439, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 415 + }, + { + "epoch": 0.5129469790382244, + "grad_norm": 2.090098149989323, + "learning_rate": 1.886443699990736e-05, + "loss": 1.1153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 416 + }, + { + "epoch": 0.5141800246609125, + "grad_norm": 2.0517637089595246, + "learning_rate": 1.885832756303745e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 417 + }, + { + "epoch": 0.5154130702836005, + "grad_norm": 3.0738713157895674, + "learning_rate": 1.885220273032951e-05, + "loss": 1.0222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 418 + }, + { + "epoch": 0.5166461159062885, + "grad_norm": 2.6677283836123005, + "learning_rate": 1.884606251242855e-05, + "loss": 1.1038, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 419 + }, + { + "epoch": 0.5178791615289766, + "grad_norm": 2.125547466007061, + "learning_rate": 1.8839906920006315e-05, + "loss": 1.2128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 420 + }, + { + "epoch": 0.5191122071516646, + "grad_norm": 2.1599202425686013, + "learning_rate": 1.883373596376126e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 421 + }, + { + "epoch": 0.5203452527743526, + "grad_norm": 2.1229927533556436, + "learning_rate": 1.8827549654418566e-05, + "loss": 1.1353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 422 + }, + { + "epoch": 0.5215782983970407, + "grad_norm": 1.9567023805342967, + "learning_rate": 1.8821348002730072e-05, + "loss": 1.0584, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 423 + }, + { + "epoch": 0.5228113440197287, + "grad_norm": 1.8989916861837828, + "learning_rate": 1.88151310194743e-05, + "loss": 0.9837, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 424 + }, + { + "epoch": 0.5240443896424167, + "grad_norm": 2.1067680957156014, + "learning_rate": 1.8808898715456415e-05, + "loss": 1.0673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 425 + }, + { + "epoch": 0.5252774352651048, + "grad_norm": 2.6013242993173926, + "learning_rate": 1.88026511015082e-05, + "loss": 1.112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 426 + }, + { + "epoch": 0.5265104808877928, + "grad_norm": 2.0539914273851907, + "learning_rate": 1.8796388188488055e-05, + "loss": 1.1031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 427 + }, + { + "epoch": 0.5277435265104808, + "grad_norm": 2.0386084337690584, + "learning_rate": 1.8790109987280972e-05, + "loss": 1.1556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 428 + }, + { + "epoch": 0.528976572133169, + "grad_norm": 2.093419206457373, + "learning_rate": 1.878381650879851e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 429 + }, + { + "epoch": 0.530209617755857, + "grad_norm": 2.2549735586348985, + "learning_rate": 1.877750776397878e-05, + "loss": 1.0663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 430 + }, + { + "epoch": 0.531442663378545, + "grad_norm": 2.1062036310291266, + "learning_rate": 1.877118376378642e-05, + "loss": 1.0455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 431 + }, + { + "epoch": 0.5326757090012331, + "grad_norm": 1.995564286989001, + "learning_rate": 1.87648445192126e-05, + "loss": 0.982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 432 + }, + { + "epoch": 0.5339087546239211, + "grad_norm": 2.41307031725252, + "learning_rate": 1.8758490041274963e-05, + "loss": 1.0778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 433 + }, + { + "epoch": 0.5351418002466092, + "grad_norm": 2.831604603209417, + "learning_rate": 1.875212034101765e-05, + "loss": 1.1345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 434 + }, + { + "epoch": 0.5363748458692972, + "grad_norm": 2.2244728383006604, + "learning_rate": 1.8745735429511236e-05, + "loss": 0.938, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 435 + }, + { + "epoch": 0.5376078914919852, + "grad_norm": 2.0679863467245823, + "learning_rate": 1.8739335317852747e-05, + "loss": 1.0763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 436 + }, + { + "epoch": 0.5388409371146733, + "grad_norm": 2.2454972633071915, + "learning_rate": 1.8732920017165624e-05, + "loss": 1.1207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 437 + }, + { + "epoch": 0.5400739827373613, + "grad_norm": 1.952229569146223, + "learning_rate": 1.8726489538599707e-05, + "loss": 1.0255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 438 + }, + { + "epoch": 0.5413070283600493, + "grad_norm": 1.9236876015939406, + "learning_rate": 1.8720043893331214e-05, + "loss": 0.9933, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 439 + }, + { + "epoch": 0.5425400739827374, + "grad_norm": 1.9736270928366435, + "learning_rate": 1.8713583092562727e-05, + "loss": 1.107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 440 + }, + { + "epoch": 0.5437731196054254, + "grad_norm": 2.1162408965440633, + "learning_rate": 1.870710714752316e-05, + "loss": 1.0441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 441 + }, + { + "epoch": 0.5450061652281134, + "grad_norm": 1.9515336502135756, + "learning_rate": 1.8700616069467753e-05, + "loss": 0.9855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 442 + }, + { + "epoch": 0.5462392108508015, + "grad_norm": 2.566164154732447, + "learning_rate": 1.8694109869678047e-05, + "loss": 1.1289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 443 + }, + { + "epoch": 0.5474722564734895, + "grad_norm": 2.3448298186890026, + "learning_rate": 1.868758855946187e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 444 + }, + { + "epoch": 0.5487053020961775, + "grad_norm": 2.076679946731619, + "learning_rate": 1.8681052150153305e-05, + "loss": 1.0494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 445 + }, + { + "epoch": 0.5499383477188656, + "grad_norm": 2.281980101143931, + "learning_rate": 1.8674500653112674e-05, + "loss": 1.0456, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 446 + }, + { + "epoch": 0.5511713933415536, + "grad_norm": 2.147469472559495, + "learning_rate": 1.8667934079726537e-05, + "loss": 1.0682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 447 + }, + { + "epoch": 0.5524044389642416, + "grad_norm": 2.0621898279926527, + "learning_rate": 1.866135244140763e-05, + "loss": 1.0527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 448 + }, + { + "epoch": 0.5536374845869297, + "grad_norm": 2.0688727890116785, + "learning_rate": 1.8654755749594905e-05, + "loss": 1.0549, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 449 + }, + { + "epoch": 0.5548705302096177, + "grad_norm": 2.0101374651739325, + "learning_rate": 1.864814401575345e-05, + "loss": 1.002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 450 + }, + { + "epoch": 0.5561035758323057, + "grad_norm": 2.2861388762596344, + "learning_rate": 1.8641517251374514e-05, + "loss": 1.0666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 451 + }, + { + "epoch": 0.5573366214549939, + "grad_norm": 2.006862831841472, + "learning_rate": 1.8634875467975456e-05, + "loss": 1.0061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 452 + }, + { + "epoch": 0.5585696670776819, + "grad_norm": 2.144270279582644, + "learning_rate": 1.862821867709975e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 453 + }, + { + "epoch": 0.55980271270037, + "grad_norm": 1.9667742618855084, + "learning_rate": 1.862154689031694e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 454 + }, + { + "epoch": 0.561035758323058, + "grad_norm": 1.8857968215229095, + "learning_rate": 1.8614860119222643e-05, + "loss": 1.0861, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 455 + }, + { + "epoch": 0.562268803945746, + "grad_norm": 1.9374803828274445, + "learning_rate": 1.860815837543852e-05, + "loss": 0.9784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 456 + }, + { + "epoch": 0.5635018495684341, + "grad_norm": 2.0764596642416433, + "learning_rate": 1.8601441670612247e-05, + "loss": 1.1005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 457 + }, + { + "epoch": 0.5647348951911221, + "grad_norm": 2.123062949997989, + "learning_rate": 1.8594710016417505e-05, + "loss": 1.1953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 458 + }, + { + "epoch": 0.5659679408138101, + "grad_norm": 2.311252498934928, + "learning_rate": 1.858796342455396e-05, + "loss": 1.1049, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 459 + }, + { + "epoch": 0.5672009864364982, + "grad_norm": 1.9780001048890374, + "learning_rate": 1.8581201906747242e-05, + "loss": 1.1393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 460 + }, + { + "epoch": 0.5684340320591862, + "grad_norm": 1.968874552657262, + "learning_rate": 1.857442547474891e-05, + "loss": 1.1578, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 461 + }, + { + "epoch": 0.5696670776818742, + "grad_norm": 1.9395225862380807, + "learning_rate": 1.8567634140336457e-05, + "loss": 1.0823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 462 + }, + { + "epoch": 0.5709001233045623, + "grad_norm": 1.9422213147988507, + "learning_rate": 1.856082791531327e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 463 + }, + { + "epoch": 0.5721331689272503, + "grad_norm": 1.9095099328124767, + "learning_rate": 1.855400681150862e-05, + "loss": 1.1041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 464 + }, + { + "epoch": 0.5733662145499383, + "grad_norm": 1.9651570734782207, + "learning_rate": 1.8547170840777635e-05, + "loss": 0.9924, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 465 + }, + { + "epoch": 0.5745992601726264, + "grad_norm": 1.9083237960346455, + "learning_rate": 1.8540320015001283e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 466 + }, + { + "epoch": 0.5758323057953144, + "grad_norm": 2.0442809672147484, + "learning_rate": 1.8533454346086345e-05, + "loss": 0.9449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 467 + }, + { + "epoch": 0.5770653514180024, + "grad_norm": 2.839273126939154, + "learning_rate": 1.8526573845965407e-05, + "loss": 1.1446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 468 + }, + { + "epoch": 0.5782983970406905, + "grad_norm": 2.8401028851750083, + "learning_rate": 1.8519678526596827e-05, + "loss": 1.0793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 469 + }, + { + "epoch": 0.5795314426633785, + "grad_norm": 2.538177485476921, + "learning_rate": 1.8512768399964725e-05, + "loss": 1.1066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 470 + }, + { + "epoch": 0.5807644882860666, + "grad_norm": 2.1457149698206526, + "learning_rate": 1.8505843478078944e-05, + "loss": 1.1138, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 471 + }, + { + "epoch": 0.5819975339087546, + "grad_norm": 1.8879825929112735, + "learning_rate": 1.8498903772975056e-05, + "loss": 1.0181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 472 + }, + { + "epoch": 0.5832305795314426, + "grad_norm": 2.083069072716887, + "learning_rate": 1.8491949296714312e-05, + "loss": 1.0307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 473 + }, + { + "epoch": 0.5844636251541308, + "grad_norm": 2.135777387695568, + "learning_rate": 1.848498006138365e-05, + "loss": 1.0346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 474 + }, + { + "epoch": 0.5856966707768188, + "grad_norm": 2.092600182986245, + "learning_rate": 1.8477996079095643e-05, + "loss": 1.041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 475 + }, + { + "epoch": 0.5869297163995068, + "grad_norm": 2.1246789769186414, + "learning_rate": 1.8470997361988514e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 476 + }, + { + "epoch": 0.5881627620221949, + "grad_norm": 2.062971539274987, + "learning_rate": 1.8463983922226076e-05, + "loss": 1.111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 477 + }, + { + "epoch": 0.5893958076448829, + "grad_norm": 2.092774168134711, + "learning_rate": 1.8456955771997744e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 478 + }, + { + "epoch": 0.5906288532675709, + "grad_norm": 2.23500231174451, + "learning_rate": 1.844991292351849e-05, + "loss": 1.1008, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 479 + }, + { + "epoch": 0.591861898890259, + "grad_norm": 2.04828338908309, + "learning_rate": 1.844285538902884e-05, + "loss": 1.0183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 480 + }, + { + "epoch": 0.593094944512947, + "grad_norm": 2.1680070152194064, + "learning_rate": 1.8435783180794836e-05, + "loss": 1.092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 481 + }, + { + "epoch": 0.594327990135635, + "grad_norm": 2.0706136769588603, + "learning_rate": 1.8428696311108033e-05, + "loss": 1.1438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 482 + }, + { + "epoch": 0.5955610357583231, + "grad_norm": 1.9925622155649279, + "learning_rate": 1.842159479228546e-05, + "loss": 1.0359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 483 + }, + { + "epoch": 0.5967940813810111, + "grad_norm": 1.9425841064456546, + "learning_rate": 1.841447863666961e-05, + "loss": 1.2084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 484 + }, + { + "epoch": 0.5980271270036991, + "grad_norm": 1.929626615569851, + "learning_rate": 1.8407347856628405e-05, + "loss": 1.0615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 485 + }, + { + "epoch": 0.5992601726263872, + "grad_norm": 2.049215260436237, + "learning_rate": 1.8400202464555202e-05, + "loss": 1.0429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 486 + }, + { + "epoch": 0.6004932182490752, + "grad_norm": 2.0464723350262624, + "learning_rate": 1.8393042472868746e-05, + "loss": 1.1504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 487 + }, + { + "epoch": 0.6017262638717632, + "grad_norm": 1.9060167485555373, + "learning_rate": 1.8385867894013144e-05, + "loss": 0.9809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 488 + }, + { + "epoch": 0.6029593094944513, + "grad_norm": 1.9897115419719036, + "learning_rate": 1.837867874045788e-05, + "loss": 1.0573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 489 + }, + { + "epoch": 0.6041923551171393, + "grad_norm": 2.241807622458982, + "learning_rate": 1.8371475024697746e-05, + "loss": 1.1149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 490 + }, + { + "epoch": 0.6054254007398274, + "grad_norm": 2.141493902645165, + "learning_rate": 1.8364256759252855e-05, + "loss": 1.0396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 491 + }, + { + "epoch": 0.6066584463625154, + "grad_norm": 2.168929995419785, + "learning_rate": 1.8357023956668606e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 492 + }, + { + "epoch": 0.6078914919852034, + "grad_norm": 2.095414950529066, + "learning_rate": 1.834977662951566e-05, + "loss": 1.0313, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 493 + }, + { + "epoch": 0.6091245376078915, + "grad_norm": 2.2287918098101516, + "learning_rate": 1.834251479038993e-05, + "loss": 1.0118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 494 + }, + { + "epoch": 0.6103575832305795, + "grad_norm": 2.0458360619004154, + "learning_rate": 1.833523845191254e-05, + "loss": 1.0644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 495 + }, + { + "epoch": 0.6115906288532675, + "grad_norm": 2.1216602719929245, + "learning_rate": 1.8327947626729825e-05, + "loss": 1.1734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 496 + }, + { + "epoch": 0.6128236744759556, + "grad_norm": 2.1445297291236947, + "learning_rate": 1.8320642327513285e-05, + "loss": 0.9661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 497 + }, + { + "epoch": 0.6140567200986436, + "grad_norm": 2.061159959033469, + "learning_rate": 1.831332256695959e-05, + "loss": 1.189, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 498 + }, + { + "epoch": 0.6152897657213316, + "grad_norm": 1.966885029141375, + "learning_rate": 1.8305988357790533e-05, + "loss": 1.0711, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 499 + }, + { + "epoch": 0.6165228113440198, + "grad_norm": 2.127061899116888, + "learning_rate": 1.8298639712753025e-05, + "loss": 0.9351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 500 + }, + { + "epoch": 0.6177558569667078, + "grad_norm": 2.2239600877101835, + "learning_rate": 1.8291276644619066e-05, + "loss": 1.0569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 501 + }, + { + "epoch": 0.6189889025893958, + "grad_norm": 2.0281028990006913, + "learning_rate": 1.8283899166185723e-05, + "loss": 1.0698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 502 + }, + { + "epoch": 0.6202219482120839, + "grad_norm": 2.1224950625771677, + "learning_rate": 1.8276507290275104e-05, + "loss": 0.9905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 503 + }, + { + "epoch": 0.6214549938347719, + "grad_norm": 2.0799215178987, + "learning_rate": 1.8269101029734343e-05, + "loss": 1.0631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 504 + }, + { + "epoch": 0.6226880394574599, + "grad_norm": 2.2495223027498388, + "learning_rate": 1.8261680397435577e-05, + "loss": 0.9869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 505 + }, + { + "epoch": 0.623921085080148, + "grad_norm": 2.036885073926771, + "learning_rate": 1.825424540627592e-05, + "loss": 1.0939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 506 + }, + { + "epoch": 0.625154130702836, + "grad_norm": 2.0423951754968064, + "learning_rate": 1.824679606917744e-05, + "loss": 0.9793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 507 + }, + { + "epoch": 0.6263871763255241, + "grad_norm": 2.118412450801216, + "learning_rate": 1.8239332399087136e-05, + "loss": 1.0214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 508 + }, + { + "epoch": 0.6276202219482121, + "grad_norm": 2.273320358602361, + "learning_rate": 1.8231854408976924e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 509 + }, + { + "epoch": 0.6288532675709001, + "grad_norm": 1.9352848815012758, + "learning_rate": 1.822436211184361e-05, + "loss": 1.0194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 510 + }, + { + "epoch": 0.6300863131935882, + "grad_norm": 2.1086688743882713, + "learning_rate": 1.8216855520708854e-05, + "loss": 0.9404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 511 + }, + { + "epoch": 0.6313193588162762, + "grad_norm": 2.268255476351323, + "learning_rate": 1.820933464861917e-05, + "loss": 0.9734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 512 + }, + { + "epoch": 0.6325524044389642, + "grad_norm": 2.3533715256185705, + "learning_rate": 1.8201799508645887e-05, + "loss": 1.1794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 513 + }, + { + "epoch": 0.6337854500616523, + "grad_norm": 2.005798886598211, + "learning_rate": 1.8194250113885134e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 514 + }, + { + "epoch": 0.6350184956843403, + "grad_norm": 1.9276850525809597, + "learning_rate": 1.8186686477457817e-05, + "loss": 1.0122, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 515 + }, + { + "epoch": 0.6362515413070283, + "grad_norm": 2.089756823449061, + "learning_rate": 1.8179108612509587e-05, + "loss": 1.1076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 516 + }, + { + "epoch": 0.6374845869297164, + "grad_norm": 2.047489618950505, + "learning_rate": 1.8171516532210825e-05, + "loss": 1.1261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 517 + }, + { + "epoch": 0.6387176325524044, + "grad_norm": 1.9819427784604287, + "learning_rate": 1.8163910249756634e-05, + "loss": 0.9829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 518 + }, + { + "epoch": 0.6399506781750924, + "grad_norm": 2.1061977657317574, + "learning_rate": 1.815628977836678e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 519 + }, + { + "epoch": 0.6411837237977805, + "grad_norm": 2.0221352739914624, + "learning_rate": 1.8148655131285697e-05, + "loss": 1.0413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 520 + }, + { + "epoch": 0.6424167694204685, + "grad_norm": 2.0126443495430033, + "learning_rate": 1.8141006321782464e-05, + "loss": 1.0737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 521 + }, + { + "epoch": 0.6436498150431565, + "grad_norm": 1.9687053123524947, + "learning_rate": 1.8133343363150762e-05, + "loss": 1.0534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 522 + }, + { + "epoch": 0.6448828606658447, + "grad_norm": 2.1885490221346036, + "learning_rate": 1.812566626870887e-05, + "loss": 1.0746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 523 + }, + { + "epoch": 0.6461159062885327, + "grad_norm": 2.279950249415201, + "learning_rate": 1.8117975051799636e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 524 + }, + { + "epoch": 0.6473489519112207, + "grad_norm": 2.2376288751244986, + "learning_rate": 1.811026972579045e-05, + "loss": 1.0655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 525 + }, + { + "epoch": 0.6485819975339088, + "grad_norm": 2.0687763759549704, + "learning_rate": 1.810255030407323e-05, + "loss": 0.9515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 526 + }, + { + "epoch": 0.6498150431565968, + "grad_norm": 1.9230312514706345, + "learning_rate": 1.8094816800064376e-05, + "loss": 1.0144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 527 + }, + { + "epoch": 0.6510480887792849, + "grad_norm": 1.9036609363684858, + "learning_rate": 1.8087069227204784e-05, + "loss": 1.1092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 528 + }, + { + "epoch": 0.6522811344019729, + "grad_norm": 2.0395351493662104, + "learning_rate": 1.807930759895979e-05, + "loss": 1.0345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 529 + }, + { + "epoch": 0.6535141800246609, + "grad_norm": 2.011352927400422, + "learning_rate": 1.8071531928819158e-05, + "loss": 0.9913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 530 + }, + { + "epoch": 0.654747225647349, + "grad_norm": 1.9222212916878523, + "learning_rate": 1.8063742230297063e-05, + "loss": 1.0643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 531 + }, + { + "epoch": 0.655980271270037, + "grad_norm": 2.203621374645629, + "learning_rate": 1.8055938516932056e-05, + "loss": 1.0912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 532 + }, + { + "epoch": 0.657213316892725, + "grad_norm": 1.9492742413222397, + "learning_rate": 1.804812080228705e-05, + "loss": 0.9613, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 533 + }, + { + "epoch": 0.6584463625154131, + "grad_norm": 2.1694867453328683, + "learning_rate": 1.8040289099949285e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 534 + }, + { + "epoch": 0.6596794081381011, + "grad_norm": 2.0272612869084474, + "learning_rate": 1.8032443423530322e-05, + "loss": 1.0473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 535 + }, + { + "epoch": 0.6609124537607891, + "grad_norm": 2.059356328787384, + "learning_rate": 1.8024583786666007e-05, + "loss": 1.1633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 536 + }, + { + "epoch": 0.6621454993834772, + "grad_norm": 1.9976930500073966, + "learning_rate": 1.8016710203016434e-05, + "loss": 1.0907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 537 + }, + { + "epoch": 0.6633785450061652, + "grad_norm": 2.2046172414513747, + "learning_rate": 1.8008822686265958e-05, + "loss": 1.0435, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 538 + }, + { + "epoch": 0.6646115906288532, + "grad_norm": 1.9172442142747625, + "learning_rate": 1.8000921250123144e-05, + "loss": 1.0905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 539 + }, + { + "epoch": 0.6658446362515413, + "grad_norm": 1.9118211475087676, + "learning_rate": 1.7993005908320737e-05, + "loss": 0.9362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 540 + }, + { + "epoch": 0.6670776818742293, + "grad_norm": 2.2793615054405096, + "learning_rate": 1.7985076674615663e-05, + "loss": 1.1407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 541 + }, + { + "epoch": 0.6683107274969173, + "grad_norm": 2.0968780040667636, + "learning_rate": 1.797713356278899e-05, + "loss": 1.0263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 542 + }, + { + "epoch": 0.6695437731196054, + "grad_norm": 2.130295025454763, + "learning_rate": 1.79691765866459e-05, + "loss": 0.99, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 543 + }, + { + "epoch": 0.6707768187422934, + "grad_norm": 2.050843680123929, + "learning_rate": 1.7961205760015678e-05, + "loss": 0.937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 544 + }, + { + "epoch": 0.6720098643649816, + "grad_norm": 2.00556261298427, + "learning_rate": 1.7953221096751678e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 545 + }, + { + "epoch": 0.6732429099876696, + "grad_norm": 2.104573565750377, + "learning_rate": 1.7945222610731306e-05, + "loss": 0.9851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 546 + }, + { + "epoch": 0.6744759556103576, + "grad_norm": 2.009986716494675, + "learning_rate": 1.793721031585599e-05, + "loss": 0.9842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 547 + }, + { + "epoch": 0.6757090012330457, + "grad_norm": 2.053047844057653, + "learning_rate": 1.792918422605115e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 548 + }, + { + "epoch": 0.6769420468557337, + "grad_norm": 2.0361666532452714, + "learning_rate": 1.7921144355266195e-05, + "loss": 1.0002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 549 + }, + { + "epoch": 0.6781750924784217, + "grad_norm": 2.02324182186846, + "learning_rate": 1.7913090717474476e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 550 + }, + { + "epoch": 0.6794081381011098, + "grad_norm": 1.9162969465242339, + "learning_rate": 1.7905023326673277e-05, + "loss": 0.9256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 551 + }, + { + "epoch": 0.6806411837237978, + "grad_norm": 2.055814853732051, + "learning_rate": 1.789694219688378e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 552 + }, + { + "epoch": 0.6818742293464858, + "grad_norm": 2.0596956840908276, + "learning_rate": 1.7888847342151047e-05, + "loss": 1.096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 553 + }, + { + "epoch": 0.6831072749691739, + "grad_norm": 1.9518592781827393, + "learning_rate": 1.7880738776543994e-05, + "loss": 0.9907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 554 + }, + { + "epoch": 0.6843403205918619, + "grad_norm": 2.0542440716859485, + "learning_rate": 1.787261651415537e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 555 + }, + { + "epoch": 0.6855733662145499, + "grad_norm": 1.9676290362610278, + "learning_rate": 1.786448056910173e-05, + "loss": 0.96, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 556 + }, + { + "epoch": 0.686806411837238, + "grad_norm": 2.0236847394224418, + "learning_rate": 1.7856330955523396e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 557 + }, + { + "epoch": 0.688039457459926, + "grad_norm": 2.166651989793048, + "learning_rate": 1.7848167687584466e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 558 + }, + { + "epoch": 0.689272503082614, + "grad_norm": 2.037401342146827, + "learning_rate": 1.783999077947275e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 559 + }, + { + "epoch": 0.6905055487053021, + "grad_norm": 2.0931620663370465, + "learning_rate": 1.783180024539978e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 560 + }, + { + "epoch": 0.6917385943279901, + "grad_norm": 2.5512794237438032, + "learning_rate": 1.7823596099600767e-05, + "loss": 1.056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 561 + }, + { + "epoch": 0.6929716399506781, + "grad_norm": 2.2145616452367305, + "learning_rate": 1.781537835633457e-05, + "loss": 1.0677, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 562 + }, + { + "epoch": 0.6942046855733662, + "grad_norm": 2.304945412618975, + "learning_rate": 1.7807147029883688e-05, + "loss": 1.0025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 563 + }, + { + "epoch": 0.6954377311960542, + "grad_norm": 2.2714171401510272, + "learning_rate": 1.7798902134554225e-05, + "loss": 0.927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 564 + }, + { + "epoch": 0.6966707768187423, + "grad_norm": 1.9934114654014257, + "learning_rate": 1.7790643684675874e-05, + "loss": 0.9226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 565 + }, + { + "epoch": 0.6979038224414303, + "grad_norm": 2.4001795539004425, + "learning_rate": 1.7782371694601882e-05, + "loss": 0.9898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 566 + }, + { + "epoch": 0.6991368680641183, + "grad_norm": 2.1543466787664944, + "learning_rate": 1.777408617870902e-05, + "loss": 0.8765, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 567 + }, + { + "epoch": 0.7003699136868065, + "grad_norm": 2.125530896627932, + "learning_rate": 1.7765787151397584e-05, + "loss": 0.9383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 568 + }, + { + "epoch": 0.7016029593094945, + "grad_norm": 2.399579117157628, + "learning_rate": 1.775747462709134e-05, + "loss": 1.116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 569 + }, + { + "epoch": 0.7028360049321825, + "grad_norm": 2.3911388414036843, + "learning_rate": 1.7749148620237515e-05, + "loss": 0.983, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 570 + }, + { + "epoch": 0.7040690505548706, + "grad_norm": 2.037432480432792, + "learning_rate": 1.7740809145306773e-05, + "loss": 1.0094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 571 + }, + { + "epoch": 0.7053020961775586, + "grad_norm": 2.0999588564507685, + "learning_rate": 1.773245621679318e-05, + "loss": 1.046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 572 + }, + { + "epoch": 0.7065351418002466, + "grad_norm": 1.9822662619027398, + "learning_rate": 1.7724089849214184e-05, + "loss": 1.0177, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 573 + }, + { + "epoch": 0.7077681874229347, + "grad_norm": 2.296463261709746, + "learning_rate": 1.77157100571106e-05, + "loss": 0.9207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 574 + }, + { + "epoch": 0.7090012330456227, + "grad_norm": 2.0434250630443045, + "learning_rate": 1.7707316855046563e-05, + "loss": 1.0088, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 575 + }, + { + "epoch": 0.7102342786683107, + "grad_norm": 2.067794231415559, + "learning_rate": 1.7698910257609524e-05, + "loss": 1.0033, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 576 + }, + { + "epoch": 0.7114673242909988, + "grad_norm": 1.9884025311346816, + "learning_rate": 1.7690490279410205e-05, + "loss": 1.0395, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 577 + }, + { + "epoch": 0.7127003699136868, + "grad_norm": 2.0663488956854676, + "learning_rate": 1.768205693508259e-05, + "loss": 1.0609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 578 + }, + { + "epoch": 0.7139334155363748, + "grad_norm": 1.9661847340747427, + "learning_rate": 1.7673610239283898e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 579 + }, + { + "epoch": 0.7151664611590629, + "grad_norm": 2.4048347386453135, + "learning_rate": 1.7665150206694543e-05, + "loss": 1.0294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 580 + }, + { + "epoch": 0.7163995067817509, + "grad_norm": 2.049926176641203, + "learning_rate": 1.765667685201813e-05, + "loss": 0.9953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 581 + }, + { + "epoch": 0.717632552404439, + "grad_norm": 2.0425901676211398, + "learning_rate": 1.7648190189981403e-05, + "loss": 1.0155, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 582 + }, + { + "epoch": 0.718865598027127, + "grad_norm": 2.111319944982802, + "learning_rate": 1.7639690235334246e-05, + "loss": 1.012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 583 + }, + { + "epoch": 0.720098643649815, + "grad_norm": 1.9625598394287143, + "learning_rate": 1.7631177002849648e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 584 + }, + { + "epoch": 0.7213316892725031, + "grad_norm": 2.0737975490297025, + "learning_rate": 1.7622650507323658e-05, + "loss": 0.9954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 585 + }, + { + "epoch": 0.7225647348951911, + "grad_norm": 2.2220755560999175, + "learning_rate": 1.7614110763575393e-05, + "loss": 1.093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 586 + }, + { + "epoch": 0.7237977805178791, + "grad_norm": 2.10369867254611, + "learning_rate": 1.7605557786446997e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 587 + }, + { + "epoch": 0.7250308261405672, + "grad_norm": 2.023199550080708, + "learning_rate": 1.7596991590803597e-05, + "loss": 1.0909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 588 + }, + { + "epoch": 0.7262638717632552, + "grad_norm": 2.1139350266949197, + "learning_rate": 1.7588412191533305e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 589 + }, + { + "epoch": 0.7274969173859432, + "grad_norm": 2.1032070896543926, + "learning_rate": 1.7579819603547187e-05, + "loss": 0.9305, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 590 + }, + { + "epoch": 0.7287299630086314, + "grad_norm": 2.1111448680444798, + "learning_rate": 1.7571213841779213e-05, + "loss": 1.0821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 591 + }, + { + "epoch": 0.7299630086313194, + "grad_norm": 2.0849758733816692, + "learning_rate": 1.7562594921186268e-05, + "loss": 0.9747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 592 + }, + { + "epoch": 0.7311960542540074, + "grad_norm": 2.7468709410129573, + "learning_rate": 1.7553962856748098e-05, + "loss": 1.0598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 593 + }, + { + "epoch": 0.7324290998766955, + "grad_norm": 2.297654831376881, + "learning_rate": 1.754531766346729e-05, + "loss": 1.0414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 594 + }, + { + "epoch": 0.7336621454993835, + "grad_norm": 2.0094458450808874, + "learning_rate": 1.7536659356369254e-05, + "loss": 0.977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 595 + }, + { + "epoch": 0.7348951911220715, + "grad_norm": 1.9967204791438438, + "learning_rate": 1.7527987950502196e-05, + "loss": 0.9274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 596 + }, + { + "epoch": 0.7361282367447596, + "grad_norm": 1.9493969416772718, + "learning_rate": 1.7519303460937073e-05, + "loss": 1.0292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 597 + }, + { + "epoch": 0.7373612823674476, + "grad_norm": 2.0023257821799247, + "learning_rate": 1.7510605902767602e-05, + "loss": 0.9574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 598 + }, + { + "epoch": 0.7385943279901356, + "grad_norm": 2.076672200643684, + "learning_rate": 1.7501895291110197e-05, + "loss": 0.9971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 599 + }, + { + "epoch": 0.7398273736128237, + "grad_norm": 2.0612367813173584, + "learning_rate": 1.7493171641103965e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 600 + }, + { + "epoch": 0.7410604192355117, + "grad_norm": 2.16071395185368, + "learning_rate": 1.7484434967910676e-05, + "loss": 1.0001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 601 + }, + { + "epoch": 0.7422934648581998, + "grad_norm": 2.0810754791890664, + "learning_rate": 1.747568528671473e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 602 + }, + { + "epoch": 0.7435265104808878, + "grad_norm": 2.0015256143479525, + "learning_rate": 1.7466922612723133e-05, + "loss": 0.9288, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 603 + }, + { + "epoch": 0.7447595561035758, + "grad_norm": 1.9574522371127243, + "learning_rate": 1.7458146961165484e-05, + "loss": 1.0365, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 604 + }, + { + "epoch": 0.7459926017262639, + "grad_norm": 1.999399085309589, + "learning_rate": 1.7449358347293925e-05, + "loss": 1.0457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 605 + }, + { + "epoch": 0.7472256473489519, + "grad_norm": 2.016778757176555, + "learning_rate": 1.7440556786383132e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 606 + }, + { + "epoch": 0.7484586929716399, + "grad_norm": 2.011184042360808, + "learning_rate": 1.743174229373028e-05, + "loss": 0.9568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 607 + }, + { + "epoch": 0.749691738594328, + "grad_norm": 1.9994468999326396, + "learning_rate": 1.7422914884655027e-05, + "loss": 0.9822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 608 + }, + { + "epoch": 0.750924784217016, + "grad_norm": 2.1373417665891163, + "learning_rate": 1.7414074574499465e-05, + "loss": 0.9128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 609 + }, + { + "epoch": 0.752157829839704, + "grad_norm": 2.414511732797334, + "learning_rate": 1.7405221378628133e-05, + "loss": 1.0866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 610 + }, + { + "epoch": 0.7533908754623921, + "grad_norm": 2.1820009401893605, + "learning_rate": 1.7396355312427935e-05, + "loss": 1.0538, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 611 + }, + { + "epoch": 0.7546239210850801, + "grad_norm": 2.0281507644839163, + "learning_rate": 1.7387476391308168e-05, + "loss": 0.9224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 612 + }, + { + "epoch": 0.7558569667077681, + "grad_norm": 2.387292229834452, + "learning_rate": 1.7378584630700454e-05, + "loss": 0.7561, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 613 + }, + { + "epoch": 0.7570900123304563, + "grad_norm": 2.0036252294718975, + "learning_rate": 1.7369680046058744e-05, + "loss": 0.9816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 614 + }, + { + "epoch": 0.7583230579531443, + "grad_norm": 2.1474265419679566, + "learning_rate": 1.7360762652859275e-05, + "loss": 0.9312, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 615 + }, + { + "epoch": 0.7595561035758323, + "grad_norm": 2.3913311043981387, + "learning_rate": 1.735183246660053e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 616 + }, + { + "epoch": 0.7607891491985204, + "grad_norm": 2.070836177310932, + "learning_rate": 1.7342889502803246e-05, + "loss": 0.8846, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 617 + }, + { + "epoch": 0.7620221948212084, + "grad_norm": 1.9909512311630846, + "learning_rate": 1.7333933777010357e-05, + "loss": 0.9683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 618 + }, + { + "epoch": 0.7632552404438965, + "grad_norm": 2.048373917294825, + "learning_rate": 1.732496530478698e-05, + "loss": 1.021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 619 + }, + { + "epoch": 0.7644882860665845, + "grad_norm": 2.046848859346405, + "learning_rate": 1.7315984101720388e-05, + "loss": 0.9949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 620 + }, + { + "epoch": 0.7657213316892725, + "grad_norm": 2.199270192761742, + "learning_rate": 1.7306990183419972e-05, + "loss": 0.9319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 621 + }, + { + "epoch": 0.7669543773119606, + "grad_norm": 2.2362149652218877, + "learning_rate": 1.729798356551723e-05, + "loss": 1.0579, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 622 + }, + { + "epoch": 0.7681874229346486, + "grad_norm": 2.072113656676678, + "learning_rate": 1.7288964263665735e-05, + "loss": 0.984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 623 + }, + { + "epoch": 0.7694204685573366, + "grad_norm": 2.2058183913371647, + "learning_rate": 1.7279932293541092e-05, + "loss": 1.0727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 624 + }, + { + "epoch": 0.7706535141800247, + "grad_norm": 2.1099223160270104, + "learning_rate": 1.7270887670840935e-05, + "loss": 0.852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 625 + }, + { + "epoch": 0.7718865598027127, + "grad_norm": 2.014275874425781, + "learning_rate": 1.7261830411284884e-05, + "loss": 1.0219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 626 + }, + { + "epoch": 0.7731196054254007, + "grad_norm": 2.1002718881715605, + "learning_rate": 1.7252760530614522e-05, + "loss": 0.9339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 627 + }, + { + "epoch": 0.7743526510480888, + "grad_norm": 2.530650182566485, + "learning_rate": 1.724367804459337e-05, + "loss": 1.1248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 628 + }, + { + "epoch": 0.7755856966707768, + "grad_norm": 2.072394234232429, + "learning_rate": 1.7234582969006847e-05, + "loss": 0.963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 629 + }, + { + "epoch": 0.7768187422934648, + "grad_norm": 2.077401892279152, + "learning_rate": 1.722547531966227e-05, + "loss": 0.9589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 630 + }, + { + "epoch": 0.7780517879161529, + "grad_norm": 2.0461500607118612, + "learning_rate": 1.7216355112388797e-05, + "loss": 0.9892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 631 + }, + { + "epoch": 0.7792848335388409, + "grad_norm": 1.9481531627295827, + "learning_rate": 1.7207222363037414e-05, + "loss": 0.9963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 632 + }, + { + "epoch": 0.7805178791615289, + "grad_norm": 2.0363659895001067, + "learning_rate": 1.7198077087480903e-05, + "loss": 0.9206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 633 + }, + { + "epoch": 0.781750924784217, + "grad_norm": 2.367528975566535, + "learning_rate": 1.718891930161383e-05, + "loss": 1.1255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 634 + }, + { + "epoch": 0.782983970406905, + "grad_norm": 2.007050047111477, + "learning_rate": 1.7179749021352482e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 635 + }, + { + "epoch": 0.7842170160295932, + "grad_norm": 1.9196848202462296, + "learning_rate": 1.7170566262634883e-05, + "loss": 0.9448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 636 + }, + { + "epoch": 0.7854500616522812, + "grad_norm": 2.145941740629715, + "learning_rate": 1.7161371041420723e-05, + "loss": 0.9428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 637 + }, + { + "epoch": 0.7866831072749692, + "grad_norm": 2.170240354212872, + "learning_rate": 1.7152163373691377e-05, + "loss": 1.0323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 638 + }, + { + "epoch": 0.7879161528976573, + "grad_norm": 2.0553179179576015, + "learning_rate": 1.714294327544983e-05, + "loss": 0.8661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 639 + }, + { + "epoch": 0.7891491985203453, + "grad_norm": 2.1113466348102397, + "learning_rate": 1.7133710762720683e-05, + "loss": 1.034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 640 + }, + { + "epoch": 0.7903822441430333, + "grad_norm": 2.146414650333374, + "learning_rate": 1.712446585155011e-05, + "loss": 1.0681, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 641 + }, + { + "epoch": 0.7916152897657214, + "grad_norm": 2.0440039273987227, + "learning_rate": 1.7115208558005834e-05, + "loss": 0.9085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 642 + }, + { + "epoch": 0.7928483353884094, + "grad_norm": 2.3653004518970535, + "learning_rate": 1.7105938898177097e-05, + "loss": 0.8514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 643 + }, + { + "epoch": 0.7940813810110974, + "grad_norm": 2.058806308472887, + "learning_rate": 1.709665688817464e-05, + "loss": 0.9258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 644 + }, + { + "epoch": 0.7953144266337855, + "grad_norm": 2.3406153888995798, + "learning_rate": 1.708736254413066e-05, + "loss": 0.9866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 645 + }, + { + "epoch": 0.7965474722564735, + "grad_norm": 3.2012831686915293, + "learning_rate": 1.70780558821988e-05, + "loss": 1.0916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 646 + }, + { + "epoch": 0.7977805178791615, + "grad_norm": 2.23457477201596, + "learning_rate": 1.7068736918554102e-05, + "loss": 0.9076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 647 + }, + { + "epoch": 0.7990135635018496, + "grad_norm": 2.255512398782474, + "learning_rate": 1.7059405669392993e-05, + "loss": 0.9113, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 648 + }, + { + "epoch": 0.8002466091245376, + "grad_norm": 2.114904513093256, + "learning_rate": 1.7050062150933252e-05, + "loss": 0.9587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 649 + }, + { + "epoch": 0.8014796547472256, + "grad_norm": 1.931643331052564, + "learning_rate": 1.7040706379413983e-05, + "loss": 0.9888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 650 + }, + { + "epoch": 0.8027127003699137, + "grad_norm": 2.0094212900142816, + "learning_rate": 1.7031338371095588e-05, + "loss": 0.9961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 651 + }, + { + "epoch": 0.8039457459926017, + "grad_norm": 2.1928733306625188, + "learning_rate": 1.702195814225973e-05, + "loss": 1.0158, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 652 + }, + { + "epoch": 0.8051787916152897, + "grad_norm": 2.2216734919656544, + "learning_rate": 1.7012565709209317e-05, + "loss": 1.0201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 653 + }, + { + "epoch": 0.8064118372379778, + "grad_norm": 2.0613969675724686, + "learning_rate": 1.700316108826847e-05, + "loss": 0.9941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 654 + }, + { + "epoch": 0.8076448828606658, + "grad_norm": 1.9372031821836655, + "learning_rate": 1.6993744295782482e-05, + "loss": 1.0026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 655 + }, + { + "epoch": 0.8088779284833539, + "grad_norm": 2.077194491232311, + "learning_rate": 1.698431534811781e-05, + "loss": 0.8491, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 656 + }, + { + "epoch": 0.8101109741060419, + "grad_norm": 2.1425857340105754, + "learning_rate": 1.6974874261662035e-05, + "loss": 0.9719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 657 + }, + { + "epoch": 0.8113440197287299, + "grad_norm": 1.8547512115782296, + "learning_rate": 1.6965421052823834e-05, + "loss": 1.0107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 658 + }, + { + "epoch": 0.812577065351418, + "grad_norm": 1.9098479812610942, + "learning_rate": 1.695595573803295e-05, + "loss": 0.95, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 659 + }, + { + "epoch": 0.813810110974106, + "grad_norm": 1.892767796833539, + "learning_rate": 1.6946478333740174e-05, + "loss": 0.86, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 660 + }, + { + "epoch": 0.815043156596794, + "grad_norm": 2.0228841723951043, + "learning_rate": 1.69369888564173e-05, + "loss": 0.9881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 661 + }, + { + "epoch": 0.8162762022194822, + "grad_norm": 2.144845399605076, + "learning_rate": 1.692748732255711e-05, + "loss": 0.8493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 662 + }, + { + "epoch": 0.8175092478421702, + "grad_norm": 2.1710135611234027, + "learning_rate": 1.6917973748673343e-05, + "loss": 0.9356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 663 + }, + { + "epoch": 0.8187422934648582, + "grad_norm": 2.200157070237925, + "learning_rate": 1.6908448151300654e-05, + "loss": 0.9922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 664 + }, + { + "epoch": 0.8199753390875463, + "grad_norm": 2.1046294162556385, + "learning_rate": 1.6898910546994605e-05, + "loss": 0.9714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 665 + }, + { + "epoch": 0.8212083847102343, + "grad_norm": 2.1921813927002622, + "learning_rate": 1.688936095233162e-05, + "loss": 1.0186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 666 + }, + { + "epoch": 0.8224414303329223, + "grad_norm": 2.5290727417605767, + "learning_rate": 1.6879799383908967e-05, + "loss": 0.9147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 667 + }, + { + "epoch": 0.8236744759556104, + "grad_norm": 2.0992091012861605, + "learning_rate": 1.6870225858344715e-05, + "loss": 1.0331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 668 + }, + { + "epoch": 0.8249075215782984, + "grad_norm": 2.228115711337027, + "learning_rate": 1.6860640392277728e-05, + "loss": 1.0372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 669 + }, + { + "epoch": 0.8261405672009864, + "grad_norm": 2.104862934581671, + "learning_rate": 1.6851043002367612e-05, + "loss": 0.9556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 670 + }, + { + "epoch": 0.8273736128236745, + "grad_norm": 2.1153449603435575, + "learning_rate": 1.6841433705294704e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 671 + }, + { + "epoch": 0.8286066584463625, + "grad_norm": 1.9113210816564241, + "learning_rate": 1.6831812517760036e-05, + "loss": 0.9553, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 672 + }, + { + "epoch": 0.8298397040690506, + "grad_norm": 2.146278639653085, + "learning_rate": 1.6822179456485292e-05, + "loss": 0.8621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 673 + }, + { + "epoch": 0.8310727496917386, + "grad_norm": 2.092521119953653, + "learning_rate": 1.681253453821281e-05, + "loss": 0.9576, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 674 + }, + { + "epoch": 0.8323057953144266, + "grad_norm": 2.108757495841555, + "learning_rate": 1.6802877779705526e-05, + "loss": 0.9349, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 675 + }, + { + "epoch": 0.8335388409371147, + "grad_norm": 2.165541838354097, + "learning_rate": 1.679320919774696e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 676 + }, + { + "epoch": 0.8347718865598027, + "grad_norm": 2.0901280484241322, + "learning_rate": 1.6783528809141175e-05, + "loss": 0.9577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 677 + }, + { + "epoch": 0.8360049321824907, + "grad_norm": 2.049770632568499, + "learning_rate": 1.6773836630712757e-05, + "loss": 0.8912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 678 + }, + { + "epoch": 0.8372379778051788, + "grad_norm": 1.9415439366071066, + "learning_rate": 1.6764132679306785e-05, + "loss": 0.9678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 679 + }, + { + "epoch": 0.8384710234278668, + "grad_norm": 2.159517279560615, + "learning_rate": 1.6754416971788795e-05, + "loss": 0.9739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 680 + }, + { + "epoch": 0.8397040690505548, + "grad_norm": 2.0578899051147745, + "learning_rate": 1.674468952504476e-05, + "loss": 0.9187, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 681 + }, + { + "epoch": 0.840937114673243, + "grad_norm": 2.1905094099573135, + "learning_rate": 1.673495035598105e-05, + "loss": 0.9753, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 682 + }, + { + "epoch": 0.842170160295931, + "grad_norm": 2.194631531971183, + "learning_rate": 1.6725199481524418e-05, + "loss": 1.0267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 683 + }, + { + "epoch": 0.843403205918619, + "grad_norm": 2.040276435918069, + "learning_rate": 1.671543691862195e-05, + "loss": 0.9703, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 684 + }, + { + "epoch": 0.8446362515413071, + "grad_norm": 1.9627382699239366, + "learning_rate": 1.6705662684241053e-05, + "loss": 0.8955, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 685 + }, + { + "epoch": 0.8458692971639951, + "grad_norm": 2.288828851032713, + "learning_rate": 1.6695876795369418e-05, + "loss": 0.946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 686 + }, + { + "epoch": 0.8471023427866831, + "grad_norm": 2.1152782339808884, + "learning_rate": 1.668607926901499e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 687 + }, + { + "epoch": 0.8483353884093712, + "grad_norm": 2.798938790414006, + "learning_rate": 1.6676270122205946e-05, + "loss": 0.8988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 688 + }, + { + "epoch": 0.8495684340320592, + "grad_norm": 2.5307103300223766, + "learning_rate": 1.6666449371990644e-05, + "loss": 0.8814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 689 + }, + { + "epoch": 0.8508014796547472, + "grad_norm": 2.2986978033861294, + "learning_rate": 1.6656617035437635e-05, + "loss": 0.9402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 690 + }, + { + "epoch": 0.8520345252774353, + "grad_norm": 2.1645514508732546, + "learning_rate": 1.664677312963558e-05, + "loss": 0.9551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 691 + }, + { + "epoch": 0.8532675709001233, + "grad_norm": 2.144781636504613, + "learning_rate": 1.6636917671693264e-05, + "loss": 1.0505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 692 + }, + { + "epoch": 0.8545006165228114, + "grad_norm": 2.068308383986026, + "learning_rate": 1.6627050678739544e-05, + "loss": 0.9237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 693 + }, + { + "epoch": 0.8557336621454994, + "grad_norm": 2.4201265499871325, + "learning_rate": 1.6617172167923327e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 694 + }, + { + "epoch": 0.8569667077681874, + "grad_norm": 1.9743124171175193, + "learning_rate": 1.660728215641353e-05, + "loss": 0.9649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 695 + }, + { + "epoch": 0.8581997533908755, + "grad_norm": 2.4976283410874847, + "learning_rate": 1.6597380661399077e-05, + "loss": 0.9445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 696 + }, + { + "epoch": 0.8594327990135635, + "grad_norm": 2.2344847337230855, + "learning_rate": 1.6587467700088834e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 697 + }, + { + "epoch": 0.8606658446362515, + "grad_norm": 2.1366592393550343, + "learning_rate": 1.6577543289711597e-05, + "loss": 0.8845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 698 + }, + { + "epoch": 0.8618988902589396, + "grad_norm": 2.0879117181580753, + "learning_rate": 1.6567607447516064e-05, + "loss": 0.7972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 699 + }, + { + "epoch": 0.8631319358816276, + "grad_norm": 2.117890235500837, + "learning_rate": 1.6557660190770803e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 700 + }, + { + "epoch": 0.8643649815043156, + "grad_norm": 2.227804015025162, + "learning_rate": 1.654770153676422e-05, + "loss": 0.9105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 701 + }, + { + "epoch": 0.8655980271270037, + "grad_norm": 2.262676552209609, + "learning_rate": 1.6537731502804525e-05, + "loss": 0.9706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 702 + }, + { + "epoch": 0.8668310727496917, + "grad_norm": 2.146446661141531, + "learning_rate": 1.6527750106219715e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 703 + }, + { + "epoch": 0.8680641183723797, + "grad_norm": 2.08167189099969, + "learning_rate": 1.6517757364357526e-05, + "loss": 0.9381, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 704 + }, + { + "epoch": 0.8692971639950678, + "grad_norm": 1.9310579355796305, + "learning_rate": 1.6507753294585417e-05, + "loss": 0.9047, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 705 + }, + { + "epoch": 0.8705302096177558, + "grad_norm": 2.0325358606748978, + "learning_rate": 1.6497737914290535e-05, + "loss": 0.9094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 706 + }, + { + "epoch": 0.8717632552404438, + "grad_norm": 2.0543982954124966, + "learning_rate": 1.648771124087969e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 707 + }, + { + "epoch": 0.872996300863132, + "grad_norm": 2.295806181212224, + "learning_rate": 1.64776732917793e-05, + "loss": 0.9444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 708 + }, + { + "epoch": 0.87422934648582, + "grad_norm": 2.1899960866925303, + "learning_rate": 1.646762408443541e-05, + "loss": 1.0083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 709 + }, + { + "epoch": 0.8754623921085081, + "grad_norm": 2.040526145894979, + "learning_rate": 1.6457563636313598e-05, + "loss": 0.8215, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 710 + }, + { + "epoch": 0.8766954377311961, + "grad_norm": 1.9967218569140772, + "learning_rate": 1.6447491964899008e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 711 + }, + { + "epoch": 0.8779284833538841, + "grad_norm": 2.131774886953309, + "learning_rate": 1.6437409087696283e-05, + "loss": 0.8786, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 712 + }, + { + "epoch": 0.8791615289765722, + "grad_norm": 2.0642410512824414, + "learning_rate": 1.6427315022229525e-05, + "loss": 0.8831, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 713 + }, + { + "epoch": 0.8803945745992602, + "grad_norm": 2.2793436000912255, + "learning_rate": 1.64172097860423e-05, + "loss": 0.8413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 714 + }, + { + "epoch": 0.8816276202219482, + "grad_norm": 1.9501864497105927, + "learning_rate": 1.640709339669758e-05, + "loss": 0.8856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 715 + }, + { + "epoch": 0.8828606658446363, + "grad_norm": 2.0279122102292977, + "learning_rate": 1.6396965871777727e-05, + "loss": 0.8835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 716 + }, + { + "epoch": 0.8840937114673243, + "grad_norm": 2.181455569803073, + "learning_rate": 1.638682722888445e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 717 + }, + { + "epoch": 0.8853267570900123, + "grad_norm": 2.1877848027614313, + "learning_rate": 1.637667748563879e-05, + "loss": 0.8314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 718 + }, + { + "epoch": 0.8865598027127004, + "grad_norm": 2.0901252578597713, + "learning_rate": 1.6366516659681067e-05, + "loss": 0.9333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 719 + }, + { + "epoch": 0.8877928483353884, + "grad_norm": 2.1846108211938673, + "learning_rate": 1.6356344768670875e-05, + "loss": 0.9309, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 720 + }, + { + "epoch": 0.8890258939580764, + "grad_norm": 2.600109718803851, + "learning_rate": 1.6346161830287034e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 721 + }, + { + "epoch": 0.8902589395807645, + "grad_norm": 2.2057915960500276, + "learning_rate": 1.6335967862227567e-05, + "loss": 0.9764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 722 + }, + { + "epoch": 0.8914919852034525, + "grad_norm": 2.1009909758008267, + "learning_rate": 1.632576288220966e-05, + "loss": 0.8741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 723 + }, + { + "epoch": 0.8927250308261405, + "grad_norm": 2.1617838023881, + "learning_rate": 1.6315546907969652e-05, + "loss": 0.9573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 724 + }, + { + "epoch": 0.8939580764488286, + "grad_norm": 2.091522410165622, + "learning_rate": 1.6305319957262972e-05, + "loss": 0.9534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 725 + }, + { + "epoch": 0.8951911220715166, + "grad_norm": 2.2845231322682418, + "learning_rate": 1.6295082047864138e-05, + "loss": 0.9156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 726 + }, + { + "epoch": 0.8964241676942046, + "grad_norm": 2.062614641816067, + "learning_rate": 1.6284833197566712e-05, + "loss": 0.9539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 727 + }, + { + "epoch": 0.8976572133168927, + "grad_norm": 2.157165674006361, + "learning_rate": 1.627457342418327e-05, + "loss": 0.9926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 728 + }, + { + "epoch": 0.8988902589395807, + "grad_norm": 2.039737473184125, + "learning_rate": 1.6264302745545374e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 729 + }, + { + "epoch": 0.9001233045622689, + "grad_norm": 2.057800120641984, + "learning_rate": 1.625402117950354e-05, + "loss": 0.9139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 730 + }, + { + "epoch": 0.9013563501849569, + "grad_norm": 2.240321306869684, + "learning_rate": 1.62437287439272e-05, + "loss": 1.0758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 731 + }, + { + "epoch": 0.9025893958076449, + "grad_norm": 2.0947466285552436, + "learning_rate": 1.6233425456704686e-05, + "loss": 0.8467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 732 + }, + { + "epoch": 0.903822441430333, + "grad_norm": 2.061121182714553, + "learning_rate": 1.6223111335743187e-05, + "loss": 0.9093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 733 + }, + { + "epoch": 0.905055487053021, + "grad_norm": 2.209914581799182, + "learning_rate": 1.6212786398968717e-05, + "loss": 0.8805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 734 + }, + { + "epoch": 0.906288532675709, + "grad_norm": 2.468473842682747, + "learning_rate": 1.6202450664326093e-05, + "loss": 0.9013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 735 + }, + { + "epoch": 0.9075215782983971, + "grad_norm": 2.3365951628381056, + "learning_rate": 1.61921041497789e-05, + "loss": 0.9112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 736 + }, + { + "epoch": 0.9087546239210851, + "grad_norm": 2.474277588242769, + "learning_rate": 1.6181746873309455e-05, + "loss": 0.9186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 737 + }, + { + "epoch": 0.9099876695437731, + "grad_norm": 2.0716027082353228, + "learning_rate": 1.6171378852918778e-05, + "loss": 0.9486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 738 + }, + { + "epoch": 0.9112207151664612, + "grad_norm": 2.165213004434134, + "learning_rate": 1.6161000106626564e-05, + "loss": 0.9726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 739 + }, + { + "epoch": 0.9124537607891492, + "grad_norm": 2.4218033902151475, + "learning_rate": 1.615061065247115e-05, + "loss": 0.8259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 740 + }, + { + "epoch": 0.9136868064118372, + "grad_norm": 1.970928985580259, + "learning_rate": 1.614021050850949e-05, + "loss": 0.9788, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 741 + }, + { + "epoch": 0.9149198520345253, + "grad_norm": 2.2639971055328934, + "learning_rate": 1.6129799692817097e-05, + "loss": 0.8745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 742 + }, + { + "epoch": 0.9161528976572133, + "grad_norm": 2.5602631751385028, + "learning_rate": 1.611937822348806e-05, + "loss": 0.9512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 743 + }, + { + "epoch": 0.9173859432799013, + "grad_norm": 2.22841865866248, + "learning_rate": 1.6108946118634957e-05, + "loss": 0.805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 744 + }, + { + "epoch": 0.9186189889025894, + "grad_norm": 2.4182686550221626, + "learning_rate": 1.6098503396388868e-05, + "loss": 1.0202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 745 + }, + { + "epoch": 0.9198520345252774, + "grad_norm": 2.2298122692949645, + "learning_rate": 1.608805007489932e-05, + "loss": 0.8972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 746 + }, + { + "epoch": 0.9210850801479655, + "grad_norm": 2.362160156268979, + "learning_rate": 1.607758617233426e-05, + "loss": 0.9039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 747 + }, + { + "epoch": 0.9223181257706535, + "grad_norm": 2.1106519602619533, + "learning_rate": 1.6067111706880027e-05, + "loss": 0.8318, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 748 + }, + { + "epoch": 0.9235511713933415, + "grad_norm": 2.0825477479243424, + "learning_rate": 1.605662669674132e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 749 + }, + { + "epoch": 0.9247842170160296, + "grad_norm": 2.036239449814601, + "learning_rate": 1.6046131160141167e-05, + "loss": 0.8727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 750 + }, + { + "epoch": 0.9260172626387176, + "grad_norm": 2.1703406452205787, + "learning_rate": 1.603562511532088e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 751 + }, + { + "epoch": 0.9272503082614056, + "grad_norm": 2.1398345899160143, + "learning_rate": 1.6025108580540046e-05, + "loss": 0.8691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 752 + }, + { + "epoch": 0.9284833538840938, + "grad_norm": 2.0888823283975375, + "learning_rate": 1.601458157407648e-05, + "loss": 0.8547, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 753 + }, + { + "epoch": 0.9297163995067818, + "grad_norm": 2.269523743217552, + "learning_rate": 1.6004044114226193e-05, + "loss": 1.0141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 754 + }, + { + "epoch": 0.9309494451294698, + "grad_norm": 2.0091635625820063, + "learning_rate": 1.5993496219303367e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 755 + }, + { + "epoch": 0.9321824907521579, + "grad_norm": 2.1386095902203612, + "learning_rate": 1.5982937907640327e-05, + "loss": 0.9388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 756 + }, + { + "epoch": 0.9334155363748459, + "grad_norm": 2.1260034823775467, + "learning_rate": 1.5972369197587486e-05, + "loss": 0.9153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 757 + }, + { + "epoch": 0.9346485819975339, + "grad_norm": 2.1225283234506156, + "learning_rate": 1.5961790107513343e-05, + "loss": 1.0178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 758 + }, + { + "epoch": 0.935881627620222, + "grad_norm": 2.0422192666840795, + "learning_rate": 1.595120065580444e-05, + "loss": 0.9856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 759 + }, + { + "epoch": 0.93711467324291, + "grad_norm": 2.051963293225036, + "learning_rate": 1.5940600860865315e-05, + "loss": 0.8559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 760 + }, + { + "epoch": 0.938347718865598, + "grad_norm": 2.127612803214104, + "learning_rate": 1.592999074111849e-05, + "loss": 0.8298, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 761 + }, + { + "epoch": 0.9395807644882861, + "grad_norm": 2.286554941831041, + "learning_rate": 1.591937031500443e-05, + "loss": 0.8796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 762 + }, + { + "epoch": 0.9408138101109741, + "grad_norm": 2.282412038316439, + "learning_rate": 1.5908739600981517e-05, + "loss": 1.022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 763 + }, + { + "epoch": 0.9420468557336621, + "grad_norm": 2.197857160398704, + "learning_rate": 1.5898098617526006e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 764 + }, + { + "epoch": 0.9432799013563502, + "grad_norm": 2.1702846778760607, + "learning_rate": 1.5887447383132008e-05, + "loss": 0.9507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 765 + }, + { + "epoch": 0.9445129469790382, + "grad_norm": 2.045700294662405, + "learning_rate": 1.587678591631144e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 766 + }, + { + "epoch": 0.9457459926017263, + "grad_norm": 2.426595174168101, + "learning_rate": 1.5866114235594014e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 767 + }, + { + "epoch": 0.9469790382244143, + "grad_norm": 2.085172001225346, + "learning_rate": 1.585543235952718e-05, + "loss": 0.9166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 768 + }, + { + "epoch": 0.9482120838471023, + "grad_norm": 2.1181076484114665, + "learning_rate": 1.5844740306676133e-05, + "loss": 0.8829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 769 + }, + { + "epoch": 0.9494451294697904, + "grad_norm": 2.64845194344853, + "learning_rate": 1.583403809562373e-05, + "loss": 0.8927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 770 + }, + { + "epoch": 0.9506781750924784, + "grad_norm": 2.4929599774250186, + "learning_rate": 1.5823325744970488e-05, + "loss": 0.9407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 771 + }, + { + "epoch": 0.9519112207151664, + "grad_norm": 2.368031870072283, + "learning_rate": 1.581260327333456e-05, + "loss": 0.9463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 772 + }, + { + "epoch": 0.9531442663378545, + "grad_norm": 2.136177985699207, + "learning_rate": 1.580187069935167e-05, + "loss": 0.8571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 773 + }, + { + "epoch": 0.9543773119605425, + "grad_norm": 2.128181510143405, + "learning_rate": 1.5791128041675114e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 774 + }, + { + "epoch": 0.9556103575832305, + "grad_norm": 2.220458022640665, + "learning_rate": 1.5780375318975715e-05, + "loss": 0.8445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 775 + }, + { + "epoch": 0.9568434032059187, + "grad_norm": 2.098330673951412, + "learning_rate": 1.5769612549941777e-05, + "loss": 0.964, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 776 + }, + { + "epoch": 0.9580764488286067, + "grad_norm": 2.094864541454194, + "learning_rate": 1.575883975327907e-05, + "loss": 0.8254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 777 + }, + { + "epoch": 0.9593094944512947, + "grad_norm": 2.3159416415455047, + "learning_rate": 1.5748056947710804e-05, + "loss": 0.9472, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 778 + }, + { + "epoch": 0.9605425400739828, + "grad_norm": 1.9950192842009455, + "learning_rate": 1.573726415197756e-05, + "loss": 0.7965, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 779 + }, + { + "epoch": 0.9617755856966708, + "grad_norm": 2.006344851533844, + "learning_rate": 1.5726461384837306e-05, + "loss": 0.8203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 780 + }, + { + "epoch": 0.9630086313193588, + "grad_norm": 2.2545503988149007, + "learning_rate": 1.5715648665065324e-05, + "loss": 0.9282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 781 + }, + { + "epoch": 0.9642416769420469, + "grad_norm": 2.2372695819528294, + "learning_rate": 1.5704826011454203e-05, + "loss": 0.9608, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 782 + }, + { + "epoch": 0.9654747225647349, + "grad_norm": 2.1710522433750357, + "learning_rate": 1.5693993442813792e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 783 + }, + { + "epoch": 0.966707768187423, + "grad_norm": 2.400731212803864, + "learning_rate": 1.568315097797118e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 784 + }, + { + "epoch": 0.967940813810111, + "grad_norm": 2.096674414566688, + "learning_rate": 1.5672298635770645e-05, + "loss": 0.6728, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 785 + }, + { + "epoch": 0.969173859432799, + "grad_norm": 2.2214704493931205, + "learning_rate": 1.566143643507364e-05, + "loss": 1.0617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 786 + }, + { + "epoch": 0.9704069050554871, + "grad_norm": 2.21278029237075, + "learning_rate": 1.5650564394758743e-05, + "loss": 0.8487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 787 + }, + { + "epoch": 0.9716399506781751, + "grad_norm": 2.3400569095138763, + "learning_rate": 1.5639682533721643e-05, + "loss": 0.9532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 788 + }, + { + "epoch": 0.9728729963008631, + "grad_norm": 2.1937723298850784, + "learning_rate": 1.5628790870875093e-05, + "loss": 0.9272, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 789 + }, + { + "epoch": 0.9741060419235512, + "grad_norm": 2.117801922125244, + "learning_rate": 1.561788942514889e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 790 + }, + { + "epoch": 0.9753390875462392, + "grad_norm": 2.436307402269952, + "learning_rate": 1.560697821548982e-05, + "loss": 0.8599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 791 + }, + { + "epoch": 0.9765721331689272, + "grad_norm": 2.2722775653462848, + "learning_rate": 1.5596057260861645e-05, + "loss": 0.8503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 792 + }, + { + "epoch": 0.9778051787916153, + "grad_norm": 2.22960384352197, + "learning_rate": 1.5585126580245066e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 793 + }, + { + "epoch": 0.9790382244143033, + "grad_norm": 2.1141630678799275, + "learning_rate": 1.557418619263768e-05, + "loss": 0.8418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 794 + }, + { + "epoch": 0.9802712700369913, + "grad_norm": 2.4590856272408224, + "learning_rate": 1.5563236117053965e-05, + "loss": 0.9197, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 795 + }, + { + "epoch": 0.9815043156596794, + "grad_norm": 2.678264319346431, + "learning_rate": 1.5552276372525237e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 796 + }, + { + "epoch": 0.9827373612823674, + "grad_norm": 2.262445020658096, + "learning_rate": 1.554130697809961e-05, + "loss": 0.9101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 797 + }, + { + "epoch": 0.9839704069050554, + "grad_norm": 2.327233920544754, + "learning_rate": 1.5530327952841965e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 798 + }, + { + "epoch": 0.9852034525277436, + "grad_norm": 2.1249691788358334, + "learning_rate": 1.551933931583393e-05, + "loss": 0.8824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 799 + }, + { + "epoch": 0.9864364981504316, + "grad_norm": 2.0648396343207325, + "learning_rate": 1.5508341086173843e-05, + "loss": 0.8895, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 800 + }, + { + "epoch": 0.9876695437731196, + "grad_norm": 2.109919443242373, + "learning_rate": 1.5497333282976698e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 801 + }, + { + "epoch": 0.9889025893958077, + "grad_norm": 2.3306692057820797, + "learning_rate": 1.5486315925374146e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 802 + }, + { + "epoch": 0.9901356350184957, + "grad_norm": 2.365009122779881, + "learning_rate": 1.547528903251443e-05, + "loss": 1.0604, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 803 + }, + { + "epoch": 0.9913686806411838, + "grad_norm": 2.2289048394713262, + "learning_rate": 1.5464252623562375e-05, + "loss": 0.8141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 804 + }, + { + "epoch": 0.9926017262638718, + "grad_norm": 2.457422539818384, + "learning_rate": 1.5453206717699334e-05, + "loss": 1.0228, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 805 + }, + { + "epoch": 0.9938347718865598, + "grad_norm": 2.0659430518171558, + "learning_rate": 1.544215133412318e-05, + "loss": 0.8641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 806 + }, + { + "epoch": 0.9950678175092479, + "grad_norm": 2.154728199437817, + "learning_rate": 1.543108649204824e-05, + "loss": 0.8029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 807 + }, + { + "epoch": 0.9963008631319359, + "grad_norm": 2.3679631548982196, + "learning_rate": 1.54200122107053e-05, + "loss": 0.866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 808 + }, + { + "epoch": 0.9975339087546239, + "grad_norm": 2.218685707693942, + "learning_rate": 1.540892850934154e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 809 + }, + { + "epoch": 0.998766954377312, + "grad_norm": 2.3274870898977187, + "learning_rate": 1.5397835407220508e-05, + "loss": 0.9724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 810 + }, + { + "epoch": 1.0, + "grad_norm": 2.159965188324255, + "learning_rate": 1.53867329236221e-05, + "loss": 0.762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 811 + }, + { + "epoch": 1.0012330456226881, + "grad_norm": 2.8883947578149782, + "learning_rate": 1.5375621077842517e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 812 + }, + { + "epoch": 1.002466091245376, + "grad_norm": 2.1490979216413963, + "learning_rate": 1.5364499889194225e-05, + "loss": 0.8653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 813 + }, + { + "epoch": 1.0036991368680641, + "grad_norm": 2.2100133362019787, + "learning_rate": 1.5353369377005928e-05, + "loss": 0.8871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 814 + }, + { + "epoch": 1.0049321824907522, + "grad_norm": 2.1549827645085444, + "learning_rate": 1.5342229560622545e-05, + "loss": 0.9146, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 815 + }, + { + "epoch": 1.0061652281134401, + "grad_norm": 2.0929823364178155, + "learning_rate": 1.5331080459405153e-05, + "loss": 0.8513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 816 + }, + { + "epoch": 1.0073982737361282, + "grad_norm": 2.220368656816853, + "learning_rate": 1.5319922092730974e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 817 + }, + { + "epoch": 1.0086313193588163, + "grad_norm": 2.1262847612815197, + "learning_rate": 1.5308754479993326e-05, + "loss": 0.8457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 818 + }, + { + "epoch": 1.0098643649815042, + "grad_norm": 2.1791530750144443, + "learning_rate": 1.5297577640601607e-05, + "loss": 0.7645, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 819 + }, + { + "epoch": 1.0110974106041923, + "grad_norm": 2.1726539474102426, + "learning_rate": 1.5286391593981243e-05, + "loss": 0.8684, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 820 + }, + { + "epoch": 1.0123304562268804, + "grad_norm": 2.193333692016209, + "learning_rate": 1.5275196359573667e-05, + "loss": 0.7566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 821 + }, + { + "epoch": 1.0135635018495683, + "grad_norm": 2.9212088946802623, + "learning_rate": 1.526399195683627e-05, + "loss": 1.0019, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 822 + }, + { + "epoch": 1.0147965474722564, + "grad_norm": 2.0580823212102417, + "learning_rate": 1.52527784052424e-05, + "loss": 0.9646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 823 + }, + { + "epoch": 1.0160295930949446, + "grad_norm": 2.19343597586527, + "learning_rate": 1.5241555724281283e-05, + "loss": 0.9453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 824 + }, + { + "epoch": 1.0172626387176325, + "grad_norm": 2.294578049001136, + "learning_rate": 1.523032393345802e-05, + "loss": 1.0273, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 825 + }, + { + "epoch": 1.0184956843403206, + "grad_norm": 1.9511593069959263, + "learning_rate": 1.5219083052293546e-05, + "loss": 0.8161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 826 + }, + { + "epoch": 1.0197287299630087, + "grad_norm": 2.421213793478604, + "learning_rate": 1.5207833100324595e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 827 + }, + { + "epoch": 1.0209617755856966, + "grad_norm": 2.1254569627132422, + "learning_rate": 1.5196574097103669e-05, + "loss": 1.0024, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 828 + }, + { + "epoch": 1.0221948212083847, + "grad_norm": 2.1741883221339213, + "learning_rate": 1.5185306062198993e-05, + "loss": 0.8694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 829 + }, + { + "epoch": 1.0234278668310728, + "grad_norm": 2.340571355784493, + "learning_rate": 1.5174029015194496e-05, + "loss": 0.9235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 830 + }, + { + "epoch": 1.024660912453761, + "grad_norm": 2.173793908678777, + "learning_rate": 1.5162742975689769e-05, + "loss": 0.8448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 831 + }, + { + "epoch": 1.0258939580764488, + "grad_norm": 2.149214329172728, + "learning_rate": 1.5151447963300027e-05, + "loss": 0.8388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 832 + }, + { + "epoch": 1.027127003699137, + "grad_norm": 2.059108722897467, + "learning_rate": 1.5140143997656092e-05, + "loss": 0.7303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 833 + }, + { + "epoch": 1.028360049321825, + "grad_norm": 2.2165337335607536, + "learning_rate": 1.512883109840433e-05, + "loss": 0.873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 834 + }, + { + "epoch": 1.029593094944513, + "grad_norm": 2.0093629352145923, + "learning_rate": 1.511750928520665e-05, + "loss": 0.7804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 835 + }, + { + "epoch": 1.030826140567201, + "grad_norm": 1.9963120627280608, + "learning_rate": 1.5106178577740442e-05, + "loss": 0.8642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 836 + }, + { + "epoch": 1.0320591861898891, + "grad_norm": 2.1963002199807384, + "learning_rate": 1.5094838995698555e-05, + "loss": 0.8688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 837 + }, + { + "epoch": 1.033292231812577, + "grad_norm": 2.1874228944498473, + "learning_rate": 1.5083490558789275e-05, + "loss": 0.9912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 838 + }, + { + "epoch": 1.0345252774352651, + "grad_norm": 2.12533056093305, + "learning_rate": 1.5072133286736261e-05, + "loss": 0.7925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 839 + }, + { + "epoch": 1.0357583230579532, + "grad_norm": 2.5000691541559745, + "learning_rate": 1.5060767199278536e-05, + "loss": 0.9544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 840 + }, + { + "epoch": 1.0369913686806411, + "grad_norm": 2.0622785615015125, + "learning_rate": 1.5049392316170443e-05, + "loss": 0.8433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 841 + }, + { + "epoch": 1.0382244143033292, + "grad_norm": 2.2009668761322563, + "learning_rate": 1.5038008657181616e-05, + "loss": 0.8157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 842 + }, + { + "epoch": 1.0394574599260173, + "grad_norm": 2.0959248263408385, + "learning_rate": 1.5026616242096932e-05, + "loss": 0.9167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 843 + }, + { + "epoch": 1.0406905055487052, + "grad_norm": 2.1893631450779893, + "learning_rate": 1.5015215090716493e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 844 + }, + { + "epoch": 1.0419235511713933, + "grad_norm": 2.077397878791412, + "learning_rate": 1.5003805222855584e-05, + "loss": 0.8383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 845 + }, + { + "epoch": 1.0431565967940815, + "grad_norm": 2.4055084927944623, + "learning_rate": 1.4992386658344643e-05, + "loss": 0.8404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 846 + }, + { + "epoch": 1.0443896424167693, + "grad_norm": 2.171777480987558, + "learning_rate": 1.4980959417029208e-05, + "loss": 0.7997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 847 + }, + { + "epoch": 1.0456226880394575, + "grad_norm": 2.157645250920109, + "learning_rate": 1.4969523518769917e-05, + "loss": 0.8738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 848 + }, + { + "epoch": 1.0468557336621456, + "grad_norm": 2.396639754777913, + "learning_rate": 1.4958078983442441e-05, + "loss": 0.6992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 849 + }, + { + "epoch": 1.0480887792848335, + "grad_norm": 2.4186334460214765, + "learning_rate": 1.4946625830937472e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 850 + }, + { + "epoch": 1.0493218249075216, + "grad_norm": 2.2977540698502996, + "learning_rate": 1.4935164081160663e-05, + "loss": 0.8683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 851 + }, + { + "epoch": 1.0505548705302097, + "grad_norm": 2.1379865919168, + "learning_rate": 1.4923693754032626e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 852 + }, + { + "epoch": 1.0517879161528976, + "grad_norm": 2.259076969791956, + "learning_rate": 1.491221486948887e-05, + "loss": 0.7691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 853 + }, + { + "epoch": 1.0530209617755857, + "grad_norm": 2.1377806626548357, + "learning_rate": 1.490072744747978e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 854 + }, + { + "epoch": 1.0542540073982738, + "grad_norm": 2.1044084023956167, + "learning_rate": 1.4889231507970581e-05, + "loss": 0.8401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 855 + }, + { + "epoch": 1.0554870530209617, + "grad_norm": 2.1271140055745286, + "learning_rate": 1.4877727070941302e-05, + "loss": 0.8084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 856 + }, + { + "epoch": 1.0567200986436498, + "grad_norm": 2.1718066261306817, + "learning_rate": 1.4866214156386734e-05, + "loss": 0.9014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 857 + }, + { + "epoch": 1.057953144266338, + "grad_norm": 2.1176596897595714, + "learning_rate": 1.4854692784316409e-05, + "loss": 0.7752, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 858 + }, + { + "epoch": 1.0591861898890258, + "grad_norm": 2.171891486084014, + "learning_rate": 1.4843162974754555e-05, + "loss": 0.6988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 859 + }, + { + "epoch": 1.060419235511714, + "grad_norm": 2.0055850881304718, + "learning_rate": 1.4831624747740065e-05, + "loss": 0.7928, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 860 + }, + { + "epoch": 1.061652281134402, + "grad_norm": 2.1272055457979624, + "learning_rate": 1.4820078123326464e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 861 + }, + { + "epoch": 1.06288532675709, + "grad_norm": 2.2409166254632416, + "learning_rate": 1.4808523121581871e-05, + "loss": 0.8428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 862 + }, + { + "epoch": 1.064118372379778, + "grad_norm": 2.167571295890406, + "learning_rate": 1.4796959762588957e-05, + "loss": 0.9326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 863 + }, + { + "epoch": 1.0653514180024661, + "grad_norm": 1.9896497336935064, + "learning_rate": 1.4785388066444928e-05, + "loss": 0.7178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 864 + }, + { + "epoch": 1.066584463625154, + "grad_norm": 1.9782830220589522, + "learning_rate": 1.4773808053261478e-05, + "loss": 0.8734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 865 + }, + { + "epoch": 1.0678175092478421, + "grad_norm": 2.1094953563881447, + "learning_rate": 1.4762219743164752e-05, + "loss": 0.72, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 866 + }, + { + "epoch": 1.0690505548705302, + "grad_norm": 2.172378157117765, + "learning_rate": 1.4750623156295316e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 867 + }, + { + "epoch": 1.0702836004932181, + "grad_norm": 2.777083857064727, + "learning_rate": 1.473901831280813e-05, + "loss": 0.887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 868 + }, + { + "epoch": 1.0715166461159062, + "grad_norm": 2.773482686862429, + "learning_rate": 1.4727405232872489e-05, + "loss": 0.9897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 869 + }, + { + "epoch": 1.0727496917385944, + "grad_norm": 2.303987298346897, + "learning_rate": 1.4715783936672017e-05, + "loss": 0.8267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 870 + }, + { + "epoch": 1.0739827373612825, + "grad_norm": 2.5090103249366718, + "learning_rate": 1.4704154444404606e-05, + "loss": 0.8297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 871 + }, + { + "epoch": 1.0752157829839704, + "grad_norm": 2.1804014393140334, + "learning_rate": 1.4692516776282404e-05, + "loss": 0.8673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 872 + }, + { + "epoch": 1.0764488286066585, + "grad_norm": 2.114741425345962, + "learning_rate": 1.4680870952531758e-05, + "loss": 0.7878, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 873 + }, + { + "epoch": 1.0776818742293466, + "grad_norm": 2.1487592065744043, + "learning_rate": 1.4669216993393195e-05, + "loss": 0.702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 874 + }, + { + "epoch": 1.0789149198520345, + "grad_norm": 2.250699104710453, + "learning_rate": 1.4657554919121388e-05, + "loss": 0.7805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 875 + }, + { + "epoch": 1.0801479654747226, + "grad_norm": 2.652130922890576, + "learning_rate": 1.46458847499851e-05, + "loss": 0.8594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 876 + }, + { + "epoch": 1.0813810110974107, + "grad_norm": 2.618911622711661, + "learning_rate": 1.4634206506267172e-05, + "loss": 0.974, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 877 + }, + { + "epoch": 1.0826140567200986, + "grad_norm": 2.104711420819804, + "learning_rate": 1.462252020826448e-05, + "loss": 0.7511, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 878 + }, + { + "epoch": 1.0838471023427867, + "grad_norm": 2.181334503374274, + "learning_rate": 1.4610825876287892e-05, + "loss": 0.8632, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 879 + }, + { + "epoch": 1.0850801479654748, + "grad_norm": 2.2281952334339126, + "learning_rate": 1.4599123530662242e-05, + "loss": 0.7623, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 880 + }, + { + "epoch": 1.0863131935881627, + "grad_norm": 2.2779707650922254, + "learning_rate": 1.4587413191726296e-05, + "loss": 0.7533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 881 + }, + { + "epoch": 1.0875462392108508, + "grad_norm": 2.227912879258747, + "learning_rate": 1.4575694879832704e-05, + "loss": 0.9742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 882 + }, + { + "epoch": 1.088779284833539, + "grad_norm": 2.200751986221906, + "learning_rate": 1.4563968615347983e-05, + "loss": 0.9859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 883 + }, + { + "epoch": 1.0900123304562268, + "grad_norm": 2.1474005643679503, + "learning_rate": 1.455223441865246e-05, + "loss": 0.885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 884 + }, + { + "epoch": 1.091245376078915, + "grad_norm": 2.5635895574555967, + "learning_rate": 1.4540492310140262e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 885 + }, + { + "epoch": 1.092478421701603, + "grad_norm": 2.086551890894098, + "learning_rate": 1.4528742310219251e-05, + "loss": 0.7838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 886 + }, + { + "epoch": 1.093711467324291, + "grad_norm": 2.100577896933077, + "learning_rate": 1.4516984439311017e-05, + "loss": 0.8245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 887 + }, + { + "epoch": 1.094944512946979, + "grad_norm": 2.0824253868885427, + "learning_rate": 1.4505218717850826e-05, + "loss": 0.8121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 888 + }, + { + "epoch": 1.0961775585696671, + "grad_norm": 2.424807773941655, + "learning_rate": 1.449344516628759e-05, + "loss": 0.6852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 889 + }, + { + "epoch": 1.097410604192355, + "grad_norm": 2.2203614431330587, + "learning_rate": 1.4481663805083823e-05, + "loss": 0.9022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 890 + }, + { + "epoch": 1.0986436498150431, + "grad_norm": 2.194684728317858, + "learning_rate": 1.4469874654715618e-05, + "loss": 0.8018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 891 + }, + { + "epoch": 1.0998766954377313, + "grad_norm": 2.2557943511918066, + "learning_rate": 1.4458077735672606e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 892 + }, + { + "epoch": 1.1011097410604191, + "grad_norm": 2.350846713741873, + "learning_rate": 1.4446273068457916e-05, + "loss": 0.8373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 893 + }, + { + "epoch": 1.1023427866831073, + "grad_norm": 2.3206912584026584, + "learning_rate": 1.4434460673588151e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 894 + }, + { + "epoch": 1.1035758323057954, + "grad_norm": 2.083251461091603, + "learning_rate": 1.4422640571593336e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 895 + }, + { + "epoch": 1.1048088779284833, + "grad_norm": 2.2429047820431456, + "learning_rate": 1.4410812783016893e-05, + "loss": 0.8066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 896 + }, + { + "epoch": 1.1060419235511714, + "grad_norm": 2.39736831788261, + "learning_rate": 1.4398977328415608e-05, + "loss": 0.8961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 897 + }, + { + "epoch": 1.1072749691738595, + "grad_norm": 2.4569397246397995, + "learning_rate": 1.4387134228359587e-05, + "loss": 0.9745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 898 + }, + { + "epoch": 1.1085080147965474, + "grad_norm": 2.194341713414294, + "learning_rate": 1.4375283503432223e-05, + "loss": 0.7642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 899 + }, + { + "epoch": 1.1097410604192355, + "grad_norm": 2.108094195937167, + "learning_rate": 1.4363425174230161e-05, + "loss": 0.7856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 900 + }, + { + "epoch": 1.1109741060419236, + "grad_norm": 2.140652113297743, + "learning_rate": 1.4351559261363267e-05, + "loss": 0.6648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 901 + }, + { + "epoch": 1.1122071516646117, + "grad_norm": 2.1936452316718063, + "learning_rate": 1.4339685785454587e-05, + "loss": 0.7336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 902 + }, + { + "epoch": 1.1134401972872996, + "grad_norm": 2.2269402261375038, + "learning_rate": 1.4327804767140303e-05, + "loss": 0.7708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 903 + }, + { + "epoch": 1.1146732429099877, + "grad_norm": 2.2989998022422453, + "learning_rate": 1.4315916227069713e-05, + "loss": 0.743, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 904 + }, + { + "epoch": 1.1159062885326758, + "grad_norm": 2.2917968867963427, + "learning_rate": 1.4304020185905189e-05, + "loss": 0.8126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 905 + }, + { + "epoch": 1.1171393341553637, + "grad_norm": 2.217875844794138, + "learning_rate": 1.4292116664322134e-05, + "loss": 0.6544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 906 + }, + { + "epoch": 1.1183723797780518, + "grad_norm": 2.4162119819935803, + "learning_rate": 1.4280205683008959e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 907 + }, + { + "epoch": 1.11960542540074, + "grad_norm": 2.3603757259614047, + "learning_rate": 1.4268287262667033e-05, + "loss": 0.7766, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 908 + }, + { + "epoch": 1.1208384710234278, + "grad_norm": 2.323141591914079, + "learning_rate": 1.4256361424010663e-05, + "loss": 0.8892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 909 + }, + { + "epoch": 1.122071516646116, + "grad_norm": 2.196302554500726, + "learning_rate": 1.4244428187767039e-05, + "loss": 0.9378, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 910 + }, + { + "epoch": 1.123304562268804, + "grad_norm": 2.2395403393552393, + "learning_rate": 1.4232487574676212e-05, + "loss": 1.0055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 911 + }, + { + "epoch": 1.124537607891492, + "grad_norm": 2.15769661287864, + "learning_rate": 1.4220539605491057e-05, + "loss": 0.748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 912 + }, + { + "epoch": 1.12577065351418, + "grad_norm": 2.0601339442187605, + "learning_rate": 1.4208584300977233e-05, + "loss": 0.7894, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 913 + }, + { + "epoch": 1.1270036991368682, + "grad_norm": 2.10447957980717, + "learning_rate": 1.4196621681913145e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 914 + }, + { + "epoch": 1.128236744759556, + "grad_norm": 2.3961229528506265, + "learning_rate": 1.4184651769089911e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 915 + }, + { + "epoch": 1.1294697903822442, + "grad_norm": 2.457287266967065, + "learning_rate": 1.417267458331133e-05, + "loss": 0.8519, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 916 + }, + { + "epoch": 1.1307028360049323, + "grad_norm": 2.023004753946474, + "learning_rate": 1.4160690145393838e-05, + "loss": 0.8449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 917 + }, + { + "epoch": 1.1319358816276202, + "grad_norm": 2.2533720475013634, + "learning_rate": 1.4148698476166471e-05, + "loss": 0.8221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 918 + }, + { + "epoch": 1.1331689272503083, + "grad_norm": 2.105957236229041, + "learning_rate": 1.4136699596470846e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 919 + }, + { + "epoch": 1.1344019728729964, + "grad_norm": 2.1652513956532946, + "learning_rate": 1.4124693527161095e-05, + "loss": 0.8322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 920 + }, + { + "epoch": 1.1356350184956843, + "grad_norm": 2.2463359048305693, + "learning_rate": 1.4112680289103858e-05, + "loss": 0.8458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 921 + }, + { + "epoch": 1.1368680641183724, + "grad_norm": 2.1946517808282784, + "learning_rate": 1.4100659903178231e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 922 + }, + { + "epoch": 1.1381011097410605, + "grad_norm": 2.116652661158196, + "learning_rate": 1.4088632390275731e-05, + "loss": 0.8496, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 923 + }, + { + "epoch": 1.1393341553637484, + "grad_norm": 2.1260376630680677, + "learning_rate": 1.4076597771300267e-05, + "loss": 0.7646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 924 + }, + { + "epoch": 1.1405672009864365, + "grad_norm": 2.1008385205849684, + "learning_rate": 1.4064556067168092e-05, + "loss": 0.8303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 925 + }, + { + "epoch": 1.1418002466091246, + "grad_norm": 2.1995475134175573, + "learning_rate": 1.4052507298807774e-05, + "loss": 0.8636, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 926 + }, + { + "epoch": 1.1430332922318125, + "grad_norm": 2.238025377523696, + "learning_rate": 1.404045148716016e-05, + "loss": 0.9095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 927 + }, + { + "epoch": 1.1442663378545006, + "grad_norm": 2.1584647692083636, + "learning_rate": 1.4028388653178346e-05, + "loss": 0.9552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 928 + }, + { + "epoch": 1.1454993834771887, + "grad_norm": 2.277913607137681, + "learning_rate": 1.4016318817827618e-05, + "loss": 0.6826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 929 + }, + { + "epoch": 1.1467324290998766, + "grad_norm": 2.1210301206048494, + "learning_rate": 1.4004242002085441e-05, + "loss": 0.7648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 930 + }, + { + "epoch": 1.1479654747225647, + "grad_norm": 2.1730216571228143, + "learning_rate": 1.3992158226941409e-05, + "loss": 0.7348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 931 + }, + { + "epoch": 1.1491985203452528, + "grad_norm": 2.2903920964212197, + "learning_rate": 1.3980067513397208e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 932 + }, + { + "epoch": 1.1504315659679407, + "grad_norm": 2.207582230014581, + "learning_rate": 1.3967969882466595e-05, + "loss": 0.8393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 933 + }, + { + "epoch": 1.1516646115906288, + "grad_norm": 2.1487993957640605, + "learning_rate": 1.3955865355175334e-05, + "loss": 0.9618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 934 + }, + { + "epoch": 1.152897657213317, + "grad_norm": 2.1100013581157673, + "learning_rate": 1.3943753952561179e-05, + "loss": 0.7185, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 935 + }, + { + "epoch": 1.1541307028360048, + "grad_norm": 2.1580493494759505, + "learning_rate": 1.3931635695673845e-05, + "loss": 0.8656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 936 + }, + { + "epoch": 1.155363748458693, + "grad_norm": 2.449600660057695, + "learning_rate": 1.3919510605574944e-05, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 937 + }, + { + "epoch": 1.156596794081381, + "grad_norm": 2.2204766170828782, + "learning_rate": 1.3907378703337977e-05, + "loss": 0.823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 938 + }, + { + "epoch": 1.157829839704069, + "grad_norm": 2.563974651134322, + "learning_rate": 1.389524001004827e-05, + "loss": 0.7685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 939 + }, + { + "epoch": 1.159062885326757, + "grad_norm": 2.258821529498267, + "learning_rate": 1.388309454680297e-05, + "loss": 0.673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 940 + }, + { + "epoch": 1.1602959309494452, + "grad_norm": 2.292191724384705, + "learning_rate": 1.3870942334710978e-05, + "loss": 0.8204, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 941 + }, + { + "epoch": 1.161528976572133, + "grad_norm": 2.283429457608797, + "learning_rate": 1.3858783394892925e-05, + "loss": 0.7777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 942 + }, + { + "epoch": 1.1627620221948212, + "grad_norm": 2.186748888333124, + "learning_rate": 1.3846617748481137e-05, + "loss": 0.7862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 943 + }, + { + "epoch": 1.1639950678175093, + "grad_norm": 2.262303399886297, + "learning_rate": 1.3834445416619602e-05, + "loss": 0.7559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 944 + }, + { + "epoch": 1.1652281134401972, + "grad_norm": 2.262832253985301, + "learning_rate": 1.3822266420463914e-05, + "loss": 0.9429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 945 + }, + { + "epoch": 1.1664611590628853, + "grad_norm": 2.150186199729938, + "learning_rate": 1.3810080781181264e-05, + "loss": 0.814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 946 + }, + { + "epoch": 1.1676942046855734, + "grad_norm": 2.343694448653775, + "learning_rate": 1.3797888519950373e-05, + "loss": 0.93, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 947 + }, + { + "epoch": 1.1689272503082615, + "grad_norm": 2.1813728957413385, + "learning_rate": 1.3785689657961494e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 948 + }, + { + "epoch": 1.1701602959309494, + "grad_norm": 2.119521364690522, + "learning_rate": 1.3773484216416326e-05, + "loss": 0.6718, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 949 + }, + { + "epoch": 1.1713933415536375, + "grad_norm": 2.2831289714651537, + "learning_rate": 1.3761272216528023e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 950 + }, + { + "epoch": 1.1726263871763256, + "grad_norm": 2.431755167028705, + "learning_rate": 1.3749053679521123e-05, + "loss": 0.8664, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 951 + }, + { + "epoch": 1.1738594327990135, + "grad_norm": 2.148435358078549, + "learning_rate": 1.3736828626631539e-05, + "loss": 0.6746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 952 + }, + { + "epoch": 1.1750924784217016, + "grad_norm": 2.124965667169116, + "learning_rate": 1.3724597079106498e-05, + "loss": 0.9202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 953 + }, + { + "epoch": 1.1763255240443897, + "grad_norm": 2.2382797156893397, + "learning_rate": 1.3712359058204525e-05, + "loss": 0.7696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 954 + }, + { + "epoch": 1.1775585696670776, + "grad_norm": 2.2207967591300632, + "learning_rate": 1.3700114585195381e-05, + "loss": 0.8654, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 955 + }, + { + "epoch": 1.1787916152897657, + "grad_norm": 2.3010582856815347, + "learning_rate": 1.3687863681360059e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 956 + }, + { + "epoch": 1.1800246609124538, + "grad_norm": 2.146527954418168, + "learning_rate": 1.367560636799071e-05, + "loss": 0.6291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 957 + }, + { + "epoch": 1.1812577065351417, + "grad_norm": 2.203217169625211, + "learning_rate": 1.3663342666390642e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 958 + }, + { + "epoch": 1.1824907521578298, + "grad_norm": 2.380280804061753, + "learning_rate": 1.365107259787425e-05, + "loss": 0.8015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 959 + }, + { + "epoch": 1.183723797780518, + "grad_norm": 2.7057879375717766, + "learning_rate": 1.3638796183767013e-05, + "loss": 0.8469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 960 + }, + { + "epoch": 1.1849568434032058, + "grad_norm": 2.0549864237362856, + "learning_rate": 1.362651344540542e-05, + "loss": 0.7724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 961 + }, + { + "epoch": 1.186189889025894, + "grad_norm": 2.165523344913771, + "learning_rate": 1.3614224404136964e-05, + "loss": 0.7119, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 962 + }, + { + "epoch": 1.187422934648582, + "grad_norm": 2.193503404484943, + "learning_rate": 1.3601929081320085e-05, + "loss": 0.769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 963 + }, + { + "epoch": 1.18865598027127, + "grad_norm": 2.30296466364579, + "learning_rate": 1.358962749832415e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 964 + }, + { + "epoch": 1.189889025893958, + "grad_norm": 2.3370321886887226, + "learning_rate": 1.3577319676529394e-05, + "loss": 0.8251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 965 + }, + { + "epoch": 1.1911220715166462, + "grad_norm": 2.1447616659612727, + "learning_rate": 1.3565005637326902e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 966 + }, + { + "epoch": 1.192355117139334, + "grad_norm": 2.1126610192998263, + "learning_rate": 1.3552685402118569e-05, + "loss": 0.893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 967 + }, + { + "epoch": 1.1935881627620222, + "grad_norm": 2.145481581859885, + "learning_rate": 1.3540358992317048e-05, + "loss": 0.8777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 968 + }, + { + "epoch": 1.1948212083847103, + "grad_norm": 2.0436843289234714, + "learning_rate": 1.3528026429345733e-05, + "loss": 0.7041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 969 + }, + { + "epoch": 1.1960542540073984, + "grad_norm": 2.3729052917065863, + "learning_rate": 1.351568773463871e-05, + "loss": 0.7822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 970 + }, + { + "epoch": 1.1972872996300863, + "grad_norm": 2.2771218973970977, + "learning_rate": 1.3503342929640713e-05, + "loss": 0.6887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 971 + }, + { + "epoch": 1.1985203452527744, + "grad_norm": 2.114743129136917, + "learning_rate": 1.3490992035807108e-05, + "loss": 0.6587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 972 + }, + { + "epoch": 1.1997533908754625, + "grad_norm": 2.210055964932895, + "learning_rate": 1.3478635074603844e-05, + "loss": 0.6986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 973 + }, + { + "epoch": 1.2009864364981504, + "grad_norm": 2.376467969342028, + "learning_rate": 1.34662720675074e-05, + "loss": 0.8992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 974 + }, + { + "epoch": 1.2022194821208385, + "grad_norm": 2.3169180290916893, + "learning_rate": 1.3453903036004784e-05, + "loss": 0.8468, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 975 + }, + { + "epoch": 1.2034525277435266, + "grad_norm": 2.4929171744348095, + "learning_rate": 1.3441528001593448e-05, + "loss": 0.8746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 976 + }, + { + "epoch": 1.2046855733662145, + "grad_norm": 2.3498012552061223, + "learning_rate": 1.3429146985781305e-05, + "loss": 0.848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 977 + }, + { + "epoch": 1.2059186189889026, + "grad_norm": 2.3519351668545223, + "learning_rate": 1.3416760010086647e-05, + "loss": 0.8045, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 978 + }, + { + "epoch": 1.2071516646115907, + "grad_norm": 2.152128272403449, + "learning_rate": 1.3404367096038127e-05, + "loss": 0.6621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 979 + }, + { + "epoch": 1.2083847102342786, + "grad_norm": 2.2310865684580126, + "learning_rate": 1.339196826517472e-05, + "loss": 0.8074, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 980 + }, + { + "epoch": 1.2096177558569667, + "grad_norm": 2.4218206259337336, + "learning_rate": 1.3379563539045688e-05, + "loss": 0.7489, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 981 + }, + { + "epoch": 1.2108508014796548, + "grad_norm": 2.3171020050526754, + "learning_rate": 1.3367152939210532e-05, + "loss": 0.6967, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 982 + }, + { + "epoch": 1.2120838471023427, + "grad_norm": 2.253666664504199, + "learning_rate": 1.3354736487238968e-05, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 983 + }, + { + "epoch": 1.2133168927250308, + "grad_norm": 2.202923113820963, + "learning_rate": 1.3342314204710878e-05, + "loss": 0.6643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 984 + }, + { + "epoch": 1.214549938347719, + "grad_norm": 2.3907506714199216, + "learning_rate": 1.332988611321628e-05, + "loss": 0.7483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 985 + }, + { + "epoch": 1.2157829839704068, + "grad_norm": 2.3288855585799726, + "learning_rate": 1.3317452234355286e-05, + "loss": 0.8775, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 986 + }, + { + "epoch": 1.217016029593095, + "grad_norm": 2.4270404362139124, + "learning_rate": 1.3305012589738075e-05, + "loss": 0.9152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 987 + }, + { + "epoch": 1.218249075215783, + "grad_norm": 2.3906498632865825, + "learning_rate": 1.3292567200984833e-05, + "loss": 0.6825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 988 + }, + { + "epoch": 1.219482120838471, + "grad_norm": 2.1776282578685495, + "learning_rate": 1.3280116089725743e-05, + "loss": 0.7098, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 989 + }, + { + "epoch": 1.220715166461159, + "grad_norm": 2.3972197284848265, + "learning_rate": 1.3267659277600921e-05, + "loss": 0.8804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 990 + }, + { + "epoch": 1.2219482120838472, + "grad_norm": 2.272454158882342, + "learning_rate": 1.3255196786260402e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 991 + }, + { + "epoch": 1.223181257706535, + "grad_norm": 2.0898886573973985, + "learning_rate": 1.3242728637364084e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 992 + }, + { + "epoch": 1.2244143033292232, + "grad_norm": 2.1283771526768422, + "learning_rate": 1.3230254852581703e-05, + "loss": 0.7541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 993 + }, + { + "epoch": 1.2256473489519113, + "grad_norm": 2.5947641083937785, + "learning_rate": 1.3217775453592786e-05, + "loss": 0.6516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 994 + }, + { + "epoch": 1.2268803945745992, + "grad_norm": 2.2994678179446155, + "learning_rate": 1.3205290462086623e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 995 + }, + { + "epoch": 1.2281134401972873, + "grad_norm": 2.419174005886699, + "learning_rate": 1.3192799899762216e-05, + "loss": 0.6653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 996 + }, + { + "epoch": 1.2293464858199754, + "grad_norm": 2.7052778802902724, + "learning_rate": 1.318030378832826e-05, + "loss": 0.8398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 997 + }, + { + "epoch": 1.2305795314426633, + "grad_norm": 2.182898548205837, + "learning_rate": 1.3167802149503079e-05, + "loss": 0.7902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 998 + }, + { + "epoch": 1.2318125770653514, + "grad_norm": 2.319601966909909, + "learning_rate": 1.315529500501462e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 999 + }, + { + "epoch": 1.2330456226880395, + "grad_norm": 2.3070852285534875, + "learning_rate": 1.3142782376600389e-05, + "loss": 0.7369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1000 + }, + { + "epoch": 1.2342786683107274, + "grad_norm": 2.3823064106232956, + "learning_rate": 1.3130264286007427e-05, + "loss": 0.8759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1001 + }, + { + "epoch": 1.2355117139334155, + "grad_norm": 2.2502915191290405, + "learning_rate": 1.3117740754992263e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1002 + }, + { + "epoch": 1.2367447595561036, + "grad_norm": 2.329022774742351, + "learning_rate": 1.310521180532089e-05, + "loss": 0.6222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1003 + }, + { + "epoch": 1.2379778051787915, + "grad_norm": 2.451176371025935, + "learning_rate": 1.3092677458768712e-05, + "loss": 0.9302, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1004 + }, + { + "epoch": 1.2392108508014796, + "grad_norm": 2.266087500621217, + "learning_rate": 1.3080137737120514e-05, + "loss": 0.7132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1005 + }, + { + "epoch": 1.2404438964241677, + "grad_norm": 2.2897262982963205, + "learning_rate": 1.3067592662170424e-05, + "loss": 0.7692, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1006 + }, + { + "epoch": 1.2416769420468556, + "grad_norm": 2.321897633330746, + "learning_rate": 1.3055042255721877e-05, + "loss": 0.8054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1007 + }, + { + "epoch": 1.2429099876695437, + "grad_norm": 2.327507620912996, + "learning_rate": 1.3042486539587566e-05, + "loss": 0.8495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1008 + }, + { + "epoch": 1.2441430332922319, + "grad_norm": 2.144625913300158, + "learning_rate": 1.3029925535589423e-05, + "loss": 0.7546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1009 + }, + { + "epoch": 1.2453760789149197, + "grad_norm": 2.1664603999352217, + "learning_rate": 1.301735926555856e-05, + "loss": 0.758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1010 + }, + { + "epoch": 1.2466091245376079, + "grad_norm": 2.2424395261695422, + "learning_rate": 1.300478775133525e-05, + "loss": 0.6794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1011 + }, + { + "epoch": 1.247842170160296, + "grad_norm": 2.135987556369877, + "learning_rate": 1.2992211014768872e-05, + "loss": 0.7814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1012 + }, + { + "epoch": 1.2490752157829839, + "grad_norm": 2.3429443823198812, + "learning_rate": 1.297962907771789e-05, + "loss": 0.756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1013 + }, + { + "epoch": 1.250308261405672, + "grad_norm": 2.3806466769436083, + "learning_rate": 1.29670419620498e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1014 + }, + { + "epoch": 1.25154130702836, + "grad_norm": 2.2614648564029536, + "learning_rate": 1.2954449689641105e-05, + "loss": 0.6227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1015 + }, + { + "epoch": 1.252774352651048, + "grad_norm": 2.690263466615422, + "learning_rate": 1.294185228237726e-05, + "loss": 0.8233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1016 + }, + { + "epoch": 1.254007398273736, + "grad_norm": 2.473463220519423, + "learning_rate": 1.2929249762152659e-05, + "loss": 0.7693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1017 + }, + { + "epoch": 1.2552404438964242, + "grad_norm": 2.203357353825905, + "learning_rate": 1.2916642150870567e-05, + "loss": 0.7486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1018 + }, + { + "epoch": 1.256473489519112, + "grad_norm": 2.407886169080358, + "learning_rate": 1.2904029470443105e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1019 + }, + { + "epoch": 1.2577065351418002, + "grad_norm": 2.416338321530697, + "learning_rate": 1.289141174279121e-05, + "loss": 0.7849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1020 + }, + { + "epoch": 1.2589395807644883, + "grad_norm": 2.1578914081370617, + "learning_rate": 1.2878788989844577e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1021 + }, + { + "epoch": 1.2601726263871762, + "grad_norm": 2.111289324604086, + "learning_rate": 1.2866161233541648e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1022 + }, + { + "epoch": 1.2614056720098643, + "grad_norm": 2.463350736417066, + "learning_rate": 1.2853528495829557e-05, + "loss": 0.7408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1023 + }, + { + "epoch": 1.2626387176325524, + "grad_norm": 2.5227574157477424, + "learning_rate": 1.284089079866409e-05, + "loss": 0.6943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1024 + }, + { + "epoch": 1.2638717632552403, + "grad_norm": 2.550177757961712, + "learning_rate": 1.2828248164009658e-05, + "loss": 0.8063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1025 + }, + { + "epoch": 1.2651048088779284, + "grad_norm": 2.279399386552855, + "learning_rate": 1.2815600613839254e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1026 + }, + { + "epoch": 1.2663378545006165, + "grad_norm": 2.403266684764464, + "learning_rate": 1.2802948170134411e-05, + "loss": 0.8159, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1027 + }, + { + "epoch": 1.2675709001233046, + "grad_norm": 2.3974372248557088, + "learning_rate": 1.2790290854885173e-05, + "loss": 0.7733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1028 + }, + { + "epoch": 1.2688039457459925, + "grad_norm": 2.75325380275428, + "learning_rate": 1.2777628690090035e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1029 + }, + { + "epoch": 1.2700369913686806, + "grad_norm": 2.253046145862575, + "learning_rate": 1.2764961697755947e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1030 + }, + { + "epoch": 1.2712700369913688, + "grad_norm": 2.0946644139817967, + "learning_rate": 1.2752289899898225e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1031 + }, + { + "epoch": 1.2725030826140566, + "grad_norm": 2.2771512156869576, + "learning_rate": 1.2739613318540546e-05, + "loss": 0.8351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1032 + }, + { + "epoch": 1.2737361282367448, + "grad_norm": 2.360325908676525, + "learning_rate": 1.2726931975714904e-05, + "loss": 0.8367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1033 + }, + { + "epoch": 1.2749691738594329, + "grad_norm": 2.310569034824901, + "learning_rate": 1.2714245893461568e-05, + "loss": 0.7346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1034 + }, + { + "epoch": 1.2762022194821208, + "grad_norm": 2.1364800896900413, + "learning_rate": 1.2701555093829034e-05, + "loss": 0.6637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1035 + }, + { + "epoch": 1.2774352651048089, + "grad_norm": 2.2320085770485734, + "learning_rate": 1.2688859598874013e-05, + "loss": 0.6849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1036 + }, + { + "epoch": 1.278668310727497, + "grad_norm": 2.2073805269530067, + "learning_rate": 1.2676159430661361e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1037 + }, + { + "epoch": 1.279901356350185, + "grad_norm": 2.2795714091180694, + "learning_rate": 1.2663454611264065e-05, + "loss": 0.8542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1038 + }, + { + "epoch": 1.281134401972873, + "grad_norm": 2.4086359220078535, + "learning_rate": 1.2650745162763195e-05, + "loss": 0.6699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1039 + }, + { + "epoch": 1.282367447595561, + "grad_norm": 2.3833290498603787, + "learning_rate": 1.2638031107247871e-05, + "loss": 0.7976, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1040 + }, + { + "epoch": 1.2836004932182492, + "grad_norm": 2.4078757121437873, + "learning_rate": 1.26253124668152e-05, + "loss": 0.7284, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1041 + }, + { + "epoch": 1.284833538840937, + "grad_norm": 2.3345693841609045, + "learning_rate": 1.2612589263570287e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1042 + }, + { + "epoch": 1.2860665844636252, + "grad_norm": 2.2804486360942633, + "learning_rate": 1.2599861519626142e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1043 + }, + { + "epoch": 1.2872996300863133, + "grad_norm": 2.219596427341039, + "learning_rate": 1.2587129257103684e-05, + "loss": 0.723, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1044 + }, + { + "epoch": 1.2885326757090012, + "grad_norm": 2.318632178358617, + "learning_rate": 1.2574392498131671e-05, + "loss": 0.7334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1045 + }, + { + "epoch": 1.2897657213316893, + "grad_norm": 2.1919076842831933, + "learning_rate": 1.2561651264846692e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1046 + }, + { + "epoch": 1.2909987669543774, + "grad_norm": 2.4787933074669186, + "learning_rate": 1.2548905579393096e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1047 + }, + { + "epoch": 1.2922318125770653, + "grad_norm": 2.4023622763250883, + "learning_rate": 1.2536155463922982e-05, + "loss": 0.826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1048 + }, + { + "epoch": 1.2934648581997534, + "grad_norm": 2.117125770604436, + "learning_rate": 1.252340094059614e-05, + "loss": 0.8052, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1049 + }, + { + "epoch": 1.2946979038224415, + "grad_norm": 2.2921041661226926, + "learning_rate": 1.251064203158003e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1050 + }, + { + "epoch": 1.2959309494451294, + "grad_norm": 2.5171575376979334, + "learning_rate": 1.2497878759049722e-05, + "loss": 0.6485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1051 + }, + { + "epoch": 1.2971639950678175, + "grad_norm": 2.1128556413546806, + "learning_rate": 1.248511114518788e-05, + "loss": 0.6076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1052 + }, + { + "epoch": 1.2983970406905057, + "grad_norm": 2.1493124843340907, + "learning_rate": 1.2472339212184712e-05, + "loss": 0.7343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1053 + }, + { + "epoch": 1.2996300863131935, + "grad_norm": 2.889368089302167, + "learning_rate": 1.2459562982237933e-05, + "loss": 0.7978, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1054 + }, + { + "epoch": 1.3008631319358817, + "grad_norm": 2.495383798802243, + "learning_rate": 1.2446782477552718e-05, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1055 + }, + { + "epoch": 1.3020961775585698, + "grad_norm": 2.517167273202911, + "learning_rate": 1.2433997720341682e-05, + "loss": 0.8663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1056 + }, + { + "epoch": 1.3033292231812577, + "grad_norm": 2.33725722509719, + "learning_rate": 1.2421208732824823e-05, + "loss": 0.7587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1057 + }, + { + "epoch": 1.3045622688039458, + "grad_norm": 2.29088652068904, + "learning_rate": 1.2408415537229497e-05, + "loss": 0.8488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1058 + }, + { + "epoch": 1.3057953144266339, + "grad_norm": 2.283233783970098, + "learning_rate": 1.239561815579037e-05, + "loss": 0.6389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1059 + }, + { + "epoch": 1.3070283600493218, + "grad_norm": 2.2928393078851568, + "learning_rate": 1.2382816610749388e-05, + "loss": 0.7261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1060 + }, + { + "epoch": 1.3082614056720099, + "grad_norm": 2.163749641191809, + "learning_rate": 1.2370010924355726e-05, + "loss": 0.6366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1061 + }, + { + "epoch": 1.309494451294698, + "grad_norm": 2.1229525047726963, + "learning_rate": 1.2357201118865762e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1062 + }, + { + "epoch": 1.3107274969173859, + "grad_norm": 2.272499973037141, + "learning_rate": 1.234438721654303e-05, + "loss": 0.6777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1063 + }, + { + "epoch": 1.311960542540074, + "grad_norm": 2.2172413085789207, + "learning_rate": 1.233156923965819e-05, + "loss": 0.8492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1064 + }, + { + "epoch": 1.313193588162762, + "grad_norm": 2.215217499299492, + "learning_rate": 1.2318747210488973e-05, + "loss": 0.816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1065 + }, + { + "epoch": 1.31442663378545, + "grad_norm": 2.2046121405476384, + "learning_rate": 1.2305921151320164e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1066 + }, + { + "epoch": 1.315659679408138, + "grad_norm": 2.410535974252942, + "learning_rate": 1.2293091084443547e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1067 + }, + { + "epoch": 1.3168927250308262, + "grad_norm": 2.2346919759512427, + "learning_rate": 1.2280257032157873e-05, + "loss": 0.5882, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1068 + }, + { + "epoch": 1.318125770653514, + "grad_norm": 2.217353695685963, + "learning_rate": 1.2267419016768814e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1069 + }, + { + "epoch": 1.3193588162762022, + "grad_norm": 2.3310973124804892, + "learning_rate": 1.225457706058894e-05, + "loss": 0.8387, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1070 + }, + { + "epoch": 1.3205918618988903, + "grad_norm": 2.363091956760975, + "learning_rate": 1.224173118593766e-05, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1071 + }, + { + "epoch": 1.3218249075215782, + "grad_norm": 2.3530303319299835, + "learning_rate": 1.22288814151412e-05, + "loss": 0.7271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1072 + }, + { + "epoch": 1.3230579531442663, + "grad_norm": 2.3412040138713137, + "learning_rate": 1.2216027770532559e-05, + "loss": 0.7482, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1073 + }, + { + "epoch": 1.3242909987669544, + "grad_norm": 2.388832200908418, + "learning_rate": 1.2203170274451462e-05, + "loss": 0.6927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1074 + }, + { + "epoch": 1.3255240443896423, + "grad_norm": 2.107265565119065, + "learning_rate": 1.219030894924433e-05, + "loss": 0.6069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1075 + }, + { + "epoch": 1.3267570900123304, + "grad_norm": 2.249447889766729, + "learning_rate": 1.2177443817264241e-05, + "loss": 0.7429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1076 + }, + { + "epoch": 1.3279901356350186, + "grad_norm": 2.337103531405435, + "learning_rate": 1.2164574900870887e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1077 + }, + { + "epoch": 1.3292231812577064, + "grad_norm": 2.4954282627925806, + "learning_rate": 1.2151702222430542e-05, + "loss": 0.781, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1078 + }, + { + "epoch": 1.3304562268803946, + "grad_norm": 2.4023586193420714, + "learning_rate": 1.2138825804316015e-05, + "loss": 0.7126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1079 + }, + { + "epoch": 1.3316892725030827, + "grad_norm": 2.293592183809771, + "learning_rate": 1.2125945668906611e-05, + "loss": 0.859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1080 + }, + { + "epoch": 1.3329223181257706, + "grad_norm": 2.280827683105619, + "learning_rate": 1.2113061838588104e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1081 + }, + { + "epoch": 1.3341553637484587, + "grad_norm": 2.2618423177043647, + "learning_rate": 1.2100174335752683e-05, + "loss": 0.8226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1082 + }, + { + "epoch": 1.3353884093711468, + "grad_norm": 2.263174374532883, + "learning_rate": 1.2087283182798923e-05, + "loss": 0.8671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1083 + }, + { + "epoch": 1.3366214549938347, + "grad_norm": 2.3088875209533875, + "learning_rate": 1.2074388402131737e-05, + "loss": 0.7575, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1084 + }, + { + "epoch": 1.3378545006165228, + "grad_norm": 2.217074039041576, + "learning_rate": 1.2061490016162357e-05, + "loss": 0.8498, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1085 + }, + { + "epoch": 1.339087546239211, + "grad_norm": 2.558767932572183, + "learning_rate": 1.2048588047308263e-05, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1086 + }, + { + "epoch": 1.3403205918618988, + "grad_norm": 2.1426590192119392, + "learning_rate": 1.203568251799318e-05, + "loss": 0.8473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1087 + }, + { + "epoch": 1.341553637484587, + "grad_norm": 2.3130116023826903, + "learning_rate": 1.2022773450647002e-05, + "loss": 0.6533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1088 + }, + { + "epoch": 1.342786683107275, + "grad_norm": 2.2474452770176505, + "learning_rate": 1.2009860867705793e-05, + "loss": 0.7731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1089 + }, + { + "epoch": 1.344019728729963, + "grad_norm": 2.553351511741815, + "learning_rate": 1.1996944791611711e-05, + "loss": 0.8881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1090 + }, + { + "epoch": 1.345252774352651, + "grad_norm": 2.4417233034036747, + "learning_rate": 1.1984025244812987e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1091 + }, + { + "epoch": 1.3464858199753391, + "grad_norm": 2.02108560427268, + "learning_rate": 1.1971102249763895e-05, + "loss": 0.5307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1092 + }, + { + "epoch": 1.347718865598027, + "grad_norm": 2.4627196027809655, + "learning_rate": 1.1958175828924692e-05, + "loss": 0.6095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1093 + }, + { + "epoch": 1.3489519112207151, + "grad_norm": 2.2087226765380703, + "learning_rate": 1.194524600476159e-05, + "loss": 0.7225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1094 + }, + { + "epoch": 1.3501849568434032, + "grad_norm": 2.268031072263066, + "learning_rate": 1.1932312799746722e-05, + "loss": 0.7492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1095 + }, + { + "epoch": 1.3514180024660911, + "grad_norm": 2.187415973980175, + "learning_rate": 1.1919376236358087e-05, + "loss": 0.6372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1096 + }, + { + "epoch": 1.3526510480887792, + "grad_norm": 2.3776853142465755, + "learning_rate": 1.1906436337079531e-05, + "loss": 0.757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1097 + }, + { + "epoch": 1.3538840937114673, + "grad_norm": 2.633083891326042, + "learning_rate": 1.189349312440069e-05, + "loss": 0.741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1098 + }, + { + "epoch": 1.3551171393341552, + "grad_norm": 2.3175112181742703, + "learning_rate": 1.1880546620816971e-05, + "loss": 0.577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1099 + }, + { + "epoch": 1.3563501849568433, + "grad_norm": 2.57847955905109, + "learning_rate": 1.1867596848829479e-05, + "loss": 0.7485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1100 + }, + { + "epoch": 1.3575832305795315, + "grad_norm": 2.2807181061723223, + "learning_rate": 1.1854643830945023e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1101 + }, + { + "epoch": 1.3588162762022196, + "grad_norm": 2.150298874669539, + "learning_rate": 1.1841687589676037e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1102 + }, + { + "epoch": 1.3600493218249075, + "grad_norm": 2.239857482190429, + "learning_rate": 1.1828728147540565e-05, + "loss": 0.5845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1103 + }, + { + "epoch": 1.3612823674475956, + "grad_norm": 2.391169760631781, + "learning_rate": 1.1815765527062207e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1104 + }, + { + "epoch": 1.3625154130702837, + "grad_norm": 2.233510553215683, + "learning_rate": 1.18027997507701e-05, + "loss": 0.687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1105 + }, + { + "epoch": 1.3637484586929716, + "grad_norm": 2.3406994321803376, + "learning_rate": 1.1789830841198854e-05, + "loss": 0.541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1106 + }, + { + "epoch": 1.3649815043156597, + "grad_norm": 3.0813520964620884, + "learning_rate": 1.1776858820888532e-05, + "loss": 0.8939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1107 + }, + { + "epoch": 1.3662145499383478, + "grad_norm": 2.289968441014776, + "learning_rate": 1.1763883712384593e-05, + "loss": 0.6178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1108 + }, + { + "epoch": 1.3674475955610357, + "grad_norm": 2.753373823273893, + "learning_rate": 1.1750905538237877e-05, + "loss": 0.8218, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1109 + }, + { + "epoch": 1.3686806411837238, + "grad_norm": 2.582931593682227, + "learning_rate": 1.1737924321004544e-05, + "loss": 0.5923, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1110 + }, + { + "epoch": 1.369913686806412, + "grad_norm": 2.577422739044973, + "learning_rate": 1.1724940083246042e-05, + "loss": 0.7151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1111 + }, + { + "epoch": 1.3711467324291, + "grad_norm": 2.4016601831106867, + "learning_rate": 1.171195284752907e-05, + "loss": 0.6719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1112 + }, + { + "epoch": 1.372379778051788, + "grad_norm": 2.2381623856871187, + "learning_rate": 1.1698962636425546e-05, + "loss": 0.7466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1113 + }, + { + "epoch": 1.373612823674476, + "grad_norm": 2.3077954380459498, + "learning_rate": 1.1685969472512544e-05, + "loss": 0.6609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1114 + }, + { + "epoch": 1.3748458692971641, + "grad_norm": 2.528220870127657, + "learning_rate": 1.1672973378372283e-05, + "loss": 0.7914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1115 + }, + { + "epoch": 1.376078914919852, + "grad_norm": 2.2640778971544946, + "learning_rate": 1.1659974376592065e-05, + "loss": 0.5914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1116 + }, + { + "epoch": 1.3773119605425401, + "grad_norm": 2.3357561361414994, + "learning_rate": 1.1646972489764254e-05, + "loss": 0.7194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1117 + }, + { + "epoch": 1.3785450061652282, + "grad_norm": 2.578566043676433, + "learning_rate": 1.163396774048622e-05, + "loss": 0.8069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1118 + }, + { + "epoch": 1.3797780517879161, + "grad_norm": 2.6394995201547853, + "learning_rate": 1.1620960151360317e-05, + "loss": 0.797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1119 + }, + { + "epoch": 1.3810110974106042, + "grad_norm": 2.2935912265432963, + "learning_rate": 1.1607949744993829e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1120 + }, + { + "epoch": 1.3822441430332923, + "grad_norm": 2.3848813510148394, + "learning_rate": 1.1594936543998933e-05, + "loss": 0.5816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1121 + }, + { + "epoch": 1.3834771886559802, + "grad_norm": 2.266951138614489, + "learning_rate": 1.1581920570992673e-05, + "loss": 0.8167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1122 + }, + { + "epoch": 1.3847102342786684, + "grad_norm": 2.0749563395142867, + "learning_rate": 1.1568901848596905e-05, + "loss": 0.6801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1123 + }, + { + "epoch": 1.3859432799013565, + "grad_norm": 2.259958476393895, + "learning_rate": 1.1555880399438257e-05, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1124 + }, + { + "epoch": 1.3871763255240444, + "grad_norm": 2.329878469976296, + "learning_rate": 1.1542856246148109e-05, + "loss": 0.7625, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1125 + }, + { + "epoch": 1.3884093711467325, + "grad_norm": 2.626476895279836, + "learning_rate": 1.1529829411362532e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1126 + }, + { + "epoch": 1.3896424167694206, + "grad_norm": 2.2360683032625297, + "learning_rate": 1.1516799917722262e-05, + "loss": 0.6588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1127 + }, + { + "epoch": 1.3908754623921085, + "grad_norm": 2.2606389128750664, + "learning_rate": 1.1503767787872652e-05, + "loss": 0.612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1128 + }, + { + "epoch": 1.3921085080147966, + "grad_norm": 2.3488413613095043, + "learning_rate": 1.149073304446364e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1129 + }, + { + "epoch": 1.3933415536374847, + "grad_norm": 2.187014482346159, + "learning_rate": 1.1477695710149706e-05, + "loss": 0.6897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1130 + }, + { + "epoch": 1.3945745992601726, + "grad_norm": 2.232874524639436, + "learning_rate": 1.1464655807589829e-05, + "loss": 0.7495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1131 + }, + { + "epoch": 1.3958076448828607, + "grad_norm": 2.2422283956051277, + "learning_rate": 1.1451613359447462e-05, + "loss": 0.7748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1132 + }, + { + "epoch": 1.3970406905055488, + "grad_norm": 2.2170691512303327, + "learning_rate": 1.1438568388390465e-05, + "loss": 0.7026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1133 + }, + { + "epoch": 1.3982737361282367, + "grad_norm": 2.3511388510197495, + "learning_rate": 1.1425520917091104e-05, + "loss": 0.7438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1134 + }, + { + "epoch": 1.3995067817509248, + "grad_norm": 2.302626316717242, + "learning_rate": 1.141247096822597e-05, + "loss": 0.6808, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1135 + }, + { + "epoch": 1.400739827373613, + "grad_norm": 2.358682446508914, + "learning_rate": 1.1399418564475977e-05, + "loss": 0.7462, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1136 + }, + { + "epoch": 1.4019728729963008, + "grad_norm": 2.4921870940266864, + "learning_rate": 1.138636372852629e-05, + "loss": 0.7565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1137 + }, + { + "epoch": 1.403205918618989, + "grad_norm": 2.881251766211438, + "learning_rate": 1.1373306483066315e-05, + "loss": 0.8531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1138 + }, + { + "epoch": 1.404438964241677, + "grad_norm": 2.374914568635526, + "learning_rate": 1.1360246850789638e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1139 + }, + { + "epoch": 1.405672009864365, + "grad_norm": 2.4892197803325478, + "learning_rate": 1.1347184854394e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1140 + }, + { + "epoch": 1.406905055487053, + "grad_norm": 2.335470361345138, + "learning_rate": 1.1334120516581239e-05, + "loss": 0.6239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1141 + }, + { + "epoch": 1.4081381011097411, + "grad_norm": 2.295703427576132, + "learning_rate": 1.1321053860057275e-05, + "loss": 0.7663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1142 + }, + { + "epoch": 1.409371146732429, + "grad_norm": 2.266519214119513, + "learning_rate": 1.1307984907532047e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1143 + }, + { + "epoch": 1.4106041923551171, + "grad_norm": 2.3854379731705917, + "learning_rate": 1.1294913681719497e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1144 + }, + { + "epoch": 1.4118372379778052, + "grad_norm": 2.3244362810232473, + "learning_rate": 1.1281840205337504e-05, + "loss": 0.694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1145 + }, + { + "epoch": 1.4130702836004931, + "grad_norm": 2.3269738125045727, + "learning_rate": 1.1268764501107872e-05, + "loss": 0.6867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1146 + }, + { + "epoch": 1.4143033292231812, + "grad_norm": 2.1038829157154533, + "learning_rate": 1.1255686591756265e-05, + "loss": 0.4939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1147 + }, + { + "epoch": 1.4155363748458694, + "grad_norm": 2.3057935606552356, + "learning_rate": 1.124260650001219e-05, + "loss": 0.697, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1148 + }, + { + "epoch": 1.4167694204685573, + "grad_norm": 2.4457435727408448, + "learning_rate": 1.1229524248608936e-05, + "loss": 0.7129, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1149 + }, + { + "epoch": 1.4180024660912454, + "grad_norm": 2.5516324512905793, + "learning_rate": 1.1216439860283556e-05, + "loss": 0.7116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1150 + }, + { + "epoch": 1.4192355117139335, + "grad_norm": 2.439191648697924, + "learning_rate": 1.120335335777681e-05, + "loss": 0.7269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1151 + }, + { + "epoch": 1.4204685573366214, + "grad_norm": 2.381736438404098, + "learning_rate": 1.1190264763833138e-05, + "loss": 0.7226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1152 + }, + { + "epoch": 1.4217016029593095, + "grad_norm": 2.326191969965805, + "learning_rate": 1.1177174101200609e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1153 + }, + { + "epoch": 1.4229346485819976, + "grad_norm": 2.4443478986341964, + "learning_rate": 1.1164081392630895e-05, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1154 + }, + { + "epoch": 1.4241676942046855, + "grad_norm": 2.4191171109083274, + "learning_rate": 1.115098666087921e-05, + "loss": 0.671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1155 + }, + { + "epoch": 1.4254007398273736, + "grad_norm": 2.331429993762429, + "learning_rate": 1.1137889928704303e-05, + "loss": 0.7875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1156 + }, + { + "epoch": 1.4266337854500617, + "grad_norm": 2.6182853389449905, + "learning_rate": 1.1124791218868383e-05, + "loss": 0.7465, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1157 + }, + { + "epoch": 1.4278668310727496, + "grad_norm": 2.3276838263801474, + "learning_rate": 1.111169055413711e-05, + "loss": 0.6731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1158 + }, + { + "epoch": 1.4290998766954377, + "grad_norm": 2.3555978375187783, + "learning_rate": 1.109858795727953e-05, + "loss": 0.7079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1159 + }, + { + "epoch": 1.4303329223181258, + "grad_norm": 2.243198455884724, + "learning_rate": 1.1085483451068056e-05, + "loss": 0.6767, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1160 + }, + { + "epoch": 1.4315659679408137, + "grad_norm": 2.426755129308782, + "learning_rate": 1.107237705827841e-05, + "loss": 0.6942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1161 + }, + { + "epoch": 1.4327990135635018, + "grad_norm": 2.310530088918712, + "learning_rate": 1.1059268801689605e-05, + "loss": 0.7414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1162 + }, + { + "epoch": 1.43403205918619, + "grad_norm": 2.3870562170497984, + "learning_rate": 1.104615870408388e-05, + "loss": 0.3904, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1163 + }, + { + "epoch": 1.4352651048088778, + "grad_norm": 2.185670591654591, + "learning_rate": 1.1033046788246684e-05, + "loss": 0.5667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1164 + }, + { + "epoch": 1.436498150431566, + "grad_norm": 2.2363567918319904, + "learning_rate": 1.1019933076966624e-05, + "loss": 0.6582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1165 + }, + { + "epoch": 1.437731196054254, + "grad_norm": 2.4989532650808917, + "learning_rate": 1.1006817593035423e-05, + "loss": 0.8567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1166 + }, + { + "epoch": 1.438964241676942, + "grad_norm": 2.3045828034357614, + "learning_rate": 1.0993700359247889e-05, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1167 + }, + { + "epoch": 1.44019728729963, + "grad_norm": 2.4408566344557263, + "learning_rate": 1.0980581398401875e-05, + "loss": 0.867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1168 + }, + { + "epoch": 1.4414303329223181, + "grad_norm": 2.433134793461258, + "learning_rate": 1.0967460733298224e-05, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1169 + }, + { + "epoch": 1.442663378545006, + "grad_norm": 2.084326240070432, + "learning_rate": 1.0954338386740746e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1170 + }, + { + "epoch": 1.4438964241676941, + "grad_norm": 2.3889216941699165, + "learning_rate": 1.0941214381536185e-05, + "loss": 0.7649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1171 + }, + { + "epoch": 1.4451294697903823, + "grad_norm": 2.3052157462538787, + "learning_rate": 1.092808874049415e-05, + "loss": 0.6548, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1172 + }, + { + "epoch": 1.4463625154130701, + "grad_norm": 2.412258190430439, + "learning_rate": 1.0914961486427103e-05, + "loss": 0.6655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1173 + }, + { + "epoch": 1.4475955610357583, + "grad_norm": 2.3820079102806755, + "learning_rate": 1.0901832642150308e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1174 + }, + { + "epoch": 1.4488286066584464, + "grad_norm": 2.826438039521343, + "learning_rate": 1.0888702230481788e-05, + "loss": 0.6666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1175 + }, + { + "epoch": 1.4500616522811345, + "grad_norm": 2.3005434823783166, + "learning_rate": 1.0875570274242305e-05, + "loss": 0.7333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1176 + }, + { + "epoch": 1.4512946979038224, + "grad_norm": 2.580920616275187, + "learning_rate": 1.0862436796255285e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1177 + }, + { + "epoch": 1.4525277435265105, + "grad_norm": 2.39619982968134, + "learning_rate": 1.0849301819346809e-05, + "loss": 0.6086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1178 + }, + { + "epoch": 1.4537607891491986, + "grad_norm": 2.4469170067316246, + "learning_rate": 1.0836165366345572e-05, + "loss": 0.7415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1179 + }, + { + "epoch": 1.4549938347718865, + "grad_norm": 2.4895765738835327, + "learning_rate": 1.0823027460082815e-05, + "loss": 0.8269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1180 + }, + { + "epoch": 1.4562268803945746, + "grad_norm": 2.1191361440164935, + "learning_rate": 1.0809888123392321e-05, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1181 + }, + { + "epoch": 1.4574599260172627, + "grad_norm": 2.0580916232206263, + "learning_rate": 1.0796747379110351e-05, + "loss": 0.5483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1182 + }, + { + "epoch": 1.4586929716399506, + "grad_norm": 2.358708614850794, + "learning_rate": 1.0783605250075615e-05, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1183 + }, + { + "epoch": 1.4599260172626387, + "grad_norm": 2.3706216835188068, + "learning_rate": 1.0770461759129225e-05, + "loss": 0.6925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1184 + }, + { + "epoch": 1.4611590628853268, + "grad_norm": 2.075953282437584, + "learning_rate": 1.0757316929114673e-05, + "loss": 0.7095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1185 + }, + { + "epoch": 1.462392108508015, + "grad_norm": 2.3425821034074175, + "learning_rate": 1.0744170782877765e-05, + "loss": 0.5685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1186 + }, + { + "epoch": 1.4636251541307028, + "grad_norm": 2.364864320197578, + "learning_rate": 1.07310233432666e-05, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1187 + }, + { + "epoch": 1.464858199753391, + "grad_norm": 2.269198583092931, + "learning_rate": 1.071787463313152e-05, + "loss": 0.5853, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1188 + }, + { + "epoch": 1.466091245376079, + "grad_norm": 2.3353973695396504, + "learning_rate": 1.0704724675325086e-05, + "loss": 0.6851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1189 + }, + { + "epoch": 1.467324290998767, + "grad_norm": 2.297514940312218, + "learning_rate": 1.0691573492702014e-05, + "loss": 0.6437, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1190 + }, + { + "epoch": 1.468557336621455, + "grad_norm": 2.347920635849606, + "learning_rate": 1.0678421108119161e-05, + "loss": 0.5607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1191 + }, + { + "epoch": 1.4697903822441432, + "grad_norm": 2.4998335942015792, + "learning_rate": 1.0665267544435464e-05, + "loss": 0.715, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1192 + }, + { + "epoch": 1.471023427866831, + "grad_norm": 2.6799025570634494, + "learning_rate": 1.065211282451192e-05, + "loss": 0.855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1193 + }, + { + "epoch": 1.4722564734895192, + "grad_norm": 2.3846645442657497, + "learning_rate": 1.0638956971211515e-05, + "loss": 0.8093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1194 + }, + { + "epoch": 1.4734895191122073, + "grad_norm": 2.289987275457904, + "learning_rate": 1.0625800007399232e-05, + "loss": 0.615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1195 + }, + { + "epoch": 1.4747225647348952, + "grad_norm": 2.398231591907997, + "learning_rate": 1.061264195594196e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1196 + }, + { + "epoch": 1.4759556103575833, + "grad_norm": 2.598957432876697, + "learning_rate": 1.0599482839708492e-05, + "loss": 0.5761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1197 + }, + { + "epoch": 1.4771886559802714, + "grad_norm": 2.3008244142116308, + "learning_rate": 1.0586322681569468e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1198 + }, + { + "epoch": 1.4784217016029593, + "grad_norm": 2.62152754882142, + "learning_rate": 1.0573161504397345e-05, + "loss": 0.9059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1199 + }, + { + "epoch": 1.4796547472256474, + "grad_norm": 2.7027129465803323, + "learning_rate": 1.0559999331066334e-05, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1200 + }, + { + "epoch": 1.4808877928483355, + "grad_norm": 2.441005731777059, + "learning_rate": 1.0546836184452397e-05, + "loss": 0.8212, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1201 + }, + { + "epoch": 1.4821208384710234, + "grad_norm": 2.4156909019245516, + "learning_rate": 1.0533672087433176e-05, + "loss": 0.6755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1202 + }, + { + "epoch": 1.4833538840937115, + "grad_norm": 2.373408895848631, + "learning_rate": 1.0520507062887966e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1203 + }, + { + "epoch": 1.4845869297163996, + "grad_norm": 2.1607701722680814, + "learning_rate": 1.050734113369768e-05, + "loss": 0.6858, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1204 + }, + { + "epoch": 1.4858199753390875, + "grad_norm": 2.211217022679158, + "learning_rate": 1.0494174322744797e-05, + "loss": 0.7383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1205 + }, + { + "epoch": 1.4870530209617756, + "grad_norm": 2.180028246293776, + "learning_rate": 1.0481006652913328e-05, + "loss": 0.5521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1206 + }, + { + "epoch": 1.4882860665844637, + "grad_norm": 2.393847834562169, + "learning_rate": 1.0467838147088785e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1207 + }, + { + "epoch": 1.4895191122071516, + "grad_norm": 2.20100213605133, + "learning_rate": 1.0454668828158122e-05, + "loss": 0.7316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1208 + }, + { + "epoch": 1.4907521578298397, + "grad_norm": 2.43405475601361, + "learning_rate": 1.0441498719009715e-05, + "loss": 0.7545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1209 + }, + { + "epoch": 1.4919852034525278, + "grad_norm": 2.3073004776119004, + "learning_rate": 1.0428327842533309e-05, + "loss": 0.5749, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1210 + }, + { + "epoch": 1.4932182490752157, + "grad_norm": 2.5013121128147535, + "learning_rate": 1.041515622161998e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1211 + }, + { + "epoch": 1.4944512946979038, + "grad_norm": 2.7168877702234537, + "learning_rate": 1.0401983879162108e-05, + "loss": 0.7342, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1212 + }, + { + "epoch": 1.495684340320592, + "grad_norm": 2.35281737411902, + "learning_rate": 1.0388810838053316e-05, + "loss": 0.6203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1213 + }, + { + "epoch": 1.4969173859432798, + "grad_norm": 2.400077140272344, + "learning_rate": 1.0375637121188447e-05, + "loss": 0.6201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1214 + }, + { + "epoch": 1.498150431565968, + "grad_norm": 2.5745039938322796, + "learning_rate": 1.0362462751463518e-05, + "loss": 0.6622, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1215 + }, + { + "epoch": 1.499383477188656, + "grad_norm": 2.4221988301394966, + "learning_rate": 1.0349287751775677e-05, + "loss": 0.7025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1216 + }, + { + "epoch": 1.500616522811344, + "grad_norm": 2.386783019524135, + "learning_rate": 1.033611214502317e-05, + "loss": 0.5671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1217 + }, + { + "epoch": 1.501849568434032, + "grad_norm": 2.539552131989461, + "learning_rate": 1.0322935954105303e-05, + "loss": 0.6469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1218 + }, + { + "epoch": 1.5030826140567202, + "grad_norm": 2.5808036138210673, + "learning_rate": 1.0309759201922386e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1219 + }, + { + "epoch": 1.504315659679408, + "grad_norm": 2.5496582442036395, + "learning_rate": 1.0296581911375712e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1220 + }, + { + "epoch": 1.5055487053020962, + "grad_norm": 2.304498549067636, + "learning_rate": 1.0283404105367512e-05, + "loss": 0.6358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1221 + }, + { + "epoch": 1.5067817509247843, + "grad_norm": 2.315506316971876, + "learning_rate": 1.0270225806800904e-05, + "loss": 0.643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1222 + }, + { + "epoch": 1.5080147965474722, + "grad_norm": 2.6544020115345726, + "learning_rate": 1.0257047038579868e-05, + "loss": 0.7594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1223 + }, + { + "epoch": 1.5092478421701603, + "grad_norm": 2.1918210902533843, + "learning_rate": 1.0243867823609207e-05, + "loss": 0.6812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1224 + }, + { + "epoch": 1.5104808877928484, + "grad_norm": 2.3728046507766316, + "learning_rate": 1.0230688184794482e-05, + "loss": 0.7418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1225 + }, + { + "epoch": 1.5117139334155363, + "grad_norm": 2.3382615280020254, + "learning_rate": 1.0217508145042007e-05, + "loss": 0.662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1226 + }, + { + "epoch": 1.5129469790382244, + "grad_norm": 2.174082577884056, + "learning_rate": 1.0204327727258787e-05, + "loss": 0.5961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1227 + }, + { + "epoch": 1.5141800246609125, + "grad_norm": 2.3119491032275596, + "learning_rate": 1.0191146954352486e-05, + "loss": 0.776, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1228 + }, + { + "epoch": 1.5154130702836004, + "grad_norm": 2.1858032851748086, + "learning_rate": 1.017796584923138e-05, + "loss": 0.5297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1229 + }, + { + "epoch": 1.5166461159062885, + "grad_norm": 2.379536502814726, + "learning_rate": 1.0164784434804327e-05, + "loss": 0.6471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1230 + }, + { + "epoch": 1.5178791615289766, + "grad_norm": 2.5211256291060655, + "learning_rate": 1.0151602733980722e-05, + "loss": 0.6299, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1231 + }, + { + "epoch": 1.5191122071516645, + "grad_norm": 2.406098151191697, + "learning_rate": 1.013842076967046e-05, + "loss": 0.6612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1232 + }, + { + "epoch": 1.5203452527743526, + "grad_norm": 2.4669460715815843, + "learning_rate": 1.0125238564783886e-05, + "loss": 0.7121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1233 + }, + { + "epoch": 1.5215782983970407, + "grad_norm": 2.420528895345573, + "learning_rate": 1.0112056142231774e-05, + "loss": 0.5687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1234 + }, + { + "epoch": 1.5228113440197286, + "grad_norm": 2.7051912038626074, + "learning_rate": 1.0098873524925268e-05, + "loss": 0.6909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1235 + }, + { + "epoch": 1.5240443896424167, + "grad_norm": 2.496413960615992, + "learning_rate": 1.008569073577585e-05, + "loss": 0.6021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1236 + }, + { + "epoch": 1.5252774352651048, + "grad_norm": 2.4865070368167808, + "learning_rate": 1.007250779769531e-05, + "loss": 0.5448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1237 + }, + { + "epoch": 1.5265104808877927, + "grad_norm": 2.4574534769181513, + "learning_rate": 1.005932473359569e-05, + "loss": 0.6773, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1238 + }, + { + "epoch": 1.5277435265104808, + "grad_norm": 2.3040474377405826, + "learning_rate": 1.0046141566389247e-05, + "loss": 0.5662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1239 + }, + { + "epoch": 1.528976572133169, + "grad_norm": 2.42558040335409, + "learning_rate": 1.0032958318988426e-05, + "loss": 0.6739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1240 + }, + { + "epoch": 1.5302096177558568, + "grad_norm": 2.3661886293872243, + "learning_rate": 1.0019775014305804e-05, + "loss": 0.6297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1241 + }, + { + "epoch": 1.531442663378545, + "grad_norm": 2.314603674785468, + "learning_rate": 1.0006591675254063e-05, + "loss": 0.5369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1242 + }, + { + "epoch": 1.532675709001233, + "grad_norm": 2.3999228551654412, + "learning_rate": 9.993408324745942e-06, + "loss": 0.6449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1243 + }, + { + "epoch": 1.533908754623921, + "grad_norm": 2.3862165327435667, + "learning_rate": 9.9802249856942e-06, + "loss": 0.6161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1244 + }, + { + "epoch": 1.5351418002466093, + "grad_norm": 2.4581866876380487, + "learning_rate": 9.96704168101158e-06, + "loss": 0.6433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1245 + }, + { + "epoch": 1.5363748458692972, + "grad_norm": 2.505484522547059, + "learning_rate": 9.953858433610756e-06, + "loss": 0.6709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1246 + }, + { + "epoch": 1.537607891491985, + "grad_norm": 2.561128967653091, + "learning_rate": 9.940675266404311e-06, + "loss": 0.7824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1247 + }, + { + "epoch": 1.5388409371146734, + "grad_norm": 2.4218451757264234, + "learning_rate": 9.927492202304693e-06, + "loss": 0.7463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1248 + }, + { + "epoch": 1.5400739827373613, + "grad_norm": 2.4125121526776154, + "learning_rate": 9.91430926422415e-06, + "loss": 0.6605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1249 + }, + { + "epoch": 1.5413070283600492, + "grad_norm": 2.445505663908, + "learning_rate": 9.901126475074737e-06, + "loss": 0.626, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1250 + }, + { + "epoch": 1.5425400739827375, + "grad_norm": 2.314619197677989, + "learning_rate": 9.887943857768228e-06, + "loss": 0.6289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1251 + }, + { + "epoch": 1.5437731196054254, + "grad_norm": 2.3232344719049967, + "learning_rate": 9.874761435216119e-06, + "loss": 0.5306, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1252 + }, + { + "epoch": 1.5450061652281133, + "grad_norm": 2.4949418638683007, + "learning_rate": 9.861579230329542e-06, + "loss": 0.5698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1253 + }, + { + "epoch": 1.5462392108508016, + "grad_norm": 2.3089453128921105, + "learning_rate": 9.84839726601928e-06, + "loss": 0.5801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1254 + }, + { + "epoch": 1.5474722564734895, + "grad_norm": 2.274576156733362, + "learning_rate": 9.835215565195676e-06, + "loss": 0.8039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1255 + }, + { + "epoch": 1.5487053020961774, + "grad_norm": 2.2604020732970818, + "learning_rate": 9.822034150768625e-06, + "loss": 0.7985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1256 + }, + { + "epoch": 1.5499383477188657, + "grad_norm": 2.1641795636087915, + "learning_rate": 9.80885304564752e-06, + "loss": 0.6705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1257 + }, + { + "epoch": 1.5511713933415536, + "grad_norm": 2.48269267754425, + "learning_rate": 9.795672272741216e-06, + "loss": 0.6985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1258 + }, + { + "epoch": 1.5524044389642415, + "grad_norm": 2.3771051242924903, + "learning_rate": 9.782491854957998e-06, + "loss": 0.6647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1259 + }, + { + "epoch": 1.5536374845869299, + "grad_norm": 2.2133943018691684, + "learning_rate": 9.769311815205522e-06, + "loss": 0.4763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1260 + }, + { + "epoch": 1.5548705302096177, + "grad_norm": 2.4689964457263605, + "learning_rate": 9.756132176390797e-06, + "loss": 0.667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1261 + }, + { + "epoch": 1.5561035758323056, + "grad_norm": 2.2973874789470936, + "learning_rate": 9.742952961420134e-06, + "loss": 0.6321, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1262 + }, + { + "epoch": 1.557336621454994, + "grad_norm": 2.368171008344279, + "learning_rate": 9.7297741931991e-06, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1263 + }, + { + "epoch": 1.5585696670776819, + "grad_norm": 2.435391702423389, + "learning_rate": 9.716595894632493e-06, + "loss": 0.5594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1264 + }, + { + "epoch": 1.55980271270037, + "grad_norm": 2.4382371041738184, + "learning_rate": 9.703418088624292e-06, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1265 + }, + { + "epoch": 1.561035758323058, + "grad_norm": 2.512447464904025, + "learning_rate": 9.690240798077616e-06, + "loss": 0.7641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1266 + }, + { + "epoch": 1.562268803945746, + "grad_norm": 2.271874085642332, + "learning_rate": 9.6770640458947e-06, + "loss": 0.6891, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1267 + }, + { + "epoch": 1.563501849568434, + "grad_norm": 2.2061279299276113, + "learning_rate": 9.66388785497683e-06, + "loss": 0.4999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1268 + }, + { + "epoch": 1.5647348951911222, + "grad_norm": 2.3058075007472416, + "learning_rate": 9.650712248224327e-06, + "loss": 0.6043, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1269 + }, + { + "epoch": 1.56596794081381, + "grad_norm": 2.444257229950801, + "learning_rate": 9.637537248536485e-06, + "loss": 0.6876, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1270 + }, + { + "epoch": 1.5672009864364982, + "grad_norm": 2.4130708431685197, + "learning_rate": 9.624362878811557e-06, + "loss": 0.5949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1271 + }, + { + "epoch": 1.5684340320591863, + "grad_norm": 2.340625006242772, + "learning_rate": 9.611189161946688e-06, + "loss": 0.6838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1272 + }, + { + "epoch": 1.5696670776818742, + "grad_norm": 2.2920633133032116, + "learning_rate": 9.598016120837894e-06, + "loss": 0.6186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1273 + }, + { + "epoch": 1.5709001233045623, + "grad_norm": 2.425912926280816, + "learning_rate": 9.584843778380022e-06, + "loss": 0.5922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1274 + }, + { + "epoch": 1.5721331689272504, + "grad_norm": 2.332147214152514, + "learning_rate": 9.571672157466695e-06, + "loss": 0.6555, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1275 + }, + { + "epoch": 1.5733662145499383, + "grad_norm": 2.398257339947792, + "learning_rate": 9.55850128099029e-06, + "loss": 0.6346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1276 + }, + { + "epoch": 1.5745992601726264, + "grad_norm": 2.142956121629841, + "learning_rate": 9.545331171841882e-06, + "loss": 0.5515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1277 + }, + { + "epoch": 1.5758323057953145, + "grad_norm": 2.407908540970335, + "learning_rate": 9.53216185291122e-06, + "loss": 0.5475, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1278 + }, + { + "epoch": 1.5770653514180024, + "grad_norm": 2.320052780540722, + "learning_rate": 9.518993347086676e-06, + "loss": 0.6044, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1279 + }, + { + "epoch": 1.5782983970406905, + "grad_norm": 2.4974178144916643, + "learning_rate": 9.505825677255205e-06, + "loss": 0.6436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1280 + }, + { + "epoch": 1.5795314426633786, + "grad_norm": 2.543298528231235, + "learning_rate": 9.492658866302325e-06, + "loss": 0.6388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1281 + }, + { + "epoch": 1.5807644882860665, + "grad_norm": 2.3663335712794917, + "learning_rate": 9.479492937112034e-06, + "loss": 0.5769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1282 + }, + { + "epoch": 1.5819975339087546, + "grad_norm": 2.301999998958795, + "learning_rate": 9.46632791256683e-06, + "loss": 0.6323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1283 + }, + { + "epoch": 1.5832305795314427, + "grad_norm": 2.3062144791204386, + "learning_rate": 9.453163815547604e-06, + "loss": 0.6727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1284 + }, + { + "epoch": 1.5844636251541306, + "grad_norm": 2.5906430609145947, + "learning_rate": 9.440000668933668e-06, + "loss": 0.7282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1285 + }, + { + "epoch": 1.5856966707768188, + "grad_norm": 2.6113606139990586, + "learning_rate": 9.426838495602659e-06, + "loss": 0.685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1286 + }, + { + "epoch": 1.5869297163995069, + "grad_norm": 2.1825712632265244, + "learning_rate": 9.413677318430532e-06, + "loss": 0.5556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1287 + }, + { + "epoch": 1.5881627620221948, + "grad_norm": 2.394162195116171, + "learning_rate": 9.400517160291511e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1288 + }, + { + "epoch": 1.5893958076448829, + "grad_norm": 2.375297383969579, + "learning_rate": 9.387358044058044e-06, + "loss": 0.7398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1289 + }, + { + "epoch": 1.590628853267571, + "grad_norm": 2.268014800016658, + "learning_rate": 9.374199992600775e-06, + "loss": 0.554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1290 + }, + { + "epoch": 1.5918618988902589, + "grad_norm": 2.273418089045509, + "learning_rate": 9.361043028788487e-06, + "loss": 0.5092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1291 + }, + { + "epoch": 1.593094944512947, + "grad_norm": 2.258287636448223, + "learning_rate": 9.347887175488086e-06, + "loss": 0.5517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1292 + }, + { + "epoch": 1.594327990135635, + "grad_norm": 2.4638493212600756, + "learning_rate": 9.334732455564538e-06, + "loss": 0.6536, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1293 + }, + { + "epoch": 1.595561035758323, + "grad_norm": 2.45854766908022, + "learning_rate": 9.321578891880838e-06, + "loss": 0.5922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1294 + }, + { + "epoch": 1.596794081381011, + "grad_norm": 2.6177980977782522, + "learning_rate": 9.30842650729799e-06, + "loss": 0.7473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1295 + }, + { + "epoch": 1.5980271270036992, + "grad_norm": 2.43914078165767, + "learning_rate": 9.295275324674915e-06, + "loss": 0.5646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1296 + }, + { + "epoch": 1.599260172626387, + "grad_norm": 2.365874131549762, + "learning_rate": 9.282125366868484e-06, + "loss": 0.5354, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1297 + }, + { + "epoch": 1.6004932182490752, + "grad_norm": 2.528721468130841, + "learning_rate": 9.268976656733402e-06, + "loss": 0.5554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1298 + }, + { + "epoch": 1.6017262638717633, + "grad_norm": 2.512092936139005, + "learning_rate": 9.255829217122237e-06, + "loss": 0.533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1299 + }, + { + "epoch": 1.6029593094944512, + "grad_norm": 2.354466866391755, + "learning_rate": 9.242683070885328e-06, + "loss": 0.4921, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1300 + }, + { + "epoch": 1.6041923551171393, + "grad_norm": 2.4883337985127936, + "learning_rate": 9.229538240870774e-06, + "loss": 0.6147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1301 + }, + { + "epoch": 1.6054254007398274, + "grad_norm": 2.4285611973298216, + "learning_rate": 9.21639474992439e-06, + "loss": 0.5665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1302 + }, + { + "epoch": 1.6066584463625153, + "grad_norm": 2.5959012805664994, + "learning_rate": 9.203252620889654e-06, + "loss": 0.6071, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1303 + }, + { + "epoch": 1.6078914919852034, + "grad_norm": 2.356030881254319, + "learning_rate": 9.190111876607684e-06, + "loss": 0.5597, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1304 + }, + { + "epoch": 1.6091245376078915, + "grad_norm": 2.4328199046350503, + "learning_rate": 9.176972539917188e-06, + "loss": 0.5543, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1305 + }, + { + "epoch": 1.6103575832305794, + "grad_norm": 2.531198496944925, + "learning_rate": 9.163834633654431e-06, + "loss": 0.7257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1306 + }, + { + "epoch": 1.6115906288532675, + "grad_norm": 2.4402117071794245, + "learning_rate": 9.150698180653193e-06, + "loss": 0.6173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1307 + }, + { + "epoch": 1.6128236744759556, + "grad_norm": 2.8953638685717547, + "learning_rate": 9.13756320374472e-06, + "loss": 0.7887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1308 + }, + { + "epoch": 1.6140567200986435, + "grad_norm": 2.459699675606036, + "learning_rate": 9.1244297257577e-06, + "loss": 0.7027, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1309 + }, + { + "epoch": 1.6152897657213316, + "grad_norm": 2.3328161668697467, + "learning_rate": 9.111297769518213e-06, + "loss": 0.5885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1310 + }, + { + "epoch": 1.6165228113440198, + "grad_norm": 2.4218060271919812, + "learning_rate": 9.098167357849699e-06, + "loss": 0.6497, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1311 + }, + { + "epoch": 1.6177558569667077, + "grad_norm": 2.294366317458183, + "learning_rate": 9.0850385135729e-06, + "loss": 0.5431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1312 + }, + { + "epoch": 1.6189889025893958, + "grad_norm": 2.3225490101301287, + "learning_rate": 9.071911259505852e-06, + "loss": 0.6644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1313 + }, + { + "epoch": 1.6202219482120839, + "grad_norm": 2.5880544489532826, + "learning_rate": 9.05878561846382e-06, + "loss": 0.517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1314 + }, + { + "epoch": 1.6214549938347718, + "grad_norm": 2.230629935973886, + "learning_rate": 9.045661613259254e-06, + "loss": 0.5261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1315 + }, + { + "epoch": 1.6226880394574599, + "grad_norm": 2.1770639270798307, + "learning_rate": 9.032539266701781e-06, + "loss": 0.4244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1316 + }, + { + "epoch": 1.623921085080148, + "grad_norm": 2.709845393499603, + "learning_rate": 9.019418601598129e-06, + "loss": 0.637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1317 + }, + { + "epoch": 1.6251541307028359, + "grad_norm": 3.460293995200764, + "learning_rate": 9.006299640752113e-06, + "loss": 0.6512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1318 + }, + { + "epoch": 1.6263871763255242, + "grad_norm": 3.5764897422793744, + "learning_rate": 8.993182406964579e-06, + "loss": 0.6457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1319 + }, + { + "epoch": 1.627620221948212, + "grad_norm": 2.5994177808868657, + "learning_rate": 8.980066923033378e-06, + "loss": 0.574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1320 + }, + { + "epoch": 1.6288532675709, + "grad_norm": 2.5757673207857135, + "learning_rate": 8.96695321175332e-06, + "loss": 0.4942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1321 + }, + { + "epoch": 1.6300863131935883, + "grad_norm": 2.453402170110872, + "learning_rate": 8.953841295916123e-06, + "loss": 0.6917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1322 + }, + { + "epoch": 1.6313193588162762, + "grad_norm": 2.654143412873541, + "learning_rate": 8.940731198310402e-06, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1323 + }, + { + "epoch": 1.632552404438964, + "grad_norm": 2.4250402090614935, + "learning_rate": 8.927622941721594e-06, + "loss": 0.6999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1324 + }, + { + "epoch": 1.6337854500616524, + "grad_norm": 2.761651051842012, + "learning_rate": 8.91451654893195e-06, + "loss": 0.6075, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1325 + }, + { + "epoch": 1.6350184956843403, + "grad_norm": 2.6957159420077894, + "learning_rate": 8.901412042720475e-06, + "loss": 0.6084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1326 + }, + { + "epoch": 1.6362515413070282, + "grad_norm": 2.2293950670646745, + "learning_rate": 8.888309445862892e-06, + "loss": 0.7359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1327 + }, + { + "epoch": 1.6374845869297165, + "grad_norm": 2.3226832325484663, + "learning_rate": 8.87520878113162e-06, + "loss": 0.5255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1328 + }, + { + "epoch": 1.6387176325524044, + "grad_norm": 2.5216170149103583, + "learning_rate": 8.8621100712957e-06, + "loss": 0.6287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1329 + }, + { + "epoch": 1.6399506781750923, + "grad_norm": 2.685061433529547, + "learning_rate": 8.849013339120793e-06, + "loss": 0.7093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1330 + }, + { + "epoch": 1.6411837237977807, + "grad_norm": 2.5648896854812975, + "learning_rate": 8.835918607369109e-06, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1331 + }, + { + "epoch": 1.6424167694204685, + "grad_norm": 2.6253257359858404, + "learning_rate": 8.822825898799393e-06, + "loss": 0.7446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1332 + }, + { + "epoch": 1.6436498150431564, + "grad_norm": 2.274148046399076, + "learning_rate": 8.809735236166864e-06, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1333 + }, + { + "epoch": 1.6448828606658448, + "grad_norm": 2.3764316480496537, + "learning_rate": 8.796646642223192e-06, + "loss": 0.6847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1334 + }, + { + "epoch": 1.6461159062885327, + "grad_norm": 2.3471176710530024, + "learning_rate": 8.783560139716449e-06, + "loss": 0.5461, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1335 + }, + { + "epoch": 1.6473489519112205, + "grad_norm": 2.4255652189713715, + "learning_rate": 8.770475751391066e-06, + "loss": 0.4423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1336 + }, + { + "epoch": 1.6485819975339089, + "grad_norm": 2.5292711557625673, + "learning_rate": 8.757393499987815e-06, + "loss": 0.7064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1337 + }, + { + "epoch": 1.6498150431565968, + "grad_norm": 2.427296462933323, + "learning_rate": 8.744313408243737e-06, + "loss": 0.6039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1338 + }, + { + "epoch": 1.6510480887792849, + "grad_norm": 2.5975517369534105, + "learning_rate": 8.731235498892127e-06, + "loss": 0.6391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1339 + }, + { + "epoch": 1.652281134401973, + "grad_norm": 2.491354995578972, + "learning_rate": 8.718159794662498e-06, + "loss": 0.6087, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1340 + }, + { + "epoch": 1.6535141800246609, + "grad_norm": 2.4884148521815526, + "learning_rate": 8.705086318280505e-06, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1341 + }, + { + "epoch": 1.654747225647349, + "grad_norm": 2.8396628177391907, + "learning_rate": 8.692015092467956e-06, + "loss": 0.7335, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1342 + }, + { + "epoch": 1.655980271270037, + "grad_norm": 2.7368157959861557, + "learning_rate": 8.678946139942728e-06, + "loss": 0.7007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1343 + }, + { + "epoch": 1.657213316892725, + "grad_norm": 2.3611404478863873, + "learning_rate": 8.665879483418766e-06, + "loss": 0.5668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1344 + }, + { + "epoch": 1.658446362515413, + "grad_norm": 2.191668075058394, + "learning_rate": 8.652815145606003e-06, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1345 + }, + { + "epoch": 1.6596794081381012, + "grad_norm": 2.14788648268729, + "learning_rate": 8.63975314921036e-06, + "loss": 0.5166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1346 + }, + { + "epoch": 1.660912453760789, + "grad_norm": 2.3068777176506963, + "learning_rate": 8.626693516933687e-06, + "loss": 0.5394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1347 + }, + { + "epoch": 1.6621454993834772, + "grad_norm": 2.49955124289679, + "learning_rate": 8.613636271473712e-06, + "loss": 0.6493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1348 + }, + { + "epoch": 1.6633785450061653, + "grad_norm": 2.1471304936125652, + "learning_rate": 8.60058143552403e-06, + "loss": 0.5408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1349 + }, + { + "epoch": 1.6646115906288532, + "grad_norm": 2.3378284654154444, + "learning_rate": 8.587529031774033e-06, + "loss": 0.5559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1350 + }, + { + "epoch": 1.6658446362515413, + "grad_norm": 2.272350354054125, + "learning_rate": 8.574479082908903e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1351 + }, + { + "epoch": 1.6670776818742294, + "grad_norm": 2.6718347888114393, + "learning_rate": 8.561431611609537e-06, + "loss": 0.6518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1352 + }, + { + "epoch": 1.6683107274969173, + "grad_norm": 2.5683646046012307, + "learning_rate": 8.548386640552542e-06, + "loss": 0.627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1353 + }, + { + "epoch": 1.6695437731196054, + "grad_norm": 2.429950074454179, + "learning_rate": 8.535344192410173e-06, + "loss": 0.6244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1354 + }, + { + "epoch": 1.6707768187422936, + "grad_norm": 2.459482147364562, + "learning_rate": 8.522304289850298e-06, + "loss": 0.5089, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1355 + }, + { + "epoch": 1.6720098643649814, + "grad_norm": 2.5078184456517802, + "learning_rate": 8.509266955536364e-06, + "loss": 0.5966, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1356 + }, + { + "epoch": 1.6732429099876696, + "grad_norm": 2.8738369967418786, + "learning_rate": 8.49623221212735e-06, + "loss": 0.7568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1357 + }, + { + "epoch": 1.6744759556103577, + "grad_norm": 2.566639808326863, + "learning_rate": 8.483200082277743e-06, + "loss": 0.6565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1358 + }, + { + "epoch": 1.6757090012330456, + "grad_norm": 2.18616861232984, + "learning_rate": 8.47017058863747e-06, + "loss": 0.5219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1359 + }, + { + "epoch": 1.6769420468557337, + "grad_norm": 2.2874183385113387, + "learning_rate": 8.457143753851893e-06, + "loss": 0.515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1360 + }, + { + "epoch": 1.6781750924784218, + "grad_norm": 2.3279389147190503, + "learning_rate": 8.444119600561746e-06, + "loss": 0.5737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1361 + }, + { + "epoch": 1.6794081381011097, + "grad_norm": 2.510085283064121, + "learning_rate": 8.431098151403098e-06, + "loss": 0.5527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1362 + }, + { + "epoch": 1.6806411837237978, + "grad_norm": 2.350766304372499, + "learning_rate": 8.41807942900733e-06, + "loss": 0.5444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1363 + }, + { + "epoch": 1.681874229346486, + "grad_norm": 2.4637839990972266, + "learning_rate": 8.405063456001069e-06, + "loss": 0.617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1364 + }, + { + "epoch": 1.6831072749691738, + "grad_norm": 2.3520486778455303, + "learning_rate": 8.392050255006172e-06, + "loss": 0.5787, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1365 + }, + { + "epoch": 1.684340320591862, + "grad_norm": 2.6951378359823064, + "learning_rate": 8.379039848639685e-06, + "loss": 0.5679, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1366 + }, + { + "epoch": 1.68557336621455, + "grad_norm": 2.5680249855790533, + "learning_rate": 8.366032259513782e-06, + "loss": 0.5366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1367 + }, + { + "epoch": 1.686806411837238, + "grad_norm": 2.7844067151133496, + "learning_rate": 8.35302751023575e-06, + "loss": 0.6181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1368 + }, + { + "epoch": 1.688039457459926, + "grad_norm": 2.3271421104987895, + "learning_rate": 8.340025623407938e-06, + "loss": 0.4529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1369 + }, + { + "epoch": 1.6892725030826141, + "grad_norm": 2.7324858411976822, + "learning_rate": 8.327026621627722e-06, + "loss": 0.7326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1370 + }, + { + "epoch": 1.690505548705302, + "grad_norm": 2.5720836300974383, + "learning_rate": 8.31403052748746e-06, + "loss": 0.6784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1371 + }, + { + "epoch": 1.6917385943279901, + "grad_norm": 2.440765245099682, + "learning_rate": 8.301037363574454e-06, + "loss": 0.6014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1372 + }, + { + "epoch": 1.6929716399506782, + "grad_norm": 2.594069817186588, + "learning_rate": 8.288047152470931e-06, + "loss": 0.5856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1373 + }, + { + "epoch": 1.6942046855733661, + "grad_norm": 2.4701026222170452, + "learning_rate": 8.27505991675396e-06, + "loss": 0.7036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1374 + }, + { + "epoch": 1.6954377311960542, + "grad_norm": 2.3893004984314365, + "learning_rate": 8.26207567899546e-06, + "loss": 0.5404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1375 + }, + { + "epoch": 1.6966707768187423, + "grad_norm": 2.3404547915407656, + "learning_rate": 8.249094461762124e-06, + "loss": 0.5142, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1376 + }, + { + "epoch": 1.6979038224414302, + "grad_norm": 2.4832770686344556, + "learning_rate": 8.236116287615409e-06, + "loss": 0.6183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1377 + }, + { + "epoch": 1.6991368680641183, + "grad_norm": 2.262237110861396, + "learning_rate": 8.223141179111471e-06, + "loss": 0.4504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1378 + }, + { + "epoch": 1.7003699136868065, + "grad_norm": 2.62998506842995, + "learning_rate": 8.210169158801146e-06, + "loss": 0.6316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1379 + }, + { + "epoch": 1.7016029593094943, + "grad_norm": 2.2853273497892164, + "learning_rate": 8.197200249229902e-06, + "loss": 0.5379, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1380 + }, + { + "epoch": 1.7028360049321825, + "grad_norm": 2.489883484033712, + "learning_rate": 8.184234472937795e-06, + "loss": 0.5635, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1381 + }, + { + "epoch": 1.7040690505548706, + "grad_norm": 2.4399096703479723, + "learning_rate": 8.171271852459442e-06, + "loss": 0.5672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1382 + }, + { + "epoch": 1.7053020961775585, + "grad_norm": 2.5877064849336673, + "learning_rate": 8.158312410323968e-06, + "loss": 0.56, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1383 + }, + { + "epoch": 1.7065351418002466, + "grad_norm": 2.5080025813914486, + "learning_rate": 8.145356169054982e-06, + "loss": 0.6415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1384 + }, + { + "epoch": 1.7077681874229347, + "grad_norm": 2.474550610636856, + "learning_rate": 8.132403151170524e-06, + "loss": 0.5209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1385 + }, + { + "epoch": 1.7090012330456226, + "grad_norm": 2.4546793222588037, + "learning_rate": 8.11945337918303e-06, + "loss": 0.6133, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1386 + }, + { + "epoch": 1.7102342786683107, + "grad_norm": 2.3960818942570623, + "learning_rate": 8.106506875599312e-06, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1387 + }, + { + "epoch": 1.7114673242909988, + "grad_norm": 2.323552505571765, + "learning_rate": 8.09356366292047e-06, + "loss": 0.5402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1388 + }, + { + "epoch": 1.7127003699136867, + "grad_norm": 2.088694863039824, + "learning_rate": 8.080623763641916e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1389 + }, + { + "epoch": 1.7139334155363748, + "grad_norm": 2.171047701838099, + "learning_rate": 8.067687200253281e-06, + "loss": 0.4947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1390 + }, + { + "epoch": 1.715166461159063, + "grad_norm": 2.4534573577067165, + "learning_rate": 8.054753995238415e-06, + "loss": 0.6494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1391 + }, + { + "epoch": 1.7163995067817508, + "grad_norm": 2.315613179007512, + "learning_rate": 8.04182417107531e-06, + "loss": 0.605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1392 + }, + { + "epoch": 1.7176325524044391, + "grad_norm": 2.382282194715041, + "learning_rate": 8.028897750236106e-06, + "loss": 0.3862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1393 + }, + { + "epoch": 1.718865598027127, + "grad_norm": 2.3689104564582606, + "learning_rate": 8.015974755187015e-06, + "loss": 0.4825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1394 + }, + { + "epoch": 1.720098643649815, + "grad_norm": 2.5004960798500098, + "learning_rate": 8.003055208388292e-06, + "loss": 0.5614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1395 + }, + { + "epoch": 1.7213316892725032, + "grad_norm": 2.376364766941311, + "learning_rate": 7.990139132294212e-06, + "loss": 0.4446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1396 + }, + { + "epoch": 1.7225647348951911, + "grad_norm": 2.6246902766612745, + "learning_rate": 7.977226549353e-06, + "loss": 0.6399, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1397 + }, + { + "epoch": 1.723797780517879, + "grad_norm": 2.484606542490508, + "learning_rate": 7.964317482006824e-06, + "loss": 0.5081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1398 + }, + { + "epoch": 1.7250308261405674, + "grad_norm": 3.38389227109005, + "learning_rate": 7.951411952691739e-06, + "loss": 0.755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1399 + }, + { + "epoch": 1.7262638717632552, + "grad_norm": 2.578597094452948, + "learning_rate": 7.938509983837647e-06, + "loss": 0.4826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1400 + }, + { + "epoch": 1.7274969173859431, + "grad_norm": 2.4802210829330105, + "learning_rate": 7.925611597868266e-06, + "loss": 0.493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1401 + }, + { + "epoch": 1.7287299630086315, + "grad_norm": 2.4523105300960517, + "learning_rate": 7.91271681720108e-06, + "loss": 0.6206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1402 + }, + { + "epoch": 1.7299630086313194, + "grad_norm": 2.386348175383016, + "learning_rate": 7.899825664247322e-06, + "loss": 0.5701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1403 + }, + { + "epoch": 1.7311960542540072, + "grad_norm": 2.4083074706971375, + "learning_rate": 7.886938161411897e-06, + "loss": 0.5884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1404 + }, + { + "epoch": 1.7324290998766956, + "grad_norm": 2.5069557743624316, + "learning_rate": 7.874054331093389e-06, + "loss": 0.4971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1405 + }, + { + "epoch": 1.7336621454993835, + "grad_norm": 2.55856356518894, + "learning_rate": 7.861174195683987e-06, + "loss": 0.5018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1406 + }, + { + "epoch": 1.7348951911220714, + "grad_norm": 2.568692520564215, + "learning_rate": 7.848297777569458e-06, + "loss": 0.6251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1407 + }, + { + "epoch": 1.7361282367447597, + "grad_norm": 2.626547699354354, + "learning_rate": 7.835425099129116e-06, + "loss": 0.4866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1408 + }, + { + "epoch": 1.7373612823674476, + "grad_norm": 2.655578554031113, + "learning_rate": 7.822556182735762e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1409 + }, + { + "epoch": 1.7385943279901355, + "grad_norm": 2.514238469731576, + "learning_rate": 7.809691050755676e-06, + "loss": 0.529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1410 + }, + { + "epoch": 1.7398273736128238, + "grad_norm": 2.486688645388684, + "learning_rate": 7.796829725548543e-06, + "loss": 0.5598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1411 + }, + { + "epoch": 1.7410604192355117, + "grad_norm": 2.5300619846568755, + "learning_rate": 7.783972229467443e-06, + "loss": 0.6431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1412 + }, + { + "epoch": 1.7422934648581998, + "grad_norm": 2.782005331836491, + "learning_rate": 7.771118584858801e-06, + "loss": 0.7072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1413 + }, + { + "epoch": 1.743526510480888, + "grad_norm": 2.4671627637221296, + "learning_rate": 7.758268814062343e-06, + "loss": 0.6404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1414 + }, + { + "epoch": 1.7447595561035758, + "grad_norm": 2.1612453871779285, + "learning_rate": 7.745422939411064e-06, + "loss": 0.4505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1415 + }, + { + "epoch": 1.745992601726264, + "grad_norm": 2.296774520080563, + "learning_rate": 7.73258098323119e-06, + "loss": 0.5345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1416 + }, + { + "epoch": 1.747225647348952, + "grad_norm": 2.6549737125780326, + "learning_rate": 7.719742967842133e-06, + "loss": 0.6232, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1417 + }, + { + "epoch": 1.74845869297164, + "grad_norm": 2.343040089887724, + "learning_rate": 7.706908915556457e-06, + "loss": 0.4859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1418 + }, + { + "epoch": 1.749691738594328, + "grad_norm": 2.459027338987637, + "learning_rate": 7.694078848679836e-06, + "loss": 0.6424, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1419 + }, + { + "epoch": 1.7509247842170161, + "grad_norm": 2.6006881140576383, + "learning_rate": 7.681252789511029e-06, + "loss": 0.6836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1420 + }, + { + "epoch": 1.752157829839704, + "grad_norm": 2.813687773910776, + "learning_rate": 7.668430760341813e-06, + "loss": 0.6596, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1421 + }, + { + "epoch": 1.7533908754623921, + "grad_norm": 2.7501988552768166, + "learning_rate": 7.655612783456973e-06, + "loss": 0.5757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1422 + }, + { + "epoch": 1.7546239210850803, + "grad_norm": 2.602564686174615, + "learning_rate": 7.64279888113424e-06, + "loss": 0.6304, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1423 + }, + { + "epoch": 1.7558569667077681, + "grad_norm": 2.2125559792606606, + "learning_rate": 7.629989075644277e-06, + "loss": 0.4345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1424 + }, + { + "epoch": 1.7570900123304563, + "grad_norm": 2.552624039521638, + "learning_rate": 7.6171833892506145e-06, + "loss": 0.6566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1425 + }, + { + "epoch": 1.7583230579531444, + "grad_norm": 2.53122973187578, + "learning_rate": 7.60438184420963e-06, + "loss": 0.5989, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1426 + }, + { + "epoch": 1.7595561035758323, + "grad_norm": 2.4914845101760834, + "learning_rate": 7.591584462770506e-06, + "loss": 0.5674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1427 + }, + { + "epoch": 1.7607891491985204, + "grad_norm": 2.71358370673903, + "learning_rate": 7.57879126717518e-06, + "loss": 0.6321, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1428 + }, + { + "epoch": 1.7620221948212085, + "grad_norm": 2.5302056277845413, + "learning_rate": 7.566002279658322e-06, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1429 + }, + { + "epoch": 1.7632552404438964, + "grad_norm": 2.371803209433193, + "learning_rate": 7.553217522447285e-06, + "loss": 0.4999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1430 + }, + { + "epoch": 1.7644882860665845, + "grad_norm": 2.527610014870494, + "learning_rate": 7.540437017762068e-06, + "loss": 0.5314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1431 + }, + { + "epoch": 1.7657213316892726, + "grad_norm": 2.440388163849852, + "learning_rate": 7.52766078781529e-06, + "loss": 0.6007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1432 + }, + { + "epoch": 1.7669543773119605, + "grad_norm": 2.369028343720501, + "learning_rate": 7.51488885481212e-06, + "loss": 0.5996, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1433 + }, + { + "epoch": 1.7681874229346486, + "grad_norm": 2.259203085653468, + "learning_rate": 7.502121240950284e-06, + "loss": 0.5022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1434 + }, + { + "epoch": 1.7694204685573367, + "grad_norm": 2.346742207074351, + "learning_rate": 7.489357968419975e-06, + "loss": 0.5232, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1435 + }, + { + "epoch": 1.7706535141800246, + "grad_norm": 2.3571911434438926, + "learning_rate": 7.476599059403866e-06, + "loss": 0.5529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1436 + }, + { + "epoch": 1.7718865598027127, + "grad_norm": 2.564577423619225, + "learning_rate": 7.463844536077021e-06, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1437 + }, + { + "epoch": 1.7731196054254008, + "grad_norm": 2.3757451812868062, + "learning_rate": 7.451094420606906e-06, + "loss": 0.5998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1438 + }, + { + "epoch": 1.7743526510480887, + "grad_norm": 2.3897936394773542, + "learning_rate": 7.438348735153311e-06, + "loss": 0.5233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1439 + }, + { + "epoch": 1.7755856966707768, + "grad_norm": 2.384296664394737, + "learning_rate": 7.425607501868328e-06, + "loss": 0.4798, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1440 + }, + { + "epoch": 1.776818742293465, + "grad_norm": 2.326128061878694, + "learning_rate": 7.41287074289632e-06, + "loss": 0.5628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1441 + }, + { + "epoch": 1.7780517879161528, + "grad_norm": 2.3448649893975646, + "learning_rate": 7.4001384803738595e-06, + "loss": 0.5539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1442 + }, + { + "epoch": 1.779284833538841, + "grad_norm": 2.335303557427351, + "learning_rate": 7.387410736429717e-06, + "loss": 0.463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1443 + }, + { + "epoch": 1.780517879161529, + "grad_norm": 2.5062756692562815, + "learning_rate": 7.374687533184801e-06, + "loss": 0.5804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1444 + }, + { + "epoch": 1.781750924784217, + "grad_norm": 2.667998142301554, + "learning_rate": 7.361968892752134e-06, + "loss": 0.5044, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1445 + }, + { + "epoch": 1.782983970406905, + "grad_norm": 2.8938425700505856, + "learning_rate": 7.349254837236808e-06, + "loss": 0.7082, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1446 + }, + { + "epoch": 1.7842170160295932, + "grad_norm": 2.638110967591072, + "learning_rate": 7.336545388735936e-06, + "loss": 0.5165, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1447 + }, + { + "epoch": 1.785450061652281, + "grad_norm": 2.5434717078771314, + "learning_rate": 7.323840569338645e-06, + "loss": 0.5274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1448 + }, + { + "epoch": 1.7866831072749692, + "grad_norm": 2.880952060159319, + "learning_rate": 7.31114040112599e-06, + "loss": 0.6319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1449 + }, + { + "epoch": 1.7879161528976573, + "grad_norm": 2.7232472895100974, + "learning_rate": 7.29844490617097e-06, + "loss": 0.6471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1450 + }, + { + "epoch": 1.7891491985203452, + "grad_norm": 2.6289292491069234, + "learning_rate": 7.285754106538435e-06, + "loss": 0.6144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1451 + }, + { + "epoch": 1.7903822441430333, + "grad_norm": 2.3571805040637117, + "learning_rate": 7.273068024285096e-06, + "loss": 0.5023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1452 + }, + { + "epoch": 1.7916152897657214, + "grad_norm": 2.5462291798096115, + "learning_rate": 7.260386681459457e-06, + "loss": 0.594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1453 + }, + { + "epoch": 1.7928483353884093, + "grad_norm": 2.364391159514722, + "learning_rate": 7.247710100101778e-06, + "loss": 0.3867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1454 + }, + { + "epoch": 1.7940813810110974, + "grad_norm": 2.557120216564179, + "learning_rate": 7.235038302244058e-06, + "loss": 0.4541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1455 + }, + { + "epoch": 1.7953144266337855, + "grad_norm": 2.3515150176521398, + "learning_rate": 7.2223713099099655e-06, + "loss": 0.5005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1456 + }, + { + "epoch": 1.7965474722564734, + "grad_norm": 3.0149925007019465, + "learning_rate": 7.209709145114834e-06, + "loss": 0.6562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1457 + }, + { + "epoch": 1.7977805178791615, + "grad_norm": 2.9385786617892484, + "learning_rate": 7.197051829865591e-06, + "loss": 0.7172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1458 + }, + { + "epoch": 1.7990135635018496, + "grad_norm": 2.4438200151823577, + "learning_rate": 7.1843993861607485e-06, + "loss": 0.545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1459 + }, + { + "epoch": 1.8002466091245375, + "grad_norm": 2.189016598422486, + "learning_rate": 7.171751835990346e-06, + "loss": 0.3945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1460 + }, + { + "epoch": 1.8014796547472256, + "grad_norm": 2.599380048646185, + "learning_rate": 7.159109201335914e-06, + "loss": 0.6221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1461 + }, + { + "epoch": 1.8027127003699137, + "grad_norm": 2.3218284667005045, + "learning_rate": 7.146471504170448e-06, + "loss": 0.4081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1462 + }, + { + "epoch": 1.8039457459926016, + "grad_norm": 2.483257454081904, + "learning_rate": 7.1338387664583545e-06, + "loss": 0.5952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1463 + }, + { + "epoch": 1.8051787916152897, + "grad_norm": 2.3060282168534267, + "learning_rate": 7.1212110101554225e-06, + "loss": 0.478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1464 + }, + { + "epoch": 1.8064118372379778, + "grad_norm": 2.353277914016821, + "learning_rate": 7.108588257208793e-06, + "loss": 0.4602, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1465 + }, + { + "epoch": 1.8076448828606657, + "grad_norm": 2.402021352720206, + "learning_rate": 7.095970529556895e-06, + "loss": 0.423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1466 + }, + { + "epoch": 1.808877928483354, + "grad_norm": 2.4089157744395187, + "learning_rate": 7.083357849129437e-06, + "loss": 0.4962, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1467 + }, + { + "epoch": 1.810110974106042, + "grad_norm": 2.4445181455084803, + "learning_rate": 7.070750237847344e-06, + "loss": 0.4237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1468 + }, + { + "epoch": 1.8113440197287298, + "grad_norm": 2.5896452836545576, + "learning_rate": 7.058147717622742e-06, + "loss": 0.6034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1469 + }, + { + "epoch": 1.8125770653514182, + "grad_norm": 2.5692451627366215, + "learning_rate": 7.0455503103588975e-06, + "loss": 0.4897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1470 + }, + { + "epoch": 1.813810110974106, + "grad_norm": 2.5193270914325185, + "learning_rate": 7.0329580379502e-06, + "loss": 0.5908, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1471 + }, + { + "epoch": 1.815043156596794, + "grad_norm": 2.6722250129620586, + "learning_rate": 7.020370922282113e-06, + "loss": 0.6295, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1472 + }, + { + "epoch": 1.8162762022194823, + "grad_norm": 2.4547108738067838, + "learning_rate": 7.00778898523113e-06, + "loss": 0.5357, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1473 + }, + { + "epoch": 1.8175092478421702, + "grad_norm": 2.437273516415158, + "learning_rate": 6.995212248664755e-06, + "loss": 0.5101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1474 + }, + { + "epoch": 1.818742293464858, + "grad_norm": 2.279547074068301, + "learning_rate": 6.982640734441443e-06, + "loss": 0.5079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1475 + }, + { + "epoch": 1.8199753390875464, + "grad_norm": 2.435539711925499, + "learning_rate": 6.9700744644105815e-06, + "loss": 0.5245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1476 + }, + { + "epoch": 1.8212083847102343, + "grad_norm": 2.420190205704888, + "learning_rate": 6.9575134604124364e-06, + "loss": 0.5115, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1477 + }, + { + "epoch": 1.8224414303329222, + "grad_norm": 2.6220274719559424, + "learning_rate": 6.9449577442781246e-06, + "loss": 0.5708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1478 + }, + { + "epoch": 1.8236744759556105, + "grad_norm": 2.4198151981250624, + "learning_rate": 6.932407337829579e-06, + "loss": 0.5779, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1479 + }, + { + "epoch": 1.8249075215782984, + "grad_norm": 2.710527668536825, + "learning_rate": 6.919862262879488e-06, + "loss": 0.706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1480 + }, + { + "epoch": 1.8261405672009863, + "grad_norm": 2.260314185301613, + "learning_rate": 6.907322541231294e-06, + "loss": 0.3152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1481 + }, + { + "epoch": 1.8273736128236746, + "grad_norm": 2.4880601560363087, + "learning_rate": 6.894788194679112e-06, + "loss": 0.4559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1482 + }, + { + "epoch": 1.8286066584463625, + "grad_norm": 2.53859220201902, + "learning_rate": 6.882259245007741e-06, + "loss": 0.6073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1483 + }, + { + "epoch": 1.8298397040690506, + "grad_norm": 2.406753699086694, + "learning_rate": 6.869735713992576e-06, + "loss": 0.5376, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1484 + }, + { + "epoch": 1.8310727496917387, + "grad_norm": 2.5573243142415616, + "learning_rate": 6.857217623399613e-06, + "loss": 0.5161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1485 + }, + { + "epoch": 1.8323057953144266, + "grad_norm": 2.4701507919473964, + "learning_rate": 6.844704994985383e-06, + "loss": 0.4962, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1486 + }, + { + "epoch": 1.8335388409371147, + "grad_norm": 2.505190115969386, + "learning_rate": 6.832197850496922e-06, + "loss": 0.5941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1487 + }, + { + "epoch": 1.8347718865598028, + "grad_norm": 2.8094520017647766, + "learning_rate": 6.819696211671745e-06, + "loss": 0.6998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1488 + }, + { + "epoch": 1.8360049321824907, + "grad_norm": 2.7116637712221685, + "learning_rate": 6.807200100237786e-06, + "loss": 0.5916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1489 + }, + { + "epoch": 1.8372379778051788, + "grad_norm": 2.4742606262555538, + "learning_rate": 6.794709537913382e-06, + "loss": 0.4571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1490 + }, + { + "epoch": 1.838471023427867, + "grad_norm": 2.3388179882861038, + "learning_rate": 6.782224546407217e-06, + "loss": 0.4336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1491 + }, + { + "epoch": 1.8397040690505548, + "grad_norm": 2.5535368655522723, + "learning_rate": 6.769745147418298e-06, + "loss": 0.5513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1492 + }, + { + "epoch": 1.840937114673243, + "grad_norm": 2.3844296720019136, + "learning_rate": 6.75727136263592e-06, + "loss": 0.4769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1493 + }, + { + "epoch": 1.842170160295931, + "grad_norm": 2.430982238426814, + "learning_rate": 6.744803213739599e-06, + "loss": 0.6191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1494 + }, + { + "epoch": 1.843403205918619, + "grad_norm": 2.5719309543455484, + "learning_rate": 6.732340722399083e-06, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1495 + }, + { + "epoch": 1.844636251541307, + "grad_norm": 2.4358254754188247, + "learning_rate": 6.719883910274258e-06, + "loss": 0.6404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1496 + }, + { + "epoch": 1.8458692971639952, + "grad_norm": 2.2277422785314385, + "learning_rate": 6.7074327990151666e-06, + "loss": 0.4467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1497 + }, + { + "epoch": 1.847102342786683, + "grad_norm": 2.3553522334648234, + "learning_rate": 6.694987410261927e-06, + "loss": 0.4733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1498 + }, + { + "epoch": 1.8483353884093712, + "grad_norm": 2.305268243528217, + "learning_rate": 6.6825477656447135e-06, + "loss": 0.5213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1499 + }, + { + "epoch": 1.8495684340320593, + "grad_norm": 2.476358939717476, + "learning_rate": 6.6701138867837245e-06, + "loss": 0.5979, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1500 + }, + { + "epoch": 1.8508014796547472, + "grad_norm": 2.5413097090235897, + "learning_rate": 6.657685795289126e-06, + "loss": 0.464, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1501 + }, + { + "epoch": 1.8520345252774353, + "grad_norm": 2.581561221486171, + "learning_rate": 6.6452635127610366e-06, + "loss": 0.5169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1502 + }, + { + "epoch": 1.8532675709001234, + "grad_norm": 2.752939514877913, + "learning_rate": 6.632847060789471e-06, + "loss": 0.5458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1503 + }, + { + "epoch": 1.8545006165228113, + "grad_norm": 2.393983614231186, + "learning_rate": 6.620436460954314e-06, + "loss": 0.5606, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1504 + }, + { + "epoch": 1.8557336621454994, + "grad_norm": 2.2296540462723415, + "learning_rate": 6.608031734825282e-06, + "loss": 0.5184, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1505 + }, + { + "epoch": 1.8569667077681875, + "grad_norm": 2.3150895635322377, + "learning_rate": 6.595632903961876e-06, + "loss": 0.3391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1506 + }, + { + "epoch": 1.8581997533908754, + "grad_norm": 2.4812557469058687, + "learning_rate": 6.583239989913357e-06, + "loss": 0.5283, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1507 + }, + { + "epoch": 1.8594327990135635, + "grad_norm": 2.240249157372891, + "learning_rate": 6.570853014218699e-06, + "loss": 0.4267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1508 + }, + { + "epoch": 1.8606658446362516, + "grad_norm": 2.418565073853559, + "learning_rate": 6.558471998406556e-06, + "loss": 0.5317, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1509 + }, + { + "epoch": 1.8618988902589395, + "grad_norm": 2.509766709851647, + "learning_rate": 6.546096963995221e-06, + "loss": 0.5639, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1510 + }, + { + "epoch": 1.8631319358816276, + "grad_norm": 2.6229591577559606, + "learning_rate": 6.5337279324925995e-06, + "loss": 0.6702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1511 + }, + { + "epoch": 1.8643649815043157, + "grad_norm": 2.8436863684007605, + "learning_rate": 6.521364925396159e-06, + "loss": 0.6629, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1512 + }, + { + "epoch": 1.8655980271270036, + "grad_norm": 2.88285582890902, + "learning_rate": 6.50900796419289e-06, + "loss": 0.6007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1513 + }, + { + "epoch": 1.8668310727496917, + "grad_norm": 2.4591483237884533, + "learning_rate": 6.496657070359289e-06, + "loss": 0.5649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1514 + }, + { + "epoch": 1.8680641183723798, + "grad_norm": 2.588461757043, + "learning_rate": 6.484312265361294e-06, + "loss": 0.6014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1515 + }, + { + "epoch": 1.8692971639950677, + "grad_norm": 2.3915380277510256, + "learning_rate": 6.47197357065427e-06, + "loss": 0.5105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1516 + }, + { + "epoch": 1.8705302096177558, + "grad_norm": 2.4853884668402597, + "learning_rate": 6.459641007682953e-06, + "loss": 0.5631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1517 + }, + { + "epoch": 1.871763255240444, + "grad_norm": 2.1904070749761546, + "learning_rate": 6.447314597881433e-06, + "loss": 0.4229, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1518 + }, + { + "epoch": 1.8729963008631318, + "grad_norm": 2.392397858425607, + "learning_rate": 6.4349943626731e-06, + "loss": 0.5323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1519 + }, + { + "epoch": 1.87422934648582, + "grad_norm": 2.5868871798442936, + "learning_rate": 6.422680323470609e-06, + "loss": 0.5943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1520 + }, + { + "epoch": 1.875462392108508, + "grad_norm": 2.8171109508645515, + "learning_rate": 6.410372501675854e-06, + "loss": 0.6011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1521 + }, + { + "epoch": 1.876695437731196, + "grad_norm": 2.5043723972276406, + "learning_rate": 6.398070918679917e-06, + "loss": 0.6401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1522 + }, + { + "epoch": 1.877928483353884, + "grad_norm": 2.2319121044157075, + "learning_rate": 6.3857755958630415e-06, + "loss": 0.5526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1523 + }, + { + "epoch": 1.8791615289765722, + "grad_norm": 2.668249861016908, + "learning_rate": 6.3734865545945834e-06, + "loss": 0.6208, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1524 + }, + { + "epoch": 1.88039457459926, + "grad_norm": 2.328498736662863, + "learning_rate": 6.361203816232988e-06, + "loss": 0.4526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1525 + }, + { + "epoch": 1.8816276202219482, + "grad_norm": 2.532493722492898, + "learning_rate": 6.348927402125753e-06, + "loss": 0.5397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1526 + }, + { + "epoch": 1.8828606658446363, + "grad_norm": 2.4124669695000023, + "learning_rate": 6.3366573336093605e-06, + "loss": 0.3821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1527 + }, + { + "epoch": 1.8840937114673242, + "grad_norm": 2.328411541073336, + "learning_rate": 6.324393632009295e-06, + "loss": 0.5093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1528 + }, + { + "epoch": 1.8853267570900123, + "grad_norm": 2.5365227172099196, + "learning_rate": 6.312136318639945e-06, + "loss": 0.5598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1529 + }, + { + "epoch": 1.8865598027127004, + "grad_norm": 2.607793778617426, + "learning_rate": 6.299885414804618e-06, + "loss": 0.5194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1530 + }, + { + "epoch": 1.8877928483353883, + "grad_norm": 2.335857326269118, + "learning_rate": 6.287640941795479e-06, + "loss": 0.3871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1531 + }, + { + "epoch": 1.8890258939580764, + "grad_norm": 2.647531529113461, + "learning_rate": 6.275402920893503e-06, + "loss": 0.521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1532 + }, + { + "epoch": 1.8902589395807645, + "grad_norm": 2.9804935503220285, + "learning_rate": 6.263171373368464e-06, + "loss": 0.5535, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1533 + }, + { + "epoch": 1.8914919852034524, + "grad_norm": 2.4613526931496215, + "learning_rate": 6.250946320478879e-06, + "loss": 0.3987, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1534 + }, + { + "epoch": 1.8927250308261405, + "grad_norm": 2.694644539489535, + "learning_rate": 6.238727783471983e-06, + "loss": 0.5107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1535 + }, + { + "epoch": 1.8939580764488286, + "grad_norm": 2.516795775084936, + "learning_rate": 6.2265157835836775e-06, + "loss": 0.3624, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1536 + }, + { + "epoch": 1.8951911220715165, + "grad_norm": 2.893994117314153, + "learning_rate": 6.214310342038507e-06, + "loss": 0.5791, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1537 + }, + { + "epoch": 1.8964241676942046, + "grad_norm": 2.773382488051452, + "learning_rate": 6.202111480049628e-06, + "loss": 0.6339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1538 + }, + { + "epoch": 1.8976572133168927, + "grad_norm": 2.797457073115434, + "learning_rate": 6.1899192188187385e-06, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1539 + }, + { + "epoch": 1.8988902589395806, + "grad_norm": 2.5101307384888343, + "learning_rate": 6.177733579536091e-06, + "loss": 0.4852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1540 + }, + { + "epoch": 1.900123304562269, + "grad_norm": 2.711866598306881, + "learning_rate": 6.165554583380401e-06, + "loss": 0.5412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1541 + }, + { + "epoch": 1.9013563501849569, + "grad_norm": 2.41897150751434, + "learning_rate": 6.1533822515188665e-06, + "loss": 0.4917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1542 + }, + { + "epoch": 1.9025893958076447, + "grad_norm": 2.4935923340546875, + "learning_rate": 6.141216605107077e-06, + "loss": 0.4712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1543 + }, + { + "epoch": 1.903822441430333, + "grad_norm": 2.3408793732304356, + "learning_rate": 6.129057665289024e-06, + "loss": 0.4915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1544 + }, + { + "epoch": 1.905055487053021, + "grad_norm": 2.49420888585378, + "learning_rate": 6.116905453197032e-06, + "loss": 0.4243, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1545 + }, + { + "epoch": 1.9062885326757089, + "grad_norm": 2.3882774651201517, + "learning_rate": 6.1047599899517305e-06, + "loss": 0.3567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1546 + }, + { + "epoch": 1.9075215782983972, + "grad_norm": 2.412746547072853, + "learning_rate": 6.092621296662028e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1547 + }, + { + "epoch": 1.908754623921085, + "grad_norm": 2.793611618535343, + "learning_rate": 6.080489394425058e-06, + "loss": 0.4794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1548 + }, + { + "epoch": 1.909987669543773, + "grad_norm": 2.86759252402197, + "learning_rate": 6.06836430432616e-06, + "loss": 0.5258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1549 + }, + { + "epoch": 1.9112207151664613, + "grad_norm": 3.5964778408959197, + "learning_rate": 6.056246047438823e-06, + "loss": 0.6431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1550 + }, + { + "epoch": 1.9124537607891492, + "grad_norm": 3.004291710626235, + "learning_rate": 6.044134644824671e-06, + "loss": 0.5851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1551 + }, + { + "epoch": 1.913686806411837, + "grad_norm": 2.553994453235498, + "learning_rate": 6.0320301175334096e-06, + "loss": 0.4947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1552 + }, + { + "epoch": 1.9149198520345254, + "grad_norm": 2.3305952965305723, + "learning_rate": 6.019932486602793e-06, + "loss": 0.4291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1553 + }, + { + "epoch": 1.9161528976572133, + "grad_norm": 2.70796232056951, + "learning_rate": 6.007841773058596e-06, + "loss": 0.5056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1554 + }, + { + "epoch": 1.9173859432799012, + "grad_norm": 2.401542735999577, + "learning_rate": 5.9957579979145595e-06, + "loss": 0.4256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1555 + }, + { + "epoch": 1.9186189889025895, + "grad_norm": 2.6832304024737783, + "learning_rate": 5.983681182172387e-06, + "loss": 0.6448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1556 + }, + { + "epoch": 1.9198520345252774, + "grad_norm": 2.5250444762365314, + "learning_rate": 5.971611346821657e-06, + "loss": 0.5107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1557 + }, + { + "epoch": 1.9210850801479655, + "grad_norm": 2.487361580061221, + "learning_rate": 5.959548512839839e-06, + "loss": 0.4747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1558 + }, + { + "epoch": 1.9223181257706536, + "grad_norm": 2.5305388152078963, + "learning_rate": 5.947492701192229e-06, + "loss": 0.4648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1559 + }, + { + "epoch": 1.9235511713933415, + "grad_norm": 2.455837488266591, + "learning_rate": 5.9354439328319105e-06, + "loss": 0.4423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1560 + }, + { + "epoch": 1.9247842170160296, + "grad_norm": 2.468256257896946, + "learning_rate": 5.923402228699736e-06, + "loss": 0.4569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1561 + }, + { + "epoch": 1.9260172626387178, + "grad_norm": 2.42939124280653, + "learning_rate": 5.91136760972427e-06, + "loss": 0.3932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1562 + }, + { + "epoch": 1.9272503082614056, + "grad_norm": 2.858568888655988, + "learning_rate": 5.89934009682177e-06, + "loss": 0.484, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1563 + }, + { + "epoch": 1.9284833538840938, + "grad_norm": 2.61347238372933, + "learning_rate": 5.887319710896144e-06, + "loss": 0.467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1564 + }, + { + "epoch": 1.9297163995067819, + "grad_norm": 2.7599324090566926, + "learning_rate": 5.875306472838908e-06, + "loss": 0.6094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1565 + }, + { + "epoch": 1.9309494451294698, + "grad_norm": 2.6103338708006505, + "learning_rate": 5.863300403529159e-06, + "loss": 0.4638, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1566 + }, + { + "epoch": 1.9321824907521579, + "grad_norm": 2.47982724255905, + "learning_rate": 5.85130152383353e-06, + "loss": 0.4826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1567 + }, + { + "epoch": 1.933415536374846, + "grad_norm": 2.578853751626434, + "learning_rate": 5.839309854606167e-06, + "loss": 0.5235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1568 + }, + { + "epoch": 1.9346485819975339, + "grad_norm": 2.4399405587351146, + "learning_rate": 5.827325416688671e-06, + "loss": 0.4686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1569 + }, + { + "epoch": 1.935881627620222, + "grad_norm": 2.5935511573767585, + "learning_rate": 5.81534823091009e-06, + "loss": 0.4941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1570 + }, + { + "epoch": 1.93711467324291, + "grad_norm": 2.3108330384724596, + "learning_rate": 5.803378318086859e-06, + "loss": 0.4216, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1571 + }, + { + "epoch": 1.938347718865598, + "grad_norm": 2.542118633939692, + "learning_rate": 5.791415699022768e-06, + "loss": 0.5643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1572 + }, + { + "epoch": 1.939580764488286, + "grad_norm": 2.6343709127133526, + "learning_rate": 5.779460394508944e-06, + "loss": 0.5811, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1573 + }, + { + "epoch": 1.9408138101109742, + "grad_norm": 2.476140626898723, + "learning_rate": 5.767512425323791e-06, + "loss": 0.4819, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1574 + }, + { + "epoch": 1.942046855733662, + "grad_norm": 2.6259462878787634, + "learning_rate": 5.755571812232967e-06, + "loss": 0.6072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1575 + }, + { + "epoch": 1.9432799013563502, + "grad_norm": 2.4454523827907804, + "learning_rate": 5.74363857598934e-06, + "loss": 0.4893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1576 + }, + { + "epoch": 1.9445129469790383, + "grad_norm": 2.5545835071067575, + "learning_rate": 5.731712737332965e-06, + "loss": 0.5415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1577 + }, + { + "epoch": 1.9457459926017262, + "grad_norm": 2.790586346292402, + "learning_rate": 5.7197943169910455e-06, + "loss": 0.619, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1578 + }, + { + "epoch": 1.9469790382244143, + "grad_norm": 2.5655789543184913, + "learning_rate": 5.707883335677867e-06, + "loss": 0.3993, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1579 + }, + { + "epoch": 1.9482120838471024, + "grad_norm": 2.3868281881615063, + "learning_rate": 5.695979814094814e-06, + "loss": 0.4513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1580 + }, + { + "epoch": 1.9494451294697903, + "grad_norm": 2.5119777174691684, + "learning_rate": 5.6840837729302865e-06, + "loss": 0.403, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1581 + }, + { + "epoch": 1.9506781750924784, + "grad_norm": 2.454297741118528, + "learning_rate": 5.672195232859703e-06, + "loss": 0.4341, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1582 + }, + { + "epoch": 1.9519112207151665, + "grad_norm": 2.173279017356807, + "learning_rate": 5.660314214545417e-06, + "loss": 0.4009, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1583 + }, + { + "epoch": 1.9531442663378544, + "grad_norm": 2.41693418545067, + "learning_rate": 5.648440738636731e-06, + "loss": 0.5145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1584 + }, + { + "epoch": 1.9543773119605425, + "grad_norm": 2.451502353773675, + "learning_rate": 5.63657482576984e-06, + "loss": 0.4148, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1585 + }, + { + "epoch": 1.9556103575832307, + "grad_norm": 2.5952977561456083, + "learning_rate": 5.624716496567782e-06, + "loss": 0.4059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1586 + }, + { + "epoch": 1.9568434032059185, + "grad_norm": 2.6119219402093647, + "learning_rate": 5.612865771640419e-06, + "loss": 0.4455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1587 + }, + { + "epoch": 1.9580764488286067, + "grad_norm": 2.5564882849476294, + "learning_rate": 5.601022671584394e-06, + "loss": 0.4633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1588 + }, + { + "epoch": 1.9593094944512948, + "grad_norm": 2.8188713009613244, + "learning_rate": 5.58918721698311e-06, + "loss": 0.565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1589 + }, + { + "epoch": 1.9605425400739827, + "grad_norm": 2.4162766103238895, + "learning_rate": 5.577359428406669e-06, + "loss": 0.5188, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1590 + }, + { + "epoch": 1.9617755856966708, + "grad_norm": 2.50589027882338, + "learning_rate": 5.56553932641185e-06, + "loss": 0.4671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1591 + }, + { + "epoch": 1.9630086313193589, + "grad_norm": 2.338580770590297, + "learning_rate": 5.553726931542085e-06, + "loss": 0.4271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1592 + }, + { + "epoch": 1.9642416769420468, + "grad_norm": 2.727731476438198, + "learning_rate": 5.541922264327395e-06, + "loss": 0.5796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1593 + }, + { + "epoch": 1.9654747225647349, + "grad_norm": 2.372879579889336, + "learning_rate": 5.530125345284388e-06, + "loss": 0.4791, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1594 + }, + { + "epoch": 1.966707768187423, + "grad_norm": 2.402358128750275, + "learning_rate": 5.5183361949161804e-06, + "loss": 0.5589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1595 + }, + { + "epoch": 1.9679408138101109, + "grad_norm": 2.521073091835758, + "learning_rate": 5.506554833712412e-06, + "loss": 0.5612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1596 + }, + { + "epoch": 1.969173859432799, + "grad_norm": 2.475424445052322, + "learning_rate": 5.494781282149175e-06, + "loss": 0.4164, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1597 + }, + { + "epoch": 1.970406905055487, + "grad_norm": 2.4762092606289388, + "learning_rate": 5.4830155606889855e-06, + "loss": 0.6031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1598 + }, + { + "epoch": 1.971639950678175, + "grad_norm": 2.3075398162125857, + "learning_rate": 5.471257689780753e-06, + "loss": 0.3457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1599 + }, + { + "epoch": 1.972872996300863, + "grad_norm": 2.8308395918813405, + "learning_rate": 5.459507689859742e-06, + "loss": 0.5926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1600 + }, + { + "epoch": 1.9741060419235512, + "grad_norm": 2.695737068994709, + "learning_rate": 5.447765581347542e-06, + "loss": 0.5015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1601 + }, + { + "epoch": 1.975339087546239, + "grad_norm": 2.88226095679846, + "learning_rate": 5.436031384652021e-06, + "loss": 0.5126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1602 + }, + { + "epoch": 1.9765721331689272, + "grad_norm": 2.466798154495963, + "learning_rate": 5.424305120167295e-06, + "loss": 0.3504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1603 + }, + { + "epoch": 1.9778051787916153, + "grad_norm": 2.999551007485252, + "learning_rate": 5.412586808273705e-06, + "loss": 0.6638, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1604 + }, + { + "epoch": 1.9790382244143032, + "grad_norm": 2.687785997003729, + "learning_rate": 5.400876469337759e-06, + "loss": 0.4778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1605 + }, + { + "epoch": 1.9802712700369913, + "grad_norm": 2.697985425334561, + "learning_rate": 5.389174123712113e-06, + "loss": 0.4376, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1606 + }, + { + "epoch": 1.9815043156596794, + "grad_norm": 2.2270783908633094, + "learning_rate": 5.377479791735522e-06, + "loss": 0.3055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1607 + }, + { + "epoch": 1.9827373612823673, + "grad_norm": 2.535575049691839, + "learning_rate": 5.3657934937328296e-06, + "loss": 0.4307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1608 + }, + { + "epoch": 1.9839704069050554, + "grad_norm": 2.6549657486777063, + "learning_rate": 5.354115250014904e-06, + "loss": 0.5848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1609 + }, + { + "epoch": 1.9852034525277436, + "grad_norm": 2.83855139958114, + "learning_rate": 5.342445080878614e-06, + "loss": 0.5668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1610 + }, + { + "epoch": 1.9864364981504314, + "grad_norm": 2.5321393689709284, + "learning_rate": 5.330783006606805e-06, + "loss": 0.3667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1611 + }, + { + "epoch": 1.9876695437731196, + "grad_norm": 3.1246050900975657, + "learning_rate": 5.319129047468244e-06, + "loss": 0.5487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1612 + }, + { + "epoch": 1.9889025893958077, + "grad_norm": 2.661845983771021, + "learning_rate": 5.307483223717598e-06, + "loss": 0.4961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1613 + }, + { + "epoch": 1.9901356350184956, + "grad_norm": 2.567032419410714, + "learning_rate": 5.295845555595396e-06, + "loss": 0.4989, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1614 + }, + { + "epoch": 1.9913686806411839, + "grad_norm": 2.715329364090932, + "learning_rate": 5.284216063327987e-06, + "loss": 0.55, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1615 + }, + { + "epoch": 1.9926017262638718, + "grad_norm": 2.4247843342398565, + "learning_rate": 5.27259476712751e-06, + "loss": 0.4278, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1616 + }, + { + "epoch": 1.9938347718865597, + "grad_norm": 2.586736358421805, + "learning_rate": 5.260981687191873e-06, + "loss": 0.3932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1617 + }, + { + "epoch": 1.995067817509248, + "grad_norm": 2.689699358030808, + "learning_rate": 5.249376843704686e-06, + "loss": 0.6152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1618 + }, + { + "epoch": 1.9963008631319359, + "grad_norm": 2.393722648406475, + "learning_rate": 5.2377802568352495e-06, + "loss": 0.4516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1619 + }, + { + "epoch": 1.9975339087546238, + "grad_norm": 2.151941518652342, + "learning_rate": 5.226191946738526e-06, + "loss": 0.4433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1620 + }, + { + "epoch": 1.998766954377312, + "grad_norm": 2.7987504690673988, + "learning_rate": 5.214611933555076e-06, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1621 + }, + { + "epoch": 2.0, + "grad_norm": 2.59776624758183, + "learning_rate": 5.20304023741105e-06, + "loss": 0.619, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1622 + }, + { + "epoch": 2.001233045622688, + "grad_norm": 2.43048324510223, + "learning_rate": 5.191476878418133e-06, + "loss": 0.4669, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1623 + }, + { + "epoch": 2.0024660912453762, + "grad_norm": 2.3352351017918402, + "learning_rate": 5.179921876673535e-06, + "loss": 0.4957, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1624 + } + ], + "logging_steps": 1, + "max_steps": 2433, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 406, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2667130398965760.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1624/training_args.bin b/checkpoint-1624/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7b675d785639ed4b99af767b6c87a1227c13ec1b --- /dev/null +++ b/checkpoint-1624/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e80ea310ffc6c44543f32aa82ecc00b1bf9d3e329881647443a895f041e17955 +size 9208 diff --git a/checkpoint-2030/chat_template.jinja b/checkpoint-2030/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..870322b8f52314199bfd74bf68b8b8f1126395ae --- /dev/null +++ b/checkpoint-2030/chat_template.jinja @@ -0,0 +1,5 @@ +{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} diff --git a/checkpoint-2030/config.json b/checkpoint-2030/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4c0f53b6c66a0b8a2a4eb621c949412e2a7f353a --- /dev/null +++ b/checkpoint-2030/config.json @@ -0,0 +1,35 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.2", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/checkpoint-2030/generation_config.json b/checkpoint-2030/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b513e54e3195b917260c9a8a04c9f3683f19de35 --- /dev/null +++ b/checkpoint-2030/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.55.2" +} diff --git a/checkpoint-2030/model-00004-of-00004.safetensors b/checkpoint-2030/model-00004-of-00004.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0222ea4db2e17199b003ee7161d887d2d1793bc1 --- /dev/null +++ b/checkpoint-2030/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffeeea9feb8010f427aba5333ea181a007a4ad3247593be9ff1550922ae7db8e +size 1168138808 diff --git a/checkpoint-2030/model.safetensors.index.json b/checkpoint-2030/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..58a7ef40d52ab3340cf131cf33f8923f5144aaf9 --- /dev/null +++ b/checkpoint-2030/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-2030/special_tokens_map.json b/checkpoint-2030/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3c1d04911c269b925af977a3151c9704e990e4d0 --- /dev/null +++ b/checkpoint-2030/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-2030/tokenizer.json b/checkpoint-2030/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-2030/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-2030/tokenizer_config.json b/checkpoint-2030/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..c68051fe3c4d23234a59316bc52d21f6e3a4182c --- /dev/null +++ b/checkpoint-2030/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|finetune_right_pad_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-2030/trainer_state.json b/checkpoint-2030/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..3e2f095e04087e592a3657b0a3726e082b78f64c --- /dev/null +++ b/checkpoint-2030/trainer_state.json @@ -0,0 +1,20334 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.50308261405672, + "eval_steps": 500, + "global_step": 2030, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0012330456226880395, + "grad_norm": 13.955555906863093, + "learning_rate": 0.0, + "loss": 1.9039, + "memory/device_mem_reserved(gib)": 19.83, + "memory/max_mem_active(gib)": 16.16, + "memory/max_mem_allocated(gib)": 16.16, + "step": 1 + }, + { + "epoch": 0.002466091245376079, + "grad_norm": 14.68412675273428, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.9395, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 2 + }, + { + "epoch": 0.0036991368680641184, + "grad_norm": 15.404314959825989, + "learning_rate": 8.000000000000001e-07, + "loss": 1.9524, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 3 + }, + { + "epoch": 0.004932182490752158, + "grad_norm": 14.099502171385124, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.8527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 4 + }, + { + "epoch": 0.006165228113440197, + "grad_norm": 12.213499070556583, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.8853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 5 + }, + { + "epoch": 0.007398273736128237, + "grad_norm": 7.4228290511653725, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.7686, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 6 + }, + { + "epoch": 0.008631319358816275, + "grad_norm": 6.989334002600767, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.7976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 7 + }, + { + "epoch": 0.009864364981504316, + "grad_norm": 7.330479156896915, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6029, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 8 + }, + { + "epoch": 0.011097410604192354, + "grad_norm": 7.091571204851302, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.6097, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 9 + }, + { + "epoch": 0.012330456226880395, + "grad_norm": 5.900225931668668, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.6512, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 10 + }, + { + "epoch": 0.013563501849568433, + "grad_norm": 12.562784448457464, + "learning_rate": 4.000000000000001e-06, + "loss": 1.4824, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 11 + }, + { + "epoch": 0.014796547472256474, + "grad_norm": 6.056074456731366, + "learning_rate": 4.4e-06, + "loss": 1.5581, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 12 + }, + { + "epoch": 0.016029593094944512, + "grad_norm": 5.051119502724206, + "learning_rate": 4.800000000000001e-06, + "loss": 1.5275, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 13 + }, + { + "epoch": 0.01726263871763255, + "grad_norm": 4.022180485081632, + "learning_rate": 5.2e-06, + "loss": 1.5075, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 14 + }, + { + "epoch": 0.018495684340320593, + "grad_norm": 4.5104551135107815, + "learning_rate": 5.600000000000001e-06, + "loss": 1.4638, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 15 + }, + { + "epoch": 0.01972872996300863, + "grad_norm": 4.3429897237996276, + "learning_rate": 6e-06, + "loss": 1.3996, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 16 + }, + { + "epoch": 0.02096177558569667, + "grad_norm": 3.9791138623880764, + "learning_rate": 6.4000000000000006e-06, + "loss": 1.4059, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 17 + }, + { + "epoch": 0.02219482120838471, + "grad_norm": 4.194505204288471, + "learning_rate": 6.800000000000001e-06, + "loss": 1.3747, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 18 + }, + { + "epoch": 0.02342786683107275, + "grad_norm": 3.6309957760327713, + "learning_rate": 7.2000000000000005e-06, + "loss": 1.3565, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 19 + }, + { + "epoch": 0.02466091245376079, + "grad_norm": 3.431355025722372, + "learning_rate": 7.600000000000001e-06, + "loss": 1.4019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 20 + }, + { + "epoch": 0.025893958076448828, + "grad_norm": 3.258474422309909, + "learning_rate": 8.000000000000001e-06, + "loss": 1.364, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 21 + }, + { + "epoch": 0.027127003699136867, + "grad_norm": 3.7750095693616736, + "learning_rate": 8.400000000000001e-06, + "loss": 1.3798, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 22 + }, + { + "epoch": 0.02836004932182491, + "grad_norm": 3.1192450402705125, + "learning_rate": 8.8e-06, + "loss": 1.2677, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 23 + }, + { + "epoch": 0.029593094944512947, + "grad_norm": 3.34669710403596, + "learning_rate": 9.200000000000002e-06, + "loss": 1.3465, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 24 + }, + { + "epoch": 0.030826140567200986, + "grad_norm": 3.0969841766129638, + "learning_rate": 9.600000000000001e-06, + "loss": 1.3662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 25 + }, + { + "epoch": 0.032059186189889025, + "grad_norm": 2.9521843506112466, + "learning_rate": 1e-05, + "loss": 1.3295, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 26 + }, + { + "epoch": 0.03329223181257707, + "grad_norm": 2.9117498945141085, + "learning_rate": 1.04e-05, + "loss": 1.3066, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 27 + }, + { + "epoch": 0.0345252774352651, + "grad_norm": 2.9215831263409013, + "learning_rate": 1.0800000000000002e-05, + "loss": 1.3221, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 28 + }, + { + "epoch": 0.035758323057953144, + "grad_norm": 2.97321184472987, + "learning_rate": 1.1200000000000001e-05, + "loss": 1.3247, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 29 + }, + { + "epoch": 0.036991368680641186, + "grad_norm": 3.293476484963883, + "learning_rate": 1.16e-05, + "loss": 1.2582, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 30 + }, + { + "epoch": 0.03822441430332922, + "grad_norm": 2.8642108219246745, + "learning_rate": 1.2e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 31 + }, + { + "epoch": 0.03945745992601726, + "grad_norm": 2.8916726488438824, + "learning_rate": 1.2400000000000002e-05, + "loss": 1.2628, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 32 + }, + { + "epoch": 0.040690505548705305, + "grad_norm": 3.17999982306396, + "learning_rate": 1.2800000000000001e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 33 + }, + { + "epoch": 0.04192355117139334, + "grad_norm": 2.6965579635696453, + "learning_rate": 1.3200000000000002e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 34 + }, + { + "epoch": 0.04315659679408138, + "grad_norm": 2.9489592197594896, + "learning_rate": 1.3600000000000002e-05, + "loss": 1.3113, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 35 + }, + { + "epoch": 0.04438964241676942, + "grad_norm": 2.8898479257030734, + "learning_rate": 1.4e-05, + "loss": 1.3387, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 36 + }, + { + "epoch": 0.04562268803945746, + "grad_norm": 2.6370539695770425, + "learning_rate": 1.4400000000000001e-05, + "loss": 1.3314, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 37 + }, + { + "epoch": 0.0468557336621455, + "grad_norm": 2.7915119494225813, + "learning_rate": 1.48e-05, + "loss": 1.2603, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 38 + }, + { + "epoch": 0.04808877928483354, + "grad_norm": 2.602735589719503, + "learning_rate": 1.5200000000000002e-05, + "loss": 1.3492, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 39 + }, + { + "epoch": 0.04932182490752158, + "grad_norm": 2.9123309438390113, + "learning_rate": 1.5600000000000003e-05, + "loss": 1.2291, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 40 + }, + { + "epoch": 0.05055487053020962, + "grad_norm": 2.577807086473754, + "learning_rate": 1.6000000000000003e-05, + "loss": 1.3098, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 41 + }, + { + "epoch": 0.051787916152897656, + "grad_norm": 2.548383978805174, + "learning_rate": 1.64e-05, + "loss": 1.2893, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 42 + }, + { + "epoch": 0.0530209617755857, + "grad_norm": 2.6372216084932814, + "learning_rate": 1.6800000000000002e-05, + "loss": 1.2701, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 43 + }, + { + "epoch": 0.05425400739827373, + "grad_norm": 2.60286437866227, + "learning_rate": 1.72e-05, + "loss": 1.3579, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 44 + }, + { + "epoch": 0.055487053020961775, + "grad_norm": 2.536071820500707, + "learning_rate": 1.76e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 45 + }, + { + "epoch": 0.05672009864364982, + "grad_norm": 2.8143481627767963, + "learning_rate": 1.8e-05, + "loss": 1.1729, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 46 + }, + { + "epoch": 0.05795314426633785, + "grad_norm": 2.8600643284872596, + "learning_rate": 1.8400000000000003e-05, + "loss": 1.3246, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 47 + }, + { + "epoch": 0.059186189889025895, + "grad_norm": 2.618463019216554, + "learning_rate": 1.88e-05, + "loss": 1.2878, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 48 + }, + { + "epoch": 0.06041923551171394, + "grad_norm": 2.757496450593693, + "learning_rate": 1.9200000000000003e-05, + "loss": 1.3454, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 49 + }, + { + "epoch": 0.06165228113440197, + "grad_norm": 2.5356573576135126, + "learning_rate": 1.9600000000000002e-05, + "loss": 1.2719, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 50 + }, + { + "epoch": 0.06288532675709001, + "grad_norm": 2.532419474064049, + "learning_rate": 2e-05, + "loss": 1.332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 51 + }, + { + "epoch": 0.06411837237977805, + "grad_norm": 2.711902763412396, + "learning_rate": 1.999999130996347e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 52 + }, + { + "epoch": 0.06535141800246609, + "grad_norm": 2.654767047589697, + "learning_rate": 1.9999965239868983e-05, + "loss": 1.3092, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 53 + }, + { + "epoch": 0.06658446362515413, + "grad_norm": 3.2372229064932454, + "learning_rate": 1.9999921789761843e-05, + "loss": 1.3428, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 54 + }, + { + "epoch": 0.06781750924784218, + "grad_norm": 2.723103510772695, + "learning_rate": 1.999986095971757e-05, + "loss": 1.3918, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 55 + }, + { + "epoch": 0.0690505548705302, + "grad_norm": 2.849098835881172, + "learning_rate": 1.9999782749841895e-05, + "loss": 1.3304, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 56 + }, + { + "epoch": 0.07028360049321825, + "grad_norm": 2.4115082044348557, + "learning_rate": 1.9999687160270735e-05, + "loss": 1.2818, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 57 + }, + { + "epoch": 0.07151664611590629, + "grad_norm": 2.7720437269939495, + "learning_rate": 1.9999574191170233e-05, + "loss": 1.2799, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 58 + }, + { + "epoch": 0.07274969173859433, + "grad_norm": 2.4891158758553042, + "learning_rate": 1.9999443842736723e-05, + "loss": 1.2408, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 59 + }, + { + "epoch": 0.07398273736128237, + "grad_norm": 2.8066935220153737, + "learning_rate": 1.999929611519676e-05, + "loss": 1.3157, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 60 + }, + { + "epoch": 0.0752157829839704, + "grad_norm": 2.6715709033170705, + "learning_rate": 1.9999131008807094e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 61 + }, + { + "epoch": 0.07644882860665844, + "grad_norm": 2.586596297320694, + "learning_rate": 1.9998948523854676e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 62 + }, + { + "epoch": 0.07768187422934648, + "grad_norm": 2.646887045495792, + "learning_rate": 1.999874866065667e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 63 + }, + { + "epoch": 0.07891491985203453, + "grad_norm": 2.419730330481743, + "learning_rate": 1.999853141956044e-05, + "loss": 1.2829, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 64 + }, + { + "epoch": 0.08014796547472257, + "grad_norm": 2.5201815842521245, + "learning_rate": 1.999829680094355e-05, + "loss": 1.3243, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 65 + }, + { + "epoch": 0.08138101109741061, + "grad_norm": 2.503429794099987, + "learning_rate": 1.9998044805213773e-05, + "loss": 1.2681, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 66 + }, + { + "epoch": 0.08261405672009864, + "grad_norm": 2.348439339795846, + "learning_rate": 1.9997775432809074e-05, + "loss": 1.3359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 67 + }, + { + "epoch": 0.08384710234278668, + "grad_norm": 2.3828095918277357, + "learning_rate": 1.999748868419763e-05, + "loss": 1.3526, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 68 + }, + { + "epoch": 0.08508014796547472, + "grad_norm": 2.5076969489265273, + "learning_rate": 1.9997184559877806e-05, + "loss": 1.3531, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 69 + }, + { + "epoch": 0.08631319358816276, + "grad_norm": 2.388938276200468, + "learning_rate": 1.9996863060378176e-05, + "loss": 1.334, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 70 + }, + { + "epoch": 0.08754623921085081, + "grad_norm": 2.283154945083281, + "learning_rate": 1.999652418625751e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 71 + }, + { + "epoch": 0.08877928483353884, + "grad_norm": 2.2613574311048734, + "learning_rate": 1.9996167938104767e-05, + "loss": 1.2858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 72 + }, + { + "epoch": 0.09001233045622688, + "grad_norm": 2.3931162231369076, + "learning_rate": 1.999579431653912e-05, + "loss": 1.3089, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 73 + }, + { + "epoch": 0.09124537607891492, + "grad_norm": 2.5073647819033194, + "learning_rate": 1.9995403322209913e-05, + "loss": 1.382, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 74 + }, + { + "epoch": 0.09247842170160296, + "grad_norm": 2.5388011932611536, + "learning_rate": 1.999499495579671e-05, + "loss": 1.402, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 75 + }, + { + "epoch": 0.093711467324291, + "grad_norm": 2.3884809904992252, + "learning_rate": 1.9994569218009247e-05, + "loss": 1.3079, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 76 + }, + { + "epoch": 0.09494451294697903, + "grad_norm": 2.3454483578272094, + "learning_rate": 1.999412610958746e-05, + "loss": 1.2653, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 77 + }, + { + "epoch": 0.09617755856966707, + "grad_norm": 2.186902835408518, + "learning_rate": 1.9993665631301474e-05, + "loss": 1.2069, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 78 + }, + { + "epoch": 0.09741060419235512, + "grad_norm": 2.3845058131797368, + "learning_rate": 1.9993187783951603e-05, + "loss": 1.3541, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 79 + }, + { + "epoch": 0.09864364981504316, + "grad_norm": 2.256399889979939, + "learning_rate": 1.9992692568368354e-05, + "loss": 1.2119, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 80 + }, + { + "epoch": 0.0998766954377312, + "grad_norm": 2.3361092398369903, + "learning_rate": 1.9992179985412416e-05, + "loss": 1.2585, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 81 + }, + { + "epoch": 0.10110974106041924, + "grad_norm": 2.289061082197673, + "learning_rate": 1.999165003597465e-05, + "loss": 1.3459, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 82 + }, + { + "epoch": 0.10234278668310727, + "grad_norm": 2.228187448538031, + "learning_rate": 1.999110272097613e-05, + "loss": 1.3739, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 83 + }, + { + "epoch": 0.10357583230579531, + "grad_norm": 2.157361593613978, + "learning_rate": 1.9990538041368078e-05, + "loss": 1.2575, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 84 + }, + { + "epoch": 0.10480887792848335, + "grad_norm": 2.294809937254642, + "learning_rate": 1.998995599813192e-05, + "loss": 1.2834, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 85 + }, + { + "epoch": 0.1060419235511714, + "grad_norm": 2.1831743941808925, + "learning_rate": 1.998935659227925e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 86 + }, + { + "epoch": 0.10727496917385944, + "grad_norm": 2.247228611971362, + "learning_rate": 1.998873982485183e-05, + "loss": 1.2654, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 87 + }, + { + "epoch": 0.10850801479654747, + "grad_norm": 2.2844560193856416, + "learning_rate": 1.9988105696921622e-05, + "loss": 1.2927, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 88 + }, + { + "epoch": 0.10974106041923551, + "grad_norm": 2.3780157649218157, + "learning_rate": 1.9987454209590736e-05, + "loss": 1.2621, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 89 + }, + { + "epoch": 0.11097410604192355, + "grad_norm": 2.3017268441414003, + "learning_rate": 1.9986785363991466e-05, + "loss": 1.2676, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 90 + }, + { + "epoch": 0.11220715166461159, + "grad_norm": 2.1906552279173503, + "learning_rate": 1.9986099161286266e-05, + "loss": 1.2281, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 91 + }, + { + "epoch": 0.11344019728729964, + "grad_norm": 2.4009419002081276, + "learning_rate": 1.9985395602667764e-05, + "loss": 1.2267, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 92 + }, + { + "epoch": 0.11467324290998766, + "grad_norm": 2.190217063330867, + "learning_rate": 1.9984674689358744e-05, + "loss": 1.2778, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 93 + }, + { + "epoch": 0.1159062885326757, + "grad_norm": 2.305624793209019, + "learning_rate": 1.998393642261217e-05, + "loss": 1.2821, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 94 + }, + { + "epoch": 0.11713933415536375, + "grad_norm": 2.162537568697834, + "learning_rate": 1.9983180803711146e-05, + "loss": 1.3775, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 95 + }, + { + "epoch": 0.11837237977805179, + "grad_norm": 2.2632096292316706, + "learning_rate": 1.9982407833968947e-05, + "loss": 1.3358, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 96 + }, + { + "epoch": 0.11960542540073983, + "grad_norm": 2.268508818243908, + "learning_rate": 1.9981617514728998e-05, + "loss": 1.2368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 97 + }, + { + "epoch": 0.12083847102342787, + "grad_norm": 2.2721363895167377, + "learning_rate": 1.9980809847364884e-05, + "loss": 1.2774, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 98 + }, + { + "epoch": 0.1220715166461159, + "grad_norm": 2.1661250921439974, + "learning_rate": 1.9979984833280332e-05, + "loss": 1.359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 99 + }, + { + "epoch": 0.12330456226880394, + "grad_norm": 2.2521676172131873, + "learning_rate": 1.997914247390922e-05, + "loss": 1.216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 100 + }, + { + "epoch": 0.12453760789149199, + "grad_norm": 2.3907655242108095, + "learning_rate": 1.9978282770715586e-05, + "loss": 1.2468, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 101 + }, + { + "epoch": 0.12577065351418001, + "grad_norm": 2.2007275768445074, + "learning_rate": 1.9977405725193587e-05, + "loss": 1.2415, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 102 + }, + { + "epoch": 0.12700369913686807, + "grad_norm": 2.3304230330598346, + "learning_rate": 1.9976511338867542e-05, + "loss": 1.2438, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 103 + }, + { + "epoch": 0.1282367447595561, + "grad_norm": 2.346171964705753, + "learning_rate": 1.99755996132919e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 104 + }, + { + "epoch": 0.12946979038224415, + "grad_norm": 2.286623534979458, + "learning_rate": 1.9974670550051248e-05, + "loss": 1.2897, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 105 + }, + { + "epoch": 0.13070283600493218, + "grad_norm": 2.245774575899364, + "learning_rate": 1.99737241507603e-05, + "loss": 1.2923, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 106 + }, + { + "epoch": 0.1319358816276202, + "grad_norm": 2.2352297973588953, + "learning_rate": 1.9972760417063912e-05, + "loss": 1.1434, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 107 + }, + { + "epoch": 0.13316892725030827, + "grad_norm": 2.169802916370051, + "learning_rate": 1.9971779350637054e-05, + "loss": 1.2353, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 108 + }, + { + "epoch": 0.1344019728729963, + "grad_norm": 2.333533512175979, + "learning_rate": 1.9970780953184828e-05, + "loss": 1.2641, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 109 + }, + { + "epoch": 0.13563501849568435, + "grad_norm": 2.2985803581198674, + "learning_rate": 1.996976522644246e-05, + "loss": 1.3036, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 110 + }, + { + "epoch": 0.13686806411837238, + "grad_norm": 2.0992498120010272, + "learning_rate": 1.996873217217528e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 111 + }, + { + "epoch": 0.1381011097410604, + "grad_norm": 2.396895991067544, + "learning_rate": 1.9967681792178758e-05, + "loss": 1.2533, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 112 + }, + { + "epoch": 0.13933415536374846, + "grad_norm": 2.0738932505082768, + "learning_rate": 1.9966614088278456e-05, + "loss": 1.3332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 113 + }, + { + "epoch": 0.1405672009864365, + "grad_norm": 2.1049156004628395, + "learning_rate": 1.9965529062330048e-05, + "loss": 1.2793, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 114 + }, + { + "epoch": 0.14180024660912455, + "grad_norm": 2.2137820282577847, + "learning_rate": 1.9964426716219316e-05, + "loss": 1.1932, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 115 + }, + { + "epoch": 0.14303329223181258, + "grad_norm": 2.1545610182661976, + "learning_rate": 1.9963307051862152e-05, + "loss": 1.2662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 116 + }, + { + "epoch": 0.1442663378545006, + "grad_norm": 2.151286882718536, + "learning_rate": 1.9962170071204542e-05, + "loss": 1.205, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 117 + }, + { + "epoch": 0.14549938347718866, + "grad_norm": 2.026608825482131, + "learning_rate": 1.996101577622256e-05, + "loss": 1.195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 118 + }, + { + "epoch": 0.1467324290998767, + "grad_norm": 2.156975981742732, + "learning_rate": 1.9959844168922377e-05, + "loss": 1.3284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 119 + }, + { + "epoch": 0.14796547472256474, + "grad_norm": 2.2311865782919025, + "learning_rate": 1.9958655251340265e-05, + "loss": 1.2183, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 120 + }, + { + "epoch": 0.14919852034525277, + "grad_norm": 2.2940740150536376, + "learning_rate": 1.9957449025542567e-05, + "loss": 1.2625, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 121 + }, + { + "epoch": 0.1504315659679408, + "grad_norm": 2.214218654766027, + "learning_rate": 1.9956225493625708e-05, + "loss": 1.2177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 122 + }, + { + "epoch": 0.15166461159062886, + "grad_norm": 2.1690434056776775, + "learning_rate": 1.9954984657716202e-05, + "loss": 1.2049, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 123 + }, + { + "epoch": 0.15289765721331688, + "grad_norm": 2.1450511034984046, + "learning_rate": 1.9953726519970625e-05, + "loss": 1.232, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 124 + }, + { + "epoch": 0.15413070283600494, + "grad_norm": 2.0475086343318543, + "learning_rate": 1.995245108257563e-05, + "loss": 1.2665, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 125 + }, + { + "epoch": 0.15536374845869297, + "grad_norm": 2.1232594967339318, + "learning_rate": 1.9951158347747945e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 126 + }, + { + "epoch": 0.15659679408138102, + "grad_norm": 2.1915516478247308, + "learning_rate": 1.9949848317734345e-05, + "loss": 1.177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 127 + }, + { + "epoch": 0.15782983970406905, + "grad_norm": 2.2848682922250205, + "learning_rate": 1.9948520994811674e-05, + "loss": 1.2406, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 128 + }, + { + "epoch": 0.15906288532675708, + "grad_norm": 2.2223107161107505, + "learning_rate": 1.9947176381286825e-05, + "loss": 1.327, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 129 + }, + { + "epoch": 0.16029593094944514, + "grad_norm": 2.1247307489434837, + "learning_rate": 1.994581447949675e-05, + "loss": 1.1634, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 130 + }, + { + "epoch": 0.16152897657213316, + "grad_norm": 1.926245615401859, + "learning_rate": 1.9944435291808445e-05, + "loss": 1.1655, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 131 + }, + { + "epoch": 0.16276202219482122, + "grad_norm": 2.2531819769934858, + "learning_rate": 1.9943038820618946e-05, + "loss": 1.2216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 132 + }, + { + "epoch": 0.16399506781750925, + "grad_norm": 2.2183949614858305, + "learning_rate": 1.994162506835533e-05, + "loss": 1.2202, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 133 + }, + { + "epoch": 0.16522811344019728, + "grad_norm": 2.164554433266706, + "learning_rate": 1.994019403747471e-05, + "loss": 1.2245, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 134 + }, + { + "epoch": 0.16646115906288533, + "grad_norm": 2.17653642461108, + "learning_rate": 1.9938745730464228e-05, + "loss": 1.1899, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 135 + }, + { + "epoch": 0.16769420468557336, + "grad_norm": 1.9936998341266101, + "learning_rate": 1.9937280149841055e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 136 + }, + { + "epoch": 0.16892725030826142, + "grad_norm": 2.27069520590907, + "learning_rate": 1.9935797298152374e-05, + "loss": 1.3023, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 137 + }, + { + "epoch": 0.17016029593094945, + "grad_norm": 2.244074718609361, + "learning_rate": 1.9934297177975397e-05, + "loss": 1.3292, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 138 + }, + { + "epoch": 0.17139334155363747, + "grad_norm": 2.2121305803685094, + "learning_rate": 1.9932779791917343e-05, + "loss": 1.3028, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 139 + }, + { + "epoch": 0.17262638717632553, + "grad_norm": 2.7262731821688697, + "learning_rate": 1.9931245142615442e-05, + "loss": 1.2861, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 140 + }, + { + "epoch": 0.17385943279901356, + "grad_norm": 2.2051993935523497, + "learning_rate": 1.992969323273692e-05, + "loss": 1.3134, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 141 + }, + { + "epoch": 0.17509247842170161, + "grad_norm": 2.088688387254833, + "learning_rate": 1.992812406497901e-05, + "loss": 1.262, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 142 + }, + { + "epoch": 0.17632552404438964, + "grad_norm": 2.19652156002121, + "learning_rate": 1.9926537642068942e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 143 + }, + { + "epoch": 0.17755856966707767, + "grad_norm": 1.887489027726817, + "learning_rate": 1.9924933966763924e-05, + "loss": 1.2847, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 144 + }, + { + "epoch": 0.17879161528976573, + "grad_norm": 2.088506357330406, + "learning_rate": 1.992331304185116e-05, + "loss": 1.2218, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 145 + }, + { + "epoch": 0.18002466091245375, + "grad_norm": 2.102792679186299, + "learning_rate": 1.9921674870147827e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 146 + }, + { + "epoch": 0.1812577065351418, + "grad_norm": 2.0847095844816015, + "learning_rate": 1.992001945450108e-05, + "loss": 1.2986, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 147 + }, + { + "epoch": 0.18249075215782984, + "grad_norm": 2.344867998715345, + "learning_rate": 1.9918346797788045e-05, + "loss": 1.1347, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 148 + }, + { + "epoch": 0.18372379778051787, + "grad_norm": 2.1013292606380602, + "learning_rate": 1.991665690291581e-05, + "loss": 1.1319, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 149 + }, + { + "epoch": 0.18495684340320592, + "grad_norm": 2.2475258887436538, + "learning_rate": 1.9914949772821418e-05, + "loss": 1.1648, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 150 + }, + { + "epoch": 0.18618988902589395, + "grad_norm": 2.223126056771945, + "learning_rate": 1.9913225410471888e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 151 + }, + { + "epoch": 0.187422934648582, + "grad_norm": 2.102849036750559, + "learning_rate": 1.9911483818864163e-05, + "loss": 1.2062, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 152 + }, + { + "epoch": 0.18865598027127004, + "grad_norm": 2.2107888348018805, + "learning_rate": 1.9909725001025146e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 153 + }, + { + "epoch": 0.18988902589395806, + "grad_norm": 2.1730720194510456, + "learning_rate": 1.9907948960011677e-05, + "loss": 1.2133, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 154 + }, + { + "epoch": 0.19112207151664612, + "grad_norm": 2.0874944224993115, + "learning_rate": 1.9906155698910523e-05, + "loss": 1.1966, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 155 + }, + { + "epoch": 0.19235511713933415, + "grad_norm": 2.047788318815714, + "learning_rate": 1.990434522083839e-05, + "loss": 1.26, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 156 + }, + { + "epoch": 0.1935881627620222, + "grad_norm": 2.0992674413368415, + "learning_rate": 1.99025175289419e-05, + "loss": 1.2544, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 157 + }, + { + "epoch": 0.19482120838471023, + "grad_norm": 2.279539056045867, + "learning_rate": 1.9900672626397593e-05, + "loss": 1.3123, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 158 + }, + { + "epoch": 0.1960542540073983, + "grad_norm": 1.9940111001673004, + "learning_rate": 1.9898810516411926e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 159 + }, + { + "epoch": 0.19728729963008632, + "grad_norm": 2.278268154556979, + "learning_rate": 1.989693120222126e-05, + "loss": 1.236, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 160 + }, + { + "epoch": 0.19852034525277434, + "grad_norm": 2.3184527515323796, + "learning_rate": 1.9895034687091854e-05, + "loss": 1.1691, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 161 + }, + { + "epoch": 0.1997533908754624, + "grad_norm": 2.094943513595831, + "learning_rate": 1.9893120974319867e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 162 + }, + { + "epoch": 0.20098643649815043, + "grad_norm": 2.296308593300257, + "learning_rate": 1.9891190067231346e-05, + "loss": 1.2976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 163 + }, + { + "epoch": 0.20221948212083848, + "grad_norm": 2.131546792333376, + "learning_rate": 1.988924196918222e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 164 + }, + { + "epoch": 0.2034525277435265, + "grad_norm": 1.981669864872828, + "learning_rate": 1.9887276683558302e-05, + "loss": 1.2934, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 165 + }, + { + "epoch": 0.20468557336621454, + "grad_norm": 2.0382654138005423, + "learning_rate": 1.9885294213775267e-05, + "loss": 1.3207, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 166 + }, + { + "epoch": 0.2059186189889026, + "grad_norm": 2.074242050982157, + "learning_rate": 1.9883294563278665e-05, + "loss": 1.1832, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 167 + }, + { + "epoch": 0.20715166461159062, + "grad_norm": 2.1994142392378317, + "learning_rate": 1.9881277735543902e-05, + "loss": 1.2284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 168 + }, + { + "epoch": 0.20838471023427868, + "grad_norm": 2.1893568284451987, + "learning_rate": 1.987924373407624e-05, + "loss": 1.2572, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 169 + }, + { + "epoch": 0.2096177558569667, + "grad_norm": 2.0490903431710352, + "learning_rate": 1.987719256241079e-05, + "loss": 1.2808, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 170 + }, + { + "epoch": 0.21085080147965474, + "grad_norm": 2.3424605103097176, + "learning_rate": 1.9875124224112498e-05, + "loss": 1.2924, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 171 + }, + { + "epoch": 0.2120838471023428, + "grad_norm": 2.100563128603325, + "learning_rate": 1.987303872277616e-05, + "loss": 1.1368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 172 + }, + { + "epoch": 0.21331689272503082, + "grad_norm": 2.303295425870437, + "learning_rate": 1.9870936062026385e-05, + "loss": 1.3425, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 173 + }, + { + "epoch": 0.21454993834771888, + "grad_norm": 1.9108233071934615, + "learning_rate": 1.9868816245517615e-05, + "loss": 1.1926, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 174 + }, + { + "epoch": 0.2157829839704069, + "grad_norm": 2.2539584509521173, + "learning_rate": 1.986667927693411e-05, + "loss": 1.222, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 175 + }, + { + "epoch": 0.21701602959309493, + "grad_norm": 2.006165392274905, + "learning_rate": 1.986452515998993e-05, + "loss": 1.2853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 176 + }, + { + "epoch": 0.218249075215783, + "grad_norm": 2.1337815004505454, + "learning_rate": 1.9862353898428948e-05, + "loss": 1.2671, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 177 + }, + { + "epoch": 0.21948212083847102, + "grad_norm": 2.062911126777883, + "learning_rate": 1.986016549602484e-05, + "loss": 1.2397, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 178 + }, + { + "epoch": 0.22071516646115907, + "grad_norm": 2.115740291239205, + "learning_rate": 1.9857959956581053e-05, + "loss": 1.2534, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 179 + }, + { + "epoch": 0.2219482120838471, + "grad_norm": 2.3733626044353837, + "learning_rate": 1.9855737283930843e-05, + "loss": 1.2082, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 180 + }, + { + "epoch": 0.22318125770653513, + "grad_norm": 2.0972006372187373, + "learning_rate": 1.985349748193722e-05, + "loss": 1.1091, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 181 + }, + { + "epoch": 0.22441430332922319, + "grad_norm": 2.0727561431265635, + "learning_rate": 1.9851240554492985e-05, + "loss": 1.2153, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 182 + }, + { + "epoch": 0.22564734895191121, + "grad_norm": 2.094605431898615, + "learning_rate": 1.9848966505520694e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 183 + }, + { + "epoch": 0.22688039457459927, + "grad_norm": 2.1524188959905866, + "learning_rate": 1.9846675338972654e-05, + "loss": 1.219, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 184 + }, + { + "epoch": 0.2281134401972873, + "grad_norm": 2.008037870627797, + "learning_rate": 1.9844367058830933e-05, + "loss": 1.2635, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 185 + }, + { + "epoch": 0.22934648581997533, + "grad_norm": 2.050448462235658, + "learning_rate": 1.984204166910734e-05, + "loss": 1.187, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 186 + }, + { + "epoch": 0.23057953144266338, + "grad_norm": 2.2727997887406066, + "learning_rate": 1.983969917384342e-05, + "loss": 1.2709, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 187 + }, + { + "epoch": 0.2318125770653514, + "grad_norm": 2.0295960774209414, + "learning_rate": 1.9837339577110443e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 188 + }, + { + "epoch": 0.23304562268803947, + "grad_norm": 2.250775161795276, + "learning_rate": 1.9834962883009406e-05, + "loss": 1.2566, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 189 + }, + { + "epoch": 0.2342786683107275, + "grad_norm": 2.221930554057359, + "learning_rate": 1.9832569095671025e-05, + "loss": 1.2547, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 190 + }, + { + "epoch": 0.23551171393341552, + "grad_norm": 2.071699318441252, + "learning_rate": 1.9830158219255715e-05, + "loss": 1.1712, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 191 + }, + { + "epoch": 0.23674475955610358, + "grad_norm": 1.9444133453118777, + "learning_rate": 1.98277302579536e-05, + "loss": 1.1858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 192 + }, + { + "epoch": 0.2379778051787916, + "grad_norm": 2.134049888070869, + "learning_rate": 1.9825285215984495e-05, + "loss": 1.2212, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 193 + }, + { + "epoch": 0.23921085080147966, + "grad_norm": 2.1418697560189988, + "learning_rate": 1.9822823097597893e-05, + "loss": 1.2602, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 194 + }, + { + "epoch": 0.2404438964241677, + "grad_norm": 2.049709663111986, + "learning_rate": 1.9820343907072982e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 195 + }, + { + "epoch": 0.24167694204685575, + "grad_norm": 2.1074579631320605, + "learning_rate": 1.9817847648718615e-05, + "loss": 1.1863, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 196 + }, + { + "epoch": 0.24290998766954378, + "grad_norm": 2.279482709995341, + "learning_rate": 1.9815334326873304e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 197 + }, + { + "epoch": 0.2441430332922318, + "grad_norm": 2.055596994912923, + "learning_rate": 1.9812803945905215e-05, + "loss": 1.2952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 198 + }, + { + "epoch": 0.24537607891491986, + "grad_norm": 2.132145299143716, + "learning_rate": 1.9810256510212178e-05, + "loss": 1.2276, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 199 + }, + { + "epoch": 0.2466091245376079, + "grad_norm": 2.077623874214478, + "learning_rate": 1.9807692024221647e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 200 + }, + { + "epoch": 0.24784217016029594, + "grad_norm": 2.0797643511906356, + "learning_rate": 1.9805110492390723e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 201 + }, + { + "epoch": 0.24907521578298397, + "grad_norm": 2.0930244790953854, + "learning_rate": 1.9802511919206123e-05, + "loss": 1.1984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 202 + }, + { + "epoch": 0.25030826140567203, + "grad_norm": 1.9804465011478614, + "learning_rate": 1.9799896309184192e-05, + "loss": 1.1436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 203 + }, + { + "epoch": 0.25154130702836003, + "grad_norm": 2.054252771935451, + "learning_rate": 1.9797263666870876e-05, + "loss": 1.1913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 204 + }, + { + "epoch": 0.2527743526510481, + "grad_norm": 1.9941485664509049, + "learning_rate": 1.9794613996841723e-05, + "loss": 1.1524, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 205 + }, + { + "epoch": 0.25400739827373614, + "grad_norm": 2.051706932133756, + "learning_rate": 1.9791947303701882e-05, + "loss": 1.2003, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 206 + }, + { + "epoch": 0.25524044389642414, + "grad_norm": 1.9992360549131547, + "learning_rate": 1.9789263592086085e-05, + "loss": 1.2392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 207 + }, + { + "epoch": 0.2564734895191122, + "grad_norm": 2.0558176906965375, + "learning_rate": 1.9786562866658644e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 208 + }, + { + "epoch": 0.25770653514180025, + "grad_norm": 2.2225484776914426, + "learning_rate": 1.9783845132113438e-05, + "loss": 1.1771, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 209 + }, + { + "epoch": 0.2589395807644883, + "grad_norm": 2.2832873135247516, + "learning_rate": 1.978111039317391e-05, + "loss": 1.2226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 210 + }, + { + "epoch": 0.2601726263871763, + "grad_norm": 1.9887162353358598, + "learning_rate": 1.9778358654593054e-05, + "loss": 1.1936, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 211 + }, + { + "epoch": 0.26140567200986436, + "grad_norm": 1.9738209615627333, + "learning_rate": 1.9775589921153417e-05, + "loss": 1.2666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 212 + }, + { + "epoch": 0.2626387176325524, + "grad_norm": 1.8817044406816048, + "learning_rate": 1.9772804197667074e-05, + "loss": 1.1842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 213 + }, + { + "epoch": 0.2638717632552404, + "grad_norm": 2.013799438378123, + "learning_rate": 1.9770001488975632e-05, + "loss": 1.2242, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 214 + }, + { + "epoch": 0.2651048088779285, + "grad_norm": 1.9119015873636653, + "learning_rate": 1.9767181799950222e-05, + "loss": 1.0898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 215 + }, + { + "epoch": 0.26633785450061653, + "grad_norm": 1.9353703261833344, + "learning_rate": 1.976434513549148e-05, + "loss": 1.2241, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 216 + }, + { + "epoch": 0.2675709001233046, + "grad_norm": 1.8684279347954125, + "learning_rate": 1.9761491500529555e-05, + "loss": 1.209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 217 + }, + { + "epoch": 0.2688039457459926, + "grad_norm": 2.1880109676776556, + "learning_rate": 1.9758620900024083e-05, + "loss": 1.1956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 218 + }, + { + "epoch": 0.27003699136868065, + "grad_norm": 2.083271319194177, + "learning_rate": 1.9755733338964186e-05, + "loss": 1.3062, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 219 + }, + { + "epoch": 0.2712700369913687, + "grad_norm": 2.093319969744569, + "learning_rate": 1.9752828822368468e-05, + "loss": 1.2083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 220 + }, + { + "epoch": 0.2725030826140567, + "grad_norm": 1.8811834241518346, + "learning_rate": 1.9749907355285e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 221 + }, + { + "epoch": 0.27373612823674476, + "grad_norm": 2.012899236904161, + "learning_rate": 1.9746968942791314e-05, + "loss": 1.1476, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 222 + }, + { + "epoch": 0.2749691738594328, + "grad_norm": 1.9990139044183095, + "learning_rate": 1.9744013589994397e-05, + "loss": 1.1257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 223 + }, + { + "epoch": 0.2762022194821208, + "grad_norm": 1.9924345221184356, + "learning_rate": 1.9741041302030666e-05, + "loss": 1.224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 224 + }, + { + "epoch": 0.27743526510480887, + "grad_norm": 2.2180474451395344, + "learning_rate": 1.9738052084065982e-05, + "loss": 1.1713, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 225 + }, + { + "epoch": 0.2786683107274969, + "grad_norm": 1.983690334551248, + "learning_rate": 1.973504594129563e-05, + "loss": 1.179, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 226 + }, + { + "epoch": 0.279901356350185, + "grad_norm": 2.283777738854751, + "learning_rate": 1.97320228789443e-05, + "loss": 1.2869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 227 + }, + { + "epoch": 0.281134401972873, + "grad_norm": 1.943369966527272, + "learning_rate": 1.9728982902266107e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 228 + }, + { + "epoch": 0.28236744759556104, + "grad_norm": 2.0711638809016333, + "learning_rate": 1.9725926016544544e-05, + "loss": 1.16, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 229 + }, + { + "epoch": 0.2836004932182491, + "grad_norm": 1.9786842049852629, + "learning_rate": 1.972285222709251e-05, + "loss": 1.2338, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 230 + }, + { + "epoch": 0.2848335388409371, + "grad_norm": 1.9447068528661613, + "learning_rate": 1.9719761539252268e-05, + "loss": 1.1827, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 231 + }, + { + "epoch": 0.28606658446362515, + "grad_norm": 2.0161169975225475, + "learning_rate": 1.971665395839545e-05, + "loss": 1.117, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 232 + }, + { + "epoch": 0.2872996300863132, + "grad_norm": 2.181216178694371, + "learning_rate": 1.9713529489923067e-05, + "loss": 1.1807, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 233 + }, + { + "epoch": 0.2885326757090012, + "grad_norm": 2.254001309068264, + "learning_rate": 1.9710388139265458e-05, + "loss": 1.0036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 234 + }, + { + "epoch": 0.28976572133168926, + "grad_norm": 2.028042430685245, + "learning_rate": 1.9707229911882318e-05, + "loss": 1.1175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 235 + }, + { + "epoch": 0.2909987669543773, + "grad_norm": 2.0601619279255177, + "learning_rate": 1.9704054813262666e-05, + "loss": 1.2758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 236 + }, + { + "epoch": 0.2922318125770654, + "grad_norm": 2.125107661538684, + "learning_rate": 1.9700862848924852e-05, + "loss": 1.2611, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 237 + }, + { + "epoch": 0.2934648581997534, + "grad_norm": 2.0416663193733746, + "learning_rate": 1.9697654024416528e-05, + "loss": 1.1986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 238 + }, + { + "epoch": 0.29469790382244143, + "grad_norm": 2.0193702602947554, + "learning_rate": 1.9694428345314657e-05, + "loss": 1.181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 239 + }, + { + "epoch": 0.2959309494451295, + "grad_norm": 2.0150819278141148, + "learning_rate": 1.9691185817225493e-05, + "loss": 1.1118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 240 + }, + { + "epoch": 0.2971639950678175, + "grad_norm": 2.355208602911491, + "learning_rate": 1.9687926445784576e-05, + "loss": 1.1013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 241 + }, + { + "epoch": 0.29839704069050554, + "grad_norm": 1.9770821951357245, + "learning_rate": 1.968465023665671e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 242 + }, + { + "epoch": 0.2996300863131936, + "grad_norm": 1.992866162682536, + "learning_rate": 1.9681357195535974e-05, + "loss": 1.2264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 243 + }, + { + "epoch": 0.3008631319358816, + "grad_norm": 1.9560049490402185, + "learning_rate": 1.9678047328145702e-05, + "loss": 1.1903, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 244 + }, + { + "epoch": 0.30209617755856966, + "grad_norm": 2.0070855957858114, + "learning_rate": 1.967472064023846e-05, + "loss": 1.1793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 245 + }, + { + "epoch": 0.3033292231812577, + "grad_norm": 2.0661988646186313, + "learning_rate": 1.9671377137596064e-05, + "loss": 1.2222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 246 + }, + { + "epoch": 0.30456226880394577, + "grad_norm": 2.1056352825098816, + "learning_rate": 1.9668016826029538e-05, + "loss": 1.2107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 247 + }, + { + "epoch": 0.30579531442663377, + "grad_norm": 1.98815122638957, + "learning_rate": 1.9664639711379135e-05, + "loss": 1.2816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 248 + }, + { + "epoch": 0.3070283600493218, + "grad_norm": 2.1440514280412737, + "learning_rate": 1.9661245799514298e-05, + "loss": 1.2021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 249 + }, + { + "epoch": 0.3082614056720099, + "grad_norm": 2.0231899704902805, + "learning_rate": 1.9657835096333676e-05, + "loss": 1.2358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 250 + }, + { + "epoch": 0.3094944512946979, + "grad_norm": 1.9883606430802652, + "learning_rate": 1.9654407607765098e-05, + "loss": 1.1847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 251 + }, + { + "epoch": 0.31072749691738594, + "grad_norm": 2.1163621969666786, + "learning_rate": 1.965096333976556e-05, + "loss": 1.2595, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 252 + }, + { + "epoch": 0.311960542540074, + "grad_norm": 1.9898069180734743, + "learning_rate": 1.9647502298321223e-05, + "loss": 1.2286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 253 + }, + { + "epoch": 0.31319358816276205, + "grad_norm": 1.929003891101706, + "learning_rate": 1.9644024489447403e-05, + "loss": 1.2394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 254 + }, + { + "epoch": 0.31442663378545005, + "grad_norm": 1.98661541473719, + "learning_rate": 1.9640529919188562e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 255 + }, + { + "epoch": 0.3156596794081381, + "grad_norm": 1.9483621179175639, + "learning_rate": 1.9637018593618285e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 256 + }, + { + "epoch": 0.31689272503082616, + "grad_norm": 1.8370569270185535, + "learning_rate": 1.9633490518839285e-05, + "loss": 1.1627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 257 + }, + { + "epoch": 0.31812577065351416, + "grad_norm": 2.2350844635149376, + "learning_rate": 1.9629945700983376e-05, + "loss": 1.1717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 258 + }, + { + "epoch": 0.3193588162762022, + "grad_norm": 2.3938085963940847, + "learning_rate": 1.9626384146211482e-05, + "loss": 1.2982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 259 + }, + { + "epoch": 0.3205918618988903, + "grad_norm": 1.9382250145807791, + "learning_rate": 1.962280586071361e-05, + "loss": 1.1414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 260 + }, + { + "epoch": 0.3218249075215783, + "grad_norm": 2.048034832679974, + "learning_rate": 1.9619210850708844e-05, + "loss": 1.2367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 261 + }, + { + "epoch": 0.32305795314426633, + "grad_norm": 1.868035630300947, + "learning_rate": 1.9615599122445338e-05, + "loss": 1.1582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 262 + }, + { + "epoch": 0.3242909987669544, + "grad_norm": 1.9842993509323321, + "learning_rate": 1.9611970682200308e-05, + "loss": 1.1214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 263 + }, + { + "epoch": 0.32552404438964244, + "grad_norm": 2.009754214636891, + "learning_rate": 1.9608325536280003e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 264 + }, + { + "epoch": 0.32675709001233044, + "grad_norm": 2.0322676023608097, + "learning_rate": 1.9604663691019715e-05, + "loss": 1.1514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 265 + }, + { + "epoch": 0.3279901356350185, + "grad_norm": 1.9898336979024067, + "learning_rate": 1.960098515278376e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 266 + }, + { + "epoch": 0.32922318125770655, + "grad_norm": 2.1120578466218007, + "learning_rate": 1.959728992796546e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 267 + }, + { + "epoch": 0.33045622688039455, + "grad_norm": 2.230681018136028, + "learning_rate": 1.959357802298715e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 268 + }, + { + "epoch": 0.3316892725030826, + "grad_norm": 2.121990574572694, + "learning_rate": 1.9589849444300142e-05, + "loss": 1.1686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 269 + }, + { + "epoch": 0.33292231812577067, + "grad_norm": 1.9793276124371497, + "learning_rate": 1.958610419838473e-05, + "loss": 1.2605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 270 + }, + { + "epoch": 0.33415536374845867, + "grad_norm": 2.160304448339165, + "learning_rate": 1.958234229175019e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 271 + }, + { + "epoch": 0.3353884093711467, + "grad_norm": 2.0375075579110646, + "learning_rate": 1.9578563730934733e-05, + "loss": 1.1282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 272 + }, + { + "epoch": 0.3366214549938348, + "grad_norm": 2.0496730732316313, + "learning_rate": 1.9574768522505528e-05, + "loss": 1.2336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 273 + }, + { + "epoch": 0.33785450061652283, + "grad_norm": 2.0096053009394987, + "learning_rate": 1.9570956673058678e-05, + "loss": 1.1737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 274 + }, + { + "epoch": 0.33908754623921084, + "grad_norm": 2.064991805627586, + "learning_rate": 1.9567128189219203e-05, + "loss": 1.2067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 275 + }, + { + "epoch": 0.3403205918618989, + "grad_norm": 2.016629291677699, + "learning_rate": 1.9563283077641032e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 276 + }, + { + "epoch": 0.34155363748458695, + "grad_norm": 2.026571915003702, + "learning_rate": 1.9559421345007004e-05, + "loss": 0.9722, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 277 + }, + { + "epoch": 0.34278668310727495, + "grad_norm": 2.0408528020000896, + "learning_rate": 1.9555542998028836e-05, + "loss": 1.0951, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 278 + }, + { + "epoch": 0.344019728729963, + "grad_norm": 2.4490550716870336, + "learning_rate": 1.955164804344712e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 279 + }, + { + "epoch": 0.34525277435265106, + "grad_norm": 2.0298228033571517, + "learning_rate": 1.9547736488031322e-05, + "loss": 1.2091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 280 + }, + { + "epoch": 0.34648581997533906, + "grad_norm": 1.928172339548102, + "learning_rate": 1.9543808338579746e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 281 + }, + { + "epoch": 0.3477188655980271, + "grad_norm": 1.9880687290891643, + "learning_rate": 1.953986360191955e-05, + "loss": 1.0397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 282 + }, + { + "epoch": 0.34895191122071517, + "grad_norm": 2.3895656693876823, + "learning_rate": 1.9535902284906717e-05, + "loss": 1.2004, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 283 + }, + { + "epoch": 0.35018495684340323, + "grad_norm": 1.9995303874089259, + "learning_rate": 1.9531924394426036e-05, + "loss": 1.2001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 284 + }, + { + "epoch": 0.35141800246609123, + "grad_norm": 1.9751518073159213, + "learning_rate": 1.9527929937391118e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 285 + }, + { + "epoch": 0.3526510480887793, + "grad_norm": 2.002111448708798, + "learning_rate": 1.9523918920744357e-05, + "loss": 1.0845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 286 + }, + { + "epoch": 0.35388409371146734, + "grad_norm": 2.0131766303020036, + "learning_rate": 1.9519891351456926e-05, + "loss": 1.1764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 287 + }, + { + "epoch": 0.35511713933415534, + "grad_norm": 2.2331356025573785, + "learning_rate": 1.9515847236528772e-05, + "loss": 1.225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 288 + }, + { + "epoch": 0.3563501849568434, + "grad_norm": 2.0628294266012204, + "learning_rate": 1.9511786582988597e-05, + "loss": 1.2259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 289 + }, + { + "epoch": 0.35758323057953145, + "grad_norm": 1.9151442433897523, + "learning_rate": 1.9507709397893846e-05, + "loss": 1.2103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 290 + }, + { + "epoch": 0.3588162762022195, + "grad_norm": 1.958231535128413, + "learning_rate": 1.9503615688330698e-05, + "loss": 1.105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 291 + }, + { + "epoch": 0.3600493218249075, + "grad_norm": 2.149244076290205, + "learning_rate": 1.9499505461414045e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 292 + }, + { + "epoch": 0.36128236744759556, + "grad_norm": 2.017117422238215, + "learning_rate": 1.9495378724287498e-05, + "loss": 1.1007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 293 + }, + { + "epoch": 0.3625154130702836, + "grad_norm": 1.9210343338897897, + "learning_rate": 1.949123548412335e-05, + "loss": 1.0755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 294 + }, + { + "epoch": 0.3637484586929716, + "grad_norm": 1.9955568267380113, + "learning_rate": 1.9487075748122586e-05, + "loss": 1.1702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 295 + }, + { + "epoch": 0.3649815043156597, + "grad_norm": 2.014663690647869, + "learning_rate": 1.948289952351486e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 296 + }, + { + "epoch": 0.36621454993834773, + "grad_norm": 2.0560814736523922, + "learning_rate": 1.9478706817558474e-05, + "loss": 1.1981, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 297 + }, + { + "epoch": 0.36744759556103573, + "grad_norm": 2.072244571774935, + "learning_rate": 1.947449763754039e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 298 + }, + { + "epoch": 0.3686806411837238, + "grad_norm": 2.3134145282189724, + "learning_rate": 1.947027199077619e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 299 + }, + { + "epoch": 0.36991368680641185, + "grad_norm": 1.9505648569157141, + "learning_rate": 1.9466029884610072e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 300 + }, + { + "epoch": 0.3711467324290999, + "grad_norm": 1.9850708859825466, + "learning_rate": 1.946177132641486e-05, + "loss": 1.0863, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 301 + }, + { + "epoch": 0.3723797780517879, + "grad_norm": 2.3267730903818755, + "learning_rate": 1.945749632359195e-05, + "loss": 1.1405, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 302 + }, + { + "epoch": 0.37361282367447596, + "grad_norm": 2.403438763479749, + "learning_rate": 1.9453204883571333e-05, + "loss": 1.1977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 303 + }, + { + "epoch": 0.374845869297164, + "grad_norm": 2.028737664163498, + "learning_rate": 1.9448897013811558e-05, + "loss": 1.2029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 304 + }, + { + "epoch": 0.376078914919852, + "grad_norm": 1.9981856925053472, + "learning_rate": 1.9444572721799743e-05, + "loss": 1.1873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 305 + }, + { + "epoch": 0.37731196054254007, + "grad_norm": 1.968701546642257, + "learning_rate": 1.9440232015051526e-05, + "loss": 1.1721, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 306 + }, + { + "epoch": 0.3785450061652281, + "grad_norm": 1.864390281708884, + "learning_rate": 1.9435874901111102e-05, + "loss": 1.1797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 307 + }, + { + "epoch": 0.3797780517879161, + "grad_norm": 2.0705929640726577, + "learning_rate": 1.9431501387551158e-05, + "loss": 1.1453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 308 + }, + { + "epoch": 0.3810110974106042, + "grad_norm": 1.884359293663656, + "learning_rate": 1.942711148197289e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 309 + }, + { + "epoch": 0.38224414303329224, + "grad_norm": 1.936787839699547, + "learning_rate": 1.9422705192005994e-05, + "loss": 1.2022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 310 + }, + { + "epoch": 0.3834771886559803, + "grad_norm": 1.9636401987324612, + "learning_rate": 1.941828252530863e-05, + "loss": 1.2015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 311 + }, + { + "epoch": 0.3847102342786683, + "grad_norm": 2.0693459412185495, + "learning_rate": 1.9413843489567423e-05, + "loss": 1.054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 312 + }, + { + "epoch": 0.38594327990135635, + "grad_norm": 2.0182958064572256, + "learning_rate": 1.9409388092497453e-05, + "loss": 1.1086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 313 + }, + { + "epoch": 0.3871763255240444, + "grad_norm": 2.093433911405551, + "learning_rate": 1.940491634184223e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 314 + }, + { + "epoch": 0.3884093711467324, + "grad_norm": 2.018372038783884, + "learning_rate": 1.9400428245373693e-05, + "loss": 1.0813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 315 + }, + { + "epoch": 0.38964241676942046, + "grad_norm": 1.9703329339612685, + "learning_rate": 1.9395923810892184e-05, + "loss": 1.0768, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 316 + }, + { + "epoch": 0.3908754623921085, + "grad_norm": 1.9327973664586777, + "learning_rate": 1.939140304622644e-05, + "loss": 1.1157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 317 + }, + { + "epoch": 0.3921085080147966, + "grad_norm": 2.019497002973341, + "learning_rate": 1.938686595923359e-05, + "loss": 0.9835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 318 + }, + { + "epoch": 0.3933415536374846, + "grad_norm": 2.089968778066188, + "learning_rate": 1.9382312557799116e-05, + "loss": 1.0213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 319 + }, + { + "epoch": 0.39457459926017263, + "grad_norm": 2.0510099518796006, + "learning_rate": 1.937774284983687e-05, + "loss": 1.099, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 320 + }, + { + "epoch": 0.3958076448828607, + "grad_norm": 2.0175024562414063, + "learning_rate": 1.9373156843289037e-05, + "loss": 1.0764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 321 + }, + { + "epoch": 0.3970406905055487, + "grad_norm": 2.020977606778504, + "learning_rate": 1.9368554546126126e-05, + "loss": 1.0562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 322 + }, + { + "epoch": 0.39827373612823674, + "grad_norm": 2.2138761252967125, + "learning_rate": 1.936393596634697e-05, + "loss": 1.196, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 323 + }, + { + "epoch": 0.3995067817509248, + "grad_norm": 1.855560397827274, + "learning_rate": 1.9359301111978683e-05, + "loss": 1.0725, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 324 + }, + { + "epoch": 0.4007398273736128, + "grad_norm": 2.0805790476522636, + "learning_rate": 1.9354649991076684e-05, + "loss": 1.1665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 325 + }, + { + "epoch": 0.40197287299630086, + "grad_norm": 1.8807037569919354, + "learning_rate": 1.9349982611724654e-05, + "loss": 1.1366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 326 + }, + { + "epoch": 0.4032059186189889, + "grad_norm": 1.9286740986289053, + "learning_rate": 1.9345298982034536e-05, + "loss": 1.132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 327 + }, + { + "epoch": 0.40443896424167697, + "grad_norm": 2.0846402787959786, + "learning_rate": 1.9340599110146504e-05, + "loss": 1.1055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 328 + }, + { + "epoch": 0.40567200986436497, + "grad_norm": 2.2183587793468993, + "learning_rate": 1.9335883004228974e-05, + "loss": 1.1544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 329 + }, + { + "epoch": 0.406905055487053, + "grad_norm": 2.2213015226303643, + "learning_rate": 1.9331150672478575e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 330 + }, + { + "epoch": 0.4081381011097411, + "grad_norm": 2.2277113238319726, + "learning_rate": 1.9326402123120132e-05, + "loss": 1.0451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 331 + }, + { + "epoch": 0.4093711467324291, + "grad_norm": 2.259911506493339, + "learning_rate": 1.9321637364406658e-05, + "loss": 1.0236, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 332 + }, + { + "epoch": 0.41060419235511714, + "grad_norm": 2.3573090703399417, + "learning_rate": 1.931685640461934e-05, + "loss": 1.1412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 333 + }, + { + "epoch": 0.4118372379778052, + "grad_norm": 2.1097456297814468, + "learning_rate": 1.9312059252067517e-05, + "loss": 1.1901, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 334 + }, + { + "epoch": 0.4130702836004932, + "grad_norm": 2.121952527251297, + "learning_rate": 1.930724591508868e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 335 + }, + { + "epoch": 0.41430332922318125, + "grad_norm": 2.091327214317948, + "learning_rate": 1.930241640204844e-05, + "loss": 1.0135, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 336 + }, + { + "epoch": 0.4155363748458693, + "grad_norm": 2.1272216173587255, + "learning_rate": 1.929757072134053e-05, + "loss": 1.1754, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 337 + }, + { + "epoch": 0.41676942046855736, + "grad_norm": 1.9973226929736474, + "learning_rate": 1.929270888138678e-05, + "loss": 1.1392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 338 + }, + { + "epoch": 0.41800246609124536, + "grad_norm": 1.9318995412408386, + "learning_rate": 1.9287830890637093e-05, + "loss": 1.0762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 339 + }, + { + "epoch": 0.4192355117139334, + "grad_norm": 2.785840304521055, + "learning_rate": 1.9282936757569463e-05, + "loss": 1.1699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 340 + }, + { + "epoch": 0.4204685573366215, + "grad_norm": 2.171910512782072, + "learning_rate": 1.9278026490689925e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 341 + }, + { + "epoch": 0.4217016029593095, + "grad_norm": 2.052852932277261, + "learning_rate": 1.927310009853256e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 342 + }, + { + "epoch": 0.42293464858199753, + "grad_norm": 1.9538275366336955, + "learning_rate": 1.926815758965947e-05, + "loss": 1.1934, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 343 + }, + { + "epoch": 0.4241676942046856, + "grad_norm": 1.9772612426660126, + "learning_rate": 1.9263198972660774e-05, + "loss": 1.1061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 344 + }, + { + "epoch": 0.4254007398273736, + "grad_norm": 1.9522364508331842, + "learning_rate": 1.9258224256154588e-05, + "loss": 1.101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 345 + }, + { + "epoch": 0.42663378545006164, + "grad_norm": 2.1060276682972714, + "learning_rate": 1.9253233448787e-05, + "loss": 1.2094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 346 + }, + { + "epoch": 0.4278668310727497, + "grad_norm": 2.020183264992464, + "learning_rate": 1.9248226559232074e-05, + "loss": 1.2012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 347 + }, + { + "epoch": 0.42909987669543775, + "grad_norm": 2.1361451948859256, + "learning_rate": 1.9243203596191815e-05, + "loss": 1.2714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 348 + }, + { + "epoch": 0.43033292231812575, + "grad_norm": 1.890346669461953, + "learning_rate": 1.9238164568396177e-05, + "loss": 1.067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 349 + }, + { + "epoch": 0.4315659679408138, + "grad_norm": 1.908858157385325, + "learning_rate": 1.9233109484603025e-05, + "loss": 1.1123, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 350 + }, + { + "epoch": 0.43279901356350187, + "grad_norm": 1.9095667461482675, + "learning_rate": 1.9228038353598122e-05, + "loss": 1.0834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 351 + }, + { + "epoch": 0.43403205918618987, + "grad_norm": 1.8950946192756821, + "learning_rate": 1.9222951184195146e-05, + "loss": 1.0686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 352 + }, + { + "epoch": 0.4352651048088779, + "grad_norm": 2.5409486063614066, + "learning_rate": 1.921784798523562e-05, + "loss": 1.1168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 353 + }, + { + "epoch": 0.436498150431566, + "grad_norm": 1.9251760877216109, + "learning_rate": 1.9212728765588954e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 354 + }, + { + "epoch": 0.43773119605425403, + "grad_norm": 2.12327469473301, + "learning_rate": 1.9207593534152383e-05, + "loss": 1.1871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 355 + }, + { + "epoch": 0.43896424167694204, + "grad_norm": 2.0404259513933214, + "learning_rate": 1.9202442299850977e-05, + "loss": 1.0985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 356 + }, + { + "epoch": 0.4401972872996301, + "grad_norm": 2.10787306308974, + "learning_rate": 1.9197275071637617e-05, + "loss": 1.0063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 357 + }, + { + "epoch": 0.44143033292231815, + "grad_norm": 2.2402685683097996, + "learning_rate": 1.919209185849299e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 358 + }, + { + "epoch": 0.44266337854500615, + "grad_norm": 2.148922705064427, + "learning_rate": 1.9186892669425546e-05, + "loss": 1.0889, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 359 + }, + { + "epoch": 0.4438964241676942, + "grad_norm": 1.8789248361497928, + "learning_rate": 1.9181677513471525e-05, + "loss": 1.1796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 360 + }, + { + "epoch": 0.44512946979038226, + "grad_norm": 1.8886466190263609, + "learning_rate": 1.9176446399694903e-05, + "loss": 1.1494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 361 + }, + { + "epoch": 0.44636251541307026, + "grad_norm": 1.9432580511904547, + "learning_rate": 1.9171199337187395e-05, + "loss": 1.2599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 362 + }, + { + "epoch": 0.4475955610357583, + "grad_norm": 1.8820277177030855, + "learning_rate": 1.916593633506843e-05, + "loss": 1.1884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 363 + }, + { + "epoch": 0.44882860665844637, + "grad_norm": 2.249700768154195, + "learning_rate": 1.9160657402485147e-05, + "loss": 1.0503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 364 + }, + { + "epoch": 0.45006165228113443, + "grad_norm": 2.082324666968816, + "learning_rate": 1.9155362548612366e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 365 + }, + { + "epoch": 0.45129469790382243, + "grad_norm": 2.2090950658394193, + "learning_rate": 1.915005178265259e-05, + "loss": 1.1742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 366 + }, + { + "epoch": 0.4525277435265105, + "grad_norm": 2.2007793765172683, + "learning_rate": 1.914472511383596e-05, + "loss": 1.2617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 367 + }, + { + "epoch": 0.45376078914919854, + "grad_norm": 1.9166813253050292, + "learning_rate": 1.913938255142027e-05, + "loss": 1.1444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 368 + }, + { + "epoch": 0.45499383477188654, + "grad_norm": 1.908891419449054, + "learning_rate": 1.913402410469093e-05, + "loss": 1.1139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 369 + }, + { + "epoch": 0.4562268803945746, + "grad_norm": 2.0434795360025566, + "learning_rate": 1.9128649782960965e-05, + "loss": 1.1761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 370 + }, + { + "epoch": 0.45745992601726265, + "grad_norm": 2.2319795236629165, + "learning_rate": 1.9123259595570978e-05, + "loss": 0.9997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 371 + }, + { + "epoch": 0.45869297163995065, + "grad_norm": 1.9325225410679332, + "learning_rate": 1.911785355188916e-05, + "loss": 1.1191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 372 + }, + { + "epoch": 0.4599260172626387, + "grad_norm": 2.3718583427313384, + "learning_rate": 1.9112431661311248e-05, + "loss": 1.1801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 373 + }, + { + "epoch": 0.46115906288532676, + "grad_norm": 2.2778458302261853, + "learning_rate": 1.9106993933260535e-05, + "loss": 1.1973, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 374 + }, + { + "epoch": 0.4623921085080148, + "grad_norm": 2.104485888382934, + "learning_rate": 1.910154037718783e-05, + "loss": 1.206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 375 + }, + { + "epoch": 0.4636251541307028, + "grad_norm": 2.149849667802547, + "learning_rate": 1.909607100257145e-05, + "loss": 1.0479, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 376 + }, + { + "epoch": 0.4648581997533909, + "grad_norm": 2.242058517922103, + "learning_rate": 1.9090585818917205e-05, + "loss": 1.0902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 377 + }, + { + "epoch": 0.46609124537607893, + "grad_norm": 2.0943438214587307, + "learning_rate": 1.9085084835758394e-05, + "loss": 1.2428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 378 + }, + { + "epoch": 0.46732429099876693, + "grad_norm": 2.1534146036573647, + "learning_rate": 1.907956806265576e-05, + "loss": 1.0392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 379 + }, + { + "epoch": 0.468557336621455, + "grad_norm": 2.2361240083173324, + "learning_rate": 1.9074035509197495e-05, + "loss": 1.1337, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 380 + }, + { + "epoch": 0.46979038224414305, + "grad_norm": 2.0666798517773795, + "learning_rate": 1.906848718499922e-05, + "loss": 1.0574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 381 + }, + { + "epoch": 0.47102342786683105, + "grad_norm": 2.0056597359218094, + "learning_rate": 1.906292309970396e-05, + "loss": 1.1982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 382 + }, + { + "epoch": 0.4722564734895191, + "grad_norm": 1.931784648165126, + "learning_rate": 1.9057343262982136e-05, + "loss": 1.0671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 383 + }, + { + "epoch": 0.47348951911220716, + "grad_norm": 2.2650382784813408, + "learning_rate": 1.905174768453155e-05, + "loss": 1.1384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 384 + }, + { + "epoch": 0.4747225647348952, + "grad_norm": 2.2386337800038594, + "learning_rate": 1.904613637407735e-05, + "loss": 1.169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 385 + }, + { + "epoch": 0.4759556103575832, + "grad_norm": 1.985439502238856, + "learning_rate": 1.9040509341372045e-05, + "loss": 1.0544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 386 + }, + { + "epoch": 0.47718865598027127, + "grad_norm": 1.9991995362972141, + "learning_rate": 1.903486659619545e-05, + "loss": 1.1048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 387 + }, + { + "epoch": 0.4784217016029593, + "grad_norm": 1.938313304762426, + "learning_rate": 1.90292081483547e-05, + "loss": 1.0217, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 388 + }, + { + "epoch": 0.4796547472256473, + "grad_norm": 2.0593580639988605, + "learning_rate": 1.9023534007684217e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 389 + }, + { + "epoch": 0.4808877928483354, + "grad_norm": 2.0113392094384723, + "learning_rate": 1.90178441840457e-05, + "loss": 1.2114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 390 + }, + { + "epoch": 0.48212083847102344, + "grad_norm": 1.9865461362024583, + "learning_rate": 1.901213868732811e-05, + "loss": 1.0818, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 391 + }, + { + "epoch": 0.4833538840937115, + "grad_norm": 1.9044551829036014, + "learning_rate": 1.9006417527447634e-05, + "loss": 1.1046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 392 + }, + { + "epoch": 0.4845869297163995, + "grad_norm": 1.8508018232087575, + "learning_rate": 1.9000680714347694e-05, + "loss": 1.1411, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 393 + }, + { + "epoch": 0.48581997533908755, + "grad_norm": 1.98419367965867, + "learning_rate": 1.8994928257998907e-05, + "loss": 0.9804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 394 + }, + { + "epoch": 0.4870530209617756, + "grad_norm": 2.1671038349199567, + "learning_rate": 1.8989160168399093e-05, + "loss": 1.1656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 395 + }, + { + "epoch": 0.4882860665844636, + "grad_norm": 2.0570349889938284, + "learning_rate": 1.898337645557323e-05, + "loss": 1.2111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 396 + }, + { + "epoch": 0.48951911220715166, + "grad_norm": 1.9619760913338724, + "learning_rate": 1.8977577129573452e-05, + "loss": 1.0932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 397 + }, + { + "epoch": 0.4907521578298397, + "grad_norm": 1.9608046473112009, + "learning_rate": 1.897176220047903e-05, + "loss": 1.1389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 398 + }, + { + "epoch": 0.4919852034525277, + "grad_norm": 1.9645035824539876, + "learning_rate": 1.8965931678396357e-05, + "loss": 1.1056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 399 + }, + { + "epoch": 0.4932182490752158, + "grad_norm": 1.961734329156117, + "learning_rate": 1.896008557345892e-05, + "loss": 1.1322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 400 + }, + { + "epoch": 0.49445129469790383, + "grad_norm": 1.844560254648902, + "learning_rate": 1.8954223895827288e-05, + "loss": 1.0712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 401 + }, + { + "epoch": 0.4956843403205919, + "grad_norm": 2.1465454734369933, + "learning_rate": 1.8948346655689108e-05, + "loss": 1.1294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 402 + }, + { + "epoch": 0.4969173859432799, + "grad_norm": 2.078244569426143, + "learning_rate": 1.8942453863259058e-05, + "loss": 1.0546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 403 + }, + { + "epoch": 0.49815043156596794, + "grad_norm": 2.264508341916039, + "learning_rate": 1.893654552877886e-05, + "loss": 1.0568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 404 + }, + { + "epoch": 0.499383477188656, + "grad_norm": 2.0051680479216696, + "learning_rate": 1.893062166251724e-05, + "loss": 1.0156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 405 + }, + { + "epoch": 0.5006165228113441, + "grad_norm": 2.1569911537953725, + "learning_rate": 1.8924682274769925e-05, + "loss": 1.173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 406 + }, + { + "epoch": 0.5018495684340321, + "grad_norm": 2.002983438517559, + "learning_rate": 1.8918727375859608e-05, + "loss": 1.0073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 407 + }, + { + "epoch": 0.5030826140567201, + "grad_norm": 1.9516175334775212, + "learning_rate": 1.891275697613595e-05, + "loss": 1.1404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 408 + }, + { + "epoch": 0.5043156596794082, + "grad_norm": 2.1494580236712832, + "learning_rate": 1.890677108597555e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 409 + }, + { + "epoch": 0.5055487053020962, + "grad_norm": 2.025395779691303, + "learning_rate": 1.8900769715781928e-05, + "loss": 0.9756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 410 + }, + { + "epoch": 0.5067817509247842, + "grad_norm": 1.9927648271180456, + "learning_rate": 1.889475287598551e-05, + "loss": 1.1394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 411 + }, + { + "epoch": 0.5080147965474723, + "grad_norm": 1.9818065701766985, + "learning_rate": 1.8888720577043603e-05, + "loss": 1.1683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 412 + }, + { + "epoch": 0.5092478421701603, + "grad_norm": 1.9410101081952555, + "learning_rate": 1.8882672829440393e-05, + "loss": 1.0839, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 413 + }, + { + "epoch": 0.5104808877928483, + "grad_norm": 1.8396805813100958, + "learning_rate": 1.8876609643686906e-05, + "loss": 1.114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 414 + }, + { + "epoch": 0.5117139334155364, + "grad_norm": 2.0105399328568905, + "learning_rate": 1.8870531030321004e-05, + "loss": 1.1439, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 415 + }, + { + "epoch": 0.5129469790382244, + "grad_norm": 2.090098149989323, + "learning_rate": 1.886443699990736e-05, + "loss": 1.1153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 416 + }, + { + "epoch": 0.5141800246609125, + "grad_norm": 2.0517637089595246, + "learning_rate": 1.885832756303745e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 417 + }, + { + "epoch": 0.5154130702836005, + "grad_norm": 3.0738713157895674, + "learning_rate": 1.885220273032951e-05, + "loss": 1.0222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 418 + }, + { + "epoch": 0.5166461159062885, + "grad_norm": 2.6677283836123005, + "learning_rate": 1.884606251242855e-05, + "loss": 1.1038, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 419 + }, + { + "epoch": 0.5178791615289766, + "grad_norm": 2.125547466007061, + "learning_rate": 1.8839906920006315e-05, + "loss": 1.2128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 420 + }, + { + "epoch": 0.5191122071516646, + "grad_norm": 2.1599202425686013, + "learning_rate": 1.883373596376126e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 421 + }, + { + "epoch": 0.5203452527743526, + "grad_norm": 2.1229927533556436, + "learning_rate": 1.8827549654418566e-05, + "loss": 1.1353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 422 + }, + { + "epoch": 0.5215782983970407, + "grad_norm": 1.9567023805342967, + "learning_rate": 1.8821348002730072e-05, + "loss": 1.0584, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 423 + }, + { + "epoch": 0.5228113440197287, + "grad_norm": 1.8989916861837828, + "learning_rate": 1.88151310194743e-05, + "loss": 0.9837, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 424 + }, + { + "epoch": 0.5240443896424167, + "grad_norm": 2.1067680957156014, + "learning_rate": 1.8808898715456415e-05, + "loss": 1.0673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 425 + }, + { + "epoch": 0.5252774352651048, + "grad_norm": 2.6013242993173926, + "learning_rate": 1.88026511015082e-05, + "loss": 1.112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 426 + }, + { + "epoch": 0.5265104808877928, + "grad_norm": 2.0539914273851907, + "learning_rate": 1.8796388188488055e-05, + "loss": 1.1031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 427 + }, + { + "epoch": 0.5277435265104808, + "grad_norm": 2.0386084337690584, + "learning_rate": 1.8790109987280972e-05, + "loss": 1.1556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 428 + }, + { + "epoch": 0.528976572133169, + "grad_norm": 2.093419206457373, + "learning_rate": 1.878381650879851e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 429 + }, + { + "epoch": 0.530209617755857, + "grad_norm": 2.2549735586348985, + "learning_rate": 1.877750776397878e-05, + "loss": 1.0663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 430 + }, + { + "epoch": 0.531442663378545, + "grad_norm": 2.1062036310291266, + "learning_rate": 1.877118376378642e-05, + "loss": 1.0455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 431 + }, + { + "epoch": 0.5326757090012331, + "grad_norm": 1.995564286989001, + "learning_rate": 1.87648445192126e-05, + "loss": 0.982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 432 + }, + { + "epoch": 0.5339087546239211, + "grad_norm": 2.41307031725252, + "learning_rate": 1.8758490041274963e-05, + "loss": 1.0778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 433 + }, + { + "epoch": 0.5351418002466092, + "grad_norm": 2.831604603209417, + "learning_rate": 1.875212034101765e-05, + "loss": 1.1345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 434 + }, + { + "epoch": 0.5363748458692972, + "grad_norm": 2.2244728383006604, + "learning_rate": 1.8745735429511236e-05, + "loss": 0.938, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 435 + }, + { + "epoch": 0.5376078914919852, + "grad_norm": 2.0679863467245823, + "learning_rate": 1.8739335317852747e-05, + "loss": 1.0763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 436 + }, + { + "epoch": 0.5388409371146733, + "grad_norm": 2.2454972633071915, + "learning_rate": 1.8732920017165624e-05, + "loss": 1.1207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 437 + }, + { + "epoch": 0.5400739827373613, + "grad_norm": 1.952229569146223, + "learning_rate": 1.8726489538599707e-05, + "loss": 1.0255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 438 + }, + { + "epoch": 0.5413070283600493, + "grad_norm": 1.9236876015939406, + "learning_rate": 1.8720043893331214e-05, + "loss": 0.9933, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 439 + }, + { + "epoch": 0.5425400739827374, + "grad_norm": 1.9736270928366435, + "learning_rate": 1.8713583092562727e-05, + "loss": 1.107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 440 + }, + { + "epoch": 0.5437731196054254, + "grad_norm": 2.1162408965440633, + "learning_rate": 1.870710714752316e-05, + "loss": 1.0441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 441 + }, + { + "epoch": 0.5450061652281134, + "grad_norm": 1.9515336502135756, + "learning_rate": 1.8700616069467753e-05, + "loss": 0.9855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 442 + }, + { + "epoch": 0.5462392108508015, + "grad_norm": 2.566164154732447, + "learning_rate": 1.8694109869678047e-05, + "loss": 1.1289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 443 + }, + { + "epoch": 0.5474722564734895, + "grad_norm": 2.3448298186890026, + "learning_rate": 1.868758855946187e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 444 + }, + { + "epoch": 0.5487053020961775, + "grad_norm": 2.076679946731619, + "learning_rate": 1.8681052150153305e-05, + "loss": 1.0494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 445 + }, + { + "epoch": 0.5499383477188656, + "grad_norm": 2.281980101143931, + "learning_rate": 1.8674500653112674e-05, + "loss": 1.0456, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 446 + }, + { + "epoch": 0.5511713933415536, + "grad_norm": 2.147469472559495, + "learning_rate": 1.8667934079726537e-05, + "loss": 1.0682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 447 + }, + { + "epoch": 0.5524044389642416, + "grad_norm": 2.0621898279926527, + "learning_rate": 1.866135244140763e-05, + "loss": 1.0527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 448 + }, + { + "epoch": 0.5536374845869297, + "grad_norm": 2.0688727890116785, + "learning_rate": 1.8654755749594905e-05, + "loss": 1.0549, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 449 + }, + { + "epoch": 0.5548705302096177, + "grad_norm": 2.0101374651739325, + "learning_rate": 1.864814401575345e-05, + "loss": 1.002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 450 + }, + { + "epoch": 0.5561035758323057, + "grad_norm": 2.2861388762596344, + "learning_rate": 1.8641517251374514e-05, + "loss": 1.0666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 451 + }, + { + "epoch": 0.5573366214549939, + "grad_norm": 2.006862831841472, + "learning_rate": 1.8634875467975456e-05, + "loss": 1.0061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 452 + }, + { + "epoch": 0.5585696670776819, + "grad_norm": 2.144270279582644, + "learning_rate": 1.862821867709975e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 453 + }, + { + "epoch": 0.55980271270037, + "grad_norm": 1.9667742618855084, + "learning_rate": 1.862154689031694e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 454 + }, + { + "epoch": 0.561035758323058, + "grad_norm": 1.8857968215229095, + "learning_rate": 1.8614860119222643e-05, + "loss": 1.0861, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 455 + }, + { + "epoch": 0.562268803945746, + "grad_norm": 1.9374803828274445, + "learning_rate": 1.860815837543852e-05, + "loss": 0.9784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 456 + }, + { + "epoch": 0.5635018495684341, + "grad_norm": 2.0764596642416433, + "learning_rate": 1.8601441670612247e-05, + "loss": 1.1005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 457 + }, + { + "epoch": 0.5647348951911221, + "grad_norm": 2.123062949997989, + "learning_rate": 1.8594710016417505e-05, + "loss": 1.1953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 458 + }, + { + "epoch": 0.5659679408138101, + "grad_norm": 2.311252498934928, + "learning_rate": 1.858796342455396e-05, + "loss": 1.1049, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 459 + }, + { + "epoch": 0.5672009864364982, + "grad_norm": 1.9780001048890374, + "learning_rate": 1.8581201906747242e-05, + "loss": 1.1393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 460 + }, + { + "epoch": 0.5684340320591862, + "grad_norm": 1.968874552657262, + "learning_rate": 1.857442547474891e-05, + "loss": 1.1578, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 461 + }, + { + "epoch": 0.5696670776818742, + "grad_norm": 1.9395225862380807, + "learning_rate": 1.8567634140336457e-05, + "loss": 1.0823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 462 + }, + { + "epoch": 0.5709001233045623, + "grad_norm": 1.9422213147988507, + "learning_rate": 1.856082791531327e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 463 + }, + { + "epoch": 0.5721331689272503, + "grad_norm": 1.9095099328124767, + "learning_rate": 1.855400681150862e-05, + "loss": 1.1041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 464 + }, + { + "epoch": 0.5733662145499383, + "grad_norm": 1.9651570734782207, + "learning_rate": 1.8547170840777635e-05, + "loss": 0.9924, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 465 + }, + { + "epoch": 0.5745992601726264, + "grad_norm": 1.9083237960346455, + "learning_rate": 1.8540320015001283e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 466 + }, + { + "epoch": 0.5758323057953144, + "grad_norm": 2.0442809672147484, + "learning_rate": 1.8533454346086345e-05, + "loss": 0.9449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 467 + }, + { + "epoch": 0.5770653514180024, + "grad_norm": 2.839273126939154, + "learning_rate": 1.8526573845965407e-05, + "loss": 1.1446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 468 + }, + { + "epoch": 0.5782983970406905, + "grad_norm": 2.8401028851750083, + "learning_rate": 1.8519678526596827e-05, + "loss": 1.0793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 469 + }, + { + "epoch": 0.5795314426633785, + "grad_norm": 2.538177485476921, + "learning_rate": 1.8512768399964725e-05, + "loss": 1.1066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 470 + }, + { + "epoch": 0.5807644882860666, + "grad_norm": 2.1457149698206526, + "learning_rate": 1.8505843478078944e-05, + "loss": 1.1138, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 471 + }, + { + "epoch": 0.5819975339087546, + "grad_norm": 1.8879825929112735, + "learning_rate": 1.8498903772975056e-05, + "loss": 1.0181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 472 + }, + { + "epoch": 0.5832305795314426, + "grad_norm": 2.083069072716887, + "learning_rate": 1.8491949296714312e-05, + "loss": 1.0307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 473 + }, + { + "epoch": 0.5844636251541308, + "grad_norm": 2.135777387695568, + "learning_rate": 1.848498006138365e-05, + "loss": 1.0346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 474 + }, + { + "epoch": 0.5856966707768188, + "grad_norm": 2.092600182986245, + "learning_rate": 1.8477996079095643e-05, + "loss": 1.041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 475 + }, + { + "epoch": 0.5869297163995068, + "grad_norm": 2.1246789769186414, + "learning_rate": 1.8470997361988514e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 476 + }, + { + "epoch": 0.5881627620221949, + "grad_norm": 2.062971539274987, + "learning_rate": 1.8463983922226076e-05, + "loss": 1.111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 477 + }, + { + "epoch": 0.5893958076448829, + "grad_norm": 2.092774168134711, + "learning_rate": 1.8456955771997744e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 478 + }, + { + "epoch": 0.5906288532675709, + "grad_norm": 2.23500231174451, + "learning_rate": 1.844991292351849e-05, + "loss": 1.1008, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 479 + }, + { + "epoch": 0.591861898890259, + "grad_norm": 2.04828338908309, + "learning_rate": 1.844285538902884e-05, + "loss": 1.0183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 480 + }, + { + "epoch": 0.593094944512947, + "grad_norm": 2.1680070152194064, + "learning_rate": 1.8435783180794836e-05, + "loss": 1.092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 481 + }, + { + "epoch": 0.594327990135635, + "grad_norm": 2.0706136769588603, + "learning_rate": 1.8428696311108033e-05, + "loss": 1.1438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 482 + }, + { + "epoch": 0.5955610357583231, + "grad_norm": 1.9925622155649279, + "learning_rate": 1.842159479228546e-05, + "loss": 1.0359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 483 + }, + { + "epoch": 0.5967940813810111, + "grad_norm": 1.9425841064456546, + "learning_rate": 1.841447863666961e-05, + "loss": 1.2084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 484 + }, + { + "epoch": 0.5980271270036991, + "grad_norm": 1.929626615569851, + "learning_rate": 1.8407347856628405e-05, + "loss": 1.0615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 485 + }, + { + "epoch": 0.5992601726263872, + "grad_norm": 2.049215260436237, + "learning_rate": 1.8400202464555202e-05, + "loss": 1.0429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 486 + }, + { + "epoch": 0.6004932182490752, + "grad_norm": 2.0464723350262624, + "learning_rate": 1.8393042472868746e-05, + "loss": 1.1504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 487 + }, + { + "epoch": 0.6017262638717632, + "grad_norm": 1.9060167485555373, + "learning_rate": 1.8385867894013144e-05, + "loss": 0.9809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 488 + }, + { + "epoch": 0.6029593094944513, + "grad_norm": 1.9897115419719036, + "learning_rate": 1.837867874045788e-05, + "loss": 1.0573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 489 + }, + { + "epoch": 0.6041923551171393, + "grad_norm": 2.241807622458982, + "learning_rate": 1.8371475024697746e-05, + "loss": 1.1149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 490 + }, + { + "epoch": 0.6054254007398274, + "grad_norm": 2.141493902645165, + "learning_rate": 1.8364256759252855e-05, + "loss": 1.0396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 491 + }, + { + "epoch": 0.6066584463625154, + "grad_norm": 2.168929995419785, + "learning_rate": 1.8357023956668606e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 492 + }, + { + "epoch": 0.6078914919852034, + "grad_norm": 2.095414950529066, + "learning_rate": 1.834977662951566e-05, + "loss": 1.0313, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 493 + }, + { + "epoch": 0.6091245376078915, + "grad_norm": 2.2287918098101516, + "learning_rate": 1.834251479038993e-05, + "loss": 1.0118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 494 + }, + { + "epoch": 0.6103575832305795, + "grad_norm": 2.0458360619004154, + "learning_rate": 1.833523845191254e-05, + "loss": 1.0644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 495 + }, + { + "epoch": 0.6115906288532675, + "grad_norm": 2.1216602719929245, + "learning_rate": 1.8327947626729825e-05, + "loss": 1.1734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 496 + }, + { + "epoch": 0.6128236744759556, + "grad_norm": 2.1445297291236947, + "learning_rate": 1.8320642327513285e-05, + "loss": 0.9661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 497 + }, + { + "epoch": 0.6140567200986436, + "grad_norm": 2.061159959033469, + "learning_rate": 1.831332256695959e-05, + "loss": 1.189, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 498 + }, + { + "epoch": 0.6152897657213316, + "grad_norm": 1.966885029141375, + "learning_rate": 1.8305988357790533e-05, + "loss": 1.0711, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 499 + }, + { + "epoch": 0.6165228113440198, + "grad_norm": 2.127061899116888, + "learning_rate": 1.8298639712753025e-05, + "loss": 0.9351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 500 + }, + { + "epoch": 0.6177558569667078, + "grad_norm": 2.2239600877101835, + "learning_rate": 1.8291276644619066e-05, + "loss": 1.0569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 501 + }, + { + "epoch": 0.6189889025893958, + "grad_norm": 2.0281028990006913, + "learning_rate": 1.8283899166185723e-05, + "loss": 1.0698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 502 + }, + { + "epoch": 0.6202219482120839, + "grad_norm": 2.1224950625771677, + "learning_rate": 1.8276507290275104e-05, + "loss": 0.9905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 503 + }, + { + "epoch": 0.6214549938347719, + "grad_norm": 2.0799215178987, + "learning_rate": 1.8269101029734343e-05, + "loss": 1.0631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 504 + }, + { + "epoch": 0.6226880394574599, + "grad_norm": 2.2495223027498388, + "learning_rate": 1.8261680397435577e-05, + "loss": 0.9869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 505 + }, + { + "epoch": 0.623921085080148, + "grad_norm": 2.036885073926771, + "learning_rate": 1.825424540627592e-05, + "loss": 1.0939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 506 + }, + { + "epoch": 0.625154130702836, + "grad_norm": 2.0423951754968064, + "learning_rate": 1.824679606917744e-05, + "loss": 0.9793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 507 + }, + { + "epoch": 0.6263871763255241, + "grad_norm": 2.118412450801216, + "learning_rate": 1.8239332399087136e-05, + "loss": 1.0214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 508 + }, + { + "epoch": 0.6276202219482121, + "grad_norm": 2.273320358602361, + "learning_rate": 1.8231854408976924e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 509 + }, + { + "epoch": 0.6288532675709001, + "grad_norm": 1.9352848815012758, + "learning_rate": 1.822436211184361e-05, + "loss": 1.0194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 510 + }, + { + "epoch": 0.6300863131935882, + "grad_norm": 2.1086688743882713, + "learning_rate": 1.8216855520708854e-05, + "loss": 0.9404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 511 + }, + { + "epoch": 0.6313193588162762, + "grad_norm": 2.268255476351323, + "learning_rate": 1.820933464861917e-05, + "loss": 0.9734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 512 + }, + { + "epoch": 0.6325524044389642, + "grad_norm": 2.3533715256185705, + "learning_rate": 1.8201799508645887e-05, + "loss": 1.1794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 513 + }, + { + "epoch": 0.6337854500616523, + "grad_norm": 2.005798886598211, + "learning_rate": 1.8194250113885134e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 514 + }, + { + "epoch": 0.6350184956843403, + "grad_norm": 1.9276850525809597, + "learning_rate": 1.8186686477457817e-05, + "loss": 1.0122, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 515 + }, + { + "epoch": 0.6362515413070283, + "grad_norm": 2.089756823449061, + "learning_rate": 1.8179108612509587e-05, + "loss": 1.1076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 516 + }, + { + "epoch": 0.6374845869297164, + "grad_norm": 2.047489618950505, + "learning_rate": 1.8171516532210825e-05, + "loss": 1.1261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 517 + }, + { + "epoch": 0.6387176325524044, + "grad_norm": 1.9819427784604287, + "learning_rate": 1.8163910249756634e-05, + "loss": 0.9829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 518 + }, + { + "epoch": 0.6399506781750924, + "grad_norm": 2.1061977657317574, + "learning_rate": 1.815628977836678e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 519 + }, + { + "epoch": 0.6411837237977805, + "grad_norm": 2.0221352739914624, + "learning_rate": 1.8148655131285697e-05, + "loss": 1.0413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 520 + }, + { + "epoch": 0.6424167694204685, + "grad_norm": 2.0126443495430033, + "learning_rate": 1.8141006321782464e-05, + "loss": 1.0737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 521 + }, + { + "epoch": 0.6436498150431565, + "grad_norm": 1.9687053123524947, + "learning_rate": 1.8133343363150762e-05, + "loss": 1.0534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 522 + }, + { + "epoch": 0.6448828606658447, + "grad_norm": 2.1885490221346036, + "learning_rate": 1.812566626870887e-05, + "loss": 1.0746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 523 + }, + { + "epoch": 0.6461159062885327, + "grad_norm": 2.279950249415201, + "learning_rate": 1.8117975051799636e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 524 + }, + { + "epoch": 0.6473489519112207, + "grad_norm": 2.2376288751244986, + "learning_rate": 1.811026972579045e-05, + "loss": 1.0655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 525 + }, + { + "epoch": 0.6485819975339088, + "grad_norm": 2.0687763759549704, + "learning_rate": 1.810255030407323e-05, + "loss": 0.9515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 526 + }, + { + "epoch": 0.6498150431565968, + "grad_norm": 1.9230312514706345, + "learning_rate": 1.8094816800064376e-05, + "loss": 1.0144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 527 + }, + { + "epoch": 0.6510480887792849, + "grad_norm": 1.9036609363684858, + "learning_rate": 1.8087069227204784e-05, + "loss": 1.1092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 528 + }, + { + "epoch": 0.6522811344019729, + "grad_norm": 2.0395351493662104, + "learning_rate": 1.807930759895979e-05, + "loss": 1.0345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 529 + }, + { + "epoch": 0.6535141800246609, + "grad_norm": 2.011352927400422, + "learning_rate": 1.8071531928819158e-05, + "loss": 0.9913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 530 + }, + { + "epoch": 0.654747225647349, + "grad_norm": 1.9222212916878523, + "learning_rate": 1.8063742230297063e-05, + "loss": 1.0643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 531 + }, + { + "epoch": 0.655980271270037, + "grad_norm": 2.203621374645629, + "learning_rate": 1.8055938516932056e-05, + "loss": 1.0912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 532 + }, + { + "epoch": 0.657213316892725, + "grad_norm": 1.9492742413222397, + "learning_rate": 1.804812080228705e-05, + "loss": 0.9613, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 533 + }, + { + "epoch": 0.6584463625154131, + "grad_norm": 2.1694867453328683, + "learning_rate": 1.8040289099949285e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 534 + }, + { + "epoch": 0.6596794081381011, + "grad_norm": 2.0272612869084474, + "learning_rate": 1.8032443423530322e-05, + "loss": 1.0473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 535 + }, + { + "epoch": 0.6609124537607891, + "grad_norm": 2.059356328787384, + "learning_rate": 1.8024583786666007e-05, + "loss": 1.1633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 536 + }, + { + "epoch": 0.6621454993834772, + "grad_norm": 1.9976930500073966, + "learning_rate": 1.8016710203016434e-05, + "loss": 1.0907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 537 + }, + { + "epoch": 0.6633785450061652, + "grad_norm": 2.2046172414513747, + "learning_rate": 1.8008822686265958e-05, + "loss": 1.0435, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 538 + }, + { + "epoch": 0.6646115906288532, + "grad_norm": 1.9172442142747625, + "learning_rate": 1.8000921250123144e-05, + "loss": 1.0905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 539 + }, + { + "epoch": 0.6658446362515413, + "grad_norm": 1.9118211475087676, + "learning_rate": 1.7993005908320737e-05, + "loss": 0.9362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 540 + }, + { + "epoch": 0.6670776818742293, + "grad_norm": 2.2793615054405096, + "learning_rate": 1.7985076674615663e-05, + "loss": 1.1407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 541 + }, + { + "epoch": 0.6683107274969173, + "grad_norm": 2.0968780040667636, + "learning_rate": 1.797713356278899e-05, + "loss": 1.0263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 542 + }, + { + "epoch": 0.6695437731196054, + "grad_norm": 2.130295025454763, + "learning_rate": 1.79691765866459e-05, + "loss": 0.99, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 543 + }, + { + "epoch": 0.6707768187422934, + "grad_norm": 2.050843680123929, + "learning_rate": 1.7961205760015678e-05, + "loss": 0.937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 544 + }, + { + "epoch": 0.6720098643649816, + "grad_norm": 2.00556261298427, + "learning_rate": 1.7953221096751678e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 545 + }, + { + "epoch": 0.6732429099876696, + "grad_norm": 2.104573565750377, + "learning_rate": 1.7945222610731306e-05, + "loss": 0.9851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 546 + }, + { + "epoch": 0.6744759556103576, + "grad_norm": 2.009986716494675, + "learning_rate": 1.793721031585599e-05, + "loss": 0.9842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 547 + }, + { + "epoch": 0.6757090012330457, + "grad_norm": 2.053047844057653, + "learning_rate": 1.792918422605115e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 548 + }, + { + "epoch": 0.6769420468557337, + "grad_norm": 2.0361666532452714, + "learning_rate": 1.7921144355266195e-05, + "loss": 1.0002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 549 + }, + { + "epoch": 0.6781750924784217, + "grad_norm": 2.02324182186846, + "learning_rate": 1.7913090717474476e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 550 + }, + { + "epoch": 0.6794081381011098, + "grad_norm": 1.9162969465242339, + "learning_rate": 1.7905023326673277e-05, + "loss": 0.9256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 551 + }, + { + "epoch": 0.6806411837237978, + "grad_norm": 2.055814853732051, + "learning_rate": 1.789694219688378e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 552 + }, + { + "epoch": 0.6818742293464858, + "grad_norm": 2.0596956840908276, + "learning_rate": 1.7888847342151047e-05, + "loss": 1.096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 553 + }, + { + "epoch": 0.6831072749691739, + "grad_norm": 1.9518592781827393, + "learning_rate": 1.7880738776543994e-05, + "loss": 0.9907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 554 + }, + { + "epoch": 0.6843403205918619, + "grad_norm": 2.0542440716859485, + "learning_rate": 1.787261651415537e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 555 + }, + { + "epoch": 0.6855733662145499, + "grad_norm": 1.9676290362610278, + "learning_rate": 1.786448056910173e-05, + "loss": 0.96, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 556 + }, + { + "epoch": 0.686806411837238, + "grad_norm": 2.0236847394224418, + "learning_rate": 1.7856330955523396e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 557 + }, + { + "epoch": 0.688039457459926, + "grad_norm": 2.166651989793048, + "learning_rate": 1.7848167687584466e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 558 + }, + { + "epoch": 0.689272503082614, + "grad_norm": 2.037401342146827, + "learning_rate": 1.783999077947275e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 559 + }, + { + "epoch": 0.6905055487053021, + "grad_norm": 2.0931620663370465, + "learning_rate": 1.783180024539978e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 560 + }, + { + "epoch": 0.6917385943279901, + "grad_norm": 2.5512794237438032, + "learning_rate": 1.7823596099600767e-05, + "loss": 1.056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 561 + }, + { + "epoch": 0.6929716399506781, + "grad_norm": 2.2145616452367305, + "learning_rate": 1.781537835633457e-05, + "loss": 1.0677, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 562 + }, + { + "epoch": 0.6942046855733662, + "grad_norm": 2.304945412618975, + "learning_rate": 1.7807147029883688e-05, + "loss": 1.0025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 563 + }, + { + "epoch": 0.6954377311960542, + "grad_norm": 2.2714171401510272, + "learning_rate": 1.7798902134554225e-05, + "loss": 0.927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 564 + }, + { + "epoch": 0.6966707768187423, + "grad_norm": 1.9934114654014257, + "learning_rate": 1.7790643684675874e-05, + "loss": 0.9226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 565 + }, + { + "epoch": 0.6979038224414303, + "grad_norm": 2.4001795539004425, + "learning_rate": 1.7782371694601882e-05, + "loss": 0.9898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 566 + }, + { + "epoch": 0.6991368680641183, + "grad_norm": 2.1543466787664944, + "learning_rate": 1.777408617870902e-05, + "loss": 0.8765, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 567 + }, + { + "epoch": 0.7003699136868065, + "grad_norm": 2.125530896627932, + "learning_rate": 1.7765787151397584e-05, + "loss": 0.9383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 568 + }, + { + "epoch": 0.7016029593094945, + "grad_norm": 2.399579117157628, + "learning_rate": 1.775747462709134e-05, + "loss": 1.116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 569 + }, + { + "epoch": 0.7028360049321825, + "grad_norm": 2.3911388414036843, + "learning_rate": 1.7749148620237515e-05, + "loss": 0.983, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 570 + }, + { + "epoch": 0.7040690505548706, + "grad_norm": 2.037432480432792, + "learning_rate": 1.7740809145306773e-05, + "loss": 1.0094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 571 + }, + { + "epoch": 0.7053020961775586, + "grad_norm": 2.0999588564507685, + "learning_rate": 1.773245621679318e-05, + "loss": 1.046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 572 + }, + { + "epoch": 0.7065351418002466, + "grad_norm": 1.9822662619027398, + "learning_rate": 1.7724089849214184e-05, + "loss": 1.0177, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 573 + }, + { + "epoch": 0.7077681874229347, + "grad_norm": 2.296463261709746, + "learning_rate": 1.77157100571106e-05, + "loss": 0.9207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 574 + }, + { + "epoch": 0.7090012330456227, + "grad_norm": 2.0434250630443045, + "learning_rate": 1.7707316855046563e-05, + "loss": 1.0088, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 575 + }, + { + "epoch": 0.7102342786683107, + "grad_norm": 2.067794231415559, + "learning_rate": 1.7698910257609524e-05, + "loss": 1.0033, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 576 + }, + { + "epoch": 0.7114673242909988, + "grad_norm": 1.9884025311346816, + "learning_rate": 1.7690490279410205e-05, + "loss": 1.0395, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 577 + }, + { + "epoch": 0.7127003699136868, + "grad_norm": 2.0663488956854676, + "learning_rate": 1.768205693508259e-05, + "loss": 1.0609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 578 + }, + { + "epoch": 0.7139334155363748, + "grad_norm": 1.9661847340747427, + "learning_rate": 1.7673610239283898e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 579 + }, + { + "epoch": 0.7151664611590629, + "grad_norm": 2.4048347386453135, + "learning_rate": 1.7665150206694543e-05, + "loss": 1.0294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 580 + }, + { + "epoch": 0.7163995067817509, + "grad_norm": 2.049926176641203, + "learning_rate": 1.765667685201813e-05, + "loss": 0.9953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 581 + }, + { + "epoch": 0.717632552404439, + "grad_norm": 2.0425901676211398, + "learning_rate": 1.7648190189981403e-05, + "loss": 1.0155, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 582 + }, + { + "epoch": 0.718865598027127, + "grad_norm": 2.111319944982802, + "learning_rate": 1.7639690235334246e-05, + "loss": 1.012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 583 + }, + { + "epoch": 0.720098643649815, + "grad_norm": 1.9625598394287143, + "learning_rate": 1.7631177002849648e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 584 + }, + { + "epoch": 0.7213316892725031, + "grad_norm": 2.0737975490297025, + "learning_rate": 1.7622650507323658e-05, + "loss": 0.9954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 585 + }, + { + "epoch": 0.7225647348951911, + "grad_norm": 2.2220755560999175, + "learning_rate": 1.7614110763575393e-05, + "loss": 1.093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 586 + }, + { + "epoch": 0.7237977805178791, + "grad_norm": 2.10369867254611, + "learning_rate": 1.7605557786446997e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 587 + }, + { + "epoch": 0.7250308261405672, + "grad_norm": 2.023199550080708, + "learning_rate": 1.7596991590803597e-05, + "loss": 1.0909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 588 + }, + { + "epoch": 0.7262638717632552, + "grad_norm": 2.1139350266949197, + "learning_rate": 1.7588412191533305e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 589 + }, + { + "epoch": 0.7274969173859432, + "grad_norm": 2.1032070896543926, + "learning_rate": 1.7579819603547187e-05, + "loss": 0.9305, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 590 + }, + { + "epoch": 0.7287299630086314, + "grad_norm": 2.1111448680444798, + "learning_rate": 1.7571213841779213e-05, + "loss": 1.0821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 591 + }, + { + "epoch": 0.7299630086313194, + "grad_norm": 2.0849758733816692, + "learning_rate": 1.7562594921186268e-05, + "loss": 0.9747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 592 + }, + { + "epoch": 0.7311960542540074, + "grad_norm": 2.7468709410129573, + "learning_rate": 1.7553962856748098e-05, + "loss": 1.0598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 593 + }, + { + "epoch": 0.7324290998766955, + "grad_norm": 2.297654831376881, + "learning_rate": 1.754531766346729e-05, + "loss": 1.0414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 594 + }, + { + "epoch": 0.7336621454993835, + "grad_norm": 2.0094458450808874, + "learning_rate": 1.7536659356369254e-05, + "loss": 0.977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 595 + }, + { + "epoch": 0.7348951911220715, + "grad_norm": 1.9967204791438438, + "learning_rate": 1.7527987950502196e-05, + "loss": 0.9274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 596 + }, + { + "epoch": 0.7361282367447596, + "grad_norm": 1.9493969416772718, + "learning_rate": 1.7519303460937073e-05, + "loss": 1.0292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 597 + }, + { + "epoch": 0.7373612823674476, + "grad_norm": 2.0023257821799247, + "learning_rate": 1.7510605902767602e-05, + "loss": 0.9574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 598 + }, + { + "epoch": 0.7385943279901356, + "grad_norm": 2.076672200643684, + "learning_rate": 1.7501895291110197e-05, + "loss": 0.9971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 599 + }, + { + "epoch": 0.7398273736128237, + "grad_norm": 2.0612367813173584, + "learning_rate": 1.7493171641103965e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 600 + }, + { + "epoch": 0.7410604192355117, + "grad_norm": 2.16071395185368, + "learning_rate": 1.7484434967910676e-05, + "loss": 1.0001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 601 + }, + { + "epoch": 0.7422934648581998, + "grad_norm": 2.0810754791890664, + "learning_rate": 1.747568528671473e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 602 + }, + { + "epoch": 0.7435265104808878, + "grad_norm": 2.0015256143479525, + "learning_rate": 1.7466922612723133e-05, + "loss": 0.9288, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 603 + }, + { + "epoch": 0.7447595561035758, + "grad_norm": 1.9574522371127243, + "learning_rate": 1.7458146961165484e-05, + "loss": 1.0365, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 604 + }, + { + "epoch": 0.7459926017262639, + "grad_norm": 1.999399085309589, + "learning_rate": 1.7449358347293925e-05, + "loss": 1.0457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 605 + }, + { + "epoch": 0.7472256473489519, + "grad_norm": 2.016778757176555, + "learning_rate": 1.7440556786383132e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 606 + }, + { + "epoch": 0.7484586929716399, + "grad_norm": 2.011184042360808, + "learning_rate": 1.743174229373028e-05, + "loss": 0.9568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 607 + }, + { + "epoch": 0.749691738594328, + "grad_norm": 1.9994468999326396, + "learning_rate": 1.7422914884655027e-05, + "loss": 0.9822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 608 + }, + { + "epoch": 0.750924784217016, + "grad_norm": 2.1373417665891163, + "learning_rate": 1.7414074574499465e-05, + "loss": 0.9128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 609 + }, + { + "epoch": 0.752157829839704, + "grad_norm": 2.414511732797334, + "learning_rate": 1.7405221378628133e-05, + "loss": 1.0866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 610 + }, + { + "epoch": 0.7533908754623921, + "grad_norm": 2.1820009401893605, + "learning_rate": 1.7396355312427935e-05, + "loss": 1.0538, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 611 + }, + { + "epoch": 0.7546239210850801, + "grad_norm": 2.0281507644839163, + "learning_rate": 1.7387476391308168e-05, + "loss": 0.9224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 612 + }, + { + "epoch": 0.7558569667077681, + "grad_norm": 2.387292229834452, + "learning_rate": 1.7378584630700454e-05, + "loss": 0.7561, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 613 + }, + { + "epoch": 0.7570900123304563, + "grad_norm": 2.0036252294718975, + "learning_rate": 1.7369680046058744e-05, + "loss": 0.9816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 614 + }, + { + "epoch": 0.7583230579531443, + "grad_norm": 2.1474265419679566, + "learning_rate": 1.7360762652859275e-05, + "loss": 0.9312, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 615 + }, + { + "epoch": 0.7595561035758323, + "grad_norm": 2.3913311043981387, + "learning_rate": 1.735183246660053e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 616 + }, + { + "epoch": 0.7607891491985204, + "grad_norm": 2.070836177310932, + "learning_rate": 1.7342889502803246e-05, + "loss": 0.8846, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 617 + }, + { + "epoch": 0.7620221948212084, + "grad_norm": 1.9909512311630846, + "learning_rate": 1.7333933777010357e-05, + "loss": 0.9683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 618 + }, + { + "epoch": 0.7632552404438965, + "grad_norm": 2.048373917294825, + "learning_rate": 1.732496530478698e-05, + "loss": 1.021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 619 + }, + { + "epoch": 0.7644882860665845, + "grad_norm": 2.046848859346405, + "learning_rate": 1.7315984101720388e-05, + "loss": 0.9949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 620 + }, + { + "epoch": 0.7657213316892725, + "grad_norm": 2.199270192761742, + "learning_rate": 1.7306990183419972e-05, + "loss": 0.9319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 621 + }, + { + "epoch": 0.7669543773119606, + "grad_norm": 2.2362149652218877, + "learning_rate": 1.729798356551723e-05, + "loss": 1.0579, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 622 + }, + { + "epoch": 0.7681874229346486, + "grad_norm": 2.072113656676678, + "learning_rate": 1.7288964263665735e-05, + "loss": 0.984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 623 + }, + { + "epoch": 0.7694204685573366, + "grad_norm": 2.2058183913371647, + "learning_rate": 1.7279932293541092e-05, + "loss": 1.0727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 624 + }, + { + "epoch": 0.7706535141800247, + "grad_norm": 2.1099223160270104, + "learning_rate": 1.7270887670840935e-05, + "loss": 0.852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 625 + }, + { + "epoch": 0.7718865598027127, + "grad_norm": 2.014275874425781, + "learning_rate": 1.7261830411284884e-05, + "loss": 1.0219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 626 + }, + { + "epoch": 0.7731196054254007, + "grad_norm": 2.1002718881715605, + "learning_rate": 1.7252760530614522e-05, + "loss": 0.9339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 627 + }, + { + "epoch": 0.7743526510480888, + "grad_norm": 2.530650182566485, + "learning_rate": 1.724367804459337e-05, + "loss": 1.1248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 628 + }, + { + "epoch": 0.7755856966707768, + "grad_norm": 2.072394234232429, + "learning_rate": 1.7234582969006847e-05, + "loss": 0.963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 629 + }, + { + "epoch": 0.7768187422934648, + "grad_norm": 2.077401892279152, + "learning_rate": 1.722547531966227e-05, + "loss": 0.9589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 630 + }, + { + "epoch": 0.7780517879161529, + "grad_norm": 2.0461500607118612, + "learning_rate": 1.7216355112388797e-05, + "loss": 0.9892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 631 + }, + { + "epoch": 0.7792848335388409, + "grad_norm": 1.9481531627295827, + "learning_rate": 1.7207222363037414e-05, + "loss": 0.9963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 632 + }, + { + "epoch": 0.7805178791615289, + "grad_norm": 2.0363659895001067, + "learning_rate": 1.7198077087480903e-05, + "loss": 0.9206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 633 + }, + { + "epoch": 0.781750924784217, + "grad_norm": 2.367528975566535, + "learning_rate": 1.718891930161383e-05, + "loss": 1.1255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 634 + }, + { + "epoch": 0.782983970406905, + "grad_norm": 2.007050047111477, + "learning_rate": 1.7179749021352482e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 635 + }, + { + "epoch": 0.7842170160295932, + "grad_norm": 1.9196848202462296, + "learning_rate": 1.7170566262634883e-05, + "loss": 0.9448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 636 + }, + { + "epoch": 0.7854500616522812, + "grad_norm": 2.145941740629715, + "learning_rate": 1.7161371041420723e-05, + "loss": 0.9428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 637 + }, + { + "epoch": 0.7866831072749692, + "grad_norm": 2.170240354212872, + "learning_rate": 1.7152163373691377e-05, + "loss": 1.0323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 638 + }, + { + "epoch": 0.7879161528976573, + "grad_norm": 2.0553179179576015, + "learning_rate": 1.714294327544983e-05, + "loss": 0.8661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 639 + }, + { + "epoch": 0.7891491985203453, + "grad_norm": 2.1113466348102397, + "learning_rate": 1.7133710762720683e-05, + "loss": 1.034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 640 + }, + { + "epoch": 0.7903822441430333, + "grad_norm": 2.146414650333374, + "learning_rate": 1.712446585155011e-05, + "loss": 1.0681, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 641 + }, + { + "epoch": 0.7916152897657214, + "grad_norm": 2.0440039273987227, + "learning_rate": 1.7115208558005834e-05, + "loss": 0.9085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 642 + }, + { + "epoch": 0.7928483353884094, + "grad_norm": 2.3653004518970535, + "learning_rate": 1.7105938898177097e-05, + "loss": 0.8514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 643 + }, + { + "epoch": 0.7940813810110974, + "grad_norm": 2.058806308472887, + "learning_rate": 1.709665688817464e-05, + "loss": 0.9258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 644 + }, + { + "epoch": 0.7953144266337855, + "grad_norm": 2.3406153888995798, + "learning_rate": 1.708736254413066e-05, + "loss": 0.9866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 645 + }, + { + "epoch": 0.7965474722564735, + "grad_norm": 3.2012831686915293, + "learning_rate": 1.70780558821988e-05, + "loss": 1.0916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 646 + }, + { + "epoch": 0.7977805178791615, + "grad_norm": 2.23457477201596, + "learning_rate": 1.7068736918554102e-05, + "loss": 0.9076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 647 + }, + { + "epoch": 0.7990135635018496, + "grad_norm": 2.255512398782474, + "learning_rate": 1.7059405669392993e-05, + "loss": 0.9113, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 648 + }, + { + "epoch": 0.8002466091245376, + "grad_norm": 2.114904513093256, + "learning_rate": 1.7050062150933252e-05, + "loss": 0.9587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 649 + }, + { + "epoch": 0.8014796547472256, + "grad_norm": 1.931643331052564, + "learning_rate": 1.7040706379413983e-05, + "loss": 0.9888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 650 + }, + { + "epoch": 0.8027127003699137, + "grad_norm": 2.0094212900142816, + "learning_rate": 1.7031338371095588e-05, + "loss": 0.9961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 651 + }, + { + "epoch": 0.8039457459926017, + "grad_norm": 2.1928733306625188, + "learning_rate": 1.702195814225973e-05, + "loss": 1.0158, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 652 + }, + { + "epoch": 0.8051787916152897, + "grad_norm": 2.2216734919656544, + "learning_rate": 1.7012565709209317e-05, + "loss": 1.0201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 653 + }, + { + "epoch": 0.8064118372379778, + "grad_norm": 2.0613969675724686, + "learning_rate": 1.700316108826847e-05, + "loss": 0.9941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 654 + }, + { + "epoch": 0.8076448828606658, + "grad_norm": 1.9372031821836655, + "learning_rate": 1.6993744295782482e-05, + "loss": 1.0026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 655 + }, + { + "epoch": 0.8088779284833539, + "grad_norm": 2.077194491232311, + "learning_rate": 1.698431534811781e-05, + "loss": 0.8491, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 656 + }, + { + "epoch": 0.8101109741060419, + "grad_norm": 2.1425857340105754, + "learning_rate": 1.6974874261662035e-05, + "loss": 0.9719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 657 + }, + { + "epoch": 0.8113440197287299, + "grad_norm": 1.8547512115782296, + "learning_rate": 1.6965421052823834e-05, + "loss": 1.0107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 658 + }, + { + "epoch": 0.812577065351418, + "grad_norm": 1.9098479812610942, + "learning_rate": 1.695595573803295e-05, + "loss": 0.95, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 659 + }, + { + "epoch": 0.813810110974106, + "grad_norm": 1.892767796833539, + "learning_rate": 1.6946478333740174e-05, + "loss": 0.86, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 660 + }, + { + "epoch": 0.815043156596794, + "grad_norm": 2.0228841723951043, + "learning_rate": 1.69369888564173e-05, + "loss": 0.9881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 661 + }, + { + "epoch": 0.8162762022194822, + "grad_norm": 2.144845399605076, + "learning_rate": 1.692748732255711e-05, + "loss": 0.8493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 662 + }, + { + "epoch": 0.8175092478421702, + "grad_norm": 2.1710135611234027, + "learning_rate": 1.6917973748673343e-05, + "loss": 0.9356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 663 + }, + { + "epoch": 0.8187422934648582, + "grad_norm": 2.200157070237925, + "learning_rate": 1.6908448151300654e-05, + "loss": 0.9922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 664 + }, + { + "epoch": 0.8199753390875463, + "grad_norm": 2.1046294162556385, + "learning_rate": 1.6898910546994605e-05, + "loss": 0.9714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 665 + }, + { + "epoch": 0.8212083847102343, + "grad_norm": 2.1921813927002622, + "learning_rate": 1.688936095233162e-05, + "loss": 1.0186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 666 + }, + { + "epoch": 0.8224414303329223, + "grad_norm": 2.5290727417605767, + "learning_rate": 1.6879799383908967e-05, + "loss": 0.9147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 667 + }, + { + "epoch": 0.8236744759556104, + "grad_norm": 2.0992091012861605, + "learning_rate": 1.6870225858344715e-05, + "loss": 1.0331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 668 + }, + { + "epoch": 0.8249075215782984, + "grad_norm": 2.228115711337027, + "learning_rate": 1.6860640392277728e-05, + "loss": 1.0372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 669 + }, + { + "epoch": 0.8261405672009864, + "grad_norm": 2.104862934581671, + "learning_rate": 1.6851043002367612e-05, + "loss": 0.9556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 670 + }, + { + "epoch": 0.8273736128236745, + "grad_norm": 2.1153449603435575, + "learning_rate": 1.6841433705294704e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 671 + }, + { + "epoch": 0.8286066584463625, + "grad_norm": 1.9113210816564241, + "learning_rate": 1.6831812517760036e-05, + "loss": 0.9553, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 672 + }, + { + "epoch": 0.8298397040690506, + "grad_norm": 2.146278639653085, + "learning_rate": 1.6822179456485292e-05, + "loss": 0.8621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 673 + }, + { + "epoch": 0.8310727496917386, + "grad_norm": 2.092521119953653, + "learning_rate": 1.681253453821281e-05, + "loss": 0.9576, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 674 + }, + { + "epoch": 0.8323057953144266, + "grad_norm": 2.108757495841555, + "learning_rate": 1.6802877779705526e-05, + "loss": 0.9349, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 675 + }, + { + "epoch": 0.8335388409371147, + "grad_norm": 2.165541838354097, + "learning_rate": 1.679320919774696e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 676 + }, + { + "epoch": 0.8347718865598027, + "grad_norm": 2.0901280484241322, + "learning_rate": 1.6783528809141175e-05, + "loss": 0.9577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 677 + }, + { + "epoch": 0.8360049321824907, + "grad_norm": 2.049770632568499, + "learning_rate": 1.6773836630712757e-05, + "loss": 0.8912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 678 + }, + { + "epoch": 0.8372379778051788, + "grad_norm": 1.9415439366071066, + "learning_rate": 1.6764132679306785e-05, + "loss": 0.9678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 679 + }, + { + "epoch": 0.8384710234278668, + "grad_norm": 2.159517279560615, + "learning_rate": 1.6754416971788795e-05, + "loss": 0.9739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 680 + }, + { + "epoch": 0.8397040690505548, + "grad_norm": 2.0578899051147745, + "learning_rate": 1.674468952504476e-05, + "loss": 0.9187, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 681 + }, + { + "epoch": 0.840937114673243, + "grad_norm": 2.1905094099573135, + "learning_rate": 1.673495035598105e-05, + "loss": 0.9753, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 682 + }, + { + "epoch": 0.842170160295931, + "grad_norm": 2.194631531971183, + "learning_rate": 1.6725199481524418e-05, + "loss": 1.0267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 683 + }, + { + "epoch": 0.843403205918619, + "grad_norm": 2.040276435918069, + "learning_rate": 1.671543691862195e-05, + "loss": 0.9703, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 684 + }, + { + "epoch": 0.8446362515413071, + "grad_norm": 1.9627382699239366, + "learning_rate": 1.6705662684241053e-05, + "loss": 0.8955, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 685 + }, + { + "epoch": 0.8458692971639951, + "grad_norm": 2.288828851032713, + "learning_rate": 1.6695876795369418e-05, + "loss": 0.946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 686 + }, + { + "epoch": 0.8471023427866831, + "grad_norm": 2.1152782339808884, + "learning_rate": 1.668607926901499e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 687 + }, + { + "epoch": 0.8483353884093712, + "grad_norm": 2.798938790414006, + "learning_rate": 1.6676270122205946e-05, + "loss": 0.8988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 688 + }, + { + "epoch": 0.8495684340320592, + "grad_norm": 2.5307103300223766, + "learning_rate": 1.6666449371990644e-05, + "loss": 0.8814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 689 + }, + { + "epoch": 0.8508014796547472, + "grad_norm": 2.2986978033861294, + "learning_rate": 1.6656617035437635e-05, + "loss": 0.9402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 690 + }, + { + "epoch": 0.8520345252774353, + "grad_norm": 2.1645514508732546, + "learning_rate": 1.664677312963558e-05, + "loss": 0.9551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 691 + }, + { + "epoch": 0.8532675709001233, + "grad_norm": 2.144781636504613, + "learning_rate": 1.6636917671693264e-05, + "loss": 1.0505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 692 + }, + { + "epoch": 0.8545006165228114, + "grad_norm": 2.068308383986026, + "learning_rate": 1.6627050678739544e-05, + "loss": 0.9237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 693 + }, + { + "epoch": 0.8557336621454994, + "grad_norm": 2.4201265499871325, + "learning_rate": 1.6617172167923327e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 694 + }, + { + "epoch": 0.8569667077681874, + "grad_norm": 1.9743124171175193, + "learning_rate": 1.660728215641353e-05, + "loss": 0.9649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 695 + }, + { + "epoch": 0.8581997533908755, + "grad_norm": 2.4976283410874847, + "learning_rate": 1.6597380661399077e-05, + "loss": 0.9445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 696 + }, + { + "epoch": 0.8594327990135635, + "grad_norm": 2.2344847337230855, + "learning_rate": 1.6587467700088834e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 697 + }, + { + "epoch": 0.8606658446362515, + "grad_norm": 2.1366592393550343, + "learning_rate": 1.6577543289711597e-05, + "loss": 0.8845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 698 + }, + { + "epoch": 0.8618988902589396, + "grad_norm": 2.0879117181580753, + "learning_rate": 1.6567607447516064e-05, + "loss": 0.7972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 699 + }, + { + "epoch": 0.8631319358816276, + "grad_norm": 2.117890235500837, + "learning_rate": 1.6557660190770803e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 700 + }, + { + "epoch": 0.8643649815043156, + "grad_norm": 2.227804015025162, + "learning_rate": 1.654770153676422e-05, + "loss": 0.9105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 701 + }, + { + "epoch": 0.8655980271270037, + "grad_norm": 2.262676552209609, + "learning_rate": 1.6537731502804525e-05, + "loss": 0.9706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 702 + }, + { + "epoch": 0.8668310727496917, + "grad_norm": 2.146446661141531, + "learning_rate": 1.6527750106219715e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 703 + }, + { + "epoch": 0.8680641183723797, + "grad_norm": 2.08167189099969, + "learning_rate": 1.6517757364357526e-05, + "loss": 0.9381, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 704 + }, + { + "epoch": 0.8692971639950678, + "grad_norm": 1.9310579355796305, + "learning_rate": 1.6507753294585417e-05, + "loss": 0.9047, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 705 + }, + { + "epoch": 0.8705302096177558, + "grad_norm": 2.0325358606748978, + "learning_rate": 1.6497737914290535e-05, + "loss": 0.9094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 706 + }, + { + "epoch": 0.8717632552404438, + "grad_norm": 2.0543982954124966, + "learning_rate": 1.648771124087969e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 707 + }, + { + "epoch": 0.872996300863132, + "grad_norm": 2.295806181212224, + "learning_rate": 1.64776732917793e-05, + "loss": 0.9444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 708 + }, + { + "epoch": 0.87422934648582, + "grad_norm": 2.1899960866925303, + "learning_rate": 1.646762408443541e-05, + "loss": 1.0083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 709 + }, + { + "epoch": 0.8754623921085081, + "grad_norm": 2.040526145894979, + "learning_rate": 1.6457563636313598e-05, + "loss": 0.8215, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 710 + }, + { + "epoch": 0.8766954377311961, + "grad_norm": 1.9967218569140772, + "learning_rate": 1.6447491964899008e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 711 + }, + { + "epoch": 0.8779284833538841, + "grad_norm": 2.131774886953309, + "learning_rate": 1.6437409087696283e-05, + "loss": 0.8786, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 712 + }, + { + "epoch": 0.8791615289765722, + "grad_norm": 2.0642410512824414, + "learning_rate": 1.6427315022229525e-05, + "loss": 0.8831, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 713 + }, + { + "epoch": 0.8803945745992602, + "grad_norm": 2.2793436000912255, + "learning_rate": 1.64172097860423e-05, + "loss": 0.8413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 714 + }, + { + "epoch": 0.8816276202219482, + "grad_norm": 1.9501864497105927, + "learning_rate": 1.640709339669758e-05, + "loss": 0.8856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 715 + }, + { + "epoch": 0.8828606658446363, + "grad_norm": 2.0279122102292977, + "learning_rate": 1.6396965871777727e-05, + "loss": 0.8835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 716 + }, + { + "epoch": 0.8840937114673243, + "grad_norm": 2.181455569803073, + "learning_rate": 1.638682722888445e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 717 + }, + { + "epoch": 0.8853267570900123, + "grad_norm": 2.1877848027614313, + "learning_rate": 1.637667748563879e-05, + "loss": 0.8314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 718 + }, + { + "epoch": 0.8865598027127004, + "grad_norm": 2.0901252578597713, + "learning_rate": 1.6366516659681067e-05, + "loss": 0.9333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 719 + }, + { + "epoch": 0.8877928483353884, + "grad_norm": 2.1846108211938673, + "learning_rate": 1.6356344768670875e-05, + "loss": 0.9309, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 720 + }, + { + "epoch": 0.8890258939580764, + "grad_norm": 2.600109718803851, + "learning_rate": 1.6346161830287034e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 721 + }, + { + "epoch": 0.8902589395807645, + "grad_norm": 2.2057915960500276, + "learning_rate": 1.6335967862227567e-05, + "loss": 0.9764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 722 + }, + { + "epoch": 0.8914919852034525, + "grad_norm": 2.1009909758008267, + "learning_rate": 1.632576288220966e-05, + "loss": 0.8741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 723 + }, + { + "epoch": 0.8927250308261405, + "grad_norm": 2.1617838023881, + "learning_rate": 1.6315546907969652e-05, + "loss": 0.9573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 724 + }, + { + "epoch": 0.8939580764488286, + "grad_norm": 2.091522410165622, + "learning_rate": 1.6305319957262972e-05, + "loss": 0.9534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 725 + }, + { + "epoch": 0.8951911220715166, + "grad_norm": 2.2845231322682418, + "learning_rate": 1.6295082047864138e-05, + "loss": 0.9156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 726 + }, + { + "epoch": 0.8964241676942046, + "grad_norm": 2.062614641816067, + "learning_rate": 1.6284833197566712e-05, + "loss": 0.9539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 727 + }, + { + "epoch": 0.8976572133168927, + "grad_norm": 2.157165674006361, + "learning_rate": 1.627457342418327e-05, + "loss": 0.9926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 728 + }, + { + "epoch": 0.8988902589395807, + "grad_norm": 2.039737473184125, + "learning_rate": 1.6264302745545374e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 729 + }, + { + "epoch": 0.9001233045622689, + "grad_norm": 2.057800120641984, + "learning_rate": 1.625402117950354e-05, + "loss": 0.9139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 730 + }, + { + "epoch": 0.9013563501849569, + "grad_norm": 2.240321306869684, + "learning_rate": 1.62437287439272e-05, + "loss": 1.0758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 731 + }, + { + "epoch": 0.9025893958076449, + "grad_norm": 2.0947466285552436, + "learning_rate": 1.6233425456704686e-05, + "loss": 0.8467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 732 + }, + { + "epoch": 0.903822441430333, + "grad_norm": 2.061121182714553, + "learning_rate": 1.6223111335743187e-05, + "loss": 0.9093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 733 + }, + { + "epoch": 0.905055487053021, + "grad_norm": 2.209914581799182, + "learning_rate": 1.6212786398968717e-05, + "loss": 0.8805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 734 + }, + { + "epoch": 0.906288532675709, + "grad_norm": 2.468473842682747, + "learning_rate": 1.6202450664326093e-05, + "loss": 0.9013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 735 + }, + { + "epoch": 0.9075215782983971, + "grad_norm": 2.3365951628381056, + "learning_rate": 1.61921041497789e-05, + "loss": 0.9112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 736 + }, + { + "epoch": 0.9087546239210851, + "grad_norm": 2.474277588242769, + "learning_rate": 1.6181746873309455e-05, + "loss": 0.9186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 737 + }, + { + "epoch": 0.9099876695437731, + "grad_norm": 2.0716027082353228, + "learning_rate": 1.6171378852918778e-05, + "loss": 0.9486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 738 + }, + { + "epoch": 0.9112207151664612, + "grad_norm": 2.165213004434134, + "learning_rate": 1.6161000106626564e-05, + "loss": 0.9726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 739 + }, + { + "epoch": 0.9124537607891492, + "grad_norm": 2.4218033902151475, + "learning_rate": 1.615061065247115e-05, + "loss": 0.8259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 740 + }, + { + "epoch": 0.9136868064118372, + "grad_norm": 1.970928985580259, + "learning_rate": 1.614021050850949e-05, + "loss": 0.9788, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 741 + }, + { + "epoch": 0.9149198520345253, + "grad_norm": 2.2639971055328934, + "learning_rate": 1.6129799692817097e-05, + "loss": 0.8745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 742 + }, + { + "epoch": 0.9161528976572133, + "grad_norm": 2.5602631751385028, + "learning_rate": 1.611937822348806e-05, + "loss": 0.9512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 743 + }, + { + "epoch": 0.9173859432799013, + "grad_norm": 2.22841865866248, + "learning_rate": 1.6108946118634957e-05, + "loss": 0.805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 744 + }, + { + "epoch": 0.9186189889025894, + "grad_norm": 2.4182686550221626, + "learning_rate": 1.6098503396388868e-05, + "loss": 1.0202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 745 + }, + { + "epoch": 0.9198520345252774, + "grad_norm": 2.2298122692949645, + "learning_rate": 1.608805007489932e-05, + "loss": 0.8972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 746 + }, + { + "epoch": 0.9210850801479655, + "grad_norm": 2.362160156268979, + "learning_rate": 1.607758617233426e-05, + "loss": 0.9039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 747 + }, + { + "epoch": 0.9223181257706535, + "grad_norm": 2.1106519602619533, + "learning_rate": 1.6067111706880027e-05, + "loss": 0.8318, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 748 + }, + { + "epoch": 0.9235511713933415, + "grad_norm": 2.0825477479243424, + "learning_rate": 1.605662669674132e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 749 + }, + { + "epoch": 0.9247842170160296, + "grad_norm": 2.036239449814601, + "learning_rate": 1.6046131160141167e-05, + "loss": 0.8727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 750 + }, + { + "epoch": 0.9260172626387176, + "grad_norm": 2.1703406452205787, + "learning_rate": 1.603562511532088e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 751 + }, + { + "epoch": 0.9272503082614056, + "grad_norm": 2.1398345899160143, + "learning_rate": 1.6025108580540046e-05, + "loss": 0.8691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 752 + }, + { + "epoch": 0.9284833538840938, + "grad_norm": 2.0888823283975375, + "learning_rate": 1.601458157407648e-05, + "loss": 0.8547, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 753 + }, + { + "epoch": 0.9297163995067818, + "grad_norm": 2.269523743217552, + "learning_rate": 1.6004044114226193e-05, + "loss": 1.0141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 754 + }, + { + "epoch": 0.9309494451294698, + "grad_norm": 2.0091635625820063, + "learning_rate": 1.5993496219303367e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 755 + }, + { + "epoch": 0.9321824907521579, + "grad_norm": 2.1386095902203612, + "learning_rate": 1.5982937907640327e-05, + "loss": 0.9388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 756 + }, + { + "epoch": 0.9334155363748459, + "grad_norm": 2.1260034823775467, + "learning_rate": 1.5972369197587486e-05, + "loss": 0.9153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 757 + }, + { + "epoch": 0.9346485819975339, + "grad_norm": 2.1225283234506156, + "learning_rate": 1.5961790107513343e-05, + "loss": 1.0178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 758 + }, + { + "epoch": 0.935881627620222, + "grad_norm": 2.0422192666840795, + "learning_rate": 1.595120065580444e-05, + "loss": 0.9856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 759 + }, + { + "epoch": 0.93711467324291, + "grad_norm": 2.051963293225036, + "learning_rate": 1.5940600860865315e-05, + "loss": 0.8559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 760 + }, + { + "epoch": 0.938347718865598, + "grad_norm": 2.127612803214104, + "learning_rate": 1.592999074111849e-05, + "loss": 0.8298, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 761 + }, + { + "epoch": 0.9395807644882861, + "grad_norm": 2.286554941831041, + "learning_rate": 1.591937031500443e-05, + "loss": 0.8796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 762 + }, + { + "epoch": 0.9408138101109741, + "grad_norm": 2.282412038316439, + "learning_rate": 1.5908739600981517e-05, + "loss": 1.022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 763 + }, + { + "epoch": 0.9420468557336621, + "grad_norm": 2.197857160398704, + "learning_rate": 1.5898098617526006e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 764 + }, + { + "epoch": 0.9432799013563502, + "grad_norm": 2.1702846778760607, + "learning_rate": 1.5887447383132008e-05, + "loss": 0.9507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 765 + }, + { + "epoch": 0.9445129469790382, + "grad_norm": 2.045700294662405, + "learning_rate": 1.587678591631144e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 766 + }, + { + "epoch": 0.9457459926017263, + "grad_norm": 2.426595174168101, + "learning_rate": 1.5866114235594014e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 767 + }, + { + "epoch": 0.9469790382244143, + "grad_norm": 2.085172001225346, + "learning_rate": 1.585543235952718e-05, + "loss": 0.9166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 768 + }, + { + "epoch": 0.9482120838471023, + "grad_norm": 2.1181076484114665, + "learning_rate": 1.5844740306676133e-05, + "loss": 0.8829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 769 + }, + { + "epoch": 0.9494451294697904, + "grad_norm": 2.64845194344853, + "learning_rate": 1.583403809562373e-05, + "loss": 0.8927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 770 + }, + { + "epoch": 0.9506781750924784, + "grad_norm": 2.4929599774250186, + "learning_rate": 1.5823325744970488e-05, + "loss": 0.9407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 771 + }, + { + "epoch": 0.9519112207151664, + "grad_norm": 2.368031870072283, + "learning_rate": 1.581260327333456e-05, + "loss": 0.9463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 772 + }, + { + "epoch": 0.9531442663378545, + "grad_norm": 2.136177985699207, + "learning_rate": 1.580187069935167e-05, + "loss": 0.8571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 773 + }, + { + "epoch": 0.9543773119605425, + "grad_norm": 2.128181510143405, + "learning_rate": 1.5791128041675114e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 774 + }, + { + "epoch": 0.9556103575832305, + "grad_norm": 2.220458022640665, + "learning_rate": 1.5780375318975715e-05, + "loss": 0.8445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 775 + }, + { + "epoch": 0.9568434032059187, + "grad_norm": 2.098330673951412, + "learning_rate": 1.5769612549941777e-05, + "loss": 0.964, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 776 + }, + { + "epoch": 0.9580764488286067, + "grad_norm": 2.094864541454194, + "learning_rate": 1.575883975327907e-05, + "loss": 0.8254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 777 + }, + { + "epoch": 0.9593094944512947, + "grad_norm": 2.3159416415455047, + "learning_rate": 1.5748056947710804e-05, + "loss": 0.9472, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 778 + }, + { + "epoch": 0.9605425400739828, + "grad_norm": 1.9950192842009455, + "learning_rate": 1.573726415197756e-05, + "loss": 0.7965, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 779 + }, + { + "epoch": 0.9617755856966708, + "grad_norm": 2.006344851533844, + "learning_rate": 1.5726461384837306e-05, + "loss": 0.8203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 780 + }, + { + "epoch": 0.9630086313193588, + "grad_norm": 2.2545503988149007, + "learning_rate": 1.5715648665065324e-05, + "loss": 0.9282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 781 + }, + { + "epoch": 0.9642416769420469, + "grad_norm": 2.2372695819528294, + "learning_rate": 1.5704826011454203e-05, + "loss": 0.9608, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 782 + }, + { + "epoch": 0.9654747225647349, + "grad_norm": 2.1710522433750357, + "learning_rate": 1.5693993442813792e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 783 + }, + { + "epoch": 0.966707768187423, + "grad_norm": 2.400731212803864, + "learning_rate": 1.568315097797118e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 784 + }, + { + "epoch": 0.967940813810111, + "grad_norm": 2.096674414566688, + "learning_rate": 1.5672298635770645e-05, + "loss": 0.6728, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 785 + }, + { + "epoch": 0.969173859432799, + "grad_norm": 2.2214704493931205, + "learning_rate": 1.566143643507364e-05, + "loss": 1.0617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 786 + }, + { + "epoch": 0.9704069050554871, + "grad_norm": 2.21278029237075, + "learning_rate": 1.5650564394758743e-05, + "loss": 0.8487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 787 + }, + { + "epoch": 0.9716399506781751, + "grad_norm": 2.3400569095138763, + "learning_rate": 1.5639682533721643e-05, + "loss": 0.9532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 788 + }, + { + "epoch": 0.9728729963008631, + "grad_norm": 2.1937723298850784, + "learning_rate": 1.5628790870875093e-05, + "loss": 0.9272, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 789 + }, + { + "epoch": 0.9741060419235512, + "grad_norm": 2.117801922125244, + "learning_rate": 1.561788942514889e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 790 + }, + { + "epoch": 0.9753390875462392, + "grad_norm": 2.436307402269952, + "learning_rate": 1.560697821548982e-05, + "loss": 0.8599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 791 + }, + { + "epoch": 0.9765721331689272, + "grad_norm": 2.2722775653462848, + "learning_rate": 1.5596057260861645e-05, + "loss": 0.8503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 792 + }, + { + "epoch": 0.9778051787916153, + "grad_norm": 2.22960384352197, + "learning_rate": 1.5585126580245066e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 793 + }, + { + "epoch": 0.9790382244143033, + "grad_norm": 2.1141630678799275, + "learning_rate": 1.557418619263768e-05, + "loss": 0.8418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 794 + }, + { + "epoch": 0.9802712700369913, + "grad_norm": 2.4590856272408224, + "learning_rate": 1.5563236117053965e-05, + "loss": 0.9197, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 795 + }, + { + "epoch": 0.9815043156596794, + "grad_norm": 2.678264319346431, + "learning_rate": 1.5552276372525237e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 796 + }, + { + "epoch": 0.9827373612823674, + "grad_norm": 2.262445020658096, + "learning_rate": 1.554130697809961e-05, + "loss": 0.9101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 797 + }, + { + "epoch": 0.9839704069050554, + "grad_norm": 2.327233920544754, + "learning_rate": 1.5530327952841965e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 798 + }, + { + "epoch": 0.9852034525277436, + "grad_norm": 2.1249691788358334, + "learning_rate": 1.551933931583393e-05, + "loss": 0.8824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 799 + }, + { + "epoch": 0.9864364981504316, + "grad_norm": 2.0648396343207325, + "learning_rate": 1.5508341086173843e-05, + "loss": 0.8895, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 800 + }, + { + "epoch": 0.9876695437731196, + "grad_norm": 2.109919443242373, + "learning_rate": 1.5497333282976698e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 801 + }, + { + "epoch": 0.9889025893958077, + "grad_norm": 2.3306692057820797, + "learning_rate": 1.5486315925374146e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 802 + }, + { + "epoch": 0.9901356350184957, + "grad_norm": 2.365009122779881, + "learning_rate": 1.547528903251443e-05, + "loss": 1.0604, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 803 + }, + { + "epoch": 0.9913686806411838, + "grad_norm": 2.2289048394713262, + "learning_rate": 1.5464252623562375e-05, + "loss": 0.8141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 804 + }, + { + "epoch": 0.9926017262638718, + "grad_norm": 2.457422539818384, + "learning_rate": 1.5453206717699334e-05, + "loss": 1.0228, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 805 + }, + { + "epoch": 0.9938347718865598, + "grad_norm": 2.0659430518171558, + "learning_rate": 1.544215133412318e-05, + "loss": 0.8641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 806 + }, + { + "epoch": 0.9950678175092479, + "grad_norm": 2.154728199437817, + "learning_rate": 1.543108649204824e-05, + "loss": 0.8029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 807 + }, + { + "epoch": 0.9963008631319359, + "grad_norm": 2.3679631548982196, + "learning_rate": 1.54200122107053e-05, + "loss": 0.866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 808 + }, + { + "epoch": 0.9975339087546239, + "grad_norm": 2.218685707693942, + "learning_rate": 1.540892850934154e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 809 + }, + { + "epoch": 0.998766954377312, + "grad_norm": 2.3274870898977187, + "learning_rate": 1.5397835407220508e-05, + "loss": 0.9724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 810 + }, + { + "epoch": 1.0, + "grad_norm": 2.159965188324255, + "learning_rate": 1.53867329236221e-05, + "loss": 0.762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 811 + }, + { + "epoch": 1.0012330456226881, + "grad_norm": 2.8883947578149782, + "learning_rate": 1.5375621077842517e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 812 + }, + { + "epoch": 1.002466091245376, + "grad_norm": 2.1490979216413963, + "learning_rate": 1.5364499889194225e-05, + "loss": 0.8653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 813 + }, + { + "epoch": 1.0036991368680641, + "grad_norm": 2.2100133362019787, + "learning_rate": 1.5353369377005928e-05, + "loss": 0.8871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 814 + }, + { + "epoch": 1.0049321824907522, + "grad_norm": 2.1549827645085444, + "learning_rate": 1.5342229560622545e-05, + "loss": 0.9146, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 815 + }, + { + "epoch": 1.0061652281134401, + "grad_norm": 2.0929823364178155, + "learning_rate": 1.5331080459405153e-05, + "loss": 0.8513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 816 + }, + { + "epoch": 1.0073982737361282, + "grad_norm": 2.220368656816853, + "learning_rate": 1.5319922092730974e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 817 + }, + { + "epoch": 1.0086313193588163, + "grad_norm": 2.1262847612815197, + "learning_rate": 1.5308754479993326e-05, + "loss": 0.8457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 818 + }, + { + "epoch": 1.0098643649815042, + "grad_norm": 2.1791530750144443, + "learning_rate": 1.5297577640601607e-05, + "loss": 0.7645, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 819 + }, + { + "epoch": 1.0110974106041923, + "grad_norm": 2.1726539474102426, + "learning_rate": 1.5286391593981243e-05, + "loss": 0.8684, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 820 + }, + { + "epoch": 1.0123304562268804, + "grad_norm": 2.193333692016209, + "learning_rate": 1.5275196359573667e-05, + "loss": 0.7566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 821 + }, + { + "epoch": 1.0135635018495683, + "grad_norm": 2.9212088946802623, + "learning_rate": 1.526399195683627e-05, + "loss": 1.0019, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 822 + }, + { + "epoch": 1.0147965474722564, + "grad_norm": 2.0580823212102417, + "learning_rate": 1.52527784052424e-05, + "loss": 0.9646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 823 + }, + { + "epoch": 1.0160295930949446, + "grad_norm": 2.19343597586527, + "learning_rate": 1.5241555724281283e-05, + "loss": 0.9453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 824 + }, + { + "epoch": 1.0172626387176325, + "grad_norm": 2.294578049001136, + "learning_rate": 1.523032393345802e-05, + "loss": 1.0273, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 825 + }, + { + "epoch": 1.0184956843403206, + "grad_norm": 1.9511593069959263, + "learning_rate": 1.5219083052293546e-05, + "loss": 0.8161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 826 + }, + { + "epoch": 1.0197287299630087, + "grad_norm": 2.421213793478604, + "learning_rate": 1.5207833100324595e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 827 + }, + { + "epoch": 1.0209617755856966, + "grad_norm": 2.1254569627132422, + "learning_rate": 1.5196574097103669e-05, + "loss": 1.0024, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 828 + }, + { + "epoch": 1.0221948212083847, + "grad_norm": 2.1741883221339213, + "learning_rate": 1.5185306062198993e-05, + "loss": 0.8694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 829 + }, + { + "epoch": 1.0234278668310728, + "grad_norm": 2.340571355784493, + "learning_rate": 1.5174029015194496e-05, + "loss": 0.9235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 830 + }, + { + "epoch": 1.024660912453761, + "grad_norm": 2.173793908678777, + "learning_rate": 1.5162742975689769e-05, + "loss": 0.8448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 831 + }, + { + "epoch": 1.0258939580764488, + "grad_norm": 2.149214329172728, + "learning_rate": 1.5151447963300027e-05, + "loss": 0.8388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 832 + }, + { + "epoch": 1.027127003699137, + "grad_norm": 2.059108722897467, + "learning_rate": 1.5140143997656092e-05, + "loss": 0.7303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 833 + }, + { + "epoch": 1.028360049321825, + "grad_norm": 2.2165337335607536, + "learning_rate": 1.512883109840433e-05, + "loss": 0.873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 834 + }, + { + "epoch": 1.029593094944513, + "grad_norm": 2.0093629352145923, + "learning_rate": 1.511750928520665e-05, + "loss": 0.7804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 835 + }, + { + "epoch": 1.030826140567201, + "grad_norm": 1.9963120627280608, + "learning_rate": 1.5106178577740442e-05, + "loss": 0.8642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 836 + }, + { + "epoch": 1.0320591861898891, + "grad_norm": 2.1963002199807384, + "learning_rate": 1.5094838995698555e-05, + "loss": 0.8688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 837 + }, + { + "epoch": 1.033292231812577, + "grad_norm": 2.1874228944498473, + "learning_rate": 1.5083490558789275e-05, + "loss": 0.9912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 838 + }, + { + "epoch": 1.0345252774352651, + "grad_norm": 2.12533056093305, + "learning_rate": 1.5072133286736261e-05, + "loss": 0.7925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 839 + }, + { + "epoch": 1.0357583230579532, + "grad_norm": 2.5000691541559745, + "learning_rate": 1.5060767199278536e-05, + "loss": 0.9544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 840 + }, + { + "epoch": 1.0369913686806411, + "grad_norm": 2.0622785615015125, + "learning_rate": 1.5049392316170443e-05, + "loss": 0.8433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 841 + }, + { + "epoch": 1.0382244143033292, + "grad_norm": 2.2009668761322563, + "learning_rate": 1.5038008657181616e-05, + "loss": 0.8157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 842 + }, + { + "epoch": 1.0394574599260173, + "grad_norm": 2.0959248263408385, + "learning_rate": 1.5026616242096932e-05, + "loss": 0.9167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 843 + }, + { + "epoch": 1.0406905055487052, + "grad_norm": 2.1893631450779893, + "learning_rate": 1.5015215090716493e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 844 + }, + { + "epoch": 1.0419235511713933, + "grad_norm": 2.077397878791412, + "learning_rate": 1.5003805222855584e-05, + "loss": 0.8383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 845 + }, + { + "epoch": 1.0431565967940815, + "grad_norm": 2.4055084927944623, + "learning_rate": 1.4992386658344643e-05, + "loss": 0.8404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 846 + }, + { + "epoch": 1.0443896424167693, + "grad_norm": 2.171777480987558, + "learning_rate": 1.4980959417029208e-05, + "loss": 0.7997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 847 + }, + { + "epoch": 1.0456226880394575, + "grad_norm": 2.157645250920109, + "learning_rate": 1.4969523518769917e-05, + "loss": 0.8738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 848 + }, + { + "epoch": 1.0468557336621456, + "grad_norm": 2.396639754777913, + "learning_rate": 1.4958078983442441e-05, + "loss": 0.6992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 849 + }, + { + "epoch": 1.0480887792848335, + "grad_norm": 2.4186334460214765, + "learning_rate": 1.4946625830937472e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 850 + }, + { + "epoch": 1.0493218249075216, + "grad_norm": 2.2977540698502996, + "learning_rate": 1.4935164081160663e-05, + "loss": 0.8683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 851 + }, + { + "epoch": 1.0505548705302097, + "grad_norm": 2.1379865919168, + "learning_rate": 1.4923693754032626e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 852 + }, + { + "epoch": 1.0517879161528976, + "grad_norm": 2.259076969791956, + "learning_rate": 1.491221486948887e-05, + "loss": 0.7691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 853 + }, + { + "epoch": 1.0530209617755857, + "grad_norm": 2.1377806626548357, + "learning_rate": 1.490072744747978e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 854 + }, + { + "epoch": 1.0542540073982738, + "grad_norm": 2.1044084023956167, + "learning_rate": 1.4889231507970581e-05, + "loss": 0.8401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 855 + }, + { + "epoch": 1.0554870530209617, + "grad_norm": 2.1271140055745286, + "learning_rate": 1.4877727070941302e-05, + "loss": 0.8084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 856 + }, + { + "epoch": 1.0567200986436498, + "grad_norm": 2.1718066261306817, + "learning_rate": 1.4866214156386734e-05, + "loss": 0.9014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 857 + }, + { + "epoch": 1.057953144266338, + "grad_norm": 2.1176596897595714, + "learning_rate": 1.4854692784316409e-05, + "loss": 0.7752, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 858 + }, + { + "epoch": 1.0591861898890258, + "grad_norm": 2.171891486084014, + "learning_rate": 1.4843162974754555e-05, + "loss": 0.6988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 859 + }, + { + "epoch": 1.060419235511714, + "grad_norm": 2.0055850881304718, + "learning_rate": 1.4831624747740065e-05, + "loss": 0.7928, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 860 + }, + { + "epoch": 1.061652281134402, + "grad_norm": 2.1272055457979624, + "learning_rate": 1.4820078123326464e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 861 + }, + { + "epoch": 1.06288532675709, + "grad_norm": 2.2409166254632416, + "learning_rate": 1.4808523121581871e-05, + "loss": 0.8428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 862 + }, + { + "epoch": 1.064118372379778, + "grad_norm": 2.167571295890406, + "learning_rate": 1.4796959762588957e-05, + "loss": 0.9326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 863 + }, + { + "epoch": 1.0653514180024661, + "grad_norm": 1.9896497336935064, + "learning_rate": 1.4785388066444928e-05, + "loss": 0.7178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 864 + }, + { + "epoch": 1.066584463625154, + "grad_norm": 1.9782830220589522, + "learning_rate": 1.4773808053261478e-05, + "loss": 0.8734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 865 + }, + { + "epoch": 1.0678175092478421, + "grad_norm": 2.1094953563881447, + "learning_rate": 1.4762219743164752e-05, + "loss": 0.72, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 866 + }, + { + "epoch": 1.0690505548705302, + "grad_norm": 2.172378157117765, + "learning_rate": 1.4750623156295316e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 867 + }, + { + "epoch": 1.0702836004932181, + "grad_norm": 2.777083857064727, + "learning_rate": 1.473901831280813e-05, + "loss": 0.887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 868 + }, + { + "epoch": 1.0715166461159062, + "grad_norm": 2.773482686862429, + "learning_rate": 1.4727405232872489e-05, + "loss": 0.9897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 869 + }, + { + "epoch": 1.0727496917385944, + "grad_norm": 2.303987298346897, + "learning_rate": 1.4715783936672017e-05, + "loss": 0.8267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 870 + }, + { + "epoch": 1.0739827373612825, + "grad_norm": 2.5090103249366718, + "learning_rate": 1.4704154444404606e-05, + "loss": 0.8297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 871 + }, + { + "epoch": 1.0752157829839704, + "grad_norm": 2.1804014393140334, + "learning_rate": 1.4692516776282404e-05, + "loss": 0.8673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 872 + }, + { + "epoch": 1.0764488286066585, + "grad_norm": 2.114741425345962, + "learning_rate": 1.4680870952531758e-05, + "loss": 0.7878, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 873 + }, + { + "epoch": 1.0776818742293466, + "grad_norm": 2.1487592065744043, + "learning_rate": 1.4669216993393195e-05, + "loss": 0.702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 874 + }, + { + "epoch": 1.0789149198520345, + "grad_norm": 2.250699104710453, + "learning_rate": 1.4657554919121388e-05, + "loss": 0.7805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 875 + }, + { + "epoch": 1.0801479654747226, + "grad_norm": 2.652130922890576, + "learning_rate": 1.46458847499851e-05, + "loss": 0.8594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 876 + }, + { + "epoch": 1.0813810110974107, + "grad_norm": 2.618911622711661, + "learning_rate": 1.4634206506267172e-05, + "loss": 0.974, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 877 + }, + { + "epoch": 1.0826140567200986, + "grad_norm": 2.104711420819804, + "learning_rate": 1.462252020826448e-05, + "loss": 0.7511, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 878 + }, + { + "epoch": 1.0838471023427867, + "grad_norm": 2.181334503374274, + "learning_rate": 1.4610825876287892e-05, + "loss": 0.8632, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 879 + }, + { + "epoch": 1.0850801479654748, + "grad_norm": 2.2281952334339126, + "learning_rate": 1.4599123530662242e-05, + "loss": 0.7623, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 880 + }, + { + "epoch": 1.0863131935881627, + "grad_norm": 2.2779707650922254, + "learning_rate": 1.4587413191726296e-05, + "loss": 0.7533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 881 + }, + { + "epoch": 1.0875462392108508, + "grad_norm": 2.227912879258747, + "learning_rate": 1.4575694879832704e-05, + "loss": 0.9742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 882 + }, + { + "epoch": 1.088779284833539, + "grad_norm": 2.200751986221906, + "learning_rate": 1.4563968615347983e-05, + "loss": 0.9859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 883 + }, + { + "epoch": 1.0900123304562268, + "grad_norm": 2.1474005643679503, + "learning_rate": 1.455223441865246e-05, + "loss": 0.885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 884 + }, + { + "epoch": 1.091245376078915, + "grad_norm": 2.5635895574555967, + "learning_rate": 1.4540492310140262e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 885 + }, + { + "epoch": 1.092478421701603, + "grad_norm": 2.086551890894098, + "learning_rate": 1.4528742310219251e-05, + "loss": 0.7838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 886 + }, + { + "epoch": 1.093711467324291, + "grad_norm": 2.100577896933077, + "learning_rate": 1.4516984439311017e-05, + "loss": 0.8245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 887 + }, + { + "epoch": 1.094944512946979, + "grad_norm": 2.0824253868885427, + "learning_rate": 1.4505218717850826e-05, + "loss": 0.8121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 888 + }, + { + "epoch": 1.0961775585696671, + "grad_norm": 2.424807773941655, + "learning_rate": 1.449344516628759e-05, + "loss": 0.6852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 889 + }, + { + "epoch": 1.097410604192355, + "grad_norm": 2.2203614431330587, + "learning_rate": 1.4481663805083823e-05, + "loss": 0.9022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 890 + }, + { + "epoch": 1.0986436498150431, + "grad_norm": 2.194684728317858, + "learning_rate": 1.4469874654715618e-05, + "loss": 0.8018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 891 + }, + { + "epoch": 1.0998766954377313, + "grad_norm": 2.2557943511918066, + "learning_rate": 1.4458077735672606e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 892 + }, + { + "epoch": 1.1011097410604191, + "grad_norm": 2.350846713741873, + "learning_rate": 1.4446273068457916e-05, + "loss": 0.8373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 893 + }, + { + "epoch": 1.1023427866831073, + "grad_norm": 2.3206912584026584, + "learning_rate": 1.4434460673588151e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 894 + }, + { + "epoch": 1.1035758323057954, + "grad_norm": 2.083251461091603, + "learning_rate": 1.4422640571593336e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 895 + }, + { + "epoch": 1.1048088779284833, + "grad_norm": 2.2429047820431456, + "learning_rate": 1.4410812783016893e-05, + "loss": 0.8066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 896 + }, + { + "epoch": 1.1060419235511714, + "grad_norm": 2.39736831788261, + "learning_rate": 1.4398977328415608e-05, + "loss": 0.8961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 897 + }, + { + "epoch": 1.1072749691738595, + "grad_norm": 2.4569397246397995, + "learning_rate": 1.4387134228359587e-05, + "loss": 0.9745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 898 + }, + { + "epoch": 1.1085080147965474, + "grad_norm": 2.194341713414294, + "learning_rate": 1.4375283503432223e-05, + "loss": 0.7642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 899 + }, + { + "epoch": 1.1097410604192355, + "grad_norm": 2.108094195937167, + "learning_rate": 1.4363425174230161e-05, + "loss": 0.7856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 900 + }, + { + "epoch": 1.1109741060419236, + "grad_norm": 2.140652113297743, + "learning_rate": 1.4351559261363267e-05, + "loss": 0.6648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 901 + }, + { + "epoch": 1.1122071516646117, + "grad_norm": 2.1936452316718063, + "learning_rate": 1.4339685785454587e-05, + "loss": 0.7336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 902 + }, + { + "epoch": 1.1134401972872996, + "grad_norm": 2.2269402261375038, + "learning_rate": 1.4327804767140303e-05, + "loss": 0.7708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 903 + }, + { + "epoch": 1.1146732429099877, + "grad_norm": 2.2989998022422453, + "learning_rate": 1.4315916227069713e-05, + "loss": 0.743, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 904 + }, + { + "epoch": 1.1159062885326758, + "grad_norm": 2.2917968867963427, + "learning_rate": 1.4304020185905189e-05, + "loss": 0.8126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 905 + }, + { + "epoch": 1.1171393341553637, + "grad_norm": 2.217875844794138, + "learning_rate": 1.4292116664322134e-05, + "loss": 0.6544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 906 + }, + { + "epoch": 1.1183723797780518, + "grad_norm": 2.4162119819935803, + "learning_rate": 1.4280205683008959e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 907 + }, + { + "epoch": 1.11960542540074, + "grad_norm": 2.3603757259614047, + "learning_rate": 1.4268287262667033e-05, + "loss": 0.7766, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 908 + }, + { + "epoch": 1.1208384710234278, + "grad_norm": 2.323141591914079, + "learning_rate": 1.4256361424010663e-05, + "loss": 0.8892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 909 + }, + { + "epoch": 1.122071516646116, + "grad_norm": 2.196302554500726, + "learning_rate": 1.4244428187767039e-05, + "loss": 0.9378, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 910 + }, + { + "epoch": 1.123304562268804, + "grad_norm": 2.2395403393552393, + "learning_rate": 1.4232487574676212e-05, + "loss": 1.0055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 911 + }, + { + "epoch": 1.124537607891492, + "grad_norm": 2.15769661287864, + "learning_rate": 1.4220539605491057e-05, + "loss": 0.748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 912 + }, + { + "epoch": 1.12577065351418, + "grad_norm": 2.0601339442187605, + "learning_rate": 1.4208584300977233e-05, + "loss": 0.7894, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 913 + }, + { + "epoch": 1.1270036991368682, + "grad_norm": 2.10447957980717, + "learning_rate": 1.4196621681913145e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 914 + }, + { + "epoch": 1.128236744759556, + "grad_norm": 2.3961229528506265, + "learning_rate": 1.4184651769089911e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 915 + }, + { + "epoch": 1.1294697903822442, + "grad_norm": 2.457287266967065, + "learning_rate": 1.417267458331133e-05, + "loss": 0.8519, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 916 + }, + { + "epoch": 1.1307028360049323, + "grad_norm": 2.023004753946474, + "learning_rate": 1.4160690145393838e-05, + "loss": 0.8449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 917 + }, + { + "epoch": 1.1319358816276202, + "grad_norm": 2.2533720475013634, + "learning_rate": 1.4148698476166471e-05, + "loss": 0.8221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 918 + }, + { + "epoch": 1.1331689272503083, + "grad_norm": 2.105957236229041, + "learning_rate": 1.4136699596470846e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 919 + }, + { + "epoch": 1.1344019728729964, + "grad_norm": 2.1652513956532946, + "learning_rate": 1.4124693527161095e-05, + "loss": 0.8322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 920 + }, + { + "epoch": 1.1356350184956843, + "grad_norm": 2.2463359048305693, + "learning_rate": 1.4112680289103858e-05, + "loss": 0.8458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 921 + }, + { + "epoch": 1.1368680641183724, + "grad_norm": 2.1946517808282784, + "learning_rate": 1.4100659903178231e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 922 + }, + { + "epoch": 1.1381011097410605, + "grad_norm": 2.116652661158196, + "learning_rate": 1.4088632390275731e-05, + "loss": 0.8496, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 923 + }, + { + "epoch": 1.1393341553637484, + "grad_norm": 2.1260376630680677, + "learning_rate": 1.4076597771300267e-05, + "loss": 0.7646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 924 + }, + { + "epoch": 1.1405672009864365, + "grad_norm": 2.1008385205849684, + "learning_rate": 1.4064556067168092e-05, + "loss": 0.8303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 925 + }, + { + "epoch": 1.1418002466091246, + "grad_norm": 2.1995475134175573, + "learning_rate": 1.4052507298807774e-05, + "loss": 0.8636, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 926 + }, + { + "epoch": 1.1430332922318125, + "grad_norm": 2.238025377523696, + "learning_rate": 1.404045148716016e-05, + "loss": 0.9095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 927 + }, + { + "epoch": 1.1442663378545006, + "grad_norm": 2.1584647692083636, + "learning_rate": 1.4028388653178346e-05, + "loss": 0.9552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 928 + }, + { + "epoch": 1.1454993834771887, + "grad_norm": 2.277913607137681, + "learning_rate": 1.4016318817827618e-05, + "loss": 0.6826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 929 + }, + { + "epoch": 1.1467324290998766, + "grad_norm": 2.1210301206048494, + "learning_rate": 1.4004242002085441e-05, + "loss": 0.7648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 930 + }, + { + "epoch": 1.1479654747225647, + "grad_norm": 2.1730216571228143, + "learning_rate": 1.3992158226941409e-05, + "loss": 0.7348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 931 + }, + { + "epoch": 1.1491985203452528, + "grad_norm": 2.2903920964212197, + "learning_rate": 1.3980067513397208e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 932 + }, + { + "epoch": 1.1504315659679407, + "grad_norm": 2.207582230014581, + "learning_rate": 1.3967969882466595e-05, + "loss": 0.8393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 933 + }, + { + "epoch": 1.1516646115906288, + "grad_norm": 2.1487993957640605, + "learning_rate": 1.3955865355175334e-05, + "loss": 0.9618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 934 + }, + { + "epoch": 1.152897657213317, + "grad_norm": 2.1100013581157673, + "learning_rate": 1.3943753952561179e-05, + "loss": 0.7185, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 935 + }, + { + "epoch": 1.1541307028360048, + "grad_norm": 2.1580493494759505, + "learning_rate": 1.3931635695673845e-05, + "loss": 0.8656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 936 + }, + { + "epoch": 1.155363748458693, + "grad_norm": 2.449600660057695, + "learning_rate": 1.3919510605574944e-05, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 937 + }, + { + "epoch": 1.156596794081381, + "grad_norm": 2.2204766170828782, + "learning_rate": 1.3907378703337977e-05, + "loss": 0.823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 938 + }, + { + "epoch": 1.157829839704069, + "grad_norm": 2.563974651134322, + "learning_rate": 1.389524001004827e-05, + "loss": 0.7685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 939 + }, + { + "epoch": 1.159062885326757, + "grad_norm": 2.258821529498267, + "learning_rate": 1.388309454680297e-05, + "loss": 0.673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 940 + }, + { + "epoch": 1.1602959309494452, + "grad_norm": 2.292191724384705, + "learning_rate": 1.3870942334710978e-05, + "loss": 0.8204, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 941 + }, + { + "epoch": 1.161528976572133, + "grad_norm": 2.283429457608797, + "learning_rate": 1.3858783394892925e-05, + "loss": 0.7777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 942 + }, + { + "epoch": 1.1627620221948212, + "grad_norm": 2.186748888333124, + "learning_rate": 1.3846617748481137e-05, + "loss": 0.7862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 943 + }, + { + "epoch": 1.1639950678175093, + "grad_norm": 2.262303399886297, + "learning_rate": 1.3834445416619602e-05, + "loss": 0.7559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 944 + }, + { + "epoch": 1.1652281134401972, + "grad_norm": 2.262832253985301, + "learning_rate": 1.3822266420463914e-05, + "loss": 0.9429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 945 + }, + { + "epoch": 1.1664611590628853, + "grad_norm": 2.150186199729938, + "learning_rate": 1.3810080781181264e-05, + "loss": 0.814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 946 + }, + { + "epoch": 1.1676942046855734, + "grad_norm": 2.343694448653775, + "learning_rate": 1.3797888519950373e-05, + "loss": 0.93, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 947 + }, + { + "epoch": 1.1689272503082615, + "grad_norm": 2.1813728957413385, + "learning_rate": 1.3785689657961494e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 948 + }, + { + "epoch": 1.1701602959309494, + "grad_norm": 2.119521364690522, + "learning_rate": 1.3773484216416326e-05, + "loss": 0.6718, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 949 + }, + { + "epoch": 1.1713933415536375, + "grad_norm": 2.2831289714651537, + "learning_rate": 1.3761272216528023e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 950 + }, + { + "epoch": 1.1726263871763256, + "grad_norm": 2.431755167028705, + "learning_rate": 1.3749053679521123e-05, + "loss": 0.8664, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 951 + }, + { + "epoch": 1.1738594327990135, + "grad_norm": 2.148435358078549, + "learning_rate": 1.3736828626631539e-05, + "loss": 0.6746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 952 + }, + { + "epoch": 1.1750924784217016, + "grad_norm": 2.124965667169116, + "learning_rate": 1.3724597079106498e-05, + "loss": 0.9202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 953 + }, + { + "epoch": 1.1763255240443897, + "grad_norm": 2.2382797156893397, + "learning_rate": 1.3712359058204525e-05, + "loss": 0.7696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 954 + }, + { + "epoch": 1.1775585696670776, + "grad_norm": 2.2207967591300632, + "learning_rate": 1.3700114585195381e-05, + "loss": 0.8654, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 955 + }, + { + "epoch": 1.1787916152897657, + "grad_norm": 2.3010582856815347, + "learning_rate": 1.3687863681360059e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 956 + }, + { + "epoch": 1.1800246609124538, + "grad_norm": 2.146527954418168, + "learning_rate": 1.367560636799071e-05, + "loss": 0.6291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 957 + }, + { + "epoch": 1.1812577065351417, + "grad_norm": 2.203217169625211, + "learning_rate": 1.3663342666390642e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 958 + }, + { + "epoch": 1.1824907521578298, + "grad_norm": 2.380280804061753, + "learning_rate": 1.365107259787425e-05, + "loss": 0.8015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 959 + }, + { + "epoch": 1.183723797780518, + "grad_norm": 2.7057879375717766, + "learning_rate": 1.3638796183767013e-05, + "loss": 0.8469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 960 + }, + { + "epoch": 1.1849568434032058, + "grad_norm": 2.0549864237362856, + "learning_rate": 1.362651344540542e-05, + "loss": 0.7724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 961 + }, + { + "epoch": 1.186189889025894, + "grad_norm": 2.165523344913771, + "learning_rate": 1.3614224404136964e-05, + "loss": 0.7119, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 962 + }, + { + "epoch": 1.187422934648582, + "grad_norm": 2.193503404484943, + "learning_rate": 1.3601929081320085e-05, + "loss": 0.769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 963 + }, + { + "epoch": 1.18865598027127, + "grad_norm": 2.30296466364579, + "learning_rate": 1.358962749832415e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 964 + }, + { + "epoch": 1.189889025893958, + "grad_norm": 2.3370321886887226, + "learning_rate": 1.3577319676529394e-05, + "loss": 0.8251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 965 + }, + { + "epoch": 1.1911220715166462, + "grad_norm": 2.1447616659612727, + "learning_rate": 1.3565005637326902e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 966 + }, + { + "epoch": 1.192355117139334, + "grad_norm": 2.1126610192998263, + "learning_rate": 1.3552685402118569e-05, + "loss": 0.893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 967 + }, + { + "epoch": 1.1935881627620222, + "grad_norm": 2.145481581859885, + "learning_rate": 1.3540358992317048e-05, + "loss": 0.8777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 968 + }, + { + "epoch": 1.1948212083847103, + "grad_norm": 2.0436843289234714, + "learning_rate": 1.3528026429345733e-05, + "loss": 0.7041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 969 + }, + { + "epoch": 1.1960542540073984, + "grad_norm": 2.3729052917065863, + "learning_rate": 1.351568773463871e-05, + "loss": 0.7822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 970 + }, + { + "epoch": 1.1972872996300863, + "grad_norm": 2.2771218973970977, + "learning_rate": 1.3503342929640713e-05, + "loss": 0.6887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 971 + }, + { + "epoch": 1.1985203452527744, + "grad_norm": 2.114743129136917, + "learning_rate": 1.3490992035807108e-05, + "loss": 0.6587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 972 + }, + { + "epoch": 1.1997533908754625, + "grad_norm": 2.210055964932895, + "learning_rate": 1.3478635074603844e-05, + "loss": 0.6986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 973 + }, + { + "epoch": 1.2009864364981504, + "grad_norm": 2.376467969342028, + "learning_rate": 1.34662720675074e-05, + "loss": 0.8992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 974 + }, + { + "epoch": 1.2022194821208385, + "grad_norm": 2.3169180290916893, + "learning_rate": 1.3453903036004784e-05, + "loss": 0.8468, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 975 + }, + { + "epoch": 1.2034525277435266, + "grad_norm": 2.4929171744348095, + "learning_rate": 1.3441528001593448e-05, + "loss": 0.8746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 976 + }, + { + "epoch": 1.2046855733662145, + "grad_norm": 2.3498012552061223, + "learning_rate": 1.3429146985781305e-05, + "loss": 0.848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 977 + }, + { + "epoch": 1.2059186189889026, + "grad_norm": 2.3519351668545223, + "learning_rate": 1.3416760010086647e-05, + "loss": 0.8045, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 978 + }, + { + "epoch": 1.2071516646115907, + "grad_norm": 2.152128272403449, + "learning_rate": 1.3404367096038127e-05, + "loss": 0.6621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 979 + }, + { + "epoch": 1.2083847102342786, + "grad_norm": 2.2310865684580126, + "learning_rate": 1.339196826517472e-05, + "loss": 0.8074, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 980 + }, + { + "epoch": 1.2096177558569667, + "grad_norm": 2.4218206259337336, + "learning_rate": 1.3379563539045688e-05, + "loss": 0.7489, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 981 + }, + { + "epoch": 1.2108508014796548, + "grad_norm": 2.3171020050526754, + "learning_rate": 1.3367152939210532e-05, + "loss": 0.6967, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 982 + }, + { + "epoch": 1.2120838471023427, + "grad_norm": 2.253666664504199, + "learning_rate": 1.3354736487238968e-05, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 983 + }, + { + "epoch": 1.2133168927250308, + "grad_norm": 2.202923113820963, + "learning_rate": 1.3342314204710878e-05, + "loss": 0.6643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 984 + }, + { + "epoch": 1.214549938347719, + "grad_norm": 2.3907506714199216, + "learning_rate": 1.332988611321628e-05, + "loss": 0.7483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 985 + }, + { + "epoch": 1.2157829839704068, + "grad_norm": 2.3288855585799726, + "learning_rate": 1.3317452234355286e-05, + "loss": 0.8775, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 986 + }, + { + "epoch": 1.217016029593095, + "grad_norm": 2.4270404362139124, + "learning_rate": 1.3305012589738075e-05, + "loss": 0.9152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 987 + }, + { + "epoch": 1.218249075215783, + "grad_norm": 2.3906498632865825, + "learning_rate": 1.3292567200984833e-05, + "loss": 0.6825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 988 + }, + { + "epoch": 1.219482120838471, + "grad_norm": 2.1776282578685495, + "learning_rate": 1.3280116089725743e-05, + "loss": 0.7098, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 989 + }, + { + "epoch": 1.220715166461159, + "grad_norm": 2.3972197284848265, + "learning_rate": 1.3267659277600921e-05, + "loss": 0.8804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 990 + }, + { + "epoch": 1.2219482120838472, + "grad_norm": 2.272454158882342, + "learning_rate": 1.3255196786260402e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 991 + }, + { + "epoch": 1.223181257706535, + "grad_norm": 2.0898886573973985, + "learning_rate": 1.3242728637364084e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 992 + }, + { + "epoch": 1.2244143033292232, + "grad_norm": 2.1283771526768422, + "learning_rate": 1.3230254852581703e-05, + "loss": 0.7541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 993 + }, + { + "epoch": 1.2256473489519113, + "grad_norm": 2.5947641083937785, + "learning_rate": 1.3217775453592786e-05, + "loss": 0.6516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 994 + }, + { + "epoch": 1.2268803945745992, + "grad_norm": 2.2994678179446155, + "learning_rate": 1.3205290462086623e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 995 + }, + { + "epoch": 1.2281134401972873, + "grad_norm": 2.419174005886699, + "learning_rate": 1.3192799899762216e-05, + "loss": 0.6653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 996 + }, + { + "epoch": 1.2293464858199754, + "grad_norm": 2.7052778802902724, + "learning_rate": 1.318030378832826e-05, + "loss": 0.8398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 997 + }, + { + "epoch": 1.2305795314426633, + "grad_norm": 2.182898548205837, + "learning_rate": 1.3167802149503079e-05, + "loss": 0.7902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 998 + }, + { + "epoch": 1.2318125770653514, + "grad_norm": 2.319601966909909, + "learning_rate": 1.315529500501462e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 999 + }, + { + "epoch": 1.2330456226880395, + "grad_norm": 2.3070852285534875, + "learning_rate": 1.3142782376600389e-05, + "loss": 0.7369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1000 + }, + { + "epoch": 1.2342786683107274, + "grad_norm": 2.3823064106232956, + "learning_rate": 1.3130264286007427e-05, + "loss": 0.8759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1001 + }, + { + "epoch": 1.2355117139334155, + "grad_norm": 2.2502915191290405, + "learning_rate": 1.3117740754992263e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1002 + }, + { + "epoch": 1.2367447595561036, + "grad_norm": 2.329022774742351, + "learning_rate": 1.310521180532089e-05, + "loss": 0.6222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1003 + }, + { + "epoch": 1.2379778051787915, + "grad_norm": 2.451176371025935, + "learning_rate": 1.3092677458768712e-05, + "loss": 0.9302, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1004 + }, + { + "epoch": 1.2392108508014796, + "grad_norm": 2.266087500621217, + "learning_rate": 1.3080137737120514e-05, + "loss": 0.7132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1005 + }, + { + "epoch": 1.2404438964241677, + "grad_norm": 2.2897262982963205, + "learning_rate": 1.3067592662170424e-05, + "loss": 0.7692, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1006 + }, + { + "epoch": 1.2416769420468556, + "grad_norm": 2.321897633330746, + "learning_rate": 1.3055042255721877e-05, + "loss": 0.8054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1007 + }, + { + "epoch": 1.2429099876695437, + "grad_norm": 2.327507620912996, + "learning_rate": 1.3042486539587566e-05, + "loss": 0.8495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1008 + }, + { + "epoch": 1.2441430332922319, + "grad_norm": 2.144625913300158, + "learning_rate": 1.3029925535589423e-05, + "loss": 0.7546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1009 + }, + { + "epoch": 1.2453760789149197, + "grad_norm": 2.1664603999352217, + "learning_rate": 1.301735926555856e-05, + "loss": 0.758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1010 + }, + { + "epoch": 1.2466091245376079, + "grad_norm": 2.2424395261695422, + "learning_rate": 1.300478775133525e-05, + "loss": 0.6794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1011 + }, + { + "epoch": 1.247842170160296, + "grad_norm": 2.135987556369877, + "learning_rate": 1.2992211014768872e-05, + "loss": 0.7814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1012 + }, + { + "epoch": 1.2490752157829839, + "grad_norm": 2.3429443823198812, + "learning_rate": 1.297962907771789e-05, + "loss": 0.756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1013 + }, + { + "epoch": 1.250308261405672, + "grad_norm": 2.3806466769436083, + "learning_rate": 1.29670419620498e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1014 + }, + { + "epoch": 1.25154130702836, + "grad_norm": 2.2614648564029536, + "learning_rate": 1.2954449689641105e-05, + "loss": 0.6227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1015 + }, + { + "epoch": 1.252774352651048, + "grad_norm": 2.690263466615422, + "learning_rate": 1.294185228237726e-05, + "loss": 0.8233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1016 + }, + { + "epoch": 1.254007398273736, + "grad_norm": 2.473463220519423, + "learning_rate": 1.2929249762152659e-05, + "loss": 0.7693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1017 + }, + { + "epoch": 1.2552404438964242, + "grad_norm": 2.203357353825905, + "learning_rate": 1.2916642150870567e-05, + "loss": 0.7486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1018 + }, + { + "epoch": 1.256473489519112, + "grad_norm": 2.407886169080358, + "learning_rate": 1.2904029470443105e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1019 + }, + { + "epoch": 1.2577065351418002, + "grad_norm": 2.416338321530697, + "learning_rate": 1.289141174279121e-05, + "loss": 0.7849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1020 + }, + { + "epoch": 1.2589395807644883, + "grad_norm": 2.1578914081370617, + "learning_rate": 1.2878788989844577e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1021 + }, + { + "epoch": 1.2601726263871762, + "grad_norm": 2.111289324604086, + "learning_rate": 1.2866161233541648e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1022 + }, + { + "epoch": 1.2614056720098643, + "grad_norm": 2.463350736417066, + "learning_rate": 1.2853528495829557e-05, + "loss": 0.7408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1023 + }, + { + "epoch": 1.2626387176325524, + "grad_norm": 2.5227574157477424, + "learning_rate": 1.284089079866409e-05, + "loss": 0.6943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1024 + }, + { + "epoch": 1.2638717632552403, + "grad_norm": 2.550177757961712, + "learning_rate": 1.2828248164009658e-05, + "loss": 0.8063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1025 + }, + { + "epoch": 1.2651048088779284, + "grad_norm": 2.279399386552855, + "learning_rate": 1.2815600613839254e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1026 + }, + { + "epoch": 1.2663378545006165, + "grad_norm": 2.403266684764464, + "learning_rate": 1.2802948170134411e-05, + "loss": 0.8159, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1027 + }, + { + "epoch": 1.2675709001233046, + "grad_norm": 2.3974372248557088, + "learning_rate": 1.2790290854885173e-05, + "loss": 0.7733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1028 + }, + { + "epoch": 1.2688039457459925, + "grad_norm": 2.75325380275428, + "learning_rate": 1.2777628690090035e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1029 + }, + { + "epoch": 1.2700369913686806, + "grad_norm": 2.253046145862575, + "learning_rate": 1.2764961697755947e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1030 + }, + { + "epoch": 1.2712700369913688, + "grad_norm": 2.0946644139817967, + "learning_rate": 1.2752289899898225e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1031 + }, + { + "epoch": 1.2725030826140566, + "grad_norm": 2.2771512156869576, + "learning_rate": 1.2739613318540546e-05, + "loss": 0.8351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1032 + }, + { + "epoch": 1.2737361282367448, + "grad_norm": 2.360325908676525, + "learning_rate": 1.2726931975714904e-05, + "loss": 0.8367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1033 + }, + { + "epoch": 1.2749691738594329, + "grad_norm": 2.310569034824901, + "learning_rate": 1.2714245893461568e-05, + "loss": 0.7346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1034 + }, + { + "epoch": 1.2762022194821208, + "grad_norm": 2.1364800896900413, + "learning_rate": 1.2701555093829034e-05, + "loss": 0.6637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1035 + }, + { + "epoch": 1.2774352651048089, + "grad_norm": 2.2320085770485734, + "learning_rate": 1.2688859598874013e-05, + "loss": 0.6849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1036 + }, + { + "epoch": 1.278668310727497, + "grad_norm": 2.2073805269530067, + "learning_rate": 1.2676159430661361e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1037 + }, + { + "epoch": 1.279901356350185, + "grad_norm": 2.2795714091180694, + "learning_rate": 1.2663454611264065e-05, + "loss": 0.8542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1038 + }, + { + "epoch": 1.281134401972873, + "grad_norm": 2.4086359220078535, + "learning_rate": 1.2650745162763195e-05, + "loss": 0.6699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1039 + }, + { + "epoch": 1.282367447595561, + "grad_norm": 2.3833290498603787, + "learning_rate": 1.2638031107247871e-05, + "loss": 0.7976, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1040 + }, + { + "epoch": 1.2836004932182492, + "grad_norm": 2.4078757121437873, + "learning_rate": 1.26253124668152e-05, + "loss": 0.7284, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1041 + }, + { + "epoch": 1.284833538840937, + "grad_norm": 2.3345693841609045, + "learning_rate": 1.2612589263570287e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1042 + }, + { + "epoch": 1.2860665844636252, + "grad_norm": 2.2804486360942633, + "learning_rate": 1.2599861519626142e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1043 + }, + { + "epoch": 1.2872996300863133, + "grad_norm": 2.219596427341039, + "learning_rate": 1.2587129257103684e-05, + "loss": 0.723, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1044 + }, + { + "epoch": 1.2885326757090012, + "grad_norm": 2.318632178358617, + "learning_rate": 1.2574392498131671e-05, + "loss": 0.7334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1045 + }, + { + "epoch": 1.2897657213316893, + "grad_norm": 2.1919076842831933, + "learning_rate": 1.2561651264846692e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1046 + }, + { + "epoch": 1.2909987669543774, + "grad_norm": 2.4787933074669186, + "learning_rate": 1.2548905579393096e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1047 + }, + { + "epoch": 1.2922318125770653, + "grad_norm": 2.4023622763250883, + "learning_rate": 1.2536155463922982e-05, + "loss": 0.826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1048 + }, + { + "epoch": 1.2934648581997534, + "grad_norm": 2.117125770604436, + "learning_rate": 1.252340094059614e-05, + "loss": 0.8052, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1049 + }, + { + "epoch": 1.2946979038224415, + "grad_norm": 2.2921041661226926, + "learning_rate": 1.251064203158003e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1050 + }, + { + "epoch": 1.2959309494451294, + "grad_norm": 2.5171575376979334, + "learning_rate": 1.2497878759049722e-05, + "loss": 0.6485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1051 + }, + { + "epoch": 1.2971639950678175, + "grad_norm": 2.1128556413546806, + "learning_rate": 1.248511114518788e-05, + "loss": 0.6076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1052 + }, + { + "epoch": 1.2983970406905057, + "grad_norm": 2.1493124843340907, + "learning_rate": 1.2472339212184712e-05, + "loss": 0.7343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1053 + }, + { + "epoch": 1.2996300863131935, + "grad_norm": 2.889368089302167, + "learning_rate": 1.2459562982237933e-05, + "loss": 0.7978, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1054 + }, + { + "epoch": 1.3008631319358817, + "grad_norm": 2.495383798802243, + "learning_rate": 1.2446782477552718e-05, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1055 + }, + { + "epoch": 1.3020961775585698, + "grad_norm": 2.517167273202911, + "learning_rate": 1.2433997720341682e-05, + "loss": 0.8663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1056 + }, + { + "epoch": 1.3033292231812577, + "grad_norm": 2.33725722509719, + "learning_rate": 1.2421208732824823e-05, + "loss": 0.7587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1057 + }, + { + "epoch": 1.3045622688039458, + "grad_norm": 2.29088652068904, + "learning_rate": 1.2408415537229497e-05, + "loss": 0.8488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1058 + }, + { + "epoch": 1.3057953144266339, + "grad_norm": 2.283233783970098, + "learning_rate": 1.239561815579037e-05, + "loss": 0.6389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1059 + }, + { + "epoch": 1.3070283600493218, + "grad_norm": 2.2928393078851568, + "learning_rate": 1.2382816610749388e-05, + "loss": 0.7261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1060 + }, + { + "epoch": 1.3082614056720099, + "grad_norm": 2.163749641191809, + "learning_rate": 1.2370010924355726e-05, + "loss": 0.6366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1061 + }, + { + "epoch": 1.309494451294698, + "grad_norm": 2.1229525047726963, + "learning_rate": 1.2357201118865762e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1062 + }, + { + "epoch": 1.3107274969173859, + "grad_norm": 2.272499973037141, + "learning_rate": 1.234438721654303e-05, + "loss": 0.6777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1063 + }, + { + "epoch": 1.311960542540074, + "grad_norm": 2.2172413085789207, + "learning_rate": 1.233156923965819e-05, + "loss": 0.8492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1064 + }, + { + "epoch": 1.313193588162762, + "grad_norm": 2.215217499299492, + "learning_rate": 1.2318747210488973e-05, + "loss": 0.816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1065 + }, + { + "epoch": 1.31442663378545, + "grad_norm": 2.2046121405476384, + "learning_rate": 1.2305921151320164e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1066 + }, + { + "epoch": 1.315659679408138, + "grad_norm": 2.410535974252942, + "learning_rate": 1.2293091084443547e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1067 + }, + { + "epoch": 1.3168927250308262, + "grad_norm": 2.2346919759512427, + "learning_rate": 1.2280257032157873e-05, + "loss": 0.5882, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1068 + }, + { + "epoch": 1.318125770653514, + "grad_norm": 2.217353695685963, + "learning_rate": 1.2267419016768814e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1069 + }, + { + "epoch": 1.3193588162762022, + "grad_norm": 2.3310973124804892, + "learning_rate": 1.225457706058894e-05, + "loss": 0.8387, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1070 + }, + { + "epoch": 1.3205918618988903, + "grad_norm": 2.363091956760975, + "learning_rate": 1.224173118593766e-05, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1071 + }, + { + "epoch": 1.3218249075215782, + "grad_norm": 2.3530303319299835, + "learning_rate": 1.22288814151412e-05, + "loss": 0.7271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1072 + }, + { + "epoch": 1.3230579531442663, + "grad_norm": 2.3412040138713137, + "learning_rate": 1.2216027770532559e-05, + "loss": 0.7482, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1073 + }, + { + "epoch": 1.3242909987669544, + "grad_norm": 2.388832200908418, + "learning_rate": 1.2203170274451462e-05, + "loss": 0.6927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1074 + }, + { + "epoch": 1.3255240443896423, + "grad_norm": 2.107265565119065, + "learning_rate": 1.219030894924433e-05, + "loss": 0.6069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1075 + }, + { + "epoch": 1.3267570900123304, + "grad_norm": 2.249447889766729, + "learning_rate": 1.2177443817264241e-05, + "loss": 0.7429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1076 + }, + { + "epoch": 1.3279901356350186, + "grad_norm": 2.337103531405435, + "learning_rate": 1.2164574900870887e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1077 + }, + { + "epoch": 1.3292231812577064, + "grad_norm": 2.4954282627925806, + "learning_rate": 1.2151702222430542e-05, + "loss": 0.781, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1078 + }, + { + "epoch": 1.3304562268803946, + "grad_norm": 2.4023586193420714, + "learning_rate": 1.2138825804316015e-05, + "loss": 0.7126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1079 + }, + { + "epoch": 1.3316892725030827, + "grad_norm": 2.293592183809771, + "learning_rate": 1.2125945668906611e-05, + "loss": 0.859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1080 + }, + { + "epoch": 1.3329223181257706, + "grad_norm": 2.280827683105619, + "learning_rate": 1.2113061838588104e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1081 + }, + { + "epoch": 1.3341553637484587, + "grad_norm": 2.2618423177043647, + "learning_rate": 1.2100174335752683e-05, + "loss": 0.8226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1082 + }, + { + "epoch": 1.3353884093711468, + "grad_norm": 2.263174374532883, + "learning_rate": 1.2087283182798923e-05, + "loss": 0.8671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1083 + }, + { + "epoch": 1.3366214549938347, + "grad_norm": 2.3088875209533875, + "learning_rate": 1.2074388402131737e-05, + "loss": 0.7575, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1084 + }, + { + "epoch": 1.3378545006165228, + "grad_norm": 2.217074039041576, + "learning_rate": 1.2061490016162357e-05, + "loss": 0.8498, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1085 + }, + { + "epoch": 1.339087546239211, + "grad_norm": 2.558767932572183, + "learning_rate": 1.2048588047308263e-05, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1086 + }, + { + "epoch": 1.3403205918618988, + "grad_norm": 2.1426590192119392, + "learning_rate": 1.203568251799318e-05, + "loss": 0.8473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1087 + }, + { + "epoch": 1.341553637484587, + "grad_norm": 2.3130116023826903, + "learning_rate": 1.2022773450647002e-05, + "loss": 0.6533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1088 + }, + { + "epoch": 1.342786683107275, + "grad_norm": 2.2474452770176505, + "learning_rate": 1.2009860867705793e-05, + "loss": 0.7731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1089 + }, + { + "epoch": 1.344019728729963, + "grad_norm": 2.553351511741815, + "learning_rate": 1.1996944791611711e-05, + "loss": 0.8881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1090 + }, + { + "epoch": 1.345252774352651, + "grad_norm": 2.4417233034036747, + "learning_rate": 1.1984025244812987e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1091 + }, + { + "epoch": 1.3464858199753391, + "grad_norm": 2.02108560427268, + "learning_rate": 1.1971102249763895e-05, + "loss": 0.5307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1092 + }, + { + "epoch": 1.347718865598027, + "grad_norm": 2.4627196027809655, + "learning_rate": 1.1958175828924692e-05, + "loss": 0.6095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1093 + }, + { + "epoch": 1.3489519112207151, + "grad_norm": 2.2087226765380703, + "learning_rate": 1.194524600476159e-05, + "loss": 0.7225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1094 + }, + { + "epoch": 1.3501849568434032, + "grad_norm": 2.268031072263066, + "learning_rate": 1.1932312799746722e-05, + "loss": 0.7492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1095 + }, + { + "epoch": 1.3514180024660911, + "grad_norm": 2.187415973980175, + "learning_rate": 1.1919376236358087e-05, + "loss": 0.6372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1096 + }, + { + "epoch": 1.3526510480887792, + "grad_norm": 2.3776853142465755, + "learning_rate": 1.1906436337079531e-05, + "loss": 0.757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1097 + }, + { + "epoch": 1.3538840937114673, + "grad_norm": 2.633083891326042, + "learning_rate": 1.189349312440069e-05, + "loss": 0.741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1098 + }, + { + "epoch": 1.3551171393341552, + "grad_norm": 2.3175112181742703, + "learning_rate": 1.1880546620816971e-05, + "loss": 0.577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1099 + }, + { + "epoch": 1.3563501849568433, + "grad_norm": 2.57847955905109, + "learning_rate": 1.1867596848829479e-05, + "loss": 0.7485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1100 + }, + { + "epoch": 1.3575832305795315, + "grad_norm": 2.2807181061723223, + "learning_rate": 1.1854643830945023e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1101 + }, + { + "epoch": 1.3588162762022196, + "grad_norm": 2.150298874669539, + "learning_rate": 1.1841687589676037e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1102 + }, + { + "epoch": 1.3600493218249075, + "grad_norm": 2.239857482190429, + "learning_rate": 1.1828728147540565e-05, + "loss": 0.5845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1103 + }, + { + "epoch": 1.3612823674475956, + "grad_norm": 2.391169760631781, + "learning_rate": 1.1815765527062207e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1104 + }, + { + "epoch": 1.3625154130702837, + "grad_norm": 2.233510553215683, + "learning_rate": 1.18027997507701e-05, + "loss": 0.687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1105 + }, + { + "epoch": 1.3637484586929716, + "grad_norm": 2.3406994321803376, + "learning_rate": 1.1789830841198854e-05, + "loss": 0.541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1106 + }, + { + "epoch": 1.3649815043156597, + "grad_norm": 3.0813520964620884, + "learning_rate": 1.1776858820888532e-05, + "loss": 0.8939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1107 + }, + { + "epoch": 1.3662145499383478, + "grad_norm": 2.289968441014776, + "learning_rate": 1.1763883712384593e-05, + "loss": 0.6178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1108 + }, + { + "epoch": 1.3674475955610357, + "grad_norm": 2.753373823273893, + "learning_rate": 1.1750905538237877e-05, + "loss": 0.8218, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1109 + }, + { + "epoch": 1.3686806411837238, + "grad_norm": 2.582931593682227, + "learning_rate": 1.1737924321004544e-05, + "loss": 0.5923, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1110 + }, + { + "epoch": 1.369913686806412, + "grad_norm": 2.577422739044973, + "learning_rate": 1.1724940083246042e-05, + "loss": 0.7151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1111 + }, + { + "epoch": 1.3711467324291, + "grad_norm": 2.4016601831106867, + "learning_rate": 1.171195284752907e-05, + "loss": 0.6719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1112 + }, + { + "epoch": 1.372379778051788, + "grad_norm": 2.2381623856871187, + "learning_rate": 1.1698962636425546e-05, + "loss": 0.7466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1113 + }, + { + "epoch": 1.373612823674476, + "grad_norm": 2.3077954380459498, + "learning_rate": 1.1685969472512544e-05, + "loss": 0.6609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1114 + }, + { + "epoch": 1.3748458692971641, + "grad_norm": 2.528220870127657, + "learning_rate": 1.1672973378372283e-05, + "loss": 0.7914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1115 + }, + { + "epoch": 1.376078914919852, + "grad_norm": 2.2640778971544946, + "learning_rate": 1.1659974376592065e-05, + "loss": 0.5914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1116 + }, + { + "epoch": 1.3773119605425401, + "grad_norm": 2.3357561361414994, + "learning_rate": 1.1646972489764254e-05, + "loss": 0.7194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1117 + }, + { + "epoch": 1.3785450061652282, + "grad_norm": 2.578566043676433, + "learning_rate": 1.163396774048622e-05, + "loss": 0.8069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1118 + }, + { + "epoch": 1.3797780517879161, + "grad_norm": 2.6394995201547853, + "learning_rate": 1.1620960151360317e-05, + "loss": 0.797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1119 + }, + { + "epoch": 1.3810110974106042, + "grad_norm": 2.2935912265432963, + "learning_rate": 1.1607949744993829e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1120 + }, + { + "epoch": 1.3822441430332923, + "grad_norm": 2.3848813510148394, + "learning_rate": 1.1594936543998933e-05, + "loss": 0.5816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1121 + }, + { + "epoch": 1.3834771886559802, + "grad_norm": 2.266951138614489, + "learning_rate": 1.1581920570992673e-05, + "loss": 0.8167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1122 + }, + { + "epoch": 1.3847102342786684, + "grad_norm": 2.0749563395142867, + "learning_rate": 1.1568901848596905e-05, + "loss": 0.6801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1123 + }, + { + "epoch": 1.3859432799013565, + "grad_norm": 2.259958476393895, + "learning_rate": 1.1555880399438257e-05, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1124 + }, + { + "epoch": 1.3871763255240444, + "grad_norm": 2.329878469976296, + "learning_rate": 1.1542856246148109e-05, + "loss": 0.7625, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1125 + }, + { + "epoch": 1.3884093711467325, + "grad_norm": 2.626476895279836, + "learning_rate": 1.1529829411362532e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1126 + }, + { + "epoch": 1.3896424167694206, + "grad_norm": 2.2360683032625297, + "learning_rate": 1.1516799917722262e-05, + "loss": 0.6588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1127 + }, + { + "epoch": 1.3908754623921085, + "grad_norm": 2.2606389128750664, + "learning_rate": 1.1503767787872652e-05, + "loss": 0.612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1128 + }, + { + "epoch": 1.3921085080147966, + "grad_norm": 2.3488413613095043, + "learning_rate": 1.149073304446364e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1129 + }, + { + "epoch": 1.3933415536374847, + "grad_norm": 2.187014482346159, + "learning_rate": 1.1477695710149706e-05, + "loss": 0.6897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1130 + }, + { + "epoch": 1.3945745992601726, + "grad_norm": 2.232874524639436, + "learning_rate": 1.1464655807589829e-05, + "loss": 0.7495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1131 + }, + { + "epoch": 1.3958076448828607, + "grad_norm": 2.2422283956051277, + "learning_rate": 1.1451613359447462e-05, + "loss": 0.7748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1132 + }, + { + "epoch": 1.3970406905055488, + "grad_norm": 2.2170691512303327, + "learning_rate": 1.1438568388390465e-05, + "loss": 0.7026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1133 + }, + { + "epoch": 1.3982737361282367, + "grad_norm": 2.3511388510197495, + "learning_rate": 1.1425520917091104e-05, + "loss": 0.7438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1134 + }, + { + "epoch": 1.3995067817509248, + "grad_norm": 2.302626316717242, + "learning_rate": 1.141247096822597e-05, + "loss": 0.6808, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1135 + }, + { + "epoch": 1.400739827373613, + "grad_norm": 2.358682446508914, + "learning_rate": 1.1399418564475977e-05, + "loss": 0.7462, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1136 + }, + { + "epoch": 1.4019728729963008, + "grad_norm": 2.4921870940266864, + "learning_rate": 1.138636372852629e-05, + "loss": 0.7565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1137 + }, + { + "epoch": 1.403205918618989, + "grad_norm": 2.881251766211438, + "learning_rate": 1.1373306483066315e-05, + "loss": 0.8531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1138 + }, + { + "epoch": 1.404438964241677, + "grad_norm": 2.374914568635526, + "learning_rate": 1.1360246850789638e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1139 + }, + { + "epoch": 1.405672009864365, + "grad_norm": 2.4892197803325478, + "learning_rate": 1.1347184854394e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1140 + }, + { + "epoch": 1.406905055487053, + "grad_norm": 2.335470361345138, + "learning_rate": 1.1334120516581239e-05, + "loss": 0.6239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1141 + }, + { + "epoch": 1.4081381011097411, + "grad_norm": 2.295703427576132, + "learning_rate": 1.1321053860057275e-05, + "loss": 0.7663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1142 + }, + { + "epoch": 1.409371146732429, + "grad_norm": 2.266519214119513, + "learning_rate": 1.1307984907532047e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1143 + }, + { + "epoch": 1.4106041923551171, + "grad_norm": 2.3854379731705917, + "learning_rate": 1.1294913681719497e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1144 + }, + { + "epoch": 1.4118372379778052, + "grad_norm": 2.3244362810232473, + "learning_rate": 1.1281840205337504e-05, + "loss": 0.694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1145 + }, + { + "epoch": 1.4130702836004931, + "grad_norm": 2.3269738125045727, + "learning_rate": 1.1268764501107872e-05, + "loss": 0.6867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1146 + }, + { + "epoch": 1.4143033292231812, + "grad_norm": 2.1038829157154533, + "learning_rate": 1.1255686591756265e-05, + "loss": 0.4939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1147 + }, + { + "epoch": 1.4155363748458694, + "grad_norm": 2.3057935606552356, + "learning_rate": 1.124260650001219e-05, + "loss": 0.697, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1148 + }, + { + "epoch": 1.4167694204685573, + "grad_norm": 2.4457435727408448, + "learning_rate": 1.1229524248608936e-05, + "loss": 0.7129, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1149 + }, + { + "epoch": 1.4180024660912454, + "grad_norm": 2.5516324512905793, + "learning_rate": 1.1216439860283556e-05, + "loss": 0.7116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1150 + }, + { + "epoch": 1.4192355117139335, + "grad_norm": 2.439191648697924, + "learning_rate": 1.120335335777681e-05, + "loss": 0.7269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1151 + }, + { + "epoch": 1.4204685573366214, + "grad_norm": 2.381736438404098, + "learning_rate": 1.1190264763833138e-05, + "loss": 0.7226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1152 + }, + { + "epoch": 1.4217016029593095, + "grad_norm": 2.326191969965805, + "learning_rate": 1.1177174101200609e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1153 + }, + { + "epoch": 1.4229346485819976, + "grad_norm": 2.4443478986341964, + "learning_rate": 1.1164081392630895e-05, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1154 + }, + { + "epoch": 1.4241676942046855, + "grad_norm": 2.4191171109083274, + "learning_rate": 1.115098666087921e-05, + "loss": 0.671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1155 + }, + { + "epoch": 1.4254007398273736, + "grad_norm": 2.331429993762429, + "learning_rate": 1.1137889928704303e-05, + "loss": 0.7875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1156 + }, + { + "epoch": 1.4266337854500617, + "grad_norm": 2.6182853389449905, + "learning_rate": 1.1124791218868383e-05, + "loss": 0.7465, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1157 + }, + { + "epoch": 1.4278668310727496, + "grad_norm": 2.3276838263801474, + "learning_rate": 1.111169055413711e-05, + "loss": 0.6731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1158 + }, + { + "epoch": 1.4290998766954377, + "grad_norm": 2.3555978375187783, + "learning_rate": 1.109858795727953e-05, + "loss": 0.7079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1159 + }, + { + "epoch": 1.4303329223181258, + "grad_norm": 2.243198455884724, + "learning_rate": 1.1085483451068056e-05, + "loss": 0.6767, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1160 + }, + { + "epoch": 1.4315659679408137, + "grad_norm": 2.426755129308782, + "learning_rate": 1.107237705827841e-05, + "loss": 0.6942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1161 + }, + { + "epoch": 1.4327990135635018, + "grad_norm": 2.310530088918712, + "learning_rate": 1.1059268801689605e-05, + "loss": 0.7414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1162 + }, + { + "epoch": 1.43403205918619, + "grad_norm": 2.3870562170497984, + "learning_rate": 1.104615870408388e-05, + "loss": 0.3904, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1163 + }, + { + "epoch": 1.4352651048088778, + "grad_norm": 2.185670591654591, + "learning_rate": 1.1033046788246684e-05, + "loss": 0.5667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1164 + }, + { + "epoch": 1.436498150431566, + "grad_norm": 2.2363567918319904, + "learning_rate": 1.1019933076966624e-05, + "loss": 0.6582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1165 + }, + { + "epoch": 1.437731196054254, + "grad_norm": 2.4989532650808917, + "learning_rate": 1.1006817593035423e-05, + "loss": 0.8567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1166 + }, + { + "epoch": 1.438964241676942, + "grad_norm": 2.3045828034357614, + "learning_rate": 1.0993700359247889e-05, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1167 + }, + { + "epoch": 1.44019728729963, + "grad_norm": 2.4408566344557263, + "learning_rate": 1.0980581398401875e-05, + "loss": 0.867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1168 + }, + { + "epoch": 1.4414303329223181, + "grad_norm": 2.433134793461258, + "learning_rate": 1.0967460733298224e-05, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1169 + }, + { + "epoch": 1.442663378545006, + "grad_norm": 2.084326240070432, + "learning_rate": 1.0954338386740746e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1170 + }, + { + "epoch": 1.4438964241676941, + "grad_norm": 2.3889216941699165, + "learning_rate": 1.0941214381536185e-05, + "loss": 0.7649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1171 + }, + { + "epoch": 1.4451294697903823, + "grad_norm": 2.3052157462538787, + "learning_rate": 1.092808874049415e-05, + "loss": 0.6548, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1172 + }, + { + "epoch": 1.4463625154130701, + "grad_norm": 2.412258190430439, + "learning_rate": 1.0914961486427103e-05, + "loss": 0.6655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1173 + }, + { + "epoch": 1.4475955610357583, + "grad_norm": 2.3820079102806755, + "learning_rate": 1.0901832642150308e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1174 + }, + { + "epoch": 1.4488286066584464, + "grad_norm": 2.826438039521343, + "learning_rate": 1.0888702230481788e-05, + "loss": 0.6666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1175 + }, + { + "epoch": 1.4500616522811345, + "grad_norm": 2.3005434823783166, + "learning_rate": 1.0875570274242305e-05, + "loss": 0.7333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1176 + }, + { + "epoch": 1.4512946979038224, + "grad_norm": 2.580920616275187, + "learning_rate": 1.0862436796255285e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1177 + }, + { + "epoch": 1.4525277435265105, + "grad_norm": 2.39619982968134, + "learning_rate": 1.0849301819346809e-05, + "loss": 0.6086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1178 + }, + { + "epoch": 1.4537607891491986, + "grad_norm": 2.4469170067316246, + "learning_rate": 1.0836165366345572e-05, + "loss": 0.7415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1179 + }, + { + "epoch": 1.4549938347718865, + "grad_norm": 2.4895765738835327, + "learning_rate": 1.0823027460082815e-05, + "loss": 0.8269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1180 + }, + { + "epoch": 1.4562268803945746, + "grad_norm": 2.1191361440164935, + "learning_rate": 1.0809888123392321e-05, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1181 + }, + { + "epoch": 1.4574599260172627, + "grad_norm": 2.0580916232206263, + "learning_rate": 1.0796747379110351e-05, + "loss": 0.5483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1182 + }, + { + "epoch": 1.4586929716399506, + "grad_norm": 2.358708614850794, + "learning_rate": 1.0783605250075615e-05, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1183 + }, + { + "epoch": 1.4599260172626387, + "grad_norm": 2.3706216835188068, + "learning_rate": 1.0770461759129225e-05, + "loss": 0.6925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1184 + }, + { + "epoch": 1.4611590628853268, + "grad_norm": 2.075953282437584, + "learning_rate": 1.0757316929114673e-05, + "loss": 0.7095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1185 + }, + { + "epoch": 1.462392108508015, + "grad_norm": 2.3425821034074175, + "learning_rate": 1.0744170782877765e-05, + "loss": 0.5685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1186 + }, + { + "epoch": 1.4636251541307028, + "grad_norm": 2.364864320197578, + "learning_rate": 1.07310233432666e-05, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1187 + }, + { + "epoch": 1.464858199753391, + "grad_norm": 2.269198583092931, + "learning_rate": 1.071787463313152e-05, + "loss": 0.5853, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1188 + }, + { + "epoch": 1.466091245376079, + "grad_norm": 2.3353973695396504, + "learning_rate": 1.0704724675325086e-05, + "loss": 0.6851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1189 + }, + { + "epoch": 1.467324290998767, + "grad_norm": 2.297514940312218, + "learning_rate": 1.0691573492702014e-05, + "loss": 0.6437, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1190 + }, + { + "epoch": 1.468557336621455, + "grad_norm": 2.347920635849606, + "learning_rate": 1.0678421108119161e-05, + "loss": 0.5607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1191 + }, + { + "epoch": 1.4697903822441432, + "grad_norm": 2.4998335942015792, + "learning_rate": 1.0665267544435464e-05, + "loss": 0.715, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1192 + }, + { + "epoch": 1.471023427866831, + "grad_norm": 2.6799025570634494, + "learning_rate": 1.065211282451192e-05, + "loss": 0.855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1193 + }, + { + "epoch": 1.4722564734895192, + "grad_norm": 2.3846645442657497, + "learning_rate": 1.0638956971211515e-05, + "loss": 0.8093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1194 + }, + { + "epoch": 1.4734895191122073, + "grad_norm": 2.289987275457904, + "learning_rate": 1.0625800007399232e-05, + "loss": 0.615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1195 + }, + { + "epoch": 1.4747225647348952, + "grad_norm": 2.398231591907997, + "learning_rate": 1.061264195594196e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1196 + }, + { + "epoch": 1.4759556103575833, + "grad_norm": 2.598957432876697, + "learning_rate": 1.0599482839708492e-05, + "loss": 0.5761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1197 + }, + { + "epoch": 1.4771886559802714, + "grad_norm": 2.3008244142116308, + "learning_rate": 1.0586322681569468e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1198 + }, + { + "epoch": 1.4784217016029593, + "grad_norm": 2.62152754882142, + "learning_rate": 1.0573161504397345e-05, + "loss": 0.9059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1199 + }, + { + "epoch": 1.4796547472256474, + "grad_norm": 2.7027129465803323, + "learning_rate": 1.0559999331066334e-05, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1200 + }, + { + "epoch": 1.4808877928483355, + "grad_norm": 2.441005731777059, + "learning_rate": 1.0546836184452397e-05, + "loss": 0.8212, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1201 + }, + { + "epoch": 1.4821208384710234, + "grad_norm": 2.4156909019245516, + "learning_rate": 1.0533672087433176e-05, + "loss": 0.6755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1202 + }, + { + "epoch": 1.4833538840937115, + "grad_norm": 2.373408895848631, + "learning_rate": 1.0520507062887966e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1203 + }, + { + "epoch": 1.4845869297163996, + "grad_norm": 2.1607701722680814, + "learning_rate": 1.050734113369768e-05, + "loss": 0.6858, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1204 + }, + { + "epoch": 1.4858199753390875, + "grad_norm": 2.211217022679158, + "learning_rate": 1.0494174322744797e-05, + "loss": 0.7383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1205 + }, + { + "epoch": 1.4870530209617756, + "grad_norm": 2.180028246293776, + "learning_rate": 1.0481006652913328e-05, + "loss": 0.5521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1206 + }, + { + "epoch": 1.4882860665844637, + "grad_norm": 2.393847834562169, + "learning_rate": 1.0467838147088785e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1207 + }, + { + "epoch": 1.4895191122071516, + "grad_norm": 2.20100213605133, + "learning_rate": 1.0454668828158122e-05, + "loss": 0.7316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1208 + }, + { + "epoch": 1.4907521578298397, + "grad_norm": 2.43405475601361, + "learning_rate": 1.0441498719009715e-05, + "loss": 0.7545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1209 + }, + { + "epoch": 1.4919852034525278, + "grad_norm": 2.3073004776119004, + "learning_rate": 1.0428327842533309e-05, + "loss": 0.5749, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1210 + }, + { + "epoch": 1.4932182490752157, + "grad_norm": 2.5013121128147535, + "learning_rate": 1.041515622161998e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1211 + }, + { + "epoch": 1.4944512946979038, + "grad_norm": 2.7168877702234537, + "learning_rate": 1.0401983879162108e-05, + "loss": 0.7342, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1212 + }, + { + "epoch": 1.495684340320592, + "grad_norm": 2.35281737411902, + "learning_rate": 1.0388810838053316e-05, + "loss": 0.6203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1213 + }, + { + "epoch": 1.4969173859432798, + "grad_norm": 2.400077140272344, + "learning_rate": 1.0375637121188447e-05, + "loss": 0.6201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1214 + }, + { + "epoch": 1.498150431565968, + "grad_norm": 2.5745039938322796, + "learning_rate": 1.0362462751463518e-05, + "loss": 0.6622, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1215 + }, + { + "epoch": 1.499383477188656, + "grad_norm": 2.4221988301394966, + "learning_rate": 1.0349287751775677e-05, + "loss": 0.7025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1216 + }, + { + "epoch": 1.500616522811344, + "grad_norm": 2.386783019524135, + "learning_rate": 1.033611214502317e-05, + "loss": 0.5671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1217 + }, + { + "epoch": 1.501849568434032, + "grad_norm": 2.539552131989461, + "learning_rate": 1.0322935954105303e-05, + "loss": 0.6469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1218 + }, + { + "epoch": 1.5030826140567202, + "grad_norm": 2.5808036138210673, + "learning_rate": 1.0309759201922386e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1219 + }, + { + "epoch": 1.504315659679408, + "grad_norm": 2.5496582442036395, + "learning_rate": 1.0296581911375712e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1220 + }, + { + "epoch": 1.5055487053020962, + "grad_norm": 2.304498549067636, + "learning_rate": 1.0283404105367512e-05, + "loss": 0.6358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1221 + }, + { + "epoch": 1.5067817509247843, + "grad_norm": 2.315506316971876, + "learning_rate": 1.0270225806800904e-05, + "loss": 0.643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1222 + }, + { + "epoch": 1.5080147965474722, + "grad_norm": 2.6544020115345726, + "learning_rate": 1.0257047038579868e-05, + "loss": 0.7594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1223 + }, + { + "epoch": 1.5092478421701603, + "grad_norm": 2.1918210902533843, + "learning_rate": 1.0243867823609207e-05, + "loss": 0.6812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1224 + }, + { + "epoch": 1.5104808877928484, + "grad_norm": 2.3728046507766316, + "learning_rate": 1.0230688184794482e-05, + "loss": 0.7418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1225 + }, + { + "epoch": 1.5117139334155363, + "grad_norm": 2.3382615280020254, + "learning_rate": 1.0217508145042007e-05, + "loss": 0.662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1226 + }, + { + "epoch": 1.5129469790382244, + "grad_norm": 2.174082577884056, + "learning_rate": 1.0204327727258787e-05, + "loss": 0.5961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1227 + }, + { + "epoch": 1.5141800246609125, + "grad_norm": 2.3119491032275596, + "learning_rate": 1.0191146954352486e-05, + "loss": 0.776, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1228 + }, + { + "epoch": 1.5154130702836004, + "grad_norm": 2.1858032851748086, + "learning_rate": 1.017796584923138e-05, + "loss": 0.5297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1229 + }, + { + "epoch": 1.5166461159062885, + "grad_norm": 2.379536502814726, + "learning_rate": 1.0164784434804327e-05, + "loss": 0.6471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1230 + }, + { + "epoch": 1.5178791615289766, + "grad_norm": 2.5211256291060655, + "learning_rate": 1.0151602733980722e-05, + "loss": 0.6299, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1231 + }, + { + "epoch": 1.5191122071516645, + "grad_norm": 2.406098151191697, + "learning_rate": 1.013842076967046e-05, + "loss": 0.6612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1232 + }, + { + "epoch": 1.5203452527743526, + "grad_norm": 2.4669460715815843, + "learning_rate": 1.0125238564783886e-05, + "loss": 0.7121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1233 + }, + { + "epoch": 1.5215782983970407, + "grad_norm": 2.420528895345573, + "learning_rate": 1.0112056142231774e-05, + "loss": 0.5687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1234 + }, + { + "epoch": 1.5228113440197286, + "grad_norm": 2.7051912038626074, + "learning_rate": 1.0098873524925268e-05, + "loss": 0.6909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1235 + }, + { + "epoch": 1.5240443896424167, + "grad_norm": 2.496413960615992, + "learning_rate": 1.008569073577585e-05, + "loss": 0.6021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1236 + }, + { + "epoch": 1.5252774352651048, + "grad_norm": 2.4865070368167808, + "learning_rate": 1.007250779769531e-05, + "loss": 0.5448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1237 + }, + { + "epoch": 1.5265104808877927, + "grad_norm": 2.4574534769181513, + "learning_rate": 1.005932473359569e-05, + "loss": 0.6773, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1238 + }, + { + "epoch": 1.5277435265104808, + "grad_norm": 2.3040474377405826, + "learning_rate": 1.0046141566389247e-05, + "loss": 0.5662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1239 + }, + { + "epoch": 1.528976572133169, + "grad_norm": 2.42558040335409, + "learning_rate": 1.0032958318988426e-05, + "loss": 0.6739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1240 + }, + { + "epoch": 1.5302096177558568, + "grad_norm": 2.3661886293872243, + "learning_rate": 1.0019775014305804e-05, + "loss": 0.6297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1241 + }, + { + "epoch": 1.531442663378545, + "grad_norm": 2.314603674785468, + "learning_rate": 1.0006591675254063e-05, + "loss": 0.5369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1242 + }, + { + "epoch": 1.532675709001233, + "grad_norm": 2.3999228551654412, + "learning_rate": 9.993408324745942e-06, + "loss": 0.6449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1243 + }, + { + "epoch": 1.533908754623921, + "grad_norm": 2.3862165327435667, + "learning_rate": 9.9802249856942e-06, + "loss": 0.6161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1244 + }, + { + "epoch": 1.5351418002466093, + "grad_norm": 2.4581866876380487, + "learning_rate": 9.96704168101158e-06, + "loss": 0.6433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1245 + }, + { + "epoch": 1.5363748458692972, + "grad_norm": 2.505484522547059, + "learning_rate": 9.953858433610756e-06, + "loss": 0.6709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1246 + }, + { + "epoch": 1.537607891491985, + "grad_norm": 2.561128967653091, + "learning_rate": 9.940675266404311e-06, + "loss": 0.7824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1247 + }, + { + "epoch": 1.5388409371146734, + "grad_norm": 2.4218451757264234, + "learning_rate": 9.927492202304693e-06, + "loss": 0.7463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1248 + }, + { + "epoch": 1.5400739827373613, + "grad_norm": 2.4125121526776154, + "learning_rate": 9.91430926422415e-06, + "loss": 0.6605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1249 + }, + { + "epoch": 1.5413070283600492, + "grad_norm": 2.445505663908, + "learning_rate": 9.901126475074737e-06, + "loss": 0.626, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1250 + }, + { + "epoch": 1.5425400739827375, + "grad_norm": 2.314619197677989, + "learning_rate": 9.887943857768228e-06, + "loss": 0.6289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1251 + }, + { + "epoch": 1.5437731196054254, + "grad_norm": 2.3232344719049967, + "learning_rate": 9.874761435216119e-06, + "loss": 0.5306, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1252 + }, + { + "epoch": 1.5450061652281133, + "grad_norm": 2.4949418638683007, + "learning_rate": 9.861579230329542e-06, + "loss": 0.5698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1253 + }, + { + "epoch": 1.5462392108508016, + "grad_norm": 2.3089453128921105, + "learning_rate": 9.84839726601928e-06, + "loss": 0.5801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1254 + }, + { + "epoch": 1.5474722564734895, + "grad_norm": 2.274576156733362, + "learning_rate": 9.835215565195676e-06, + "loss": 0.8039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1255 + }, + { + "epoch": 1.5487053020961774, + "grad_norm": 2.2604020732970818, + "learning_rate": 9.822034150768625e-06, + "loss": 0.7985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1256 + }, + { + "epoch": 1.5499383477188657, + "grad_norm": 2.1641795636087915, + "learning_rate": 9.80885304564752e-06, + "loss": 0.6705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1257 + }, + { + "epoch": 1.5511713933415536, + "grad_norm": 2.48269267754425, + "learning_rate": 9.795672272741216e-06, + "loss": 0.6985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1258 + }, + { + "epoch": 1.5524044389642415, + "grad_norm": 2.3771051242924903, + "learning_rate": 9.782491854957998e-06, + "loss": 0.6647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1259 + }, + { + "epoch": 1.5536374845869299, + "grad_norm": 2.2133943018691684, + "learning_rate": 9.769311815205522e-06, + "loss": 0.4763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1260 + }, + { + "epoch": 1.5548705302096177, + "grad_norm": 2.4689964457263605, + "learning_rate": 9.756132176390797e-06, + "loss": 0.667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1261 + }, + { + "epoch": 1.5561035758323056, + "grad_norm": 2.2973874789470936, + "learning_rate": 9.742952961420134e-06, + "loss": 0.6321, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1262 + }, + { + "epoch": 1.557336621454994, + "grad_norm": 2.368171008344279, + "learning_rate": 9.7297741931991e-06, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1263 + }, + { + "epoch": 1.5585696670776819, + "grad_norm": 2.435391702423389, + "learning_rate": 9.716595894632493e-06, + "loss": 0.5594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1264 + }, + { + "epoch": 1.55980271270037, + "grad_norm": 2.4382371041738184, + "learning_rate": 9.703418088624292e-06, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1265 + }, + { + "epoch": 1.561035758323058, + "grad_norm": 2.512447464904025, + "learning_rate": 9.690240798077616e-06, + "loss": 0.7641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1266 + }, + { + "epoch": 1.562268803945746, + "grad_norm": 2.271874085642332, + "learning_rate": 9.6770640458947e-06, + "loss": 0.6891, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1267 + }, + { + "epoch": 1.563501849568434, + "grad_norm": 2.2061279299276113, + "learning_rate": 9.66388785497683e-06, + "loss": 0.4999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1268 + }, + { + "epoch": 1.5647348951911222, + "grad_norm": 2.3058075007472416, + "learning_rate": 9.650712248224327e-06, + "loss": 0.6043, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1269 + }, + { + "epoch": 1.56596794081381, + "grad_norm": 2.444257229950801, + "learning_rate": 9.637537248536485e-06, + "loss": 0.6876, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1270 + }, + { + "epoch": 1.5672009864364982, + "grad_norm": 2.4130708431685197, + "learning_rate": 9.624362878811557e-06, + "loss": 0.5949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1271 + }, + { + "epoch": 1.5684340320591863, + "grad_norm": 2.340625006242772, + "learning_rate": 9.611189161946688e-06, + "loss": 0.6838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1272 + }, + { + "epoch": 1.5696670776818742, + "grad_norm": 2.2920633133032116, + "learning_rate": 9.598016120837894e-06, + "loss": 0.6186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1273 + }, + { + "epoch": 1.5709001233045623, + "grad_norm": 2.425912926280816, + "learning_rate": 9.584843778380022e-06, + "loss": 0.5922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1274 + }, + { + "epoch": 1.5721331689272504, + "grad_norm": 2.332147214152514, + "learning_rate": 9.571672157466695e-06, + "loss": 0.6555, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1275 + }, + { + "epoch": 1.5733662145499383, + "grad_norm": 2.398257339947792, + "learning_rate": 9.55850128099029e-06, + "loss": 0.6346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1276 + }, + { + "epoch": 1.5745992601726264, + "grad_norm": 2.142956121629841, + "learning_rate": 9.545331171841882e-06, + "loss": 0.5515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1277 + }, + { + "epoch": 1.5758323057953145, + "grad_norm": 2.407908540970335, + "learning_rate": 9.53216185291122e-06, + "loss": 0.5475, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1278 + }, + { + "epoch": 1.5770653514180024, + "grad_norm": 2.320052780540722, + "learning_rate": 9.518993347086676e-06, + "loss": 0.6044, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1279 + }, + { + "epoch": 1.5782983970406905, + "grad_norm": 2.4974178144916643, + "learning_rate": 9.505825677255205e-06, + "loss": 0.6436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1280 + }, + { + "epoch": 1.5795314426633786, + "grad_norm": 2.543298528231235, + "learning_rate": 9.492658866302325e-06, + "loss": 0.6388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1281 + }, + { + "epoch": 1.5807644882860665, + "grad_norm": 2.3663335712794917, + "learning_rate": 9.479492937112034e-06, + "loss": 0.5769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1282 + }, + { + "epoch": 1.5819975339087546, + "grad_norm": 2.301999998958795, + "learning_rate": 9.46632791256683e-06, + "loss": 0.6323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1283 + }, + { + "epoch": 1.5832305795314427, + "grad_norm": 2.3062144791204386, + "learning_rate": 9.453163815547604e-06, + "loss": 0.6727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1284 + }, + { + "epoch": 1.5844636251541306, + "grad_norm": 2.5906430609145947, + "learning_rate": 9.440000668933668e-06, + "loss": 0.7282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1285 + }, + { + "epoch": 1.5856966707768188, + "grad_norm": 2.6113606139990586, + "learning_rate": 9.426838495602659e-06, + "loss": 0.685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1286 + }, + { + "epoch": 1.5869297163995069, + "grad_norm": 2.1825712632265244, + "learning_rate": 9.413677318430532e-06, + "loss": 0.5556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1287 + }, + { + "epoch": 1.5881627620221948, + "grad_norm": 2.394162195116171, + "learning_rate": 9.400517160291511e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1288 + }, + { + "epoch": 1.5893958076448829, + "grad_norm": 2.375297383969579, + "learning_rate": 9.387358044058044e-06, + "loss": 0.7398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1289 + }, + { + "epoch": 1.590628853267571, + "grad_norm": 2.268014800016658, + "learning_rate": 9.374199992600775e-06, + "loss": 0.554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1290 + }, + { + "epoch": 1.5918618988902589, + "grad_norm": 2.273418089045509, + "learning_rate": 9.361043028788487e-06, + "loss": 0.5092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1291 + }, + { + "epoch": 1.593094944512947, + "grad_norm": 2.258287636448223, + "learning_rate": 9.347887175488086e-06, + "loss": 0.5517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1292 + }, + { + "epoch": 1.594327990135635, + "grad_norm": 2.4638493212600756, + "learning_rate": 9.334732455564538e-06, + "loss": 0.6536, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1293 + }, + { + "epoch": 1.595561035758323, + "grad_norm": 2.45854766908022, + "learning_rate": 9.321578891880838e-06, + "loss": 0.5922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1294 + }, + { + "epoch": 1.596794081381011, + "grad_norm": 2.6177980977782522, + "learning_rate": 9.30842650729799e-06, + "loss": 0.7473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1295 + }, + { + "epoch": 1.5980271270036992, + "grad_norm": 2.43914078165767, + "learning_rate": 9.295275324674915e-06, + "loss": 0.5646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1296 + }, + { + "epoch": 1.599260172626387, + "grad_norm": 2.365874131549762, + "learning_rate": 9.282125366868484e-06, + "loss": 0.5354, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1297 + }, + { + "epoch": 1.6004932182490752, + "grad_norm": 2.528721468130841, + "learning_rate": 9.268976656733402e-06, + "loss": 0.5554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1298 + }, + { + "epoch": 1.6017262638717633, + "grad_norm": 2.512092936139005, + "learning_rate": 9.255829217122237e-06, + "loss": 0.533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1299 + }, + { + "epoch": 1.6029593094944512, + "grad_norm": 2.354466866391755, + "learning_rate": 9.242683070885328e-06, + "loss": 0.4921, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1300 + }, + { + "epoch": 1.6041923551171393, + "grad_norm": 2.4883337985127936, + "learning_rate": 9.229538240870774e-06, + "loss": 0.6147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1301 + }, + { + "epoch": 1.6054254007398274, + "grad_norm": 2.4285611973298216, + "learning_rate": 9.21639474992439e-06, + "loss": 0.5665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1302 + }, + { + "epoch": 1.6066584463625153, + "grad_norm": 2.5959012805664994, + "learning_rate": 9.203252620889654e-06, + "loss": 0.6071, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1303 + }, + { + "epoch": 1.6078914919852034, + "grad_norm": 2.356030881254319, + "learning_rate": 9.190111876607684e-06, + "loss": 0.5597, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1304 + }, + { + "epoch": 1.6091245376078915, + "grad_norm": 2.4328199046350503, + "learning_rate": 9.176972539917188e-06, + "loss": 0.5543, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1305 + }, + { + "epoch": 1.6103575832305794, + "grad_norm": 2.531198496944925, + "learning_rate": 9.163834633654431e-06, + "loss": 0.7257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1306 + }, + { + "epoch": 1.6115906288532675, + "grad_norm": 2.4402117071794245, + "learning_rate": 9.150698180653193e-06, + "loss": 0.6173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1307 + }, + { + "epoch": 1.6128236744759556, + "grad_norm": 2.8953638685717547, + "learning_rate": 9.13756320374472e-06, + "loss": 0.7887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1308 + }, + { + "epoch": 1.6140567200986435, + "grad_norm": 2.459699675606036, + "learning_rate": 9.1244297257577e-06, + "loss": 0.7027, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1309 + }, + { + "epoch": 1.6152897657213316, + "grad_norm": 2.3328161668697467, + "learning_rate": 9.111297769518213e-06, + "loss": 0.5885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1310 + }, + { + "epoch": 1.6165228113440198, + "grad_norm": 2.4218060271919812, + "learning_rate": 9.098167357849699e-06, + "loss": 0.6497, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1311 + }, + { + "epoch": 1.6177558569667077, + "grad_norm": 2.294366317458183, + "learning_rate": 9.0850385135729e-06, + "loss": 0.5431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1312 + }, + { + "epoch": 1.6189889025893958, + "grad_norm": 2.3225490101301287, + "learning_rate": 9.071911259505852e-06, + "loss": 0.6644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1313 + }, + { + "epoch": 1.6202219482120839, + "grad_norm": 2.5880544489532826, + "learning_rate": 9.05878561846382e-06, + "loss": 0.517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1314 + }, + { + "epoch": 1.6214549938347718, + "grad_norm": 2.230629935973886, + "learning_rate": 9.045661613259254e-06, + "loss": 0.5261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1315 + }, + { + "epoch": 1.6226880394574599, + "grad_norm": 2.1770639270798307, + "learning_rate": 9.032539266701781e-06, + "loss": 0.4244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1316 + }, + { + "epoch": 1.623921085080148, + "grad_norm": 2.709845393499603, + "learning_rate": 9.019418601598129e-06, + "loss": 0.637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1317 + }, + { + "epoch": 1.6251541307028359, + "grad_norm": 3.460293995200764, + "learning_rate": 9.006299640752113e-06, + "loss": 0.6512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1318 + }, + { + "epoch": 1.6263871763255242, + "grad_norm": 3.5764897422793744, + "learning_rate": 8.993182406964579e-06, + "loss": 0.6457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1319 + }, + { + "epoch": 1.627620221948212, + "grad_norm": 2.5994177808868657, + "learning_rate": 8.980066923033378e-06, + "loss": 0.574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1320 + }, + { + "epoch": 1.6288532675709, + "grad_norm": 2.5757673207857135, + "learning_rate": 8.96695321175332e-06, + "loss": 0.4942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1321 + }, + { + "epoch": 1.6300863131935883, + "grad_norm": 2.453402170110872, + "learning_rate": 8.953841295916123e-06, + "loss": 0.6917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1322 + }, + { + "epoch": 1.6313193588162762, + "grad_norm": 2.654143412873541, + "learning_rate": 8.940731198310402e-06, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1323 + }, + { + "epoch": 1.632552404438964, + "grad_norm": 2.4250402090614935, + "learning_rate": 8.927622941721594e-06, + "loss": 0.6999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1324 + }, + { + "epoch": 1.6337854500616524, + "grad_norm": 2.761651051842012, + "learning_rate": 8.91451654893195e-06, + "loss": 0.6075, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1325 + }, + { + "epoch": 1.6350184956843403, + "grad_norm": 2.6957159420077894, + "learning_rate": 8.901412042720475e-06, + "loss": 0.6084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1326 + }, + { + "epoch": 1.6362515413070282, + "grad_norm": 2.2293950670646745, + "learning_rate": 8.888309445862892e-06, + "loss": 0.7359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1327 + }, + { + "epoch": 1.6374845869297165, + "grad_norm": 2.3226832325484663, + "learning_rate": 8.87520878113162e-06, + "loss": 0.5255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1328 + }, + { + "epoch": 1.6387176325524044, + "grad_norm": 2.5216170149103583, + "learning_rate": 8.8621100712957e-06, + "loss": 0.6287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1329 + }, + { + "epoch": 1.6399506781750923, + "grad_norm": 2.685061433529547, + "learning_rate": 8.849013339120793e-06, + "loss": 0.7093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1330 + }, + { + "epoch": 1.6411837237977807, + "grad_norm": 2.5648896854812975, + "learning_rate": 8.835918607369109e-06, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1331 + }, + { + "epoch": 1.6424167694204685, + "grad_norm": 2.6253257359858404, + "learning_rate": 8.822825898799393e-06, + "loss": 0.7446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1332 + }, + { + "epoch": 1.6436498150431564, + "grad_norm": 2.274148046399076, + "learning_rate": 8.809735236166864e-06, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1333 + }, + { + "epoch": 1.6448828606658448, + "grad_norm": 2.3764316480496537, + "learning_rate": 8.796646642223192e-06, + "loss": 0.6847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1334 + }, + { + "epoch": 1.6461159062885327, + "grad_norm": 2.3471176710530024, + "learning_rate": 8.783560139716449e-06, + "loss": 0.5461, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1335 + }, + { + "epoch": 1.6473489519112205, + "grad_norm": 2.4255652189713715, + "learning_rate": 8.770475751391066e-06, + "loss": 0.4423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1336 + }, + { + "epoch": 1.6485819975339089, + "grad_norm": 2.5292711557625673, + "learning_rate": 8.757393499987815e-06, + "loss": 0.7064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1337 + }, + { + "epoch": 1.6498150431565968, + "grad_norm": 2.427296462933323, + "learning_rate": 8.744313408243737e-06, + "loss": 0.6039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1338 + }, + { + "epoch": 1.6510480887792849, + "grad_norm": 2.5975517369534105, + "learning_rate": 8.731235498892127e-06, + "loss": 0.6391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1339 + }, + { + "epoch": 1.652281134401973, + "grad_norm": 2.491354995578972, + "learning_rate": 8.718159794662498e-06, + "loss": 0.6087, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1340 + }, + { + "epoch": 1.6535141800246609, + "grad_norm": 2.4884148521815526, + "learning_rate": 8.705086318280505e-06, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1341 + }, + { + "epoch": 1.654747225647349, + "grad_norm": 2.8396628177391907, + "learning_rate": 8.692015092467956e-06, + "loss": 0.7335, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1342 + }, + { + "epoch": 1.655980271270037, + "grad_norm": 2.7368157959861557, + "learning_rate": 8.678946139942728e-06, + "loss": 0.7007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1343 + }, + { + "epoch": 1.657213316892725, + "grad_norm": 2.3611404478863873, + "learning_rate": 8.665879483418766e-06, + "loss": 0.5668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1344 + }, + { + "epoch": 1.658446362515413, + "grad_norm": 2.191668075058394, + "learning_rate": 8.652815145606003e-06, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1345 + }, + { + "epoch": 1.6596794081381012, + "grad_norm": 2.14788648268729, + "learning_rate": 8.63975314921036e-06, + "loss": 0.5166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1346 + }, + { + "epoch": 1.660912453760789, + "grad_norm": 2.3068777176506963, + "learning_rate": 8.626693516933687e-06, + "loss": 0.5394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1347 + }, + { + "epoch": 1.6621454993834772, + "grad_norm": 2.49955124289679, + "learning_rate": 8.613636271473712e-06, + "loss": 0.6493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1348 + }, + { + "epoch": 1.6633785450061653, + "grad_norm": 2.1471304936125652, + "learning_rate": 8.60058143552403e-06, + "loss": 0.5408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1349 + }, + { + "epoch": 1.6646115906288532, + "grad_norm": 2.3378284654154444, + "learning_rate": 8.587529031774033e-06, + "loss": 0.5559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1350 + }, + { + "epoch": 1.6658446362515413, + "grad_norm": 2.272350354054125, + "learning_rate": 8.574479082908903e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1351 + }, + { + "epoch": 1.6670776818742294, + "grad_norm": 2.6718347888114393, + "learning_rate": 8.561431611609537e-06, + "loss": 0.6518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1352 + }, + { + "epoch": 1.6683107274969173, + "grad_norm": 2.5683646046012307, + "learning_rate": 8.548386640552542e-06, + "loss": 0.627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1353 + }, + { + "epoch": 1.6695437731196054, + "grad_norm": 2.429950074454179, + "learning_rate": 8.535344192410173e-06, + "loss": 0.6244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1354 + }, + { + "epoch": 1.6707768187422936, + "grad_norm": 2.459482147364562, + "learning_rate": 8.522304289850298e-06, + "loss": 0.5089, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1355 + }, + { + "epoch": 1.6720098643649814, + "grad_norm": 2.5078184456517802, + "learning_rate": 8.509266955536364e-06, + "loss": 0.5966, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1356 + }, + { + "epoch": 1.6732429099876696, + "grad_norm": 2.8738369967418786, + "learning_rate": 8.49623221212735e-06, + "loss": 0.7568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1357 + }, + { + "epoch": 1.6744759556103577, + "grad_norm": 2.566639808326863, + "learning_rate": 8.483200082277743e-06, + "loss": 0.6565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1358 + }, + { + "epoch": 1.6757090012330456, + "grad_norm": 2.18616861232984, + "learning_rate": 8.47017058863747e-06, + "loss": 0.5219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1359 + }, + { + "epoch": 1.6769420468557337, + "grad_norm": 2.2874183385113387, + "learning_rate": 8.457143753851893e-06, + "loss": 0.515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1360 + }, + { + "epoch": 1.6781750924784218, + "grad_norm": 2.3279389147190503, + "learning_rate": 8.444119600561746e-06, + "loss": 0.5737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1361 + }, + { + "epoch": 1.6794081381011097, + "grad_norm": 2.510085283064121, + "learning_rate": 8.431098151403098e-06, + "loss": 0.5527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1362 + }, + { + "epoch": 1.6806411837237978, + "grad_norm": 2.350766304372499, + "learning_rate": 8.41807942900733e-06, + "loss": 0.5444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1363 + }, + { + "epoch": 1.681874229346486, + "grad_norm": 2.4637839990972266, + "learning_rate": 8.405063456001069e-06, + "loss": 0.617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1364 + }, + { + "epoch": 1.6831072749691738, + "grad_norm": 2.3520486778455303, + "learning_rate": 8.392050255006172e-06, + "loss": 0.5787, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1365 + }, + { + "epoch": 1.684340320591862, + "grad_norm": 2.6951378359823064, + "learning_rate": 8.379039848639685e-06, + "loss": 0.5679, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1366 + }, + { + "epoch": 1.68557336621455, + "grad_norm": 2.5680249855790533, + "learning_rate": 8.366032259513782e-06, + "loss": 0.5366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1367 + }, + { + "epoch": 1.686806411837238, + "grad_norm": 2.7844067151133496, + "learning_rate": 8.35302751023575e-06, + "loss": 0.6181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1368 + }, + { + "epoch": 1.688039457459926, + "grad_norm": 2.3271421104987895, + "learning_rate": 8.340025623407938e-06, + "loss": 0.4529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1369 + }, + { + "epoch": 1.6892725030826141, + "grad_norm": 2.7324858411976822, + "learning_rate": 8.327026621627722e-06, + "loss": 0.7326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1370 + }, + { + "epoch": 1.690505548705302, + "grad_norm": 2.5720836300974383, + "learning_rate": 8.31403052748746e-06, + "loss": 0.6784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1371 + }, + { + "epoch": 1.6917385943279901, + "grad_norm": 2.440765245099682, + "learning_rate": 8.301037363574454e-06, + "loss": 0.6014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1372 + }, + { + "epoch": 1.6929716399506782, + "grad_norm": 2.594069817186588, + "learning_rate": 8.288047152470931e-06, + "loss": 0.5856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1373 + }, + { + "epoch": 1.6942046855733661, + "grad_norm": 2.4701026222170452, + "learning_rate": 8.27505991675396e-06, + "loss": 0.7036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1374 + }, + { + "epoch": 1.6954377311960542, + "grad_norm": 2.3893004984314365, + "learning_rate": 8.26207567899546e-06, + "loss": 0.5404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1375 + }, + { + "epoch": 1.6966707768187423, + "grad_norm": 2.3404547915407656, + "learning_rate": 8.249094461762124e-06, + "loss": 0.5142, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1376 + }, + { + "epoch": 1.6979038224414302, + "grad_norm": 2.4832770686344556, + "learning_rate": 8.236116287615409e-06, + "loss": 0.6183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1377 + }, + { + "epoch": 1.6991368680641183, + "grad_norm": 2.262237110861396, + "learning_rate": 8.223141179111471e-06, + "loss": 0.4504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1378 + }, + { + "epoch": 1.7003699136868065, + "grad_norm": 2.62998506842995, + "learning_rate": 8.210169158801146e-06, + "loss": 0.6316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1379 + }, + { + "epoch": 1.7016029593094943, + "grad_norm": 2.2853273497892164, + "learning_rate": 8.197200249229902e-06, + "loss": 0.5379, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1380 + }, + { + "epoch": 1.7028360049321825, + "grad_norm": 2.489883484033712, + "learning_rate": 8.184234472937795e-06, + "loss": 0.5635, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1381 + }, + { + "epoch": 1.7040690505548706, + "grad_norm": 2.4399096703479723, + "learning_rate": 8.171271852459442e-06, + "loss": 0.5672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1382 + }, + { + "epoch": 1.7053020961775585, + "grad_norm": 2.5877064849336673, + "learning_rate": 8.158312410323968e-06, + "loss": 0.56, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1383 + }, + { + "epoch": 1.7065351418002466, + "grad_norm": 2.5080025813914486, + "learning_rate": 8.145356169054982e-06, + "loss": 0.6415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1384 + }, + { + "epoch": 1.7077681874229347, + "grad_norm": 2.474550610636856, + "learning_rate": 8.132403151170524e-06, + "loss": 0.5209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1385 + }, + { + "epoch": 1.7090012330456226, + "grad_norm": 2.4546793222588037, + "learning_rate": 8.11945337918303e-06, + "loss": 0.6133, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1386 + }, + { + "epoch": 1.7102342786683107, + "grad_norm": 2.3960818942570623, + "learning_rate": 8.106506875599312e-06, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1387 + }, + { + "epoch": 1.7114673242909988, + "grad_norm": 2.323552505571765, + "learning_rate": 8.09356366292047e-06, + "loss": 0.5402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1388 + }, + { + "epoch": 1.7127003699136867, + "grad_norm": 2.088694863039824, + "learning_rate": 8.080623763641916e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1389 + }, + { + "epoch": 1.7139334155363748, + "grad_norm": 2.171047701838099, + "learning_rate": 8.067687200253281e-06, + "loss": 0.4947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1390 + }, + { + "epoch": 1.715166461159063, + "grad_norm": 2.4534573577067165, + "learning_rate": 8.054753995238415e-06, + "loss": 0.6494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1391 + }, + { + "epoch": 1.7163995067817508, + "grad_norm": 2.315613179007512, + "learning_rate": 8.04182417107531e-06, + "loss": 0.605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1392 + }, + { + "epoch": 1.7176325524044391, + "grad_norm": 2.382282194715041, + "learning_rate": 8.028897750236106e-06, + "loss": 0.3862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1393 + }, + { + "epoch": 1.718865598027127, + "grad_norm": 2.3689104564582606, + "learning_rate": 8.015974755187015e-06, + "loss": 0.4825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1394 + }, + { + "epoch": 1.720098643649815, + "grad_norm": 2.5004960798500098, + "learning_rate": 8.003055208388292e-06, + "loss": 0.5614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1395 + }, + { + "epoch": 1.7213316892725032, + "grad_norm": 2.376364766941311, + "learning_rate": 7.990139132294212e-06, + "loss": 0.4446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1396 + }, + { + "epoch": 1.7225647348951911, + "grad_norm": 2.6246902766612745, + "learning_rate": 7.977226549353e-06, + "loss": 0.6399, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1397 + }, + { + "epoch": 1.723797780517879, + "grad_norm": 2.484606542490508, + "learning_rate": 7.964317482006824e-06, + "loss": 0.5081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1398 + }, + { + "epoch": 1.7250308261405674, + "grad_norm": 3.38389227109005, + "learning_rate": 7.951411952691739e-06, + "loss": 0.755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1399 + }, + { + "epoch": 1.7262638717632552, + "grad_norm": 2.578597094452948, + "learning_rate": 7.938509983837647e-06, + "loss": 0.4826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1400 + }, + { + "epoch": 1.7274969173859431, + "grad_norm": 2.4802210829330105, + "learning_rate": 7.925611597868266e-06, + "loss": 0.493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1401 + }, + { + "epoch": 1.7287299630086315, + "grad_norm": 2.4523105300960517, + "learning_rate": 7.91271681720108e-06, + "loss": 0.6206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1402 + }, + { + "epoch": 1.7299630086313194, + "grad_norm": 2.386348175383016, + "learning_rate": 7.899825664247322e-06, + "loss": 0.5701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1403 + }, + { + "epoch": 1.7311960542540072, + "grad_norm": 2.4083074706971375, + "learning_rate": 7.886938161411897e-06, + "loss": 0.5884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1404 + }, + { + "epoch": 1.7324290998766956, + "grad_norm": 2.5069557743624316, + "learning_rate": 7.874054331093389e-06, + "loss": 0.4971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1405 + }, + { + "epoch": 1.7336621454993835, + "grad_norm": 2.55856356518894, + "learning_rate": 7.861174195683987e-06, + "loss": 0.5018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1406 + }, + { + "epoch": 1.7348951911220714, + "grad_norm": 2.568692520564215, + "learning_rate": 7.848297777569458e-06, + "loss": 0.6251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1407 + }, + { + "epoch": 1.7361282367447597, + "grad_norm": 2.626547699354354, + "learning_rate": 7.835425099129116e-06, + "loss": 0.4866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1408 + }, + { + "epoch": 1.7373612823674476, + "grad_norm": 2.655578554031113, + "learning_rate": 7.822556182735762e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1409 + }, + { + "epoch": 1.7385943279901355, + "grad_norm": 2.514238469731576, + "learning_rate": 7.809691050755676e-06, + "loss": 0.529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1410 + }, + { + "epoch": 1.7398273736128238, + "grad_norm": 2.486688645388684, + "learning_rate": 7.796829725548543e-06, + "loss": 0.5598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1411 + }, + { + "epoch": 1.7410604192355117, + "grad_norm": 2.5300619846568755, + "learning_rate": 7.783972229467443e-06, + "loss": 0.6431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1412 + }, + { + "epoch": 1.7422934648581998, + "grad_norm": 2.782005331836491, + "learning_rate": 7.771118584858801e-06, + "loss": 0.7072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1413 + }, + { + "epoch": 1.743526510480888, + "grad_norm": 2.4671627637221296, + "learning_rate": 7.758268814062343e-06, + "loss": 0.6404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1414 + }, + { + "epoch": 1.7447595561035758, + "grad_norm": 2.1612453871779285, + "learning_rate": 7.745422939411064e-06, + "loss": 0.4505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1415 + }, + { + "epoch": 1.745992601726264, + "grad_norm": 2.296774520080563, + "learning_rate": 7.73258098323119e-06, + "loss": 0.5345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1416 + }, + { + "epoch": 1.747225647348952, + "grad_norm": 2.6549737125780326, + "learning_rate": 7.719742967842133e-06, + "loss": 0.6232, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1417 + }, + { + "epoch": 1.74845869297164, + "grad_norm": 2.343040089887724, + "learning_rate": 7.706908915556457e-06, + "loss": 0.4859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1418 + }, + { + "epoch": 1.749691738594328, + "grad_norm": 2.459027338987637, + "learning_rate": 7.694078848679836e-06, + "loss": 0.6424, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1419 + }, + { + "epoch": 1.7509247842170161, + "grad_norm": 2.6006881140576383, + "learning_rate": 7.681252789511029e-06, + "loss": 0.6836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1420 + }, + { + "epoch": 1.752157829839704, + "grad_norm": 2.813687773910776, + "learning_rate": 7.668430760341813e-06, + "loss": 0.6596, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1421 + }, + { + "epoch": 1.7533908754623921, + "grad_norm": 2.7501988552768166, + "learning_rate": 7.655612783456973e-06, + "loss": 0.5757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1422 + }, + { + "epoch": 1.7546239210850803, + "grad_norm": 2.602564686174615, + "learning_rate": 7.64279888113424e-06, + "loss": 0.6304, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1423 + }, + { + "epoch": 1.7558569667077681, + "grad_norm": 2.2125559792606606, + "learning_rate": 7.629989075644277e-06, + "loss": 0.4345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1424 + }, + { + "epoch": 1.7570900123304563, + "grad_norm": 2.552624039521638, + "learning_rate": 7.6171833892506145e-06, + "loss": 0.6566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1425 + }, + { + "epoch": 1.7583230579531444, + "grad_norm": 2.53122973187578, + "learning_rate": 7.60438184420963e-06, + "loss": 0.5989, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1426 + }, + { + "epoch": 1.7595561035758323, + "grad_norm": 2.4914845101760834, + "learning_rate": 7.591584462770506e-06, + "loss": 0.5674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1427 + }, + { + "epoch": 1.7607891491985204, + "grad_norm": 2.71358370673903, + "learning_rate": 7.57879126717518e-06, + "loss": 0.6321, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1428 + }, + { + "epoch": 1.7620221948212085, + "grad_norm": 2.5302056277845413, + "learning_rate": 7.566002279658322e-06, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1429 + }, + { + "epoch": 1.7632552404438964, + "grad_norm": 2.371803209433193, + "learning_rate": 7.553217522447285e-06, + "loss": 0.4999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1430 + }, + { + "epoch": 1.7644882860665845, + "grad_norm": 2.527610014870494, + "learning_rate": 7.540437017762068e-06, + "loss": 0.5314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1431 + }, + { + "epoch": 1.7657213316892726, + "grad_norm": 2.440388163849852, + "learning_rate": 7.52766078781529e-06, + "loss": 0.6007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1432 + }, + { + "epoch": 1.7669543773119605, + "grad_norm": 2.369028343720501, + "learning_rate": 7.51488885481212e-06, + "loss": 0.5996, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1433 + }, + { + "epoch": 1.7681874229346486, + "grad_norm": 2.259203085653468, + "learning_rate": 7.502121240950284e-06, + "loss": 0.5022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1434 + }, + { + "epoch": 1.7694204685573367, + "grad_norm": 2.346742207074351, + "learning_rate": 7.489357968419975e-06, + "loss": 0.5232, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1435 + }, + { + "epoch": 1.7706535141800246, + "grad_norm": 2.3571911434438926, + "learning_rate": 7.476599059403866e-06, + "loss": 0.5529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1436 + }, + { + "epoch": 1.7718865598027127, + "grad_norm": 2.564577423619225, + "learning_rate": 7.463844536077021e-06, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1437 + }, + { + "epoch": 1.7731196054254008, + "grad_norm": 2.3757451812868062, + "learning_rate": 7.451094420606906e-06, + "loss": 0.5998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1438 + }, + { + "epoch": 1.7743526510480887, + "grad_norm": 2.3897936394773542, + "learning_rate": 7.438348735153311e-06, + "loss": 0.5233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1439 + }, + { + "epoch": 1.7755856966707768, + "grad_norm": 2.384296664394737, + "learning_rate": 7.425607501868328e-06, + "loss": 0.4798, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1440 + }, + { + "epoch": 1.776818742293465, + "grad_norm": 2.326128061878694, + "learning_rate": 7.41287074289632e-06, + "loss": 0.5628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1441 + }, + { + "epoch": 1.7780517879161528, + "grad_norm": 2.3448649893975646, + "learning_rate": 7.4001384803738595e-06, + "loss": 0.5539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1442 + }, + { + "epoch": 1.779284833538841, + "grad_norm": 2.335303557427351, + "learning_rate": 7.387410736429717e-06, + "loss": 0.463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1443 + }, + { + "epoch": 1.780517879161529, + "grad_norm": 2.5062756692562815, + "learning_rate": 7.374687533184801e-06, + "loss": 0.5804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1444 + }, + { + "epoch": 1.781750924784217, + "grad_norm": 2.667998142301554, + "learning_rate": 7.361968892752134e-06, + "loss": 0.5044, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1445 + }, + { + "epoch": 1.782983970406905, + "grad_norm": 2.8938425700505856, + "learning_rate": 7.349254837236808e-06, + "loss": 0.7082, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1446 + }, + { + "epoch": 1.7842170160295932, + "grad_norm": 2.638110967591072, + "learning_rate": 7.336545388735936e-06, + "loss": 0.5165, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1447 + }, + { + "epoch": 1.785450061652281, + "grad_norm": 2.5434717078771314, + "learning_rate": 7.323840569338645e-06, + "loss": 0.5274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1448 + }, + { + "epoch": 1.7866831072749692, + "grad_norm": 2.880952060159319, + "learning_rate": 7.31114040112599e-06, + "loss": 0.6319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1449 + }, + { + "epoch": 1.7879161528976573, + "grad_norm": 2.7232472895100974, + "learning_rate": 7.29844490617097e-06, + "loss": 0.6471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1450 + }, + { + "epoch": 1.7891491985203452, + "grad_norm": 2.6289292491069234, + "learning_rate": 7.285754106538435e-06, + "loss": 0.6144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1451 + }, + { + "epoch": 1.7903822441430333, + "grad_norm": 2.3571805040637117, + "learning_rate": 7.273068024285096e-06, + "loss": 0.5023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1452 + }, + { + "epoch": 1.7916152897657214, + "grad_norm": 2.5462291798096115, + "learning_rate": 7.260386681459457e-06, + "loss": 0.594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1453 + }, + { + "epoch": 1.7928483353884093, + "grad_norm": 2.364391159514722, + "learning_rate": 7.247710100101778e-06, + "loss": 0.3867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1454 + }, + { + "epoch": 1.7940813810110974, + "grad_norm": 2.557120216564179, + "learning_rate": 7.235038302244058e-06, + "loss": 0.4541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1455 + }, + { + "epoch": 1.7953144266337855, + "grad_norm": 2.3515150176521398, + "learning_rate": 7.2223713099099655e-06, + "loss": 0.5005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1456 + }, + { + "epoch": 1.7965474722564734, + "grad_norm": 3.0149925007019465, + "learning_rate": 7.209709145114834e-06, + "loss": 0.6562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1457 + }, + { + "epoch": 1.7977805178791615, + "grad_norm": 2.9385786617892484, + "learning_rate": 7.197051829865591e-06, + "loss": 0.7172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1458 + }, + { + "epoch": 1.7990135635018496, + "grad_norm": 2.4438200151823577, + "learning_rate": 7.1843993861607485e-06, + "loss": 0.545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1459 + }, + { + "epoch": 1.8002466091245375, + "grad_norm": 2.189016598422486, + "learning_rate": 7.171751835990346e-06, + "loss": 0.3945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1460 + }, + { + "epoch": 1.8014796547472256, + "grad_norm": 2.599380048646185, + "learning_rate": 7.159109201335914e-06, + "loss": 0.6221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1461 + }, + { + "epoch": 1.8027127003699137, + "grad_norm": 2.3218284667005045, + "learning_rate": 7.146471504170448e-06, + "loss": 0.4081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1462 + }, + { + "epoch": 1.8039457459926016, + "grad_norm": 2.483257454081904, + "learning_rate": 7.1338387664583545e-06, + "loss": 0.5952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1463 + }, + { + "epoch": 1.8051787916152897, + "grad_norm": 2.3060282168534267, + "learning_rate": 7.1212110101554225e-06, + "loss": 0.478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1464 + }, + { + "epoch": 1.8064118372379778, + "grad_norm": 2.353277914016821, + "learning_rate": 7.108588257208793e-06, + "loss": 0.4602, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1465 + }, + { + "epoch": 1.8076448828606657, + "grad_norm": 2.402021352720206, + "learning_rate": 7.095970529556895e-06, + "loss": 0.423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1466 + }, + { + "epoch": 1.808877928483354, + "grad_norm": 2.4089157744395187, + "learning_rate": 7.083357849129437e-06, + "loss": 0.4962, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1467 + }, + { + "epoch": 1.810110974106042, + "grad_norm": 2.4445181455084803, + "learning_rate": 7.070750237847344e-06, + "loss": 0.4237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1468 + }, + { + "epoch": 1.8113440197287298, + "grad_norm": 2.5896452836545576, + "learning_rate": 7.058147717622742e-06, + "loss": 0.6034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1469 + }, + { + "epoch": 1.8125770653514182, + "grad_norm": 2.5692451627366215, + "learning_rate": 7.0455503103588975e-06, + "loss": 0.4897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1470 + }, + { + "epoch": 1.813810110974106, + "grad_norm": 2.5193270914325185, + "learning_rate": 7.0329580379502e-06, + "loss": 0.5908, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1471 + }, + { + "epoch": 1.815043156596794, + "grad_norm": 2.6722250129620586, + "learning_rate": 7.020370922282113e-06, + "loss": 0.6295, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1472 + }, + { + "epoch": 1.8162762022194823, + "grad_norm": 2.4547108738067838, + "learning_rate": 7.00778898523113e-06, + "loss": 0.5357, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1473 + }, + { + "epoch": 1.8175092478421702, + "grad_norm": 2.437273516415158, + "learning_rate": 6.995212248664755e-06, + "loss": 0.5101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1474 + }, + { + "epoch": 1.818742293464858, + "grad_norm": 2.279547074068301, + "learning_rate": 6.982640734441443e-06, + "loss": 0.5079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1475 + }, + { + "epoch": 1.8199753390875464, + "grad_norm": 2.435539711925499, + "learning_rate": 6.9700744644105815e-06, + "loss": 0.5245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1476 + }, + { + "epoch": 1.8212083847102343, + "grad_norm": 2.420190205704888, + "learning_rate": 6.9575134604124364e-06, + "loss": 0.5115, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1477 + }, + { + "epoch": 1.8224414303329222, + "grad_norm": 2.6220274719559424, + "learning_rate": 6.9449577442781246e-06, + "loss": 0.5708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1478 + }, + { + "epoch": 1.8236744759556105, + "grad_norm": 2.4198151981250624, + "learning_rate": 6.932407337829579e-06, + "loss": 0.5779, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1479 + }, + { + "epoch": 1.8249075215782984, + "grad_norm": 2.710527668536825, + "learning_rate": 6.919862262879488e-06, + "loss": 0.706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1480 + }, + { + "epoch": 1.8261405672009863, + "grad_norm": 2.260314185301613, + "learning_rate": 6.907322541231294e-06, + "loss": 0.3152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1481 + }, + { + "epoch": 1.8273736128236746, + "grad_norm": 2.4880601560363087, + "learning_rate": 6.894788194679112e-06, + "loss": 0.4559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1482 + }, + { + "epoch": 1.8286066584463625, + "grad_norm": 2.53859220201902, + "learning_rate": 6.882259245007741e-06, + "loss": 0.6073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1483 + }, + { + "epoch": 1.8298397040690506, + "grad_norm": 2.406753699086694, + "learning_rate": 6.869735713992576e-06, + "loss": 0.5376, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1484 + }, + { + "epoch": 1.8310727496917387, + "grad_norm": 2.5573243142415616, + "learning_rate": 6.857217623399613e-06, + "loss": 0.5161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1485 + }, + { + "epoch": 1.8323057953144266, + "grad_norm": 2.4701507919473964, + "learning_rate": 6.844704994985383e-06, + "loss": 0.4962, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1486 + }, + { + "epoch": 1.8335388409371147, + "grad_norm": 2.505190115969386, + "learning_rate": 6.832197850496922e-06, + "loss": 0.5941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1487 + }, + { + "epoch": 1.8347718865598028, + "grad_norm": 2.8094520017647766, + "learning_rate": 6.819696211671745e-06, + "loss": 0.6998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1488 + }, + { + "epoch": 1.8360049321824907, + "grad_norm": 2.7116637712221685, + "learning_rate": 6.807200100237786e-06, + "loss": 0.5916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1489 + }, + { + "epoch": 1.8372379778051788, + "grad_norm": 2.4742606262555538, + "learning_rate": 6.794709537913382e-06, + "loss": 0.4571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1490 + }, + { + "epoch": 1.838471023427867, + "grad_norm": 2.3388179882861038, + "learning_rate": 6.782224546407217e-06, + "loss": 0.4336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1491 + }, + { + "epoch": 1.8397040690505548, + "grad_norm": 2.5535368655522723, + "learning_rate": 6.769745147418298e-06, + "loss": 0.5513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1492 + }, + { + "epoch": 1.840937114673243, + "grad_norm": 2.3844296720019136, + "learning_rate": 6.75727136263592e-06, + "loss": 0.4769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1493 + }, + { + "epoch": 1.842170160295931, + "grad_norm": 2.430982238426814, + "learning_rate": 6.744803213739599e-06, + "loss": 0.6191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1494 + }, + { + "epoch": 1.843403205918619, + "grad_norm": 2.5719309543455484, + "learning_rate": 6.732340722399083e-06, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1495 + }, + { + "epoch": 1.844636251541307, + "grad_norm": 2.4358254754188247, + "learning_rate": 6.719883910274258e-06, + "loss": 0.6404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1496 + }, + { + "epoch": 1.8458692971639952, + "grad_norm": 2.2277422785314385, + "learning_rate": 6.7074327990151666e-06, + "loss": 0.4467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1497 + }, + { + "epoch": 1.847102342786683, + "grad_norm": 2.3553522334648234, + "learning_rate": 6.694987410261927e-06, + "loss": 0.4733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1498 + }, + { + "epoch": 1.8483353884093712, + "grad_norm": 2.305268243528217, + "learning_rate": 6.6825477656447135e-06, + "loss": 0.5213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1499 + }, + { + "epoch": 1.8495684340320593, + "grad_norm": 2.476358939717476, + "learning_rate": 6.6701138867837245e-06, + "loss": 0.5979, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1500 + }, + { + "epoch": 1.8508014796547472, + "grad_norm": 2.5413097090235897, + "learning_rate": 6.657685795289126e-06, + "loss": 0.464, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1501 + }, + { + "epoch": 1.8520345252774353, + "grad_norm": 2.581561221486171, + "learning_rate": 6.6452635127610366e-06, + "loss": 0.5169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1502 + }, + { + "epoch": 1.8532675709001234, + "grad_norm": 2.752939514877913, + "learning_rate": 6.632847060789471e-06, + "loss": 0.5458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1503 + }, + { + "epoch": 1.8545006165228113, + "grad_norm": 2.393983614231186, + "learning_rate": 6.620436460954314e-06, + "loss": 0.5606, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1504 + }, + { + "epoch": 1.8557336621454994, + "grad_norm": 2.2296540462723415, + "learning_rate": 6.608031734825282e-06, + "loss": 0.5184, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1505 + }, + { + "epoch": 1.8569667077681875, + "grad_norm": 2.3150895635322377, + "learning_rate": 6.595632903961876e-06, + "loss": 0.3391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1506 + }, + { + "epoch": 1.8581997533908754, + "grad_norm": 2.4812557469058687, + "learning_rate": 6.583239989913357e-06, + "loss": 0.5283, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1507 + }, + { + "epoch": 1.8594327990135635, + "grad_norm": 2.240249157372891, + "learning_rate": 6.570853014218699e-06, + "loss": 0.4267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1508 + }, + { + "epoch": 1.8606658446362516, + "grad_norm": 2.418565073853559, + "learning_rate": 6.558471998406556e-06, + "loss": 0.5317, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1509 + }, + { + "epoch": 1.8618988902589395, + "grad_norm": 2.509766709851647, + "learning_rate": 6.546096963995221e-06, + "loss": 0.5639, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1510 + }, + { + "epoch": 1.8631319358816276, + "grad_norm": 2.6229591577559606, + "learning_rate": 6.5337279324925995e-06, + "loss": 0.6702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1511 + }, + { + "epoch": 1.8643649815043157, + "grad_norm": 2.8436863684007605, + "learning_rate": 6.521364925396159e-06, + "loss": 0.6629, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1512 + }, + { + "epoch": 1.8655980271270036, + "grad_norm": 2.88285582890902, + "learning_rate": 6.50900796419289e-06, + "loss": 0.6007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1513 + }, + { + "epoch": 1.8668310727496917, + "grad_norm": 2.4591483237884533, + "learning_rate": 6.496657070359289e-06, + "loss": 0.5649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1514 + }, + { + "epoch": 1.8680641183723798, + "grad_norm": 2.588461757043, + "learning_rate": 6.484312265361294e-06, + "loss": 0.6014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1515 + }, + { + "epoch": 1.8692971639950677, + "grad_norm": 2.3915380277510256, + "learning_rate": 6.47197357065427e-06, + "loss": 0.5105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1516 + }, + { + "epoch": 1.8705302096177558, + "grad_norm": 2.4853884668402597, + "learning_rate": 6.459641007682953e-06, + "loss": 0.5631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1517 + }, + { + "epoch": 1.871763255240444, + "grad_norm": 2.1904070749761546, + "learning_rate": 6.447314597881433e-06, + "loss": 0.4229, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1518 + }, + { + "epoch": 1.8729963008631318, + "grad_norm": 2.392397858425607, + "learning_rate": 6.4349943626731e-06, + "loss": 0.5323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1519 + }, + { + "epoch": 1.87422934648582, + "grad_norm": 2.5868871798442936, + "learning_rate": 6.422680323470609e-06, + "loss": 0.5943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1520 + }, + { + "epoch": 1.875462392108508, + "grad_norm": 2.8171109508645515, + "learning_rate": 6.410372501675854e-06, + "loss": 0.6011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1521 + }, + { + "epoch": 1.876695437731196, + "grad_norm": 2.5043723972276406, + "learning_rate": 6.398070918679917e-06, + "loss": 0.6401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1522 + }, + { + "epoch": 1.877928483353884, + "grad_norm": 2.2319121044157075, + "learning_rate": 6.3857755958630415e-06, + "loss": 0.5526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1523 + }, + { + "epoch": 1.8791615289765722, + "grad_norm": 2.668249861016908, + "learning_rate": 6.3734865545945834e-06, + "loss": 0.6208, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1524 + }, + { + "epoch": 1.88039457459926, + "grad_norm": 2.328498736662863, + "learning_rate": 6.361203816232988e-06, + "loss": 0.4526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1525 + }, + { + "epoch": 1.8816276202219482, + "grad_norm": 2.532493722492898, + "learning_rate": 6.348927402125753e-06, + "loss": 0.5397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1526 + }, + { + "epoch": 1.8828606658446363, + "grad_norm": 2.4124669695000023, + "learning_rate": 6.3366573336093605e-06, + "loss": 0.3821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1527 + }, + { + "epoch": 1.8840937114673242, + "grad_norm": 2.328411541073336, + "learning_rate": 6.324393632009295e-06, + "loss": 0.5093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1528 + }, + { + "epoch": 1.8853267570900123, + "grad_norm": 2.5365227172099196, + "learning_rate": 6.312136318639945e-06, + "loss": 0.5598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1529 + }, + { + "epoch": 1.8865598027127004, + "grad_norm": 2.607793778617426, + "learning_rate": 6.299885414804618e-06, + "loss": 0.5194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1530 + }, + { + "epoch": 1.8877928483353883, + "grad_norm": 2.335857326269118, + "learning_rate": 6.287640941795479e-06, + "loss": 0.3871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1531 + }, + { + "epoch": 1.8890258939580764, + "grad_norm": 2.647531529113461, + "learning_rate": 6.275402920893503e-06, + "loss": 0.521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1532 + }, + { + "epoch": 1.8902589395807645, + "grad_norm": 2.9804935503220285, + "learning_rate": 6.263171373368464e-06, + "loss": 0.5535, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1533 + }, + { + "epoch": 1.8914919852034524, + "grad_norm": 2.4613526931496215, + "learning_rate": 6.250946320478879e-06, + "loss": 0.3987, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1534 + }, + { + "epoch": 1.8927250308261405, + "grad_norm": 2.694644539489535, + "learning_rate": 6.238727783471983e-06, + "loss": 0.5107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1535 + }, + { + "epoch": 1.8939580764488286, + "grad_norm": 2.516795775084936, + "learning_rate": 6.2265157835836775e-06, + "loss": 0.3624, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1536 + }, + { + "epoch": 1.8951911220715165, + "grad_norm": 2.893994117314153, + "learning_rate": 6.214310342038507e-06, + "loss": 0.5791, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1537 + }, + { + "epoch": 1.8964241676942046, + "grad_norm": 2.773382488051452, + "learning_rate": 6.202111480049628e-06, + "loss": 0.6339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1538 + }, + { + "epoch": 1.8976572133168927, + "grad_norm": 2.797457073115434, + "learning_rate": 6.1899192188187385e-06, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1539 + }, + { + "epoch": 1.8988902589395806, + "grad_norm": 2.5101307384888343, + "learning_rate": 6.177733579536091e-06, + "loss": 0.4852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1540 + }, + { + "epoch": 1.900123304562269, + "grad_norm": 2.711866598306881, + "learning_rate": 6.165554583380401e-06, + "loss": 0.5412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1541 + }, + { + "epoch": 1.9013563501849569, + "grad_norm": 2.41897150751434, + "learning_rate": 6.1533822515188665e-06, + "loss": 0.4917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1542 + }, + { + "epoch": 1.9025893958076447, + "grad_norm": 2.4935923340546875, + "learning_rate": 6.141216605107077e-06, + "loss": 0.4712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1543 + }, + { + "epoch": 1.903822441430333, + "grad_norm": 2.3408793732304356, + "learning_rate": 6.129057665289024e-06, + "loss": 0.4915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1544 + }, + { + "epoch": 1.905055487053021, + "grad_norm": 2.49420888585378, + "learning_rate": 6.116905453197032e-06, + "loss": 0.4243, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1545 + }, + { + "epoch": 1.9062885326757089, + "grad_norm": 2.3882774651201517, + "learning_rate": 6.1047599899517305e-06, + "loss": 0.3567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1546 + }, + { + "epoch": 1.9075215782983972, + "grad_norm": 2.412746547072853, + "learning_rate": 6.092621296662028e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1547 + }, + { + "epoch": 1.908754623921085, + "grad_norm": 2.793611618535343, + "learning_rate": 6.080489394425058e-06, + "loss": 0.4794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1548 + }, + { + "epoch": 1.909987669543773, + "grad_norm": 2.86759252402197, + "learning_rate": 6.06836430432616e-06, + "loss": 0.5258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1549 + }, + { + "epoch": 1.9112207151664613, + "grad_norm": 3.5964778408959197, + "learning_rate": 6.056246047438823e-06, + "loss": 0.6431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1550 + }, + { + "epoch": 1.9124537607891492, + "grad_norm": 3.004291710626235, + "learning_rate": 6.044134644824671e-06, + "loss": 0.5851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1551 + }, + { + "epoch": 1.913686806411837, + "grad_norm": 2.553994453235498, + "learning_rate": 6.0320301175334096e-06, + "loss": 0.4947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1552 + }, + { + "epoch": 1.9149198520345254, + "grad_norm": 2.3305952965305723, + "learning_rate": 6.019932486602793e-06, + "loss": 0.4291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1553 + }, + { + "epoch": 1.9161528976572133, + "grad_norm": 2.70796232056951, + "learning_rate": 6.007841773058596e-06, + "loss": 0.5056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1554 + }, + { + "epoch": 1.9173859432799012, + "grad_norm": 2.401542735999577, + "learning_rate": 5.9957579979145595e-06, + "loss": 0.4256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1555 + }, + { + "epoch": 1.9186189889025895, + "grad_norm": 2.6832304024737783, + "learning_rate": 5.983681182172387e-06, + "loss": 0.6448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1556 + }, + { + "epoch": 1.9198520345252774, + "grad_norm": 2.5250444762365314, + "learning_rate": 5.971611346821657e-06, + "loss": 0.5107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1557 + }, + { + "epoch": 1.9210850801479655, + "grad_norm": 2.487361580061221, + "learning_rate": 5.959548512839839e-06, + "loss": 0.4747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1558 + }, + { + "epoch": 1.9223181257706536, + "grad_norm": 2.5305388152078963, + "learning_rate": 5.947492701192229e-06, + "loss": 0.4648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1559 + }, + { + "epoch": 1.9235511713933415, + "grad_norm": 2.455837488266591, + "learning_rate": 5.9354439328319105e-06, + "loss": 0.4423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1560 + }, + { + "epoch": 1.9247842170160296, + "grad_norm": 2.468256257896946, + "learning_rate": 5.923402228699736e-06, + "loss": 0.4569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1561 + }, + { + "epoch": 1.9260172626387178, + "grad_norm": 2.42939124280653, + "learning_rate": 5.91136760972427e-06, + "loss": 0.3932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1562 + }, + { + "epoch": 1.9272503082614056, + "grad_norm": 2.858568888655988, + "learning_rate": 5.89934009682177e-06, + "loss": 0.484, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1563 + }, + { + "epoch": 1.9284833538840938, + "grad_norm": 2.61347238372933, + "learning_rate": 5.887319710896144e-06, + "loss": 0.467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1564 + }, + { + "epoch": 1.9297163995067819, + "grad_norm": 2.7599324090566926, + "learning_rate": 5.875306472838908e-06, + "loss": 0.6094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1565 + }, + { + "epoch": 1.9309494451294698, + "grad_norm": 2.6103338708006505, + "learning_rate": 5.863300403529159e-06, + "loss": 0.4638, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1566 + }, + { + "epoch": 1.9321824907521579, + "grad_norm": 2.47982724255905, + "learning_rate": 5.85130152383353e-06, + "loss": 0.4826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1567 + }, + { + "epoch": 1.933415536374846, + "grad_norm": 2.578853751626434, + "learning_rate": 5.839309854606167e-06, + "loss": 0.5235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1568 + }, + { + "epoch": 1.9346485819975339, + "grad_norm": 2.4399405587351146, + "learning_rate": 5.827325416688671e-06, + "loss": 0.4686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1569 + }, + { + "epoch": 1.935881627620222, + "grad_norm": 2.5935511573767585, + "learning_rate": 5.81534823091009e-06, + "loss": 0.4941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1570 + }, + { + "epoch": 1.93711467324291, + "grad_norm": 2.3108330384724596, + "learning_rate": 5.803378318086859e-06, + "loss": 0.4216, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1571 + }, + { + "epoch": 1.938347718865598, + "grad_norm": 2.542118633939692, + "learning_rate": 5.791415699022768e-06, + "loss": 0.5643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1572 + }, + { + "epoch": 1.939580764488286, + "grad_norm": 2.6343709127133526, + "learning_rate": 5.779460394508944e-06, + "loss": 0.5811, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1573 + }, + { + "epoch": 1.9408138101109742, + "grad_norm": 2.476140626898723, + "learning_rate": 5.767512425323791e-06, + "loss": 0.4819, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1574 + }, + { + "epoch": 1.942046855733662, + "grad_norm": 2.6259462878787634, + "learning_rate": 5.755571812232967e-06, + "loss": 0.6072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1575 + }, + { + "epoch": 1.9432799013563502, + "grad_norm": 2.4454523827907804, + "learning_rate": 5.74363857598934e-06, + "loss": 0.4893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1576 + }, + { + "epoch": 1.9445129469790383, + "grad_norm": 2.5545835071067575, + "learning_rate": 5.731712737332965e-06, + "loss": 0.5415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1577 + }, + { + "epoch": 1.9457459926017262, + "grad_norm": 2.790586346292402, + "learning_rate": 5.7197943169910455e-06, + "loss": 0.619, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1578 + }, + { + "epoch": 1.9469790382244143, + "grad_norm": 2.5655789543184913, + "learning_rate": 5.707883335677867e-06, + "loss": 0.3993, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1579 + }, + { + "epoch": 1.9482120838471024, + "grad_norm": 2.3868281881615063, + "learning_rate": 5.695979814094814e-06, + "loss": 0.4513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1580 + }, + { + "epoch": 1.9494451294697903, + "grad_norm": 2.5119777174691684, + "learning_rate": 5.6840837729302865e-06, + "loss": 0.403, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1581 + }, + { + "epoch": 1.9506781750924784, + "grad_norm": 2.454297741118528, + "learning_rate": 5.672195232859703e-06, + "loss": 0.4341, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1582 + }, + { + "epoch": 1.9519112207151665, + "grad_norm": 2.173279017356807, + "learning_rate": 5.660314214545417e-06, + "loss": 0.4009, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1583 + }, + { + "epoch": 1.9531442663378544, + "grad_norm": 2.41693418545067, + "learning_rate": 5.648440738636731e-06, + "loss": 0.5145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1584 + }, + { + "epoch": 1.9543773119605425, + "grad_norm": 2.451502353773675, + "learning_rate": 5.63657482576984e-06, + "loss": 0.4148, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1585 + }, + { + "epoch": 1.9556103575832307, + "grad_norm": 2.5952977561456083, + "learning_rate": 5.624716496567782e-06, + "loss": 0.4059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1586 + }, + { + "epoch": 1.9568434032059185, + "grad_norm": 2.6119219402093647, + "learning_rate": 5.612865771640419e-06, + "loss": 0.4455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1587 + }, + { + "epoch": 1.9580764488286067, + "grad_norm": 2.5564882849476294, + "learning_rate": 5.601022671584394e-06, + "loss": 0.4633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1588 + }, + { + "epoch": 1.9593094944512948, + "grad_norm": 2.8188713009613244, + "learning_rate": 5.58918721698311e-06, + "loss": 0.565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1589 + }, + { + "epoch": 1.9605425400739827, + "grad_norm": 2.4162766103238895, + "learning_rate": 5.577359428406669e-06, + "loss": 0.5188, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1590 + }, + { + "epoch": 1.9617755856966708, + "grad_norm": 2.50589027882338, + "learning_rate": 5.56553932641185e-06, + "loss": 0.4671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1591 + }, + { + "epoch": 1.9630086313193589, + "grad_norm": 2.338580770590297, + "learning_rate": 5.553726931542085e-06, + "loss": 0.4271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1592 + }, + { + "epoch": 1.9642416769420468, + "grad_norm": 2.727731476438198, + "learning_rate": 5.541922264327395e-06, + "loss": 0.5796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1593 + }, + { + "epoch": 1.9654747225647349, + "grad_norm": 2.372879579889336, + "learning_rate": 5.530125345284388e-06, + "loss": 0.4791, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1594 + }, + { + "epoch": 1.966707768187423, + "grad_norm": 2.402358128750275, + "learning_rate": 5.5183361949161804e-06, + "loss": 0.5589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1595 + }, + { + "epoch": 1.9679408138101109, + "grad_norm": 2.521073091835758, + "learning_rate": 5.506554833712412e-06, + "loss": 0.5612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1596 + }, + { + "epoch": 1.969173859432799, + "grad_norm": 2.475424445052322, + "learning_rate": 5.494781282149175e-06, + "loss": 0.4164, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1597 + }, + { + "epoch": 1.970406905055487, + "grad_norm": 2.4762092606289388, + "learning_rate": 5.4830155606889855e-06, + "loss": 0.6031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1598 + }, + { + "epoch": 1.971639950678175, + "grad_norm": 2.3075398162125857, + "learning_rate": 5.471257689780753e-06, + "loss": 0.3457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1599 + }, + { + "epoch": 1.972872996300863, + "grad_norm": 2.8308395918813405, + "learning_rate": 5.459507689859742e-06, + "loss": 0.5926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1600 + }, + { + "epoch": 1.9741060419235512, + "grad_norm": 2.695737068994709, + "learning_rate": 5.447765581347542e-06, + "loss": 0.5015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1601 + }, + { + "epoch": 1.975339087546239, + "grad_norm": 2.88226095679846, + "learning_rate": 5.436031384652021e-06, + "loss": 0.5126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1602 + }, + { + "epoch": 1.9765721331689272, + "grad_norm": 2.466798154495963, + "learning_rate": 5.424305120167295e-06, + "loss": 0.3504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1603 + }, + { + "epoch": 1.9778051787916153, + "grad_norm": 2.999551007485252, + "learning_rate": 5.412586808273705e-06, + "loss": 0.6638, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1604 + }, + { + "epoch": 1.9790382244143032, + "grad_norm": 2.687785997003729, + "learning_rate": 5.400876469337759e-06, + "loss": 0.4778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1605 + }, + { + "epoch": 1.9802712700369913, + "grad_norm": 2.697985425334561, + "learning_rate": 5.389174123712113e-06, + "loss": 0.4376, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1606 + }, + { + "epoch": 1.9815043156596794, + "grad_norm": 2.2270783908633094, + "learning_rate": 5.377479791735522e-06, + "loss": 0.3055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1607 + }, + { + "epoch": 1.9827373612823673, + "grad_norm": 2.535575049691839, + "learning_rate": 5.3657934937328296e-06, + "loss": 0.4307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1608 + }, + { + "epoch": 1.9839704069050554, + "grad_norm": 2.6549657486777063, + "learning_rate": 5.354115250014904e-06, + "loss": 0.5848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1609 + }, + { + "epoch": 1.9852034525277436, + "grad_norm": 2.83855139958114, + "learning_rate": 5.342445080878614e-06, + "loss": 0.5668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1610 + }, + { + "epoch": 1.9864364981504314, + "grad_norm": 2.5321393689709284, + "learning_rate": 5.330783006606805e-06, + "loss": 0.3667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1611 + }, + { + "epoch": 1.9876695437731196, + "grad_norm": 3.1246050900975657, + "learning_rate": 5.319129047468244e-06, + "loss": 0.5487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1612 + }, + { + "epoch": 1.9889025893958077, + "grad_norm": 2.661845983771021, + "learning_rate": 5.307483223717598e-06, + "loss": 0.4961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1613 + }, + { + "epoch": 1.9901356350184956, + "grad_norm": 2.567032419410714, + "learning_rate": 5.295845555595396e-06, + "loss": 0.4989, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1614 + }, + { + "epoch": 1.9913686806411839, + "grad_norm": 2.715329364090932, + "learning_rate": 5.284216063327987e-06, + "loss": 0.55, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1615 + }, + { + "epoch": 1.9926017262638718, + "grad_norm": 2.4247843342398565, + "learning_rate": 5.27259476712751e-06, + "loss": 0.4278, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1616 + }, + { + "epoch": 1.9938347718865597, + "grad_norm": 2.586736358421805, + "learning_rate": 5.260981687191873e-06, + "loss": 0.3932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1617 + }, + { + "epoch": 1.995067817509248, + "grad_norm": 2.689699358030808, + "learning_rate": 5.249376843704686e-06, + "loss": 0.6152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1618 + }, + { + "epoch": 1.9963008631319359, + "grad_norm": 2.393722648406475, + "learning_rate": 5.2377802568352495e-06, + "loss": 0.4516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1619 + }, + { + "epoch": 1.9975339087546238, + "grad_norm": 2.151941518652342, + "learning_rate": 5.226191946738526e-06, + "loss": 0.4433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1620 + }, + { + "epoch": 1.998766954377312, + "grad_norm": 2.7987504690673988, + "learning_rate": 5.214611933555076e-06, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1621 + }, + { + "epoch": 2.0, + "grad_norm": 2.59776624758183, + "learning_rate": 5.20304023741105e-06, + "loss": 0.619, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1622 + }, + { + "epoch": 2.001233045622688, + "grad_norm": 2.43048324510223, + "learning_rate": 5.191476878418133e-06, + "loss": 0.4669, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1623 + }, + { + "epoch": 2.0024660912453762, + "grad_norm": 2.3352351017918402, + "learning_rate": 5.179921876673535e-06, + "loss": 0.4957, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1624 + }, + { + "epoch": 2.003699136868064, + "grad_norm": 2.395114246328322, + "learning_rate": 5.168375252259938e-06, + "loss": 0.4676, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1625 + }, + { + "epoch": 2.004932182490752, + "grad_norm": 2.2731002396640685, + "learning_rate": 5.156837025245447e-06, + "loss": 0.4332, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1626 + }, + { + "epoch": 2.0061652281134403, + "grad_norm": 2.7712994823928057, + "learning_rate": 5.145307215683595e-06, + "loss": 0.5389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1627 + }, + { + "epoch": 2.0073982737361282, + "grad_norm": 2.4056856515117993, + "learning_rate": 5.133785843613268e-06, + "loss": 0.4502, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1628 + }, + { + "epoch": 2.008631319358816, + "grad_norm": 2.2752234585471305, + "learning_rate": 5.122272929058701e-06, + "loss": 0.3659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1629 + }, + { + "epoch": 2.0098643649815044, + "grad_norm": 2.388137760527975, + "learning_rate": 5.110768492029421e-06, + "loss": 0.4334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1630 + }, + { + "epoch": 2.0110974106041923, + "grad_norm": 2.5706000208497217, + "learning_rate": 5.0992725525202205e-06, + "loss": 0.4949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1631 + }, + { + "epoch": 2.0123304562268802, + "grad_norm": 2.6004403792326274, + "learning_rate": 5.0877851305111334e-06, + "loss": 0.5269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1632 + }, + { + "epoch": 2.0135635018495686, + "grad_norm": 3.0312932980382468, + "learning_rate": 5.076306245967378e-06, + "loss": 0.5396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1633 + }, + { + "epoch": 2.0147965474722564, + "grad_norm": 2.439419139508058, + "learning_rate": 5.064835918839341e-06, + "loss": 0.3857, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1634 + }, + { + "epoch": 2.0160295930949443, + "grad_norm": 2.5177486175717974, + "learning_rate": 5.053374169062532e-06, + "loss": 0.4622, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1635 + }, + { + "epoch": 2.0172626387176327, + "grad_norm": 2.506250151038749, + "learning_rate": 5.041921016557556e-06, + "loss": 0.4383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1636 + }, + { + "epoch": 2.0184956843403206, + "grad_norm": 2.704760808949749, + "learning_rate": 5.030476481230086e-06, + "loss": 0.6166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1637 + }, + { + "epoch": 2.0197287299630085, + "grad_norm": 2.4695715254758337, + "learning_rate": 5.0190405829707935e-06, + "loss": 0.561, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1638 + }, + { + "epoch": 2.020961775585697, + "grad_norm": 2.7241602595851973, + "learning_rate": 5.007613341655363e-06, + "loss": 0.445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1639 + }, + { + "epoch": 2.0221948212083847, + "grad_norm": 2.3233960707374854, + "learning_rate": 4.996194777144416e-06, + "loss": 0.3948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1640 + }, + { + "epoch": 2.0234278668310726, + "grad_norm": 2.2442713678355823, + "learning_rate": 4.984784909283513e-06, + "loss": 0.3532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1641 + }, + { + "epoch": 2.024660912453761, + "grad_norm": 2.634392676529801, + "learning_rate": 4.973383757903072e-06, + "loss": 0.523, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1642 + }, + { + "epoch": 2.025893958076449, + "grad_norm": 2.8398388568834543, + "learning_rate": 4.961991342818385e-06, + "loss": 0.5234, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1643 + }, + { + "epoch": 2.0271270036991367, + "grad_norm": 2.6473433580844534, + "learning_rate": 4.950607683829557e-06, + "loss": 0.5612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1644 + }, + { + "epoch": 2.028360049321825, + "grad_norm": 2.523266270766798, + "learning_rate": 4.939232800721466e-06, + "loss": 0.5246, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1645 + }, + { + "epoch": 2.029593094944513, + "grad_norm": 2.5900711083492407, + "learning_rate": 4.927866713263744e-06, + "loss": 0.5103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1646 + }, + { + "epoch": 2.030826140567201, + "grad_norm": 2.463325250265772, + "learning_rate": 4.9165094412107275e-06, + "loss": 0.3759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1647 + }, + { + "epoch": 2.032059186189889, + "grad_norm": 2.494410303111952, + "learning_rate": 4.905161004301446e-06, + "loss": 0.4665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1648 + }, + { + "epoch": 2.033292231812577, + "grad_norm": 2.5651504375469987, + "learning_rate": 4.8938214222595636e-06, + "loss": 0.4896, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1649 + }, + { + "epoch": 2.034525277435265, + "grad_norm": 2.429966304010776, + "learning_rate": 4.882490714793352e-06, + "loss": 0.4589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1650 + }, + { + "epoch": 2.0357583230579532, + "grad_norm": 2.5231453655050653, + "learning_rate": 4.871168901595672e-06, + "loss": 0.5689, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1651 + }, + { + "epoch": 2.036991368680641, + "grad_norm": 2.6634619932788177, + "learning_rate": 4.859856002343913e-06, + "loss": 0.5838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1652 + }, + { + "epoch": 2.038224414303329, + "grad_norm": 2.5914302041981654, + "learning_rate": 4.8485520366999776e-06, + "loss": 0.4551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1653 + }, + { + "epoch": 2.0394574599260173, + "grad_norm": 2.22524078497228, + "learning_rate": 4.8372570243102345e-06, + "loss": 0.3869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1654 + }, + { + "epoch": 2.0406905055487052, + "grad_norm": 2.4814059043859804, + "learning_rate": 4.825970984805508e-06, + "loss": 0.4673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1655 + }, + { + "epoch": 2.041923551171393, + "grad_norm": 2.376833385977507, + "learning_rate": 4.814693937801008e-06, + "loss": 0.5249, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1656 + }, + { + "epoch": 2.0431565967940815, + "grad_norm": 2.522951679239216, + "learning_rate": 4.803425902896333e-06, + "loss": 0.4087, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1657 + }, + { + "epoch": 2.0443896424167693, + "grad_norm": 2.572178268061794, + "learning_rate": 4.792166899675407e-06, + "loss": 0.4922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1658 + }, + { + "epoch": 2.0456226880394572, + "grad_norm": 2.2642013364057383, + "learning_rate": 4.780916947706454e-06, + "loss": 0.3774, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1659 + }, + { + "epoch": 2.0468557336621456, + "grad_norm": 2.462040224339899, + "learning_rate": 4.769676066541981e-06, + "loss": 0.3919, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1660 + }, + { + "epoch": 2.0480887792848335, + "grad_norm": 2.5189042773825925, + "learning_rate": 4.75844427571872e-06, + "loss": 0.4466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1661 + }, + { + "epoch": 2.049321824907522, + "grad_norm": 2.6470599682008316, + "learning_rate": 4.747221594757598e-06, + "loss": 0.4843, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1662 + }, + { + "epoch": 2.0505548705302097, + "grad_norm": 2.543427335219999, + "learning_rate": 4.736008043163727e-06, + "loss": 0.5011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1663 + }, + { + "epoch": 2.0517879161528976, + "grad_norm": 2.832308169983458, + "learning_rate": 4.724803640426337e-06, + "loss": 0.4082, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1664 + }, + { + "epoch": 2.053020961775586, + "grad_norm": 2.448298632296988, + "learning_rate": 4.71360840601876e-06, + "loss": 0.3741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1665 + }, + { + "epoch": 2.054254007398274, + "grad_norm": 2.883720119682108, + "learning_rate": 4.702422359398394e-06, + "loss": 0.5057, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1666 + }, + { + "epoch": 2.0554870530209617, + "grad_norm": 2.392751477441128, + "learning_rate": 4.691245520006676e-06, + "loss": 0.351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1667 + }, + { + "epoch": 2.05672009864365, + "grad_norm": 2.2055087345825757, + "learning_rate": 4.680077907269032e-06, + "loss": 0.2998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1668 + }, + { + "epoch": 2.057953144266338, + "grad_norm": 2.473248669807766, + "learning_rate": 4.668919540594849e-06, + "loss": 0.5238, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1669 + }, + { + "epoch": 2.059186189889026, + "grad_norm": 2.3174240557182295, + "learning_rate": 4.657770439377458e-06, + "loss": 0.3478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1670 + }, + { + "epoch": 2.060419235511714, + "grad_norm": 2.2018582212758817, + "learning_rate": 4.646630622994072e-06, + "loss": 0.3582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1671 + }, + { + "epoch": 2.061652281134402, + "grad_norm": 2.537278150516378, + "learning_rate": 4.635500110805782e-06, + "loss": 0.442, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1672 + }, + { + "epoch": 2.06288532675709, + "grad_norm": 2.7734825628365094, + "learning_rate": 4.624378922157486e-06, + "loss": 0.5095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1673 + }, + { + "epoch": 2.0641183723797782, + "grad_norm": 2.5606426040553547, + "learning_rate": 4.6132670763779045e-06, + "loss": 0.3809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1674 + }, + { + "epoch": 2.065351418002466, + "grad_norm": 2.6820946526966045, + "learning_rate": 4.602164592779495e-06, + "loss": 0.4306, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1675 + }, + { + "epoch": 2.066584463625154, + "grad_norm": 2.5728870468376153, + "learning_rate": 4.591071490658465e-06, + "loss": 0.3495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1676 + }, + { + "epoch": 2.0678175092478424, + "grad_norm": 2.9950959525162113, + "learning_rate": 4.579987789294705e-06, + "loss": 0.5589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1677 + }, + { + "epoch": 2.0690505548705302, + "grad_norm": 2.6084748423929947, + "learning_rate": 4.56891350795176e-06, + "loss": 0.4674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1678 + }, + { + "epoch": 2.070283600493218, + "grad_norm": 2.558954194636351, + "learning_rate": 4.5578486658768236e-06, + "loss": 0.5383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1679 + }, + { + "epoch": 2.0715166461159065, + "grad_norm": 2.1592602256682083, + "learning_rate": 4.546793282300669e-06, + "loss": 0.3016, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1680 + }, + { + "epoch": 2.0727496917385944, + "grad_norm": 2.6206188628458915, + "learning_rate": 4.53574737643763e-06, + "loss": 0.3556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1681 + }, + { + "epoch": 2.0739827373612822, + "grad_norm": 2.5923118457963974, + "learning_rate": 4.524710967485571e-06, + "loss": 0.3413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1682 + }, + { + "epoch": 2.0752157829839706, + "grad_norm": 2.50571139289609, + "learning_rate": 4.513684074625853e-06, + "loss": 0.3758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1683 + }, + { + "epoch": 2.0764488286066585, + "grad_norm": 2.738099209536595, + "learning_rate": 4.5026667170233065e-06, + "loss": 0.5091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1684 + }, + { + "epoch": 2.0776818742293464, + "grad_norm": 2.8102966517380294, + "learning_rate": 4.491658913826161e-06, + "loss": 0.5011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1685 + }, + { + "epoch": 2.0789149198520347, + "grad_norm": 2.7957945330332117, + "learning_rate": 4.480660684166074e-06, + "loss": 0.3797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1686 + }, + { + "epoch": 2.0801479654747226, + "grad_norm": 3.0246754194205074, + "learning_rate": 4.469672047158039e-06, + "loss": 0.5499, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1687 + }, + { + "epoch": 2.0813810110974105, + "grad_norm": 2.7422346079651865, + "learning_rate": 4.458693021900399e-06, + "loss": 0.5348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1688 + }, + { + "epoch": 2.082614056720099, + "grad_norm": 2.6624901301648087, + "learning_rate": 4.4477236274747646e-06, + "loss": 0.4238, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1689 + }, + { + "epoch": 2.0838471023427867, + "grad_norm": 2.4655736338315104, + "learning_rate": 4.436763882946032e-06, + "loss": 0.4601, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1690 + }, + { + "epoch": 2.0850801479654746, + "grad_norm": 2.5201507017265534, + "learning_rate": 4.4258138073623216e-06, + "loss": 0.4972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1691 + }, + { + "epoch": 2.086313193588163, + "grad_norm": 2.375598788079959, + "learning_rate": 4.414873419754939e-06, + "loss": 0.4604, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1692 + }, + { + "epoch": 2.087546239210851, + "grad_norm": 2.6310260435763464, + "learning_rate": 4.403942739138359e-06, + "loss": 0.4896, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1693 + }, + { + "epoch": 2.0887792848335387, + "grad_norm": 2.7962256448727465, + "learning_rate": 4.393021784510182e-06, + "loss": 0.4518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1694 + }, + { + "epoch": 2.090012330456227, + "grad_norm": 2.5173184764135264, + "learning_rate": 4.382110574851112e-06, + "loss": 0.5738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1695 + }, + { + "epoch": 2.091245376078915, + "grad_norm": 2.3857792383387415, + "learning_rate": 4.371209129124908e-06, + "loss": 0.2941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1696 + }, + { + "epoch": 2.092478421701603, + "grad_norm": 2.3196034427524386, + "learning_rate": 4.36031746627836e-06, + "loss": 0.3398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1697 + }, + { + "epoch": 2.093711467324291, + "grad_norm": 2.674529578743137, + "learning_rate": 4.349435605241263e-06, + "loss": 0.5239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1698 + }, + { + "epoch": 2.094944512946979, + "grad_norm": 2.69239620455149, + "learning_rate": 4.338563564926368e-06, + "loss": 0.4752, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1699 + }, + { + "epoch": 2.096177558569667, + "grad_norm": 2.2696516232736665, + "learning_rate": 4.327701364229361e-06, + "loss": 0.3404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1700 + }, + { + "epoch": 2.0974106041923553, + "grad_norm": 2.94419875596226, + "learning_rate": 4.316849022028823e-06, + "loss": 0.5584, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1701 + }, + { + "epoch": 2.098643649815043, + "grad_norm": 2.63116353614959, + "learning_rate": 4.3060065571862065e-06, + "loss": 0.5006, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1702 + }, + { + "epoch": 2.099876695437731, + "grad_norm": 2.6481372890887016, + "learning_rate": 4.295173988545798e-06, + "loss": 0.4588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1703 + }, + { + "epoch": 2.1011097410604194, + "grad_norm": 2.654639166541817, + "learning_rate": 4.284351334934679e-06, + "loss": 0.4244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1704 + }, + { + "epoch": 2.1023427866831073, + "grad_norm": 2.580352134378634, + "learning_rate": 4.273538615162699e-06, + "loss": 0.4331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1705 + }, + { + "epoch": 2.103575832305795, + "grad_norm": 2.7091649413696732, + "learning_rate": 4.262735848022442e-06, + "loss": 0.6151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1706 + }, + { + "epoch": 2.1048088779284835, + "grad_norm": 2.5378400192827333, + "learning_rate": 4.2519430522892006e-06, + "loss": 0.3985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1707 + }, + { + "epoch": 2.1060419235511714, + "grad_norm": 2.5807185135055315, + "learning_rate": 4.241160246720931e-06, + "loss": 0.5335, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1708 + }, + { + "epoch": 2.1072749691738593, + "grad_norm": 2.5355635545223016, + "learning_rate": 4.230387450058225e-06, + "loss": 0.4811, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1709 + }, + { + "epoch": 2.1085080147965476, + "grad_norm": 2.6493461999852688, + "learning_rate": 4.219624681024288e-06, + "loss": 0.5149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1710 + }, + { + "epoch": 2.1097410604192355, + "grad_norm": 2.28405593070979, + "learning_rate": 4.2088719583248886e-06, + "loss": 0.3647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1711 + }, + { + "epoch": 2.1109741060419234, + "grad_norm": 2.6503235283708233, + "learning_rate": 4.198129300648336e-06, + "loss": 0.4812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1712 + }, + { + "epoch": 2.1122071516646117, + "grad_norm": 2.8066242646378274, + "learning_rate": 4.1873967266654444e-06, + "loss": 0.4478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1713 + }, + { + "epoch": 2.1134401972872996, + "grad_norm": 2.5301643915382, + "learning_rate": 4.1766742550295145e-06, + "loss": 0.3663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1714 + }, + { + "epoch": 2.1146732429099875, + "grad_norm": 2.6643383692422344, + "learning_rate": 4.165961904376275e-06, + "loss": 0.4954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1715 + }, + { + "epoch": 2.115906288532676, + "grad_norm": 2.687328638943086, + "learning_rate": 4.155259693323869e-06, + "loss": 0.5286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1716 + }, + { + "epoch": 2.1171393341553637, + "grad_norm": 2.3051275379782323, + "learning_rate": 4.144567640472819e-06, + "loss": 0.4043, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1717 + }, + { + "epoch": 2.1183723797780516, + "grad_norm": 2.6933982791775635, + "learning_rate": 4.13388576440599e-06, + "loss": 0.6243, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1718 + }, + { + "epoch": 2.11960542540074, + "grad_norm": 2.2088880651767173, + "learning_rate": 4.123214083688567e-06, + "loss": 0.3343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1719 + }, + { + "epoch": 2.120838471023428, + "grad_norm": 2.47172384136877, + "learning_rate": 4.112552616867997e-06, + "loss": 0.4237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1720 + }, + { + "epoch": 2.1220715166461157, + "grad_norm": 2.7748339806468176, + "learning_rate": 4.101901382473997e-06, + "loss": 0.5573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1721 + }, + { + "epoch": 2.123304562268804, + "grad_norm": 2.424501283224548, + "learning_rate": 4.091260399018484e-06, + "loss": 0.4265, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1722 + }, + { + "epoch": 2.124537607891492, + "grad_norm": 2.650174524506375, + "learning_rate": 4.08062968499557e-06, + "loss": 0.3477, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1723 + }, + { + "epoch": 2.12577065351418, + "grad_norm": 2.1938914180986817, + "learning_rate": 4.070009258881514e-06, + "loss": 0.2546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1724 + }, + { + "epoch": 2.127003699136868, + "grad_norm": 2.427206552496681, + "learning_rate": 4.059399139134687e-06, + "loss": 0.4614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1725 + }, + { + "epoch": 2.128236744759556, + "grad_norm": 2.600029453320743, + "learning_rate": 4.048799344195563e-06, + "loss": 0.5003, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1726 + }, + { + "epoch": 2.1294697903822444, + "grad_norm": 2.525899460269773, + "learning_rate": 4.038209892486659e-06, + "loss": 0.4207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1727 + }, + { + "epoch": 2.1307028360049323, + "grad_norm": 2.360301952652528, + "learning_rate": 4.027630802412517e-06, + "loss": 0.3161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1728 + }, + { + "epoch": 2.13193588162762, + "grad_norm": 2.3181572601045337, + "learning_rate": 4.017062092359678e-06, + "loss": 0.3167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1729 + }, + { + "epoch": 2.133168927250308, + "grad_norm": 2.7352618597650915, + "learning_rate": 4.006503780696632e-06, + "loss": 0.4445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1730 + }, + { + "epoch": 2.1344019728729964, + "grad_norm": 2.2207856211599877, + "learning_rate": 3.995955885773813e-06, + "loss": 0.3265, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1731 + }, + { + "epoch": 2.1356350184956843, + "grad_norm": 2.6621075014327413, + "learning_rate": 3.985418425923524e-06, + "loss": 0.4374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1732 + }, + { + "epoch": 2.1368680641183726, + "grad_norm": 2.5556232841368036, + "learning_rate": 3.974891419459958e-06, + "loss": 0.4089, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1733 + }, + { + "epoch": 2.1381011097410605, + "grad_norm": 2.6065282615267513, + "learning_rate": 3.964374884679121e-06, + "loss": 0.4171, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1734 + }, + { + "epoch": 2.1393341553637484, + "grad_norm": 2.373815079862734, + "learning_rate": 3.953868839858835e-06, + "loss": 0.3434, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1735 + }, + { + "epoch": 2.1405672009864363, + "grad_norm": 2.9153003666503645, + "learning_rate": 3.943373303258681e-06, + "loss": 0.4485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1736 + }, + { + "epoch": 2.1418002466091246, + "grad_norm": 2.6549134296354744, + "learning_rate": 3.932888293119973e-06, + "loss": 0.3891, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1737 + }, + { + "epoch": 2.1430332922318125, + "grad_norm": 2.910217909489204, + "learning_rate": 3.922413827665743e-06, + "loss": 0.4802, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1738 + }, + { + "epoch": 2.144266337854501, + "grad_norm": 2.9357454721384166, + "learning_rate": 3.911949925100685e-06, + "loss": 0.5186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1739 + }, + { + "epoch": 2.1454993834771887, + "grad_norm": 2.2618509660304387, + "learning_rate": 3.9014966036111354e-06, + "loss": 0.3406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1740 + }, + { + "epoch": 2.1467324290998766, + "grad_norm": 2.435460049819525, + "learning_rate": 3.891053881365043e-06, + "loss": 0.2776, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1741 + }, + { + "epoch": 2.147965474722565, + "grad_norm": 2.7650110493406896, + "learning_rate": 3.8806217765119424e-06, + "loss": 0.4867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1742 + }, + { + "epoch": 2.149198520345253, + "grad_norm": 2.5854743717015416, + "learning_rate": 3.870200307182903e-06, + "loss": 0.4048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1743 + }, + { + "epoch": 2.1504315659679407, + "grad_norm": 2.3719198940009254, + "learning_rate": 3.859789491490512e-06, + "loss": 0.4134, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1744 + }, + { + "epoch": 2.151664611590629, + "grad_norm": 2.868930160159961, + "learning_rate": 3.8493893475288505e-06, + "loss": 0.4194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1745 + }, + { + "epoch": 2.152897657213317, + "grad_norm": 2.6468667044836782, + "learning_rate": 3.838999893373436e-06, + "loss": 0.3691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1746 + }, + { + "epoch": 2.154130702836005, + "grad_norm": 2.604955712378186, + "learning_rate": 3.828621147081228e-06, + "loss": 0.3841, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1747 + }, + { + "epoch": 2.155363748458693, + "grad_norm": 2.6929978968474337, + "learning_rate": 3.818253126690549e-06, + "loss": 0.4759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1748 + }, + { + "epoch": 2.156596794081381, + "grad_norm": 2.616836044439998, + "learning_rate": 3.8078958502210993e-06, + "loss": 0.4427, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1749 + }, + { + "epoch": 2.157829839704069, + "grad_norm": 2.333588490179514, + "learning_rate": 3.797549335673908e-06, + "loss": 0.3661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1750 + }, + { + "epoch": 2.1590628853267573, + "grad_norm": 2.3306365994040337, + "learning_rate": 3.7872136010312865e-06, + "loss": 0.3125, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1751 + }, + { + "epoch": 2.160295930949445, + "grad_norm": 2.6439633169753463, + "learning_rate": 3.776888664256818e-06, + "loss": 0.3674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1752 + }, + { + "epoch": 2.161528976572133, + "grad_norm": 2.3752280229105494, + "learning_rate": 3.766574543295316e-06, + "loss": 0.3564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1753 + }, + { + "epoch": 2.1627620221948214, + "grad_norm": 2.3414215129379397, + "learning_rate": 3.756271256072803e-06, + "loss": 0.3079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1754 + }, + { + "epoch": 2.1639950678175093, + "grad_norm": 2.516659620108233, + "learning_rate": 3.7459788204964655e-06, + "loss": 0.4656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1755 + }, + { + "epoch": 2.165228113440197, + "grad_norm": 2.544764929665444, + "learning_rate": 3.735697254454627e-06, + "loss": 0.4453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1756 + }, + { + "epoch": 2.1664611590628855, + "grad_norm": 2.497553122385638, + "learning_rate": 3.725426575816732e-06, + "loss": 0.4223, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1757 + }, + { + "epoch": 2.1676942046855734, + "grad_norm": 2.278810104971583, + "learning_rate": 3.715166802433292e-06, + "loss": 0.3068, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1758 + }, + { + "epoch": 2.1689272503082613, + "grad_norm": 2.5107937194021446, + "learning_rate": 3.704917952135868e-06, + "loss": 0.3118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1759 + }, + { + "epoch": 2.1701602959309496, + "grad_norm": 2.7208144861944437, + "learning_rate": 3.6946800427370324e-06, + "loss": 0.3651, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1760 + }, + { + "epoch": 2.1713933415536375, + "grad_norm": 2.667315336204, + "learning_rate": 3.68445309203035e-06, + "loss": 0.4358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1761 + }, + { + "epoch": 2.1726263871763254, + "grad_norm": 2.512467743621635, + "learning_rate": 3.6742371177903433e-06, + "loss": 0.4263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1762 + }, + { + "epoch": 2.1738594327990137, + "grad_norm": 2.5887977603248773, + "learning_rate": 3.6640321377724364e-06, + "loss": 0.4037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1763 + }, + { + "epoch": 2.1750924784217016, + "grad_norm": 2.499047622832279, + "learning_rate": 3.65383816971297e-06, + "loss": 0.3269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1764 + }, + { + "epoch": 2.1763255240443895, + "grad_norm": 2.9728127490819247, + "learning_rate": 3.6436552313291264e-06, + "loss": 0.4693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1765 + }, + { + "epoch": 2.177558569667078, + "grad_norm": 2.9085704760310147, + "learning_rate": 3.633483340318935e-06, + "loss": 0.5007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1766 + }, + { + "epoch": 2.1787916152897657, + "grad_norm": 2.5551125679125106, + "learning_rate": 3.623322514361214e-06, + "loss": 0.3223, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1767 + }, + { + "epoch": 2.1800246609124536, + "grad_norm": 2.5216862821943584, + "learning_rate": 3.6131727711155484e-06, + "loss": 0.4088, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1768 + }, + { + "epoch": 2.181257706535142, + "grad_norm": 2.9196091577731518, + "learning_rate": 3.6030341282222735e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1769 + }, + { + "epoch": 2.18249075215783, + "grad_norm": 2.8463716887797625, + "learning_rate": 3.592906603302422e-06, + "loss": 0.5664, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1770 + }, + { + "epoch": 2.1837237977805177, + "grad_norm": 2.6993188839356352, + "learning_rate": 3.5827902139577044e-06, + "loss": 0.5555, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1771 + }, + { + "epoch": 2.184956843403206, + "grad_norm": 2.682693106322052, + "learning_rate": 3.5726849777704774e-06, + "loss": 0.4845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1772 + }, + { + "epoch": 2.186189889025894, + "grad_norm": 2.7032999448673274, + "learning_rate": 3.5625909123037215e-06, + "loss": 0.4596, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1773 + }, + { + "epoch": 2.187422934648582, + "grad_norm": 2.476182150321705, + "learning_rate": 3.5525080351009933e-06, + "loss": 0.4628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1774 + }, + { + "epoch": 2.18865598027127, + "grad_norm": 2.499458338772572, + "learning_rate": 3.5424363636864034e-06, + "loss": 0.438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1775 + }, + { + "epoch": 2.189889025893958, + "grad_norm": 2.3950343929132067, + "learning_rate": 3.5323759155645964e-06, + "loss": 0.3968, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1776 + }, + { + "epoch": 2.191122071516646, + "grad_norm": 2.7292045770887996, + "learning_rate": 3.522326708220699e-06, + "loss": 0.3672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1777 + }, + { + "epoch": 2.1923551171393343, + "grad_norm": 2.5822142468134293, + "learning_rate": 3.512288759120317e-06, + "loss": 0.4035, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1778 + }, + { + "epoch": 2.193588162762022, + "grad_norm": 2.4201356770012135, + "learning_rate": 3.5022620857094645e-06, + "loss": 0.5251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1779 + }, + { + "epoch": 2.19482120838471, + "grad_norm": 2.468897467978063, + "learning_rate": 3.4922467054145846e-06, + "loss": 0.4279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1780 + }, + { + "epoch": 2.1960542540073984, + "grad_norm": 2.567136894287482, + "learning_rate": 3.4822426356424754e-06, + "loss": 0.4641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1781 + }, + { + "epoch": 2.1972872996300863, + "grad_norm": 2.409074019271962, + "learning_rate": 3.472249893780287e-06, + "loss": 0.3393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1782 + }, + { + "epoch": 2.198520345252774, + "grad_norm": 2.5475663022568473, + "learning_rate": 3.4622684971954777e-06, + "loss": 0.4396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1783 + }, + { + "epoch": 2.1997533908754625, + "grad_norm": 2.654800541696009, + "learning_rate": 3.4522984632357826e-06, + "loss": 0.4128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1784 + }, + { + "epoch": 2.2009864364981504, + "grad_norm": 2.251545490229221, + "learning_rate": 3.4423398092292003e-06, + "loss": 0.3661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1785 + }, + { + "epoch": 2.2022194821208383, + "grad_norm": 2.560355670972336, + "learning_rate": 3.432392552483941e-06, + "loss": 0.506, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1786 + }, + { + "epoch": 2.2034525277435266, + "grad_norm": 2.957755450148777, + "learning_rate": 3.4224567102884098e-06, + "loss": 0.4471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1787 + }, + { + "epoch": 2.2046855733662145, + "grad_norm": 2.727694808021836, + "learning_rate": 3.4125322999111698e-06, + "loss": 0.4985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1788 + }, + { + "epoch": 2.2059186189889024, + "grad_norm": 2.5078793281615646, + "learning_rate": 3.4026193386009244e-06, + "loss": 0.3459, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1789 + }, + { + "epoch": 2.2071516646115907, + "grad_norm": 2.4663933369954356, + "learning_rate": 3.392717843586472e-06, + "loss": 0.2747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1790 + }, + { + "epoch": 2.2083847102342786, + "grad_norm": 2.336841544925504, + "learning_rate": 3.382827832076676e-06, + "loss": 0.3025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1791 + }, + { + "epoch": 2.2096177558569665, + "grad_norm": 2.5259006303278353, + "learning_rate": 3.3729493212604593e-06, + "loss": 0.3807, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1792 + }, + { + "epoch": 2.210850801479655, + "grad_norm": 2.4611507324991355, + "learning_rate": 3.363082328306737e-06, + "loss": 0.3556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1793 + }, + { + "epoch": 2.2120838471023427, + "grad_norm": 2.6218201591114174, + "learning_rate": 3.353226870364422e-06, + "loss": 0.3614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1794 + }, + { + "epoch": 2.2133168927250306, + "grad_norm": 2.823653063437215, + "learning_rate": 3.343382964562368e-06, + "loss": 0.4235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1795 + }, + { + "epoch": 2.214549938347719, + "grad_norm": 2.5370302947596524, + "learning_rate": 3.3335506280093544e-06, + "loss": 0.3716, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1796 + }, + { + "epoch": 2.215782983970407, + "grad_norm": 2.6056377036330183, + "learning_rate": 3.3237298777940586e-06, + "loss": 0.4282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1797 + }, + { + "epoch": 2.2170160295930947, + "grad_norm": 2.4707774591274014, + "learning_rate": 3.313920730985013e-06, + "loss": 0.3865, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1798 + }, + { + "epoch": 2.218249075215783, + "grad_norm": 2.6353955376812133, + "learning_rate": 3.304123204630587e-06, + "loss": 0.4042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1799 + }, + { + "epoch": 2.219482120838471, + "grad_norm": 2.9175053608530988, + "learning_rate": 3.29433731575895e-06, + "loss": 0.4431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1800 + }, + { + "epoch": 2.220715166461159, + "grad_norm": 2.4251950790112655, + "learning_rate": 3.2845630813780538e-06, + "loss": 0.3509, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1801 + }, + { + "epoch": 2.221948212083847, + "grad_norm": 2.6971411900024203, + "learning_rate": 3.274800518475586e-06, + "loss": 0.3719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1802 + }, + { + "epoch": 2.223181257706535, + "grad_norm": 3.083984775257449, + "learning_rate": 3.2650496440189504e-06, + "loss": 0.4078, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1803 + }, + { + "epoch": 2.2244143033292234, + "grad_norm": 2.668941643126659, + "learning_rate": 3.255310474955243e-06, + "loss": 0.4439, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1804 + }, + { + "epoch": 2.2256473489519113, + "grad_norm": 2.976039030354018, + "learning_rate": 3.2455830282112087e-06, + "loss": 0.4573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1805 + }, + { + "epoch": 2.226880394574599, + "grad_norm": 2.6167153610229126, + "learning_rate": 3.23586732069322e-06, + "loss": 0.4224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1806 + }, + { + "epoch": 2.228113440197287, + "grad_norm": 2.734527135614152, + "learning_rate": 3.226163369287245e-06, + "loss": 0.3709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1807 + }, + { + "epoch": 2.2293464858199754, + "grad_norm": 3.0042658435735454, + "learning_rate": 3.2164711908588253e-06, + "loss": 0.5491, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1808 + }, + { + "epoch": 2.2305795314426633, + "grad_norm": 2.7454305809971693, + "learning_rate": 3.2067908022530446e-06, + "loss": 0.4983, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1809 + }, + { + "epoch": 2.2318125770653516, + "grad_norm": 2.5164374780175436, + "learning_rate": 3.1971222202944752e-06, + "loss": 0.402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1810 + }, + { + "epoch": 2.2330456226880395, + "grad_norm": 2.5784808191967095, + "learning_rate": 3.187465461787194e-06, + "loss": 0.3722, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1811 + }, + { + "epoch": 2.2342786683107274, + "grad_norm": 2.6856695561609323, + "learning_rate": 3.1778205435147093e-06, + "loss": 0.4127, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1812 + }, + { + "epoch": 2.2355117139334153, + "grad_norm": 2.6325754846210727, + "learning_rate": 3.1681874822399684e-06, + "loss": 0.3554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1813 + }, + { + "epoch": 2.2367447595561036, + "grad_norm": 2.471040398086649, + "learning_rate": 3.1585662947052977e-06, + "loss": 0.4473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1814 + }, + { + "epoch": 2.2379778051787915, + "grad_norm": 2.4419144129307924, + "learning_rate": 3.1489569976323885e-06, + "loss": 0.3648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1815 + }, + { + "epoch": 2.23921085080148, + "grad_norm": 2.551971628620259, + "learning_rate": 3.1393596077222752e-06, + "loss": 0.3954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1816 + }, + { + "epoch": 2.2404438964241677, + "grad_norm": 2.7230280726188942, + "learning_rate": 3.1297741416552895e-06, + "loss": 0.4502, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1817 + }, + { + "epoch": 2.2416769420468556, + "grad_norm": 2.5986833977870774, + "learning_rate": 3.1202006160910403e-06, + "loss": 0.5191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1818 + }, + { + "epoch": 2.242909987669544, + "grad_norm": 2.7125045113343034, + "learning_rate": 3.1106390476683833e-06, + "loss": 0.5319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1819 + }, + { + "epoch": 2.244143033292232, + "grad_norm": 2.4605911006490726, + "learning_rate": 3.101089453005399e-06, + "loss": 0.3073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1820 + }, + { + "epoch": 2.2453760789149197, + "grad_norm": 2.786082762868325, + "learning_rate": 3.09155184869935e-06, + "loss": 0.5374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1821 + }, + { + "epoch": 2.246609124537608, + "grad_norm": 3.0739075984030095, + "learning_rate": 3.08202625132666e-06, + "loss": 0.5128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1822 + }, + { + "epoch": 2.247842170160296, + "grad_norm": 2.400418914937858, + "learning_rate": 3.0725126774428915e-06, + "loss": 0.3134, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1823 + }, + { + "epoch": 2.249075215782984, + "grad_norm": 2.45119216482281, + "learning_rate": 3.0630111435827015e-06, + "loss": 0.3013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1824 + }, + { + "epoch": 2.250308261405672, + "grad_norm": 2.9921352173355182, + "learning_rate": 3.053521666259832e-06, + "loss": 0.5281, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1825 + }, + { + "epoch": 2.25154130702836, + "grad_norm": 2.2410327144875986, + "learning_rate": 3.044044261967053e-06, + "loss": 0.2617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1826 + }, + { + "epoch": 2.252774352651048, + "grad_norm": 2.6733315687315415, + "learning_rate": 3.0345789471761677e-06, + "loss": 0.4527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1827 + }, + { + "epoch": 2.2540073982737363, + "grad_norm": 2.9860637695259538, + "learning_rate": 3.0251257383379675e-06, + "loss": 0.5013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1828 + }, + { + "epoch": 2.255240443896424, + "grad_norm": 2.5671494331238476, + "learning_rate": 3.0156846518821925e-06, + "loss": 0.3735, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1829 + }, + { + "epoch": 2.256473489519112, + "grad_norm": 2.680154991425641, + "learning_rate": 3.006255704217522e-06, + "loss": 0.3478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1830 + }, + { + "epoch": 2.2577065351418004, + "grad_norm": 2.4302954673272894, + "learning_rate": 2.996838911731532e-06, + "loss": 0.2961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1831 + }, + { + "epoch": 2.2589395807644883, + "grad_norm": 2.3839702852195344, + "learning_rate": 2.987434290790683e-06, + "loss": 0.2412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1832 + }, + { + "epoch": 2.260172626387176, + "grad_norm": 2.3513764104654076, + "learning_rate": 2.9780418577402727e-06, + "loss": 0.4039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1833 + }, + { + "epoch": 2.2614056720098645, + "grad_norm": 2.5365870091486844, + "learning_rate": 2.9686616289044133e-06, + "loss": 0.4231, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1834 + }, + { + "epoch": 2.2626387176325524, + "grad_norm": 2.6943065218303346, + "learning_rate": 2.959293620586018e-06, + "loss": 0.419, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1835 + }, + { + "epoch": 2.2638717632552403, + "grad_norm": 2.67535698552442, + "learning_rate": 2.9499378490667486e-06, + "loss": 0.3214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1836 + }, + { + "epoch": 2.2651048088779286, + "grad_norm": 2.648573827419777, + "learning_rate": 2.940594330607013e-06, + "loss": 0.3894, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1837 + }, + { + "epoch": 2.2663378545006165, + "grad_norm": 2.718965497393186, + "learning_rate": 2.931263081445902e-06, + "loss": 0.3538, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1838 + }, + { + "epoch": 2.2675709001233044, + "grad_norm": 2.8944057958849636, + "learning_rate": 2.921944117801203e-06, + "loss": 0.3672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1839 + }, + { + "epoch": 2.2688039457459928, + "grad_norm": 3.275410695882557, + "learning_rate": 2.9126374558693394e-06, + "loss": 0.506, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1840 + }, + { + "epoch": 2.2700369913686806, + "grad_norm": 2.794818093317088, + "learning_rate": 2.903343111825362e-06, + "loss": 0.4013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1841 + }, + { + "epoch": 2.2712700369913685, + "grad_norm": 2.8408389229850255, + "learning_rate": 2.894061101822905e-06, + "loss": 0.4814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1842 + }, + { + "epoch": 2.272503082614057, + "grad_norm": 2.8213904379609707, + "learning_rate": 2.884791441994168e-06, + "loss": 0.4325, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1843 + }, + { + "epoch": 2.2737361282367448, + "grad_norm": 2.318478307851654, + "learning_rate": 2.8755341484498923e-06, + "loss": 0.2796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1844 + }, + { + "epoch": 2.2749691738594326, + "grad_norm": 2.7101825130130712, + "learning_rate": 2.8662892372793203e-06, + "loss": 0.368, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1845 + }, + { + "epoch": 2.276202219482121, + "grad_norm": 2.232410112947641, + "learning_rate": 2.8570567245501736e-06, + "loss": 0.2401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1846 + }, + { + "epoch": 2.277435265104809, + "grad_norm": 2.7730682747492046, + "learning_rate": 2.8478366263086242e-06, + "loss": 0.458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1847 + }, + { + "epoch": 2.2786683107274968, + "grad_norm": 2.7088179960724026, + "learning_rate": 2.838628958579277e-06, + "loss": 0.3906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1848 + }, + { + "epoch": 2.279901356350185, + "grad_norm": 2.642992601836881, + "learning_rate": 2.8294337373651227e-06, + "loss": 0.4505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1849 + }, + { + "epoch": 2.281134401972873, + "grad_norm": 2.566770724267774, + "learning_rate": 2.8202509786475183e-06, + "loss": 0.4041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1850 + }, + { + "epoch": 2.282367447595561, + "grad_norm": 2.8325370719966334, + "learning_rate": 2.8110806983861726e-06, + "loss": 0.3838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1851 + }, + { + "epoch": 2.283600493218249, + "grad_norm": 2.7305618513803034, + "learning_rate": 2.801922912519098e-06, + "loss": 0.4741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1852 + }, + { + "epoch": 2.284833538840937, + "grad_norm": 2.7167758355736717, + "learning_rate": 2.7927776369625915e-06, + "loss": 0.4109, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1853 + }, + { + "epoch": 2.286066584463625, + "grad_norm": 2.1944056474153566, + "learning_rate": 2.7836448876112065e-06, + "loss": 0.2787, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1854 + }, + { + "epoch": 2.2872996300863133, + "grad_norm": 2.206944335103158, + "learning_rate": 2.7745246803377303e-06, + "loss": 0.3027, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1855 + }, + { + "epoch": 2.288532675709001, + "grad_norm": 2.3467437676570677, + "learning_rate": 2.7654170309931537e-06, + "loss": 0.3316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1856 + }, + { + "epoch": 2.289765721331689, + "grad_norm": 2.8197181669039253, + "learning_rate": 2.7563219554066346e-06, + "loss": 0.4353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1857 + }, + { + "epoch": 2.2909987669543774, + "grad_norm": 2.6463694243329297, + "learning_rate": 2.747239469385481e-06, + "loss": 0.4461, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1858 + }, + { + "epoch": 2.2922318125770653, + "grad_norm": 2.5693847166016495, + "learning_rate": 2.738169588715116e-06, + "loss": 0.4154, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1859 + }, + { + "epoch": 2.293464858199753, + "grad_norm": 2.9084887582207224, + "learning_rate": 2.729112329159066e-06, + "loss": 0.4642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1860 + }, + { + "epoch": 2.2946979038224415, + "grad_norm": 2.661155216526391, + "learning_rate": 2.720067706458911e-06, + "loss": 0.4911, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1861 + }, + { + "epoch": 2.2959309494451294, + "grad_norm": 2.249438410721065, + "learning_rate": 2.7110357363342667e-06, + "loss": 0.2955, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1862 + }, + { + "epoch": 2.2971639950678173, + "grad_norm": 2.4248296466556605, + "learning_rate": 2.7020164344827705e-06, + "loss": 0.3078, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1863 + }, + { + "epoch": 2.2983970406905057, + "grad_norm": 2.8178210324627604, + "learning_rate": 2.6930098165800324e-06, + "loss": 0.5657, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1864 + }, + { + "epoch": 2.2996300863131935, + "grad_norm": 2.1492514900867867, + "learning_rate": 2.684015898279618e-06, + "loss": 0.2542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1865 + }, + { + "epoch": 2.3008631319358814, + "grad_norm": 2.5767111705244097, + "learning_rate": 2.6750346952130224e-06, + "loss": 0.2911, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1866 + }, + { + "epoch": 2.3020961775585698, + "grad_norm": 2.243951773213573, + "learning_rate": 2.666066222989644e-06, + "loss": 0.2932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1867 + }, + { + "epoch": 2.3033292231812577, + "grad_norm": 2.3067727425907587, + "learning_rate": 2.65711049719676e-06, + "loss": 0.2374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1868 + }, + { + "epoch": 2.304562268803946, + "grad_norm": 2.415737071109258, + "learning_rate": 2.6481675333994728e-06, + "loss": 0.3661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1869 + }, + { + "epoch": 2.305795314426634, + "grad_norm": 2.296552712112862, + "learning_rate": 2.639237347140731e-06, + "loss": 0.2565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1870 + }, + { + "epoch": 2.3070283600493218, + "grad_norm": 2.5905345710770766, + "learning_rate": 2.630319953941255e-06, + "loss": 0.2994, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1871 + }, + { + "epoch": 2.3082614056720097, + "grad_norm": 2.9421711380538755, + "learning_rate": 2.6214153692995504e-06, + "loss": 0.3917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1872 + }, + { + "epoch": 2.309494451294698, + "grad_norm": 2.24918012711347, + "learning_rate": 2.612523608691837e-06, + "loss": 0.29, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1873 + }, + { + "epoch": 2.310727496917386, + "grad_norm": 2.6468263977498396, + "learning_rate": 2.6036446875720657e-06, + "loss": 0.3579, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1874 + }, + { + "epoch": 2.311960542540074, + "grad_norm": 2.9869717640588616, + "learning_rate": 2.5947786213718695e-06, + "loss": 0.4606, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1875 + }, + { + "epoch": 2.313193588162762, + "grad_norm": 2.7765364835681083, + "learning_rate": 2.585925425500534e-06, + "loss": 0.4072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1876 + }, + { + "epoch": 2.31442663378545, + "grad_norm": 2.5297254099675426, + "learning_rate": 2.5770851153449785e-06, + "loss": 0.3614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1877 + }, + { + "epoch": 2.315659679408138, + "grad_norm": 2.7275298517849924, + "learning_rate": 2.5682577062697213e-06, + "loss": 0.4544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1878 + }, + { + "epoch": 2.316892725030826, + "grad_norm": 3.1515080991733235, + "learning_rate": 2.559443213616871e-06, + "loss": 0.49, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1879 + }, + { + "epoch": 2.318125770653514, + "grad_norm": 3.1665702629969315, + "learning_rate": 2.5506416527060796e-06, + "loss": 0.5849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1880 + }, + { + "epoch": 2.3193588162762024, + "grad_norm": 2.5623146927718254, + "learning_rate": 2.5418530388345175e-06, + "loss": 0.3085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1881 + }, + { + "epoch": 2.3205918618988903, + "grad_norm": 2.46776926089478, + "learning_rate": 2.533077387276869e-06, + "loss": 0.2888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1882 + }, + { + "epoch": 2.321824907521578, + "grad_norm": 2.775069601858482, + "learning_rate": 2.524314713285272e-06, + "loss": 0.3817, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1883 + }, + { + "epoch": 2.323057953144266, + "grad_norm": 2.670954218210571, + "learning_rate": 2.515565032089329e-06, + "loss": 0.4478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1884 + }, + { + "epoch": 2.3242909987669544, + "grad_norm": 2.5238477153530665, + "learning_rate": 2.5068283588960372e-06, + "loss": 0.4481, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1885 + }, + { + "epoch": 2.3255240443896423, + "grad_norm": 2.6849323631816295, + "learning_rate": 2.498104708889807e-06, + "loss": 0.4133, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1886 + }, + { + "epoch": 2.3267570900123307, + "grad_norm": 2.4377081666404075, + "learning_rate": 2.4893940972323995e-06, + "loss": 0.332, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1887 + }, + { + "epoch": 2.3279901356350186, + "grad_norm": 2.7114394525033063, + "learning_rate": 2.480696539062928e-06, + "loss": 0.4282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1888 + }, + { + "epoch": 2.3292231812577064, + "grad_norm": 2.603372665925758, + "learning_rate": 2.4720120494978095e-06, + "loss": 0.3697, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1889 + }, + { + "epoch": 2.3304562268803943, + "grad_norm": 2.805458776057501, + "learning_rate": 2.4633406436307473e-06, + "loss": 0.326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1890 + }, + { + "epoch": 2.3316892725030827, + "grad_norm": 3.008639998561586, + "learning_rate": 2.4546823365327133e-06, + "loss": 0.479, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1891 + }, + { + "epoch": 2.3329223181257706, + "grad_norm": 2.848060710470996, + "learning_rate": 2.446037143251907e-06, + "loss": 0.4492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1892 + }, + { + "epoch": 2.334155363748459, + "grad_norm": 2.6972361605525172, + "learning_rate": 2.437405078813733e-06, + "loss": 0.4651, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1893 + }, + { + "epoch": 2.335388409371147, + "grad_norm": 2.6444772328432764, + "learning_rate": 2.428786158220787e-06, + "loss": 0.3531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1894 + }, + { + "epoch": 2.3366214549938347, + "grad_norm": 2.825025155857256, + "learning_rate": 2.420180396452817e-06, + "loss": 0.4398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1895 + }, + { + "epoch": 2.337854500616523, + "grad_norm": 2.847774141381515, + "learning_rate": 2.4115878084666967e-06, + "loss": 0.4774, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1896 + }, + { + "epoch": 2.339087546239211, + "grad_norm": 2.7240100074211626, + "learning_rate": 2.4030084091964058e-06, + "loss": 0.5387, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1897 + }, + { + "epoch": 2.340320591861899, + "grad_norm": 2.6596947077013238, + "learning_rate": 2.3944422135530067e-06, + "loss": 0.3699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1898 + }, + { + "epoch": 2.341553637484587, + "grad_norm": 2.3195986093996828, + "learning_rate": 2.385889236424609e-06, + "loss": 0.2357, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1899 + }, + { + "epoch": 2.342786683107275, + "grad_norm": 3.138152832229155, + "learning_rate": 2.3773494926763453e-06, + "loss": 0.5716, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1900 + }, + { + "epoch": 2.344019728729963, + "grad_norm": 2.4353439616422317, + "learning_rate": 2.3688229971503585e-06, + "loss": 0.307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1901 + }, + { + "epoch": 2.3452527743526512, + "grad_norm": 2.7479264877441305, + "learning_rate": 2.3603097646657547e-06, + "loss": 0.4065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1902 + }, + { + "epoch": 2.346485819975339, + "grad_norm": 2.7608377722996464, + "learning_rate": 2.3518098100186003e-06, + "loss": 0.4474, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1903 + }, + { + "epoch": 2.347718865598027, + "grad_norm": 2.8009720711063983, + "learning_rate": 2.3433231479818754e-06, + "loss": 0.3542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1904 + }, + { + "epoch": 2.3489519112207153, + "grad_norm": 2.6450074204233727, + "learning_rate": 2.3348497933054593e-06, + "loss": 0.2626, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1905 + }, + { + "epoch": 2.3501849568434032, + "grad_norm": 2.718286299118251, + "learning_rate": 2.3263897607161045e-06, + "loss": 0.3783, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1906 + }, + { + "epoch": 2.351418002466091, + "grad_norm": 3.1635623479202004, + "learning_rate": 2.317943064917412e-06, + "loss": 0.458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1907 + }, + { + "epoch": 2.3526510480887795, + "grad_norm": 3.023923451231218, + "learning_rate": 2.3095097205898e-06, + "loss": 0.3445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1908 + }, + { + "epoch": 2.3538840937114673, + "grad_norm": 2.7560067491256754, + "learning_rate": 2.301089742390479e-06, + "loss": 0.3877, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1909 + }, + { + "epoch": 2.3551171393341552, + "grad_norm": 2.4987053182394185, + "learning_rate": 2.292683144953438e-06, + "loss": 0.2485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1910 + }, + { + "epoch": 2.3563501849568436, + "grad_norm": 3.028553166873761, + "learning_rate": 2.284289942889403e-06, + "loss": 0.5015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1911 + }, + { + "epoch": 2.3575832305795315, + "grad_norm": 2.6107906939757033, + "learning_rate": 2.275910150785818e-06, + "loss": 0.356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1912 + }, + { + "epoch": 2.3588162762022193, + "grad_norm": 2.4846901180876872, + "learning_rate": 2.267543783206824e-06, + "loss": 0.2767, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1913 + }, + { + "epoch": 2.3600493218249077, + "grad_norm": 2.924247536298395, + "learning_rate": 2.2591908546932283e-06, + "loss": 0.4083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1914 + }, + { + "epoch": 2.3612823674475956, + "grad_norm": 2.715130315338188, + "learning_rate": 2.2508513797624887e-06, + "loss": 0.4079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1915 + }, + { + "epoch": 2.3625154130702835, + "grad_norm": 2.261533556908912, + "learning_rate": 2.2425253729086625e-06, + "loss": 0.2094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1916 + }, + { + "epoch": 2.363748458692972, + "grad_norm": 2.9340455954588918, + "learning_rate": 2.2342128486024193e-06, + "loss": 0.5613, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1917 + }, + { + "epoch": 2.3649815043156597, + "grad_norm": 2.759033032828826, + "learning_rate": 2.22591382129098e-06, + "loss": 0.4658, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1918 + }, + { + "epoch": 2.3662145499383476, + "grad_norm": 2.9330784187057812, + "learning_rate": 2.217628305398122e-06, + "loss": 0.4703, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1919 + }, + { + "epoch": 2.367447595561036, + "grad_norm": 2.926964264297683, + "learning_rate": 2.209356315324128e-06, + "loss": 0.3734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1920 + }, + { + "epoch": 2.368680641183724, + "grad_norm": 2.7140676889542674, + "learning_rate": 2.201097865445776e-06, + "loss": 0.4358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1921 + }, + { + "epoch": 2.3699136868064117, + "grad_norm": 2.3950079044896504, + "learning_rate": 2.192852970116316e-06, + "loss": 0.3264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1922 + }, + { + "epoch": 2.3711467324291, + "grad_norm": 2.4698900190893847, + "learning_rate": 2.1846216436654342e-06, + "loss": 0.2551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1923 + }, + { + "epoch": 2.372379778051788, + "grad_norm": 2.57908607952626, + "learning_rate": 2.1764039003992377e-06, + "loss": 0.2718, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1924 + }, + { + "epoch": 2.373612823674476, + "grad_norm": 2.7443866869071747, + "learning_rate": 2.16819975460022e-06, + "loss": 0.3698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1925 + }, + { + "epoch": 2.374845869297164, + "grad_norm": 2.4541746426775424, + "learning_rate": 2.1600092205272493e-06, + "loss": 0.3114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1926 + }, + { + "epoch": 2.376078914919852, + "grad_norm": 2.6764807552617094, + "learning_rate": 2.1518323124155393e-06, + "loss": 0.3946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1927 + }, + { + "epoch": 2.37731196054254, + "grad_norm": 2.232502294129531, + "learning_rate": 2.143669044476605e-06, + "loss": 0.2064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1928 + }, + { + "epoch": 2.3785450061652282, + "grad_norm": 2.465612079035968, + "learning_rate": 2.135519430898274e-06, + "loss": 0.2784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1929 + }, + { + "epoch": 2.379778051787916, + "grad_norm": 2.827852473582675, + "learning_rate": 2.1273834858446285e-06, + "loss": 0.3079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1930 + }, + { + "epoch": 2.381011097410604, + "grad_norm": 2.633449394986566, + "learning_rate": 2.1192612234560093e-06, + "loss": 0.3227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1931 + }, + { + "epoch": 2.3822441430332923, + "grad_norm": 2.7760612390178325, + "learning_rate": 2.1111526578489573e-06, + "loss": 0.3875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1932 + }, + { + "epoch": 2.3834771886559802, + "grad_norm": 2.543518865851238, + "learning_rate": 2.103057803116223e-06, + "loss": 0.2956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1933 + }, + { + "epoch": 2.384710234278668, + "grad_norm": 2.5903161811570716, + "learning_rate": 2.0949766733267263e-06, + "loss": 0.3226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1934 + }, + { + "epoch": 2.3859432799013565, + "grad_norm": 2.6196746717343307, + "learning_rate": 2.0869092825255266e-06, + "loss": 0.3737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1935 + }, + { + "epoch": 2.3871763255240444, + "grad_norm": 2.4200712742066726, + "learning_rate": 2.078855644733808e-06, + "loss": 0.4331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1936 + }, + { + "epoch": 2.3884093711467322, + "grad_norm": 2.97992031816307, + "learning_rate": 2.070815773948851e-06, + "loss": 0.4552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1937 + }, + { + "epoch": 2.3896424167694206, + "grad_norm": 2.868889423356795, + "learning_rate": 2.062789684144013e-06, + "loss": 0.3954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1938 + }, + { + "epoch": 2.3908754623921085, + "grad_norm": 2.587694442487643, + "learning_rate": 2.054777389268695e-06, + "loss": 0.4605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1939 + }, + { + "epoch": 2.392108508014797, + "grad_norm": 2.8211137836092632, + "learning_rate": 2.046778903248321e-06, + "loss": 0.3973, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1940 + }, + { + "epoch": 2.3933415536374847, + "grad_norm": 2.468677079748375, + "learning_rate": 2.038794239984324e-06, + "loss": 0.3423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1941 + }, + { + "epoch": 2.3945745992601726, + "grad_norm": 2.5859956743331343, + "learning_rate": 2.0308234133541036e-06, + "loss": 0.3282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1942 + }, + { + "epoch": 2.3958076448828605, + "grad_norm": 2.5722903261889027, + "learning_rate": 2.0228664372110152e-06, + "loss": 0.3676, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1943 + }, + { + "epoch": 2.397040690505549, + "grad_norm": 2.56754893044426, + "learning_rate": 2.0149233253843393e-06, + "loss": 0.3648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1944 + }, + { + "epoch": 2.3982737361282367, + "grad_norm": 2.5487534871682835, + "learning_rate": 2.0069940916792665e-06, + "loss": 0.3583, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1945 + }, + { + "epoch": 2.399506781750925, + "grad_norm": 2.765885050624665, + "learning_rate": 1.999078749876858e-06, + "loss": 0.3466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1946 + }, + { + "epoch": 2.400739827373613, + "grad_norm": 2.8412723371615347, + "learning_rate": 1.9911773137340416e-06, + "loss": 0.4694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1947 + }, + { + "epoch": 2.401972872996301, + "grad_norm": 2.7637883479587417, + "learning_rate": 1.9832897969835687e-06, + "loss": 0.3952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1948 + }, + { + "epoch": 2.4032059186189887, + "grad_norm": 2.466586812071735, + "learning_rate": 1.975416213333997e-06, + "loss": 0.3669, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1949 + }, + { + "epoch": 2.404438964241677, + "grad_norm": 2.7661522361602904, + "learning_rate": 1.967556576469678e-06, + "loss": 0.3284, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1950 + }, + { + "epoch": 2.405672009864365, + "grad_norm": 2.5114022421240056, + "learning_rate": 1.9597109000507163e-06, + "loss": 0.3843, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1951 + }, + { + "epoch": 2.4069050554870532, + "grad_norm": 2.6057762001360336, + "learning_rate": 1.9518791977129537e-06, + "loss": 0.3851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1952 + }, + { + "epoch": 2.408138101109741, + "grad_norm": 2.415051279016423, + "learning_rate": 1.944061483067945e-06, + "loss": 0.3463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1953 + }, + { + "epoch": 2.409371146732429, + "grad_norm": 2.7762579065761734, + "learning_rate": 1.9362577697029383e-06, + "loss": 0.3868, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1954 + }, + { + "epoch": 2.410604192355117, + "grad_norm": 3.063713670502854, + "learning_rate": 1.9284680711808447e-06, + "loss": 0.3809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1955 + }, + { + "epoch": 2.4118372379778052, + "grad_norm": 2.7071449186380336, + "learning_rate": 1.9206924010402116e-06, + "loss": 0.3126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1956 + }, + { + "epoch": 2.413070283600493, + "grad_norm": 2.99690461715501, + "learning_rate": 1.9129307727952184e-06, + "loss": 0.3971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1957 + }, + { + "epoch": 2.4143033292231815, + "grad_norm": 2.538996577415216, + "learning_rate": 1.9051831999356274e-06, + "loss": 0.308, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1958 + }, + { + "epoch": 2.4155363748458694, + "grad_norm": 2.598169922298, + "learning_rate": 1.897449695926774e-06, + "loss": 0.3779, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1959 + }, + { + "epoch": 2.4167694204685573, + "grad_norm": 2.785021708374361, + "learning_rate": 1.8897302742095513e-06, + "loss": 0.4517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1960 + }, + { + "epoch": 2.418002466091245, + "grad_norm": 2.755401413560113, + "learning_rate": 1.8820249482003638e-06, + "loss": 0.4856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1961 + }, + { + "epoch": 2.4192355117139335, + "grad_norm": 2.596435825974963, + "learning_rate": 1.8743337312911335e-06, + "loss": 0.353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1962 + }, + { + "epoch": 2.4204685573366214, + "grad_norm": 2.786096288352132, + "learning_rate": 1.866656636849241e-06, + "loss": 0.4245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1963 + }, + { + "epoch": 2.4217016029593097, + "grad_norm": 2.439658478811596, + "learning_rate": 1.85899367821754e-06, + "loss": 0.2639, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1964 + }, + { + "epoch": 2.4229346485819976, + "grad_norm": 2.544176056739862, + "learning_rate": 1.8513448687143032e-06, + "loss": 0.3492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1965 + }, + { + "epoch": 2.4241676942046855, + "grad_norm": 3.333459945880387, + "learning_rate": 1.8437102216332236e-06, + "loss": 0.6163, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1966 + }, + { + "epoch": 2.4254007398273734, + "grad_norm": 2.419327956697947, + "learning_rate": 1.8360897502433695e-06, + "loss": 0.209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1967 + }, + { + "epoch": 2.4266337854500617, + "grad_norm": 2.570872200928867, + "learning_rate": 1.8284834677891749e-06, + "loss": 0.3493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1968 + }, + { + "epoch": 2.4278668310727496, + "grad_norm": 2.6222413870421786, + "learning_rate": 1.8208913874904177e-06, + "loss": 0.3, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1969 + }, + { + "epoch": 2.429099876695438, + "grad_norm": 2.6311595401436976, + "learning_rate": 1.8133135225421873e-06, + "loss": 0.3072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1970 + }, + { + "epoch": 2.430332922318126, + "grad_norm": 2.5866942216903235, + "learning_rate": 1.8057498861148691e-06, + "loss": 0.2551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1971 + }, + { + "epoch": 2.4315659679408137, + "grad_norm": 2.868334596003309, + "learning_rate": 1.798200491354114e-06, + "loss": 0.3534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1972 + }, + { + "epoch": 2.432799013563502, + "grad_norm": 2.636368307683731, + "learning_rate": 1.7906653513808303e-06, + "loss": 0.3249, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1973 + }, + { + "epoch": 2.43403205918619, + "grad_norm": 2.2980539591250118, + "learning_rate": 1.783144479291149e-06, + "loss": 0.3174, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1974 + }, + { + "epoch": 2.435265104808878, + "grad_norm": 2.7793989257875253, + "learning_rate": 1.7756378881563917e-06, + "loss": 0.4124, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1975 + }, + { + "epoch": 2.436498150431566, + "grad_norm": 2.3284324040289186, + "learning_rate": 1.768145591023076e-06, + "loss": 0.2354, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1976 + }, + { + "epoch": 2.437731196054254, + "grad_norm": 2.7961580556831405, + "learning_rate": 1.7606676009128643e-06, + "loss": 0.3147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1977 + }, + { + "epoch": 2.438964241676942, + "grad_norm": 2.78394209814451, + "learning_rate": 1.7532039308225656e-06, + "loss": 0.4261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1978 + }, + { + "epoch": 2.4401972872996303, + "grad_norm": 2.7644595713287634, + "learning_rate": 1.7457545937240827e-06, + "loss": 0.3966, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1979 + }, + { + "epoch": 2.441430332922318, + "grad_norm": 2.6754626629785148, + "learning_rate": 1.7383196025644233e-06, + "loss": 0.3258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1980 + }, + { + "epoch": 2.442663378545006, + "grad_norm": 2.5894501951666817, + "learning_rate": 1.7308989702656586e-06, + "loss": 0.3238, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1981 + }, + { + "epoch": 2.4438964241676944, + "grad_norm": 2.8255899965176385, + "learning_rate": 1.7234927097249e-06, + "loss": 0.3219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1982 + }, + { + "epoch": 2.4451294697903823, + "grad_norm": 2.957389703887107, + "learning_rate": 1.7161008338142804e-06, + "loss": 0.4042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1983 + }, + { + "epoch": 2.44636251541307, + "grad_norm": 2.7253630268917317, + "learning_rate": 1.7087233553809335e-06, + "loss": 0.351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1984 + }, + { + "epoch": 2.4475955610357585, + "grad_norm": 2.6055125133787387, + "learning_rate": 1.7013602872469749e-06, + "loss": 0.3155, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1985 + }, + { + "epoch": 2.4488286066584464, + "grad_norm": 2.9002281578070526, + "learning_rate": 1.6940116422094688e-06, + "loss": 0.3765, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1986 + }, + { + "epoch": 2.4500616522811343, + "grad_norm": 2.8603120212321222, + "learning_rate": 1.6866774330404112e-06, + "loss": 0.485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1987 + }, + { + "epoch": 2.4512946979038226, + "grad_norm": 2.750919681788328, + "learning_rate": 1.6793576724867167e-06, + "loss": 0.3956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1988 + }, + { + "epoch": 2.4525277435265105, + "grad_norm": 2.5911381734037167, + "learning_rate": 1.6720523732701787e-06, + "loss": 0.2877, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1989 + }, + { + "epoch": 2.4537607891491984, + "grad_norm": 2.711257130090504, + "learning_rate": 1.6647615480874623e-06, + "loss": 0.3308, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1990 + }, + { + "epoch": 2.4549938347718867, + "grad_norm": 2.824061791832042, + "learning_rate": 1.6574852096100723e-06, + "loss": 0.4287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1991 + }, + { + "epoch": 2.4562268803945746, + "grad_norm": 2.887945699620393, + "learning_rate": 1.6502233704843395e-06, + "loss": 0.3861, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1992 + }, + { + "epoch": 2.4574599260172625, + "grad_norm": 2.6357336450913955, + "learning_rate": 1.6429760433313957e-06, + "loss": 0.3608, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1993 + }, + { + "epoch": 2.458692971639951, + "grad_norm": 2.672435780241681, + "learning_rate": 1.6357432407471474e-06, + "loss": 0.4074, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1994 + }, + { + "epoch": 2.4599260172626387, + "grad_norm": 2.359546711744168, + "learning_rate": 1.628524975302258e-06, + "loss": 0.3301, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1995 + }, + { + "epoch": 2.4611590628853266, + "grad_norm": 2.663889849020532, + "learning_rate": 1.6213212595421225e-06, + "loss": 0.261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1996 + }, + { + "epoch": 2.462392108508015, + "grad_norm": 2.3712367457855597, + "learning_rate": 1.6141321059868565e-06, + "loss": 0.206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1997 + }, + { + "epoch": 2.463625154130703, + "grad_norm": 2.3333104602294545, + "learning_rate": 1.6069575271312588e-06, + "loss": 0.2662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1998 + }, + { + "epoch": 2.4648581997533907, + "grad_norm": 2.3951475290329673, + "learning_rate": 1.5997975354447981e-06, + "loss": 0.3517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1999 + }, + { + "epoch": 2.466091245376079, + "grad_norm": 2.547897971468787, + "learning_rate": 1.592652143371597e-06, + "loss": 0.3334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2000 + }, + { + "epoch": 2.467324290998767, + "grad_norm": 2.4595176510202954, + "learning_rate": 1.5855213633303956e-06, + "loss": 0.3423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2001 + }, + { + "epoch": 2.468557336621455, + "grad_norm": 2.9582915458097245, + "learning_rate": 1.578405207714543e-06, + "loss": 0.3455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2002 + }, + { + "epoch": 2.469790382244143, + "grad_norm": 2.428673296135474, + "learning_rate": 1.5713036888919675e-06, + "loss": 0.2992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2003 + }, + { + "epoch": 2.471023427866831, + "grad_norm": 2.743398790088527, + "learning_rate": 1.5642168192051644e-06, + "loss": 0.3322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2004 + }, + { + "epoch": 2.472256473489519, + "grad_norm": 3.324065459022819, + "learning_rate": 1.557144610971163e-06, + "loss": 0.4465, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2005 + }, + { + "epoch": 2.4734895191122073, + "grad_norm": 2.162373213734668, + "learning_rate": 1.5500870764815102e-06, + "loss": 0.2274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2006 + }, + { + "epoch": 2.474722564734895, + "grad_norm": 2.6083953204344366, + "learning_rate": 1.5430442280022584e-06, + "loss": 0.3268, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2007 + }, + { + "epoch": 2.475955610357583, + "grad_norm": 2.1764316712749787, + "learning_rate": 1.536016077773923e-06, + "loss": 0.1526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2008 + }, + { + "epoch": 2.4771886559802714, + "grad_norm": 2.704429097110481, + "learning_rate": 1.5290026380114876e-06, + "loss": 0.2536, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2009 + }, + { + "epoch": 2.4784217016029593, + "grad_norm": 3.0958298333409857, + "learning_rate": 1.522003920904358e-06, + "loss": 0.4273, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2010 + }, + { + "epoch": 2.479654747225647, + "grad_norm": 2.9092939112809577, + "learning_rate": 1.5150199386163543e-06, + "loss": 0.3945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2011 + }, + { + "epoch": 2.4808877928483355, + "grad_norm": 3.116797881391071, + "learning_rate": 1.5080507032856894e-06, + "loss": 0.4497, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2012 + }, + { + "epoch": 2.4821208384710234, + "grad_norm": 2.760889317748852, + "learning_rate": 1.5010962270249486e-06, + "loss": 0.2909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2013 + }, + { + "epoch": 2.4833538840937113, + "grad_norm": 2.6074546180794833, + "learning_rate": 1.4941565219210585e-06, + "loss": 0.2939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2014 + }, + { + "epoch": 2.4845869297163996, + "grad_norm": 3.0837456899974827, + "learning_rate": 1.4872316000352783e-06, + "loss": 0.4267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2015 + }, + { + "epoch": 2.4858199753390875, + "grad_norm": 2.925304791109893, + "learning_rate": 1.4803214734031734e-06, + "loss": 0.3992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2016 + }, + { + "epoch": 2.487053020961776, + "grad_norm": 2.873066706251663, + "learning_rate": 1.4734261540345951e-06, + "loss": 0.3851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2017 + }, + { + "epoch": 2.4882860665844637, + "grad_norm": 3.0170897183556535, + "learning_rate": 1.466545653913659e-06, + "loss": 0.3999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2018 + }, + { + "epoch": 2.4895191122071516, + "grad_norm": 2.8171232756409337, + "learning_rate": 1.4596799849987198e-06, + "loss": 0.409, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2019 + }, + { + "epoch": 2.4907521578298395, + "grad_norm": 3.1825430295716157, + "learning_rate": 1.4528291592223642e-06, + "loss": 0.4256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2020 + }, + { + "epoch": 2.491985203452528, + "grad_norm": 2.7388479608941725, + "learning_rate": 1.4459931884913825e-06, + "loss": 0.3645, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2021 + }, + { + "epoch": 2.4932182490752157, + "grad_norm": 2.7221846310700437, + "learning_rate": 1.4391720846867308e-06, + "loss": 0.3802, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2022 + }, + { + "epoch": 2.494451294697904, + "grad_norm": 2.794004204773263, + "learning_rate": 1.4323658596635459e-06, + "loss": 0.3532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2023 + }, + { + "epoch": 2.495684340320592, + "grad_norm": 2.9987018690091087, + "learning_rate": 1.425574525251092e-06, + "loss": 0.4432, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2024 + }, + { + "epoch": 2.49691738594328, + "grad_norm": 2.6488734906053693, + "learning_rate": 1.4187980932527612e-06, + "loss": 0.3241, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2025 + }, + { + "epoch": 2.4981504315659677, + "grad_norm": 2.433220382902042, + "learning_rate": 1.4120365754460408e-06, + "loss": 0.2895, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2026 + }, + { + "epoch": 2.499383477188656, + "grad_norm": 2.434542396902984, + "learning_rate": 1.4052899835824951e-06, + "loss": 0.2108, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2027 + }, + { + "epoch": 2.500616522811344, + "grad_norm": 2.524311609378393, + "learning_rate": 1.3985583293877547e-06, + "loss": 0.3472, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2028 + }, + { + "epoch": 2.5018495684340323, + "grad_norm": 3.1734464121061396, + "learning_rate": 1.3918416245614818e-06, + "loss": 0.5071, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2029 + }, + { + "epoch": 2.50308261405672, + "grad_norm": 2.576082693205846, + "learning_rate": 1.3851398807773587e-06, + "loss": 0.3994, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2030 + } + ], + "logging_steps": 1, + "max_steps": 2433, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 406, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3333912998707200.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-2030/training_args.bin b/checkpoint-2030/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7b675d785639ed4b99af767b6c87a1227c13ec1b --- /dev/null +++ b/checkpoint-2030/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e80ea310ffc6c44543f32aa82ecc00b1bf9d3e329881647443a895f041e17955 +size 9208 diff --git a/checkpoint-2433/chat_template.jinja b/checkpoint-2433/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..870322b8f52314199bfd74bf68b8b8f1126395ae --- /dev/null +++ b/checkpoint-2433/chat_template.jinja @@ -0,0 +1,5 @@ +{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} diff --git a/checkpoint-2433/config.json b/checkpoint-2433/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4c0f53b6c66a0b8a2a4eb621c949412e2a7f353a --- /dev/null +++ b/checkpoint-2433/config.json @@ -0,0 +1,35 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.2", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/checkpoint-2433/generation_config.json b/checkpoint-2433/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b513e54e3195b917260c9a8a04c9f3683f19de35 --- /dev/null +++ b/checkpoint-2433/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.55.2" +} diff --git a/checkpoint-2433/model-00004-of-00004.safetensors b/checkpoint-2433/model-00004-of-00004.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2de9c89cd2fed0f3ae65ea5c2bafc2c1da5de87b --- /dev/null +++ b/checkpoint-2433/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d00ab5dcfd3b949feab0bfb18da3457ad431f69815f9971015d3a83d6d7ffb2b +size 1168138808 diff --git a/checkpoint-2433/model.safetensors.index.json b/checkpoint-2433/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..58a7ef40d52ab3340cf131cf33f8923f5144aaf9 --- /dev/null +++ b/checkpoint-2433/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-2433/special_tokens_map.json b/checkpoint-2433/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3c1d04911c269b925af977a3151c9704e990e4d0 --- /dev/null +++ b/checkpoint-2433/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-2433/tokenizer.json b/checkpoint-2433/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-2433/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-2433/tokenizer_config.json b/checkpoint-2433/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..c68051fe3c4d23234a59316bc52d21f6e3a4182c --- /dev/null +++ b/checkpoint-2433/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|finetune_right_pad_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-2433/trainer_state.json b/checkpoint-2433/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..2f6e61981224501086f07a277b58fcc998b51255 --- /dev/null +++ b/checkpoint-2433/trainer_state.json @@ -0,0 +1,24364 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.0, + "eval_steps": 500, + "global_step": 2433, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0012330456226880395, + "grad_norm": 13.955555906863093, + "learning_rate": 0.0, + "loss": 1.9039, + "memory/device_mem_reserved(gib)": 19.83, + "memory/max_mem_active(gib)": 16.16, + "memory/max_mem_allocated(gib)": 16.16, + "step": 1 + }, + { + "epoch": 0.002466091245376079, + "grad_norm": 14.68412675273428, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.9395, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 2 + }, + { + "epoch": 0.0036991368680641184, + "grad_norm": 15.404314959825989, + "learning_rate": 8.000000000000001e-07, + "loss": 1.9524, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 3 + }, + { + "epoch": 0.004932182490752158, + "grad_norm": 14.099502171385124, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.8527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 4 + }, + { + "epoch": 0.006165228113440197, + "grad_norm": 12.213499070556583, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.8853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 5 + }, + { + "epoch": 0.007398273736128237, + "grad_norm": 7.4228290511653725, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.7686, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 6 + }, + { + "epoch": 0.008631319358816275, + "grad_norm": 6.989334002600767, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.7976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 7 + }, + { + "epoch": 0.009864364981504316, + "grad_norm": 7.330479156896915, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6029, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 8 + }, + { + "epoch": 0.011097410604192354, + "grad_norm": 7.091571204851302, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.6097, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 9 + }, + { + "epoch": 0.012330456226880395, + "grad_norm": 5.900225931668668, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.6512, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 10 + }, + { + "epoch": 0.013563501849568433, + "grad_norm": 12.562784448457464, + "learning_rate": 4.000000000000001e-06, + "loss": 1.4824, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 11 + }, + { + "epoch": 0.014796547472256474, + "grad_norm": 6.056074456731366, + "learning_rate": 4.4e-06, + "loss": 1.5581, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 12 + }, + { + "epoch": 0.016029593094944512, + "grad_norm": 5.051119502724206, + "learning_rate": 4.800000000000001e-06, + "loss": 1.5275, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.11, + "memory/max_mem_allocated(gib)": 18.13, + "step": 13 + }, + { + "epoch": 0.01726263871763255, + "grad_norm": 4.022180485081632, + "learning_rate": 5.2e-06, + "loss": 1.5075, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 14 + }, + { + "epoch": 0.018495684340320593, + "grad_norm": 4.5104551135107815, + "learning_rate": 5.600000000000001e-06, + "loss": 1.4638, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 15 + }, + { + "epoch": 0.01972872996300863, + "grad_norm": 4.3429897237996276, + "learning_rate": 6e-06, + "loss": 1.3996, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 16 + }, + { + "epoch": 0.02096177558569667, + "grad_norm": 3.9791138623880764, + "learning_rate": 6.4000000000000006e-06, + "loss": 1.4059, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 17 + }, + { + "epoch": 0.02219482120838471, + "grad_norm": 4.194505204288471, + "learning_rate": 6.800000000000001e-06, + "loss": 1.3747, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 18 + }, + { + "epoch": 0.02342786683107275, + "grad_norm": 3.6309957760327713, + "learning_rate": 7.2000000000000005e-06, + "loss": 1.3565, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 19 + }, + { + "epoch": 0.02466091245376079, + "grad_norm": 3.431355025722372, + "learning_rate": 7.600000000000001e-06, + "loss": 1.4019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 20 + }, + { + "epoch": 0.025893958076448828, + "grad_norm": 3.258474422309909, + "learning_rate": 8.000000000000001e-06, + "loss": 1.364, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 21 + }, + { + "epoch": 0.027127003699136867, + "grad_norm": 3.7750095693616736, + "learning_rate": 8.400000000000001e-06, + "loss": 1.3798, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 22 + }, + { + "epoch": 0.02836004932182491, + "grad_norm": 3.1192450402705125, + "learning_rate": 8.8e-06, + "loss": 1.2677, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 23 + }, + { + "epoch": 0.029593094944512947, + "grad_norm": 3.34669710403596, + "learning_rate": 9.200000000000002e-06, + "loss": 1.3465, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 24 + }, + { + "epoch": 0.030826140567200986, + "grad_norm": 3.0969841766129638, + "learning_rate": 9.600000000000001e-06, + "loss": 1.3662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 25 + }, + { + "epoch": 0.032059186189889025, + "grad_norm": 2.9521843506112466, + "learning_rate": 1e-05, + "loss": 1.3295, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 26 + }, + { + "epoch": 0.03329223181257707, + "grad_norm": 2.9117498945141085, + "learning_rate": 1.04e-05, + "loss": 1.3066, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.14, + "memory/max_mem_allocated(gib)": 18.16, + "step": 27 + }, + { + "epoch": 0.0345252774352651, + "grad_norm": 2.9215831263409013, + "learning_rate": 1.0800000000000002e-05, + "loss": 1.3221, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 28 + }, + { + "epoch": 0.035758323057953144, + "grad_norm": 2.97321184472987, + "learning_rate": 1.1200000000000001e-05, + "loss": 1.3247, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 29 + }, + { + "epoch": 0.036991368680641186, + "grad_norm": 3.293476484963883, + "learning_rate": 1.16e-05, + "loss": 1.2582, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 30 + }, + { + "epoch": 0.03822441430332922, + "grad_norm": 2.8642108219246745, + "learning_rate": 1.2e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 31 + }, + { + "epoch": 0.03945745992601726, + "grad_norm": 2.8916726488438824, + "learning_rate": 1.2400000000000002e-05, + "loss": 1.2628, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 32 + }, + { + "epoch": 0.040690505548705305, + "grad_norm": 3.17999982306396, + "learning_rate": 1.2800000000000001e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 33 + }, + { + "epoch": 0.04192355117139334, + "grad_norm": 2.6965579635696453, + "learning_rate": 1.3200000000000002e-05, + "loss": 1.4046, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 34 + }, + { + "epoch": 0.04315659679408138, + "grad_norm": 2.9489592197594896, + "learning_rate": 1.3600000000000002e-05, + "loss": 1.3113, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 35 + }, + { + "epoch": 0.04438964241676942, + "grad_norm": 2.8898479257030734, + "learning_rate": 1.4e-05, + "loss": 1.3387, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 36 + }, + { + "epoch": 0.04562268803945746, + "grad_norm": 2.6370539695770425, + "learning_rate": 1.4400000000000001e-05, + "loss": 1.3314, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 37 + }, + { + "epoch": 0.0468557336621455, + "grad_norm": 2.7915119494225813, + "learning_rate": 1.48e-05, + "loss": 1.2603, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 38 + }, + { + "epoch": 0.04808877928483354, + "grad_norm": 2.602735589719503, + "learning_rate": 1.5200000000000002e-05, + "loss": 1.3492, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 39 + }, + { + "epoch": 0.04932182490752158, + "grad_norm": 2.9123309438390113, + "learning_rate": 1.5600000000000003e-05, + "loss": 1.2291, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 40 + }, + { + "epoch": 0.05055487053020962, + "grad_norm": 2.577807086473754, + "learning_rate": 1.6000000000000003e-05, + "loss": 1.3098, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 41 + }, + { + "epoch": 0.051787916152897656, + "grad_norm": 2.548383978805174, + "learning_rate": 1.64e-05, + "loss": 1.2893, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 42 + }, + { + "epoch": 0.0530209617755857, + "grad_norm": 2.6372216084932814, + "learning_rate": 1.6800000000000002e-05, + "loss": 1.2701, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 43 + }, + { + "epoch": 0.05425400739827373, + "grad_norm": 2.60286437866227, + "learning_rate": 1.72e-05, + "loss": 1.3579, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.15, + "memory/max_mem_allocated(gib)": 18.18, + "step": 44 + }, + { + "epoch": 0.055487053020961775, + "grad_norm": 2.536071820500707, + "learning_rate": 1.76e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 45 + }, + { + "epoch": 0.05672009864364982, + "grad_norm": 2.8143481627767963, + "learning_rate": 1.8e-05, + "loss": 1.1729, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 46 + }, + { + "epoch": 0.05795314426633785, + "grad_norm": 2.8600643284872596, + "learning_rate": 1.8400000000000003e-05, + "loss": 1.3246, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 47 + }, + { + "epoch": 0.059186189889025895, + "grad_norm": 2.618463019216554, + "learning_rate": 1.88e-05, + "loss": 1.2878, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 48 + }, + { + "epoch": 0.06041923551171394, + "grad_norm": 2.757496450593693, + "learning_rate": 1.9200000000000003e-05, + "loss": 1.3454, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 49 + }, + { + "epoch": 0.06165228113440197, + "grad_norm": 2.5356573576135126, + "learning_rate": 1.9600000000000002e-05, + "loss": 1.2719, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 50 + }, + { + "epoch": 0.06288532675709001, + "grad_norm": 2.532419474064049, + "learning_rate": 2e-05, + "loss": 1.332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 51 + }, + { + "epoch": 0.06411837237977805, + "grad_norm": 2.711902763412396, + "learning_rate": 1.999999130996347e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 52 + }, + { + "epoch": 0.06535141800246609, + "grad_norm": 2.654767047589697, + "learning_rate": 1.9999965239868983e-05, + "loss": 1.3092, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 53 + }, + { + "epoch": 0.06658446362515413, + "grad_norm": 3.2372229064932454, + "learning_rate": 1.9999921789761843e-05, + "loss": 1.3428, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 54 + }, + { + "epoch": 0.06781750924784218, + "grad_norm": 2.723103510772695, + "learning_rate": 1.999986095971757e-05, + "loss": 1.3918, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 55 + }, + { + "epoch": 0.0690505548705302, + "grad_norm": 2.849098835881172, + "learning_rate": 1.9999782749841895e-05, + "loss": 1.3304, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 56 + }, + { + "epoch": 0.07028360049321825, + "grad_norm": 2.4115082044348557, + "learning_rate": 1.9999687160270735e-05, + "loss": 1.2818, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 57 + }, + { + "epoch": 0.07151664611590629, + "grad_norm": 2.7720437269939495, + "learning_rate": 1.9999574191170233e-05, + "loss": 1.2799, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 58 + }, + { + "epoch": 0.07274969173859433, + "grad_norm": 2.4891158758553042, + "learning_rate": 1.9999443842736723e-05, + "loss": 1.2408, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 59 + }, + { + "epoch": 0.07398273736128237, + "grad_norm": 2.8066935220153737, + "learning_rate": 1.999929611519676e-05, + "loss": 1.3157, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 60 + }, + { + "epoch": 0.0752157829839704, + "grad_norm": 2.6715709033170705, + "learning_rate": 1.9999131008807094e-05, + "loss": 1.3185, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 61 + }, + { + "epoch": 0.07644882860665844, + "grad_norm": 2.586596297320694, + "learning_rate": 1.9998948523854676e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 62 + }, + { + "epoch": 0.07768187422934648, + "grad_norm": 2.646887045495792, + "learning_rate": 1.999874866065667e-05, + "loss": 1.349, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 63 + }, + { + "epoch": 0.07891491985203453, + "grad_norm": 2.419730330481743, + "learning_rate": 1.999853141956044e-05, + "loss": 1.2829, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 64 + }, + { + "epoch": 0.08014796547472257, + "grad_norm": 2.5201815842521245, + "learning_rate": 1.999829680094355e-05, + "loss": 1.3243, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 65 + }, + { + "epoch": 0.08138101109741061, + "grad_norm": 2.503429794099987, + "learning_rate": 1.9998044805213773e-05, + "loss": 1.2681, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 66 + }, + { + "epoch": 0.08261405672009864, + "grad_norm": 2.348439339795846, + "learning_rate": 1.9997775432809074e-05, + "loss": 1.3359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 67 + }, + { + "epoch": 0.08384710234278668, + "grad_norm": 2.3828095918277357, + "learning_rate": 1.999748868419763e-05, + "loss": 1.3526, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 68 + }, + { + "epoch": 0.08508014796547472, + "grad_norm": 2.5076969489265273, + "learning_rate": 1.9997184559877806e-05, + "loss": 1.3531, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 69 + }, + { + "epoch": 0.08631319358816276, + "grad_norm": 2.388938276200468, + "learning_rate": 1.9996863060378176e-05, + "loss": 1.334, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 70 + }, + { + "epoch": 0.08754623921085081, + "grad_norm": 2.283154945083281, + "learning_rate": 1.999652418625751e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 71 + }, + { + "epoch": 0.08877928483353884, + "grad_norm": 2.2613574311048734, + "learning_rate": 1.9996167938104767e-05, + "loss": 1.2858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 72 + }, + { + "epoch": 0.09001233045622688, + "grad_norm": 2.3931162231369076, + "learning_rate": 1.999579431653912e-05, + "loss": 1.3089, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 73 + }, + { + "epoch": 0.09124537607891492, + "grad_norm": 2.5073647819033194, + "learning_rate": 1.9995403322209913e-05, + "loss": 1.382, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 74 + }, + { + "epoch": 0.09247842170160296, + "grad_norm": 2.5388011932611536, + "learning_rate": 1.999499495579671e-05, + "loss": 1.402, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 75 + }, + { + "epoch": 0.093711467324291, + "grad_norm": 2.3884809904992252, + "learning_rate": 1.9994569218009247e-05, + "loss": 1.3079, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 76 + }, + { + "epoch": 0.09494451294697903, + "grad_norm": 2.3454483578272094, + "learning_rate": 1.999412610958746e-05, + "loss": 1.2653, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 77 + }, + { + "epoch": 0.09617755856966707, + "grad_norm": 2.186902835408518, + "learning_rate": 1.9993665631301474e-05, + "loss": 1.2069, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 78 + }, + { + "epoch": 0.09741060419235512, + "grad_norm": 2.3845058131797368, + "learning_rate": 1.9993187783951603e-05, + "loss": 1.3541, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 79 + }, + { + "epoch": 0.09864364981504316, + "grad_norm": 2.256399889979939, + "learning_rate": 1.9992692568368354e-05, + "loss": 1.2119, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 80 + }, + { + "epoch": 0.0998766954377312, + "grad_norm": 2.3361092398369903, + "learning_rate": 1.9992179985412416e-05, + "loss": 1.2585, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 81 + }, + { + "epoch": 0.10110974106041924, + "grad_norm": 2.289061082197673, + "learning_rate": 1.999165003597465e-05, + "loss": 1.3459, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 82 + }, + { + "epoch": 0.10234278668310727, + "grad_norm": 2.228187448538031, + "learning_rate": 1.999110272097613e-05, + "loss": 1.3739, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 83 + }, + { + "epoch": 0.10357583230579531, + "grad_norm": 2.157361593613978, + "learning_rate": 1.9990538041368078e-05, + "loss": 1.2575, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 84 + }, + { + "epoch": 0.10480887792848335, + "grad_norm": 2.294809937254642, + "learning_rate": 1.998995599813192e-05, + "loss": 1.2834, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 85 + }, + { + "epoch": 0.1060419235511714, + "grad_norm": 2.1831743941808925, + "learning_rate": 1.998935659227925e-05, + "loss": 1.2699, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 86 + }, + { + "epoch": 0.10727496917385944, + "grad_norm": 2.247228611971362, + "learning_rate": 1.998873982485183e-05, + "loss": 1.2654, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 87 + }, + { + "epoch": 0.10850801479654747, + "grad_norm": 2.2844560193856416, + "learning_rate": 1.9988105696921622e-05, + "loss": 1.2927, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 88 + }, + { + "epoch": 0.10974106041923551, + "grad_norm": 2.3780157649218157, + "learning_rate": 1.9987454209590736e-05, + "loss": 1.2621, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 89 + }, + { + "epoch": 0.11097410604192355, + "grad_norm": 2.3017268441414003, + "learning_rate": 1.9986785363991466e-05, + "loss": 1.2676, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 90 + }, + { + "epoch": 0.11220715166461159, + "grad_norm": 2.1906552279173503, + "learning_rate": 1.9986099161286266e-05, + "loss": 1.2281, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 91 + }, + { + "epoch": 0.11344019728729964, + "grad_norm": 2.4009419002081276, + "learning_rate": 1.9985395602667764e-05, + "loss": 1.2267, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 92 + }, + { + "epoch": 0.11467324290998766, + "grad_norm": 2.190217063330867, + "learning_rate": 1.9984674689358744e-05, + "loss": 1.2778, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 93 + }, + { + "epoch": 0.1159062885326757, + "grad_norm": 2.305624793209019, + "learning_rate": 1.998393642261217e-05, + "loss": 1.2821, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.16, + "memory/max_mem_allocated(gib)": 18.18, + "step": 94 + }, + { + "epoch": 0.11713933415536375, + "grad_norm": 2.162537568697834, + "learning_rate": 1.9983180803711146e-05, + "loss": 1.3775, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 95 + }, + { + "epoch": 0.11837237977805179, + "grad_norm": 2.2632096292316706, + "learning_rate": 1.9982407833968947e-05, + "loss": 1.3358, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 96 + }, + { + "epoch": 0.11960542540073983, + "grad_norm": 2.268508818243908, + "learning_rate": 1.9981617514728998e-05, + "loss": 1.2368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 97 + }, + { + "epoch": 0.12083847102342787, + "grad_norm": 2.2721363895167377, + "learning_rate": 1.9980809847364884e-05, + "loss": 1.2774, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 98 + }, + { + "epoch": 0.1220715166461159, + "grad_norm": 2.1661250921439974, + "learning_rate": 1.9979984833280332e-05, + "loss": 1.359, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 99 + }, + { + "epoch": 0.12330456226880394, + "grad_norm": 2.2521676172131873, + "learning_rate": 1.997914247390922e-05, + "loss": 1.216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 100 + }, + { + "epoch": 0.12453760789149199, + "grad_norm": 2.3907655242108095, + "learning_rate": 1.9978282770715586e-05, + "loss": 1.2468, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 101 + }, + { + "epoch": 0.12577065351418001, + "grad_norm": 2.2007275768445074, + "learning_rate": 1.9977405725193587e-05, + "loss": 1.2415, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 102 + }, + { + "epoch": 0.12700369913686807, + "grad_norm": 2.3304230330598346, + "learning_rate": 1.9976511338867542e-05, + "loss": 1.2438, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 103 + }, + { + "epoch": 0.1282367447595561, + "grad_norm": 2.346171964705753, + "learning_rate": 1.99755996132919e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 104 + }, + { + "epoch": 0.12946979038224415, + "grad_norm": 2.286623534979458, + "learning_rate": 1.9974670550051248e-05, + "loss": 1.2897, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 105 + }, + { + "epoch": 0.13070283600493218, + "grad_norm": 2.245774575899364, + "learning_rate": 1.99737241507603e-05, + "loss": 1.2923, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 106 + }, + { + "epoch": 0.1319358816276202, + "grad_norm": 2.2352297973588953, + "learning_rate": 1.9972760417063912e-05, + "loss": 1.1434, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 107 + }, + { + "epoch": 0.13316892725030827, + "grad_norm": 2.169802916370051, + "learning_rate": 1.9971779350637054e-05, + "loss": 1.2353, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 108 + }, + { + "epoch": 0.1344019728729963, + "grad_norm": 2.333533512175979, + "learning_rate": 1.9970780953184828e-05, + "loss": 1.2641, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 109 + }, + { + "epoch": 0.13563501849568435, + "grad_norm": 2.2985803581198674, + "learning_rate": 1.996976522644246e-05, + "loss": 1.3036, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 110 + }, + { + "epoch": 0.13686806411837238, + "grad_norm": 2.0992498120010272, + "learning_rate": 1.996873217217528e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 111 + }, + { + "epoch": 0.1381011097410604, + "grad_norm": 2.396895991067544, + "learning_rate": 1.9967681792178758e-05, + "loss": 1.2533, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 112 + }, + { + "epoch": 0.13933415536374846, + "grad_norm": 2.0738932505082768, + "learning_rate": 1.9966614088278456e-05, + "loss": 1.3332, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 113 + }, + { + "epoch": 0.1405672009864365, + "grad_norm": 2.1049156004628395, + "learning_rate": 1.9965529062330048e-05, + "loss": 1.2793, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 114 + }, + { + "epoch": 0.14180024660912455, + "grad_norm": 2.2137820282577847, + "learning_rate": 1.9964426716219316e-05, + "loss": 1.1932, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 115 + }, + { + "epoch": 0.14303329223181258, + "grad_norm": 2.1545610182661976, + "learning_rate": 1.9963307051862152e-05, + "loss": 1.2662, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 116 + }, + { + "epoch": 0.1442663378545006, + "grad_norm": 2.151286882718536, + "learning_rate": 1.9962170071204542e-05, + "loss": 1.205, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 117 + }, + { + "epoch": 0.14549938347718866, + "grad_norm": 2.026608825482131, + "learning_rate": 1.996101577622256e-05, + "loss": 1.195, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 118 + }, + { + "epoch": 0.1467324290998767, + "grad_norm": 2.156975981742732, + "learning_rate": 1.9959844168922377e-05, + "loss": 1.3284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 119 + }, + { + "epoch": 0.14796547472256474, + "grad_norm": 2.2311865782919025, + "learning_rate": 1.9958655251340265e-05, + "loss": 1.2183, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 120 + }, + { + "epoch": 0.14919852034525277, + "grad_norm": 2.2940740150536376, + "learning_rate": 1.9957449025542567e-05, + "loss": 1.2625, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 121 + }, + { + "epoch": 0.1504315659679408, + "grad_norm": 2.214218654766027, + "learning_rate": 1.9956225493625708e-05, + "loss": 1.2177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 122 + }, + { + "epoch": 0.15166461159062886, + "grad_norm": 2.1690434056776775, + "learning_rate": 1.9954984657716202e-05, + "loss": 1.2049, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 123 + }, + { + "epoch": 0.15289765721331688, + "grad_norm": 2.1450511034984046, + "learning_rate": 1.9953726519970625e-05, + "loss": 1.232, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 124 + }, + { + "epoch": 0.15413070283600494, + "grad_norm": 2.0475086343318543, + "learning_rate": 1.995245108257563e-05, + "loss": 1.2665, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 125 + }, + { + "epoch": 0.15536374845869297, + "grad_norm": 2.1232594967339318, + "learning_rate": 1.9951158347747945e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 126 + }, + { + "epoch": 0.15659679408138102, + "grad_norm": 2.1915516478247308, + "learning_rate": 1.9949848317734345e-05, + "loss": 1.177, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 127 + }, + { + "epoch": 0.15782983970406905, + "grad_norm": 2.2848682922250205, + "learning_rate": 1.9948520994811674e-05, + "loss": 1.2406, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 128 + }, + { + "epoch": 0.15906288532675708, + "grad_norm": 2.2223107161107505, + "learning_rate": 1.9947176381286825e-05, + "loss": 1.327, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 129 + }, + { + "epoch": 0.16029593094944514, + "grad_norm": 2.1247307489434837, + "learning_rate": 1.994581447949675e-05, + "loss": 1.1634, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 130 + }, + { + "epoch": 0.16152897657213316, + "grad_norm": 1.926245615401859, + "learning_rate": 1.9944435291808445e-05, + "loss": 1.1655, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 131 + }, + { + "epoch": 0.16276202219482122, + "grad_norm": 2.2531819769934858, + "learning_rate": 1.9943038820618946e-05, + "loss": 1.2216, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 132 + }, + { + "epoch": 0.16399506781750925, + "grad_norm": 2.2183949614858305, + "learning_rate": 1.994162506835533e-05, + "loss": 1.2202, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 133 + }, + { + "epoch": 0.16522811344019728, + "grad_norm": 2.164554433266706, + "learning_rate": 1.994019403747471e-05, + "loss": 1.2245, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 134 + }, + { + "epoch": 0.16646115906288533, + "grad_norm": 2.17653642461108, + "learning_rate": 1.9938745730464228e-05, + "loss": 1.1899, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 135 + }, + { + "epoch": 0.16769420468557336, + "grad_norm": 1.9936998341266101, + "learning_rate": 1.9937280149841055e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 136 + }, + { + "epoch": 0.16892725030826142, + "grad_norm": 2.27069520590907, + "learning_rate": 1.9935797298152374e-05, + "loss": 1.3023, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 137 + }, + { + "epoch": 0.17016029593094945, + "grad_norm": 2.244074718609361, + "learning_rate": 1.9934297177975397e-05, + "loss": 1.3292, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 138 + }, + { + "epoch": 0.17139334155363747, + "grad_norm": 2.2121305803685094, + "learning_rate": 1.9932779791917343e-05, + "loss": 1.3028, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 139 + }, + { + "epoch": 0.17262638717632553, + "grad_norm": 2.7262731821688697, + "learning_rate": 1.9931245142615442e-05, + "loss": 1.2861, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.17, + "memory/max_mem_allocated(gib)": 18.19, + "step": 140 + }, + { + "epoch": 0.17385943279901356, + "grad_norm": 2.2051993935523497, + "learning_rate": 1.992969323273692e-05, + "loss": 1.3134, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 141 + }, + { + "epoch": 0.17509247842170161, + "grad_norm": 2.088688387254833, + "learning_rate": 1.992812406497901e-05, + "loss": 1.262, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 142 + }, + { + "epoch": 0.17632552404438964, + "grad_norm": 2.19652156002121, + "learning_rate": 1.9926537642068942e-05, + "loss": 1.3039, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 143 + }, + { + "epoch": 0.17755856966707767, + "grad_norm": 1.887489027726817, + "learning_rate": 1.9924933966763924e-05, + "loss": 1.2847, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 144 + }, + { + "epoch": 0.17879161528976573, + "grad_norm": 2.088506357330406, + "learning_rate": 1.992331304185116e-05, + "loss": 1.2218, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 145 + }, + { + "epoch": 0.18002466091245375, + "grad_norm": 2.102792679186299, + "learning_rate": 1.9921674870147827e-05, + "loss": 1.242, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 146 + }, + { + "epoch": 0.1812577065351418, + "grad_norm": 2.0847095844816015, + "learning_rate": 1.992001945450108e-05, + "loss": 1.2986, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 147 + }, + { + "epoch": 0.18249075215782984, + "grad_norm": 2.344867998715345, + "learning_rate": 1.9918346797788045e-05, + "loss": 1.1347, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 148 + }, + { + "epoch": 0.18372379778051787, + "grad_norm": 2.1013292606380602, + "learning_rate": 1.991665690291581e-05, + "loss": 1.1319, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 149 + }, + { + "epoch": 0.18495684340320592, + "grad_norm": 2.2475258887436538, + "learning_rate": 1.9914949772821418e-05, + "loss": 1.1648, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 150 + }, + { + "epoch": 0.18618988902589395, + "grad_norm": 2.223126056771945, + "learning_rate": 1.9913225410471888e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 151 + }, + { + "epoch": 0.187422934648582, + "grad_norm": 2.102849036750559, + "learning_rate": 1.9911483818864163e-05, + "loss": 1.2062, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 152 + }, + { + "epoch": 0.18865598027127004, + "grad_norm": 2.2107888348018805, + "learning_rate": 1.9909725001025146e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 153 + }, + { + "epoch": 0.18988902589395806, + "grad_norm": 2.1730720194510456, + "learning_rate": 1.9907948960011677e-05, + "loss": 1.2133, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 154 + }, + { + "epoch": 0.19112207151664612, + "grad_norm": 2.0874944224993115, + "learning_rate": 1.9906155698910523e-05, + "loss": 1.1966, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 155 + }, + { + "epoch": 0.19235511713933415, + "grad_norm": 2.047788318815714, + "learning_rate": 1.990434522083839e-05, + "loss": 1.26, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 156 + }, + { + "epoch": 0.1935881627620222, + "grad_norm": 2.0992674413368415, + "learning_rate": 1.99025175289419e-05, + "loss": 1.2544, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 157 + }, + { + "epoch": 0.19482120838471023, + "grad_norm": 2.279539056045867, + "learning_rate": 1.9900672626397593e-05, + "loss": 1.3123, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 158 + }, + { + "epoch": 0.1960542540073983, + "grad_norm": 1.9940111001673004, + "learning_rate": 1.9898810516411926e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 159 + }, + { + "epoch": 0.19728729963008632, + "grad_norm": 2.278268154556979, + "learning_rate": 1.989693120222126e-05, + "loss": 1.236, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 160 + }, + { + "epoch": 0.19852034525277434, + "grad_norm": 2.3184527515323796, + "learning_rate": 1.9895034687091854e-05, + "loss": 1.1691, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 161 + }, + { + "epoch": 0.1997533908754624, + "grad_norm": 2.094943513595831, + "learning_rate": 1.9893120974319867e-05, + "loss": 1.3019, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 162 + }, + { + "epoch": 0.20098643649815043, + "grad_norm": 2.296308593300257, + "learning_rate": 1.9891190067231346e-05, + "loss": 1.2976, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 163 + }, + { + "epoch": 0.20221948212083848, + "grad_norm": 2.131546792333376, + "learning_rate": 1.988924196918222e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 164 + }, + { + "epoch": 0.2034525277435265, + "grad_norm": 1.981669864872828, + "learning_rate": 1.9887276683558302e-05, + "loss": 1.2934, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 165 + }, + { + "epoch": 0.20468557336621454, + "grad_norm": 2.0382654138005423, + "learning_rate": 1.9885294213775267e-05, + "loss": 1.3207, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 166 + }, + { + "epoch": 0.2059186189889026, + "grad_norm": 2.074242050982157, + "learning_rate": 1.9883294563278665e-05, + "loss": 1.1832, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 167 + }, + { + "epoch": 0.20715166461159062, + "grad_norm": 2.1994142392378317, + "learning_rate": 1.9881277735543902e-05, + "loss": 1.2284, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 168 + }, + { + "epoch": 0.20838471023427868, + "grad_norm": 2.1893568284451987, + "learning_rate": 1.987924373407624e-05, + "loss": 1.2572, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 169 + }, + { + "epoch": 0.2096177558569667, + "grad_norm": 2.0490903431710352, + "learning_rate": 1.987719256241079e-05, + "loss": 1.2808, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 170 + }, + { + "epoch": 0.21085080147965474, + "grad_norm": 2.3424605103097176, + "learning_rate": 1.9875124224112498e-05, + "loss": 1.2924, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 171 + }, + { + "epoch": 0.2120838471023428, + "grad_norm": 2.100563128603325, + "learning_rate": 1.987303872277616e-05, + "loss": 1.1368, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 172 + }, + { + "epoch": 0.21331689272503082, + "grad_norm": 2.303295425870437, + "learning_rate": 1.9870936062026385e-05, + "loss": 1.3425, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 173 + }, + { + "epoch": 0.21454993834771888, + "grad_norm": 1.9108233071934615, + "learning_rate": 1.9868816245517615e-05, + "loss": 1.1926, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 174 + }, + { + "epoch": 0.2157829839704069, + "grad_norm": 2.2539584509521173, + "learning_rate": 1.986667927693411e-05, + "loss": 1.222, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 175 + }, + { + "epoch": 0.21701602959309493, + "grad_norm": 2.006165392274905, + "learning_rate": 1.986452515998993e-05, + "loss": 1.2853, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 176 + }, + { + "epoch": 0.218249075215783, + "grad_norm": 2.1337815004505454, + "learning_rate": 1.9862353898428948e-05, + "loss": 1.2671, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 177 + }, + { + "epoch": 0.21948212083847102, + "grad_norm": 2.062911126777883, + "learning_rate": 1.986016549602484e-05, + "loss": 1.2397, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 178 + }, + { + "epoch": 0.22071516646115907, + "grad_norm": 2.115740291239205, + "learning_rate": 1.9857959956581053e-05, + "loss": 1.2534, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 179 + }, + { + "epoch": 0.2219482120838471, + "grad_norm": 2.3733626044353837, + "learning_rate": 1.9855737283930843e-05, + "loss": 1.2082, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 180 + }, + { + "epoch": 0.22318125770653513, + "grad_norm": 2.0972006372187373, + "learning_rate": 1.985349748193722e-05, + "loss": 1.1091, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 181 + }, + { + "epoch": 0.22441430332922319, + "grad_norm": 2.0727561431265635, + "learning_rate": 1.9851240554492985e-05, + "loss": 1.2153, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 182 + }, + { + "epoch": 0.22564734895191121, + "grad_norm": 2.094605431898615, + "learning_rate": 1.9848966505520694e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 183 + }, + { + "epoch": 0.22688039457459927, + "grad_norm": 2.1524188959905866, + "learning_rate": 1.9846675338972654e-05, + "loss": 1.219, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 184 + }, + { + "epoch": 0.2281134401972873, + "grad_norm": 2.008037870627797, + "learning_rate": 1.9844367058830933e-05, + "loss": 1.2635, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 185 + }, + { + "epoch": 0.22934648581997533, + "grad_norm": 2.050448462235658, + "learning_rate": 1.984204166910734e-05, + "loss": 1.187, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 186 + }, + { + "epoch": 0.23057953144266338, + "grad_norm": 2.2727997887406066, + "learning_rate": 1.983969917384342e-05, + "loss": 1.2709, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 187 + }, + { + "epoch": 0.2318125770653514, + "grad_norm": 2.0295960774209414, + "learning_rate": 1.9837339577110443e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 188 + }, + { + "epoch": 0.23304562268803947, + "grad_norm": 2.250775161795276, + "learning_rate": 1.9834962883009406e-05, + "loss": 1.2566, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 189 + }, + { + "epoch": 0.2342786683107275, + "grad_norm": 2.221930554057359, + "learning_rate": 1.9832569095671025e-05, + "loss": 1.2547, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 190 + }, + { + "epoch": 0.23551171393341552, + "grad_norm": 2.071699318441252, + "learning_rate": 1.9830158219255715e-05, + "loss": 1.1712, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 191 + }, + { + "epoch": 0.23674475955610358, + "grad_norm": 1.9444133453118777, + "learning_rate": 1.98277302579536e-05, + "loss": 1.1858, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 192 + }, + { + "epoch": 0.2379778051787916, + "grad_norm": 2.134049888070869, + "learning_rate": 1.9825285215984495e-05, + "loss": 1.2212, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 193 + }, + { + "epoch": 0.23921085080147966, + "grad_norm": 2.1418697560189988, + "learning_rate": 1.9822823097597893e-05, + "loss": 1.2602, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 194 + }, + { + "epoch": 0.2404438964241677, + "grad_norm": 2.049709663111986, + "learning_rate": 1.9820343907072982e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 195 + }, + { + "epoch": 0.24167694204685575, + "grad_norm": 2.1074579631320605, + "learning_rate": 1.9817847648718615e-05, + "loss": 1.1863, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 196 + }, + { + "epoch": 0.24290998766954378, + "grad_norm": 2.279482709995341, + "learning_rate": 1.9815334326873304e-05, + "loss": 1.2225, + "memory/device_mem_reserved(gib)": 22.41, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 197 + }, + { + "epoch": 0.2441430332922318, + "grad_norm": 2.055596994912923, + "learning_rate": 1.9812803945905215e-05, + "loss": 1.2952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 198 + }, + { + "epoch": 0.24537607891491986, + "grad_norm": 2.132145299143716, + "learning_rate": 1.9810256510212178e-05, + "loss": 1.2276, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 199 + }, + { + "epoch": 0.2466091245376079, + "grad_norm": 2.077623874214478, + "learning_rate": 1.9807692024221647e-05, + "loss": 1.226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 200 + }, + { + "epoch": 0.24784217016029594, + "grad_norm": 2.0797643511906356, + "learning_rate": 1.9805110492390723e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 201 + }, + { + "epoch": 0.24907521578298397, + "grad_norm": 2.0930244790953854, + "learning_rate": 1.9802511919206123e-05, + "loss": 1.1984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 202 + }, + { + "epoch": 0.25030826140567203, + "grad_norm": 1.9804465011478614, + "learning_rate": 1.9799896309184192e-05, + "loss": 1.1436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 203 + }, + { + "epoch": 0.25154130702836003, + "grad_norm": 2.054252771935451, + "learning_rate": 1.9797263666870876e-05, + "loss": 1.1913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 204 + }, + { + "epoch": 0.2527743526510481, + "grad_norm": 1.9941485664509049, + "learning_rate": 1.9794613996841723e-05, + "loss": 1.1524, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 205 + }, + { + "epoch": 0.25400739827373614, + "grad_norm": 2.051706932133756, + "learning_rate": 1.9791947303701882e-05, + "loss": 1.2003, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 206 + }, + { + "epoch": 0.25524044389642414, + "grad_norm": 1.9992360549131547, + "learning_rate": 1.9789263592086085e-05, + "loss": 1.2392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 207 + }, + { + "epoch": 0.2564734895191122, + "grad_norm": 2.0558176906965375, + "learning_rate": 1.9786562866658644e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 208 + }, + { + "epoch": 0.25770653514180025, + "grad_norm": 2.2225484776914426, + "learning_rate": 1.9783845132113438e-05, + "loss": 1.1771, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 209 + }, + { + "epoch": 0.2589395807644883, + "grad_norm": 2.2832873135247516, + "learning_rate": 1.978111039317391e-05, + "loss": 1.2226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 210 + }, + { + "epoch": 0.2601726263871763, + "grad_norm": 1.9887162353358598, + "learning_rate": 1.9778358654593054e-05, + "loss": 1.1936, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 211 + }, + { + "epoch": 0.26140567200986436, + "grad_norm": 1.9738209615627333, + "learning_rate": 1.9775589921153417e-05, + "loss": 1.2666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 212 + }, + { + "epoch": 0.2626387176325524, + "grad_norm": 1.8817044406816048, + "learning_rate": 1.9772804197667074e-05, + "loss": 1.1842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 213 + }, + { + "epoch": 0.2638717632552404, + "grad_norm": 2.013799438378123, + "learning_rate": 1.9770001488975632e-05, + "loss": 1.2242, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 214 + }, + { + "epoch": 0.2651048088779285, + "grad_norm": 1.9119015873636653, + "learning_rate": 1.9767181799950222e-05, + "loss": 1.0898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 215 + }, + { + "epoch": 0.26633785450061653, + "grad_norm": 1.9353703261833344, + "learning_rate": 1.976434513549148e-05, + "loss": 1.2241, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 216 + }, + { + "epoch": 0.2675709001233046, + "grad_norm": 1.8684279347954125, + "learning_rate": 1.9761491500529555e-05, + "loss": 1.209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 217 + }, + { + "epoch": 0.2688039457459926, + "grad_norm": 2.1880109676776556, + "learning_rate": 1.9758620900024083e-05, + "loss": 1.1956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 218 + }, + { + "epoch": 0.27003699136868065, + "grad_norm": 2.083271319194177, + "learning_rate": 1.9755733338964186e-05, + "loss": 1.3062, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 219 + }, + { + "epoch": 0.2712700369913687, + "grad_norm": 2.093319969744569, + "learning_rate": 1.9752828822368468e-05, + "loss": 1.2083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 220 + }, + { + "epoch": 0.2725030826140567, + "grad_norm": 1.8811834241518346, + "learning_rate": 1.9749907355285e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 221 + }, + { + "epoch": 0.27373612823674476, + "grad_norm": 2.012899236904161, + "learning_rate": 1.9746968942791314e-05, + "loss": 1.1476, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 222 + }, + { + "epoch": 0.2749691738594328, + "grad_norm": 1.9990139044183095, + "learning_rate": 1.9744013589994397e-05, + "loss": 1.1257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 223 + }, + { + "epoch": 0.2762022194821208, + "grad_norm": 1.9924345221184356, + "learning_rate": 1.9741041302030666e-05, + "loss": 1.224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 224 + }, + { + "epoch": 0.27743526510480887, + "grad_norm": 2.2180474451395344, + "learning_rate": 1.9738052084065982e-05, + "loss": 1.1713, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 225 + }, + { + "epoch": 0.2786683107274969, + "grad_norm": 1.983690334551248, + "learning_rate": 1.973504594129563e-05, + "loss": 1.179, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 226 + }, + { + "epoch": 0.279901356350185, + "grad_norm": 2.283777738854751, + "learning_rate": 1.97320228789443e-05, + "loss": 1.2869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 227 + }, + { + "epoch": 0.281134401972873, + "grad_norm": 1.943369966527272, + "learning_rate": 1.9728982902266107e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 228 + }, + { + "epoch": 0.28236744759556104, + "grad_norm": 2.0711638809016333, + "learning_rate": 1.9725926016544544e-05, + "loss": 1.16, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 229 + }, + { + "epoch": 0.2836004932182491, + "grad_norm": 1.9786842049852629, + "learning_rate": 1.972285222709251e-05, + "loss": 1.2338, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 230 + }, + { + "epoch": 0.2848335388409371, + "grad_norm": 1.9447068528661613, + "learning_rate": 1.9719761539252268e-05, + "loss": 1.1827, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 231 + }, + { + "epoch": 0.28606658446362515, + "grad_norm": 2.0161169975225475, + "learning_rate": 1.971665395839545e-05, + "loss": 1.117, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 232 + }, + { + "epoch": 0.2872996300863132, + "grad_norm": 2.181216178694371, + "learning_rate": 1.9713529489923067e-05, + "loss": 1.1807, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 233 + }, + { + "epoch": 0.2885326757090012, + "grad_norm": 2.254001309068264, + "learning_rate": 1.9710388139265458e-05, + "loss": 1.0036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 234 + }, + { + "epoch": 0.28976572133168926, + "grad_norm": 2.028042430685245, + "learning_rate": 1.9707229911882318e-05, + "loss": 1.1175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 235 + }, + { + "epoch": 0.2909987669543773, + "grad_norm": 2.0601619279255177, + "learning_rate": 1.9704054813262666e-05, + "loss": 1.2758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 236 + }, + { + "epoch": 0.2922318125770654, + "grad_norm": 2.125107661538684, + "learning_rate": 1.9700862848924852e-05, + "loss": 1.2611, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 237 + }, + { + "epoch": 0.2934648581997534, + "grad_norm": 2.0416663193733746, + "learning_rate": 1.9697654024416528e-05, + "loss": 1.1986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 238 + }, + { + "epoch": 0.29469790382244143, + "grad_norm": 2.0193702602947554, + "learning_rate": 1.9694428345314657e-05, + "loss": 1.181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 239 + }, + { + "epoch": 0.2959309494451295, + "grad_norm": 2.0150819278141148, + "learning_rate": 1.9691185817225493e-05, + "loss": 1.1118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 240 + }, + { + "epoch": 0.2971639950678175, + "grad_norm": 2.355208602911491, + "learning_rate": 1.9687926445784576e-05, + "loss": 1.1013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 241 + }, + { + "epoch": 0.29839704069050554, + "grad_norm": 1.9770821951357245, + "learning_rate": 1.968465023665671e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 242 + }, + { + "epoch": 0.2996300863131936, + "grad_norm": 1.992866162682536, + "learning_rate": 1.9681357195535974e-05, + "loss": 1.2264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 243 + }, + { + "epoch": 0.3008631319358816, + "grad_norm": 1.9560049490402185, + "learning_rate": 1.9678047328145702e-05, + "loss": 1.1903, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 244 + }, + { + "epoch": 0.30209617755856966, + "grad_norm": 2.0070855957858114, + "learning_rate": 1.967472064023846e-05, + "loss": 1.1793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 245 + }, + { + "epoch": 0.3033292231812577, + "grad_norm": 2.0661988646186313, + "learning_rate": 1.9671377137596064e-05, + "loss": 1.2222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 246 + }, + { + "epoch": 0.30456226880394577, + "grad_norm": 2.1056352825098816, + "learning_rate": 1.9668016826029538e-05, + "loss": 1.2107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 247 + }, + { + "epoch": 0.30579531442663377, + "grad_norm": 1.98815122638957, + "learning_rate": 1.9664639711379135e-05, + "loss": 1.2816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 248 + }, + { + "epoch": 0.3070283600493218, + "grad_norm": 2.1440514280412737, + "learning_rate": 1.9661245799514298e-05, + "loss": 1.2021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 249 + }, + { + "epoch": 0.3082614056720099, + "grad_norm": 2.0231899704902805, + "learning_rate": 1.9657835096333676e-05, + "loss": 1.2358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 250 + }, + { + "epoch": 0.3094944512946979, + "grad_norm": 1.9883606430802652, + "learning_rate": 1.9654407607765098e-05, + "loss": 1.1847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 251 + }, + { + "epoch": 0.31072749691738594, + "grad_norm": 2.1163621969666786, + "learning_rate": 1.965096333976556e-05, + "loss": 1.2595, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 252 + }, + { + "epoch": 0.311960542540074, + "grad_norm": 1.9898069180734743, + "learning_rate": 1.9647502298321223e-05, + "loss": 1.2286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 253 + }, + { + "epoch": 0.31319358816276205, + "grad_norm": 1.929003891101706, + "learning_rate": 1.9644024489447403e-05, + "loss": 1.2394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 254 + }, + { + "epoch": 0.31442663378545005, + "grad_norm": 1.98661541473719, + "learning_rate": 1.9640529919188562e-05, + "loss": 1.1945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 255 + }, + { + "epoch": 0.3156596794081381, + "grad_norm": 1.9483621179175639, + "learning_rate": 1.9637018593618285e-05, + "loss": 1.1458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 256 + }, + { + "epoch": 0.31689272503082616, + "grad_norm": 1.8370569270185535, + "learning_rate": 1.9633490518839285e-05, + "loss": 1.1627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 257 + }, + { + "epoch": 0.31812577065351416, + "grad_norm": 2.2350844635149376, + "learning_rate": 1.9629945700983376e-05, + "loss": 1.1717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 258 + }, + { + "epoch": 0.3193588162762022, + "grad_norm": 2.3938085963940847, + "learning_rate": 1.9626384146211482e-05, + "loss": 1.2982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 259 + }, + { + "epoch": 0.3205918618988903, + "grad_norm": 1.9382250145807791, + "learning_rate": 1.962280586071361e-05, + "loss": 1.1414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 260 + }, + { + "epoch": 0.3218249075215783, + "grad_norm": 2.048034832679974, + "learning_rate": 1.9619210850708844e-05, + "loss": 1.2367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 261 + }, + { + "epoch": 0.32305795314426633, + "grad_norm": 1.868035630300947, + "learning_rate": 1.9615599122445338e-05, + "loss": 1.1582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 262 + }, + { + "epoch": 0.3242909987669544, + "grad_norm": 1.9842993509323321, + "learning_rate": 1.9611970682200308e-05, + "loss": 1.1214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 263 + }, + { + "epoch": 0.32552404438964244, + "grad_norm": 2.009754214636891, + "learning_rate": 1.9608325536280003e-05, + "loss": 1.2488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 264 + }, + { + "epoch": 0.32675709001233044, + "grad_norm": 2.0322676023608097, + "learning_rate": 1.9604663691019715e-05, + "loss": 1.1514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 265 + }, + { + "epoch": 0.3279901356350185, + "grad_norm": 1.9898336979024067, + "learning_rate": 1.960098515278376e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 266 + }, + { + "epoch": 0.32922318125770655, + "grad_norm": 2.1120578466218007, + "learning_rate": 1.959728992796546e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 267 + }, + { + "epoch": 0.33045622688039455, + "grad_norm": 2.230681018136028, + "learning_rate": 1.959357802298715e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 268 + }, + { + "epoch": 0.3316892725030826, + "grad_norm": 2.121990574572694, + "learning_rate": 1.9589849444300142e-05, + "loss": 1.1686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 269 + }, + { + "epoch": 0.33292231812577067, + "grad_norm": 1.9793276124371497, + "learning_rate": 1.958610419838473e-05, + "loss": 1.2605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 270 + }, + { + "epoch": 0.33415536374845867, + "grad_norm": 2.160304448339165, + "learning_rate": 1.958234229175019e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 271 + }, + { + "epoch": 0.3353884093711467, + "grad_norm": 2.0375075579110646, + "learning_rate": 1.9578563730934733e-05, + "loss": 1.1282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 272 + }, + { + "epoch": 0.3366214549938348, + "grad_norm": 2.0496730732316313, + "learning_rate": 1.9574768522505528e-05, + "loss": 1.2336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 273 + }, + { + "epoch": 0.33785450061652283, + "grad_norm": 2.0096053009394987, + "learning_rate": 1.9570956673058678e-05, + "loss": 1.1737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 274 + }, + { + "epoch": 0.33908754623921084, + "grad_norm": 2.064991805627586, + "learning_rate": 1.9567128189219203e-05, + "loss": 1.2067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 275 + }, + { + "epoch": 0.3403205918618989, + "grad_norm": 2.016629291677699, + "learning_rate": 1.9563283077641032e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 276 + }, + { + "epoch": 0.34155363748458695, + "grad_norm": 2.026571915003702, + "learning_rate": 1.9559421345007004e-05, + "loss": 0.9722, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 277 + }, + { + "epoch": 0.34278668310727495, + "grad_norm": 2.0408528020000896, + "learning_rate": 1.9555542998028836e-05, + "loss": 1.0951, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 278 + }, + { + "epoch": 0.344019728729963, + "grad_norm": 2.4490550716870336, + "learning_rate": 1.955164804344712e-05, + "loss": 1.2195, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 279 + }, + { + "epoch": 0.34525277435265106, + "grad_norm": 2.0298228033571517, + "learning_rate": 1.9547736488031322e-05, + "loss": 1.2091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 280 + }, + { + "epoch": 0.34648581997533906, + "grad_norm": 1.928172339548102, + "learning_rate": 1.9543808338579746e-05, + "loss": 1.2527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 281 + }, + { + "epoch": 0.3477188655980271, + "grad_norm": 1.9880687290891643, + "learning_rate": 1.953986360191955e-05, + "loss": 1.0397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 282 + }, + { + "epoch": 0.34895191122071517, + "grad_norm": 2.3895656693876823, + "learning_rate": 1.9535902284906717e-05, + "loss": 1.2004, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 283 + }, + { + "epoch": 0.35018495684340323, + "grad_norm": 1.9995303874089259, + "learning_rate": 1.9531924394426036e-05, + "loss": 1.2001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 284 + }, + { + "epoch": 0.35141800246609123, + "grad_norm": 1.9751518073159213, + "learning_rate": 1.9527929937391118e-05, + "loss": 1.0906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 285 + }, + { + "epoch": 0.3526510480887793, + "grad_norm": 2.002111448708798, + "learning_rate": 1.9523918920744357e-05, + "loss": 1.0845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 286 + }, + { + "epoch": 0.35388409371146734, + "grad_norm": 2.0131766303020036, + "learning_rate": 1.9519891351456926e-05, + "loss": 1.1764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 287 + }, + { + "epoch": 0.35511713933415534, + "grad_norm": 2.2331356025573785, + "learning_rate": 1.9515847236528772e-05, + "loss": 1.225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 288 + }, + { + "epoch": 0.3563501849568434, + "grad_norm": 2.0628294266012204, + "learning_rate": 1.9511786582988597e-05, + "loss": 1.2259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 289 + }, + { + "epoch": 0.35758323057953145, + "grad_norm": 1.9151442433897523, + "learning_rate": 1.9507709397893846e-05, + "loss": 1.2103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 290 + }, + { + "epoch": 0.3588162762022195, + "grad_norm": 1.958231535128413, + "learning_rate": 1.9503615688330698e-05, + "loss": 1.105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 291 + }, + { + "epoch": 0.3600493218249075, + "grad_norm": 2.149244076290205, + "learning_rate": 1.9499505461414045e-05, + "loss": 1.2441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 292 + }, + { + "epoch": 0.36128236744759556, + "grad_norm": 2.017117422238215, + "learning_rate": 1.9495378724287498e-05, + "loss": 1.1007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 293 + }, + { + "epoch": 0.3625154130702836, + "grad_norm": 1.9210343338897897, + "learning_rate": 1.949123548412335e-05, + "loss": 1.0755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 294 + }, + { + "epoch": 0.3637484586929716, + "grad_norm": 1.9955568267380113, + "learning_rate": 1.9487075748122586e-05, + "loss": 1.1702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 295 + }, + { + "epoch": 0.3649815043156597, + "grad_norm": 2.014663690647869, + "learning_rate": 1.948289952351486e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 296 + }, + { + "epoch": 0.36621454993834773, + "grad_norm": 2.0560814736523922, + "learning_rate": 1.9478706817558474e-05, + "loss": 1.1981, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 297 + }, + { + "epoch": 0.36744759556103573, + "grad_norm": 2.072244571774935, + "learning_rate": 1.947449763754039e-05, + "loss": 1.2293, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 298 + }, + { + "epoch": 0.3686806411837238, + "grad_norm": 2.3134145282189724, + "learning_rate": 1.947027199077619e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 299 + }, + { + "epoch": 0.36991368680641185, + "grad_norm": 1.9505648569157141, + "learning_rate": 1.9466029884610072e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 300 + }, + { + "epoch": 0.3711467324290999, + "grad_norm": 1.9850708859825466, + "learning_rate": 1.946177132641486e-05, + "loss": 1.0863, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 301 + }, + { + "epoch": 0.3723797780517879, + "grad_norm": 2.3267730903818755, + "learning_rate": 1.945749632359195e-05, + "loss": 1.1405, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 302 + }, + { + "epoch": 0.37361282367447596, + "grad_norm": 2.403438763479749, + "learning_rate": 1.9453204883571333e-05, + "loss": 1.1977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 303 + }, + { + "epoch": 0.374845869297164, + "grad_norm": 2.028737664163498, + "learning_rate": 1.9448897013811558e-05, + "loss": 1.2029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 304 + }, + { + "epoch": 0.376078914919852, + "grad_norm": 1.9981856925053472, + "learning_rate": 1.9444572721799743e-05, + "loss": 1.1873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 305 + }, + { + "epoch": 0.37731196054254007, + "grad_norm": 1.968701546642257, + "learning_rate": 1.9440232015051526e-05, + "loss": 1.1721, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 306 + }, + { + "epoch": 0.3785450061652281, + "grad_norm": 1.864390281708884, + "learning_rate": 1.9435874901111102e-05, + "loss": 1.1797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 307 + }, + { + "epoch": 0.3797780517879161, + "grad_norm": 2.0705929640726577, + "learning_rate": 1.9431501387551158e-05, + "loss": 1.1453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 308 + }, + { + "epoch": 0.3810110974106042, + "grad_norm": 1.884359293663656, + "learning_rate": 1.942711148197289e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 309 + }, + { + "epoch": 0.38224414303329224, + "grad_norm": 1.936787839699547, + "learning_rate": 1.9422705192005994e-05, + "loss": 1.2022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 310 + }, + { + "epoch": 0.3834771886559803, + "grad_norm": 1.9636401987324612, + "learning_rate": 1.941828252530863e-05, + "loss": 1.2015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 311 + }, + { + "epoch": 0.3847102342786683, + "grad_norm": 2.0693459412185495, + "learning_rate": 1.9413843489567423e-05, + "loss": 1.054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 312 + }, + { + "epoch": 0.38594327990135635, + "grad_norm": 2.0182958064572256, + "learning_rate": 1.9409388092497453e-05, + "loss": 1.1086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 313 + }, + { + "epoch": 0.3871763255240444, + "grad_norm": 2.093433911405551, + "learning_rate": 1.940491634184223e-05, + "loss": 1.1617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 314 + }, + { + "epoch": 0.3884093711467324, + "grad_norm": 2.018372038783884, + "learning_rate": 1.9400428245373693e-05, + "loss": 1.0813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 315 + }, + { + "epoch": 0.38964241676942046, + "grad_norm": 1.9703329339612685, + "learning_rate": 1.9395923810892184e-05, + "loss": 1.0768, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 316 + }, + { + "epoch": 0.3908754623921085, + "grad_norm": 1.9327973664586777, + "learning_rate": 1.939140304622644e-05, + "loss": 1.1157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 317 + }, + { + "epoch": 0.3921085080147966, + "grad_norm": 2.019497002973341, + "learning_rate": 1.938686595923359e-05, + "loss": 0.9835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 318 + }, + { + "epoch": 0.3933415536374846, + "grad_norm": 2.089968778066188, + "learning_rate": 1.9382312557799116e-05, + "loss": 1.0213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 319 + }, + { + "epoch": 0.39457459926017263, + "grad_norm": 2.0510099518796006, + "learning_rate": 1.937774284983687e-05, + "loss": 1.099, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 320 + }, + { + "epoch": 0.3958076448828607, + "grad_norm": 2.0175024562414063, + "learning_rate": 1.9373156843289037e-05, + "loss": 1.0764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 321 + }, + { + "epoch": 0.3970406905055487, + "grad_norm": 2.020977606778504, + "learning_rate": 1.9368554546126126e-05, + "loss": 1.0562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 322 + }, + { + "epoch": 0.39827373612823674, + "grad_norm": 2.2138761252967125, + "learning_rate": 1.936393596634697e-05, + "loss": 1.196, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 323 + }, + { + "epoch": 0.3995067817509248, + "grad_norm": 1.855560397827274, + "learning_rate": 1.9359301111978683e-05, + "loss": 1.0725, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 324 + }, + { + "epoch": 0.4007398273736128, + "grad_norm": 2.0805790476522636, + "learning_rate": 1.9354649991076684e-05, + "loss": 1.1665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 325 + }, + { + "epoch": 0.40197287299630086, + "grad_norm": 1.8807037569919354, + "learning_rate": 1.9349982611724654e-05, + "loss": 1.1366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 326 + }, + { + "epoch": 0.4032059186189889, + "grad_norm": 1.9286740986289053, + "learning_rate": 1.9345298982034536e-05, + "loss": 1.132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 327 + }, + { + "epoch": 0.40443896424167697, + "grad_norm": 2.0846402787959786, + "learning_rate": 1.9340599110146504e-05, + "loss": 1.1055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 328 + }, + { + "epoch": 0.40567200986436497, + "grad_norm": 2.2183587793468993, + "learning_rate": 1.9335883004228974e-05, + "loss": 1.1544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 329 + }, + { + "epoch": 0.406905055487053, + "grad_norm": 2.2213015226303643, + "learning_rate": 1.9331150672478575e-05, + "loss": 1.1111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 330 + }, + { + "epoch": 0.4081381011097411, + "grad_norm": 2.2277113238319726, + "learning_rate": 1.9326402123120132e-05, + "loss": 1.0451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 331 + }, + { + "epoch": 0.4093711467324291, + "grad_norm": 2.259911506493339, + "learning_rate": 1.9321637364406658e-05, + "loss": 1.0236, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 332 + }, + { + "epoch": 0.41060419235511714, + "grad_norm": 2.3573090703399417, + "learning_rate": 1.931685640461934e-05, + "loss": 1.1412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 333 + }, + { + "epoch": 0.4118372379778052, + "grad_norm": 2.1097456297814468, + "learning_rate": 1.9312059252067517e-05, + "loss": 1.1901, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 334 + }, + { + "epoch": 0.4130702836004932, + "grad_norm": 2.121952527251297, + "learning_rate": 1.930724591508868e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 335 + }, + { + "epoch": 0.41430332922318125, + "grad_norm": 2.091327214317948, + "learning_rate": 1.930241640204844e-05, + "loss": 1.0135, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 336 + }, + { + "epoch": 0.4155363748458693, + "grad_norm": 2.1272216173587255, + "learning_rate": 1.929757072134053e-05, + "loss": 1.1754, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 337 + }, + { + "epoch": 0.41676942046855736, + "grad_norm": 1.9973226929736474, + "learning_rate": 1.929270888138678e-05, + "loss": 1.1392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 338 + }, + { + "epoch": 0.41800246609124536, + "grad_norm": 1.9318995412408386, + "learning_rate": 1.9287830890637093e-05, + "loss": 1.0762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 339 + }, + { + "epoch": 0.4192355117139334, + "grad_norm": 2.785840304521055, + "learning_rate": 1.9282936757569463e-05, + "loss": 1.1699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 340 + }, + { + "epoch": 0.4204685573366215, + "grad_norm": 2.171910512782072, + "learning_rate": 1.9278026490689925e-05, + "loss": 1.1675, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 341 + }, + { + "epoch": 0.4217016029593095, + "grad_norm": 2.052852932277261, + "learning_rate": 1.927310009853256e-05, + "loss": 1.1501, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 342 + }, + { + "epoch": 0.42293464858199753, + "grad_norm": 1.9538275366336955, + "learning_rate": 1.926815758965947e-05, + "loss": 1.1934, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 343 + }, + { + "epoch": 0.4241676942046856, + "grad_norm": 1.9772612426660126, + "learning_rate": 1.9263198972660774e-05, + "loss": 1.1061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 344 + }, + { + "epoch": 0.4254007398273736, + "grad_norm": 1.9522364508331842, + "learning_rate": 1.9258224256154588e-05, + "loss": 1.101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 345 + }, + { + "epoch": 0.42663378545006164, + "grad_norm": 2.1060276682972714, + "learning_rate": 1.9253233448787e-05, + "loss": 1.2094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 346 + }, + { + "epoch": 0.4278668310727497, + "grad_norm": 2.020183264992464, + "learning_rate": 1.9248226559232074e-05, + "loss": 1.2012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 347 + }, + { + "epoch": 0.42909987669543775, + "grad_norm": 2.1361451948859256, + "learning_rate": 1.9243203596191815e-05, + "loss": 1.2714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 348 + }, + { + "epoch": 0.43033292231812575, + "grad_norm": 1.890346669461953, + "learning_rate": 1.9238164568396177e-05, + "loss": 1.067, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 349 + }, + { + "epoch": 0.4315659679408138, + "grad_norm": 1.908858157385325, + "learning_rate": 1.9233109484603025e-05, + "loss": 1.1123, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 350 + }, + { + "epoch": 0.43279901356350187, + "grad_norm": 1.9095667461482675, + "learning_rate": 1.9228038353598122e-05, + "loss": 1.0834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 351 + }, + { + "epoch": 0.43403205918618987, + "grad_norm": 1.8950946192756821, + "learning_rate": 1.9222951184195146e-05, + "loss": 1.0686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 352 + }, + { + "epoch": 0.4352651048088779, + "grad_norm": 2.5409486063614066, + "learning_rate": 1.921784798523562e-05, + "loss": 1.1168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 353 + }, + { + "epoch": 0.436498150431566, + "grad_norm": 1.9251760877216109, + "learning_rate": 1.9212728765588954e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 354 + }, + { + "epoch": 0.43773119605425403, + "grad_norm": 2.12327469473301, + "learning_rate": 1.9207593534152383e-05, + "loss": 1.1871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 355 + }, + { + "epoch": 0.43896424167694204, + "grad_norm": 2.0404259513933214, + "learning_rate": 1.9202442299850977e-05, + "loss": 1.0985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 356 + }, + { + "epoch": 0.4401972872996301, + "grad_norm": 2.10787306308974, + "learning_rate": 1.9197275071637617e-05, + "loss": 1.0063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 357 + }, + { + "epoch": 0.44143033292231815, + "grad_norm": 2.2402685683097996, + "learning_rate": 1.919209185849299e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 358 + }, + { + "epoch": 0.44266337854500615, + "grad_norm": 2.148922705064427, + "learning_rate": 1.9186892669425546e-05, + "loss": 1.0889, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 359 + }, + { + "epoch": 0.4438964241676942, + "grad_norm": 1.8789248361497928, + "learning_rate": 1.9181677513471525e-05, + "loss": 1.1796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 360 + }, + { + "epoch": 0.44512946979038226, + "grad_norm": 1.8886466190263609, + "learning_rate": 1.9176446399694903e-05, + "loss": 1.1494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 361 + }, + { + "epoch": 0.44636251541307026, + "grad_norm": 1.9432580511904547, + "learning_rate": 1.9171199337187395e-05, + "loss": 1.2599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 362 + }, + { + "epoch": 0.4475955610357583, + "grad_norm": 1.8820277177030855, + "learning_rate": 1.916593633506843e-05, + "loss": 1.1884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 363 + }, + { + "epoch": 0.44882860665844637, + "grad_norm": 2.249700768154195, + "learning_rate": 1.9160657402485147e-05, + "loss": 1.0503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 364 + }, + { + "epoch": 0.45006165228113443, + "grad_norm": 2.082324666968816, + "learning_rate": 1.9155362548612366e-05, + "loss": 1.2017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 365 + }, + { + "epoch": 0.45129469790382243, + "grad_norm": 2.2090950658394193, + "learning_rate": 1.915005178265259e-05, + "loss": 1.1742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 366 + }, + { + "epoch": 0.4525277435265105, + "grad_norm": 2.2007793765172683, + "learning_rate": 1.914472511383596e-05, + "loss": 1.2617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 367 + }, + { + "epoch": 0.45376078914919854, + "grad_norm": 1.9166813253050292, + "learning_rate": 1.913938255142027e-05, + "loss": 1.1444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 368 + }, + { + "epoch": 0.45499383477188654, + "grad_norm": 1.908891419449054, + "learning_rate": 1.913402410469093e-05, + "loss": 1.1139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 369 + }, + { + "epoch": 0.4562268803945746, + "grad_norm": 2.0434795360025566, + "learning_rate": 1.9128649782960965e-05, + "loss": 1.1761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 370 + }, + { + "epoch": 0.45745992601726265, + "grad_norm": 2.2319795236629165, + "learning_rate": 1.9123259595570978e-05, + "loss": 0.9997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 371 + }, + { + "epoch": 0.45869297163995065, + "grad_norm": 1.9325225410679332, + "learning_rate": 1.911785355188916e-05, + "loss": 1.1191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 372 + }, + { + "epoch": 0.4599260172626387, + "grad_norm": 2.3718583427313384, + "learning_rate": 1.9112431661311248e-05, + "loss": 1.1801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 373 + }, + { + "epoch": 0.46115906288532676, + "grad_norm": 2.2778458302261853, + "learning_rate": 1.9106993933260535e-05, + "loss": 1.1973, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 374 + }, + { + "epoch": 0.4623921085080148, + "grad_norm": 2.104485888382934, + "learning_rate": 1.910154037718783e-05, + "loss": 1.206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 375 + }, + { + "epoch": 0.4636251541307028, + "grad_norm": 2.149849667802547, + "learning_rate": 1.909607100257145e-05, + "loss": 1.0479, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 376 + }, + { + "epoch": 0.4648581997533909, + "grad_norm": 2.242058517922103, + "learning_rate": 1.9090585818917205e-05, + "loss": 1.0902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 377 + }, + { + "epoch": 0.46609124537607893, + "grad_norm": 2.0943438214587307, + "learning_rate": 1.9085084835758394e-05, + "loss": 1.2428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 378 + }, + { + "epoch": 0.46732429099876693, + "grad_norm": 2.1534146036573647, + "learning_rate": 1.907956806265576e-05, + "loss": 1.0392, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 379 + }, + { + "epoch": 0.468557336621455, + "grad_norm": 2.2361240083173324, + "learning_rate": 1.9074035509197495e-05, + "loss": 1.1337, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 380 + }, + { + "epoch": 0.46979038224414305, + "grad_norm": 2.0666798517773795, + "learning_rate": 1.906848718499922e-05, + "loss": 1.0574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 381 + }, + { + "epoch": 0.47102342786683105, + "grad_norm": 2.0056597359218094, + "learning_rate": 1.906292309970396e-05, + "loss": 1.1982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 382 + }, + { + "epoch": 0.4722564734895191, + "grad_norm": 1.931784648165126, + "learning_rate": 1.9057343262982136e-05, + "loss": 1.0671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 383 + }, + { + "epoch": 0.47348951911220716, + "grad_norm": 2.2650382784813408, + "learning_rate": 1.905174768453155e-05, + "loss": 1.1384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 384 + }, + { + "epoch": 0.4747225647348952, + "grad_norm": 2.2386337800038594, + "learning_rate": 1.904613637407735e-05, + "loss": 1.169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 385 + }, + { + "epoch": 0.4759556103575832, + "grad_norm": 1.985439502238856, + "learning_rate": 1.9040509341372045e-05, + "loss": 1.0544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 386 + }, + { + "epoch": 0.47718865598027127, + "grad_norm": 1.9991995362972141, + "learning_rate": 1.903486659619545e-05, + "loss": 1.1048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 387 + }, + { + "epoch": 0.4784217016029593, + "grad_norm": 1.938313304762426, + "learning_rate": 1.90292081483547e-05, + "loss": 1.0217, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 388 + }, + { + "epoch": 0.4796547472256473, + "grad_norm": 2.0593580639988605, + "learning_rate": 1.9023534007684217e-05, + "loss": 1.168, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 389 + }, + { + "epoch": 0.4808877928483354, + "grad_norm": 2.0113392094384723, + "learning_rate": 1.90178441840457e-05, + "loss": 1.2114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 390 + }, + { + "epoch": 0.48212083847102344, + "grad_norm": 1.9865461362024583, + "learning_rate": 1.901213868732811e-05, + "loss": 1.0818, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 391 + }, + { + "epoch": 0.4833538840937115, + "grad_norm": 1.9044551829036014, + "learning_rate": 1.9006417527447634e-05, + "loss": 1.1046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 392 + }, + { + "epoch": 0.4845869297163995, + "grad_norm": 1.8508018232087575, + "learning_rate": 1.9000680714347694e-05, + "loss": 1.1411, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 393 + }, + { + "epoch": 0.48581997533908755, + "grad_norm": 1.98419367965867, + "learning_rate": 1.8994928257998907e-05, + "loss": 0.9804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 394 + }, + { + "epoch": 0.4870530209617756, + "grad_norm": 2.1671038349199567, + "learning_rate": 1.8989160168399093e-05, + "loss": 1.1656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 395 + }, + { + "epoch": 0.4882860665844636, + "grad_norm": 2.0570349889938284, + "learning_rate": 1.898337645557323e-05, + "loss": 1.2111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 396 + }, + { + "epoch": 0.48951911220715166, + "grad_norm": 1.9619760913338724, + "learning_rate": 1.8977577129573452e-05, + "loss": 1.0932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 397 + }, + { + "epoch": 0.4907521578298397, + "grad_norm": 1.9608046473112009, + "learning_rate": 1.897176220047903e-05, + "loss": 1.1389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 398 + }, + { + "epoch": 0.4919852034525277, + "grad_norm": 1.9645035824539876, + "learning_rate": 1.8965931678396357e-05, + "loss": 1.1056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 399 + }, + { + "epoch": 0.4932182490752158, + "grad_norm": 1.961734329156117, + "learning_rate": 1.896008557345892e-05, + "loss": 1.1322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 400 + }, + { + "epoch": 0.49445129469790383, + "grad_norm": 1.844560254648902, + "learning_rate": 1.8954223895827288e-05, + "loss": 1.0712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 401 + }, + { + "epoch": 0.4956843403205919, + "grad_norm": 2.1465454734369933, + "learning_rate": 1.8948346655689108e-05, + "loss": 1.1294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 402 + }, + { + "epoch": 0.4969173859432799, + "grad_norm": 2.078244569426143, + "learning_rate": 1.8942453863259058e-05, + "loss": 1.0546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 403 + }, + { + "epoch": 0.49815043156596794, + "grad_norm": 2.264508341916039, + "learning_rate": 1.893654552877886e-05, + "loss": 1.0568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 404 + }, + { + "epoch": 0.499383477188656, + "grad_norm": 2.0051680479216696, + "learning_rate": 1.893062166251724e-05, + "loss": 1.0156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 405 + }, + { + "epoch": 0.5006165228113441, + "grad_norm": 2.1569911537953725, + "learning_rate": 1.8924682274769925e-05, + "loss": 1.173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 406 + }, + { + "epoch": 0.5018495684340321, + "grad_norm": 2.002983438517559, + "learning_rate": 1.8918727375859608e-05, + "loss": 1.0073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 407 + }, + { + "epoch": 0.5030826140567201, + "grad_norm": 1.9516175334775212, + "learning_rate": 1.891275697613595e-05, + "loss": 1.1404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 408 + }, + { + "epoch": 0.5043156596794082, + "grad_norm": 2.1494580236712832, + "learning_rate": 1.890677108597555e-05, + "loss": 1.1688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 409 + }, + { + "epoch": 0.5055487053020962, + "grad_norm": 2.025395779691303, + "learning_rate": 1.8900769715781928e-05, + "loss": 0.9756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 410 + }, + { + "epoch": 0.5067817509247842, + "grad_norm": 1.9927648271180456, + "learning_rate": 1.889475287598551e-05, + "loss": 1.1394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 411 + }, + { + "epoch": 0.5080147965474723, + "grad_norm": 1.9818065701766985, + "learning_rate": 1.8888720577043603e-05, + "loss": 1.1683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 412 + }, + { + "epoch": 0.5092478421701603, + "grad_norm": 1.9410101081952555, + "learning_rate": 1.8882672829440393e-05, + "loss": 1.0839, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 413 + }, + { + "epoch": 0.5104808877928483, + "grad_norm": 1.8396805813100958, + "learning_rate": 1.8876609643686906e-05, + "loss": 1.114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 414 + }, + { + "epoch": 0.5117139334155364, + "grad_norm": 2.0105399328568905, + "learning_rate": 1.8870531030321004e-05, + "loss": 1.1439, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 415 + }, + { + "epoch": 0.5129469790382244, + "grad_norm": 2.090098149989323, + "learning_rate": 1.886443699990736e-05, + "loss": 1.1153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 416 + }, + { + "epoch": 0.5141800246609125, + "grad_norm": 2.0517637089595246, + "learning_rate": 1.885832756303745e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 417 + }, + { + "epoch": 0.5154130702836005, + "grad_norm": 3.0738713157895674, + "learning_rate": 1.885220273032951e-05, + "loss": 1.0222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 418 + }, + { + "epoch": 0.5166461159062885, + "grad_norm": 2.6677283836123005, + "learning_rate": 1.884606251242855e-05, + "loss": 1.1038, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 419 + }, + { + "epoch": 0.5178791615289766, + "grad_norm": 2.125547466007061, + "learning_rate": 1.8839906920006315e-05, + "loss": 1.2128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 420 + }, + { + "epoch": 0.5191122071516646, + "grad_norm": 2.1599202425686013, + "learning_rate": 1.883373596376126e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 421 + }, + { + "epoch": 0.5203452527743526, + "grad_norm": 2.1229927533556436, + "learning_rate": 1.8827549654418566e-05, + "loss": 1.1353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 422 + }, + { + "epoch": 0.5215782983970407, + "grad_norm": 1.9567023805342967, + "learning_rate": 1.8821348002730072e-05, + "loss": 1.0584, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 423 + }, + { + "epoch": 0.5228113440197287, + "grad_norm": 1.8989916861837828, + "learning_rate": 1.88151310194743e-05, + "loss": 0.9837, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 424 + }, + { + "epoch": 0.5240443896424167, + "grad_norm": 2.1067680957156014, + "learning_rate": 1.8808898715456415e-05, + "loss": 1.0673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 425 + }, + { + "epoch": 0.5252774352651048, + "grad_norm": 2.6013242993173926, + "learning_rate": 1.88026511015082e-05, + "loss": 1.112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 426 + }, + { + "epoch": 0.5265104808877928, + "grad_norm": 2.0539914273851907, + "learning_rate": 1.8796388188488055e-05, + "loss": 1.1031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 427 + }, + { + "epoch": 0.5277435265104808, + "grad_norm": 2.0386084337690584, + "learning_rate": 1.8790109987280972e-05, + "loss": 1.1556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 428 + }, + { + "epoch": 0.528976572133169, + "grad_norm": 2.093419206457373, + "learning_rate": 1.878381650879851e-05, + "loss": 1.1813, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 429 + }, + { + "epoch": 0.530209617755857, + "grad_norm": 2.2549735586348985, + "learning_rate": 1.877750776397878e-05, + "loss": 1.0663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 430 + }, + { + "epoch": 0.531442663378545, + "grad_norm": 2.1062036310291266, + "learning_rate": 1.877118376378642e-05, + "loss": 1.0455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 431 + }, + { + "epoch": 0.5326757090012331, + "grad_norm": 1.995564286989001, + "learning_rate": 1.87648445192126e-05, + "loss": 0.982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 432 + }, + { + "epoch": 0.5339087546239211, + "grad_norm": 2.41307031725252, + "learning_rate": 1.8758490041274963e-05, + "loss": 1.0778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 433 + }, + { + "epoch": 0.5351418002466092, + "grad_norm": 2.831604603209417, + "learning_rate": 1.875212034101765e-05, + "loss": 1.1345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 434 + }, + { + "epoch": 0.5363748458692972, + "grad_norm": 2.2244728383006604, + "learning_rate": 1.8745735429511236e-05, + "loss": 0.938, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 435 + }, + { + "epoch": 0.5376078914919852, + "grad_norm": 2.0679863467245823, + "learning_rate": 1.8739335317852747e-05, + "loss": 1.0763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 436 + }, + { + "epoch": 0.5388409371146733, + "grad_norm": 2.2454972633071915, + "learning_rate": 1.8732920017165624e-05, + "loss": 1.1207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 437 + }, + { + "epoch": 0.5400739827373613, + "grad_norm": 1.952229569146223, + "learning_rate": 1.8726489538599707e-05, + "loss": 1.0255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 438 + }, + { + "epoch": 0.5413070283600493, + "grad_norm": 1.9236876015939406, + "learning_rate": 1.8720043893331214e-05, + "loss": 0.9933, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 439 + }, + { + "epoch": 0.5425400739827374, + "grad_norm": 1.9736270928366435, + "learning_rate": 1.8713583092562727e-05, + "loss": 1.107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 440 + }, + { + "epoch": 0.5437731196054254, + "grad_norm": 2.1162408965440633, + "learning_rate": 1.870710714752316e-05, + "loss": 1.0441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 441 + }, + { + "epoch": 0.5450061652281134, + "grad_norm": 1.9515336502135756, + "learning_rate": 1.8700616069467753e-05, + "loss": 0.9855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 442 + }, + { + "epoch": 0.5462392108508015, + "grad_norm": 2.566164154732447, + "learning_rate": 1.8694109869678047e-05, + "loss": 1.1289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 443 + }, + { + "epoch": 0.5474722564734895, + "grad_norm": 2.3448298186890026, + "learning_rate": 1.868758855946187e-05, + "loss": 1.0559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 444 + }, + { + "epoch": 0.5487053020961775, + "grad_norm": 2.076679946731619, + "learning_rate": 1.8681052150153305e-05, + "loss": 1.0494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 445 + }, + { + "epoch": 0.5499383477188656, + "grad_norm": 2.281980101143931, + "learning_rate": 1.8674500653112674e-05, + "loss": 1.0456, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 446 + }, + { + "epoch": 0.5511713933415536, + "grad_norm": 2.147469472559495, + "learning_rate": 1.8667934079726537e-05, + "loss": 1.0682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 447 + }, + { + "epoch": 0.5524044389642416, + "grad_norm": 2.0621898279926527, + "learning_rate": 1.866135244140763e-05, + "loss": 1.0527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 448 + }, + { + "epoch": 0.5536374845869297, + "grad_norm": 2.0688727890116785, + "learning_rate": 1.8654755749594905e-05, + "loss": 1.0549, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 449 + }, + { + "epoch": 0.5548705302096177, + "grad_norm": 2.0101374651739325, + "learning_rate": 1.864814401575345e-05, + "loss": 1.002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 450 + }, + { + "epoch": 0.5561035758323057, + "grad_norm": 2.2861388762596344, + "learning_rate": 1.8641517251374514e-05, + "loss": 1.0666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 451 + }, + { + "epoch": 0.5573366214549939, + "grad_norm": 2.006862831841472, + "learning_rate": 1.8634875467975456e-05, + "loss": 1.0061, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 452 + }, + { + "epoch": 0.5585696670776819, + "grad_norm": 2.144270279582644, + "learning_rate": 1.862821867709975e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 453 + }, + { + "epoch": 0.55980271270037, + "grad_norm": 1.9667742618855084, + "learning_rate": 1.862154689031694e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 454 + }, + { + "epoch": 0.561035758323058, + "grad_norm": 1.8857968215229095, + "learning_rate": 1.8614860119222643e-05, + "loss": 1.0861, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 455 + }, + { + "epoch": 0.562268803945746, + "grad_norm": 1.9374803828274445, + "learning_rate": 1.860815837543852e-05, + "loss": 0.9784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 456 + }, + { + "epoch": 0.5635018495684341, + "grad_norm": 2.0764596642416433, + "learning_rate": 1.8601441670612247e-05, + "loss": 1.1005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 457 + }, + { + "epoch": 0.5647348951911221, + "grad_norm": 2.123062949997989, + "learning_rate": 1.8594710016417505e-05, + "loss": 1.1953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 458 + }, + { + "epoch": 0.5659679408138101, + "grad_norm": 2.311252498934928, + "learning_rate": 1.858796342455396e-05, + "loss": 1.1049, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 459 + }, + { + "epoch": 0.5672009864364982, + "grad_norm": 1.9780001048890374, + "learning_rate": 1.8581201906747242e-05, + "loss": 1.1393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 460 + }, + { + "epoch": 0.5684340320591862, + "grad_norm": 1.968874552657262, + "learning_rate": 1.857442547474891e-05, + "loss": 1.1578, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 461 + }, + { + "epoch": 0.5696670776818742, + "grad_norm": 1.9395225862380807, + "learning_rate": 1.8567634140336457e-05, + "loss": 1.0823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 462 + }, + { + "epoch": 0.5709001233045623, + "grad_norm": 1.9422213147988507, + "learning_rate": 1.856082791531327e-05, + "loss": 1.085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 463 + }, + { + "epoch": 0.5721331689272503, + "grad_norm": 1.9095099328124767, + "learning_rate": 1.855400681150862e-05, + "loss": 1.1041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 464 + }, + { + "epoch": 0.5733662145499383, + "grad_norm": 1.9651570734782207, + "learning_rate": 1.8547170840777635e-05, + "loss": 0.9924, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 465 + }, + { + "epoch": 0.5745992601726264, + "grad_norm": 1.9083237960346455, + "learning_rate": 1.8540320015001283e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 466 + }, + { + "epoch": 0.5758323057953144, + "grad_norm": 2.0442809672147484, + "learning_rate": 1.8533454346086345e-05, + "loss": 0.9449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 467 + }, + { + "epoch": 0.5770653514180024, + "grad_norm": 2.839273126939154, + "learning_rate": 1.8526573845965407e-05, + "loss": 1.1446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 468 + }, + { + "epoch": 0.5782983970406905, + "grad_norm": 2.8401028851750083, + "learning_rate": 1.8519678526596827e-05, + "loss": 1.0793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 469 + }, + { + "epoch": 0.5795314426633785, + "grad_norm": 2.538177485476921, + "learning_rate": 1.8512768399964725e-05, + "loss": 1.1066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 470 + }, + { + "epoch": 0.5807644882860666, + "grad_norm": 2.1457149698206526, + "learning_rate": 1.8505843478078944e-05, + "loss": 1.1138, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 471 + }, + { + "epoch": 0.5819975339087546, + "grad_norm": 1.8879825929112735, + "learning_rate": 1.8498903772975056e-05, + "loss": 1.0181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 472 + }, + { + "epoch": 0.5832305795314426, + "grad_norm": 2.083069072716887, + "learning_rate": 1.8491949296714312e-05, + "loss": 1.0307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 473 + }, + { + "epoch": 0.5844636251541308, + "grad_norm": 2.135777387695568, + "learning_rate": 1.848498006138365e-05, + "loss": 1.0346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 474 + }, + { + "epoch": 0.5856966707768188, + "grad_norm": 2.092600182986245, + "learning_rate": 1.8477996079095643e-05, + "loss": 1.041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 475 + }, + { + "epoch": 0.5869297163995068, + "grad_norm": 2.1246789769186414, + "learning_rate": 1.8470997361988514e-05, + "loss": 1.0887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 476 + }, + { + "epoch": 0.5881627620221949, + "grad_norm": 2.062971539274987, + "learning_rate": 1.8463983922226076e-05, + "loss": 1.111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 477 + }, + { + "epoch": 0.5893958076448829, + "grad_norm": 2.092774168134711, + "learning_rate": 1.8456955771997744e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 478 + }, + { + "epoch": 0.5906288532675709, + "grad_norm": 2.23500231174451, + "learning_rate": 1.844991292351849e-05, + "loss": 1.1008, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 479 + }, + { + "epoch": 0.591861898890259, + "grad_norm": 2.04828338908309, + "learning_rate": 1.844285538902884e-05, + "loss": 1.0183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 480 + }, + { + "epoch": 0.593094944512947, + "grad_norm": 2.1680070152194064, + "learning_rate": 1.8435783180794836e-05, + "loss": 1.092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 481 + }, + { + "epoch": 0.594327990135635, + "grad_norm": 2.0706136769588603, + "learning_rate": 1.8428696311108033e-05, + "loss": 1.1438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 482 + }, + { + "epoch": 0.5955610357583231, + "grad_norm": 1.9925622155649279, + "learning_rate": 1.842159479228546e-05, + "loss": 1.0359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 483 + }, + { + "epoch": 0.5967940813810111, + "grad_norm": 1.9425841064456546, + "learning_rate": 1.841447863666961e-05, + "loss": 1.2084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 484 + }, + { + "epoch": 0.5980271270036991, + "grad_norm": 1.929626615569851, + "learning_rate": 1.8407347856628405e-05, + "loss": 1.0615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 485 + }, + { + "epoch": 0.5992601726263872, + "grad_norm": 2.049215260436237, + "learning_rate": 1.8400202464555202e-05, + "loss": 1.0429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 486 + }, + { + "epoch": 0.6004932182490752, + "grad_norm": 2.0464723350262624, + "learning_rate": 1.8393042472868746e-05, + "loss": 1.1504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 487 + }, + { + "epoch": 0.6017262638717632, + "grad_norm": 1.9060167485555373, + "learning_rate": 1.8385867894013144e-05, + "loss": 0.9809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 488 + }, + { + "epoch": 0.6029593094944513, + "grad_norm": 1.9897115419719036, + "learning_rate": 1.837867874045788e-05, + "loss": 1.0573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 489 + }, + { + "epoch": 0.6041923551171393, + "grad_norm": 2.241807622458982, + "learning_rate": 1.8371475024697746e-05, + "loss": 1.1149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 490 + }, + { + "epoch": 0.6054254007398274, + "grad_norm": 2.141493902645165, + "learning_rate": 1.8364256759252855e-05, + "loss": 1.0396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 491 + }, + { + "epoch": 0.6066584463625154, + "grad_norm": 2.168929995419785, + "learning_rate": 1.8357023956668606e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 492 + }, + { + "epoch": 0.6078914919852034, + "grad_norm": 2.095414950529066, + "learning_rate": 1.834977662951566e-05, + "loss": 1.0313, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 493 + }, + { + "epoch": 0.6091245376078915, + "grad_norm": 2.2287918098101516, + "learning_rate": 1.834251479038993e-05, + "loss": 1.0118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 494 + }, + { + "epoch": 0.6103575832305795, + "grad_norm": 2.0458360619004154, + "learning_rate": 1.833523845191254e-05, + "loss": 1.0644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 495 + }, + { + "epoch": 0.6115906288532675, + "grad_norm": 2.1216602719929245, + "learning_rate": 1.8327947626729825e-05, + "loss": 1.1734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 496 + }, + { + "epoch": 0.6128236744759556, + "grad_norm": 2.1445297291236947, + "learning_rate": 1.8320642327513285e-05, + "loss": 0.9661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 497 + }, + { + "epoch": 0.6140567200986436, + "grad_norm": 2.061159959033469, + "learning_rate": 1.831332256695959e-05, + "loss": 1.189, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 498 + }, + { + "epoch": 0.6152897657213316, + "grad_norm": 1.966885029141375, + "learning_rate": 1.8305988357790533e-05, + "loss": 1.0711, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 499 + }, + { + "epoch": 0.6165228113440198, + "grad_norm": 2.127061899116888, + "learning_rate": 1.8298639712753025e-05, + "loss": 0.9351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 500 + }, + { + "epoch": 0.6177558569667078, + "grad_norm": 2.2239600877101835, + "learning_rate": 1.8291276644619066e-05, + "loss": 1.0569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 501 + }, + { + "epoch": 0.6189889025893958, + "grad_norm": 2.0281028990006913, + "learning_rate": 1.8283899166185723e-05, + "loss": 1.0698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 502 + }, + { + "epoch": 0.6202219482120839, + "grad_norm": 2.1224950625771677, + "learning_rate": 1.8276507290275104e-05, + "loss": 0.9905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 503 + }, + { + "epoch": 0.6214549938347719, + "grad_norm": 2.0799215178987, + "learning_rate": 1.8269101029734343e-05, + "loss": 1.0631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 504 + }, + { + "epoch": 0.6226880394574599, + "grad_norm": 2.2495223027498388, + "learning_rate": 1.8261680397435577e-05, + "loss": 0.9869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 505 + }, + { + "epoch": 0.623921085080148, + "grad_norm": 2.036885073926771, + "learning_rate": 1.825424540627592e-05, + "loss": 1.0939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 506 + }, + { + "epoch": 0.625154130702836, + "grad_norm": 2.0423951754968064, + "learning_rate": 1.824679606917744e-05, + "loss": 0.9793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 507 + }, + { + "epoch": 0.6263871763255241, + "grad_norm": 2.118412450801216, + "learning_rate": 1.8239332399087136e-05, + "loss": 1.0214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 508 + }, + { + "epoch": 0.6276202219482121, + "grad_norm": 2.273320358602361, + "learning_rate": 1.8231854408976924e-05, + "loss": 1.1746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 509 + }, + { + "epoch": 0.6288532675709001, + "grad_norm": 1.9352848815012758, + "learning_rate": 1.822436211184361e-05, + "loss": 1.0194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 510 + }, + { + "epoch": 0.6300863131935882, + "grad_norm": 2.1086688743882713, + "learning_rate": 1.8216855520708854e-05, + "loss": 0.9404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 511 + }, + { + "epoch": 0.6313193588162762, + "grad_norm": 2.268255476351323, + "learning_rate": 1.820933464861917e-05, + "loss": 0.9734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 512 + }, + { + "epoch": 0.6325524044389642, + "grad_norm": 2.3533715256185705, + "learning_rate": 1.8201799508645887e-05, + "loss": 1.1794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 513 + }, + { + "epoch": 0.6337854500616523, + "grad_norm": 2.005798886598211, + "learning_rate": 1.8194250113885134e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 514 + }, + { + "epoch": 0.6350184956843403, + "grad_norm": 1.9276850525809597, + "learning_rate": 1.8186686477457817e-05, + "loss": 1.0122, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 515 + }, + { + "epoch": 0.6362515413070283, + "grad_norm": 2.089756823449061, + "learning_rate": 1.8179108612509587e-05, + "loss": 1.1076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 516 + }, + { + "epoch": 0.6374845869297164, + "grad_norm": 2.047489618950505, + "learning_rate": 1.8171516532210825e-05, + "loss": 1.1261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 517 + }, + { + "epoch": 0.6387176325524044, + "grad_norm": 1.9819427784604287, + "learning_rate": 1.8163910249756634e-05, + "loss": 0.9829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 518 + }, + { + "epoch": 0.6399506781750924, + "grad_norm": 2.1061977657317574, + "learning_rate": 1.815628977836678e-05, + "loss": 1.0139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 519 + }, + { + "epoch": 0.6411837237977805, + "grad_norm": 2.0221352739914624, + "learning_rate": 1.8148655131285697e-05, + "loss": 1.0413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 520 + }, + { + "epoch": 0.6424167694204685, + "grad_norm": 2.0126443495430033, + "learning_rate": 1.8141006321782464e-05, + "loss": 1.0737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 521 + }, + { + "epoch": 0.6436498150431565, + "grad_norm": 1.9687053123524947, + "learning_rate": 1.8133343363150762e-05, + "loss": 1.0534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 522 + }, + { + "epoch": 0.6448828606658447, + "grad_norm": 2.1885490221346036, + "learning_rate": 1.812566626870887e-05, + "loss": 1.0746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 523 + }, + { + "epoch": 0.6461159062885327, + "grad_norm": 2.279950249415201, + "learning_rate": 1.8117975051799636e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 524 + }, + { + "epoch": 0.6473489519112207, + "grad_norm": 2.2376288751244986, + "learning_rate": 1.811026972579045e-05, + "loss": 1.0655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 525 + }, + { + "epoch": 0.6485819975339088, + "grad_norm": 2.0687763759549704, + "learning_rate": 1.810255030407323e-05, + "loss": 0.9515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 526 + }, + { + "epoch": 0.6498150431565968, + "grad_norm": 1.9230312514706345, + "learning_rate": 1.8094816800064376e-05, + "loss": 1.0144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 527 + }, + { + "epoch": 0.6510480887792849, + "grad_norm": 1.9036609363684858, + "learning_rate": 1.8087069227204784e-05, + "loss": 1.1092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 528 + }, + { + "epoch": 0.6522811344019729, + "grad_norm": 2.0395351493662104, + "learning_rate": 1.807930759895979e-05, + "loss": 1.0345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 529 + }, + { + "epoch": 0.6535141800246609, + "grad_norm": 2.011352927400422, + "learning_rate": 1.8071531928819158e-05, + "loss": 0.9913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 530 + }, + { + "epoch": 0.654747225647349, + "grad_norm": 1.9222212916878523, + "learning_rate": 1.8063742230297063e-05, + "loss": 1.0643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 531 + }, + { + "epoch": 0.655980271270037, + "grad_norm": 2.203621374645629, + "learning_rate": 1.8055938516932056e-05, + "loss": 1.0912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 532 + }, + { + "epoch": 0.657213316892725, + "grad_norm": 1.9492742413222397, + "learning_rate": 1.804812080228705e-05, + "loss": 0.9613, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 533 + }, + { + "epoch": 0.6584463625154131, + "grad_norm": 2.1694867453328683, + "learning_rate": 1.8040289099949285e-05, + "loss": 1.1292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 534 + }, + { + "epoch": 0.6596794081381011, + "grad_norm": 2.0272612869084474, + "learning_rate": 1.8032443423530322e-05, + "loss": 1.0473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 535 + }, + { + "epoch": 0.6609124537607891, + "grad_norm": 2.059356328787384, + "learning_rate": 1.8024583786666007e-05, + "loss": 1.1633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 536 + }, + { + "epoch": 0.6621454993834772, + "grad_norm": 1.9976930500073966, + "learning_rate": 1.8016710203016434e-05, + "loss": 1.0907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 537 + }, + { + "epoch": 0.6633785450061652, + "grad_norm": 2.2046172414513747, + "learning_rate": 1.8008822686265958e-05, + "loss": 1.0435, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 538 + }, + { + "epoch": 0.6646115906288532, + "grad_norm": 1.9172442142747625, + "learning_rate": 1.8000921250123144e-05, + "loss": 1.0905, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 539 + }, + { + "epoch": 0.6658446362515413, + "grad_norm": 1.9118211475087676, + "learning_rate": 1.7993005908320737e-05, + "loss": 0.9362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 540 + }, + { + "epoch": 0.6670776818742293, + "grad_norm": 2.2793615054405096, + "learning_rate": 1.7985076674615663e-05, + "loss": 1.1407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 541 + }, + { + "epoch": 0.6683107274969173, + "grad_norm": 2.0968780040667636, + "learning_rate": 1.797713356278899e-05, + "loss": 1.0263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 542 + }, + { + "epoch": 0.6695437731196054, + "grad_norm": 2.130295025454763, + "learning_rate": 1.79691765866459e-05, + "loss": 0.99, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 543 + }, + { + "epoch": 0.6707768187422934, + "grad_norm": 2.050843680123929, + "learning_rate": 1.7961205760015678e-05, + "loss": 0.937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 544 + }, + { + "epoch": 0.6720098643649816, + "grad_norm": 2.00556261298427, + "learning_rate": 1.7953221096751678e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 545 + }, + { + "epoch": 0.6732429099876696, + "grad_norm": 2.104573565750377, + "learning_rate": 1.7945222610731306e-05, + "loss": 0.9851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 546 + }, + { + "epoch": 0.6744759556103576, + "grad_norm": 2.009986716494675, + "learning_rate": 1.793721031585599e-05, + "loss": 0.9842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 547 + }, + { + "epoch": 0.6757090012330457, + "grad_norm": 2.053047844057653, + "learning_rate": 1.792918422605115e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 548 + }, + { + "epoch": 0.6769420468557337, + "grad_norm": 2.0361666532452714, + "learning_rate": 1.7921144355266195e-05, + "loss": 1.0002, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 549 + }, + { + "epoch": 0.6781750924784217, + "grad_norm": 2.02324182186846, + "learning_rate": 1.7913090717474476e-05, + "loss": 1.0915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 550 + }, + { + "epoch": 0.6794081381011098, + "grad_norm": 1.9162969465242339, + "learning_rate": 1.7905023326673277e-05, + "loss": 0.9256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 551 + }, + { + "epoch": 0.6806411837237978, + "grad_norm": 2.055814853732051, + "learning_rate": 1.789694219688378e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 552 + }, + { + "epoch": 0.6818742293464858, + "grad_norm": 2.0596956840908276, + "learning_rate": 1.7888847342151047e-05, + "loss": 1.096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 553 + }, + { + "epoch": 0.6831072749691739, + "grad_norm": 1.9518592781827393, + "learning_rate": 1.7880738776543994e-05, + "loss": 0.9907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 554 + }, + { + "epoch": 0.6843403205918619, + "grad_norm": 2.0542440716859485, + "learning_rate": 1.787261651415537e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 555 + }, + { + "epoch": 0.6855733662145499, + "grad_norm": 1.9676290362610278, + "learning_rate": 1.786448056910173e-05, + "loss": 0.96, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 556 + }, + { + "epoch": 0.686806411837238, + "grad_norm": 2.0236847394224418, + "learning_rate": 1.7856330955523396e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 557 + }, + { + "epoch": 0.688039457459926, + "grad_norm": 2.166651989793048, + "learning_rate": 1.7848167687584466e-05, + "loss": 1.0152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 558 + }, + { + "epoch": 0.689272503082614, + "grad_norm": 2.037401342146827, + "learning_rate": 1.783999077947275e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 559 + }, + { + "epoch": 0.6905055487053021, + "grad_norm": 2.0931620663370465, + "learning_rate": 1.783180024539978e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 560 + }, + { + "epoch": 0.6917385943279901, + "grad_norm": 2.5512794237438032, + "learning_rate": 1.7823596099600767e-05, + "loss": 1.056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 561 + }, + { + "epoch": 0.6929716399506781, + "grad_norm": 2.2145616452367305, + "learning_rate": 1.781537835633457e-05, + "loss": 1.0677, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 562 + }, + { + "epoch": 0.6942046855733662, + "grad_norm": 2.304945412618975, + "learning_rate": 1.7807147029883688e-05, + "loss": 1.0025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 563 + }, + { + "epoch": 0.6954377311960542, + "grad_norm": 2.2714171401510272, + "learning_rate": 1.7798902134554225e-05, + "loss": 0.927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 564 + }, + { + "epoch": 0.6966707768187423, + "grad_norm": 1.9934114654014257, + "learning_rate": 1.7790643684675874e-05, + "loss": 0.9226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 565 + }, + { + "epoch": 0.6979038224414303, + "grad_norm": 2.4001795539004425, + "learning_rate": 1.7782371694601882e-05, + "loss": 0.9898, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 566 + }, + { + "epoch": 0.6991368680641183, + "grad_norm": 2.1543466787664944, + "learning_rate": 1.777408617870902e-05, + "loss": 0.8765, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 567 + }, + { + "epoch": 0.7003699136868065, + "grad_norm": 2.125530896627932, + "learning_rate": 1.7765787151397584e-05, + "loss": 0.9383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 568 + }, + { + "epoch": 0.7016029593094945, + "grad_norm": 2.399579117157628, + "learning_rate": 1.775747462709134e-05, + "loss": 1.116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 569 + }, + { + "epoch": 0.7028360049321825, + "grad_norm": 2.3911388414036843, + "learning_rate": 1.7749148620237515e-05, + "loss": 0.983, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 570 + }, + { + "epoch": 0.7040690505548706, + "grad_norm": 2.037432480432792, + "learning_rate": 1.7740809145306773e-05, + "loss": 1.0094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 571 + }, + { + "epoch": 0.7053020961775586, + "grad_norm": 2.0999588564507685, + "learning_rate": 1.773245621679318e-05, + "loss": 1.046, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 572 + }, + { + "epoch": 0.7065351418002466, + "grad_norm": 1.9822662619027398, + "learning_rate": 1.7724089849214184e-05, + "loss": 1.0177, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 573 + }, + { + "epoch": 0.7077681874229347, + "grad_norm": 2.296463261709746, + "learning_rate": 1.77157100571106e-05, + "loss": 0.9207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 574 + }, + { + "epoch": 0.7090012330456227, + "grad_norm": 2.0434250630443045, + "learning_rate": 1.7707316855046563e-05, + "loss": 1.0088, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 575 + }, + { + "epoch": 0.7102342786683107, + "grad_norm": 2.067794231415559, + "learning_rate": 1.7698910257609524e-05, + "loss": 1.0033, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 576 + }, + { + "epoch": 0.7114673242909988, + "grad_norm": 1.9884025311346816, + "learning_rate": 1.7690490279410205e-05, + "loss": 1.0395, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 577 + }, + { + "epoch": 0.7127003699136868, + "grad_norm": 2.0663488956854676, + "learning_rate": 1.768205693508259e-05, + "loss": 1.0609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 578 + }, + { + "epoch": 0.7139334155363748, + "grad_norm": 1.9661847340747427, + "learning_rate": 1.7673610239283898e-05, + "loss": 0.9852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 579 + }, + { + "epoch": 0.7151664611590629, + "grad_norm": 2.4048347386453135, + "learning_rate": 1.7665150206694543e-05, + "loss": 1.0294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 580 + }, + { + "epoch": 0.7163995067817509, + "grad_norm": 2.049926176641203, + "learning_rate": 1.765667685201813e-05, + "loss": 0.9953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 581 + }, + { + "epoch": 0.717632552404439, + "grad_norm": 2.0425901676211398, + "learning_rate": 1.7648190189981403e-05, + "loss": 1.0155, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 582 + }, + { + "epoch": 0.718865598027127, + "grad_norm": 2.111319944982802, + "learning_rate": 1.7639690235334246e-05, + "loss": 1.012, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 583 + }, + { + "epoch": 0.720098643649815, + "grad_norm": 1.9625598394287143, + "learning_rate": 1.7631177002849648e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 584 + }, + { + "epoch": 0.7213316892725031, + "grad_norm": 2.0737975490297025, + "learning_rate": 1.7622650507323658e-05, + "loss": 0.9954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 585 + }, + { + "epoch": 0.7225647348951911, + "grad_norm": 2.2220755560999175, + "learning_rate": 1.7614110763575393e-05, + "loss": 1.093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 586 + }, + { + "epoch": 0.7237977805178791, + "grad_norm": 2.10369867254611, + "learning_rate": 1.7605557786446997e-05, + "loss": 0.9514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 587 + }, + { + "epoch": 0.7250308261405672, + "grad_norm": 2.023199550080708, + "learning_rate": 1.7596991590803597e-05, + "loss": 1.0909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 588 + }, + { + "epoch": 0.7262638717632552, + "grad_norm": 2.1139350266949197, + "learning_rate": 1.7588412191533305e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 589 + }, + { + "epoch": 0.7274969173859432, + "grad_norm": 2.1032070896543926, + "learning_rate": 1.7579819603547187e-05, + "loss": 0.9305, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 590 + }, + { + "epoch": 0.7287299630086314, + "grad_norm": 2.1111448680444798, + "learning_rate": 1.7571213841779213e-05, + "loss": 1.0821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 591 + }, + { + "epoch": 0.7299630086313194, + "grad_norm": 2.0849758733816692, + "learning_rate": 1.7562594921186268e-05, + "loss": 0.9747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 592 + }, + { + "epoch": 0.7311960542540074, + "grad_norm": 2.7468709410129573, + "learning_rate": 1.7553962856748098e-05, + "loss": 1.0598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 593 + }, + { + "epoch": 0.7324290998766955, + "grad_norm": 2.297654831376881, + "learning_rate": 1.754531766346729e-05, + "loss": 1.0414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 594 + }, + { + "epoch": 0.7336621454993835, + "grad_norm": 2.0094458450808874, + "learning_rate": 1.7536659356369254e-05, + "loss": 0.977, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 595 + }, + { + "epoch": 0.7348951911220715, + "grad_norm": 1.9967204791438438, + "learning_rate": 1.7527987950502196e-05, + "loss": 0.9274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 596 + }, + { + "epoch": 0.7361282367447596, + "grad_norm": 1.9493969416772718, + "learning_rate": 1.7519303460937073e-05, + "loss": 1.0292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 597 + }, + { + "epoch": 0.7373612823674476, + "grad_norm": 2.0023257821799247, + "learning_rate": 1.7510605902767602e-05, + "loss": 0.9574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 598 + }, + { + "epoch": 0.7385943279901356, + "grad_norm": 2.076672200643684, + "learning_rate": 1.7501895291110197e-05, + "loss": 0.9971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 599 + }, + { + "epoch": 0.7398273736128237, + "grad_norm": 2.0612367813173584, + "learning_rate": 1.7493171641103965e-05, + "loss": 0.9958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 600 + }, + { + "epoch": 0.7410604192355117, + "grad_norm": 2.16071395185368, + "learning_rate": 1.7484434967910676e-05, + "loss": 1.0001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 601 + }, + { + "epoch": 0.7422934648581998, + "grad_norm": 2.0810754791890664, + "learning_rate": 1.747568528671473e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 602 + }, + { + "epoch": 0.7435265104808878, + "grad_norm": 2.0015256143479525, + "learning_rate": 1.7466922612723133e-05, + "loss": 0.9288, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 603 + }, + { + "epoch": 0.7447595561035758, + "grad_norm": 1.9574522371127243, + "learning_rate": 1.7458146961165484e-05, + "loss": 1.0365, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 604 + }, + { + "epoch": 0.7459926017262639, + "grad_norm": 1.999399085309589, + "learning_rate": 1.7449358347293925e-05, + "loss": 1.0457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 605 + }, + { + "epoch": 0.7472256473489519, + "grad_norm": 2.016778757176555, + "learning_rate": 1.7440556786383132e-05, + "loss": 1.0172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 606 + }, + { + "epoch": 0.7484586929716399, + "grad_norm": 2.011184042360808, + "learning_rate": 1.743174229373028e-05, + "loss": 0.9568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 607 + }, + { + "epoch": 0.749691738594328, + "grad_norm": 1.9994468999326396, + "learning_rate": 1.7422914884655027e-05, + "loss": 0.9822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 608 + }, + { + "epoch": 0.750924784217016, + "grad_norm": 2.1373417665891163, + "learning_rate": 1.7414074574499465e-05, + "loss": 0.9128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 609 + }, + { + "epoch": 0.752157829839704, + "grad_norm": 2.414511732797334, + "learning_rate": 1.7405221378628133e-05, + "loss": 1.0866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 610 + }, + { + "epoch": 0.7533908754623921, + "grad_norm": 2.1820009401893605, + "learning_rate": 1.7396355312427935e-05, + "loss": 1.0538, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 611 + }, + { + "epoch": 0.7546239210850801, + "grad_norm": 2.0281507644839163, + "learning_rate": 1.7387476391308168e-05, + "loss": 0.9224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 612 + }, + { + "epoch": 0.7558569667077681, + "grad_norm": 2.387292229834452, + "learning_rate": 1.7378584630700454e-05, + "loss": 0.7561, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 613 + }, + { + "epoch": 0.7570900123304563, + "grad_norm": 2.0036252294718975, + "learning_rate": 1.7369680046058744e-05, + "loss": 0.9816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 614 + }, + { + "epoch": 0.7583230579531443, + "grad_norm": 2.1474265419679566, + "learning_rate": 1.7360762652859275e-05, + "loss": 0.9312, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 615 + }, + { + "epoch": 0.7595561035758323, + "grad_norm": 2.3913311043981387, + "learning_rate": 1.735183246660053e-05, + "loss": 1.1671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 616 + }, + { + "epoch": 0.7607891491985204, + "grad_norm": 2.070836177310932, + "learning_rate": 1.7342889502803246e-05, + "loss": 0.8846, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 617 + }, + { + "epoch": 0.7620221948212084, + "grad_norm": 1.9909512311630846, + "learning_rate": 1.7333933777010357e-05, + "loss": 0.9683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 618 + }, + { + "epoch": 0.7632552404438965, + "grad_norm": 2.048373917294825, + "learning_rate": 1.732496530478698e-05, + "loss": 1.021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 619 + }, + { + "epoch": 0.7644882860665845, + "grad_norm": 2.046848859346405, + "learning_rate": 1.7315984101720388e-05, + "loss": 0.9949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 620 + }, + { + "epoch": 0.7657213316892725, + "grad_norm": 2.199270192761742, + "learning_rate": 1.7306990183419972e-05, + "loss": 0.9319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 621 + }, + { + "epoch": 0.7669543773119606, + "grad_norm": 2.2362149652218877, + "learning_rate": 1.729798356551723e-05, + "loss": 1.0579, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 622 + }, + { + "epoch": 0.7681874229346486, + "grad_norm": 2.072113656676678, + "learning_rate": 1.7288964263665735e-05, + "loss": 0.984, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 623 + }, + { + "epoch": 0.7694204685573366, + "grad_norm": 2.2058183913371647, + "learning_rate": 1.7279932293541092e-05, + "loss": 1.0727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 624 + }, + { + "epoch": 0.7706535141800247, + "grad_norm": 2.1099223160270104, + "learning_rate": 1.7270887670840935e-05, + "loss": 0.852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 625 + }, + { + "epoch": 0.7718865598027127, + "grad_norm": 2.014275874425781, + "learning_rate": 1.7261830411284884e-05, + "loss": 1.0219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 626 + }, + { + "epoch": 0.7731196054254007, + "grad_norm": 2.1002718881715605, + "learning_rate": 1.7252760530614522e-05, + "loss": 0.9339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 627 + }, + { + "epoch": 0.7743526510480888, + "grad_norm": 2.530650182566485, + "learning_rate": 1.724367804459337e-05, + "loss": 1.1248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 628 + }, + { + "epoch": 0.7755856966707768, + "grad_norm": 2.072394234232429, + "learning_rate": 1.7234582969006847e-05, + "loss": 0.963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 629 + }, + { + "epoch": 0.7768187422934648, + "grad_norm": 2.077401892279152, + "learning_rate": 1.722547531966227e-05, + "loss": 0.9589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 630 + }, + { + "epoch": 0.7780517879161529, + "grad_norm": 2.0461500607118612, + "learning_rate": 1.7216355112388797e-05, + "loss": 0.9892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 631 + }, + { + "epoch": 0.7792848335388409, + "grad_norm": 1.9481531627295827, + "learning_rate": 1.7207222363037414e-05, + "loss": 0.9963, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 632 + }, + { + "epoch": 0.7805178791615289, + "grad_norm": 2.0363659895001067, + "learning_rate": 1.7198077087480903e-05, + "loss": 0.9206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 633 + }, + { + "epoch": 0.781750924784217, + "grad_norm": 2.367528975566535, + "learning_rate": 1.718891930161383e-05, + "loss": 1.1255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 634 + }, + { + "epoch": 0.782983970406905, + "grad_norm": 2.007050047111477, + "learning_rate": 1.7179749021352482e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 635 + }, + { + "epoch": 0.7842170160295932, + "grad_norm": 1.9196848202462296, + "learning_rate": 1.7170566262634883e-05, + "loss": 0.9448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 636 + }, + { + "epoch": 0.7854500616522812, + "grad_norm": 2.145941740629715, + "learning_rate": 1.7161371041420723e-05, + "loss": 0.9428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 637 + }, + { + "epoch": 0.7866831072749692, + "grad_norm": 2.170240354212872, + "learning_rate": 1.7152163373691377e-05, + "loss": 1.0323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 638 + }, + { + "epoch": 0.7879161528976573, + "grad_norm": 2.0553179179576015, + "learning_rate": 1.714294327544983e-05, + "loss": 0.8661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 639 + }, + { + "epoch": 0.7891491985203453, + "grad_norm": 2.1113466348102397, + "learning_rate": 1.7133710762720683e-05, + "loss": 1.034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 640 + }, + { + "epoch": 0.7903822441430333, + "grad_norm": 2.146414650333374, + "learning_rate": 1.712446585155011e-05, + "loss": 1.0681, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 641 + }, + { + "epoch": 0.7916152897657214, + "grad_norm": 2.0440039273987227, + "learning_rate": 1.7115208558005834e-05, + "loss": 0.9085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 642 + }, + { + "epoch": 0.7928483353884094, + "grad_norm": 2.3653004518970535, + "learning_rate": 1.7105938898177097e-05, + "loss": 0.8514, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 643 + }, + { + "epoch": 0.7940813810110974, + "grad_norm": 2.058806308472887, + "learning_rate": 1.709665688817464e-05, + "loss": 0.9258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 644 + }, + { + "epoch": 0.7953144266337855, + "grad_norm": 2.3406153888995798, + "learning_rate": 1.708736254413066e-05, + "loss": 0.9866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 645 + }, + { + "epoch": 0.7965474722564735, + "grad_norm": 3.2012831686915293, + "learning_rate": 1.70780558821988e-05, + "loss": 1.0916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 646 + }, + { + "epoch": 0.7977805178791615, + "grad_norm": 2.23457477201596, + "learning_rate": 1.7068736918554102e-05, + "loss": 0.9076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 647 + }, + { + "epoch": 0.7990135635018496, + "grad_norm": 2.255512398782474, + "learning_rate": 1.7059405669392993e-05, + "loss": 0.9113, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 648 + }, + { + "epoch": 0.8002466091245376, + "grad_norm": 2.114904513093256, + "learning_rate": 1.7050062150933252e-05, + "loss": 0.9587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 649 + }, + { + "epoch": 0.8014796547472256, + "grad_norm": 1.931643331052564, + "learning_rate": 1.7040706379413983e-05, + "loss": 0.9888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 650 + }, + { + "epoch": 0.8027127003699137, + "grad_norm": 2.0094212900142816, + "learning_rate": 1.7031338371095588e-05, + "loss": 0.9961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 651 + }, + { + "epoch": 0.8039457459926017, + "grad_norm": 2.1928733306625188, + "learning_rate": 1.702195814225973e-05, + "loss": 1.0158, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 652 + }, + { + "epoch": 0.8051787916152897, + "grad_norm": 2.2216734919656544, + "learning_rate": 1.7012565709209317e-05, + "loss": 1.0201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 653 + }, + { + "epoch": 0.8064118372379778, + "grad_norm": 2.0613969675724686, + "learning_rate": 1.700316108826847e-05, + "loss": 0.9941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 654 + }, + { + "epoch": 0.8076448828606658, + "grad_norm": 1.9372031821836655, + "learning_rate": 1.6993744295782482e-05, + "loss": 1.0026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 655 + }, + { + "epoch": 0.8088779284833539, + "grad_norm": 2.077194491232311, + "learning_rate": 1.698431534811781e-05, + "loss": 0.8491, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 656 + }, + { + "epoch": 0.8101109741060419, + "grad_norm": 2.1425857340105754, + "learning_rate": 1.6974874261662035e-05, + "loss": 0.9719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 657 + }, + { + "epoch": 0.8113440197287299, + "grad_norm": 1.8547512115782296, + "learning_rate": 1.6965421052823834e-05, + "loss": 1.0107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 658 + }, + { + "epoch": 0.812577065351418, + "grad_norm": 1.9098479812610942, + "learning_rate": 1.695595573803295e-05, + "loss": 0.95, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 659 + }, + { + "epoch": 0.813810110974106, + "grad_norm": 1.892767796833539, + "learning_rate": 1.6946478333740174e-05, + "loss": 0.86, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 660 + }, + { + "epoch": 0.815043156596794, + "grad_norm": 2.0228841723951043, + "learning_rate": 1.69369888564173e-05, + "loss": 0.9881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 661 + }, + { + "epoch": 0.8162762022194822, + "grad_norm": 2.144845399605076, + "learning_rate": 1.692748732255711e-05, + "loss": 0.8493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 662 + }, + { + "epoch": 0.8175092478421702, + "grad_norm": 2.1710135611234027, + "learning_rate": 1.6917973748673343e-05, + "loss": 0.9356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 663 + }, + { + "epoch": 0.8187422934648582, + "grad_norm": 2.200157070237925, + "learning_rate": 1.6908448151300654e-05, + "loss": 0.9922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 664 + }, + { + "epoch": 0.8199753390875463, + "grad_norm": 2.1046294162556385, + "learning_rate": 1.6898910546994605e-05, + "loss": 0.9714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 665 + }, + { + "epoch": 0.8212083847102343, + "grad_norm": 2.1921813927002622, + "learning_rate": 1.688936095233162e-05, + "loss": 1.0186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 666 + }, + { + "epoch": 0.8224414303329223, + "grad_norm": 2.5290727417605767, + "learning_rate": 1.6879799383908967e-05, + "loss": 0.9147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 667 + }, + { + "epoch": 0.8236744759556104, + "grad_norm": 2.0992091012861605, + "learning_rate": 1.6870225858344715e-05, + "loss": 1.0331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 668 + }, + { + "epoch": 0.8249075215782984, + "grad_norm": 2.228115711337027, + "learning_rate": 1.6860640392277728e-05, + "loss": 1.0372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 669 + }, + { + "epoch": 0.8261405672009864, + "grad_norm": 2.104862934581671, + "learning_rate": 1.6851043002367612e-05, + "loss": 0.9556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 670 + }, + { + "epoch": 0.8273736128236745, + "grad_norm": 2.1153449603435575, + "learning_rate": 1.6841433705294704e-05, + "loss": 0.9848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 671 + }, + { + "epoch": 0.8286066584463625, + "grad_norm": 1.9113210816564241, + "learning_rate": 1.6831812517760036e-05, + "loss": 0.9553, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 672 + }, + { + "epoch": 0.8298397040690506, + "grad_norm": 2.146278639653085, + "learning_rate": 1.6822179456485292e-05, + "loss": 0.8621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 673 + }, + { + "epoch": 0.8310727496917386, + "grad_norm": 2.092521119953653, + "learning_rate": 1.681253453821281e-05, + "loss": 0.9576, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 674 + }, + { + "epoch": 0.8323057953144266, + "grad_norm": 2.108757495841555, + "learning_rate": 1.6802877779705526e-05, + "loss": 0.9349, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 675 + }, + { + "epoch": 0.8335388409371147, + "grad_norm": 2.165541838354097, + "learning_rate": 1.679320919774696e-05, + "loss": 0.9607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 676 + }, + { + "epoch": 0.8347718865598027, + "grad_norm": 2.0901280484241322, + "learning_rate": 1.6783528809141175e-05, + "loss": 0.9577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 677 + }, + { + "epoch": 0.8360049321824907, + "grad_norm": 2.049770632568499, + "learning_rate": 1.6773836630712757e-05, + "loss": 0.8912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 678 + }, + { + "epoch": 0.8372379778051788, + "grad_norm": 1.9415439366071066, + "learning_rate": 1.6764132679306785e-05, + "loss": 0.9678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 679 + }, + { + "epoch": 0.8384710234278668, + "grad_norm": 2.159517279560615, + "learning_rate": 1.6754416971788795e-05, + "loss": 0.9739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 680 + }, + { + "epoch": 0.8397040690505548, + "grad_norm": 2.0578899051147745, + "learning_rate": 1.674468952504476e-05, + "loss": 0.9187, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 681 + }, + { + "epoch": 0.840937114673243, + "grad_norm": 2.1905094099573135, + "learning_rate": 1.673495035598105e-05, + "loss": 0.9753, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 682 + }, + { + "epoch": 0.842170160295931, + "grad_norm": 2.194631531971183, + "learning_rate": 1.6725199481524418e-05, + "loss": 1.0267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 683 + }, + { + "epoch": 0.843403205918619, + "grad_norm": 2.040276435918069, + "learning_rate": 1.671543691862195e-05, + "loss": 0.9703, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 684 + }, + { + "epoch": 0.8446362515413071, + "grad_norm": 1.9627382699239366, + "learning_rate": 1.6705662684241053e-05, + "loss": 0.8955, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 685 + }, + { + "epoch": 0.8458692971639951, + "grad_norm": 2.288828851032713, + "learning_rate": 1.6695876795369418e-05, + "loss": 0.946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 686 + }, + { + "epoch": 0.8471023427866831, + "grad_norm": 2.1152782339808884, + "learning_rate": 1.668607926901499e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 687 + }, + { + "epoch": 0.8483353884093712, + "grad_norm": 2.798938790414006, + "learning_rate": 1.6676270122205946e-05, + "loss": 0.8988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 688 + }, + { + "epoch": 0.8495684340320592, + "grad_norm": 2.5307103300223766, + "learning_rate": 1.6666449371990644e-05, + "loss": 0.8814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 689 + }, + { + "epoch": 0.8508014796547472, + "grad_norm": 2.2986978033861294, + "learning_rate": 1.6656617035437635e-05, + "loss": 0.9402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 690 + }, + { + "epoch": 0.8520345252774353, + "grad_norm": 2.1645514508732546, + "learning_rate": 1.664677312963558e-05, + "loss": 0.9551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 691 + }, + { + "epoch": 0.8532675709001233, + "grad_norm": 2.144781636504613, + "learning_rate": 1.6636917671693264e-05, + "loss": 1.0505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 692 + }, + { + "epoch": 0.8545006165228114, + "grad_norm": 2.068308383986026, + "learning_rate": 1.6627050678739544e-05, + "loss": 0.9237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 693 + }, + { + "epoch": 0.8557336621454994, + "grad_norm": 2.4201265499871325, + "learning_rate": 1.6617172167923327e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 694 + }, + { + "epoch": 0.8569667077681874, + "grad_norm": 1.9743124171175193, + "learning_rate": 1.660728215641353e-05, + "loss": 0.9649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 695 + }, + { + "epoch": 0.8581997533908755, + "grad_norm": 2.4976283410874847, + "learning_rate": 1.6597380661399077e-05, + "loss": 0.9445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 696 + }, + { + "epoch": 0.8594327990135635, + "grad_norm": 2.2344847337230855, + "learning_rate": 1.6587467700088834e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 697 + }, + { + "epoch": 0.8606658446362515, + "grad_norm": 2.1366592393550343, + "learning_rate": 1.6577543289711597e-05, + "loss": 0.8845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 698 + }, + { + "epoch": 0.8618988902589396, + "grad_norm": 2.0879117181580753, + "learning_rate": 1.6567607447516064e-05, + "loss": 0.7972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 699 + }, + { + "epoch": 0.8631319358816276, + "grad_norm": 2.117890235500837, + "learning_rate": 1.6557660190770803e-05, + "loss": 0.9598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 700 + }, + { + "epoch": 0.8643649815043156, + "grad_norm": 2.227804015025162, + "learning_rate": 1.654770153676422e-05, + "loss": 0.9105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 701 + }, + { + "epoch": 0.8655980271270037, + "grad_norm": 2.262676552209609, + "learning_rate": 1.6537731502804525e-05, + "loss": 0.9706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 702 + }, + { + "epoch": 0.8668310727496917, + "grad_norm": 2.146446661141531, + "learning_rate": 1.6527750106219715e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 703 + }, + { + "epoch": 0.8680641183723797, + "grad_norm": 2.08167189099969, + "learning_rate": 1.6517757364357526e-05, + "loss": 0.9381, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 704 + }, + { + "epoch": 0.8692971639950678, + "grad_norm": 1.9310579355796305, + "learning_rate": 1.6507753294585417e-05, + "loss": 0.9047, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 705 + }, + { + "epoch": 0.8705302096177558, + "grad_norm": 2.0325358606748978, + "learning_rate": 1.6497737914290535e-05, + "loss": 0.9094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 706 + }, + { + "epoch": 0.8717632552404438, + "grad_norm": 2.0543982954124966, + "learning_rate": 1.648771124087969e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 707 + }, + { + "epoch": 0.872996300863132, + "grad_norm": 2.295806181212224, + "learning_rate": 1.64776732917793e-05, + "loss": 0.9444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 708 + }, + { + "epoch": 0.87422934648582, + "grad_norm": 2.1899960866925303, + "learning_rate": 1.646762408443541e-05, + "loss": 1.0083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 709 + }, + { + "epoch": 0.8754623921085081, + "grad_norm": 2.040526145894979, + "learning_rate": 1.6457563636313598e-05, + "loss": 0.8215, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 710 + }, + { + "epoch": 0.8766954377311961, + "grad_norm": 1.9967218569140772, + "learning_rate": 1.6447491964899008e-05, + "loss": 0.9227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 711 + }, + { + "epoch": 0.8779284833538841, + "grad_norm": 2.131774886953309, + "learning_rate": 1.6437409087696283e-05, + "loss": 0.8786, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 712 + }, + { + "epoch": 0.8791615289765722, + "grad_norm": 2.0642410512824414, + "learning_rate": 1.6427315022229525e-05, + "loss": 0.8831, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 713 + }, + { + "epoch": 0.8803945745992602, + "grad_norm": 2.2793436000912255, + "learning_rate": 1.64172097860423e-05, + "loss": 0.8413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 714 + }, + { + "epoch": 0.8816276202219482, + "grad_norm": 1.9501864497105927, + "learning_rate": 1.640709339669758e-05, + "loss": 0.8856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 715 + }, + { + "epoch": 0.8828606658446363, + "grad_norm": 2.0279122102292977, + "learning_rate": 1.6396965871777727e-05, + "loss": 0.8835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 716 + }, + { + "epoch": 0.8840937114673243, + "grad_norm": 2.181455569803073, + "learning_rate": 1.638682722888445e-05, + "loss": 0.9995, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 717 + }, + { + "epoch": 0.8853267570900123, + "grad_norm": 2.1877848027614313, + "learning_rate": 1.637667748563879e-05, + "loss": 0.8314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 718 + }, + { + "epoch": 0.8865598027127004, + "grad_norm": 2.0901252578597713, + "learning_rate": 1.6366516659681067e-05, + "loss": 0.9333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 719 + }, + { + "epoch": 0.8877928483353884, + "grad_norm": 2.1846108211938673, + "learning_rate": 1.6356344768670875e-05, + "loss": 0.9309, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 720 + }, + { + "epoch": 0.8890258939580764, + "grad_norm": 2.600109718803851, + "learning_rate": 1.6346161830287034e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 721 + }, + { + "epoch": 0.8902589395807645, + "grad_norm": 2.2057915960500276, + "learning_rate": 1.6335967862227567e-05, + "loss": 0.9764, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 722 + }, + { + "epoch": 0.8914919852034525, + "grad_norm": 2.1009909758008267, + "learning_rate": 1.632576288220966e-05, + "loss": 0.8741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 723 + }, + { + "epoch": 0.8927250308261405, + "grad_norm": 2.1617838023881, + "learning_rate": 1.6315546907969652e-05, + "loss": 0.9573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 724 + }, + { + "epoch": 0.8939580764488286, + "grad_norm": 2.091522410165622, + "learning_rate": 1.6305319957262972e-05, + "loss": 0.9534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 725 + }, + { + "epoch": 0.8951911220715166, + "grad_norm": 2.2845231322682418, + "learning_rate": 1.6295082047864138e-05, + "loss": 0.9156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 726 + }, + { + "epoch": 0.8964241676942046, + "grad_norm": 2.062614641816067, + "learning_rate": 1.6284833197566712e-05, + "loss": 0.9539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 727 + }, + { + "epoch": 0.8976572133168927, + "grad_norm": 2.157165674006361, + "learning_rate": 1.627457342418327e-05, + "loss": 0.9926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 728 + }, + { + "epoch": 0.8988902589395807, + "grad_norm": 2.039737473184125, + "learning_rate": 1.6264302745545374e-05, + "loss": 0.9406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 729 + }, + { + "epoch": 0.9001233045622689, + "grad_norm": 2.057800120641984, + "learning_rate": 1.625402117950354e-05, + "loss": 0.9139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 730 + }, + { + "epoch": 0.9013563501849569, + "grad_norm": 2.240321306869684, + "learning_rate": 1.62437287439272e-05, + "loss": 1.0758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 731 + }, + { + "epoch": 0.9025893958076449, + "grad_norm": 2.0947466285552436, + "learning_rate": 1.6233425456704686e-05, + "loss": 0.8467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 732 + }, + { + "epoch": 0.903822441430333, + "grad_norm": 2.061121182714553, + "learning_rate": 1.6223111335743187e-05, + "loss": 0.9093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 733 + }, + { + "epoch": 0.905055487053021, + "grad_norm": 2.209914581799182, + "learning_rate": 1.6212786398968717e-05, + "loss": 0.8805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 734 + }, + { + "epoch": 0.906288532675709, + "grad_norm": 2.468473842682747, + "learning_rate": 1.6202450664326093e-05, + "loss": 0.9013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 735 + }, + { + "epoch": 0.9075215782983971, + "grad_norm": 2.3365951628381056, + "learning_rate": 1.61921041497789e-05, + "loss": 0.9112, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 736 + }, + { + "epoch": 0.9087546239210851, + "grad_norm": 2.474277588242769, + "learning_rate": 1.6181746873309455e-05, + "loss": 0.9186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 737 + }, + { + "epoch": 0.9099876695437731, + "grad_norm": 2.0716027082353228, + "learning_rate": 1.6171378852918778e-05, + "loss": 0.9486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 738 + }, + { + "epoch": 0.9112207151664612, + "grad_norm": 2.165213004434134, + "learning_rate": 1.6161000106626564e-05, + "loss": 0.9726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 739 + }, + { + "epoch": 0.9124537607891492, + "grad_norm": 2.4218033902151475, + "learning_rate": 1.615061065247115e-05, + "loss": 0.8259, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 740 + }, + { + "epoch": 0.9136868064118372, + "grad_norm": 1.970928985580259, + "learning_rate": 1.614021050850949e-05, + "loss": 0.9788, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 741 + }, + { + "epoch": 0.9149198520345253, + "grad_norm": 2.2639971055328934, + "learning_rate": 1.6129799692817097e-05, + "loss": 0.8745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 742 + }, + { + "epoch": 0.9161528976572133, + "grad_norm": 2.5602631751385028, + "learning_rate": 1.611937822348806e-05, + "loss": 0.9512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 743 + }, + { + "epoch": 0.9173859432799013, + "grad_norm": 2.22841865866248, + "learning_rate": 1.6108946118634957e-05, + "loss": 0.805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 744 + }, + { + "epoch": 0.9186189889025894, + "grad_norm": 2.4182686550221626, + "learning_rate": 1.6098503396388868e-05, + "loss": 1.0202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 745 + }, + { + "epoch": 0.9198520345252774, + "grad_norm": 2.2298122692949645, + "learning_rate": 1.608805007489932e-05, + "loss": 0.8972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 746 + }, + { + "epoch": 0.9210850801479655, + "grad_norm": 2.362160156268979, + "learning_rate": 1.607758617233426e-05, + "loss": 0.9039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 747 + }, + { + "epoch": 0.9223181257706535, + "grad_norm": 2.1106519602619533, + "learning_rate": 1.6067111706880027e-05, + "loss": 0.8318, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 748 + }, + { + "epoch": 0.9235511713933415, + "grad_norm": 2.0825477479243424, + "learning_rate": 1.605662669674132e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 749 + }, + { + "epoch": 0.9247842170160296, + "grad_norm": 2.036239449814601, + "learning_rate": 1.6046131160141167e-05, + "loss": 0.8727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 750 + }, + { + "epoch": 0.9260172626387176, + "grad_norm": 2.1703406452205787, + "learning_rate": 1.603562511532088e-05, + "loss": 0.9373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 751 + }, + { + "epoch": 0.9272503082614056, + "grad_norm": 2.1398345899160143, + "learning_rate": 1.6025108580540046e-05, + "loss": 0.8691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 752 + }, + { + "epoch": 0.9284833538840938, + "grad_norm": 2.0888823283975375, + "learning_rate": 1.601458157407648e-05, + "loss": 0.8547, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 753 + }, + { + "epoch": 0.9297163995067818, + "grad_norm": 2.269523743217552, + "learning_rate": 1.6004044114226193e-05, + "loss": 1.0141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 754 + }, + { + "epoch": 0.9309494451294698, + "grad_norm": 2.0091635625820063, + "learning_rate": 1.5993496219303367e-05, + "loss": 0.9972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 755 + }, + { + "epoch": 0.9321824907521579, + "grad_norm": 2.1386095902203612, + "learning_rate": 1.5982937907640327e-05, + "loss": 0.9388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 756 + }, + { + "epoch": 0.9334155363748459, + "grad_norm": 2.1260034823775467, + "learning_rate": 1.5972369197587486e-05, + "loss": 0.9153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 757 + }, + { + "epoch": 0.9346485819975339, + "grad_norm": 2.1225283234506156, + "learning_rate": 1.5961790107513343e-05, + "loss": 1.0178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 758 + }, + { + "epoch": 0.935881627620222, + "grad_norm": 2.0422192666840795, + "learning_rate": 1.595120065580444e-05, + "loss": 0.9856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 759 + }, + { + "epoch": 0.93711467324291, + "grad_norm": 2.051963293225036, + "learning_rate": 1.5940600860865315e-05, + "loss": 0.8559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 760 + }, + { + "epoch": 0.938347718865598, + "grad_norm": 2.127612803214104, + "learning_rate": 1.592999074111849e-05, + "loss": 0.8298, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 761 + }, + { + "epoch": 0.9395807644882861, + "grad_norm": 2.286554941831041, + "learning_rate": 1.591937031500443e-05, + "loss": 0.8796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 762 + }, + { + "epoch": 0.9408138101109741, + "grad_norm": 2.282412038316439, + "learning_rate": 1.5908739600981517e-05, + "loss": 1.022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 763 + }, + { + "epoch": 0.9420468557336621, + "grad_norm": 2.197857160398704, + "learning_rate": 1.5898098617526006e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 764 + }, + { + "epoch": 0.9432799013563502, + "grad_norm": 2.1702846778760607, + "learning_rate": 1.5887447383132008e-05, + "loss": 0.9507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 765 + }, + { + "epoch": 0.9445129469790382, + "grad_norm": 2.045700294662405, + "learning_rate": 1.587678591631144e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 766 + }, + { + "epoch": 0.9457459926017263, + "grad_norm": 2.426595174168101, + "learning_rate": 1.5866114235594014e-05, + "loss": 0.9599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 767 + }, + { + "epoch": 0.9469790382244143, + "grad_norm": 2.085172001225346, + "learning_rate": 1.585543235952718e-05, + "loss": 0.9166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 768 + }, + { + "epoch": 0.9482120838471023, + "grad_norm": 2.1181076484114665, + "learning_rate": 1.5844740306676133e-05, + "loss": 0.8829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 769 + }, + { + "epoch": 0.9494451294697904, + "grad_norm": 2.64845194344853, + "learning_rate": 1.583403809562373e-05, + "loss": 0.8927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 770 + }, + { + "epoch": 0.9506781750924784, + "grad_norm": 2.4929599774250186, + "learning_rate": 1.5823325744970488e-05, + "loss": 0.9407, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 771 + }, + { + "epoch": 0.9519112207151664, + "grad_norm": 2.368031870072283, + "learning_rate": 1.581260327333456e-05, + "loss": 0.9463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 772 + }, + { + "epoch": 0.9531442663378545, + "grad_norm": 2.136177985699207, + "learning_rate": 1.580187069935167e-05, + "loss": 0.8571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 773 + }, + { + "epoch": 0.9543773119605425, + "grad_norm": 2.128181510143405, + "learning_rate": 1.5791128041675114e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 774 + }, + { + "epoch": 0.9556103575832305, + "grad_norm": 2.220458022640665, + "learning_rate": 1.5780375318975715e-05, + "loss": 0.8445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 775 + }, + { + "epoch": 0.9568434032059187, + "grad_norm": 2.098330673951412, + "learning_rate": 1.5769612549941777e-05, + "loss": 0.964, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 776 + }, + { + "epoch": 0.9580764488286067, + "grad_norm": 2.094864541454194, + "learning_rate": 1.575883975327907e-05, + "loss": 0.8254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 777 + }, + { + "epoch": 0.9593094944512947, + "grad_norm": 2.3159416415455047, + "learning_rate": 1.5748056947710804e-05, + "loss": 0.9472, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.28, + "step": 778 + }, + { + "epoch": 0.9605425400739828, + "grad_norm": 1.9950192842009455, + "learning_rate": 1.573726415197756e-05, + "loss": 0.7965, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 779 + }, + { + "epoch": 0.9617755856966708, + "grad_norm": 2.006344851533844, + "learning_rate": 1.5726461384837306e-05, + "loss": 0.8203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 780 + }, + { + "epoch": 0.9630086313193588, + "grad_norm": 2.2545503988149007, + "learning_rate": 1.5715648665065324e-05, + "loss": 0.9282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 781 + }, + { + "epoch": 0.9642416769420469, + "grad_norm": 2.2372695819528294, + "learning_rate": 1.5704826011454203e-05, + "loss": 0.9608, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 782 + }, + { + "epoch": 0.9654747225647349, + "grad_norm": 2.1710522433750357, + "learning_rate": 1.5693993442813792e-05, + "loss": 0.7704, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 783 + }, + { + "epoch": 0.966707768187423, + "grad_norm": 2.400731212803864, + "learning_rate": 1.568315097797118e-05, + "loss": 0.9696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 784 + }, + { + "epoch": 0.967940813810111, + "grad_norm": 2.096674414566688, + "learning_rate": 1.5672298635770645e-05, + "loss": 0.6728, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 785 + }, + { + "epoch": 0.969173859432799, + "grad_norm": 2.2214704493931205, + "learning_rate": 1.566143643507364e-05, + "loss": 1.0617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 786 + }, + { + "epoch": 0.9704069050554871, + "grad_norm": 2.21278029237075, + "learning_rate": 1.5650564394758743e-05, + "loss": 0.8487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 787 + }, + { + "epoch": 0.9716399506781751, + "grad_norm": 2.3400569095138763, + "learning_rate": 1.5639682533721643e-05, + "loss": 0.9532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 788 + }, + { + "epoch": 0.9728729963008631, + "grad_norm": 2.1937723298850784, + "learning_rate": 1.5628790870875093e-05, + "loss": 0.9272, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 789 + }, + { + "epoch": 0.9741060419235512, + "grad_norm": 2.117801922125244, + "learning_rate": 1.561788942514889e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 790 + }, + { + "epoch": 0.9753390875462392, + "grad_norm": 2.436307402269952, + "learning_rate": 1.560697821548982e-05, + "loss": 0.8599, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 791 + }, + { + "epoch": 0.9765721331689272, + "grad_norm": 2.2722775653462848, + "learning_rate": 1.5596057260861645e-05, + "loss": 0.8503, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 792 + }, + { + "epoch": 0.9778051787916153, + "grad_norm": 2.22960384352197, + "learning_rate": 1.5585126580245066e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 793 + }, + { + "epoch": 0.9790382244143033, + "grad_norm": 2.1141630678799275, + "learning_rate": 1.557418619263768e-05, + "loss": 0.8418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 794 + }, + { + "epoch": 0.9802712700369913, + "grad_norm": 2.4590856272408224, + "learning_rate": 1.5563236117053965e-05, + "loss": 0.9197, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 795 + }, + { + "epoch": 0.9815043156596794, + "grad_norm": 2.678264319346431, + "learning_rate": 1.5552276372525237e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 796 + }, + { + "epoch": 0.9827373612823674, + "grad_norm": 2.262445020658096, + "learning_rate": 1.554130697809961e-05, + "loss": 0.9101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 797 + }, + { + "epoch": 0.9839704069050554, + "grad_norm": 2.327233920544754, + "learning_rate": 1.5530327952841965e-05, + "loss": 0.9812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 798 + }, + { + "epoch": 0.9852034525277436, + "grad_norm": 2.1249691788358334, + "learning_rate": 1.551933931583393e-05, + "loss": 0.8824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 799 + }, + { + "epoch": 0.9864364981504316, + "grad_norm": 2.0648396343207325, + "learning_rate": 1.5508341086173843e-05, + "loss": 0.8895, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 800 + }, + { + "epoch": 0.9876695437731196, + "grad_norm": 2.109919443242373, + "learning_rate": 1.5497333282976698e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 801 + }, + { + "epoch": 0.9889025893958077, + "grad_norm": 2.3306692057820797, + "learning_rate": 1.5486315925374146e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 802 + }, + { + "epoch": 0.9901356350184957, + "grad_norm": 2.365009122779881, + "learning_rate": 1.547528903251443e-05, + "loss": 1.0604, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 803 + }, + { + "epoch": 0.9913686806411838, + "grad_norm": 2.2289048394713262, + "learning_rate": 1.5464252623562375e-05, + "loss": 0.8141, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 804 + }, + { + "epoch": 0.9926017262638718, + "grad_norm": 2.457422539818384, + "learning_rate": 1.5453206717699334e-05, + "loss": 1.0228, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 805 + }, + { + "epoch": 0.9938347718865598, + "grad_norm": 2.0659430518171558, + "learning_rate": 1.544215133412318e-05, + "loss": 0.8641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 806 + }, + { + "epoch": 0.9950678175092479, + "grad_norm": 2.154728199437817, + "learning_rate": 1.543108649204824e-05, + "loss": 0.8029, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 807 + }, + { + "epoch": 0.9963008631319359, + "grad_norm": 2.3679631548982196, + "learning_rate": 1.54200122107053e-05, + "loss": 0.866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 808 + }, + { + "epoch": 0.9975339087546239, + "grad_norm": 2.218685707693942, + "learning_rate": 1.540892850934154e-05, + "loss": 0.8279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 809 + }, + { + "epoch": 0.998766954377312, + "grad_norm": 2.3274870898977187, + "learning_rate": 1.5397835407220508e-05, + "loss": 0.9724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 810 + }, + { + "epoch": 1.0, + "grad_norm": 2.159965188324255, + "learning_rate": 1.53867329236221e-05, + "loss": 0.762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 811 + }, + { + "epoch": 1.0012330456226881, + "grad_norm": 2.8883947578149782, + "learning_rate": 1.5375621077842517e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 812 + }, + { + "epoch": 1.002466091245376, + "grad_norm": 2.1490979216413963, + "learning_rate": 1.5364499889194225e-05, + "loss": 0.8653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 813 + }, + { + "epoch": 1.0036991368680641, + "grad_norm": 2.2100133362019787, + "learning_rate": 1.5353369377005928e-05, + "loss": 0.8871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 814 + }, + { + "epoch": 1.0049321824907522, + "grad_norm": 2.1549827645085444, + "learning_rate": 1.5342229560622545e-05, + "loss": 0.9146, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 815 + }, + { + "epoch": 1.0061652281134401, + "grad_norm": 2.0929823364178155, + "learning_rate": 1.5331080459405153e-05, + "loss": 0.8513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 816 + }, + { + "epoch": 1.0073982737361282, + "grad_norm": 2.220368656816853, + "learning_rate": 1.5319922092730974e-05, + "loss": 0.9209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 817 + }, + { + "epoch": 1.0086313193588163, + "grad_norm": 2.1262847612815197, + "learning_rate": 1.5308754479993326e-05, + "loss": 0.8457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 818 + }, + { + "epoch": 1.0098643649815042, + "grad_norm": 2.1791530750144443, + "learning_rate": 1.5297577640601607e-05, + "loss": 0.7645, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 819 + }, + { + "epoch": 1.0110974106041923, + "grad_norm": 2.1726539474102426, + "learning_rate": 1.5286391593981243e-05, + "loss": 0.8684, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 820 + }, + { + "epoch": 1.0123304562268804, + "grad_norm": 2.193333692016209, + "learning_rate": 1.5275196359573667e-05, + "loss": 0.7566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 821 + }, + { + "epoch": 1.0135635018495683, + "grad_norm": 2.9212088946802623, + "learning_rate": 1.526399195683627e-05, + "loss": 1.0019, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 822 + }, + { + "epoch": 1.0147965474722564, + "grad_norm": 2.0580823212102417, + "learning_rate": 1.52527784052424e-05, + "loss": 0.9646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 823 + }, + { + "epoch": 1.0160295930949446, + "grad_norm": 2.19343597586527, + "learning_rate": 1.5241555724281283e-05, + "loss": 0.9453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 824 + }, + { + "epoch": 1.0172626387176325, + "grad_norm": 2.294578049001136, + "learning_rate": 1.523032393345802e-05, + "loss": 1.0273, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 825 + }, + { + "epoch": 1.0184956843403206, + "grad_norm": 1.9511593069959263, + "learning_rate": 1.5219083052293546e-05, + "loss": 0.8161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 826 + }, + { + "epoch": 1.0197287299630087, + "grad_norm": 2.421213793478604, + "learning_rate": 1.5207833100324595e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 827 + }, + { + "epoch": 1.0209617755856966, + "grad_norm": 2.1254569627132422, + "learning_rate": 1.5196574097103669e-05, + "loss": 1.0024, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 828 + }, + { + "epoch": 1.0221948212083847, + "grad_norm": 2.1741883221339213, + "learning_rate": 1.5185306062198993e-05, + "loss": 0.8694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 829 + }, + { + "epoch": 1.0234278668310728, + "grad_norm": 2.340571355784493, + "learning_rate": 1.5174029015194496e-05, + "loss": 0.9235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 830 + }, + { + "epoch": 1.024660912453761, + "grad_norm": 2.173793908678777, + "learning_rate": 1.5162742975689769e-05, + "loss": 0.8448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 831 + }, + { + "epoch": 1.0258939580764488, + "grad_norm": 2.149214329172728, + "learning_rate": 1.5151447963300027e-05, + "loss": 0.8388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 832 + }, + { + "epoch": 1.027127003699137, + "grad_norm": 2.059108722897467, + "learning_rate": 1.5140143997656092e-05, + "loss": 0.7303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 833 + }, + { + "epoch": 1.028360049321825, + "grad_norm": 2.2165337335607536, + "learning_rate": 1.512883109840433e-05, + "loss": 0.873, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 834 + }, + { + "epoch": 1.029593094944513, + "grad_norm": 2.0093629352145923, + "learning_rate": 1.511750928520665e-05, + "loss": 0.7804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 835 + }, + { + "epoch": 1.030826140567201, + "grad_norm": 1.9963120627280608, + "learning_rate": 1.5106178577740442e-05, + "loss": 0.8642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 836 + }, + { + "epoch": 1.0320591861898891, + "grad_norm": 2.1963002199807384, + "learning_rate": 1.5094838995698555e-05, + "loss": 0.8688, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 837 + }, + { + "epoch": 1.033292231812577, + "grad_norm": 2.1874228944498473, + "learning_rate": 1.5083490558789275e-05, + "loss": 0.9912, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 838 + }, + { + "epoch": 1.0345252774352651, + "grad_norm": 2.12533056093305, + "learning_rate": 1.5072133286736261e-05, + "loss": 0.7925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 839 + }, + { + "epoch": 1.0357583230579532, + "grad_norm": 2.5000691541559745, + "learning_rate": 1.5060767199278536e-05, + "loss": 0.9544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 840 + }, + { + "epoch": 1.0369913686806411, + "grad_norm": 2.0622785615015125, + "learning_rate": 1.5049392316170443e-05, + "loss": 0.8433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 841 + }, + { + "epoch": 1.0382244143033292, + "grad_norm": 2.2009668761322563, + "learning_rate": 1.5038008657181616e-05, + "loss": 0.8157, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 842 + }, + { + "epoch": 1.0394574599260173, + "grad_norm": 2.0959248263408385, + "learning_rate": 1.5026616242096932e-05, + "loss": 0.9167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 843 + }, + { + "epoch": 1.0406905055487052, + "grad_norm": 2.1893631450779893, + "learning_rate": 1.5015215090716493e-05, + "loss": 0.9214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 844 + }, + { + "epoch": 1.0419235511713933, + "grad_norm": 2.077397878791412, + "learning_rate": 1.5003805222855584e-05, + "loss": 0.8383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 845 + }, + { + "epoch": 1.0431565967940815, + "grad_norm": 2.4055084927944623, + "learning_rate": 1.4992386658344643e-05, + "loss": 0.8404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 846 + }, + { + "epoch": 1.0443896424167693, + "grad_norm": 2.171777480987558, + "learning_rate": 1.4980959417029208e-05, + "loss": 0.7997, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 847 + }, + { + "epoch": 1.0456226880394575, + "grad_norm": 2.157645250920109, + "learning_rate": 1.4969523518769917e-05, + "loss": 0.8738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 848 + }, + { + "epoch": 1.0468557336621456, + "grad_norm": 2.396639754777913, + "learning_rate": 1.4958078983442441e-05, + "loss": 0.6992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 849 + }, + { + "epoch": 1.0480887792848335, + "grad_norm": 2.4186334460214765, + "learning_rate": 1.4946625830937472e-05, + "loss": 0.8937, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 850 + }, + { + "epoch": 1.0493218249075216, + "grad_norm": 2.2977540698502996, + "learning_rate": 1.4935164081160663e-05, + "loss": 0.8683, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 851 + }, + { + "epoch": 1.0505548705302097, + "grad_norm": 2.1379865919168, + "learning_rate": 1.4923693754032626e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 852 + }, + { + "epoch": 1.0517879161528976, + "grad_norm": 2.259076969791956, + "learning_rate": 1.491221486948887e-05, + "loss": 0.7691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 853 + }, + { + "epoch": 1.0530209617755857, + "grad_norm": 2.1377806626548357, + "learning_rate": 1.490072744747978e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 854 + }, + { + "epoch": 1.0542540073982738, + "grad_norm": 2.1044084023956167, + "learning_rate": 1.4889231507970581e-05, + "loss": 0.8401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 855 + }, + { + "epoch": 1.0554870530209617, + "grad_norm": 2.1271140055745286, + "learning_rate": 1.4877727070941302e-05, + "loss": 0.8084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 856 + }, + { + "epoch": 1.0567200986436498, + "grad_norm": 2.1718066261306817, + "learning_rate": 1.4866214156386734e-05, + "loss": 0.9014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 857 + }, + { + "epoch": 1.057953144266338, + "grad_norm": 2.1176596897595714, + "learning_rate": 1.4854692784316409e-05, + "loss": 0.7752, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 858 + }, + { + "epoch": 1.0591861898890258, + "grad_norm": 2.171891486084014, + "learning_rate": 1.4843162974754555e-05, + "loss": 0.6988, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 859 + }, + { + "epoch": 1.060419235511714, + "grad_norm": 2.0055850881304718, + "learning_rate": 1.4831624747740065e-05, + "loss": 0.7928, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 860 + }, + { + "epoch": 1.061652281134402, + "grad_norm": 2.1272055457979624, + "learning_rate": 1.4820078123326464e-05, + "loss": 0.8844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 861 + }, + { + "epoch": 1.06288532675709, + "grad_norm": 2.2409166254632416, + "learning_rate": 1.4808523121581871e-05, + "loss": 0.8428, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 862 + }, + { + "epoch": 1.064118372379778, + "grad_norm": 2.167571295890406, + "learning_rate": 1.4796959762588957e-05, + "loss": 0.9326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 863 + }, + { + "epoch": 1.0653514180024661, + "grad_norm": 1.9896497336935064, + "learning_rate": 1.4785388066444928e-05, + "loss": 0.7178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 864 + }, + { + "epoch": 1.066584463625154, + "grad_norm": 1.9782830220589522, + "learning_rate": 1.4773808053261478e-05, + "loss": 0.8734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 865 + }, + { + "epoch": 1.0678175092478421, + "grad_norm": 2.1094953563881447, + "learning_rate": 1.4762219743164752e-05, + "loss": 0.72, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 866 + }, + { + "epoch": 1.0690505548705302, + "grad_norm": 2.172378157117765, + "learning_rate": 1.4750623156295316e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 867 + }, + { + "epoch": 1.0702836004932181, + "grad_norm": 2.777083857064727, + "learning_rate": 1.473901831280813e-05, + "loss": 0.887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 868 + }, + { + "epoch": 1.0715166461159062, + "grad_norm": 2.773482686862429, + "learning_rate": 1.4727405232872489e-05, + "loss": 0.9897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 869 + }, + { + "epoch": 1.0727496917385944, + "grad_norm": 2.303987298346897, + "learning_rate": 1.4715783936672017e-05, + "loss": 0.8267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 870 + }, + { + "epoch": 1.0739827373612825, + "grad_norm": 2.5090103249366718, + "learning_rate": 1.4704154444404606e-05, + "loss": 0.8297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 871 + }, + { + "epoch": 1.0752157829839704, + "grad_norm": 2.1804014393140334, + "learning_rate": 1.4692516776282404e-05, + "loss": 0.8673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 872 + }, + { + "epoch": 1.0764488286066585, + "grad_norm": 2.114741425345962, + "learning_rate": 1.4680870952531758e-05, + "loss": 0.7878, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 873 + }, + { + "epoch": 1.0776818742293466, + "grad_norm": 2.1487592065744043, + "learning_rate": 1.4669216993393195e-05, + "loss": 0.702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 874 + }, + { + "epoch": 1.0789149198520345, + "grad_norm": 2.250699104710453, + "learning_rate": 1.4657554919121388e-05, + "loss": 0.7805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 875 + }, + { + "epoch": 1.0801479654747226, + "grad_norm": 2.652130922890576, + "learning_rate": 1.46458847499851e-05, + "loss": 0.8594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 876 + }, + { + "epoch": 1.0813810110974107, + "grad_norm": 2.618911622711661, + "learning_rate": 1.4634206506267172e-05, + "loss": 0.974, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 877 + }, + { + "epoch": 1.0826140567200986, + "grad_norm": 2.104711420819804, + "learning_rate": 1.462252020826448e-05, + "loss": 0.7511, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 878 + }, + { + "epoch": 1.0838471023427867, + "grad_norm": 2.181334503374274, + "learning_rate": 1.4610825876287892e-05, + "loss": 0.8632, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 879 + }, + { + "epoch": 1.0850801479654748, + "grad_norm": 2.2281952334339126, + "learning_rate": 1.4599123530662242e-05, + "loss": 0.7623, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 880 + }, + { + "epoch": 1.0863131935881627, + "grad_norm": 2.2779707650922254, + "learning_rate": 1.4587413191726296e-05, + "loss": 0.7533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 881 + }, + { + "epoch": 1.0875462392108508, + "grad_norm": 2.227912879258747, + "learning_rate": 1.4575694879832704e-05, + "loss": 0.9742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 882 + }, + { + "epoch": 1.088779284833539, + "grad_norm": 2.200751986221906, + "learning_rate": 1.4563968615347983e-05, + "loss": 0.9859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 883 + }, + { + "epoch": 1.0900123304562268, + "grad_norm": 2.1474005643679503, + "learning_rate": 1.455223441865246e-05, + "loss": 0.885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 884 + }, + { + "epoch": 1.091245376078915, + "grad_norm": 2.5635895574555967, + "learning_rate": 1.4540492310140262e-05, + "loss": 0.9648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 885 + }, + { + "epoch": 1.092478421701603, + "grad_norm": 2.086551890894098, + "learning_rate": 1.4528742310219251e-05, + "loss": 0.7838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 886 + }, + { + "epoch": 1.093711467324291, + "grad_norm": 2.100577896933077, + "learning_rate": 1.4516984439311017e-05, + "loss": 0.8245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 887 + }, + { + "epoch": 1.094944512946979, + "grad_norm": 2.0824253868885427, + "learning_rate": 1.4505218717850826e-05, + "loss": 0.8121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 888 + }, + { + "epoch": 1.0961775585696671, + "grad_norm": 2.424807773941655, + "learning_rate": 1.449344516628759e-05, + "loss": 0.6852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 889 + }, + { + "epoch": 1.097410604192355, + "grad_norm": 2.2203614431330587, + "learning_rate": 1.4481663805083823e-05, + "loss": 0.9022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 890 + }, + { + "epoch": 1.0986436498150431, + "grad_norm": 2.194684728317858, + "learning_rate": 1.4469874654715618e-05, + "loss": 0.8018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 891 + }, + { + "epoch": 1.0998766954377313, + "grad_norm": 2.2557943511918066, + "learning_rate": 1.4458077735672606e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 892 + }, + { + "epoch": 1.1011097410604191, + "grad_norm": 2.350846713741873, + "learning_rate": 1.4446273068457916e-05, + "loss": 0.8373, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 893 + }, + { + "epoch": 1.1023427866831073, + "grad_norm": 2.3206912584026584, + "learning_rate": 1.4434460673588151e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 894 + }, + { + "epoch": 1.1035758323057954, + "grad_norm": 2.083251461091603, + "learning_rate": 1.4422640571593336e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 895 + }, + { + "epoch": 1.1048088779284833, + "grad_norm": 2.2429047820431456, + "learning_rate": 1.4410812783016893e-05, + "loss": 0.8066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 896 + }, + { + "epoch": 1.1060419235511714, + "grad_norm": 2.39736831788261, + "learning_rate": 1.4398977328415608e-05, + "loss": 0.8961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 897 + }, + { + "epoch": 1.1072749691738595, + "grad_norm": 2.4569397246397995, + "learning_rate": 1.4387134228359587e-05, + "loss": 0.9745, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 898 + }, + { + "epoch": 1.1085080147965474, + "grad_norm": 2.194341713414294, + "learning_rate": 1.4375283503432223e-05, + "loss": 0.7642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 899 + }, + { + "epoch": 1.1097410604192355, + "grad_norm": 2.108094195937167, + "learning_rate": 1.4363425174230161e-05, + "loss": 0.7856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 900 + }, + { + "epoch": 1.1109741060419236, + "grad_norm": 2.140652113297743, + "learning_rate": 1.4351559261363267e-05, + "loss": 0.6648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 901 + }, + { + "epoch": 1.1122071516646117, + "grad_norm": 2.1936452316718063, + "learning_rate": 1.4339685785454587e-05, + "loss": 0.7336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 902 + }, + { + "epoch": 1.1134401972872996, + "grad_norm": 2.2269402261375038, + "learning_rate": 1.4327804767140303e-05, + "loss": 0.7708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 903 + }, + { + "epoch": 1.1146732429099877, + "grad_norm": 2.2989998022422453, + "learning_rate": 1.4315916227069713e-05, + "loss": 0.743, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 904 + }, + { + "epoch": 1.1159062885326758, + "grad_norm": 2.2917968867963427, + "learning_rate": 1.4304020185905189e-05, + "loss": 0.8126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 905 + }, + { + "epoch": 1.1171393341553637, + "grad_norm": 2.217875844794138, + "learning_rate": 1.4292116664322134e-05, + "loss": 0.6544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 906 + }, + { + "epoch": 1.1183723797780518, + "grad_norm": 2.4162119819935803, + "learning_rate": 1.4280205683008959e-05, + "loss": 0.9659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 907 + }, + { + "epoch": 1.11960542540074, + "grad_norm": 2.3603757259614047, + "learning_rate": 1.4268287262667033e-05, + "loss": 0.7766, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 908 + }, + { + "epoch": 1.1208384710234278, + "grad_norm": 2.323141591914079, + "learning_rate": 1.4256361424010663e-05, + "loss": 0.8892, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 909 + }, + { + "epoch": 1.122071516646116, + "grad_norm": 2.196302554500726, + "learning_rate": 1.4244428187767039e-05, + "loss": 0.9378, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 910 + }, + { + "epoch": 1.123304562268804, + "grad_norm": 2.2395403393552393, + "learning_rate": 1.4232487574676212e-05, + "loss": 1.0055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 911 + }, + { + "epoch": 1.124537607891492, + "grad_norm": 2.15769661287864, + "learning_rate": 1.4220539605491057e-05, + "loss": 0.748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 912 + }, + { + "epoch": 1.12577065351418, + "grad_norm": 2.0601339442187605, + "learning_rate": 1.4208584300977233e-05, + "loss": 0.7894, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 913 + }, + { + "epoch": 1.1270036991368682, + "grad_norm": 2.10447957980717, + "learning_rate": 1.4196621681913145e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 914 + }, + { + "epoch": 1.128236744759556, + "grad_norm": 2.3961229528506265, + "learning_rate": 1.4184651769089911e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 915 + }, + { + "epoch": 1.1294697903822442, + "grad_norm": 2.457287266967065, + "learning_rate": 1.417267458331133e-05, + "loss": 0.8519, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 916 + }, + { + "epoch": 1.1307028360049323, + "grad_norm": 2.023004753946474, + "learning_rate": 1.4160690145393838e-05, + "loss": 0.8449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 917 + }, + { + "epoch": 1.1319358816276202, + "grad_norm": 2.2533720475013634, + "learning_rate": 1.4148698476166471e-05, + "loss": 0.8221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 918 + }, + { + "epoch": 1.1331689272503083, + "grad_norm": 2.105957236229041, + "learning_rate": 1.4136699596470846e-05, + "loss": 0.8287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 919 + }, + { + "epoch": 1.1344019728729964, + "grad_norm": 2.1652513956532946, + "learning_rate": 1.4124693527161095e-05, + "loss": 0.8322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 920 + }, + { + "epoch": 1.1356350184956843, + "grad_norm": 2.2463359048305693, + "learning_rate": 1.4112680289103858e-05, + "loss": 0.8458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 921 + }, + { + "epoch": 1.1368680641183724, + "grad_norm": 2.1946517808282784, + "learning_rate": 1.4100659903178231e-05, + "loss": 0.7948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 922 + }, + { + "epoch": 1.1381011097410605, + "grad_norm": 2.116652661158196, + "learning_rate": 1.4088632390275731e-05, + "loss": 0.8496, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 923 + }, + { + "epoch": 1.1393341553637484, + "grad_norm": 2.1260376630680677, + "learning_rate": 1.4076597771300267e-05, + "loss": 0.7646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 924 + }, + { + "epoch": 1.1405672009864365, + "grad_norm": 2.1008385205849684, + "learning_rate": 1.4064556067168092e-05, + "loss": 0.8303, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 925 + }, + { + "epoch": 1.1418002466091246, + "grad_norm": 2.1995475134175573, + "learning_rate": 1.4052507298807774e-05, + "loss": 0.8636, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 926 + }, + { + "epoch": 1.1430332922318125, + "grad_norm": 2.238025377523696, + "learning_rate": 1.404045148716016e-05, + "loss": 0.9095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 927 + }, + { + "epoch": 1.1442663378545006, + "grad_norm": 2.1584647692083636, + "learning_rate": 1.4028388653178346e-05, + "loss": 0.9552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 928 + }, + { + "epoch": 1.1454993834771887, + "grad_norm": 2.277913607137681, + "learning_rate": 1.4016318817827618e-05, + "loss": 0.6826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 929 + }, + { + "epoch": 1.1467324290998766, + "grad_norm": 2.1210301206048494, + "learning_rate": 1.4004242002085441e-05, + "loss": 0.7648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 930 + }, + { + "epoch": 1.1479654747225647, + "grad_norm": 2.1730216571228143, + "learning_rate": 1.3992158226941409e-05, + "loss": 0.7348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 931 + }, + { + "epoch": 1.1491985203452528, + "grad_norm": 2.2903920964212197, + "learning_rate": 1.3980067513397208e-05, + "loss": 0.9628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 932 + }, + { + "epoch": 1.1504315659679407, + "grad_norm": 2.207582230014581, + "learning_rate": 1.3967969882466595e-05, + "loss": 0.8393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 933 + }, + { + "epoch": 1.1516646115906288, + "grad_norm": 2.1487993957640605, + "learning_rate": 1.3955865355175334e-05, + "loss": 0.9618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 934 + }, + { + "epoch": 1.152897657213317, + "grad_norm": 2.1100013581157673, + "learning_rate": 1.3943753952561179e-05, + "loss": 0.7185, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 935 + }, + { + "epoch": 1.1541307028360048, + "grad_norm": 2.1580493494759505, + "learning_rate": 1.3931635695673845e-05, + "loss": 0.8656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 936 + }, + { + "epoch": 1.155363748458693, + "grad_norm": 2.449600660057695, + "learning_rate": 1.3919510605574944e-05, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 937 + }, + { + "epoch": 1.156596794081381, + "grad_norm": 2.2204766170828782, + "learning_rate": 1.3907378703337977e-05, + "loss": 0.823, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 938 + }, + { + "epoch": 1.157829839704069, + "grad_norm": 2.563974651134322, + "learning_rate": 1.389524001004827e-05, + "loss": 0.7685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 939 + }, + { + "epoch": 1.159062885326757, + "grad_norm": 2.258821529498267, + "learning_rate": 1.388309454680297e-05, + "loss": 0.673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 940 + }, + { + "epoch": 1.1602959309494452, + "grad_norm": 2.292191724384705, + "learning_rate": 1.3870942334710978e-05, + "loss": 0.8204, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 941 + }, + { + "epoch": 1.161528976572133, + "grad_norm": 2.283429457608797, + "learning_rate": 1.3858783394892925e-05, + "loss": 0.7777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 942 + }, + { + "epoch": 1.1627620221948212, + "grad_norm": 2.186748888333124, + "learning_rate": 1.3846617748481137e-05, + "loss": 0.7862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 943 + }, + { + "epoch": 1.1639950678175093, + "grad_norm": 2.262303399886297, + "learning_rate": 1.3834445416619602e-05, + "loss": 0.7559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 944 + }, + { + "epoch": 1.1652281134401972, + "grad_norm": 2.262832253985301, + "learning_rate": 1.3822266420463914e-05, + "loss": 0.9429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 945 + }, + { + "epoch": 1.1664611590628853, + "grad_norm": 2.150186199729938, + "learning_rate": 1.3810080781181264e-05, + "loss": 0.814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 946 + }, + { + "epoch": 1.1676942046855734, + "grad_norm": 2.343694448653775, + "learning_rate": 1.3797888519950373e-05, + "loss": 0.93, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 947 + }, + { + "epoch": 1.1689272503082615, + "grad_norm": 2.1813728957413385, + "learning_rate": 1.3785689657961494e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 948 + }, + { + "epoch": 1.1701602959309494, + "grad_norm": 2.119521364690522, + "learning_rate": 1.3773484216416326e-05, + "loss": 0.6718, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 949 + }, + { + "epoch": 1.1713933415536375, + "grad_norm": 2.2831289714651537, + "learning_rate": 1.3761272216528023e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 950 + }, + { + "epoch": 1.1726263871763256, + "grad_norm": 2.431755167028705, + "learning_rate": 1.3749053679521123e-05, + "loss": 0.8664, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 951 + }, + { + "epoch": 1.1738594327990135, + "grad_norm": 2.148435358078549, + "learning_rate": 1.3736828626631539e-05, + "loss": 0.6746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 952 + }, + { + "epoch": 1.1750924784217016, + "grad_norm": 2.124965667169116, + "learning_rate": 1.3724597079106498e-05, + "loss": 0.9202, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 953 + }, + { + "epoch": 1.1763255240443897, + "grad_norm": 2.2382797156893397, + "learning_rate": 1.3712359058204525e-05, + "loss": 0.7696, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 954 + }, + { + "epoch": 1.1775585696670776, + "grad_norm": 2.2207967591300632, + "learning_rate": 1.3700114585195381e-05, + "loss": 0.8654, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 955 + }, + { + "epoch": 1.1787916152897657, + "grad_norm": 2.3010582856815347, + "learning_rate": 1.3687863681360059e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 956 + }, + { + "epoch": 1.1800246609124538, + "grad_norm": 2.146527954418168, + "learning_rate": 1.367560636799071e-05, + "loss": 0.6291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 957 + }, + { + "epoch": 1.1812577065351417, + "grad_norm": 2.203217169625211, + "learning_rate": 1.3663342666390642e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 958 + }, + { + "epoch": 1.1824907521578298, + "grad_norm": 2.380280804061753, + "learning_rate": 1.365107259787425e-05, + "loss": 0.8015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 959 + }, + { + "epoch": 1.183723797780518, + "grad_norm": 2.7057879375717766, + "learning_rate": 1.3638796183767013e-05, + "loss": 0.8469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 960 + }, + { + "epoch": 1.1849568434032058, + "grad_norm": 2.0549864237362856, + "learning_rate": 1.362651344540542e-05, + "loss": 0.7724, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 961 + }, + { + "epoch": 1.186189889025894, + "grad_norm": 2.165523344913771, + "learning_rate": 1.3614224404136964e-05, + "loss": 0.7119, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 962 + }, + { + "epoch": 1.187422934648582, + "grad_norm": 2.193503404484943, + "learning_rate": 1.3601929081320085e-05, + "loss": 0.769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 963 + }, + { + "epoch": 1.18865598027127, + "grad_norm": 2.30296466364579, + "learning_rate": 1.358962749832415e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 964 + }, + { + "epoch": 1.189889025893958, + "grad_norm": 2.3370321886887226, + "learning_rate": 1.3577319676529394e-05, + "loss": 0.8251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 965 + }, + { + "epoch": 1.1911220715166462, + "grad_norm": 2.1447616659612727, + "learning_rate": 1.3565005637326902e-05, + "loss": 0.8529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 966 + }, + { + "epoch": 1.192355117139334, + "grad_norm": 2.1126610192998263, + "learning_rate": 1.3552685402118569e-05, + "loss": 0.893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 967 + }, + { + "epoch": 1.1935881627620222, + "grad_norm": 2.145481581859885, + "learning_rate": 1.3540358992317048e-05, + "loss": 0.8777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 968 + }, + { + "epoch": 1.1948212083847103, + "grad_norm": 2.0436843289234714, + "learning_rate": 1.3528026429345733e-05, + "loss": 0.7041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 969 + }, + { + "epoch": 1.1960542540073984, + "grad_norm": 2.3729052917065863, + "learning_rate": 1.351568773463871e-05, + "loss": 0.7822, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 970 + }, + { + "epoch": 1.1972872996300863, + "grad_norm": 2.2771218973970977, + "learning_rate": 1.3503342929640713e-05, + "loss": 0.6887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 971 + }, + { + "epoch": 1.1985203452527744, + "grad_norm": 2.114743129136917, + "learning_rate": 1.3490992035807108e-05, + "loss": 0.6587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 972 + }, + { + "epoch": 1.1997533908754625, + "grad_norm": 2.210055964932895, + "learning_rate": 1.3478635074603844e-05, + "loss": 0.6986, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 973 + }, + { + "epoch": 1.2009864364981504, + "grad_norm": 2.376467969342028, + "learning_rate": 1.34662720675074e-05, + "loss": 0.8992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 974 + }, + { + "epoch": 1.2022194821208385, + "grad_norm": 2.3169180290916893, + "learning_rate": 1.3453903036004784e-05, + "loss": 0.8468, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 975 + }, + { + "epoch": 1.2034525277435266, + "grad_norm": 2.4929171744348095, + "learning_rate": 1.3441528001593448e-05, + "loss": 0.8746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 976 + }, + { + "epoch": 1.2046855733662145, + "grad_norm": 2.3498012552061223, + "learning_rate": 1.3429146985781305e-05, + "loss": 0.848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 977 + }, + { + "epoch": 1.2059186189889026, + "grad_norm": 2.3519351668545223, + "learning_rate": 1.3416760010086647e-05, + "loss": 0.8045, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 978 + }, + { + "epoch": 1.2071516646115907, + "grad_norm": 2.152128272403449, + "learning_rate": 1.3404367096038127e-05, + "loss": 0.6621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 979 + }, + { + "epoch": 1.2083847102342786, + "grad_norm": 2.2310865684580126, + "learning_rate": 1.339196826517472e-05, + "loss": 0.8074, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 980 + }, + { + "epoch": 1.2096177558569667, + "grad_norm": 2.4218206259337336, + "learning_rate": 1.3379563539045688e-05, + "loss": 0.7489, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 981 + }, + { + "epoch": 1.2108508014796548, + "grad_norm": 2.3171020050526754, + "learning_rate": 1.3367152939210532e-05, + "loss": 0.6967, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 982 + }, + { + "epoch": 1.2120838471023427, + "grad_norm": 2.253666664504199, + "learning_rate": 1.3354736487238968e-05, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 983 + }, + { + "epoch": 1.2133168927250308, + "grad_norm": 2.202923113820963, + "learning_rate": 1.3342314204710878e-05, + "loss": 0.6643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 984 + }, + { + "epoch": 1.214549938347719, + "grad_norm": 2.3907506714199216, + "learning_rate": 1.332988611321628e-05, + "loss": 0.7483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 985 + }, + { + "epoch": 1.2157829839704068, + "grad_norm": 2.3288855585799726, + "learning_rate": 1.3317452234355286e-05, + "loss": 0.8775, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 986 + }, + { + "epoch": 1.217016029593095, + "grad_norm": 2.4270404362139124, + "learning_rate": 1.3305012589738075e-05, + "loss": 0.9152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 987 + }, + { + "epoch": 1.218249075215783, + "grad_norm": 2.3906498632865825, + "learning_rate": 1.3292567200984833e-05, + "loss": 0.6825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 988 + }, + { + "epoch": 1.219482120838471, + "grad_norm": 2.1776282578685495, + "learning_rate": 1.3280116089725743e-05, + "loss": 0.7098, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 989 + }, + { + "epoch": 1.220715166461159, + "grad_norm": 2.3972197284848265, + "learning_rate": 1.3267659277600921e-05, + "loss": 0.8804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 990 + }, + { + "epoch": 1.2219482120838472, + "grad_norm": 2.272454158882342, + "learning_rate": 1.3255196786260402e-05, + "loss": 0.7391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 991 + }, + { + "epoch": 1.223181257706535, + "grad_norm": 2.0898886573973985, + "learning_rate": 1.3242728637364084e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 992 + }, + { + "epoch": 1.2244143033292232, + "grad_norm": 2.1283771526768422, + "learning_rate": 1.3230254852581703e-05, + "loss": 0.7541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 993 + }, + { + "epoch": 1.2256473489519113, + "grad_norm": 2.5947641083937785, + "learning_rate": 1.3217775453592786e-05, + "loss": 0.6516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 994 + }, + { + "epoch": 1.2268803945745992, + "grad_norm": 2.2994678179446155, + "learning_rate": 1.3205290462086623e-05, + "loss": 0.8175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 995 + }, + { + "epoch": 1.2281134401972873, + "grad_norm": 2.419174005886699, + "learning_rate": 1.3192799899762216e-05, + "loss": 0.6653, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 996 + }, + { + "epoch": 1.2293464858199754, + "grad_norm": 2.7052778802902724, + "learning_rate": 1.318030378832826e-05, + "loss": 0.8398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 997 + }, + { + "epoch": 1.2305795314426633, + "grad_norm": 2.182898548205837, + "learning_rate": 1.3167802149503079e-05, + "loss": 0.7902, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 998 + }, + { + "epoch": 1.2318125770653514, + "grad_norm": 2.319601966909909, + "learning_rate": 1.315529500501462e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 999 + }, + { + "epoch": 1.2330456226880395, + "grad_norm": 2.3070852285534875, + "learning_rate": 1.3142782376600389e-05, + "loss": 0.7369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1000 + }, + { + "epoch": 1.2342786683107274, + "grad_norm": 2.3823064106232956, + "learning_rate": 1.3130264286007427e-05, + "loss": 0.8759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1001 + }, + { + "epoch": 1.2355117139334155, + "grad_norm": 2.2502915191290405, + "learning_rate": 1.3117740754992263e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1002 + }, + { + "epoch": 1.2367447595561036, + "grad_norm": 2.329022774742351, + "learning_rate": 1.310521180532089e-05, + "loss": 0.6222, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1003 + }, + { + "epoch": 1.2379778051787915, + "grad_norm": 2.451176371025935, + "learning_rate": 1.3092677458768712e-05, + "loss": 0.9302, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1004 + }, + { + "epoch": 1.2392108508014796, + "grad_norm": 2.266087500621217, + "learning_rate": 1.3080137737120514e-05, + "loss": 0.7132, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1005 + }, + { + "epoch": 1.2404438964241677, + "grad_norm": 2.2897262982963205, + "learning_rate": 1.3067592662170424e-05, + "loss": 0.7692, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1006 + }, + { + "epoch": 1.2416769420468556, + "grad_norm": 2.321897633330746, + "learning_rate": 1.3055042255721877e-05, + "loss": 0.8054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1007 + }, + { + "epoch": 1.2429099876695437, + "grad_norm": 2.327507620912996, + "learning_rate": 1.3042486539587566e-05, + "loss": 0.8495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1008 + }, + { + "epoch": 1.2441430332922319, + "grad_norm": 2.144625913300158, + "learning_rate": 1.3029925535589423e-05, + "loss": 0.7546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1009 + }, + { + "epoch": 1.2453760789149197, + "grad_norm": 2.1664603999352217, + "learning_rate": 1.301735926555856e-05, + "loss": 0.758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1010 + }, + { + "epoch": 1.2466091245376079, + "grad_norm": 2.2424395261695422, + "learning_rate": 1.300478775133525e-05, + "loss": 0.6794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1011 + }, + { + "epoch": 1.247842170160296, + "grad_norm": 2.135987556369877, + "learning_rate": 1.2992211014768872e-05, + "loss": 0.7814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1012 + }, + { + "epoch": 1.2490752157829839, + "grad_norm": 2.3429443823198812, + "learning_rate": 1.297962907771789e-05, + "loss": 0.756, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1013 + }, + { + "epoch": 1.250308261405672, + "grad_norm": 2.3806466769436083, + "learning_rate": 1.29670419620498e-05, + "loss": 0.7614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1014 + }, + { + "epoch": 1.25154130702836, + "grad_norm": 2.2614648564029536, + "learning_rate": 1.2954449689641105e-05, + "loss": 0.6227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1015 + }, + { + "epoch": 1.252774352651048, + "grad_norm": 2.690263466615422, + "learning_rate": 1.294185228237726e-05, + "loss": 0.8233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1016 + }, + { + "epoch": 1.254007398273736, + "grad_norm": 2.473463220519423, + "learning_rate": 1.2929249762152659e-05, + "loss": 0.7693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1017 + }, + { + "epoch": 1.2552404438964242, + "grad_norm": 2.203357353825905, + "learning_rate": 1.2916642150870567e-05, + "loss": 0.7486, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1018 + }, + { + "epoch": 1.256473489519112, + "grad_norm": 2.407886169080358, + "learning_rate": 1.2904029470443105e-05, + "loss": 0.845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1019 + }, + { + "epoch": 1.2577065351418002, + "grad_norm": 2.416338321530697, + "learning_rate": 1.289141174279121e-05, + "loss": 0.7849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1020 + }, + { + "epoch": 1.2589395807644883, + "grad_norm": 2.1578914081370617, + "learning_rate": 1.2878788989844577e-05, + "loss": 0.8065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1021 + }, + { + "epoch": 1.2601726263871762, + "grad_norm": 2.111289324604086, + "learning_rate": 1.2866161233541648e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1022 + }, + { + "epoch": 1.2614056720098643, + "grad_norm": 2.463350736417066, + "learning_rate": 1.2853528495829557e-05, + "loss": 0.7408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1023 + }, + { + "epoch": 1.2626387176325524, + "grad_norm": 2.5227574157477424, + "learning_rate": 1.284089079866409e-05, + "loss": 0.6943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1024 + }, + { + "epoch": 1.2638717632552403, + "grad_norm": 2.550177757961712, + "learning_rate": 1.2828248164009658e-05, + "loss": 0.8063, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1025 + }, + { + "epoch": 1.2651048088779284, + "grad_norm": 2.279399386552855, + "learning_rate": 1.2815600613839254e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1026 + }, + { + "epoch": 1.2663378545006165, + "grad_norm": 2.403266684764464, + "learning_rate": 1.2802948170134411e-05, + "loss": 0.8159, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1027 + }, + { + "epoch": 1.2675709001233046, + "grad_norm": 2.3974372248557088, + "learning_rate": 1.2790290854885173e-05, + "loss": 0.7733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1028 + }, + { + "epoch": 1.2688039457459925, + "grad_norm": 2.75325380275428, + "learning_rate": 1.2777628690090035e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1029 + }, + { + "epoch": 1.2700369913686806, + "grad_norm": 2.253046145862575, + "learning_rate": 1.2764961697755947e-05, + "loss": 0.8037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1030 + }, + { + "epoch": 1.2712700369913688, + "grad_norm": 2.0946644139817967, + "learning_rate": 1.2752289899898225e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1031 + }, + { + "epoch": 1.2725030826140566, + "grad_norm": 2.2771512156869576, + "learning_rate": 1.2739613318540546e-05, + "loss": 0.8351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1032 + }, + { + "epoch": 1.2737361282367448, + "grad_norm": 2.360325908676525, + "learning_rate": 1.2726931975714904e-05, + "loss": 0.8367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1033 + }, + { + "epoch": 1.2749691738594329, + "grad_norm": 2.310569034824901, + "learning_rate": 1.2714245893461568e-05, + "loss": 0.7346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1034 + }, + { + "epoch": 1.2762022194821208, + "grad_norm": 2.1364800896900413, + "learning_rate": 1.2701555093829034e-05, + "loss": 0.6637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1035 + }, + { + "epoch": 1.2774352651048089, + "grad_norm": 2.2320085770485734, + "learning_rate": 1.2688859598874013e-05, + "loss": 0.6849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1036 + }, + { + "epoch": 1.278668310727497, + "grad_norm": 2.2073805269530067, + "learning_rate": 1.2676159430661361e-05, + "loss": 0.7287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1037 + }, + { + "epoch": 1.279901356350185, + "grad_norm": 2.2795714091180694, + "learning_rate": 1.2663454611264065e-05, + "loss": 0.8542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1038 + }, + { + "epoch": 1.281134401972873, + "grad_norm": 2.4086359220078535, + "learning_rate": 1.2650745162763195e-05, + "loss": 0.6699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1039 + }, + { + "epoch": 1.282367447595561, + "grad_norm": 2.3833290498603787, + "learning_rate": 1.2638031107247871e-05, + "loss": 0.7976, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1040 + }, + { + "epoch": 1.2836004932182492, + "grad_norm": 2.4078757121437873, + "learning_rate": 1.26253124668152e-05, + "loss": 0.7284, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1041 + }, + { + "epoch": 1.284833538840937, + "grad_norm": 2.3345693841609045, + "learning_rate": 1.2612589263570287e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1042 + }, + { + "epoch": 1.2860665844636252, + "grad_norm": 2.2804486360942633, + "learning_rate": 1.2599861519626142e-05, + "loss": 0.8166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1043 + }, + { + "epoch": 1.2872996300863133, + "grad_norm": 2.219596427341039, + "learning_rate": 1.2587129257103684e-05, + "loss": 0.723, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1044 + }, + { + "epoch": 1.2885326757090012, + "grad_norm": 2.318632178358617, + "learning_rate": 1.2574392498131671e-05, + "loss": 0.7334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1045 + }, + { + "epoch": 1.2897657213316893, + "grad_norm": 2.1919076842831933, + "learning_rate": 1.2561651264846692e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1046 + }, + { + "epoch": 1.2909987669543774, + "grad_norm": 2.4787933074669186, + "learning_rate": 1.2548905579393096e-05, + "loss": 0.834, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1047 + }, + { + "epoch": 1.2922318125770653, + "grad_norm": 2.4023622763250883, + "learning_rate": 1.2536155463922982e-05, + "loss": 0.826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1048 + }, + { + "epoch": 1.2934648581997534, + "grad_norm": 2.117125770604436, + "learning_rate": 1.252340094059614e-05, + "loss": 0.8052, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1049 + }, + { + "epoch": 1.2946979038224415, + "grad_norm": 2.2921041661226926, + "learning_rate": 1.251064203158003e-05, + "loss": 0.875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1050 + }, + { + "epoch": 1.2959309494451294, + "grad_norm": 2.5171575376979334, + "learning_rate": 1.2497878759049722e-05, + "loss": 0.6485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1051 + }, + { + "epoch": 1.2971639950678175, + "grad_norm": 2.1128556413546806, + "learning_rate": 1.248511114518788e-05, + "loss": 0.6076, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1052 + }, + { + "epoch": 1.2983970406905057, + "grad_norm": 2.1493124843340907, + "learning_rate": 1.2472339212184712e-05, + "loss": 0.7343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1053 + }, + { + "epoch": 1.2996300863131935, + "grad_norm": 2.889368089302167, + "learning_rate": 1.2459562982237933e-05, + "loss": 0.7978, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1054 + }, + { + "epoch": 1.3008631319358817, + "grad_norm": 2.495383798802243, + "learning_rate": 1.2446782477552718e-05, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1055 + }, + { + "epoch": 1.3020961775585698, + "grad_norm": 2.517167273202911, + "learning_rate": 1.2433997720341682e-05, + "loss": 0.8663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1056 + }, + { + "epoch": 1.3033292231812577, + "grad_norm": 2.33725722509719, + "learning_rate": 1.2421208732824823e-05, + "loss": 0.7587, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1057 + }, + { + "epoch": 1.3045622688039458, + "grad_norm": 2.29088652068904, + "learning_rate": 1.2408415537229497e-05, + "loss": 0.8488, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1058 + }, + { + "epoch": 1.3057953144266339, + "grad_norm": 2.283233783970098, + "learning_rate": 1.239561815579037e-05, + "loss": 0.6389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1059 + }, + { + "epoch": 1.3070283600493218, + "grad_norm": 2.2928393078851568, + "learning_rate": 1.2382816610749388e-05, + "loss": 0.7261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1060 + }, + { + "epoch": 1.3082614056720099, + "grad_norm": 2.163749641191809, + "learning_rate": 1.2370010924355726e-05, + "loss": 0.6366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1061 + }, + { + "epoch": 1.309494451294698, + "grad_norm": 2.1229525047726963, + "learning_rate": 1.2357201118865762e-05, + "loss": 0.7879, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1062 + }, + { + "epoch": 1.3107274969173859, + "grad_norm": 2.272499973037141, + "learning_rate": 1.234438721654303e-05, + "loss": 0.6777, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1063 + }, + { + "epoch": 1.311960542540074, + "grad_norm": 2.2172413085789207, + "learning_rate": 1.233156923965819e-05, + "loss": 0.8492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1064 + }, + { + "epoch": 1.313193588162762, + "grad_norm": 2.215217499299492, + "learning_rate": 1.2318747210488973e-05, + "loss": 0.816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1065 + }, + { + "epoch": 1.31442663378545, + "grad_norm": 2.2046121405476384, + "learning_rate": 1.2305921151320164e-05, + "loss": 0.7874, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1066 + }, + { + "epoch": 1.315659679408138, + "grad_norm": 2.410535974252942, + "learning_rate": 1.2293091084443547e-05, + "loss": 0.8263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1067 + }, + { + "epoch": 1.3168927250308262, + "grad_norm": 2.2346919759512427, + "learning_rate": 1.2280257032157873e-05, + "loss": 0.5882, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1068 + }, + { + "epoch": 1.318125770653514, + "grad_norm": 2.217353695685963, + "learning_rate": 1.2267419016768814e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1069 + }, + { + "epoch": 1.3193588162762022, + "grad_norm": 2.3310973124804892, + "learning_rate": 1.225457706058894e-05, + "loss": 0.8387, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1070 + }, + { + "epoch": 1.3205918618988903, + "grad_norm": 2.363091956760975, + "learning_rate": 1.224173118593766e-05, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1071 + }, + { + "epoch": 1.3218249075215782, + "grad_norm": 2.3530303319299835, + "learning_rate": 1.22288814151412e-05, + "loss": 0.7271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1072 + }, + { + "epoch": 1.3230579531442663, + "grad_norm": 2.3412040138713137, + "learning_rate": 1.2216027770532559e-05, + "loss": 0.7482, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1073 + }, + { + "epoch": 1.3242909987669544, + "grad_norm": 2.388832200908418, + "learning_rate": 1.2203170274451462e-05, + "loss": 0.6927, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1074 + }, + { + "epoch": 1.3255240443896423, + "grad_norm": 2.107265565119065, + "learning_rate": 1.219030894924433e-05, + "loss": 0.6069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1075 + }, + { + "epoch": 1.3267570900123304, + "grad_norm": 2.249447889766729, + "learning_rate": 1.2177443817264241e-05, + "loss": 0.7429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1076 + }, + { + "epoch": 1.3279901356350186, + "grad_norm": 2.337103531405435, + "learning_rate": 1.2164574900870887e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1077 + }, + { + "epoch": 1.3292231812577064, + "grad_norm": 2.4954282627925806, + "learning_rate": 1.2151702222430542e-05, + "loss": 0.781, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1078 + }, + { + "epoch": 1.3304562268803946, + "grad_norm": 2.4023586193420714, + "learning_rate": 1.2138825804316015e-05, + "loss": 0.7126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1079 + }, + { + "epoch": 1.3316892725030827, + "grad_norm": 2.293592183809771, + "learning_rate": 1.2125945668906611e-05, + "loss": 0.859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1080 + }, + { + "epoch": 1.3329223181257706, + "grad_norm": 2.280827683105619, + "learning_rate": 1.2113061838588104e-05, + "loss": 0.7505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1081 + }, + { + "epoch": 1.3341553637484587, + "grad_norm": 2.2618423177043647, + "learning_rate": 1.2100174335752683e-05, + "loss": 0.8226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1082 + }, + { + "epoch": 1.3353884093711468, + "grad_norm": 2.263174374532883, + "learning_rate": 1.2087283182798923e-05, + "loss": 0.8671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1083 + }, + { + "epoch": 1.3366214549938347, + "grad_norm": 2.3088875209533875, + "learning_rate": 1.2074388402131737e-05, + "loss": 0.7575, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1084 + }, + { + "epoch": 1.3378545006165228, + "grad_norm": 2.217074039041576, + "learning_rate": 1.2061490016162357e-05, + "loss": 0.8498, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1085 + }, + { + "epoch": 1.339087546239211, + "grad_norm": 2.558767932572183, + "learning_rate": 1.2048588047308263e-05, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1086 + }, + { + "epoch": 1.3403205918618988, + "grad_norm": 2.1426590192119392, + "learning_rate": 1.203568251799318e-05, + "loss": 0.8473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1087 + }, + { + "epoch": 1.341553637484587, + "grad_norm": 2.3130116023826903, + "learning_rate": 1.2022773450647002e-05, + "loss": 0.6533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1088 + }, + { + "epoch": 1.342786683107275, + "grad_norm": 2.2474452770176505, + "learning_rate": 1.2009860867705793e-05, + "loss": 0.7731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1089 + }, + { + "epoch": 1.344019728729963, + "grad_norm": 2.553351511741815, + "learning_rate": 1.1996944791611711e-05, + "loss": 0.8881, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1090 + }, + { + "epoch": 1.345252774352651, + "grad_norm": 2.4417233034036747, + "learning_rate": 1.1984025244812987e-05, + "loss": 0.7436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1091 + }, + { + "epoch": 1.3464858199753391, + "grad_norm": 2.02108560427268, + "learning_rate": 1.1971102249763895e-05, + "loss": 0.5307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1092 + }, + { + "epoch": 1.347718865598027, + "grad_norm": 2.4627196027809655, + "learning_rate": 1.1958175828924692e-05, + "loss": 0.6095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1093 + }, + { + "epoch": 1.3489519112207151, + "grad_norm": 2.2087226765380703, + "learning_rate": 1.194524600476159e-05, + "loss": 0.7225, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1094 + }, + { + "epoch": 1.3501849568434032, + "grad_norm": 2.268031072263066, + "learning_rate": 1.1932312799746722e-05, + "loss": 0.7492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1095 + }, + { + "epoch": 1.3514180024660911, + "grad_norm": 2.187415973980175, + "learning_rate": 1.1919376236358087e-05, + "loss": 0.6372, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1096 + }, + { + "epoch": 1.3526510480887792, + "grad_norm": 2.3776853142465755, + "learning_rate": 1.1906436337079531e-05, + "loss": 0.757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1097 + }, + { + "epoch": 1.3538840937114673, + "grad_norm": 2.633083891326042, + "learning_rate": 1.189349312440069e-05, + "loss": 0.741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1098 + }, + { + "epoch": 1.3551171393341552, + "grad_norm": 2.3175112181742703, + "learning_rate": 1.1880546620816971e-05, + "loss": 0.577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1099 + }, + { + "epoch": 1.3563501849568433, + "grad_norm": 2.57847955905109, + "learning_rate": 1.1867596848829479e-05, + "loss": 0.7485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1100 + }, + { + "epoch": 1.3575832305795315, + "grad_norm": 2.2807181061723223, + "learning_rate": 1.1854643830945023e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1101 + }, + { + "epoch": 1.3588162762022196, + "grad_norm": 2.150298874669539, + "learning_rate": 1.1841687589676037e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1102 + }, + { + "epoch": 1.3600493218249075, + "grad_norm": 2.239857482190429, + "learning_rate": 1.1828728147540565e-05, + "loss": 0.5845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1103 + }, + { + "epoch": 1.3612823674475956, + "grad_norm": 2.391169760631781, + "learning_rate": 1.1815765527062207e-05, + "loss": 0.836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1104 + }, + { + "epoch": 1.3625154130702837, + "grad_norm": 2.233510553215683, + "learning_rate": 1.18027997507701e-05, + "loss": 0.687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1105 + }, + { + "epoch": 1.3637484586929716, + "grad_norm": 2.3406994321803376, + "learning_rate": 1.1789830841198854e-05, + "loss": 0.541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1106 + }, + { + "epoch": 1.3649815043156597, + "grad_norm": 3.0813520964620884, + "learning_rate": 1.1776858820888532e-05, + "loss": 0.8939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1107 + }, + { + "epoch": 1.3662145499383478, + "grad_norm": 2.289968441014776, + "learning_rate": 1.1763883712384593e-05, + "loss": 0.6178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1108 + }, + { + "epoch": 1.3674475955610357, + "grad_norm": 2.753373823273893, + "learning_rate": 1.1750905538237877e-05, + "loss": 0.8218, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1109 + }, + { + "epoch": 1.3686806411837238, + "grad_norm": 2.582931593682227, + "learning_rate": 1.1737924321004544e-05, + "loss": 0.5923, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1110 + }, + { + "epoch": 1.369913686806412, + "grad_norm": 2.577422739044973, + "learning_rate": 1.1724940083246042e-05, + "loss": 0.7151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1111 + }, + { + "epoch": 1.3711467324291, + "grad_norm": 2.4016601831106867, + "learning_rate": 1.171195284752907e-05, + "loss": 0.6719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1112 + }, + { + "epoch": 1.372379778051788, + "grad_norm": 2.2381623856871187, + "learning_rate": 1.1698962636425546e-05, + "loss": 0.7466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1113 + }, + { + "epoch": 1.373612823674476, + "grad_norm": 2.3077954380459498, + "learning_rate": 1.1685969472512544e-05, + "loss": 0.6609, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1114 + }, + { + "epoch": 1.3748458692971641, + "grad_norm": 2.528220870127657, + "learning_rate": 1.1672973378372283e-05, + "loss": 0.7914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1115 + }, + { + "epoch": 1.376078914919852, + "grad_norm": 2.2640778971544946, + "learning_rate": 1.1659974376592065e-05, + "loss": 0.5914, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1116 + }, + { + "epoch": 1.3773119605425401, + "grad_norm": 2.3357561361414994, + "learning_rate": 1.1646972489764254e-05, + "loss": 0.7194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1117 + }, + { + "epoch": 1.3785450061652282, + "grad_norm": 2.578566043676433, + "learning_rate": 1.163396774048622e-05, + "loss": 0.8069, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1118 + }, + { + "epoch": 1.3797780517879161, + "grad_norm": 2.6394995201547853, + "learning_rate": 1.1620960151360317e-05, + "loss": 0.797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1119 + }, + { + "epoch": 1.3810110974106042, + "grad_norm": 2.2935912265432963, + "learning_rate": 1.1607949744993829e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1120 + }, + { + "epoch": 1.3822441430332923, + "grad_norm": 2.3848813510148394, + "learning_rate": 1.1594936543998933e-05, + "loss": 0.5816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1121 + }, + { + "epoch": 1.3834771886559802, + "grad_norm": 2.266951138614489, + "learning_rate": 1.1581920570992673e-05, + "loss": 0.8167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1122 + }, + { + "epoch": 1.3847102342786684, + "grad_norm": 2.0749563395142867, + "learning_rate": 1.1568901848596905e-05, + "loss": 0.6801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1123 + }, + { + "epoch": 1.3859432799013565, + "grad_norm": 2.259958476393895, + "learning_rate": 1.1555880399438257e-05, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1124 + }, + { + "epoch": 1.3871763255240444, + "grad_norm": 2.329878469976296, + "learning_rate": 1.1542856246148109e-05, + "loss": 0.7625, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1125 + }, + { + "epoch": 1.3884093711467325, + "grad_norm": 2.626476895279836, + "learning_rate": 1.1529829411362532e-05, + "loss": 0.8145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1126 + }, + { + "epoch": 1.3896424167694206, + "grad_norm": 2.2360683032625297, + "learning_rate": 1.1516799917722262e-05, + "loss": 0.6588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1127 + }, + { + "epoch": 1.3908754623921085, + "grad_norm": 2.2606389128750664, + "learning_rate": 1.1503767787872652e-05, + "loss": 0.612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1128 + }, + { + "epoch": 1.3921085080147966, + "grad_norm": 2.3488413613095043, + "learning_rate": 1.149073304446364e-05, + "loss": 0.7111, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1129 + }, + { + "epoch": 1.3933415536374847, + "grad_norm": 2.187014482346159, + "learning_rate": 1.1477695710149706e-05, + "loss": 0.6897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1130 + }, + { + "epoch": 1.3945745992601726, + "grad_norm": 2.232874524639436, + "learning_rate": 1.1464655807589829e-05, + "loss": 0.7495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1131 + }, + { + "epoch": 1.3958076448828607, + "grad_norm": 2.2422283956051277, + "learning_rate": 1.1451613359447462e-05, + "loss": 0.7748, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1132 + }, + { + "epoch": 1.3970406905055488, + "grad_norm": 2.2170691512303327, + "learning_rate": 1.1438568388390465e-05, + "loss": 0.7026, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1133 + }, + { + "epoch": 1.3982737361282367, + "grad_norm": 2.3511388510197495, + "learning_rate": 1.1425520917091104e-05, + "loss": 0.7438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1134 + }, + { + "epoch": 1.3995067817509248, + "grad_norm": 2.302626316717242, + "learning_rate": 1.141247096822597e-05, + "loss": 0.6808, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1135 + }, + { + "epoch": 1.400739827373613, + "grad_norm": 2.358682446508914, + "learning_rate": 1.1399418564475977e-05, + "loss": 0.7462, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1136 + }, + { + "epoch": 1.4019728729963008, + "grad_norm": 2.4921870940266864, + "learning_rate": 1.138636372852629e-05, + "loss": 0.7565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1137 + }, + { + "epoch": 1.403205918618989, + "grad_norm": 2.881251766211438, + "learning_rate": 1.1373306483066315e-05, + "loss": 0.8531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1138 + }, + { + "epoch": 1.404438964241677, + "grad_norm": 2.374914568635526, + "learning_rate": 1.1360246850789638e-05, + "loss": 0.6915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1139 + }, + { + "epoch": 1.405672009864365, + "grad_norm": 2.4892197803325478, + "learning_rate": 1.1347184854394e-05, + "loss": 0.8248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1140 + }, + { + "epoch": 1.406905055487053, + "grad_norm": 2.335470361345138, + "learning_rate": 1.1334120516581239e-05, + "loss": 0.6239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1141 + }, + { + "epoch": 1.4081381011097411, + "grad_norm": 2.295703427576132, + "learning_rate": 1.1321053860057275e-05, + "loss": 0.7663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1142 + }, + { + "epoch": 1.409371146732429, + "grad_norm": 2.266519214119513, + "learning_rate": 1.1307984907532047e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1143 + }, + { + "epoch": 1.4106041923551171, + "grad_norm": 2.3854379731705917, + "learning_rate": 1.1294913681719497e-05, + "loss": 0.8726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1144 + }, + { + "epoch": 1.4118372379778052, + "grad_norm": 2.3244362810232473, + "learning_rate": 1.1281840205337504e-05, + "loss": 0.694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1145 + }, + { + "epoch": 1.4130702836004931, + "grad_norm": 2.3269738125045727, + "learning_rate": 1.1268764501107872e-05, + "loss": 0.6867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1146 + }, + { + "epoch": 1.4143033292231812, + "grad_norm": 2.1038829157154533, + "learning_rate": 1.1255686591756265e-05, + "loss": 0.4939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1147 + }, + { + "epoch": 1.4155363748458694, + "grad_norm": 2.3057935606552356, + "learning_rate": 1.124260650001219e-05, + "loss": 0.697, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1148 + }, + { + "epoch": 1.4167694204685573, + "grad_norm": 2.4457435727408448, + "learning_rate": 1.1229524248608936e-05, + "loss": 0.7129, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1149 + }, + { + "epoch": 1.4180024660912454, + "grad_norm": 2.5516324512905793, + "learning_rate": 1.1216439860283556e-05, + "loss": 0.7116, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1150 + }, + { + "epoch": 1.4192355117139335, + "grad_norm": 2.439191648697924, + "learning_rate": 1.120335335777681e-05, + "loss": 0.7269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1151 + }, + { + "epoch": 1.4204685573366214, + "grad_norm": 2.381736438404098, + "learning_rate": 1.1190264763833138e-05, + "loss": 0.7226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1152 + }, + { + "epoch": 1.4217016029593095, + "grad_norm": 2.326191969965805, + "learning_rate": 1.1177174101200609e-05, + "loss": 0.682, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1153 + }, + { + "epoch": 1.4229346485819976, + "grad_norm": 2.4443478986341964, + "learning_rate": 1.1164081392630895e-05, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1154 + }, + { + "epoch": 1.4241676942046855, + "grad_norm": 2.4191171109083274, + "learning_rate": 1.115098666087921e-05, + "loss": 0.671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1155 + }, + { + "epoch": 1.4254007398273736, + "grad_norm": 2.331429993762429, + "learning_rate": 1.1137889928704303e-05, + "loss": 0.7875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1156 + }, + { + "epoch": 1.4266337854500617, + "grad_norm": 2.6182853389449905, + "learning_rate": 1.1124791218868383e-05, + "loss": 0.7465, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1157 + }, + { + "epoch": 1.4278668310727496, + "grad_norm": 2.3276838263801474, + "learning_rate": 1.111169055413711e-05, + "loss": 0.6731, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1158 + }, + { + "epoch": 1.4290998766954377, + "grad_norm": 2.3555978375187783, + "learning_rate": 1.109858795727953e-05, + "loss": 0.7079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1159 + }, + { + "epoch": 1.4303329223181258, + "grad_norm": 2.243198455884724, + "learning_rate": 1.1085483451068056e-05, + "loss": 0.6767, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1160 + }, + { + "epoch": 1.4315659679408137, + "grad_norm": 2.426755129308782, + "learning_rate": 1.107237705827841e-05, + "loss": 0.6942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1161 + }, + { + "epoch": 1.4327990135635018, + "grad_norm": 2.310530088918712, + "learning_rate": 1.1059268801689605e-05, + "loss": 0.7414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1162 + }, + { + "epoch": 1.43403205918619, + "grad_norm": 2.3870562170497984, + "learning_rate": 1.104615870408388e-05, + "loss": 0.3904, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1163 + }, + { + "epoch": 1.4352651048088778, + "grad_norm": 2.185670591654591, + "learning_rate": 1.1033046788246684e-05, + "loss": 0.5667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1164 + }, + { + "epoch": 1.436498150431566, + "grad_norm": 2.2363567918319904, + "learning_rate": 1.1019933076966624e-05, + "loss": 0.6582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1165 + }, + { + "epoch": 1.437731196054254, + "grad_norm": 2.4989532650808917, + "learning_rate": 1.1006817593035423e-05, + "loss": 0.8567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1166 + }, + { + "epoch": 1.438964241676942, + "grad_norm": 2.3045828034357614, + "learning_rate": 1.0993700359247889e-05, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1167 + }, + { + "epoch": 1.44019728729963, + "grad_norm": 2.4408566344557263, + "learning_rate": 1.0980581398401875e-05, + "loss": 0.867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1168 + }, + { + "epoch": 1.4414303329223181, + "grad_norm": 2.433134793461258, + "learning_rate": 1.0967460733298224e-05, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1169 + }, + { + "epoch": 1.442663378545006, + "grad_norm": 2.084326240070432, + "learning_rate": 1.0954338386740746e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1170 + }, + { + "epoch": 1.4438964241676941, + "grad_norm": 2.3889216941699165, + "learning_rate": 1.0941214381536185e-05, + "loss": 0.7649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1171 + }, + { + "epoch": 1.4451294697903823, + "grad_norm": 2.3052157462538787, + "learning_rate": 1.092808874049415e-05, + "loss": 0.6548, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1172 + }, + { + "epoch": 1.4463625154130701, + "grad_norm": 2.412258190430439, + "learning_rate": 1.0914961486427103e-05, + "loss": 0.6655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1173 + }, + { + "epoch": 1.4475955610357583, + "grad_norm": 2.3820079102806755, + "learning_rate": 1.0901832642150308e-05, + "loss": 0.6374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1174 + }, + { + "epoch": 1.4488286066584464, + "grad_norm": 2.826438039521343, + "learning_rate": 1.0888702230481788e-05, + "loss": 0.6666, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1175 + }, + { + "epoch": 1.4500616522811345, + "grad_norm": 2.3005434823783166, + "learning_rate": 1.0875570274242305e-05, + "loss": 0.7333, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1176 + }, + { + "epoch": 1.4512946979038224, + "grad_norm": 2.580920616275187, + "learning_rate": 1.0862436796255285e-05, + "loss": 0.8507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1177 + }, + { + "epoch": 1.4525277435265105, + "grad_norm": 2.39619982968134, + "learning_rate": 1.0849301819346809e-05, + "loss": 0.6086, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1178 + }, + { + "epoch": 1.4537607891491986, + "grad_norm": 2.4469170067316246, + "learning_rate": 1.0836165366345572e-05, + "loss": 0.7415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1179 + }, + { + "epoch": 1.4549938347718865, + "grad_norm": 2.4895765738835327, + "learning_rate": 1.0823027460082815e-05, + "loss": 0.8269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1180 + }, + { + "epoch": 1.4562268803945746, + "grad_norm": 2.1191361440164935, + "learning_rate": 1.0809888123392321e-05, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1181 + }, + { + "epoch": 1.4574599260172627, + "grad_norm": 2.0580916232206263, + "learning_rate": 1.0796747379110351e-05, + "loss": 0.5483, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1182 + }, + { + "epoch": 1.4586929716399506, + "grad_norm": 2.358708614850794, + "learning_rate": 1.0783605250075615e-05, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1183 + }, + { + "epoch": 1.4599260172626387, + "grad_norm": 2.3706216835188068, + "learning_rate": 1.0770461759129225e-05, + "loss": 0.6925, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1184 + }, + { + "epoch": 1.4611590628853268, + "grad_norm": 2.075953282437584, + "learning_rate": 1.0757316929114673e-05, + "loss": 0.7095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1185 + }, + { + "epoch": 1.462392108508015, + "grad_norm": 2.3425821034074175, + "learning_rate": 1.0744170782877765e-05, + "loss": 0.5685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1186 + }, + { + "epoch": 1.4636251541307028, + "grad_norm": 2.364864320197578, + "learning_rate": 1.07310233432666e-05, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1187 + }, + { + "epoch": 1.464858199753391, + "grad_norm": 2.269198583092931, + "learning_rate": 1.071787463313152e-05, + "loss": 0.5853, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1188 + }, + { + "epoch": 1.466091245376079, + "grad_norm": 2.3353973695396504, + "learning_rate": 1.0704724675325086e-05, + "loss": 0.6851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1189 + }, + { + "epoch": 1.467324290998767, + "grad_norm": 2.297514940312218, + "learning_rate": 1.0691573492702014e-05, + "loss": 0.6437, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1190 + }, + { + "epoch": 1.468557336621455, + "grad_norm": 2.347920635849606, + "learning_rate": 1.0678421108119161e-05, + "loss": 0.5607, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1191 + }, + { + "epoch": 1.4697903822441432, + "grad_norm": 2.4998335942015792, + "learning_rate": 1.0665267544435464e-05, + "loss": 0.715, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1192 + }, + { + "epoch": 1.471023427866831, + "grad_norm": 2.6799025570634494, + "learning_rate": 1.065211282451192e-05, + "loss": 0.855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1193 + }, + { + "epoch": 1.4722564734895192, + "grad_norm": 2.3846645442657497, + "learning_rate": 1.0638956971211515e-05, + "loss": 0.8093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1194 + }, + { + "epoch": 1.4734895191122073, + "grad_norm": 2.289987275457904, + "learning_rate": 1.0625800007399232e-05, + "loss": 0.615, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1195 + }, + { + "epoch": 1.4747225647348952, + "grad_norm": 2.398231591907997, + "learning_rate": 1.061264195594196e-05, + "loss": 0.7564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1196 + }, + { + "epoch": 1.4759556103575833, + "grad_norm": 2.598957432876697, + "learning_rate": 1.0599482839708492e-05, + "loss": 0.5761, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1197 + }, + { + "epoch": 1.4771886559802714, + "grad_norm": 2.3008244142116308, + "learning_rate": 1.0586322681569468e-05, + "loss": 0.7096, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1198 + }, + { + "epoch": 1.4784217016029593, + "grad_norm": 2.62152754882142, + "learning_rate": 1.0573161504397345e-05, + "loss": 0.9059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1199 + }, + { + "epoch": 1.4796547472256474, + "grad_norm": 2.7027129465803323, + "learning_rate": 1.0559999331066334e-05, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1200 + }, + { + "epoch": 1.4808877928483355, + "grad_norm": 2.441005731777059, + "learning_rate": 1.0546836184452397e-05, + "loss": 0.8212, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1201 + }, + { + "epoch": 1.4821208384710234, + "grad_norm": 2.4156909019245516, + "learning_rate": 1.0533672087433176e-05, + "loss": 0.6755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1202 + }, + { + "epoch": 1.4833538840937115, + "grad_norm": 2.373408895848631, + "learning_rate": 1.0520507062887966e-05, + "loss": 0.7558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1203 + }, + { + "epoch": 1.4845869297163996, + "grad_norm": 2.1607701722680814, + "learning_rate": 1.050734113369768e-05, + "loss": 0.6858, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1204 + }, + { + "epoch": 1.4858199753390875, + "grad_norm": 2.211217022679158, + "learning_rate": 1.0494174322744797e-05, + "loss": 0.7383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1205 + }, + { + "epoch": 1.4870530209617756, + "grad_norm": 2.180028246293776, + "learning_rate": 1.0481006652913328e-05, + "loss": 0.5521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1206 + }, + { + "epoch": 1.4882860665844637, + "grad_norm": 2.393847834562169, + "learning_rate": 1.0467838147088785e-05, + "loss": 0.6334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1207 + }, + { + "epoch": 1.4895191122071516, + "grad_norm": 2.20100213605133, + "learning_rate": 1.0454668828158122e-05, + "loss": 0.7316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1208 + }, + { + "epoch": 1.4907521578298397, + "grad_norm": 2.43405475601361, + "learning_rate": 1.0441498719009715e-05, + "loss": 0.7545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1209 + }, + { + "epoch": 1.4919852034525278, + "grad_norm": 2.3073004776119004, + "learning_rate": 1.0428327842533309e-05, + "loss": 0.5749, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1210 + }, + { + "epoch": 1.4932182490752157, + "grad_norm": 2.5013121128147535, + "learning_rate": 1.041515622161998e-05, + "loss": 0.705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1211 + }, + { + "epoch": 1.4944512946979038, + "grad_norm": 2.7168877702234537, + "learning_rate": 1.0401983879162108e-05, + "loss": 0.7342, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1212 + }, + { + "epoch": 1.495684340320592, + "grad_norm": 2.35281737411902, + "learning_rate": 1.0388810838053316e-05, + "loss": 0.6203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1213 + }, + { + "epoch": 1.4969173859432798, + "grad_norm": 2.400077140272344, + "learning_rate": 1.0375637121188447e-05, + "loss": 0.6201, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1214 + }, + { + "epoch": 1.498150431565968, + "grad_norm": 2.5745039938322796, + "learning_rate": 1.0362462751463518e-05, + "loss": 0.6622, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1215 + }, + { + "epoch": 1.499383477188656, + "grad_norm": 2.4221988301394966, + "learning_rate": 1.0349287751775677e-05, + "loss": 0.7025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1216 + }, + { + "epoch": 1.500616522811344, + "grad_norm": 2.386783019524135, + "learning_rate": 1.033611214502317e-05, + "loss": 0.5671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1217 + }, + { + "epoch": 1.501849568434032, + "grad_norm": 2.539552131989461, + "learning_rate": 1.0322935954105303e-05, + "loss": 0.6469, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1218 + }, + { + "epoch": 1.5030826140567202, + "grad_norm": 2.5808036138210673, + "learning_rate": 1.0309759201922386e-05, + "loss": 0.701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1219 + }, + { + "epoch": 1.504315659679408, + "grad_norm": 2.5496582442036395, + "learning_rate": 1.0296581911375712e-05, + "loss": 0.6054, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1220 + }, + { + "epoch": 1.5055487053020962, + "grad_norm": 2.304498549067636, + "learning_rate": 1.0283404105367512e-05, + "loss": 0.6358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1221 + }, + { + "epoch": 1.5067817509247843, + "grad_norm": 2.315506316971876, + "learning_rate": 1.0270225806800904e-05, + "loss": 0.643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1222 + }, + { + "epoch": 1.5080147965474722, + "grad_norm": 2.6544020115345726, + "learning_rate": 1.0257047038579868e-05, + "loss": 0.7594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1223 + }, + { + "epoch": 1.5092478421701603, + "grad_norm": 2.1918210902533843, + "learning_rate": 1.0243867823609207e-05, + "loss": 0.6812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1224 + }, + { + "epoch": 1.5104808877928484, + "grad_norm": 2.3728046507766316, + "learning_rate": 1.0230688184794482e-05, + "loss": 0.7418, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1225 + }, + { + "epoch": 1.5117139334155363, + "grad_norm": 2.3382615280020254, + "learning_rate": 1.0217508145042007e-05, + "loss": 0.662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1226 + }, + { + "epoch": 1.5129469790382244, + "grad_norm": 2.174082577884056, + "learning_rate": 1.0204327727258787e-05, + "loss": 0.5961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1227 + }, + { + "epoch": 1.5141800246609125, + "grad_norm": 2.3119491032275596, + "learning_rate": 1.0191146954352486e-05, + "loss": 0.776, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1228 + }, + { + "epoch": 1.5154130702836004, + "grad_norm": 2.1858032851748086, + "learning_rate": 1.017796584923138e-05, + "loss": 0.5297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1229 + }, + { + "epoch": 1.5166461159062885, + "grad_norm": 2.379536502814726, + "learning_rate": 1.0164784434804327e-05, + "loss": 0.6471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1230 + }, + { + "epoch": 1.5178791615289766, + "grad_norm": 2.5211256291060655, + "learning_rate": 1.0151602733980722e-05, + "loss": 0.6299, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1231 + }, + { + "epoch": 1.5191122071516645, + "grad_norm": 2.406098151191697, + "learning_rate": 1.013842076967046e-05, + "loss": 0.6612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1232 + }, + { + "epoch": 1.5203452527743526, + "grad_norm": 2.4669460715815843, + "learning_rate": 1.0125238564783886e-05, + "loss": 0.7121, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1233 + }, + { + "epoch": 1.5215782983970407, + "grad_norm": 2.420528895345573, + "learning_rate": 1.0112056142231774e-05, + "loss": 0.5687, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1234 + }, + { + "epoch": 1.5228113440197286, + "grad_norm": 2.7051912038626074, + "learning_rate": 1.0098873524925268e-05, + "loss": 0.6909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1235 + }, + { + "epoch": 1.5240443896424167, + "grad_norm": 2.496413960615992, + "learning_rate": 1.008569073577585e-05, + "loss": 0.6021, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1236 + }, + { + "epoch": 1.5252774352651048, + "grad_norm": 2.4865070368167808, + "learning_rate": 1.007250779769531e-05, + "loss": 0.5448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1237 + }, + { + "epoch": 1.5265104808877927, + "grad_norm": 2.4574534769181513, + "learning_rate": 1.005932473359569e-05, + "loss": 0.6773, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1238 + }, + { + "epoch": 1.5277435265104808, + "grad_norm": 2.3040474377405826, + "learning_rate": 1.0046141566389247e-05, + "loss": 0.5662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1239 + }, + { + "epoch": 1.528976572133169, + "grad_norm": 2.42558040335409, + "learning_rate": 1.0032958318988426e-05, + "loss": 0.6739, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1240 + }, + { + "epoch": 1.5302096177558568, + "grad_norm": 2.3661886293872243, + "learning_rate": 1.0019775014305804e-05, + "loss": 0.6297, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1241 + }, + { + "epoch": 1.531442663378545, + "grad_norm": 2.314603674785468, + "learning_rate": 1.0006591675254063e-05, + "loss": 0.5369, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1242 + }, + { + "epoch": 1.532675709001233, + "grad_norm": 2.3999228551654412, + "learning_rate": 9.993408324745942e-06, + "loss": 0.6449, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1243 + }, + { + "epoch": 1.533908754623921, + "grad_norm": 2.3862165327435667, + "learning_rate": 9.9802249856942e-06, + "loss": 0.6161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1244 + }, + { + "epoch": 1.5351418002466093, + "grad_norm": 2.4581866876380487, + "learning_rate": 9.96704168101158e-06, + "loss": 0.6433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1245 + }, + { + "epoch": 1.5363748458692972, + "grad_norm": 2.505484522547059, + "learning_rate": 9.953858433610756e-06, + "loss": 0.6709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1246 + }, + { + "epoch": 1.537607891491985, + "grad_norm": 2.561128967653091, + "learning_rate": 9.940675266404311e-06, + "loss": 0.7824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1247 + }, + { + "epoch": 1.5388409371146734, + "grad_norm": 2.4218451757264234, + "learning_rate": 9.927492202304693e-06, + "loss": 0.7463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1248 + }, + { + "epoch": 1.5400739827373613, + "grad_norm": 2.4125121526776154, + "learning_rate": 9.91430926422415e-06, + "loss": 0.6605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1249 + }, + { + "epoch": 1.5413070283600492, + "grad_norm": 2.445505663908, + "learning_rate": 9.901126475074737e-06, + "loss": 0.626, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1250 + }, + { + "epoch": 1.5425400739827375, + "grad_norm": 2.314619197677989, + "learning_rate": 9.887943857768228e-06, + "loss": 0.6289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1251 + }, + { + "epoch": 1.5437731196054254, + "grad_norm": 2.3232344719049967, + "learning_rate": 9.874761435216119e-06, + "loss": 0.5306, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1252 + }, + { + "epoch": 1.5450061652281133, + "grad_norm": 2.4949418638683007, + "learning_rate": 9.861579230329542e-06, + "loss": 0.5698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1253 + }, + { + "epoch": 1.5462392108508016, + "grad_norm": 2.3089453128921105, + "learning_rate": 9.84839726601928e-06, + "loss": 0.5801, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1254 + }, + { + "epoch": 1.5474722564734895, + "grad_norm": 2.274576156733362, + "learning_rate": 9.835215565195676e-06, + "loss": 0.8039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1255 + }, + { + "epoch": 1.5487053020961774, + "grad_norm": 2.2604020732970818, + "learning_rate": 9.822034150768625e-06, + "loss": 0.7985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1256 + }, + { + "epoch": 1.5499383477188657, + "grad_norm": 2.1641795636087915, + "learning_rate": 9.80885304564752e-06, + "loss": 0.6705, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1257 + }, + { + "epoch": 1.5511713933415536, + "grad_norm": 2.48269267754425, + "learning_rate": 9.795672272741216e-06, + "loss": 0.6985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1258 + }, + { + "epoch": 1.5524044389642415, + "grad_norm": 2.3771051242924903, + "learning_rate": 9.782491854957998e-06, + "loss": 0.6647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1259 + }, + { + "epoch": 1.5536374845869299, + "grad_norm": 2.2133943018691684, + "learning_rate": 9.769311815205522e-06, + "loss": 0.4763, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1260 + }, + { + "epoch": 1.5548705302096177, + "grad_norm": 2.4689964457263605, + "learning_rate": 9.756132176390797e-06, + "loss": 0.667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1261 + }, + { + "epoch": 1.5561035758323056, + "grad_norm": 2.2973874789470936, + "learning_rate": 9.742952961420134e-06, + "loss": 0.6321, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1262 + }, + { + "epoch": 1.557336621454994, + "grad_norm": 2.368171008344279, + "learning_rate": 9.7297741931991e-06, + "loss": 0.7239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1263 + }, + { + "epoch": 1.5585696670776819, + "grad_norm": 2.435391702423389, + "learning_rate": 9.716595894632493e-06, + "loss": 0.5594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1264 + }, + { + "epoch": 1.55980271270037, + "grad_norm": 2.4382371041738184, + "learning_rate": 9.703418088624292e-06, + "loss": 0.7678, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1265 + }, + { + "epoch": 1.561035758323058, + "grad_norm": 2.512447464904025, + "learning_rate": 9.690240798077616e-06, + "loss": 0.7641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1266 + }, + { + "epoch": 1.562268803945746, + "grad_norm": 2.271874085642332, + "learning_rate": 9.6770640458947e-06, + "loss": 0.6891, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1267 + }, + { + "epoch": 1.563501849568434, + "grad_norm": 2.2061279299276113, + "learning_rate": 9.66388785497683e-06, + "loss": 0.4999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1268 + }, + { + "epoch": 1.5647348951911222, + "grad_norm": 2.3058075007472416, + "learning_rate": 9.650712248224327e-06, + "loss": 0.6043, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1269 + }, + { + "epoch": 1.56596794081381, + "grad_norm": 2.444257229950801, + "learning_rate": 9.637537248536485e-06, + "loss": 0.6876, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1270 + }, + { + "epoch": 1.5672009864364982, + "grad_norm": 2.4130708431685197, + "learning_rate": 9.624362878811557e-06, + "loss": 0.5949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1271 + }, + { + "epoch": 1.5684340320591863, + "grad_norm": 2.340625006242772, + "learning_rate": 9.611189161946688e-06, + "loss": 0.6838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1272 + }, + { + "epoch": 1.5696670776818742, + "grad_norm": 2.2920633133032116, + "learning_rate": 9.598016120837894e-06, + "loss": 0.6186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1273 + }, + { + "epoch": 1.5709001233045623, + "grad_norm": 2.425912926280816, + "learning_rate": 9.584843778380022e-06, + "loss": 0.5922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1274 + }, + { + "epoch": 1.5721331689272504, + "grad_norm": 2.332147214152514, + "learning_rate": 9.571672157466695e-06, + "loss": 0.6555, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1275 + }, + { + "epoch": 1.5733662145499383, + "grad_norm": 2.398257339947792, + "learning_rate": 9.55850128099029e-06, + "loss": 0.6346, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1276 + }, + { + "epoch": 1.5745992601726264, + "grad_norm": 2.142956121629841, + "learning_rate": 9.545331171841882e-06, + "loss": 0.5515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1277 + }, + { + "epoch": 1.5758323057953145, + "grad_norm": 2.407908540970335, + "learning_rate": 9.53216185291122e-06, + "loss": 0.5475, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1278 + }, + { + "epoch": 1.5770653514180024, + "grad_norm": 2.320052780540722, + "learning_rate": 9.518993347086676e-06, + "loss": 0.6044, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1279 + }, + { + "epoch": 1.5782983970406905, + "grad_norm": 2.4974178144916643, + "learning_rate": 9.505825677255205e-06, + "loss": 0.6436, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1280 + }, + { + "epoch": 1.5795314426633786, + "grad_norm": 2.543298528231235, + "learning_rate": 9.492658866302325e-06, + "loss": 0.6388, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1281 + }, + { + "epoch": 1.5807644882860665, + "grad_norm": 2.3663335712794917, + "learning_rate": 9.479492937112034e-06, + "loss": 0.5769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1282 + }, + { + "epoch": 1.5819975339087546, + "grad_norm": 2.301999998958795, + "learning_rate": 9.46632791256683e-06, + "loss": 0.6323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1283 + }, + { + "epoch": 1.5832305795314427, + "grad_norm": 2.3062144791204386, + "learning_rate": 9.453163815547604e-06, + "loss": 0.6727, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1284 + }, + { + "epoch": 1.5844636251541306, + "grad_norm": 2.5906430609145947, + "learning_rate": 9.440000668933668e-06, + "loss": 0.7282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1285 + }, + { + "epoch": 1.5856966707768188, + "grad_norm": 2.6113606139990586, + "learning_rate": 9.426838495602659e-06, + "loss": 0.685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1286 + }, + { + "epoch": 1.5869297163995069, + "grad_norm": 2.1825712632265244, + "learning_rate": 9.413677318430532e-06, + "loss": 0.5556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1287 + }, + { + "epoch": 1.5881627620221948, + "grad_norm": 2.394162195116171, + "learning_rate": 9.400517160291511e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1288 + }, + { + "epoch": 1.5893958076448829, + "grad_norm": 2.375297383969579, + "learning_rate": 9.387358044058044e-06, + "loss": 0.7398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1289 + }, + { + "epoch": 1.590628853267571, + "grad_norm": 2.268014800016658, + "learning_rate": 9.374199992600775e-06, + "loss": 0.554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1290 + }, + { + "epoch": 1.5918618988902589, + "grad_norm": 2.273418089045509, + "learning_rate": 9.361043028788487e-06, + "loss": 0.5092, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1291 + }, + { + "epoch": 1.593094944512947, + "grad_norm": 2.258287636448223, + "learning_rate": 9.347887175488086e-06, + "loss": 0.5517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1292 + }, + { + "epoch": 1.594327990135635, + "grad_norm": 2.4638493212600756, + "learning_rate": 9.334732455564538e-06, + "loss": 0.6536, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1293 + }, + { + "epoch": 1.595561035758323, + "grad_norm": 2.45854766908022, + "learning_rate": 9.321578891880838e-06, + "loss": 0.5922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1294 + }, + { + "epoch": 1.596794081381011, + "grad_norm": 2.6177980977782522, + "learning_rate": 9.30842650729799e-06, + "loss": 0.7473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1295 + }, + { + "epoch": 1.5980271270036992, + "grad_norm": 2.43914078165767, + "learning_rate": 9.295275324674915e-06, + "loss": 0.5646, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1296 + }, + { + "epoch": 1.599260172626387, + "grad_norm": 2.365874131549762, + "learning_rate": 9.282125366868484e-06, + "loss": 0.5354, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1297 + }, + { + "epoch": 1.6004932182490752, + "grad_norm": 2.528721468130841, + "learning_rate": 9.268976656733402e-06, + "loss": 0.5554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1298 + }, + { + "epoch": 1.6017262638717633, + "grad_norm": 2.512092936139005, + "learning_rate": 9.255829217122237e-06, + "loss": 0.533, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1299 + }, + { + "epoch": 1.6029593094944512, + "grad_norm": 2.354466866391755, + "learning_rate": 9.242683070885328e-06, + "loss": 0.4921, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1300 + }, + { + "epoch": 1.6041923551171393, + "grad_norm": 2.4883337985127936, + "learning_rate": 9.229538240870774e-06, + "loss": 0.6147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1301 + }, + { + "epoch": 1.6054254007398274, + "grad_norm": 2.4285611973298216, + "learning_rate": 9.21639474992439e-06, + "loss": 0.5665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1302 + }, + { + "epoch": 1.6066584463625153, + "grad_norm": 2.5959012805664994, + "learning_rate": 9.203252620889654e-06, + "loss": 0.6071, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1303 + }, + { + "epoch": 1.6078914919852034, + "grad_norm": 2.356030881254319, + "learning_rate": 9.190111876607684e-06, + "loss": 0.5597, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1304 + }, + { + "epoch": 1.6091245376078915, + "grad_norm": 2.4328199046350503, + "learning_rate": 9.176972539917188e-06, + "loss": 0.5543, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1305 + }, + { + "epoch": 1.6103575832305794, + "grad_norm": 2.531198496944925, + "learning_rate": 9.163834633654431e-06, + "loss": 0.7257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1306 + }, + { + "epoch": 1.6115906288532675, + "grad_norm": 2.4402117071794245, + "learning_rate": 9.150698180653193e-06, + "loss": 0.6173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1307 + }, + { + "epoch": 1.6128236744759556, + "grad_norm": 2.8953638685717547, + "learning_rate": 9.13756320374472e-06, + "loss": 0.7887, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1308 + }, + { + "epoch": 1.6140567200986435, + "grad_norm": 2.459699675606036, + "learning_rate": 9.1244297257577e-06, + "loss": 0.7027, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1309 + }, + { + "epoch": 1.6152897657213316, + "grad_norm": 2.3328161668697467, + "learning_rate": 9.111297769518213e-06, + "loss": 0.5885, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1310 + }, + { + "epoch": 1.6165228113440198, + "grad_norm": 2.4218060271919812, + "learning_rate": 9.098167357849699e-06, + "loss": 0.6497, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1311 + }, + { + "epoch": 1.6177558569667077, + "grad_norm": 2.294366317458183, + "learning_rate": 9.0850385135729e-06, + "loss": 0.5431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1312 + }, + { + "epoch": 1.6189889025893958, + "grad_norm": 2.3225490101301287, + "learning_rate": 9.071911259505852e-06, + "loss": 0.6644, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1313 + }, + { + "epoch": 1.6202219482120839, + "grad_norm": 2.5880544489532826, + "learning_rate": 9.05878561846382e-06, + "loss": 0.517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1314 + }, + { + "epoch": 1.6214549938347718, + "grad_norm": 2.230629935973886, + "learning_rate": 9.045661613259254e-06, + "loss": 0.5261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1315 + }, + { + "epoch": 1.6226880394574599, + "grad_norm": 2.1770639270798307, + "learning_rate": 9.032539266701781e-06, + "loss": 0.4244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1316 + }, + { + "epoch": 1.623921085080148, + "grad_norm": 2.709845393499603, + "learning_rate": 9.019418601598129e-06, + "loss": 0.637, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1317 + }, + { + "epoch": 1.6251541307028359, + "grad_norm": 3.460293995200764, + "learning_rate": 9.006299640752113e-06, + "loss": 0.6512, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1318 + }, + { + "epoch": 1.6263871763255242, + "grad_norm": 3.5764897422793744, + "learning_rate": 8.993182406964579e-06, + "loss": 0.6457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1319 + }, + { + "epoch": 1.627620221948212, + "grad_norm": 2.5994177808868657, + "learning_rate": 8.980066923033378e-06, + "loss": 0.574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1320 + }, + { + "epoch": 1.6288532675709, + "grad_norm": 2.5757673207857135, + "learning_rate": 8.96695321175332e-06, + "loss": 0.4942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1321 + }, + { + "epoch": 1.6300863131935883, + "grad_norm": 2.453402170110872, + "learning_rate": 8.953841295916123e-06, + "loss": 0.6917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1322 + }, + { + "epoch": 1.6313193588162762, + "grad_norm": 2.654143412873541, + "learning_rate": 8.940731198310402e-06, + "loss": 0.7042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1323 + }, + { + "epoch": 1.632552404438964, + "grad_norm": 2.4250402090614935, + "learning_rate": 8.927622941721594e-06, + "loss": 0.6999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1324 + }, + { + "epoch": 1.6337854500616524, + "grad_norm": 2.761651051842012, + "learning_rate": 8.91451654893195e-06, + "loss": 0.6075, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1325 + }, + { + "epoch": 1.6350184956843403, + "grad_norm": 2.6957159420077894, + "learning_rate": 8.901412042720475e-06, + "loss": 0.6084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1326 + }, + { + "epoch": 1.6362515413070282, + "grad_norm": 2.2293950670646745, + "learning_rate": 8.888309445862892e-06, + "loss": 0.7359, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1327 + }, + { + "epoch": 1.6374845869297165, + "grad_norm": 2.3226832325484663, + "learning_rate": 8.87520878113162e-06, + "loss": 0.5255, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1328 + }, + { + "epoch": 1.6387176325524044, + "grad_norm": 2.5216170149103583, + "learning_rate": 8.8621100712957e-06, + "loss": 0.6287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1329 + }, + { + "epoch": 1.6399506781750923, + "grad_norm": 2.685061433529547, + "learning_rate": 8.849013339120793e-06, + "loss": 0.7093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1330 + }, + { + "epoch": 1.6411837237977807, + "grad_norm": 2.5648896854812975, + "learning_rate": 8.835918607369109e-06, + "loss": 0.6182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1331 + }, + { + "epoch": 1.6424167694204685, + "grad_norm": 2.6253257359858404, + "learning_rate": 8.822825898799393e-06, + "loss": 0.7446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1332 + }, + { + "epoch": 1.6436498150431564, + "grad_norm": 2.274148046399076, + "learning_rate": 8.809735236166864e-06, + "loss": 0.6534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1333 + }, + { + "epoch": 1.6448828606658448, + "grad_norm": 2.3764316480496537, + "learning_rate": 8.796646642223192e-06, + "loss": 0.6847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1334 + }, + { + "epoch": 1.6461159062885327, + "grad_norm": 2.3471176710530024, + "learning_rate": 8.783560139716449e-06, + "loss": 0.5461, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1335 + }, + { + "epoch": 1.6473489519112205, + "grad_norm": 2.4255652189713715, + "learning_rate": 8.770475751391066e-06, + "loss": 0.4423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1336 + }, + { + "epoch": 1.6485819975339089, + "grad_norm": 2.5292711557625673, + "learning_rate": 8.757393499987815e-06, + "loss": 0.7064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1337 + }, + { + "epoch": 1.6498150431565968, + "grad_norm": 2.427296462933323, + "learning_rate": 8.744313408243737e-06, + "loss": 0.6039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1338 + }, + { + "epoch": 1.6510480887792849, + "grad_norm": 2.5975517369534105, + "learning_rate": 8.731235498892127e-06, + "loss": 0.6391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1339 + }, + { + "epoch": 1.652281134401973, + "grad_norm": 2.491354995578972, + "learning_rate": 8.718159794662498e-06, + "loss": 0.6087, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1340 + }, + { + "epoch": 1.6535141800246609, + "grad_norm": 2.4884148521815526, + "learning_rate": 8.705086318280505e-06, + "loss": 0.5634, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1341 + }, + { + "epoch": 1.654747225647349, + "grad_norm": 2.8396628177391907, + "learning_rate": 8.692015092467956e-06, + "loss": 0.7335, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1342 + }, + { + "epoch": 1.655980271270037, + "grad_norm": 2.7368157959861557, + "learning_rate": 8.678946139942728e-06, + "loss": 0.7007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1343 + }, + { + "epoch": 1.657213316892725, + "grad_norm": 2.3611404478863873, + "learning_rate": 8.665879483418766e-06, + "loss": 0.5668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1344 + }, + { + "epoch": 1.658446362515413, + "grad_norm": 2.191668075058394, + "learning_rate": 8.652815145606003e-06, + "loss": 0.5518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1345 + }, + { + "epoch": 1.6596794081381012, + "grad_norm": 2.14788648268729, + "learning_rate": 8.63975314921036e-06, + "loss": 0.5166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1346 + }, + { + "epoch": 1.660912453760789, + "grad_norm": 2.3068777176506963, + "learning_rate": 8.626693516933687e-06, + "loss": 0.5394, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1347 + }, + { + "epoch": 1.6621454993834772, + "grad_norm": 2.49955124289679, + "learning_rate": 8.613636271473712e-06, + "loss": 0.6493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1348 + }, + { + "epoch": 1.6633785450061653, + "grad_norm": 2.1471304936125652, + "learning_rate": 8.60058143552403e-06, + "loss": 0.5408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1349 + }, + { + "epoch": 1.6646115906288532, + "grad_norm": 2.3378284654154444, + "learning_rate": 8.587529031774033e-06, + "loss": 0.5559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1350 + }, + { + "epoch": 1.6658446362515413, + "grad_norm": 2.272350354054125, + "learning_rate": 8.574479082908903e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1351 + }, + { + "epoch": 1.6670776818742294, + "grad_norm": 2.6718347888114393, + "learning_rate": 8.561431611609537e-06, + "loss": 0.6518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1352 + }, + { + "epoch": 1.6683107274969173, + "grad_norm": 2.5683646046012307, + "learning_rate": 8.548386640552542e-06, + "loss": 0.627, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1353 + }, + { + "epoch": 1.6695437731196054, + "grad_norm": 2.429950074454179, + "learning_rate": 8.535344192410173e-06, + "loss": 0.6244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1354 + }, + { + "epoch": 1.6707768187422936, + "grad_norm": 2.459482147364562, + "learning_rate": 8.522304289850298e-06, + "loss": 0.5089, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1355 + }, + { + "epoch": 1.6720098643649814, + "grad_norm": 2.5078184456517802, + "learning_rate": 8.509266955536364e-06, + "loss": 0.5966, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1356 + }, + { + "epoch": 1.6732429099876696, + "grad_norm": 2.8738369967418786, + "learning_rate": 8.49623221212735e-06, + "loss": 0.7568, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1357 + }, + { + "epoch": 1.6744759556103577, + "grad_norm": 2.566639808326863, + "learning_rate": 8.483200082277743e-06, + "loss": 0.6565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1358 + }, + { + "epoch": 1.6757090012330456, + "grad_norm": 2.18616861232984, + "learning_rate": 8.47017058863747e-06, + "loss": 0.5219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1359 + }, + { + "epoch": 1.6769420468557337, + "grad_norm": 2.2874183385113387, + "learning_rate": 8.457143753851893e-06, + "loss": 0.515, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1360 + }, + { + "epoch": 1.6781750924784218, + "grad_norm": 2.3279389147190503, + "learning_rate": 8.444119600561746e-06, + "loss": 0.5737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1361 + }, + { + "epoch": 1.6794081381011097, + "grad_norm": 2.510085283064121, + "learning_rate": 8.431098151403098e-06, + "loss": 0.5527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1362 + }, + { + "epoch": 1.6806411837237978, + "grad_norm": 2.350766304372499, + "learning_rate": 8.41807942900733e-06, + "loss": 0.5444, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1363 + }, + { + "epoch": 1.681874229346486, + "grad_norm": 2.4637839990972266, + "learning_rate": 8.405063456001069e-06, + "loss": 0.617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1364 + }, + { + "epoch": 1.6831072749691738, + "grad_norm": 2.3520486778455303, + "learning_rate": 8.392050255006172e-06, + "loss": 0.5787, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1365 + }, + { + "epoch": 1.684340320591862, + "grad_norm": 2.6951378359823064, + "learning_rate": 8.379039848639685e-06, + "loss": 0.5679, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1366 + }, + { + "epoch": 1.68557336621455, + "grad_norm": 2.5680249855790533, + "learning_rate": 8.366032259513782e-06, + "loss": 0.5366, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1367 + }, + { + "epoch": 1.686806411837238, + "grad_norm": 2.7844067151133496, + "learning_rate": 8.35302751023575e-06, + "loss": 0.6181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1368 + }, + { + "epoch": 1.688039457459926, + "grad_norm": 2.3271421104987895, + "learning_rate": 8.340025623407938e-06, + "loss": 0.4529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1369 + }, + { + "epoch": 1.6892725030826141, + "grad_norm": 2.7324858411976822, + "learning_rate": 8.327026621627722e-06, + "loss": 0.7326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1370 + }, + { + "epoch": 1.690505548705302, + "grad_norm": 2.5720836300974383, + "learning_rate": 8.31403052748746e-06, + "loss": 0.6784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1371 + }, + { + "epoch": 1.6917385943279901, + "grad_norm": 2.440765245099682, + "learning_rate": 8.301037363574454e-06, + "loss": 0.6014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1372 + }, + { + "epoch": 1.6929716399506782, + "grad_norm": 2.594069817186588, + "learning_rate": 8.288047152470931e-06, + "loss": 0.5856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1373 + }, + { + "epoch": 1.6942046855733661, + "grad_norm": 2.4701026222170452, + "learning_rate": 8.27505991675396e-06, + "loss": 0.7036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1374 + }, + { + "epoch": 1.6954377311960542, + "grad_norm": 2.3893004984314365, + "learning_rate": 8.26207567899546e-06, + "loss": 0.5404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1375 + }, + { + "epoch": 1.6966707768187423, + "grad_norm": 2.3404547915407656, + "learning_rate": 8.249094461762124e-06, + "loss": 0.5142, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1376 + }, + { + "epoch": 1.6979038224414302, + "grad_norm": 2.4832770686344556, + "learning_rate": 8.236116287615409e-06, + "loss": 0.6183, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1377 + }, + { + "epoch": 1.6991368680641183, + "grad_norm": 2.262237110861396, + "learning_rate": 8.223141179111471e-06, + "loss": 0.4504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1378 + }, + { + "epoch": 1.7003699136868065, + "grad_norm": 2.62998506842995, + "learning_rate": 8.210169158801146e-06, + "loss": 0.6316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1379 + }, + { + "epoch": 1.7016029593094943, + "grad_norm": 2.2853273497892164, + "learning_rate": 8.197200249229902e-06, + "loss": 0.5379, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1380 + }, + { + "epoch": 1.7028360049321825, + "grad_norm": 2.489883484033712, + "learning_rate": 8.184234472937795e-06, + "loss": 0.5635, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1381 + }, + { + "epoch": 1.7040690505548706, + "grad_norm": 2.4399096703479723, + "learning_rate": 8.171271852459442e-06, + "loss": 0.5672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1382 + }, + { + "epoch": 1.7053020961775585, + "grad_norm": 2.5877064849336673, + "learning_rate": 8.158312410323968e-06, + "loss": 0.56, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1383 + }, + { + "epoch": 1.7065351418002466, + "grad_norm": 2.5080025813914486, + "learning_rate": 8.145356169054982e-06, + "loss": 0.6415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1384 + }, + { + "epoch": 1.7077681874229347, + "grad_norm": 2.474550610636856, + "learning_rate": 8.132403151170524e-06, + "loss": 0.5209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1385 + }, + { + "epoch": 1.7090012330456226, + "grad_norm": 2.4546793222588037, + "learning_rate": 8.11945337918303e-06, + "loss": 0.6133, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1386 + }, + { + "epoch": 1.7102342786683107, + "grad_norm": 2.3960818942570623, + "learning_rate": 8.106506875599312e-06, + "loss": 0.5948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1387 + }, + { + "epoch": 1.7114673242909988, + "grad_norm": 2.323552505571765, + "learning_rate": 8.09356366292047e-06, + "loss": 0.5402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1388 + }, + { + "epoch": 1.7127003699136867, + "grad_norm": 2.088694863039824, + "learning_rate": 8.080623763641916e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1389 + }, + { + "epoch": 1.7139334155363748, + "grad_norm": 2.171047701838099, + "learning_rate": 8.067687200253281e-06, + "loss": 0.4947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1390 + }, + { + "epoch": 1.715166461159063, + "grad_norm": 2.4534573577067165, + "learning_rate": 8.054753995238415e-06, + "loss": 0.6494, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1391 + }, + { + "epoch": 1.7163995067817508, + "grad_norm": 2.315613179007512, + "learning_rate": 8.04182417107531e-06, + "loss": 0.605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1392 + }, + { + "epoch": 1.7176325524044391, + "grad_norm": 2.382282194715041, + "learning_rate": 8.028897750236106e-06, + "loss": 0.3862, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1393 + }, + { + "epoch": 1.718865598027127, + "grad_norm": 2.3689104564582606, + "learning_rate": 8.015974755187015e-06, + "loss": 0.4825, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1394 + }, + { + "epoch": 1.720098643649815, + "grad_norm": 2.5004960798500098, + "learning_rate": 8.003055208388292e-06, + "loss": 0.5614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1395 + }, + { + "epoch": 1.7213316892725032, + "grad_norm": 2.376364766941311, + "learning_rate": 7.990139132294212e-06, + "loss": 0.4446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1396 + }, + { + "epoch": 1.7225647348951911, + "grad_norm": 2.6246902766612745, + "learning_rate": 7.977226549353e-06, + "loss": 0.6399, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1397 + }, + { + "epoch": 1.723797780517879, + "grad_norm": 2.484606542490508, + "learning_rate": 7.964317482006824e-06, + "loss": 0.5081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1398 + }, + { + "epoch": 1.7250308261405674, + "grad_norm": 3.38389227109005, + "learning_rate": 7.951411952691739e-06, + "loss": 0.755, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1399 + }, + { + "epoch": 1.7262638717632552, + "grad_norm": 2.578597094452948, + "learning_rate": 7.938509983837647e-06, + "loss": 0.4826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1400 + }, + { + "epoch": 1.7274969173859431, + "grad_norm": 2.4802210829330105, + "learning_rate": 7.925611597868266e-06, + "loss": 0.493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1401 + }, + { + "epoch": 1.7287299630086315, + "grad_norm": 2.4523105300960517, + "learning_rate": 7.91271681720108e-06, + "loss": 0.6206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1402 + }, + { + "epoch": 1.7299630086313194, + "grad_norm": 2.386348175383016, + "learning_rate": 7.899825664247322e-06, + "loss": 0.5701, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1403 + }, + { + "epoch": 1.7311960542540072, + "grad_norm": 2.4083074706971375, + "learning_rate": 7.886938161411897e-06, + "loss": 0.5884, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1404 + }, + { + "epoch": 1.7324290998766956, + "grad_norm": 2.5069557743624316, + "learning_rate": 7.874054331093389e-06, + "loss": 0.4971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1405 + }, + { + "epoch": 1.7336621454993835, + "grad_norm": 2.55856356518894, + "learning_rate": 7.861174195683987e-06, + "loss": 0.5018, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1406 + }, + { + "epoch": 1.7348951911220714, + "grad_norm": 2.568692520564215, + "learning_rate": 7.848297777569458e-06, + "loss": 0.6251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1407 + }, + { + "epoch": 1.7361282367447597, + "grad_norm": 2.626547699354354, + "learning_rate": 7.835425099129116e-06, + "loss": 0.4866, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1408 + }, + { + "epoch": 1.7373612823674476, + "grad_norm": 2.655578554031113, + "learning_rate": 7.822556182735762e-06, + "loss": 0.6262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1409 + }, + { + "epoch": 1.7385943279901355, + "grad_norm": 2.514238469731576, + "learning_rate": 7.809691050755676e-06, + "loss": 0.529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1410 + }, + { + "epoch": 1.7398273736128238, + "grad_norm": 2.486688645388684, + "learning_rate": 7.796829725548543e-06, + "loss": 0.5598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1411 + }, + { + "epoch": 1.7410604192355117, + "grad_norm": 2.5300619846568755, + "learning_rate": 7.783972229467443e-06, + "loss": 0.6431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1412 + }, + { + "epoch": 1.7422934648581998, + "grad_norm": 2.782005331836491, + "learning_rate": 7.771118584858801e-06, + "loss": 0.7072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1413 + }, + { + "epoch": 1.743526510480888, + "grad_norm": 2.4671627637221296, + "learning_rate": 7.758268814062343e-06, + "loss": 0.6404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1414 + }, + { + "epoch": 1.7447595561035758, + "grad_norm": 2.1612453871779285, + "learning_rate": 7.745422939411064e-06, + "loss": 0.4505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1415 + }, + { + "epoch": 1.745992601726264, + "grad_norm": 2.296774520080563, + "learning_rate": 7.73258098323119e-06, + "loss": 0.5345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1416 + }, + { + "epoch": 1.747225647348952, + "grad_norm": 2.6549737125780326, + "learning_rate": 7.719742967842133e-06, + "loss": 0.6232, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1417 + }, + { + "epoch": 1.74845869297164, + "grad_norm": 2.343040089887724, + "learning_rate": 7.706908915556457e-06, + "loss": 0.4859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1418 + }, + { + "epoch": 1.749691738594328, + "grad_norm": 2.459027338987637, + "learning_rate": 7.694078848679836e-06, + "loss": 0.6424, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1419 + }, + { + "epoch": 1.7509247842170161, + "grad_norm": 2.6006881140576383, + "learning_rate": 7.681252789511029e-06, + "loss": 0.6836, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1420 + }, + { + "epoch": 1.752157829839704, + "grad_norm": 2.813687773910776, + "learning_rate": 7.668430760341813e-06, + "loss": 0.6596, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1421 + }, + { + "epoch": 1.7533908754623921, + "grad_norm": 2.7501988552768166, + "learning_rate": 7.655612783456973e-06, + "loss": 0.5757, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1422 + }, + { + "epoch": 1.7546239210850803, + "grad_norm": 2.602564686174615, + "learning_rate": 7.64279888113424e-06, + "loss": 0.6304, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1423 + }, + { + "epoch": 1.7558569667077681, + "grad_norm": 2.2125559792606606, + "learning_rate": 7.629989075644277e-06, + "loss": 0.4345, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1424 + }, + { + "epoch": 1.7570900123304563, + "grad_norm": 2.552624039521638, + "learning_rate": 7.6171833892506145e-06, + "loss": 0.6566, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1425 + }, + { + "epoch": 1.7583230579531444, + "grad_norm": 2.53122973187578, + "learning_rate": 7.60438184420963e-06, + "loss": 0.5989, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1426 + }, + { + "epoch": 1.7595561035758323, + "grad_norm": 2.4914845101760834, + "learning_rate": 7.591584462770506e-06, + "loss": 0.5674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1427 + }, + { + "epoch": 1.7607891491985204, + "grad_norm": 2.71358370673903, + "learning_rate": 7.57879126717518e-06, + "loss": 0.6321, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1428 + }, + { + "epoch": 1.7620221948212085, + "grad_norm": 2.5302056277845413, + "learning_rate": 7.566002279658322e-06, + "loss": 0.6473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1429 + }, + { + "epoch": 1.7632552404438964, + "grad_norm": 2.371803209433193, + "learning_rate": 7.553217522447285e-06, + "loss": 0.4999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1430 + }, + { + "epoch": 1.7644882860665845, + "grad_norm": 2.527610014870494, + "learning_rate": 7.540437017762068e-06, + "loss": 0.5314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1431 + }, + { + "epoch": 1.7657213316892726, + "grad_norm": 2.440388163849852, + "learning_rate": 7.52766078781529e-06, + "loss": 0.6007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1432 + }, + { + "epoch": 1.7669543773119605, + "grad_norm": 2.369028343720501, + "learning_rate": 7.51488885481212e-06, + "loss": 0.5996, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1433 + }, + { + "epoch": 1.7681874229346486, + "grad_norm": 2.259203085653468, + "learning_rate": 7.502121240950284e-06, + "loss": 0.5022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1434 + }, + { + "epoch": 1.7694204685573367, + "grad_norm": 2.346742207074351, + "learning_rate": 7.489357968419975e-06, + "loss": 0.5232, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1435 + }, + { + "epoch": 1.7706535141800246, + "grad_norm": 2.3571911434438926, + "learning_rate": 7.476599059403866e-06, + "loss": 0.5529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1436 + }, + { + "epoch": 1.7718865598027127, + "grad_norm": 2.564577423619225, + "learning_rate": 7.463844536077021e-06, + "loss": 0.6384, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1437 + }, + { + "epoch": 1.7731196054254008, + "grad_norm": 2.3757451812868062, + "learning_rate": 7.451094420606906e-06, + "loss": 0.5998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1438 + }, + { + "epoch": 1.7743526510480887, + "grad_norm": 2.3897936394773542, + "learning_rate": 7.438348735153311e-06, + "loss": 0.5233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1439 + }, + { + "epoch": 1.7755856966707768, + "grad_norm": 2.384296664394737, + "learning_rate": 7.425607501868328e-06, + "loss": 0.4798, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1440 + }, + { + "epoch": 1.776818742293465, + "grad_norm": 2.326128061878694, + "learning_rate": 7.41287074289632e-06, + "loss": 0.5628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1441 + }, + { + "epoch": 1.7780517879161528, + "grad_norm": 2.3448649893975646, + "learning_rate": 7.4001384803738595e-06, + "loss": 0.5539, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1442 + }, + { + "epoch": 1.779284833538841, + "grad_norm": 2.335303557427351, + "learning_rate": 7.387410736429717e-06, + "loss": 0.463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1443 + }, + { + "epoch": 1.780517879161529, + "grad_norm": 2.5062756692562815, + "learning_rate": 7.374687533184801e-06, + "loss": 0.5804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1444 + }, + { + "epoch": 1.781750924784217, + "grad_norm": 2.667998142301554, + "learning_rate": 7.361968892752134e-06, + "loss": 0.5044, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1445 + }, + { + "epoch": 1.782983970406905, + "grad_norm": 2.8938425700505856, + "learning_rate": 7.349254837236808e-06, + "loss": 0.7082, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1446 + }, + { + "epoch": 1.7842170160295932, + "grad_norm": 2.638110967591072, + "learning_rate": 7.336545388735936e-06, + "loss": 0.5165, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1447 + }, + { + "epoch": 1.785450061652281, + "grad_norm": 2.5434717078771314, + "learning_rate": 7.323840569338645e-06, + "loss": 0.5274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1448 + }, + { + "epoch": 1.7866831072749692, + "grad_norm": 2.880952060159319, + "learning_rate": 7.31114040112599e-06, + "loss": 0.6319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1449 + }, + { + "epoch": 1.7879161528976573, + "grad_norm": 2.7232472895100974, + "learning_rate": 7.29844490617097e-06, + "loss": 0.6471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1450 + }, + { + "epoch": 1.7891491985203452, + "grad_norm": 2.6289292491069234, + "learning_rate": 7.285754106538435e-06, + "loss": 0.6144, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1451 + }, + { + "epoch": 1.7903822441430333, + "grad_norm": 2.3571805040637117, + "learning_rate": 7.273068024285096e-06, + "loss": 0.5023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1452 + }, + { + "epoch": 1.7916152897657214, + "grad_norm": 2.5462291798096115, + "learning_rate": 7.260386681459457e-06, + "loss": 0.594, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1453 + }, + { + "epoch": 1.7928483353884093, + "grad_norm": 2.364391159514722, + "learning_rate": 7.247710100101778e-06, + "loss": 0.3867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1454 + }, + { + "epoch": 1.7940813810110974, + "grad_norm": 2.557120216564179, + "learning_rate": 7.235038302244058e-06, + "loss": 0.4541, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1455 + }, + { + "epoch": 1.7953144266337855, + "grad_norm": 2.3515150176521398, + "learning_rate": 7.2223713099099655e-06, + "loss": 0.5005, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1456 + }, + { + "epoch": 1.7965474722564734, + "grad_norm": 3.0149925007019465, + "learning_rate": 7.209709145114834e-06, + "loss": 0.6562, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1457 + }, + { + "epoch": 1.7977805178791615, + "grad_norm": 2.9385786617892484, + "learning_rate": 7.197051829865591e-06, + "loss": 0.7172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1458 + }, + { + "epoch": 1.7990135635018496, + "grad_norm": 2.4438200151823577, + "learning_rate": 7.1843993861607485e-06, + "loss": 0.545, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1459 + }, + { + "epoch": 1.8002466091245375, + "grad_norm": 2.189016598422486, + "learning_rate": 7.171751835990346e-06, + "loss": 0.3945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1460 + }, + { + "epoch": 1.8014796547472256, + "grad_norm": 2.599380048646185, + "learning_rate": 7.159109201335914e-06, + "loss": 0.6221, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1461 + }, + { + "epoch": 1.8027127003699137, + "grad_norm": 2.3218284667005045, + "learning_rate": 7.146471504170448e-06, + "loss": 0.4081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1462 + }, + { + "epoch": 1.8039457459926016, + "grad_norm": 2.483257454081904, + "learning_rate": 7.1338387664583545e-06, + "loss": 0.5952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1463 + }, + { + "epoch": 1.8051787916152897, + "grad_norm": 2.3060282168534267, + "learning_rate": 7.1212110101554225e-06, + "loss": 0.478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1464 + }, + { + "epoch": 1.8064118372379778, + "grad_norm": 2.353277914016821, + "learning_rate": 7.108588257208793e-06, + "loss": 0.4602, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1465 + }, + { + "epoch": 1.8076448828606657, + "grad_norm": 2.402021352720206, + "learning_rate": 7.095970529556895e-06, + "loss": 0.423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1466 + }, + { + "epoch": 1.808877928483354, + "grad_norm": 2.4089157744395187, + "learning_rate": 7.083357849129437e-06, + "loss": 0.4962, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1467 + }, + { + "epoch": 1.810110974106042, + "grad_norm": 2.4445181455084803, + "learning_rate": 7.070750237847344e-06, + "loss": 0.4237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1468 + }, + { + "epoch": 1.8113440197287298, + "grad_norm": 2.5896452836545576, + "learning_rate": 7.058147717622742e-06, + "loss": 0.6034, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1469 + }, + { + "epoch": 1.8125770653514182, + "grad_norm": 2.5692451627366215, + "learning_rate": 7.0455503103588975e-06, + "loss": 0.4897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1470 + }, + { + "epoch": 1.813810110974106, + "grad_norm": 2.5193270914325185, + "learning_rate": 7.0329580379502e-06, + "loss": 0.5908, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1471 + }, + { + "epoch": 1.815043156596794, + "grad_norm": 2.6722250129620586, + "learning_rate": 7.020370922282113e-06, + "loss": 0.6295, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1472 + }, + { + "epoch": 1.8162762022194823, + "grad_norm": 2.4547108738067838, + "learning_rate": 7.00778898523113e-06, + "loss": 0.5357, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1473 + }, + { + "epoch": 1.8175092478421702, + "grad_norm": 2.437273516415158, + "learning_rate": 6.995212248664755e-06, + "loss": 0.5101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1474 + }, + { + "epoch": 1.818742293464858, + "grad_norm": 2.279547074068301, + "learning_rate": 6.982640734441443e-06, + "loss": 0.5079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1475 + }, + { + "epoch": 1.8199753390875464, + "grad_norm": 2.435539711925499, + "learning_rate": 6.9700744644105815e-06, + "loss": 0.5245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1476 + }, + { + "epoch": 1.8212083847102343, + "grad_norm": 2.420190205704888, + "learning_rate": 6.9575134604124364e-06, + "loss": 0.5115, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1477 + }, + { + "epoch": 1.8224414303329222, + "grad_norm": 2.6220274719559424, + "learning_rate": 6.9449577442781246e-06, + "loss": 0.5708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1478 + }, + { + "epoch": 1.8236744759556105, + "grad_norm": 2.4198151981250624, + "learning_rate": 6.932407337829579e-06, + "loss": 0.5779, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1479 + }, + { + "epoch": 1.8249075215782984, + "grad_norm": 2.710527668536825, + "learning_rate": 6.919862262879488e-06, + "loss": 0.706, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1480 + }, + { + "epoch": 1.8261405672009863, + "grad_norm": 2.260314185301613, + "learning_rate": 6.907322541231294e-06, + "loss": 0.3152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1481 + }, + { + "epoch": 1.8273736128236746, + "grad_norm": 2.4880601560363087, + "learning_rate": 6.894788194679112e-06, + "loss": 0.4559, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1482 + }, + { + "epoch": 1.8286066584463625, + "grad_norm": 2.53859220201902, + "learning_rate": 6.882259245007741e-06, + "loss": 0.6073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1483 + }, + { + "epoch": 1.8298397040690506, + "grad_norm": 2.406753699086694, + "learning_rate": 6.869735713992576e-06, + "loss": 0.5376, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1484 + }, + { + "epoch": 1.8310727496917387, + "grad_norm": 2.5573243142415616, + "learning_rate": 6.857217623399613e-06, + "loss": 0.5161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1485 + }, + { + "epoch": 1.8323057953144266, + "grad_norm": 2.4701507919473964, + "learning_rate": 6.844704994985383e-06, + "loss": 0.4962, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1486 + }, + { + "epoch": 1.8335388409371147, + "grad_norm": 2.505190115969386, + "learning_rate": 6.832197850496922e-06, + "loss": 0.5941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1487 + }, + { + "epoch": 1.8347718865598028, + "grad_norm": 2.8094520017647766, + "learning_rate": 6.819696211671745e-06, + "loss": 0.6998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1488 + }, + { + "epoch": 1.8360049321824907, + "grad_norm": 2.7116637712221685, + "learning_rate": 6.807200100237786e-06, + "loss": 0.5916, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1489 + }, + { + "epoch": 1.8372379778051788, + "grad_norm": 2.4742606262555538, + "learning_rate": 6.794709537913382e-06, + "loss": 0.4571, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1490 + }, + { + "epoch": 1.838471023427867, + "grad_norm": 2.3388179882861038, + "learning_rate": 6.782224546407217e-06, + "loss": 0.4336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1491 + }, + { + "epoch": 1.8397040690505548, + "grad_norm": 2.5535368655522723, + "learning_rate": 6.769745147418298e-06, + "loss": 0.5513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1492 + }, + { + "epoch": 1.840937114673243, + "grad_norm": 2.3844296720019136, + "learning_rate": 6.75727136263592e-06, + "loss": 0.4769, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1493 + }, + { + "epoch": 1.842170160295931, + "grad_norm": 2.430982238426814, + "learning_rate": 6.744803213739599e-06, + "loss": 0.6191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1494 + }, + { + "epoch": 1.843403205918619, + "grad_norm": 2.5719309543455484, + "learning_rate": 6.732340722399083e-06, + "loss": 0.6451, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1495 + }, + { + "epoch": 1.844636251541307, + "grad_norm": 2.4358254754188247, + "learning_rate": 6.719883910274258e-06, + "loss": 0.6404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1496 + }, + { + "epoch": 1.8458692971639952, + "grad_norm": 2.2277422785314385, + "learning_rate": 6.7074327990151666e-06, + "loss": 0.4467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1497 + }, + { + "epoch": 1.847102342786683, + "grad_norm": 2.3553522334648234, + "learning_rate": 6.694987410261927e-06, + "loss": 0.4733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1498 + }, + { + "epoch": 1.8483353884093712, + "grad_norm": 2.305268243528217, + "learning_rate": 6.6825477656447135e-06, + "loss": 0.5213, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1499 + }, + { + "epoch": 1.8495684340320593, + "grad_norm": 2.476358939717476, + "learning_rate": 6.6701138867837245e-06, + "loss": 0.5979, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1500 + }, + { + "epoch": 1.8508014796547472, + "grad_norm": 2.5413097090235897, + "learning_rate": 6.657685795289126e-06, + "loss": 0.464, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1501 + }, + { + "epoch": 1.8520345252774353, + "grad_norm": 2.581561221486171, + "learning_rate": 6.6452635127610366e-06, + "loss": 0.5169, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1502 + }, + { + "epoch": 1.8532675709001234, + "grad_norm": 2.752939514877913, + "learning_rate": 6.632847060789471e-06, + "loss": 0.5458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1503 + }, + { + "epoch": 1.8545006165228113, + "grad_norm": 2.393983614231186, + "learning_rate": 6.620436460954314e-06, + "loss": 0.5606, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1504 + }, + { + "epoch": 1.8557336621454994, + "grad_norm": 2.2296540462723415, + "learning_rate": 6.608031734825282e-06, + "loss": 0.5184, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1505 + }, + { + "epoch": 1.8569667077681875, + "grad_norm": 2.3150895635322377, + "learning_rate": 6.595632903961876e-06, + "loss": 0.3391, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1506 + }, + { + "epoch": 1.8581997533908754, + "grad_norm": 2.4812557469058687, + "learning_rate": 6.583239989913357e-06, + "loss": 0.5283, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1507 + }, + { + "epoch": 1.8594327990135635, + "grad_norm": 2.240249157372891, + "learning_rate": 6.570853014218699e-06, + "loss": 0.4267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1508 + }, + { + "epoch": 1.8606658446362516, + "grad_norm": 2.418565073853559, + "learning_rate": 6.558471998406556e-06, + "loss": 0.5317, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1509 + }, + { + "epoch": 1.8618988902589395, + "grad_norm": 2.509766709851647, + "learning_rate": 6.546096963995221e-06, + "loss": 0.5639, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1510 + }, + { + "epoch": 1.8631319358816276, + "grad_norm": 2.6229591577559606, + "learning_rate": 6.5337279324925995e-06, + "loss": 0.6702, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1511 + }, + { + "epoch": 1.8643649815043157, + "grad_norm": 2.8436863684007605, + "learning_rate": 6.521364925396159e-06, + "loss": 0.6629, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1512 + }, + { + "epoch": 1.8655980271270036, + "grad_norm": 2.88285582890902, + "learning_rate": 6.50900796419289e-06, + "loss": 0.6007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1513 + }, + { + "epoch": 1.8668310727496917, + "grad_norm": 2.4591483237884533, + "learning_rate": 6.496657070359289e-06, + "loss": 0.5649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1514 + }, + { + "epoch": 1.8680641183723798, + "grad_norm": 2.588461757043, + "learning_rate": 6.484312265361294e-06, + "loss": 0.6014, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1515 + }, + { + "epoch": 1.8692971639950677, + "grad_norm": 2.3915380277510256, + "learning_rate": 6.47197357065427e-06, + "loss": 0.5105, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1516 + }, + { + "epoch": 1.8705302096177558, + "grad_norm": 2.4853884668402597, + "learning_rate": 6.459641007682953e-06, + "loss": 0.5631, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1517 + }, + { + "epoch": 1.871763255240444, + "grad_norm": 2.1904070749761546, + "learning_rate": 6.447314597881433e-06, + "loss": 0.4229, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1518 + }, + { + "epoch": 1.8729963008631318, + "grad_norm": 2.392397858425607, + "learning_rate": 6.4349943626731e-06, + "loss": 0.5323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1519 + }, + { + "epoch": 1.87422934648582, + "grad_norm": 2.5868871798442936, + "learning_rate": 6.422680323470609e-06, + "loss": 0.5943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1520 + }, + { + "epoch": 1.875462392108508, + "grad_norm": 2.8171109508645515, + "learning_rate": 6.410372501675854e-06, + "loss": 0.6011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1521 + }, + { + "epoch": 1.876695437731196, + "grad_norm": 2.5043723972276406, + "learning_rate": 6.398070918679917e-06, + "loss": 0.6401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1522 + }, + { + "epoch": 1.877928483353884, + "grad_norm": 2.2319121044157075, + "learning_rate": 6.3857755958630415e-06, + "loss": 0.5526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1523 + }, + { + "epoch": 1.8791615289765722, + "grad_norm": 2.668249861016908, + "learning_rate": 6.3734865545945834e-06, + "loss": 0.6208, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1524 + }, + { + "epoch": 1.88039457459926, + "grad_norm": 2.328498736662863, + "learning_rate": 6.361203816232988e-06, + "loss": 0.4526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1525 + }, + { + "epoch": 1.8816276202219482, + "grad_norm": 2.532493722492898, + "learning_rate": 6.348927402125753e-06, + "loss": 0.5397, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1526 + }, + { + "epoch": 1.8828606658446363, + "grad_norm": 2.4124669695000023, + "learning_rate": 6.3366573336093605e-06, + "loss": 0.3821, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1527 + }, + { + "epoch": 1.8840937114673242, + "grad_norm": 2.328411541073336, + "learning_rate": 6.324393632009295e-06, + "loss": 0.5093, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1528 + }, + { + "epoch": 1.8853267570900123, + "grad_norm": 2.5365227172099196, + "learning_rate": 6.312136318639945e-06, + "loss": 0.5598, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1529 + }, + { + "epoch": 1.8865598027127004, + "grad_norm": 2.607793778617426, + "learning_rate": 6.299885414804618e-06, + "loss": 0.5194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1530 + }, + { + "epoch": 1.8877928483353883, + "grad_norm": 2.335857326269118, + "learning_rate": 6.287640941795479e-06, + "loss": 0.3871, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1531 + }, + { + "epoch": 1.8890258939580764, + "grad_norm": 2.647531529113461, + "learning_rate": 6.275402920893503e-06, + "loss": 0.521, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1532 + }, + { + "epoch": 1.8902589395807645, + "grad_norm": 2.9804935503220285, + "learning_rate": 6.263171373368464e-06, + "loss": 0.5535, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1533 + }, + { + "epoch": 1.8914919852034524, + "grad_norm": 2.4613526931496215, + "learning_rate": 6.250946320478879e-06, + "loss": 0.3987, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1534 + }, + { + "epoch": 1.8927250308261405, + "grad_norm": 2.694644539489535, + "learning_rate": 6.238727783471983e-06, + "loss": 0.5107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1535 + }, + { + "epoch": 1.8939580764488286, + "grad_norm": 2.516795775084936, + "learning_rate": 6.2265157835836775e-06, + "loss": 0.3624, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1536 + }, + { + "epoch": 1.8951911220715165, + "grad_norm": 2.893994117314153, + "learning_rate": 6.214310342038507e-06, + "loss": 0.5791, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1537 + }, + { + "epoch": 1.8964241676942046, + "grad_norm": 2.773382488051452, + "learning_rate": 6.202111480049628e-06, + "loss": 0.6339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1538 + }, + { + "epoch": 1.8976572133168927, + "grad_norm": 2.797457073115434, + "learning_rate": 6.1899192188187385e-06, + "loss": 0.618, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1539 + }, + { + "epoch": 1.8988902589395806, + "grad_norm": 2.5101307384888343, + "learning_rate": 6.177733579536091e-06, + "loss": 0.4852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1540 + }, + { + "epoch": 1.900123304562269, + "grad_norm": 2.711866598306881, + "learning_rate": 6.165554583380401e-06, + "loss": 0.5412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1541 + }, + { + "epoch": 1.9013563501849569, + "grad_norm": 2.41897150751434, + "learning_rate": 6.1533822515188665e-06, + "loss": 0.4917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1542 + }, + { + "epoch": 1.9025893958076447, + "grad_norm": 2.4935923340546875, + "learning_rate": 6.141216605107077e-06, + "loss": 0.4712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1543 + }, + { + "epoch": 1.903822441430333, + "grad_norm": 2.3408793732304356, + "learning_rate": 6.129057665289024e-06, + "loss": 0.4915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1544 + }, + { + "epoch": 1.905055487053021, + "grad_norm": 2.49420888585378, + "learning_rate": 6.116905453197032e-06, + "loss": 0.4243, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1545 + }, + { + "epoch": 1.9062885326757089, + "grad_norm": 2.3882774651201517, + "learning_rate": 6.1047599899517305e-06, + "loss": 0.3567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1546 + }, + { + "epoch": 1.9075215782983972, + "grad_norm": 2.412746547072853, + "learning_rate": 6.092621296662028e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1547 + }, + { + "epoch": 1.908754623921085, + "grad_norm": 2.793611618535343, + "learning_rate": 6.080489394425058e-06, + "loss": 0.4794, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1548 + }, + { + "epoch": 1.909987669543773, + "grad_norm": 2.86759252402197, + "learning_rate": 6.06836430432616e-06, + "loss": 0.5258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1549 + }, + { + "epoch": 1.9112207151664613, + "grad_norm": 3.5964778408959197, + "learning_rate": 6.056246047438823e-06, + "loss": 0.6431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1550 + }, + { + "epoch": 1.9124537607891492, + "grad_norm": 3.004291710626235, + "learning_rate": 6.044134644824671e-06, + "loss": 0.5851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1551 + }, + { + "epoch": 1.913686806411837, + "grad_norm": 2.553994453235498, + "learning_rate": 6.0320301175334096e-06, + "loss": 0.4947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1552 + }, + { + "epoch": 1.9149198520345254, + "grad_norm": 2.3305952965305723, + "learning_rate": 6.019932486602793e-06, + "loss": 0.4291, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1553 + }, + { + "epoch": 1.9161528976572133, + "grad_norm": 2.70796232056951, + "learning_rate": 6.007841773058596e-06, + "loss": 0.5056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1554 + }, + { + "epoch": 1.9173859432799012, + "grad_norm": 2.401542735999577, + "learning_rate": 5.9957579979145595e-06, + "loss": 0.4256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1555 + }, + { + "epoch": 1.9186189889025895, + "grad_norm": 2.6832304024737783, + "learning_rate": 5.983681182172387e-06, + "loss": 0.6448, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1556 + }, + { + "epoch": 1.9198520345252774, + "grad_norm": 2.5250444762365314, + "learning_rate": 5.971611346821657e-06, + "loss": 0.5107, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1557 + }, + { + "epoch": 1.9210850801479655, + "grad_norm": 2.487361580061221, + "learning_rate": 5.959548512839839e-06, + "loss": 0.4747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1558 + }, + { + "epoch": 1.9223181257706536, + "grad_norm": 2.5305388152078963, + "learning_rate": 5.947492701192229e-06, + "loss": 0.4648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1559 + }, + { + "epoch": 1.9235511713933415, + "grad_norm": 2.455837488266591, + "learning_rate": 5.9354439328319105e-06, + "loss": 0.4423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1560 + }, + { + "epoch": 1.9247842170160296, + "grad_norm": 2.468256257896946, + "learning_rate": 5.923402228699736e-06, + "loss": 0.4569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1561 + }, + { + "epoch": 1.9260172626387178, + "grad_norm": 2.42939124280653, + "learning_rate": 5.91136760972427e-06, + "loss": 0.3932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1562 + }, + { + "epoch": 1.9272503082614056, + "grad_norm": 2.858568888655988, + "learning_rate": 5.89934009682177e-06, + "loss": 0.484, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1563 + }, + { + "epoch": 1.9284833538840938, + "grad_norm": 2.61347238372933, + "learning_rate": 5.887319710896144e-06, + "loss": 0.467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1564 + }, + { + "epoch": 1.9297163995067819, + "grad_norm": 2.7599324090566926, + "learning_rate": 5.875306472838908e-06, + "loss": 0.6094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1565 + }, + { + "epoch": 1.9309494451294698, + "grad_norm": 2.6103338708006505, + "learning_rate": 5.863300403529159e-06, + "loss": 0.4638, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1566 + }, + { + "epoch": 1.9321824907521579, + "grad_norm": 2.47982724255905, + "learning_rate": 5.85130152383353e-06, + "loss": 0.4826, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1567 + }, + { + "epoch": 1.933415536374846, + "grad_norm": 2.578853751626434, + "learning_rate": 5.839309854606167e-06, + "loss": 0.5235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1568 + }, + { + "epoch": 1.9346485819975339, + "grad_norm": 2.4399405587351146, + "learning_rate": 5.827325416688671e-06, + "loss": 0.4686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1569 + }, + { + "epoch": 1.935881627620222, + "grad_norm": 2.5935511573767585, + "learning_rate": 5.81534823091009e-06, + "loss": 0.4941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1570 + }, + { + "epoch": 1.93711467324291, + "grad_norm": 2.3108330384724596, + "learning_rate": 5.803378318086859e-06, + "loss": 0.4216, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1571 + }, + { + "epoch": 1.938347718865598, + "grad_norm": 2.542118633939692, + "learning_rate": 5.791415699022768e-06, + "loss": 0.5643, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1572 + }, + { + "epoch": 1.939580764488286, + "grad_norm": 2.6343709127133526, + "learning_rate": 5.779460394508944e-06, + "loss": 0.5811, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1573 + }, + { + "epoch": 1.9408138101109742, + "grad_norm": 2.476140626898723, + "learning_rate": 5.767512425323791e-06, + "loss": 0.4819, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1574 + }, + { + "epoch": 1.942046855733662, + "grad_norm": 2.6259462878787634, + "learning_rate": 5.755571812232967e-06, + "loss": 0.6072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1575 + }, + { + "epoch": 1.9432799013563502, + "grad_norm": 2.4454523827907804, + "learning_rate": 5.74363857598934e-06, + "loss": 0.4893, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1576 + }, + { + "epoch": 1.9445129469790383, + "grad_norm": 2.5545835071067575, + "learning_rate": 5.731712737332965e-06, + "loss": 0.5415, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1577 + }, + { + "epoch": 1.9457459926017262, + "grad_norm": 2.790586346292402, + "learning_rate": 5.7197943169910455e-06, + "loss": 0.619, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1578 + }, + { + "epoch": 1.9469790382244143, + "grad_norm": 2.5655789543184913, + "learning_rate": 5.707883335677867e-06, + "loss": 0.3993, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1579 + }, + { + "epoch": 1.9482120838471024, + "grad_norm": 2.3868281881615063, + "learning_rate": 5.695979814094814e-06, + "loss": 0.4513, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1580 + }, + { + "epoch": 1.9494451294697903, + "grad_norm": 2.5119777174691684, + "learning_rate": 5.6840837729302865e-06, + "loss": 0.403, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1581 + }, + { + "epoch": 1.9506781750924784, + "grad_norm": 2.454297741118528, + "learning_rate": 5.672195232859703e-06, + "loss": 0.4341, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1582 + }, + { + "epoch": 1.9519112207151665, + "grad_norm": 2.173279017356807, + "learning_rate": 5.660314214545417e-06, + "loss": 0.4009, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1583 + }, + { + "epoch": 1.9531442663378544, + "grad_norm": 2.41693418545067, + "learning_rate": 5.648440738636731e-06, + "loss": 0.5145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1584 + }, + { + "epoch": 1.9543773119605425, + "grad_norm": 2.451502353773675, + "learning_rate": 5.63657482576984e-06, + "loss": 0.4148, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1585 + }, + { + "epoch": 1.9556103575832307, + "grad_norm": 2.5952977561456083, + "learning_rate": 5.624716496567782e-06, + "loss": 0.4059, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1586 + }, + { + "epoch": 1.9568434032059185, + "grad_norm": 2.6119219402093647, + "learning_rate": 5.612865771640419e-06, + "loss": 0.4455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1587 + }, + { + "epoch": 1.9580764488286067, + "grad_norm": 2.5564882849476294, + "learning_rate": 5.601022671584394e-06, + "loss": 0.4633, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1588 + }, + { + "epoch": 1.9593094944512948, + "grad_norm": 2.8188713009613244, + "learning_rate": 5.58918721698311e-06, + "loss": 0.565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1589 + }, + { + "epoch": 1.9605425400739827, + "grad_norm": 2.4162766103238895, + "learning_rate": 5.577359428406669e-06, + "loss": 0.5188, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1590 + }, + { + "epoch": 1.9617755856966708, + "grad_norm": 2.50589027882338, + "learning_rate": 5.56553932641185e-06, + "loss": 0.4671, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1591 + }, + { + "epoch": 1.9630086313193589, + "grad_norm": 2.338580770590297, + "learning_rate": 5.553726931542085e-06, + "loss": 0.4271, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1592 + }, + { + "epoch": 1.9642416769420468, + "grad_norm": 2.727731476438198, + "learning_rate": 5.541922264327395e-06, + "loss": 0.5796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1593 + }, + { + "epoch": 1.9654747225647349, + "grad_norm": 2.372879579889336, + "learning_rate": 5.530125345284388e-06, + "loss": 0.4791, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1594 + }, + { + "epoch": 1.966707768187423, + "grad_norm": 2.402358128750275, + "learning_rate": 5.5183361949161804e-06, + "loss": 0.5589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1595 + }, + { + "epoch": 1.9679408138101109, + "grad_norm": 2.521073091835758, + "learning_rate": 5.506554833712412e-06, + "loss": 0.5612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1596 + }, + { + "epoch": 1.969173859432799, + "grad_norm": 2.475424445052322, + "learning_rate": 5.494781282149175e-06, + "loss": 0.4164, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1597 + }, + { + "epoch": 1.970406905055487, + "grad_norm": 2.4762092606289388, + "learning_rate": 5.4830155606889855e-06, + "loss": 0.6031, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1598 + }, + { + "epoch": 1.971639950678175, + "grad_norm": 2.3075398162125857, + "learning_rate": 5.471257689780753e-06, + "loss": 0.3457, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1599 + }, + { + "epoch": 1.972872996300863, + "grad_norm": 2.8308395918813405, + "learning_rate": 5.459507689859742e-06, + "loss": 0.5926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1600 + }, + { + "epoch": 1.9741060419235512, + "grad_norm": 2.695737068994709, + "learning_rate": 5.447765581347542e-06, + "loss": 0.5015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1601 + }, + { + "epoch": 1.975339087546239, + "grad_norm": 2.88226095679846, + "learning_rate": 5.436031384652021e-06, + "loss": 0.5126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1602 + }, + { + "epoch": 1.9765721331689272, + "grad_norm": 2.466798154495963, + "learning_rate": 5.424305120167295e-06, + "loss": 0.3504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1603 + }, + { + "epoch": 1.9778051787916153, + "grad_norm": 2.999551007485252, + "learning_rate": 5.412586808273705e-06, + "loss": 0.6638, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1604 + }, + { + "epoch": 1.9790382244143032, + "grad_norm": 2.687785997003729, + "learning_rate": 5.400876469337759e-06, + "loss": 0.4778, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1605 + }, + { + "epoch": 1.9802712700369913, + "grad_norm": 2.697985425334561, + "learning_rate": 5.389174123712113e-06, + "loss": 0.4376, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1606 + }, + { + "epoch": 1.9815043156596794, + "grad_norm": 2.2270783908633094, + "learning_rate": 5.377479791735522e-06, + "loss": 0.3055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1607 + }, + { + "epoch": 1.9827373612823673, + "grad_norm": 2.535575049691839, + "learning_rate": 5.3657934937328296e-06, + "loss": 0.4307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1608 + }, + { + "epoch": 1.9839704069050554, + "grad_norm": 2.6549657486777063, + "learning_rate": 5.354115250014904e-06, + "loss": 0.5848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1609 + }, + { + "epoch": 1.9852034525277436, + "grad_norm": 2.83855139958114, + "learning_rate": 5.342445080878614e-06, + "loss": 0.5668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1610 + }, + { + "epoch": 1.9864364981504314, + "grad_norm": 2.5321393689709284, + "learning_rate": 5.330783006606805e-06, + "loss": 0.3667, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1611 + }, + { + "epoch": 1.9876695437731196, + "grad_norm": 3.1246050900975657, + "learning_rate": 5.319129047468244e-06, + "loss": 0.5487, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1612 + }, + { + "epoch": 1.9889025893958077, + "grad_norm": 2.661845983771021, + "learning_rate": 5.307483223717598e-06, + "loss": 0.4961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1613 + }, + { + "epoch": 1.9901356350184956, + "grad_norm": 2.567032419410714, + "learning_rate": 5.295845555595396e-06, + "loss": 0.4989, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1614 + }, + { + "epoch": 1.9913686806411839, + "grad_norm": 2.715329364090932, + "learning_rate": 5.284216063327987e-06, + "loss": 0.55, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1615 + }, + { + "epoch": 1.9926017262638718, + "grad_norm": 2.4247843342398565, + "learning_rate": 5.27259476712751e-06, + "loss": 0.4278, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1616 + }, + { + "epoch": 1.9938347718865597, + "grad_norm": 2.586736358421805, + "learning_rate": 5.260981687191873e-06, + "loss": 0.3932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1617 + }, + { + "epoch": 1.995067817509248, + "grad_norm": 2.689699358030808, + "learning_rate": 5.249376843704686e-06, + "loss": 0.6152, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1618 + }, + { + "epoch": 1.9963008631319359, + "grad_norm": 2.393722648406475, + "learning_rate": 5.2377802568352495e-06, + "loss": 0.4516, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1619 + }, + { + "epoch": 1.9975339087546238, + "grad_norm": 2.151941518652342, + "learning_rate": 5.226191946738526e-06, + "loss": 0.4433, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1620 + }, + { + "epoch": 1.998766954377312, + "grad_norm": 2.7987504690673988, + "learning_rate": 5.214611933555076e-06, + "loss": 0.7023, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1621 + }, + { + "epoch": 2.0, + "grad_norm": 2.59776624758183, + "learning_rate": 5.20304023741105e-06, + "loss": 0.619, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1622 + }, + { + "epoch": 2.001233045622688, + "grad_norm": 2.43048324510223, + "learning_rate": 5.191476878418133e-06, + "loss": 0.4669, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1623 + }, + { + "epoch": 2.0024660912453762, + "grad_norm": 2.3352351017918402, + "learning_rate": 5.179921876673535e-06, + "loss": 0.4957, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1624 + }, + { + "epoch": 2.003699136868064, + "grad_norm": 2.395114246328322, + "learning_rate": 5.168375252259938e-06, + "loss": 0.4676, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1625 + }, + { + "epoch": 2.004932182490752, + "grad_norm": 2.2731002396640685, + "learning_rate": 5.156837025245447e-06, + "loss": 0.4332, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1626 + }, + { + "epoch": 2.0061652281134403, + "grad_norm": 2.7712994823928057, + "learning_rate": 5.145307215683595e-06, + "loss": 0.5389, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1627 + }, + { + "epoch": 2.0073982737361282, + "grad_norm": 2.4056856515117993, + "learning_rate": 5.133785843613268e-06, + "loss": 0.4502, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1628 + }, + { + "epoch": 2.008631319358816, + "grad_norm": 2.2752234585471305, + "learning_rate": 5.122272929058701e-06, + "loss": 0.3659, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1629 + }, + { + "epoch": 2.0098643649815044, + "grad_norm": 2.388137760527975, + "learning_rate": 5.110768492029421e-06, + "loss": 0.4334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1630 + }, + { + "epoch": 2.0110974106041923, + "grad_norm": 2.5706000208497217, + "learning_rate": 5.0992725525202205e-06, + "loss": 0.4949, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1631 + }, + { + "epoch": 2.0123304562268802, + "grad_norm": 2.6004403792326274, + "learning_rate": 5.0877851305111334e-06, + "loss": 0.5269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1632 + }, + { + "epoch": 2.0135635018495686, + "grad_norm": 3.0312932980382468, + "learning_rate": 5.076306245967378e-06, + "loss": 0.5396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1633 + }, + { + "epoch": 2.0147965474722564, + "grad_norm": 2.439419139508058, + "learning_rate": 5.064835918839341e-06, + "loss": 0.3857, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1634 + }, + { + "epoch": 2.0160295930949443, + "grad_norm": 2.5177486175717974, + "learning_rate": 5.053374169062532e-06, + "loss": 0.4622, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1635 + }, + { + "epoch": 2.0172626387176327, + "grad_norm": 2.506250151038749, + "learning_rate": 5.041921016557556e-06, + "loss": 0.4383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1636 + }, + { + "epoch": 2.0184956843403206, + "grad_norm": 2.704760808949749, + "learning_rate": 5.030476481230086e-06, + "loss": 0.6166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1637 + }, + { + "epoch": 2.0197287299630085, + "grad_norm": 2.4695715254758337, + "learning_rate": 5.0190405829707935e-06, + "loss": 0.561, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1638 + }, + { + "epoch": 2.020961775585697, + "grad_norm": 2.7241602595851973, + "learning_rate": 5.007613341655363e-06, + "loss": 0.445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1639 + }, + { + "epoch": 2.0221948212083847, + "grad_norm": 2.3233960707374854, + "learning_rate": 4.996194777144416e-06, + "loss": 0.3948, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1640 + }, + { + "epoch": 2.0234278668310726, + "grad_norm": 2.2442713678355823, + "learning_rate": 4.984784909283513e-06, + "loss": 0.3532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1641 + }, + { + "epoch": 2.024660912453761, + "grad_norm": 2.634392676529801, + "learning_rate": 4.973383757903072e-06, + "loss": 0.523, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1642 + }, + { + "epoch": 2.025893958076449, + "grad_norm": 2.8398388568834543, + "learning_rate": 4.961991342818385e-06, + "loss": 0.5234, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1643 + }, + { + "epoch": 2.0271270036991367, + "grad_norm": 2.6473433580844534, + "learning_rate": 4.950607683829557e-06, + "loss": 0.5612, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1644 + }, + { + "epoch": 2.028360049321825, + "grad_norm": 2.523266270766798, + "learning_rate": 4.939232800721466e-06, + "loss": 0.5246, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1645 + }, + { + "epoch": 2.029593094944513, + "grad_norm": 2.5900711083492407, + "learning_rate": 4.927866713263744e-06, + "loss": 0.5103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1646 + }, + { + "epoch": 2.030826140567201, + "grad_norm": 2.463325250265772, + "learning_rate": 4.9165094412107275e-06, + "loss": 0.3759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1647 + }, + { + "epoch": 2.032059186189889, + "grad_norm": 2.494410303111952, + "learning_rate": 4.905161004301446e-06, + "loss": 0.4665, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1648 + }, + { + "epoch": 2.033292231812577, + "grad_norm": 2.5651504375469987, + "learning_rate": 4.8938214222595636e-06, + "loss": 0.4896, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1649 + }, + { + "epoch": 2.034525277435265, + "grad_norm": 2.429966304010776, + "learning_rate": 4.882490714793352e-06, + "loss": 0.4589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1650 + }, + { + "epoch": 2.0357583230579532, + "grad_norm": 2.5231453655050653, + "learning_rate": 4.871168901595672e-06, + "loss": 0.5689, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1651 + }, + { + "epoch": 2.036991368680641, + "grad_norm": 2.6634619932788177, + "learning_rate": 4.859856002343913e-06, + "loss": 0.5838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1652 + }, + { + "epoch": 2.038224414303329, + "grad_norm": 2.5914302041981654, + "learning_rate": 4.8485520366999776e-06, + "loss": 0.4551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1653 + }, + { + "epoch": 2.0394574599260173, + "grad_norm": 2.22524078497228, + "learning_rate": 4.8372570243102345e-06, + "loss": 0.3869, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1654 + }, + { + "epoch": 2.0406905055487052, + "grad_norm": 2.4814059043859804, + "learning_rate": 4.825970984805508e-06, + "loss": 0.4673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1655 + }, + { + "epoch": 2.041923551171393, + "grad_norm": 2.376833385977507, + "learning_rate": 4.814693937801008e-06, + "loss": 0.5249, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1656 + }, + { + "epoch": 2.0431565967940815, + "grad_norm": 2.522951679239216, + "learning_rate": 4.803425902896333e-06, + "loss": 0.4087, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1657 + }, + { + "epoch": 2.0443896424167693, + "grad_norm": 2.572178268061794, + "learning_rate": 4.792166899675407e-06, + "loss": 0.4922, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1658 + }, + { + "epoch": 2.0456226880394572, + "grad_norm": 2.2642013364057383, + "learning_rate": 4.780916947706454e-06, + "loss": 0.3774, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1659 + }, + { + "epoch": 2.0468557336621456, + "grad_norm": 2.462040224339899, + "learning_rate": 4.769676066541981e-06, + "loss": 0.3919, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1660 + }, + { + "epoch": 2.0480887792848335, + "grad_norm": 2.5189042773825925, + "learning_rate": 4.75844427571872e-06, + "loss": 0.4466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1661 + }, + { + "epoch": 2.049321824907522, + "grad_norm": 2.6470599682008316, + "learning_rate": 4.747221594757598e-06, + "loss": 0.4843, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1662 + }, + { + "epoch": 2.0505548705302097, + "grad_norm": 2.543427335219999, + "learning_rate": 4.736008043163727e-06, + "loss": 0.5011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1663 + }, + { + "epoch": 2.0517879161528976, + "grad_norm": 2.832308169983458, + "learning_rate": 4.724803640426337e-06, + "loss": 0.4082, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1664 + }, + { + "epoch": 2.053020961775586, + "grad_norm": 2.448298632296988, + "learning_rate": 4.71360840601876e-06, + "loss": 0.3741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1665 + }, + { + "epoch": 2.054254007398274, + "grad_norm": 2.883720119682108, + "learning_rate": 4.702422359398394e-06, + "loss": 0.5057, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1666 + }, + { + "epoch": 2.0554870530209617, + "grad_norm": 2.392751477441128, + "learning_rate": 4.691245520006676e-06, + "loss": 0.351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1667 + }, + { + "epoch": 2.05672009864365, + "grad_norm": 2.2055087345825757, + "learning_rate": 4.680077907269032e-06, + "loss": 0.2998, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1668 + }, + { + "epoch": 2.057953144266338, + "grad_norm": 2.473248669807766, + "learning_rate": 4.668919540594849e-06, + "loss": 0.5238, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1669 + }, + { + "epoch": 2.059186189889026, + "grad_norm": 2.3174240557182295, + "learning_rate": 4.657770439377458e-06, + "loss": 0.3478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1670 + }, + { + "epoch": 2.060419235511714, + "grad_norm": 2.2018582212758817, + "learning_rate": 4.646630622994072e-06, + "loss": 0.3582, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1671 + }, + { + "epoch": 2.061652281134402, + "grad_norm": 2.537278150516378, + "learning_rate": 4.635500110805782e-06, + "loss": 0.442, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1672 + }, + { + "epoch": 2.06288532675709, + "grad_norm": 2.7734825628365094, + "learning_rate": 4.624378922157486e-06, + "loss": 0.5095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1673 + }, + { + "epoch": 2.0641183723797782, + "grad_norm": 2.5606426040553547, + "learning_rate": 4.6132670763779045e-06, + "loss": 0.3809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1674 + }, + { + "epoch": 2.065351418002466, + "grad_norm": 2.6820946526966045, + "learning_rate": 4.602164592779495e-06, + "loss": 0.4306, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1675 + }, + { + "epoch": 2.066584463625154, + "grad_norm": 2.5728870468376153, + "learning_rate": 4.591071490658465e-06, + "loss": 0.3495, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1676 + }, + { + "epoch": 2.0678175092478424, + "grad_norm": 2.9950959525162113, + "learning_rate": 4.579987789294705e-06, + "loss": 0.5589, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1677 + }, + { + "epoch": 2.0690505548705302, + "grad_norm": 2.6084748423929947, + "learning_rate": 4.56891350795176e-06, + "loss": 0.4674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1678 + }, + { + "epoch": 2.070283600493218, + "grad_norm": 2.558954194636351, + "learning_rate": 4.5578486658768236e-06, + "loss": 0.5383, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1679 + }, + { + "epoch": 2.0715166461159065, + "grad_norm": 2.1592602256682083, + "learning_rate": 4.546793282300669e-06, + "loss": 0.3016, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1680 + }, + { + "epoch": 2.0727496917385944, + "grad_norm": 2.6206188628458915, + "learning_rate": 4.53574737643763e-06, + "loss": 0.3556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1681 + }, + { + "epoch": 2.0739827373612822, + "grad_norm": 2.5923118457963974, + "learning_rate": 4.524710967485571e-06, + "loss": 0.3413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1682 + }, + { + "epoch": 2.0752157829839706, + "grad_norm": 2.50571139289609, + "learning_rate": 4.513684074625853e-06, + "loss": 0.3758, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1683 + }, + { + "epoch": 2.0764488286066585, + "grad_norm": 2.738099209536595, + "learning_rate": 4.5026667170233065e-06, + "loss": 0.5091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1684 + }, + { + "epoch": 2.0776818742293464, + "grad_norm": 2.8102966517380294, + "learning_rate": 4.491658913826161e-06, + "loss": 0.5011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1685 + }, + { + "epoch": 2.0789149198520347, + "grad_norm": 2.7957945330332117, + "learning_rate": 4.480660684166074e-06, + "loss": 0.3797, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1686 + }, + { + "epoch": 2.0801479654747226, + "grad_norm": 3.0246754194205074, + "learning_rate": 4.469672047158039e-06, + "loss": 0.5499, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1687 + }, + { + "epoch": 2.0813810110974105, + "grad_norm": 2.7422346079651865, + "learning_rate": 4.458693021900399e-06, + "loss": 0.5348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1688 + }, + { + "epoch": 2.082614056720099, + "grad_norm": 2.6624901301648087, + "learning_rate": 4.4477236274747646e-06, + "loss": 0.4238, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1689 + }, + { + "epoch": 2.0838471023427867, + "grad_norm": 2.4655736338315104, + "learning_rate": 4.436763882946032e-06, + "loss": 0.4601, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1690 + }, + { + "epoch": 2.0850801479654746, + "grad_norm": 2.5201507017265534, + "learning_rate": 4.4258138073623216e-06, + "loss": 0.4972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1691 + }, + { + "epoch": 2.086313193588163, + "grad_norm": 2.375598788079959, + "learning_rate": 4.414873419754939e-06, + "loss": 0.4604, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1692 + }, + { + "epoch": 2.087546239210851, + "grad_norm": 2.6310260435763464, + "learning_rate": 4.403942739138359e-06, + "loss": 0.4896, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1693 + }, + { + "epoch": 2.0887792848335387, + "grad_norm": 2.7962256448727465, + "learning_rate": 4.393021784510182e-06, + "loss": 0.4518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1694 + }, + { + "epoch": 2.090012330456227, + "grad_norm": 2.5173184764135264, + "learning_rate": 4.382110574851112e-06, + "loss": 0.5738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1695 + }, + { + "epoch": 2.091245376078915, + "grad_norm": 2.3857792383387415, + "learning_rate": 4.371209129124908e-06, + "loss": 0.2941, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1696 + }, + { + "epoch": 2.092478421701603, + "grad_norm": 2.3196034427524386, + "learning_rate": 4.36031746627836e-06, + "loss": 0.3398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1697 + }, + { + "epoch": 2.093711467324291, + "grad_norm": 2.674529578743137, + "learning_rate": 4.349435605241263e-06, + "loss": 0.5239, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1698 + }, + { + "epoch": 2.094944512946979, + "grad_norm": 2.69239620455149, + "learning_rate": 4.338563564926368e-06, + "loss": 0.4752, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1699 + }, + { + "epoch": 2.096177558569667, + "grad_norm": 2.2696516232736665, + "learning_rate": 4.327701364229361e-06, + "loss": 0.3404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1700 + }, + { + "epoch": 2.0974106041923553, + "grad_norm": 2.94419875596226, + "learning_rate": 4.316849022028823e-06, + "loss": 0.5584, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.26, + "memory/max_mem_allocated(gib)": 18.29, + "step": 1701 + }, + { + "epoch": 2.098643649815043, + "grad_norm": 2.63116353614959, + "learning_rate": 4.3060065571862065e-06, + "loss": 0.5006, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1702 + }, + { + "epoch": 2.099876695437731, + "grad_norm": 2.6481372890887016, + "learning_rate": 4.295173988545798e-06, + "loss": 0.4588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1703 + }, + { + "epoch": 2.1011097410604194, + "grad_norm": 2.654639166541817, + "learning_rate": 4.284351334934679e-06, + "loss": 0.4244, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1704 + }, + { + "epoch": 2.1023427866831073, + "grad_norm": 2.580352134378634, + "learning_rate": 4.273538615162699e-06, + "loss": 0.4331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1705 + }, + { + "epoch": 2.103575832305795, + "grad_norm": 2.7091649413696732, + "learning_rate": 4.262735848022442e-06, + "loss": 0.6151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1706 + }, + { + "epoch": 2.1048088779284835, + "grad_norm": 2.5378400192827333, + "learning_rate": 4.2519430522892006e-06, + "loss": 0.3985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1707 + }, + { + "epoch": 2.1060419235511714, + "grad_norm": 2.5807185135055315, + "learning_rate": 4.241160246720931e-06, + "loss": 0.5335, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1708 + }, + { + "epoch": 2.1072749691738593, + "grad_norm": 2.5355635545223016, + "learning_rate": 4.230387450058225e-06, + "loss": 0.4811, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1709 + }, + { + "epoch": 2.1085080147965476, + "grad_norm": 2.6493461999852688, + "learning_rate": 4.219624681024288e-06, + "loss": 0.5149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1710 + }, + { + "epoch": 2.1097410604192355, + "grad_norm": 2.28405593070979, + "learning_rate": 4.2088719583248886e-06, + "loss": 0.3647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1711 + }, + { + "epoch": 2.1109741060419234, + "grad_norm": 2.6503235283708233, + "learning_rate": 4.198129300648336e-06, + "loss": 0.4812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1712 + }, + { + "epoch": 2.1122071516646117, + "grad_norm": 2.8066242646378274, + "learning_rate": 4.1873967266654444e-06, + "loss": 0.4478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1713 + }, + { + "epoch": 2.1134401972872996, + "grad_norm": 2.5301643915382, + "learning_rate": 4.1766742550295145e-06, + "loss": 0.3663, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1714 + }, + { + "epoch": 2.1146732429099875, + "grad_norm": 2.6643383692422344, + "learning_rate": 4.165961904376275e-06, + "loss": 0.4954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1715 + }, + { + "epoch": 2.115906288532676, + "grad_norm": 2.687328638943086, + "learning_rate": 4.155259693323869e-06, + "loss": 0.5286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1716 + }, + { + "epoch": 2.1171393341553637, + "grad_norm": 2.3051275379782323, + "learning_rate": 4.144567640472819e-06, + "loss": 0.4043, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1717 + }, + { + "epoch": 2.1183723797780516, + "grad_norm": 2.6933982791775635, + "learning_rate": 4.13388576440599e-06, + "loss": 0.6243, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1718 + }, + { + "epoch": 2.11960542540074, + "grad_norm": 2.2088880651767173, + "learning_rate": 4.123214083688567e-06, + "loss": 0.3343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1719 + }, + { + "epoch": 2.120838471023428, + "grad_norm": 2.47172384136877, + "learning_rate": 4.112552616867997e-06, + "loss": 0.4237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1720 + }, + { + "epoch": 2.1220715166461157, + "grad_norm": 2.7748339806468176, + "learning_rate": 4.101901382473997e-06, + "loss": 0.5573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1721 + }, + { + "epoch": 2.123304562268804, + "grad_norm": 2.424501283224548, + "learning_rate": 4.091260399018484e-06, + "loss": 0.4265, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1722 + }, + { + "epoch": 2.124537607891492, + "grad_norm": 2.650174524506375, + "learning_rate": 4.08062968499557e-06, + "loss": 0.3477, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1723 + }, + { + "epoch": 2.12577065351418, + "grad_norm": 2.1938914180986817, + "learning_rate": 4.070009258881514e-06, + "loss": 0.2546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1724 + }, + { + "epoch": 2.127003699136868, + "grad_norm": 2.427206552496681, + "learning_rate": 4.059399139134687e-06, + "loss": 0.4614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1725 + }, + { + "epoch": 2.128236744759556, + "grad_norm": 2.600029453320743, + "learning_rate": 4.048799344195563e-06, + "loss": 0.5003, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1726 + }, + { + "epoch": 2.1294697903822444, + "grad_norm": 2.525899460269773, + "learning_rate": 4.038209892486659e-06, + "loss": 0.4207, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1727 + }, + { + "epoch": 2.1307028360049323, + "grad_norm": 2.360301952652528, + "learning_rate": 4.027630802412517e-06, + "loss": 0.3161, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1728 + }, + { + "epoch": 2.13193588162762, + "grad_norm": 2.3181572601045337, + "learning_rate": 4.017062092359678e-06, + "loss": 0.3167, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1729 + }, + { + "epoch": 2.133168927250308, + "grad_norm": 2.7352618597650915, + "learning_rate": 4.006503780696632e-06, + "loss": 0.4445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1730 + }, + { + "epoch": 2.1344019728729964, + "grad_norm": 2.2207856211599877, + "learning_rate": 3.995955885773813e-06, + "loss": 0.3265, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1731 + }, + { + "epoch": 2.1356350184956843, + "grad_norm": 2.6621075014327413, + "learning_rate": 3.985418425923524e-06, + "loss": 0.4374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1732 + }, + { + "epoch": 2.1368680641183726, + "grad_norm": 2.5556232841368036, + "learning_rate": 3.974891419459958e-06, + "loss": 0.4089, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1733 + }, + { + "epoch": 2.1381011097410605, + "grad_norm": 2.6065282615267513, + "learning_rate": 3.964374884679121e-06, + "loss": 0.4171, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1734 + }, + { + "epoch": 2.1393341553637484, + "grad_norm": 2.373815079862734, + "learning_rate": 3.953868839858835e-06, + "loss": 0.3434, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1735 + }, + { + "epoch": 2.1405672009864363, + "grad_norm": 2.9153003666503645, + "learning_rate": 3.943373303258681e-06, + "loss": 0.4485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1736 + }, + { + "epoch": 2.1418002466091246, + "grad_norm": 2.6549134296354744, + "learning_rate": 3.932888293119973e-06, + "loss": 0.3891, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1737 + }, + { + "epoch": 2.1430332922318125, + "grad_norm": 2.910217909489204, + "learning_rate": 3.922413827665743e-06, + "loss": 0.4802, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1738 + }, + { + "epoch": 2.144266337854501, + "grad_norm": 2.9357454721384166, + "learning_rate": 3.911949925100685e-06, + "loss": 0.5186, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1739 + }, + { + "epoch": 2.1454993834771887, + "grad_norm": 2.2618509660304387, + "learning_rate": 3.9014966036111354e-06, + "loss": 0.3406, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1740 + }, + { + "epoch": 2.1467324290998766, + "grad_norm": 2.435460049819525, + "learning_rate": 3.891053881365043e-06, + "loss": 0.2776, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1741 + }, + { + "epoch": 2.147965474722565, + "grad_norm": 2.7650110493406896, + "learning_rate": 3.8806217765119424e-06, + "loss": 0.4867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1742 + }, + { + "epoch": 2.149198520345253, + "grad_norm": 2.5854743717015416, + "learning_rate": 3.870200307182903e-06, + "loss": 0.4048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1743 + }, + { + "epoch": 2.1504315659679407, + "grad_norm": 2.3719198940009254, + "learning_rate": 3.859789491490512e-06, + "loss": 0.4134, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1744 + }, + { + "epoch": 2.151664611590629, + "grad_norm": 2.868930160159961, + "learning_rate": 3.8493893475288505e-06, + "loss": 0.4194, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1745 + }, + { + "epoch": 2.152897657213317, + "grad_norm": 2.6468667044836782, + "learning_rate": 3.838999893373436e-06, + "loss": 0.3691, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1746 + }, + { + "epoch": 2.154130702836005, + "grad_norm": 2.604955712378186, + "learning_rate": 3.828621147081228e-06, + "loss": 0.3841, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1747 + }, + { + "epoch": 2.155363748458693, + "grad_norm": 2.6929978968474337, + "learning_rate": 3.818253126690549e-06, + "loss": 0.4759, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1748 + }, + { + "epoch": 2.156596794081381, + "grad_norm": 2.616836044439998, + "learning_rate": 3.8078958502210993e-06, + "loss": 0.4427, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1749 + }, + { + "epoch": 2.157829839704069, + "grad_norm": 2.333588490179514, + "learning_rate": 3.797549335673908e-06, + "loss": 0.3661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1750 + }, + { + "epoch": 2.1590628853267573, + "grad_norm": 2.3306365994040337, + "learning_rate": 3.7872136010312865e-06, + "loss": 0.3125, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1751 + }, + { + "epoch": 2.160295930949445, + "grad_norm": 2.6439633169753463, + "learning_rate": 3.776888664256818e-06, + "loss": 0.3674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1752 + }, + { + "epoch": 2.161528976572133, + "grad_norm": 2.3752280229105494, + "learning_rate": 3.766574543295316e-06, + "loss": 0.3564, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1753 + }, + { + "epoch": 2.1627620221948214, + "grad_norm": 2.3414215129379397, + "learning_rate": 3.756271256072803e-06, + "loss": 0.3079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1754 + }, + { + "epoch": 2.1639950678175093, + "grad_norm": 2.516659620108233, + "learning_rate": 3.7459788204964655e-06, + "loss": 0.4656, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1755 + }, + { + "epoch": 2.165228113440197, + "grad_norm": 2.544764929665444, + "learning_rate": 3.735697254454627e-06, + "loss": 0.4453, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1756 + }, + { + "epoch": 2.1664611590628855, + "grad_norm": 2.497553122385638, + "learning_rate": 3.725426575816732e-06, + "loss": 0.4223, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1757 + }, + { + "epoch": 2.1676942046855734, + "grad_norm": 2.278810104971583, + "learning_rate": 3.715166802433292e-06, + "loss": 0.3068, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1758 + }, + { + "epoch": 2.1689272503082613, + "grad_norm": 2.5107937194021446, + "learning_rate": 3.704917952135868e-06, + "loss": 0.3118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1759 + }, + { + "epoch": 2.1701602959309496, + "grad_norm": 2.7208144861944437, + "learning_rate": 3.6946800427370324e-06, + "loss": 0.3651, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1760 + }, + { + "epoch": 2.1713933415536375, + "grad_norm": 2.667315336204, + "learning_rate": 3.68445309203035e-06, + "loss": 0.4358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1761 + }, + { + "epoch": 2.1726263871763254, + "grad_norm": 2.512467743621635, + "learning_rate": 3.6742371177903433e-06, + "loss": 0.4263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1762 + }, + { + "epoch": 2.1738594327990137, + "grad_norm": 2.5887977603248773, + "learning_rate": 3.6640321377724364e-06, + "loss": 0.4037, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1763 + }, + { + "epoch": 2.1750924784217016, + "grad_norm": 2.499047622832279, + "learning_rate": 3.65383816971297e-06, + "loss": 0.3269, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1764 + }, + { + "epoch": 2.1763255240443895, + "grad_norm": 2.9728127490819247, + "learning_rate": 3.6436552313291264e-06, + "loss": 0.4693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1765 + }, + { + "epoch": 2.177558569667078, + "grad_norm": 2.9085704760310147, + "learning_rate": 3.633483340318935e-06, + "loss": 0.5007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1766 + }, + { + "epoch": 2.1787916152897657, + "grad_norm": 2.5551125679125106, + "learning_rate": 3.623322514361214e-06, + "loss": 0.3223, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1767 + }, + { + "epoch": 2.1800246609124536, + "grad_norm": 2.5216862821943584, + "learning_rate": 3.6131727711155484e-06, + "loss": 0.4088, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1768 + }, + { + "epoch": 2.181257706535142, + "grad_norm": 2.9196091577731518, + "learning_rate": 3.6030341282222735e-06, + "loss": 0.5717, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1769 + }, + { + "epoch": 2.18249075215783, + "grad_norm": 2.8463716887797625, + "learning_rate": 3.592906603302422e-06, + "loss": 0.5664, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1770 + }, + { + "epoch": 2.1837237977805177, + "grad_norm": 2.6993188839356352, + "learning_rate": 3.5827902139577044e-06, + "loss": 0.5555, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1771 + }, + { + "epoch": 2.184956843403206, + "grad_norm": 2.682693106322052, + "learning_rate": 3.5726849777704774e-06, + "loss": 0.4845, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1772 + }, + { + "epoch": 2.186189889025894, + "grad_norm": 2.7032999448673274, + "learning_rate": 3.5625909123037215e-06, + "loss": 0.4596, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1773 + }, + { + "epoch": 2.187422934648582, + "grad_norm": 2.476182150321705, + "learning_rate": 3.5525080351009933e-06, + "loss": 0.4628, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1774 + }, + { + "epoch": 2.18865598027127, + "grad_norm": 2.499458338772572, + "learning_rate": 3.5424363636864034e-06, + "loss": 0.438, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1775 + }, + { + "epoch": 2.189889025893958, + "grad_norm": 2.3950343929132067, + "learning_rate": 3.5323759155645964e-06, + "loss": 0.3968, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1776 + }, + { + "epoch": 2.191122071516646, + "grad_norm": 2.7292045770887996, + "learning_rate": 3.522326708220699e-06, + "loss": 0.3672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1777 + }, + { + "epoch": 2.1923551171393343, + "grad_norm": 2.5822142468134293, + "learning_rate": 3.512288759120317e-06, + "loss": 0.4035, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1778 + }, + { + "epoch": 2.193588162762022, + "grad_norm": 2.4201356770012135, + "learning_rate": 3.5022620857094645e-06, + "loss": 0.5251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1779 + }, + { + "epoch": 2.19482120838471, + "grad_norm": 2.468897467978063, + "learning_rate": 3.4922467054145846e-06, + "loss": 0.4279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1780 + }, + { + "epoch": 2.1960542540073984, + "grad_norm": 2.567136894287482, + "learning_rate": 3.4822426356424754e-06, + "loss": 0.4641, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1781 + }, + { + "epoch": 2.1972872996300863, + "grad_norm": 2.409074019271962, + "learning_rate": 3.472249893780287e-06, + "loss": 0.3393, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1782 + }, + { + "epoch": 2.198520345252774, + "grad_norm": 2.5475663022568473, + "learning_rate": 3.4622684971954777e-06, + "loss": 0.4396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1783 + }, + { + "epoch": 2.1997533908754625, + "grad_norm": 2.654800541696009, + "learning_rate": 3.4522984632357826e-06, + "loss": 0.4128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1784 + }, + { + "epoch": 2.2009864364981504, + "grad_norm": 2.251545490229221, + "learning_rate": 3.4423398092292003e-06, + "loss": 0.3661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1785 + }, + { + "epoch": 2.2022194821208383, + "grad_norm": 2.560355670972336, + "learning_rate": 3.432392552483941e-06, + "loss": 0.506, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1786 + }, + { + "epoch": 2.2034525277435266, + "grad_norm": 2.957755450148777, + "learning_rate": 3.4224567102884098e-06, + "loss": 0.4471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1787 + }, + { + "epoch": 2.2046855733662145, + "grad_norm": 2.727694808021836, + "learning_rate": 3.4125322999111698e-06, + "loss": 0.4985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1788 + }, + { + "epoch": 2.2059186189889024, + "grad_norm": 2.5078793281615646, + "learning_rate": 3.4026193386009244e-06, + "loss": 0.3459, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1789 + }, + { + "epoch": 2.2071516646115907, + "grad_norm": 2.4663933369954356, + "learning_rate": 3.392717843586472e-06, + "loss": 0.2747, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1790 + }, + { + "epoch": 2.2083847102342786, + "grad_norm": 2.336841544925504, + "learning_rate": 3.382827832076676e-06, + "loss": 0.3025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1791 + }, + { + "epoch": 2.2096177558569665, + "grad_norm": 2.5259006303278353, + "learning_rate": 3.3729493212604593e-06, + "loss": 0.3807, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1792 + }, + { + "epoch": 2.210850801479655, + "grad_norm": 2.4611507324991355, + "learning_rate": 3.363082328306737e-06, + "loss": 0.3556, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1793 + }, + { + "epoch": 2.2120838471023427, + "grad_norm": 2.6218201591114174, + "learning_rate": 3.353226870364422e-06, + "loss": 0.3614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1794 + }, + { + "epoch": 2.2133168927250306, + "grad_norm": 2.823653063437215, + "learning_rate": 3.343382964562368e-06, + "loss": 0.4235, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1795 + }, + { + "epoch": 2.214549938347719, + "grad_norm": 2.5370302947596524, + "learning_rate": 3.3335506280093544e-06, + "loss": 0.3716, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1796 + }, + { + "epoch": 2.215782983970407, + "grad_norm": 2.6056377036330183, + "learning_rate": 3.3237298777940586e-06, + "loss": 0.4282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1797 + }, + { + "epoch": 2.2170160295930947, + "grad_norm": 2.4707774591274014, + "learning_rate": 3.313920730985013e-06, + "loss": 0.3865, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1798 + }, + { + "epoch": 2.218249075215783, + "grad_norm": 2.6353955376812133, + "learning_rate": 3.304123204630587e-06, + "loss": 0.4042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1799 + }, + { + "epoch": 2.219482120838471, + "grad_norm": 2.9175053608530988, + "learning_rate": 3.29433731575895e-06, + "loss": 0.4431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1800 + }, + { + "epoch": 2.220715166461159, + "grad_norm": 2.4251950790112655, + "learning_rate": 3.2845630813780538e-06, + "loss": 0.3509, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1801 + }, + { + "epoch": 2.221948212083847, + "grad_norm": 2.6971411900024203, + "learning_rate": 3.274800518475586e-06, + "loss": 0.3719, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1802 + }, + { + "epoch": 2.223181257706535, + "grad_norm": 3.083984775257449, + "learning_rate": 3.2650496440189504e-06, + "loss": 0.4078, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1803 + }, + { + "epoch": 2.2244143033292234, + "grad_norm": 2.668941643126659, + "learning_rate": 3.255310474955243e-06, + "loss": 0.4439, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1804 + }, + { + "epoch": 2.2256473489519113, + "grad_norm": 2.976039030354018, + "learning_rate": 3.2455830282112087e-06, + "loss": 0.4573, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1805 + }, + { + "epoch": 2.226880394574599, + "grad_norm": 2.6167153610229126, + "learning_rate": 3.23586732069322e-06, + "loss": 0.4224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1806 + }, + { + "epoch": 2.228113440197287, + "grad_norm": 2.734527135614152, + "learning_rate": 3.226163369287245e-06, + "loss": 0.3709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1807 + }, + { + "epoch": 2.2293464858199754, + "grad_norm": 3.0042658435735454, + "learning_rate": 3.2164711908588253e-06, + "loss": 0.5491, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1808 + }, + { + "epoch": 2.2305795314426633, + "grad_norm": 2.7454305809971693, + "learning_rate": 3.2067908022530446e-06, + "loss": 0.4983, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1809 + }, + { + "epoch": 2.2318125770653516, + "grad_norm": 2.5164374780175436, + "learning_rate": 3.1971222202944752e-06, + "loss": 0.402, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1810 + }, + { + "epoch": 2.2330456226880395, + "grad_norm": 2.5784808191967095, + "learning_rate": 3.187465461787194e-06, + "loss": 0.3722, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1811 + }, + { + "epoch": 2.2342786683107274, + "grad_norm": 2.6856695561609323, + "learning_rate": 3.1778205435147093e-06, + "loss": 0.4127, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1812 + }, + { + "epoch": 2.2355117139334153, + "grad_norm": 2.6325754846210727, + "learning_rate": 3.1681874822399684e-06, + "loss": 0.3554, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1813 + }, + { + "epoch": 2.2367447595561036, + "grad_norm": 2.471040398086649, + "learning_rate": 3.1585662947052977e-06, + "loss": 0.4473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1814 + }, + { + "epoch": 2.2379778051787915, + "grad_norm": 2.4419144129307924, + "learning_rate": 3.1489569976323885e-06, + "loss": 0.3648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1815 + }, + { + "epoch": 2.23921085080148, + "grad_norm": 2.551971628620259, + "learning_rate": 3.1393596077222752e-06, + "loss": 0.3954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1816 + }, + { + "epoch": 2.2404438964241677, + "grad_norm": 2.7230280726188942, + "learning_rate": 3.1297741416552895e-06, + "loss": 0.4502, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1817 + }, + { + "epoch": 2.2416769420468556, + "grad_norm": 2.5986833977870774, + "learning_rate": 3.1202006160910403e-06, + "loss": 0.5191, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1818 + }, + { + "epoch": 2.242909987669544, + "grad_norm": 2.7125045113343034, + "learning_rate": 3.1106390476683833e-06, + "loss": 0.5319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1819 + }, + { + "epoch": 2.244143033292232, + "grad_norm": 2.4605911006490726, + "learning_rate": 3.101089453005399e-06, + "loss": 0.3073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1820 + }, + { + "epoch": 2.2453760789149197, + "grad_norm": 2.786082762868325, + "learning_rate": 3.09155184869935e-06, + "loss": 0.5374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1821 + }, + { + "epoch": 2.246609124537608, + "grad_norm": 3.0739075984030095, + "learning_rate": 3.08202625132666e-06, + "loss": 0.5128, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1822 + }, + { + "epoch": 2.247842170160296, + "grad_norm": 2.400418914937858, + "learning_rate": 3.0725126774428915e-06, + "loss": 0.3134, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1823 + }, + { + "epoch": 2.249075215782984, + "grad_norm": 2.45119216482281, + "learning_rate": 3.0630111435827015e-06, + "loss": 0.3013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1824 + }, + { + "epoch": 2.250308261405672, + "grad_norm": 2.9921352173355182, + "learning_rate": 3.053521666259832e-06, + "loss": 0.5281, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1825 + }, + { + "epoch": 2.25154130702836, + "grad_norm": 2.2410327144875986, + "learning_rate": 3.044044261967053e-06, + "loss": 0.2617, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1826 + }, + { + "epoch": 2.252774352651048, + "grad_norm": 2.6733315687315415, + "learning_rate": 3.0345789471761677e-06, + "loss": 0.4527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1827 + }, + { + "epoch": 2.2540073982737363, + "grad_norm": 2.9860637695259538, + "learning_rate": 3.0251257383379675e-06, + "loss": 0.5013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1828 + }, + { + "epoch": 2.255240443896424, + "grad_norm": 2.5671494331238476, + "learning_rate": 3.0156846518821925e-06, + "loss": 0.3735, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1829 + }, + { + "epoch": 2.256473489519112, + "grad_norm": 2.680154991425641, + "learning_rate": 3.006255704217522e-06, + "loss": 0.3478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1830 + }, + { + "epoch": 2.2577065351418004, + "grad_norm": 2.4302954673272894, + "learning_rate": 2.996838911731532e-06, + "loss": 0.2961, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1831 + }, + { + "epoch": 2.2589395807644883, + "grad_norm": 2.3839702852195344, + "learning_rate": 2.987434290790683e-06, + "loss": 0.2412, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1832 + }, + { + "epoch": 2.260172626387176, + "grad_norm": 2.3513764104654076, + "learning_rate": 2.9780418577402727e-06, + "loss": 0.4039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1833 + }, + { + "epoch": 2.2614056720098645, + "grad_norm": 2.5365870091486844, + "learning_rate": 2.9686616289044133e-06, + "loss": 0.4231, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1834 + }, + { + "epoch": 2.2626387176325524, + "grad_norm": 2.6943065218303346, + "learning_rate": 2.959293620586018e-06, + "loss": 0.419, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1835 + }, + { + "epoch": 2.2638717632552403, + "grad_norm": 2.67535698552442, + "learning_rate": 2.9499378490667486e-06, + "loss": 0.3214, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1836 + }, + { + "epoch": 2.2651048088779286, + "grad_norm": 2.648573827419777, + "learning_rate": 2.940594330607013e-06, + "loss": 0.3894, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1837 + }, + { + "epoch": 2.2663378545006165, + "grad_norm": 2.718965497393186, + "learning_rate": 2.931263081445902e-06, + "loss": 0.3538, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1838 + }, + { + "epoch": 2.2675709001233044, + "grad_norm": 2.8944057958849636, + "learning_rate": 2.921944117801203e-06, + "loss": 0.3672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1839 + }, + { + "epoch": 2.2688039457459928, + "grad_norm": 3.275410695882557, + "learning_rate": 2.9126374558693394e-06, + "loss": 0.506, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1840 + }, + { + "epoch": 2.2700369913686806, + "grad_norm": 2.794818093317088, + "learning_rate": 2.903343111825362e-06, + "loss": 0.4013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1841 + }, + { + "epoch": 2.2712700369913685, + "grad_norm": 2.8408389229850255, + "learning_rate": 2.894061101822905e-06, + "loss": 0.4814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1842 + }, + { + "epoch": 2.272503082614057, + "grad_norm": 2.8213904379609707, + "learning_rate": 2.884791441994168e-06, + "loss": 0.4325, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1843 + }, + { + "epoch": 2.2737361282367448, + "grad_norm": 2.318478307851654, + "learning_rate": 2.8755341484498923e-06, + "loss": 0.2796, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1844 + }, + { + "epoch": 2.2749691738594326, + "grad_norm": 2.7101825130130712, + "learning_rate": 2.8662892372793203e-06, + "loss": 0.368, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1845 + }, + { + "epoch": 2.276202219482121, + "grad_norm": 2.232410112947641, + "learning_rate": 2.8570567245501736e-06, + "loss": 0.2401, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1846 + }, + { + "epoch": 2.277435265104809, + "grad_norm": 2.7730682747492046, + "learning_rate": 2.8478366263086242e-06, + "loss": 0.458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1847 + }, + { + "epoch": 2.2786683107274968, + "grad_norm": 2.7088179960724026, + "learning_rate": 2.838628958579277e-06, + "loss": 0.3906, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1848 + }, + { + "epoch": 2.279901356350185, + "grad_norm": 2.642992601836881, + "learning_rate": 2.8294337373651227e-06, + "loss": 0.4505, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1849 + }, + { + "epoch": 2.281134401972873, + "grad_norm": 2.566770724267774, + "learning_rate": 2.8202509786475183e-06, + "loss": 0.4041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1850 + }, + { + "epoch": 2.282367447595561, + "grad_norm": 2.8325370719966334, + "learning_rate": 2.8110806983861726e-06, + "loss": 0.3838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1851 + }, + { + "epoch": 2.283600493218249, + "grad_norm": 2.7305618513803034, + "learning_rate": 2.801922912519098e-06, + "loss": 0.4741, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1852 + }, + { + "epoch": 2.284833538840937, + "grad_norm": 2.7167758355736717, + "learning_rate": 2.7927776369625915e-06, + "loss": 0.4109, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1853 + }, + { + "epoch": 2.286066584463625, + "grad_norm": 2.1944056474153566, + "learning_rate": 2.7836448876112065e-06, + "loss": 0.2787, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1854 + }, + { + "epoch": 2.2872996300863133, + "grad_norm": 2.206944335103158, + "learning_rate": 2.7745246803377303e-06, + "loss": 0.3027, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1855 + }, + { + "epoch": 2.288532675709001, + "grad_norm": 2.3467437676570677, + "learning_rate": 2.7654170309931537e-06, + "loss": 0.3316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1856 + }, + { + "epoch": 2.289765721331689, + "grad_norm": 2.8197181669039253, + "learning_rate": 2.7563219554066346e-06, + "loss": 0.4353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1857 + }, + { + "epoch": 2.2909987669543774, + "grad_norm": 2.6463694243329297, + "learning_rate": 2.747239469385481e-06, + "loss": 0.4461, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1858 + }, + { + "epoch": 2.2922318125770653, + "grad_norm": 2.5693847166016495, + "learning_rate": 2.738169588715116e-06, + "loss": 0.4154, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1859 + }, + { + "epoch": 2.293464858199753, + "grad_norm": 2.9084887582207224, + "learning_rate": 2.729112329159066e-06, + "loss": 0.4642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1860 + }, + { + "epoch": 2.2946979038224415, + "grad_norm": 2.661155216526391, + "learning_rate": 2.720067706458911e-06, + "loss": 0.4911, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1861 + }, + { + "epoch": 2.2959309494451294, + "grad_norm": 2.249438410721065, + "learning_rate": 2.7110357363342667e-06, + "loss": 0.2955, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1862 + }, + { + "epoch": 2.2971639950678173, + "grad_norm": 2.4248296466556605, + "learning_rate": 2.7020164344827705e-06, + "loss": 0.3078, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1863 + }, + { + "epoch": 2.2983970406905057, + "grad_norm": 2.8178210324627604, + "learning_rate": 2.6930098165800324e-06, + "loss": 0.5657, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1864 + }, + { + "epoch": 2.2996300863131935, + "grad_norm": 2.1492514900867867, + "learning_rate": 2.684015898279618e-06, + "loss": 0.2542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1865 + }, + { + "epoch": 2.3008631319358814, + "grad_norm": 2.5767111705244097, + "learning_rate": 2.6750346952130224e-06, + "loss": 0.2911, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1866 + }, + { + "epoch": 2.3020961775585698, + "grad_norm": 2.243951773213573, + "learning_rate": 2.666066222989644e-06, + "loss": 0.2932, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1867 + }, + { + "epoch": 2.3033292231812577, + "grad_norm": 2.3067727425907587, + "learning_rate": 2.65711049719676e-06, + "loss": 0.2374, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1868 + }, + { + "epoch": 2.304562268803946, + "grad_norm": 2.415737071109258, + "learning_rate": 2.6481675333994728e-06, + "loss": 0.3661, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1869 + }, + { + "epoch": 2.305795314426634, + "grad_norm": 2.296552712112862, + "learning_rate": 2.639237347140731e-06, + "loss": 0.2565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1870 + }, + { + "epoch": 2.3070283600493218, + "grad_norm": 2.5905345710770766, + "learning_rate": 2.630319953941255e-06, + "loss": 0.2994, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1871 + }, + { + "epoch": 2.3082614056720097, + "grad_norm": 2.9421711380538755, + "learning_rate": 2.6214153692995504e-06, + "loss": 0.3917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1872 + }, + { + "epoch": 2.309494451294698, + "grad_norm": 2.24918012711347, + "learning_rate": 2.612523608691837e-06, + "loss": 0.29, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1873 + }, + { + "epoch": 2.310727496917386, + "grad_norm": 2.6468263977498396, + "learning_rate": 2.6036446875720657e-06, + "loss": 0.3579, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1874 + }, + { + "epoch": 2.311960542540074, + "grad_norm": 2.9869717640588616, + "learning_rate": 2.5947786213718695e-06, + "loss": 0.4606, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1875 + }, + { + "epoch": 2.313193588162762, + "grad_norm": 2.7765364835681083, + "learning_rate": 2.585925425500534e-06, + "loss": 0.4072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1876 + }, + { + "epoch": 2.31442663378545, + "grad_norm": 2.5297254099675426, + "learning_rate": 2.5770851153449785e-06, + "loss": 0.3614, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1877 + }, + { + "epoch": 2.315659679408138, + "grad_norm": 2.7275298517849924, + "learning_rate": 2.5682577062697213e-06, + "loss": 0.4544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1878 + }, + { + "epoch": 2.316892725030826, + "grad_norm": 3.1515080991733235, + "learning_rate": 2.559443213616871e-06, + "loss": 0.49, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1879 + }, + { + "epoch": 2.318125770653514, + "grad_norm": 3.1665702629969315, + "learning_rate": 2.5506416527060796e-06, + "loss": 0.5849, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1880 + }, + { + "epoch": 2.3193588162762024, + "grad_norm": 2.5623146927718254, + "learning_rate": 2.5418530388345175e-06, + "loss": 0.3085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1881 + }, + { + "epoch": 2.3205918618988903, + "grad_norm": 2.46776926089478, + "learning_rate": 2.533077387276869e-06, + "loss": 0.2888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1882 + }, + { + "epoch": 2.321824907521578, + "grad_norm": 2.775069601858482, + "learning_rate": 2.524314713285272e-06, + "loss": 0.3817, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1883 + }, + { + "epoch": 2.323057953144266, + "grad_norm": 2.670954218210571, + "learning_rate": 2.515565032089329e-06, + "loss": 0.4478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1884 + }, + { + "epoch": 2.3242909987669544, + "grad_norm": 2.5238477153530665, + "learning_rate": 2.5068283588960372e-06, + "loss": 0.4481, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1885 + }, + { + "epoch": 2.3255240443896423, + "grad_norm": 2.6849323631816295, + "learning_rate": 2.498104708889807e-06, + "loss": 0.4133, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1886 + }, + { + "epoch": 2.3267570900123307, + "grad_norm": 2.4377081666404075, + "learning_rate": 2.4893940972323995e-06, + "loss": 0.332, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1887 + }, + { + "epoch": 2.3279901356350186, + "grad_norm": 2.7114394525033063, + "learning_rate": 2.480696539062928e-06, + "loss": 0.4282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1888 + }, + { + "epoch": 2.3292231812577064, + "grad_norm": 2.603372665925758, + "learning_rate": 2.4720120494978095e-06, + "loss": 0.3697, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1889 + }, + { + "epoch": 2.3304562268803943, + "grad_norm": 2.805458776057501, + "learning_rate": 2.4633406436307473e-06, + "loss": 0.326, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1890 + }, + { + "epoch": 2.3316892725030827, + "grad_norm": 3.008639998561586, + "learning_rate": 2.4546823365327133e-06, + "loss": 0.479, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1891 + }, + { + "epoch": 2.3329223181257706, + "grad_norm": 2.848060710470996, + "learning_rate": 2.446037143251907e-06, + "loss": 0.4492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1892 + }, + { + "epoch": 2.334155363748459, + "grad_norm": 2.6972361605525172, + "learning_rate": 2.437405078813733e-06, + "loss": 0.4651, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1893 + }, + { + "epoch": 2.335388409371147, + "grad_norm": 2.6444772328432764, + "learning_rate": 2.428786158220787e-06, + "loss": 0.3531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1894 + }, + { + "epoch": 2.3366214549938347, + "grad_norm": 2.825025155857256, + "learning_rate": 2.420180396452817e-06, + "loss": 0.4398, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1895 + }, + { + "epoch": 2.337854500616523, + "grad_norm": 2.847774141381515, + "learning_rate": 2.4115878084666967e-06, + "loss": 0.4774, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1896 + }, + { + "epoch": 2.339087546239211, + "grad_norm": 2.7240100074211626, + "learning_rate": 2.4030084091964058e-06, + "loss": 0.5387, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1897 + }, + { + "epoch": 2.340320591861899, + "grad_norm": 2.6596947077013238, + "learning_rate": 2.3944422135530067e-06, + "loss": 0.3699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1898 + }, + { + "epoch": 2.341553637484587, + "grad_norm": 2.3195986093996828, + "learning_rate": 2.385889236424609e-06, + "loss": 0.2357, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1899 + }, + { + "epoch": 2.342786683107275, + "grad_norm": 3.138152832229155, + "learning_rate": 2.3773494926763453e-06, + "loss": 0.5716, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1900 + }, + { + "epoch": 2.344019728729963, + "grad_norm": 2.4353439616422317, + "learning_rate": 2.3688229971503585e-06, + "loss": 0.307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1901 + }, + { + "epoch": 2.3452527743526512, + "grad_norm": 2.7479264877441305, + "learning_rate": 2.3603097646657547e-06, + "loss": 0.4065, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1902 + }, + { + "epoch": 2.346485819975339, + "grad_norm": 2.7608377722996464, + "learning_rate": 2.3518098100186003e-06, + "loss": 0.4474, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1903 + }, + { + "epoch": 2.347718865598027, + "grad_norm": 2.8009720711063983, + "learning_rate": 2.3433231479818754e-06, + "loss": 0.3542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1904 + }, + { + "epoch": 2.3489519112207153, + "grad_norm": 2.6450074204233727, + "learning_rate": 2.3348497933054593e-06, + "loss": 0.2626, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1905 + }, + { + "epoch": 2.3501849568434032, + "grad_norm": 2.718286299118251, + "learning_rate": 2.3263897607161045e-06, + "loss": 0.3783, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1906 + }, + { + "epoch": 2.351418002466091, + "grad_norm": 3.1635623479202004, + "learning_rate": 2.317943064917412e-06, + "loss": 0.458, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1907 + }, + { + "epoch": 2.3526510480887795, + "grad_norm": 3.023923451231218, + "learning_rate": 2.3095097205898e-06, + "loss": 0.3445, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1908 + }, + { + "epoch": 2.3538840937114673, + "grad_norm": 2.7560067491256754, + "learning_rate": 2.301089742390479e-06, + "loss": 0.3877, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1909 + }, + { + "epoch": 2.3551171393341552, + "grad_norm": 2.4987053182394185, + "learning_rate": 2.292683144953438e-06, + "loss": 0.2485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1910 + }, + { + "epoch": 2.3563501849568436, + "grad_norm": 3.028553166873761, + "learning_rate": 2.284289942889403e-06, + "loss": 0.5015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1911 + }, + { + "epoch": 2.3575832305795315, + "grad_norm": 2.6107906939757033, + "learning_rate": 2.275910150785818e-06, + "loss": 0.356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1912 + }, + { + "epoch": 2.3588162762022193, + "grad_norm": 2.4846901180876872, + "learning_rate": 2.267543783206824e-06, + "loss": 0.2767, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1913 + }, + { + "epoch": 2.3600493218249077, + "grad_norm": 2.924247536298395, + "learning_rate": 2.2591908546932283e-06, + "loss": 0.4083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1914 + }, + { + "epoch": 2.3612823674475956, + "grad_norm": 2.715130315338188, + "learning_rate": 2.2508513797624887e-06, + "loss": 0.4079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1915 + }, + { + "epoch": 2.3625154130702835, + "grad_norm": 2.261533556908912, + "learning_rate": 2.2425253729086625e-06, + "loss": 0.2094, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1916 + }, + { + "epoch": 2.363748458692972, + "grad_norm": 2.9340455954588918, + "learning_rate": 2.2342128486024193e-06, + "loss": 0.5613, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1917 + }, + { + "epoch": 2.3649815043156597, + "grad_norm": 2.759033032828826, + "learning_rate": 2.22591382129098e-06, + "loss": 0.4658, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1918 + }, + { + "epoch": 2.3662145499383476, + "grad_norm": 2.9330784187057812, + "learning_rate": 2.217628305398122e-06, + "loss": 0.4703, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1919 + }, + { + "epoch": 2.367447595561036, + "grad_norm": 2.926964264297683, + "learning_rate": 2.209356315324128e-06, + "loss": 0.3734, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1920 + }, + { + "epoch": 2.368680641183724, + "grad_norm": 2.7140676889542674, + "learning_rate": 2.201097865445776e-06, + "loss": 0.4358, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1921 + }, + { + "epoch": 2.3699136868064117, + "grad_norm": 2.3950079044896504, + "learning_rate": 2.192852970116316e-06, + "loss": 0.3264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1922 + }, + { + "epoch": 2.3711467324291, + "grad_norm": 2.4698900190893847, + "learning_rate": 2.1846216436654342e-06, + "loss": 0.2551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1923 + }, + { + "epoch": 2.372379778051788, + "grad_norm": 2.57908607952626, + "learning_rate": 2.1764039003992377e-06, + "loss": 0.2718, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1924 + }, + { + "epoch": 2.373612823674476, + "grad_norm": 2.7443866869071747, + "learning_rate": 2.16819975460022e-06, + "loss": 0.3698, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1925 + }, + { + "epoch": 2.374845869297164, + "grad_norm": 2.4541746426775424, + "learning_rate": 2.1600092205272493e-06, + "loss": 0.3114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1926 + }, + { + "epoch": 2.376078914919852, + "grad_norm": 2.6764807552617094, + "learning_rate": 2.1518323124155393e-06, + "loss": 0.3946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1927 + }, + { + "epoch": 2.37731196054254, + "grad_norm": 2.232502294129531, + "learning_rate": 2.143669044476605e-06, + "loss": 0.2064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1928 + }, + { + "epoch": 2.3785450061652282, + "grad_norm": 2.465612079035968, + "learning_rate": 2.135519430898274e-06, + "loss": 0.2784, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1929 + }, + { + "epoch": 2.379778051787916, + "grad_norm": 2.827852473582675, + "learning_rate": 2.1273834858446285e-06, + "loss": 0.3079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1930 + }, + { + "epoch": 2.381011097410604, + "grad_norm": 2.633449394986566, + "learning_rate": 2.1192612234560093e-06, + "loss": 0.3227, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1931 + }, + { + "epoch": 2.3822441430332923, + "grad_norm": 2.7760612390178325, + "learning_rate": 2.1111526578489573e-06, + "loss": 0.3875, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1932 + }, + { + "epoch": 2.3834771886559802, + "grad_norm": 2.543518865851238, + "learning_rate": 2.103057803116223e-06, + "loss": 0.2956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1933 + }, + { + "epoch": 2.384710234278668, + "grad_norm": 2.5903161811570716, + "learning_rate": 2.0949766733267263e-06, + "loss": 0.3226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1934 + }, + { + "epoch": 2.3859432799013565, + "grad_norm": 2.6196746717343307, + "learning_rate": 2.0869092825255266e-06, + "loss": 0.3737, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1935 + }, + { + "epoch": 2.3871763255240444, + "grad_norm": 2.4200712742066726, + "learning_rate": 2.078855644733808e-06, + "loss": 0.4331, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1936 + }, + { + "epoch": 2.3884093711467322, + "grad_norm": 2.97992031816307, + "learning_rate": 2.070815773948851e-06, + "loss": 0.4552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1937 + }, + { + "epoch": 2.3896424167694206, + "grad_norm": 2.868889423356795, + "learning_rate": 2.062789684144013e-06, + "loss": 0.3954, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1938 + }, + { + "epoch": 2.3908754623921085, + "grad_norm": 2.587694442487643, + "learning_rate": 2.054777389268695e-06, + "loss": 0.4605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1939 + }, + { + "epoch": 2.392108508014797, + "grad_norm": 2.8211137836092632, + "learning_rate": 2.046778903248321e-06, + "loss": 0.3973, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1940 + }, + { + "epoch": 2.3933415536374847, + "grad_norm": 2.468677079748375, + "learning_rate": 2.038794239984324e-06, + "loss": 0.3423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1941 + }, + { + "epoch": 2.3945745992601726, + "grad_norm": 2.5859956743331343, + "learning_rate": 2.0308234133541036e-06, + "loss": 0.3282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1942 + }, + { + "epoch": 2.3958076448828605, + "grad_norm": 2.5722903261889027, + "learning_rate": 2.0228664372110152e-06, + "loss": 0.3676, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1943 + }, + { + "epoch": 2.397040690505549, + "grad_norm": 2.56754893044426, + "learning_rate": 2.0149233253843393e-06, + "loss": 0.3648, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1944 + }, + { + "epoch": 2.3982737361282367, + "grad_norm": 2.5487534871682835, + "learning_rate": 2.0069940916792665e-06, + "loss": 0.3583, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1945 + }, + { + "epoch": 2.399506781750925, + "grad_norm": 2.765885050624665, + "learning_rate": 1.999078749876858e-06, + "loss": 0.3466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1946 + }, + { + "epoch": 2.400739827373613, + "grad_norm": 2.8412723371615347, + "learning_rate": 1.9911773137340416e-06, + "loss": 0.4694, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1947 + }, + { + "epoch": 2.401972872996301, + "grad_norm": 2.7637883479587417, + "learning_rate": 1.9832897969835687e-06, + "loss": 0.3952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1948 + }, + { + "epoch": 2.4032059186189887, + "grad_norm": 2.466586812071735, + "learning_rate": 1.975416213333997e-06, + "loss": 0.3669, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1949 + }, + { + "epoch": 2.404438964241677, + "grad_norm": 2.7661522361602904, + "learning_rate": 1.967556576469678e-06, + "loss": 0.3284, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1950 + }, + { + "epoch": 2.405672009864365, + "grad_norm": 2.5114022421240056, + "learning_rate": 1.9597109000507163e-06, + "loss": 0.3843, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1951 + }, + { + "epoch": 2.4069050554870532, + "grad_norm": 2.6057762001360336, + "learning_rate": 1.9518791977129537e-06, + "loss": 0.3851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1952 + }, + { + "epoch": 2.408138101109741, + "grad_norm": 2.415051279016423, + "learning_rate": 1.944061483067945e-06, + "loss": 0.3463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1953 + }, + { + "epoch": 2.409371146732429, + "grad_norm": 2.7762579065761734, + "learning_rate": 1.9362577697029383e-06, + "loss": 0.3868, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1954 + }, + { + "epoch": 2.410604192355117, + "grad_norm": 3.063713670502854, + "learning_rate": 1.9284680711808447e-06, + "loss": 0.3809, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1955 + }, + { + "epoch": 2.4118372379778052, + "grad_norm": 2.7071449186380336, + "learning_rate": 1.9206924010402116e-06, + "loss": 0.3126, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1956 + }, + { + "epoch": 2.413070283600493, + "grad_norm": 2.99690461715501, + "learning_rate": 1.9129307727952184e-06, + "loss": 0.3971, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1957 + }, + { + "epoch": 2.4143033292231815, + "grad_norm": 2.538996577415216, + "learning_rate": 1.9051831999356274e-06, + "loss": 0.308, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1958 + }, + { + "epoch": 2.4155363748458694, + "grad_norm": 2.598169922298, + "learning_rate": 1.897449695926774e-06, + "loss": 0.3779, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1959 + }, + { + "epoch": 2.4167694204685573, + "grad_norm": 2.785021708374361, + "learning_rate": 1.8897302742095513e-06, + "loss": 0.4517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1960 + }, + { + "epoch": 2.418002466091245, + "grad_norm": 2.755401413560113, + "learning_rate": 1.8820249482003638e-06, + "loss": 0.4856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1961 + }, + { + "epoch": 2.4192355117139335, + "grad_norm": 2.596435825974963, + "learning_rate": 1.8743337312911335e-06, + "loss": 0.353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1962 + }, + { + "epoch": 2.4204685573366214, + "grad_norm": 2.786096288352132, + "learning_rate": 1.866656636849241e-06, + "loss": 0.4245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1963 + }, + { + "epoch": 2.4217016029593097, + "grad_norm": 2.439658478811596, + "learning_rate": 1.85899367821754e-06, + "loss": 0.2639, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1964 + }, + { + "epoch": 2.4229346485819976, + "grad_norm": 2.544176056739862, + "learning_rate": 1.8513448687143032e-06, + "loss": 0.3492, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1965 + }, + { + "epoch": 2.4241676942046855, + "grad_norm": 3.333459945880387, + "learning_rate": 1.8437102216332236e-06, + "loss": 0.6163, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1966 + }, + { + "epoch": 2.4254007398273734, + "grad_norm": 2.419327956697947, + "learning_rate": 1.8360897502433695e-06, + "loss": 0.209, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1967 + }, + { + "epoch": 2.4266337854500617, + "grad_norm": 2.570872200928867, + "learning_rate": 1.8284834677891749e-06, + "loss": 0.3493, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1968 + }, + { + "epoch": 2.4278668310727496, + "grad_norm": 2.6222413870421786, + "learning_rate": 1.8208913874904177e-06, + "loss": 0.3, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1969 + }, + { + "epoch": 2.429099876695438, + "grad_norm": 2.6311595401436976, + "learning_rate": 1.8133135225421873e-06, + "loss": 0.3072, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1970 + }, + { + "epoch": 2.430332922318126, + "grad_norm": 2.5866942216903235, + "learning_rate": 1.8057498861148691e-06, + "loss": 0.2551, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1971 + }, + { + "epoch": 2.4315659679408137, + "grad_norm": 2.868334596003309, + "learning_rate": 1.798200491354114e-06, + "loss": 0.3534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1972 + }, + { + "epoch": 2.432799013563502, + "grad_norm": 2.636368307683731, + "learning_rate": 1.7906653513808303e-06, + "loss": 0.3249, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1973 + }, + { + "epoch": 2.43403205918619, + "grad_norm": 2.2980539591250118, + "learning_rate": 1.783144479291149e-06, + "loss": 0.3174, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1974 + }, + { + "epoch": 2.435265104808878, + "grad_norm": 2.7793989257875253, + "learning_rate": 1.7756378881563917e-06, + "loss": 0.4124, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1975 + }, + { + "epoch": 2.436498150431566, + "grad_norm": 2.3284324040289186, + "learning_rate": 1.768145591023076e-06, + "loss": 0.2354, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1976 + }, + { + "epoch": 2.437731196054254, + "grad_norm": 2.7961580556831405, + "learning_rate": 1.7606676009128643e-06, + "loss": 0.3147, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1977 + }, + { + "epoch": 2.438964241676942, + "grad_norm": 2.78394209814451, + "learning_rate": 1.7532039308225656e-06, + "loss": 0.4261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1978 + }, + { + "epoch": 2.4401972872996303, + "grad_norm": 2.7644595713287634, + "learning_rate": 1.7457545937240827e-06, + "loss": 0.3966, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1979 + }, + { + "epoch": 2.441430332922318, + "grad_norm": 2.6754626629785148, + "learning_rate": 1.7383196025644233e-06, + "loss": 0.3258, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1980 + }, + { + "epoch": 2.442663378545006, + "grad_norm": 2.5894501951666817, + "learning_rate": 1.7308989702656586e-06, + "loss": 0.3238, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1981 + }, + { + "epoch": 2.4438964241676944, + "grad_norm": 2.8255899965176385, + "learning_rate": 1.7234927097249e-06, + "loss": 0.3219, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1982 + }, + { + "epoch": 2.4451294697903823, + "grad_norm": 2.957389703887107, + "learning_rate": 1.7161008338142804e-06, + "loss": 0.4042, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1983 + }, + { + "epoch": 2.44636251541307, + "grad_norm": 2.7253630268917317, + "learning_rate": 1.7087233553809335e-06, + "loss": 0.351, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1984 + }, + { + "epoch": 2.4475955610357585, + "grad_norm": 2.6055125133787387, + "learning_rate": 1.7013602872469749e-06, + "loss": 0.3155, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1985 + }, + { + "epoch": 2.4488286066584464, + "grad_norm": 2.9002281578070526, + "learning_rate": 1.6940116422094688e-06, + "loss": 0.3765, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1986 + }, + { + "epoch": 2.4500616522811343, + "grad_norm": 2.8603120212321222, + "learning_rate": 1.6866774330404112e-06, + "loss": 0.485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1987 + }, + { + "epoch": 2.4512946979038226, + "grad_norm": 2.750919681788328, + "learning_rate": 1.6793576724867167e-06, + "loss": 0.3956, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1988 + }, + { + "epoch": 2.4525277435265105, + "grad_norm": 2.5911381734037167, + "learning_rate": 1.6720523732701787e-06, + "loss": 0.2877, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1989 + }, + { + "epoch": 2.4537607891491984, + "grad_norm": 2.711257130090504, + "learning_rate": 1.6647615480874623e-06, + "loss": 0.3308, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1990 + }, + { + "epoch": 2.4549938347718867, + "grad_norm": 2.824061791832042, + "learning_rate": 1.6574852096100723e-06, + "loss": 0.4287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1991 + }, + { + "epoch": 2.4562268803945746, + "grad_norm": 2.887945699620393, + "learning_rate": 1.6502233704843395e-06, + "loss": 0.3861, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1992 + }, + { + "epoch": 2.4574599260172625, + "grad_norm": 2.6357336450913955, + "learning_rate": 1.6429760433313957e-06, + "loss": 0.3608, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1993 + }, + { + "epoch": 2.458692971639951, + "grad_norm": 2.672435780241681, + "learning_rate": 1.6357432407471474e-06, + "loss": 0.4074, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1994 + }, + { + "epoch": 2.4599260172626387, + "grad_norm": 2.359546711744168, + "learning_rate": 1.628524975302258e-06, + "loss": 0.3301, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1995 + }, + { + "epoch": 2.4611590628853266, + "grad_norm": 2.663889849020532, + "learning_rate": 1.6213212595421225e-06, + "loss": 0.261, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1996 + }, + { + "epoch": 2.462392108508015, + "grad_norm": 2.3712367457855597, + "learning_rate": 1.6141321059868565e-06, + "loss": 0.206, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1997 + }, + { + "epoch": 2.463625154130703, + "grad_norm": 2.3333104602294545, + "learning_rate": 1.6069575271312588e-06, + "loss": 0.2662, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1998 + }, + { + "epoch": 2.4648581997533907, + "grad_norm": 2.3951475290329673, + "learning_rate": 1.5997975354447981e-06, + "loss": 0.3517, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 1999 + }, + { + "epoch": 2.466091245376079, + "grad_norm": 2.547897971468787, + "learning_rate": 1.592652143371597e-06, + "loss": 0.3334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2000 + }, + { + "epoch": 2.467324290998767, + "grad_norm": 2.4595176510202954, + "learning_rate": 1.5855213633303956e-06, + "loss": 0.3423, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2001 + }, + { + "epoch": 2.468557336621455, + "grad_norm": 2.9582915458097245, + "learning_rate": 1.578405207714543e-06, + "loss": 0.3455, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2002 + }, + { + "epoch": 2.469790382244143, + "grad_norm": 2.428673296135474, + "learning_rate": 1.5713036888919675e-06, + "loss": 0.2992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2003 + }, + { + "epoch": 2.471023427866831, + "grad_norm": 2.743398790088527, + "learning_rate": 1.5642168192051644e-06, + "loss": 0.3322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2004 + }, + { + "epoch": 2.472256473489519, + "grad_norm": 3.324065459022819, + "learning_rate": 1.557144610971163e-06, + "loss": 0.4465, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2005 + }, + { + "epoch": 2.4734895191122073, + "grad_norm": 2.162373213734668, + "learning_rate": 1.5500870764815102e-06, + "loss": 0.2274, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2006 + }, + { + "epoch": 2.474722564734895, + "grad_norm": 2.6083953204344366, + "learning_rate": 1.5430442280022584e-06, + "loss": 0.3268, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2007 + }, + { + "epoch": 2.475955610357583, + "grad_norm": 2.1764316712749787, + "learning_rate": 1.536016077773923e-06, + "loss": 0.1526, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2008 + }, + { + "epoch": 2.4771886559802714, + "grad_norm": 2.704429097110481, + "learning_rate": 1.5290026380114876e-06, + "loss": 0.2536, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2009 + }, + { + "epoch": 2.4784217016029593, + "grad_norm": 3.0958298333409857, + "learning_rate": 1.522003920904358e-06, + "loss": 0.4273, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2010 + }, + { + "epoch": 2.479654747225647, + "grad_norm": 2.9092939112809577, + "learning_rate": 1.5150199386163543e-06, + "loss": 0.3945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2011 + }, + { + "epoch": 2.4808877928483355, + "grad_norm": 3.116797881391071, + "learning_rate": 1.5080507032856894e-06, + "loss": 0.4497, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2012 + }, + { + "epoch": 2.4821208384710234, + "grad_norm": 2.760889317748852, + "learning_rate": 1.5010962270249486e-06, + "loss": 0.2909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2013 + }, + { + "epoch": 2.4833538840937113, + "grad_norm": 2.6074546180794833, + "learning_rate": 1.4941565219210585e-06, + "loss": 0.2939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2014 + }, + { + "epoch": 2.4845869297163996, + "grad_norm": 3.0837456899974827, + "learning_rate": 1.4872316000352783e-06, + "loss": 0.4267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2015 + }, + { + "epoch": 2.4858199753390875, + "grad_norm": 2.925304791109893, + "learning_rate": 1.4803214734031734e-06, + "loss": 0.3992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2016 + }, + { + "epoch": 2.487053020961776, + "grad_norm": 2.873066706251663, + "learning_rate": 1.4734261540345951e-06, + "loss": 0.3851, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2017 + }, + { + "epoch": 2.4882860665844637, + "grad_norm": 3.0170897183556535, + "learning_rate": 1.466545653913659e-06, + "loss": 0.3999, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2018 + }, + { + "epoch": 2.4895191122071516, + "grad_norm": 2.8171232756409337, + "learning_rate": 1.4596799849987198e-06, + "loss": 0.409, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2019 + }, + { + "epoch": 2.4907521578298395, + "grad_norm": 3.1825430295716157, + "learning_rate": 1.4528291592223642e-06, + "loss": 0.4256, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2020 + }, + { + "epoch": 2.491985203452528, + "grad_norm": 2.7388479608941725, + "learning_rate": 1.4459931884913825e-06, + "loss": 0.3645, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2021 + }, + { + "epoch": 2.4932182490752157, + "grad_norm": 2.7221846310700437, + "learning_rate": 1.4391720846867308e-06, + "loss": 0.3802, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2022 + }, + { + "epoch": 2.494451294697904, + "grad_norm": 2.794004204773263, + "learning_rate": 1.4323658596635459e-06, + "loss": 0.3532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2023 + }, + { + "epoch": 2.495684340320592, + "grad_norm": 2.9987018690091087, + "learning_rate": 1.425574525251092e-06, + "loss": 0.4432, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2024 + }, + { + "epoch": 2.49691738594328, + "grad_norm": 2.6488734906053693, + "learning_rate": 1.4187980932527612e-06, + "loss": 0.3241, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2025 + }, + { + "epoch": 2.4981504315659677, + "grad_norm": 2.433220382902042, + "learning_rate": 1.4120365754460408e-06, + "loss": 0.2895, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2026 + }, + { + "epoch": 2.499383477188656, + "grad_norm": 2.434542396902984, + "learning_rate": 1.4052899835824951e-06, + "loss": 0.2108, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2027 + }, + { + "epoch": 2.500616522811344, + "grad_norm": 2.524311609378393, + "learning_rate": 1.3985583293877547e-06, + "loss": 0.3472, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2028 + }, + { + "epoch": 2.5018495684340323, + "grad_norm": 3.1734464121061396, + "learning_rate": 1.3918416245614818e-06, + "loss": 0.5071, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2029 + }, + { + "epoch": 2.50308261405672, + "grad_norm": 2.576082693205846, + "learning_rate": 1.3851398807773587e-06, + "loss": 0.3994, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2030 + }, + { + "epoch": 2.504315659679408, + "grad_norm": 2.3758774932628968, + "learning_rate": 1.3784531096830623e-06, + "loss": 0.3173, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2031 + }, + { + "epoch": 2.505548705302096, + "grad_norm": 2.7828574834012247, + "learning_rate": 1.371781322900254e-06, + "loss": 0.3812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2032 + }, + { + "epoch": 2.5067817509247843, + "grad_norm": 2.347396739269817, + "learning_rate": 1.3651245320245465e-06, + "loss": 0.2099, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2033 + }, + { + "epoch": 2.508014796547472, + "grad_norm": 2.9121677286050884, + "learning_rate": 1.3584827486254882e-06, + "loss": 0.3686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2034 + }, + { + "epoch": 2.5092478421701605, + "grad_norm": 2.501307650986237, + "learning_rate": 1.3518559842465507e-06, + "loss": 0.3011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2035 + }, + { + "epoch": 2.5104808877928484, + "grad_norm": 2.731824672887973, + "learning_rate": 1.3452442504050967e-06, + "loss": 0.3419, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2036 + }, + { + "epoch": 2.5117139334155363, + "grad_norm": 2.688214320582552, + "learning_rate": 1.338647558592372e-06, + "loss": 0.3189, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2037 + }, + { + "epoch": 2.512946979038224, + "grad_norm": 2.046493401656216, + "learning_rate": 1.3320659202734688e-06, + "loss": 0.1985, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2038 + }, + { + "epoch": 2.5141800246609125, + "grad_norm": 2.5689432212429772, + "learning_rate": 1.3254993468873244e-06, + "loss": 0.3313, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2039 + }, + { + "epoch": 2.5154130702836004, + "grad_norm": 3.0168905879150123, + "learning_rate": 1.3189478498466967e-06, + "loss": 0.4057, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2040 + }, + { + "epoch": 2.5166461159062887, + "grad_norm": 2.712840369505315, + "learning_rate": 1.3124114405381305e-06, + "loss": 0.2649, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2041 + }, + { + "epoch": 2.5178791615289766, + "grad_norm": 2.4860407840994685, + "learning_rate": 1.305890130321954e-06, + "loss": 0.3237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2042 + }, + { + "epoch": 2.5191122071516645, + "grad_norm": 2.761535488030773, + "learning_rate": 1.2993839305322498e-06, + "loss": 0.336, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2043 + }, + { + "epoch": 2.5203452527743524, + "grad_norm": 2.7684583540824117, + "learning_rate": 1.2928928524768447e-06, + "loss": 0.3867, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2044 + }, + { + "epoch": 2.5215782983970407, + "grad_norm": 2.764895691572842, + "learning_rate": 1.286416907437278e-06, + "loss": 0.2929, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2045 + }, + { + "epoch": 2.5228113440197286, + "grad_norm": 2.4268329926610592, + "learning_rate": 1.2799561066687872e-06, + "loss": 0.2089, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2046 + }, + { + "epoch": 2.524044389642417, + "grad_norm": 2.649854219984885, + "learning_rate": 1.2735104614002958e-06, + "loss": 0.3289, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2047 + }, + { + "epoch": 2.525277435265105, + "grad_norm": 2.713889242792587, + "learning_rate": 1.2670799828343795e-06, + "loss": 0.3292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2048 + }, + { + "epoch": 2.5265104808877927, + "grad_norm": 2.5217008951125988, + "learning_rate": 1.260664682147258e-06, + "loss": 0.2672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2049 + }, + { + "epoch": 2.5277435265104806, + "grad_norm": 2.62298635633002, + "learning_rate": 1.2542645704887679e-06, + "loss": 0.3022, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2050 + }, + { + "epoch": 2.528976572133169, + "grad_norm": 2.6522520814326414, + "learning_rate": 1.247879658982354e-06, + "loss": 0.3897, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2051 + }, + { + "epoch": 2.530209617755857, + "grad_norm": 2.5416440579601964, + "learning_rate": 1.241509958725038e-06, + "loss": 0.3163, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2052 + }, + { + "epoch": 2.531442663378545, + "grad_norm": 2.757815701292901, + "learning_rate": 1.2351554807874032e-06, + "loss": 0.3496, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2053 + }, + { + "epoch": 2.532675709001233, + "grad_norm": 2.7044209947834013, + "learning_rate": 1.2288162362135813e-06, + "loss": 0.2878, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2054 + }, + { + "epoch": 2.533908754623921, + "grad_norm": 3.8328275411269495, + "learning_rate": 1.2224922360212243e-06, + "loss": 0.5375, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2055 + }, + { + "epoch": 2.5351418002466093, + "grad_norm": 2.7336516657578094, + "learning_rate": 1.2161834912014924e-06, + "loss": 0.3119, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2056 + }, + { + "epoch": 2.536374845869297, + "grad_norm": 2.7379672510937305, + "learning_rate": 1.2098900127190304e-06, + "loss": 0.3355, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2057 + }, + { + "epoch": 2.537607891491985, + "grad_norm": 2.558160684746504, + "learning_rate": 1.2036118115119454e-06, + "loss": 0.319, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2058 + }, + { + "epoch": 2.5388409371146734, + "grad_norm": 2.749942641886583, + "learning_rate": 1.1973488984918025e-06, + "loss": 0.3198, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2059 + }, + { + "epoch": 2.5400739827373613, + "grad_norm": 2.41361021971998, + "learning_rate": 1.1911012845435875e-06, + "loss": 0.2859, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2060 + }, + { + "epoch": 2.541307028360049, + "grad_norm": 2.575831774187727, + "learning_rate": 1.1848689805257008e-06, + "loss": 0.3008, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2061 + }, + { + "epoch": 2.5425400739827375, + "grad_norm": 2.6260742266225043, + "learning_rate": 1.1786519972699272e-06, + "loss": 0.3216, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2062 + }, + { + "epoch": 2.5437731196054254, + "grad_norm": 3.104361328073134, + "learning_rate": 1.1724503455814362e-06, + "loss": 0.3802, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2063 + }, + { + "epoch": 2.5450061652281133, + "grad_norm": 2.8602370289409595, + "learning_rate": 1.16626403623874e-06, + "loss": 0.3114, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2064 + }, + { + "epoch": 2.5462392108508016, + "grad_norm": 2.5858721217627862, + "learning_rate": 1.1600930799936883e-06, + "loss": 0.2818, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2065 + }, + { + "epoch": 2.5474722564734895, + "grad_norm": 2.686218231303633, + "learning_rate": 1.1539374875714514e-06, + "loss": 0.3011, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2066 + }, + { + "epoch": 2.5487053020961774, + "grad_norm": 2.735068892658291, + "learning_rate": 1.1477972696704897e-06, + "loss": 0.3726, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2067 + }, + { + "epoch": 2.5499383477188657, + "grad_norm": 2.7783997481063927, + "learning_rate": 1.1416724369625542e-06, + "loss": 0.3676, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2068 + }, + { + "epoch": 2.5511713933415536, + "grad_norm": 2.863359548731134, + "learning_rate": 1.1355630000926398e-06, + "loss": 0.3124, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2069 + }, + { + "epoch": 2.5524044389642415, + "grad_norm": 2.910145280463402, + "learning_rate": 1.1294689696789985e-06, + "loss": 0.4079, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2070 + }, + { + "epoch": 2.55363748458693, + "grad_norm": 2.908400582916427, + "learning_rate": 1.1233903563130954e-06, + "loss": 0.3681, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2071 + }, + { + "epoch": 2.5548705302096177, + "grad_norm": 2.4646111944049154, + "learning_rate": 1.117327170559609e-06, + "loss": 0.3181, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2072 + }, + { + "epoch": 2.5561035758323056, + "grad_norm": 2.719838877595105, + "learning_rate": 1.1112794229563995e-06, + "loss": 0.3636, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2073 + }, + { + "epoch": 2.557336621454994, + "grad_norm": 2.290599002456828, + "learning_rate": 1.1052471240144934e-06, + "loss": 0.2309, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2074 + }, + { + "epoch": 2.558569667077682, + "grad_norm": 3.1367498622058414, + "learning_rate": 1.0992302842180747e-06, + "loss": 0.4262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2075 + }, + { + "epoch": 2.55980271270037, + "grad_norm": 2.4295574054990676, + "learning_rate": 1.0932289140244524e-06, + "loss": 0.2842, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2076 + }, + { + "epoch": 2.561035758323058, + "grad_norm": 3.747803117771801, + "learning_rate": 1.087243023864053e-06, + "loss": 0.647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2077 + }, + { + "epoch": 2.562268803945746, + "grad_norm": 2.8828281138426504, + "learning_rate": 1.0812726241403937e-06, + "loss": 0.4434, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2078 + }, + { + "epoch": 2.563501849568434, + "grad_norm": 2.741675458781193, + "learning_rate": 1.0753177252300784e-06, + "loss": 0.2738, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2079 + }, + { + "epoch": 2.564734895191122, + "grad_norm": 2.8394418762870766, + "learning_rate": 1.0693783374827615e-06, + "loss": 0.4459, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2080 + }, + { + "epoch": 2.56596794081381, + "grad_norm": 2.9924478193996373, + "learning_rate": 1.0634544712211414e-06, + "loss": 0.3847, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2081 + }, + { + "epoch": 2.5672009864364984, + "grad_norm": 2.932685161162636, + "learning_rate": 1.0575461367409445e-06, + "loss": 0.3084, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2082 + }, + { + "epoch": 2.5684340320591863, + "grad_norm": 2.94162911936551, + "learning_rate": 1.0516533443108945e-06, + "loss": 0.4137, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2083 + }, + { + "epoch": 2.569667077681874, + "grad_norm": 2.998437333915942, + "learning_rate": 1.0457761041727155e-06, + "loss": 0.431, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2084 + }, + { + "epoch": 2.570900123304562, + "grad_norm": 3.521824875699741, + "learning_rate": 1.0399144265410842e-06, + "loss": 0.6146, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2085 + }, + { + "epoch": 2.5721331689272504, + "grad_norm": 2.7270694133764892, + "learning_rate": 1.0340683216036441e-06, + "loss": 0.422, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2086 + }, + { + "epoch": 2.5733662145499383, + "grad_norm": 3.0431991874036566, + "learning_rate": 1.028237799520969e-06, + "loss": 0.4264, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2087 + }, + { + "epoch": 2.5745992601726266, + "grad_norm": 2.6082022216703376, + "learning_rate": 1.0224228704265494e-06, + "loss": 0.2909, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2088 + }, + { + "epoch": 2.5758323057953145, + "grad_norm": 3.1632431507662537, + "learning_rate": 1.0166235444267714e-06, + "loss": 0.5323, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2089 + }, + { + "epoch": 2.5770653514180024, + "grad_norm": 3.514164456519417, + "learning_rate": 1.010839831600907e-06, + "loss": 0.6178, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2090 + }, + { + "epoch": 2.5782983970406903, + "grad_norm": 2.813745630976113, + "learning_rate": 1.0050717420010926e-06, + "loss": 0.322, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2091 + }, + { + "epoch": 2.5795314426633786, + "grad_norm": 2.4269366735428552, + "learning_rate": 9.993192856523103e-07, + "loss": 0.2708, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2092 + }, + { + "epoch": 2.5807644882860665, + "grad_norm": 2.704856535075446, + "learning_rate": 9.93582472552368e-07, + "loss": 0.2844, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2093 + }, + { + "epoch": 2.581997533908755, + "grad_norm": 2.7165576879036344, + "learning_rate": 9.878613126718927e-07, + "loss": 0.3203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2094 + }, + { + "epoch": 2.5832305795314427, + "grad_norm": 2.933329456270621, + "learning_rate": 9.82155815954301e-07, + "loss": 0.4399, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2095 + }, + { + "epoch": 2.5844636251541306, + "grad_norm": 2.2751035220363462, + "learning_rate": 9.764659923157872e-07, + "loss": 0.3253, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2096 + }, + { + "epoch": 2.5856966707768185, + "grad_norm": 2.6561510579792755, + "learning_rate": 9.70791851645304e-07, + "loss": 0.283, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2097 + }, + { + "epoch": 2.586929716399507, + "grad_norm": 3.0205144969702546, + "learning_rate": 9.651334038045524e-07, + "loss": 0.4118, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2098 + }, + { + "epoch": 2.5881627620221948, + "grad_norm": 2.6987646607488998, + "learning_rate": 9.594906586279596e-07, + "loss": 0.3689, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2099 + }, + { + "epoch": 2.589395807644883, + "grad_norm": 3.0826532375796694, + "learning_rate": 9.538636259226497e-07, + "loss": 0.3565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2100 + }, + { + "epoch": 2.590628853267571, + "grad_norm": 2.5870567635603723, + "learning_rate": 9.48252315468452e-07, + "loss": 0.2498, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2101 + }, + { + "epoch": 2.591861898890259, + "grad_norm": 2.514738691266998, + "learning_rate": 9.426567370178629e-07, + "loss": 0.3041, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2102 + }, + { + "epoch": 2.5930949445129468, + "grad_norm": 2.345667044068714, + "learning_rate": 9.370769002960411e-07, + "loss": 0.2686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2103 + }, + { + "epoch": 2.594327990135635, + "grad_norm": 2.84429660778142, + "learning_rate": 9.315128150007823e-07, + "loss": 0.4091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2104 + }, + { + "epoch": 2.595561035758323, + "grad_norm": 2.805637022307319, + "learning_rate": 9.259644908025056e-07, + "loss": 0.3534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2105 + }, + { + "epoch": 2.5967940813810113, + "grad_norm": 3.054470236984765, + "learning_rate": 9.204319373442428e-07, + "loss": 0.3549, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2106 + }, + { + "epoch": 2.598027127003699, + "grad_norm": 2.937698854104628, + "learning_rate": 9.149151642416087e-07, + "loss": 0.334, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2107 + }, + { + "epoch": 2.599260172626387, + "grad_norm": 2.4316266495187873, + "learning_rate": 9.094141810827973e-07, + "loss": 0.3176, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2108 + }, + { + "epoch": 2.600493218249075, + "grad_norm": 2.254122822877399, + "learning_rate": 9.039289974285559e-07, + "loss": 0.2047, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2109 + }, + { + "epoch": 2.6017262638717633, + "grad_norm": 2.568438628492689, + "learning_rate": 8.984596228121755e-07, + "loss": 0.3287, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2110 + }, + { + "epoch": 2.602959309494451, + "grad_norm": 2.8283319973424974, + "learning_rate": 8.93006066739468e-07, + "loss": 0.3432, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2111 + }, + { + "epoch": 2.6041923551171395, + "grad_norm": 2.6918230220304404, + "learning_rate": 8.875683386887524e-07, + "loss": 0.2766, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2112 + }, + { + "epoch": 2.6054254007398274, + "grad_norm": 2.645231347168935, + "learning_rate": 8.821464481108443e-07, + "loss": 0.3197, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2113 + }, + { + "epoch": 2.6066584463625153, + "grad_norm": 2.72529098713838, + "learning_rate": 8.767404044290229e-07, + "loss": 0.2404, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2114 + }, + { + "epoch": 2.607891491985203, + "grad_norm": 2.542384437873485, + "learning_rate": 8.713502170390397e-07, + "loss": 0.248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2115 + }, + { + "epoch": 2.6091245376078915, + "grad_norm": 3.246141127340731, + "learning_rate": 8.659758953090714e-07, + "loss": 0.4527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2116 + }, + { + "epoch": 2.6103575832305794, + "grad_norm": 3.6904197461466537, + "learning_rate": 8.606174485797325e-07, + "loss": 0.463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2117 + }, + { + "epoch": 2.6115906288532678, + "grad_norm": 2.700124766503106, + "learning_rate": 8.552748861640414e-07, + "loss": 0.2926, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2118 + }, + { + "epoch": 2.6128236744759556, + "grad_norm": 2.6759281011209053, + "learning_rate": 8.499482173474138e-07, + "loss": 0.3204, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2119 + }, + { + "epoch": 2.6140567200986435, + "grad_norm": 2.426884378444823, + "learning_rate": 8.446374513876355e-07, + "loss": 0.2175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2120 + }, + { + "epoch": 2.6152897657213314, + "grad_norm": 2.9455204687864582, + "learning_rate": 8.393425975148562e-07, + "loss": 0.4087, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2121 + }, + { + "epoch": 2.6165228113440198, + "grad_norm": 2.9699872635936417, + "learning_rate": 8.340636649315736e-07, + "loss": 0.4025, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2122 + }, + { + "epoch": 2.6177558569667077, + "grad_norm": 2.534981188094632, + "learning_rate": 8.288006628126089e-07, + "loss": 0.3135, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2123 + }, + { + "epoch": 2.618988902589396, + "grad_norm": 2.4519907184425076, + "learning_rate": 8.235536003050982e-07, + "loss": 0.3064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2124 + }, + { + "epoch": 2.620221948212084, + "grad_norm": 3.063736416186299, + "learning_rate": 8.183224865284767e-07, + "loss": 0.4251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2125 + }, + { + "epoch": 2.6214549938347718, + "grad_norm": 3.096065296493664, + "learning_rate": 8.131073305744542e-07, + "loss": 0.4414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2126 + }, + { + "epoch": 2.6226880394574597, + "grad_norm": 2.32063595875409, + "learning_rate": 8.079081415070167e-07, + "loss": 0.237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2127 + }, + { + "epoch": 2.623921085080148, + "grad_norm": 2.448315773087064, + "learning_rate": 8.02724928362385e-07, + "loss": 0.2762, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2128 + }, + { + "epoch": 2.625154130702836, + "grad_norm": 2.7385596940488304, + "learning_rate": 7.975577001490264e-07, + "loss": 0.3421, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2129 + }, + { + "epoch": 2.626387176325524, + "grad_norm": 2.7895945525631247, + "learning_rate": 7.924064658476182e-07, + "loss": 0.3558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2130 + }, + { + "epoch": 2.627620221948212, + "grad_norm": 2.6723893074006897, + "learning_rate": 7.872712344110467e-07, + "loss": 0.3371, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2131 + }, + { + "epoch": 2.6288532675709, + "grad_norm": 2.5000122771671713, + "learning_rate": 7.821520147643802e-07, + "loss": 0.2636, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2132 + }, + { + "epoch": 2.6300863131935883, + "grad_norm": 2.301348255073216, + "learning_rate": 7.770488158048572e-07, + "loss": 0.2647, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2133 + }, + { + "epoch": 2.631319358816276, + "grad_norm": 3.3922269985212496, + "learning_rate": 7.719616464018787e-07, + "loss": 0.5083, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2134 + }, + { + "epoch": 2.632552404438964, + "grad_norm": 3.191717491215566, + "learning_rate": 7.668905153969808e-07, + "loss": 0.362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2135 + }, + { + "epoch": 2.6337854500616524, + "grad_norm": 2.6437468967942745, + "learning_rate": 7.618354316038257e-07, + "loss": 0.3017, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2136 + }, + { + "epoch": 2.6350184956843403, + "grad_norm": 3.0892646716858967, + "learning_rate": 7.567964038081855e-07, + "loss": 0.4299, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2137 + }, + { + "epoch": 2.636251541307028, + "grad_norm": 2.3684431011024087, + "learning_rate": 7.517734407679289e-07, + "loss": 0.2943, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2138 + }, + { + "epoch": 2.6374845869297165, + "grad_norm": 2.6445326991118114, + "learning_rate": 7.467665512130018e-07, + "loss": 0.3446, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2139 + }, + { + "epoch": 2.6387176325524044, + "grad_norm": 2.9778345320905393, + "learning_rate": 7.417757438454131e-07, + "loss": 0.4721, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2140 + }, + { + "epoch": 2.6399506781750923, + "grad_norm": 2.9780123735843875, + "learning_rate": 7.368010273392267e-07, + "loss": 0.4237, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2141 + }, + { + "epoch": 2.6411837237977807, + "grad_norm": 3.0783390935516475, + "learning_rate": 7.31842410340533e-07, + "loss": 0.4354, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2142 + }, + { + "epoch": 2.6424167694204685, + "grad_norm": 2.575788597385855, + "learning_rate": 7.268999014674449e-07, + "loss": 0.2829, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2143 + }, + { + "epoch": 2.6436498150431564, + "grad_norm": 2.874096380154356, + "learning_rate": 7.219735093100766e-07, + "loss": 0.3294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2144 + }, + { + "epoch": 2.6448828606658448, + "grad_norm": 2.9439671201694932, + "learning_rate": 7.170632424305379e-07, + "loss": 0.4352, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2145 + }, + { + "epoch": 2.6461159062885327, + "grad_norm": 3.0548876771514406, + "learning_rate": 7.12169109362908e-07, + "loss": 0.3814, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2146 + }, + { + "epoch": 2.6473489519112205, + "grad_norm": 2.7227655572000184, + "learning_rate": 7.072911186132236e-07, + "loss": 0.3471, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2147 + }, + { + "epoch": 2.648581997533909, + "grad_norm": 2.6127642443555934, + "learning_rate": 7.024292786594711e-07, + "loss": 0.2768, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2148 + }, + { + "epoch": 2.6498150431565968, + "grad_norm": 2.5487594534779716, + "learning_rate": 6.9758359795156e-07, + "loss": 0.2314, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2149 + }, + { + "epoch": 2.6510480887792847, + "grad_norm": 2.7838847454830873, + "learning_rate": 6.927540849113224e-07, + "loss": 0.3203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2150 + }, + { + "epoch": 2.652281134401973, + "grad_norm": 2.861427726842305, + "learning_rate": 6.879407479324851e-07, + "loss": 0.3504, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2151 + }, + { + "epoch": 2.653514180024661, + "grad_norm": 2.6828614082668065, + "learning_rate": 6.831435953806631e-07, + "loss": 0.3091, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2152 + }, + { + "epoch": 2.654747225647349, + "grad_norm": 2.65009181541632, + "learning_rate": 6.783626355933448e-07, + "loss": 0.2928, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2153 + }, + { + "epoch": 2.655980271270037, + "grad_norm": 3.0403775342787114, + "learning_rate": 6.735978768798712e-07, + "loss": 0.4531, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2154 + }, + { + "epoch": 2.657213316892725, + "grad_norm": 2.794771980738937, + "learning_rate": 6.68849327521428e-07, + "loss": 0.3226, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2155 + }, + { + "epoch": 2.658446362515413, + "grad_norm": 2.756593482580578, + "learning_rate": 6.641169957710269e-07, + "loss": 0.3468, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2156 + }, + { + "epoch": 2.659679408138101, + "grad_norm": 3.258706243877682, + "learning_rate": 6.594008898534976e-07, + "loss": 0.4603, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2157 + }, + { + "epoch": 2.660912453760789, + "grad_norm": 3.207580449518217, + "learning_rate": 6.54701017965469e-07, + "loss": 0.462, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2158 + }, + { + "epoch": 2.6621454993834774, + "grad_norm": 3.2653771336865463, + "learning_rate": 6.500173882753469e-07, + "loss": 0.441, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2159 + }, + { + "epoch": 2.6633785450061653, + "grad_norm": 3.1415827019539093, + "learning_rate": 6.453500089233178e-07, + "loss": 0.46, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2160 + }, + { + "epoch": 2.664611590628853, + "grad_norm": 2.767410566814036, + "learning_rate": 6.406988880213195e-07, + "loss": 0.4478, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2161 + }, + { + "epoch": 2.665844636251541, + "grad_norm": 2.931008689013411, + "learning_rate": 6.360640336530366e-07, + "loss": 0.3305, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2162 + }, + { + "epoch": 2.6670776818742294, + "grad_norm": 2.4294915612305785, + "learning_rate": 6.314454538738757e-07, + "loss": 0.2934, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2163 + }, + { + "epoch": 2.6683107274969173, + "grad_norm": 2.5513858609612456, + "learning_rate": 6.268431567109645e-07, + "loss": 0.2536, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2164 + }, + { + "epoch": 2.6695437731196057, + "grad_norm": 2.651100260036379, + "learning_rate": 6.222571501631302e-07, + "loss": 0.3103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2165 + }, + { + "epoch": 2.6707768187422936, + "grad_norm": 2.38456527102379, + "learning_rate": 6.17687442200885e-07, + "loss": 0.2103, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2166 + }, + { + "epoch": 2.6720098643649814, + "grad_norm": 3.001791403045041, + "learning_rate": 6.131340407664144e-07, + "loss": 0.3546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2167 + }, + { + "epoch": 2.6732429099876693, + "grad_norm": 2.52655628098101, + "learning_rate": 6.085969537735615e-07, + "loss": 0.2742, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2168 + }, + { + "epoch": 2.6744759556103577, + "grad_norm": 2.5598496428290862, + "learning_rate": 6.040761891078206e-07, + "loss": 0.2435, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2169 + }, + { + "epoch": 2.6757090012330456, + "grad_norm": 2.5663755974631113, + "learning_rate": 5.995717546263102e-07, + "loss": 0.2623, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2170 + }, + { + "epoch": 2.676942046855734, + "grad_norm": 2.4918016686508, + "learning_rate": 5.950836581577713e-07, + "loss": 0.3085, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2171 + }, + { + "epoch": 2.678175092478422, + "grad_norm": 2.9580267523860666, + "learning_rate": 5.906119075025496e-07, + "loss": 0.3952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2172 + }, + { + "epoch": 2.6794081381011097, + "grad_norm": 2.6022102383170576, + "learning_rate": 5.861565104325783e-07, + "loss": 0.3812, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2173 + }, + { + "epoch": 2.6806411837237976, + "grad_norm": 3.18289248328154, + "learning_rate": 5.817174746913745e-07, + "loss": 0.4567, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2174 + }, + { + "epoch": 2.681874229346486, + "grad_norm": 2.638699715045903, + "learning_rate": 5.772948079940077e-07, + "loss": 0.3016, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2175 + }, + { + "epoch": 2.683107274969174, + "grad_norm": 3.0706400741828452, + "learning_rate": 5.728885180271105e-07, + "loss": 0.4779, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2176 + }, + { + "epoch": 2.684340320591862, + "grad_norm": 2.94433495823207, + "learning_rate": 5.684986124488445e-07, + "loss": 0.363, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2177 + }, + { + "epoch": 2.68557336621455, + "grad_norm": 2.6522120464214556, + "learning_rate": 5.641250988889002e-07, + "loss": 0.2923, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2178 + }, + { + "epoch": 2.686806411837238, + "grad_norm": 2.6730782417287453, + "learning_rate": 5.597679849484739e-07, + "loss": 0.2437, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2179 + }, + { + "epoch": 2.688039457459926, + "grad_norm": 2.877819125843327, + "learning_rate": 5.554272782002602e-07, + "loss": 0.3789, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2180 + }, + { + "epoch": 2.689272503082614, + "grad_norm": 2.494737118535582, + "learning_rate": 5.51102986188442e-07, + "loss": 0.2563, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2181 + }, + { + "epoch": 2.690505548705302, + "grad_norm": 2.357490810329682, + "learning_rate": 5.4679511642867e-07, + "loss": 0.2295, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2182 + }, + { + "epoch": 2.6917385943279903, + "grad_norm": 2.6030983370502474, + "learning_rate": 5.425036764080527e-07, + "loss": 0.3577, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2183 + }, + { + "epoch": 2.6929716399506782, + "grad_norm": 2.721506673541932, + "learning_rate": 5.382286735851428e-07, + "loss": 0.2917, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2184 + }, + { + "epoch": 2.694204685573366, + "grad_norm": 2.4533131449489565, + "learning_rate": 5.339701153899291e-07, + "loss": 0.3143, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2185 + }, + { + "epoch": 2.695437731196054, + "grad_norm": 2.6685959681986806, + "learning_rate": 5.297280092238155e-07, + "loss": 0.2856, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2186 + }, + { + "epoch": 2.6966707768187423, + "grad_norm": 2.45466400044313, + "learning_rate": 5.255023624596122e-07, + "loss": 0.257, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2187 + }, + { + "epoch": 2.6979038224414302, + "grad_norm": 2.5189259604300807, + "learning_rate": 5.212931824415269e-07, + "loss": 0.2672, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2188 + }, + { + "epoch": 2.6991368680641186, + "grad_norm": 2.676635818452075, + "learning_rate": 5.17100476485144e-07, + "loss": 0.2876, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2189 + }, + { + "epoch": 2.7003699136868065, + "grad_norm": 2.835387458892149, + "learning_rate": 5.129242518774147e-07, + "loss": 0.3733, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2190 + }, + { + "epoch": 2.7016029593094943, + "grad_norm": 3.1714092235602975, + "learning_rate": 5.08764515876653e-07, + "loss": 0.4077, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2191 + }, + { + "epoch": 2.7028360049321822, + "grad_norm": 2.7479445461661216, + "learning_rate": 5.046212757125046e-07, + "loss": 0.2996, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2192 + }, + { + "epoch": 2.7040690505548706, + "grad_norm": 3.112272599454582, + "learning_rate": 5.004945385859561e-07, + "loss": 0.375, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2193 + }, + { + "epoch": 2.7053020961775585, + "grad_norm": 3.432818543832485, + "learning_rate": 4.963843116693057e-07, + "loss": 0.4078, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2194 + }, + { + "epoch": 2.706535141800247, + "grad_norm": 2.6816007954070997, + "learning_rate": 4.922906021061557e-07, + "loss": 0.3574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2195 + }, + { + "epoch": 2.7077681874229347, + "grad_norm": 2.791343350270728, + "learning_rate": 4.882134170114039e-07, + "loss": 0.2835, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2196 + }, + { + "epoch": 2.7090012330456226, + "grad_norm": 2.6626193488192023, + "learning_rate": 4.841527634712295e-07, + "loss": 0.2964, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2197 + }, + { + "epoch": 2.7102342786683105, + "grad_norm": 3.0470087530556684, + "learning_rate": 4.801086485430761e-07, + "loss": 0.3629, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2198 + }, + { + "epoch": 2.711467324290999, + "grad_norm": 2.672432922245955, + "learning_rate": 4.7608107925564583e-07, + "loss": 0.3081, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2199 + }, + { + "epoch": 2.7127003699136867, + "grad_norm": 2.7992002975388646, + "learning_rate": 4.720700626088837e-07, + "loss": 0.3229, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2200 + }, + { + "epoch": 2.713933415536375, + "grad_norm": 2.6268441126040916, + "learning_rate": 4.680756055739666e-07, + "loss": 0.2793, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2201 + }, + { + "epoch": 2.715166461159063, + "grad_norm": 3.127898243079573, + "learning_rate": 4.640977150932879e-07, + "loss": 0.4043, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2202 + }, + { + "epoch": 2.716399506781751, + "grad_norm": 2.678962666360488, + "learning_rate": 4.601363980804519e-07, + "loss": 0.3115, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2203 + }, + { + "epoch": 2.717632552404439, + "grad_norm": 2.086204952822806, + "learning_rate": 4.561916614202544e-07, + "loss": 0.1558, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2204 + }, + { + "epoch": 2.718865598027127, + "grad_norm": 2.8395871349795723, + "learning_rate": 4.522635119686825e-07, + "loss": 0.4355, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2205 + }, + { + "epoch": 2.720098643649815, + "grad_norm": 3.130472545404112, + "learning_rate": 4.4835195655288044e-07, + "loss": 0.3824, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2206 + }, + { + "epoch": 2.7213316892725032, + "grad_norm": 2.513592274819219, + "learning_rate": 4.444570019711669e-07, + "loss": 0.2262, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2207 + }, + { + "epoch": 2.722564734895191, + "grad_norm": 2.2860823121731273, + "learning_rate": 4.405786549929958e-07, + "loss": 0.2164, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2208 + }, + { + "epoch": 2.723797780517879, + "grad_norm": 2.7573339648742117, + "learning_rate": 4.3671692235896845e-07, + "loss": 0.2744, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2209 + }, + { + "epoch": 2.7250308261405674, + "grad_norm": 2.681707991850475, + "learning_rate": 4.328718107807994e-07, + "loss": 0.3588, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2210 + }, + { + "epoch": 2.7262638717632552, + "grad_norm": 2.7941508323953537, + "learning_rate": 4.29043326941323e-07, + "loss": 0.3362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2211 + }, + { + "epoch": 2.727496917385943, + "grad_norm": 2.6121546331706047, + "learning_rate": 4.252314774944721e-07, + "loss": 0.3621, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2212 + }, + { + "epoch": 2.7287299630086315, + "grad_norm": 2.548315199795081, + "learning_rate": 4.2143626906526845e-07, + "loss": 0.2353, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2213 + }, + { + "epoch": 2.7299630086313194, + "grad_norm": 2.679304484955262, + "learning_rate": 4.1765770824981235e-07, + "loss": 0.3502, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2214 + }, + { + "epoch": 2.7311960542540072, + "grad_norm": 2.667678407852026, + "learning_rate": 4.138958016152683e-07, + "loss": 0.2942, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2215 + }, + { + "epoch": 2.7324290998766956, + "grad_norm": 2.6834712222907853, + "learning_rate": 4.101505556998597e-07, + "loss": 0.2805, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2216 + }, + { + "epoch": 2.7336621454993835, + "grad_norm": 2.877694576929324, + "learning_rate": 4.064219770128519e-07, + "loss": 0.3806, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2217 + }, + { + "epoch": 2.7348951911220714, + "grad_norm": 2.9757482603801253, + "learning_rate": 4.027100720345389e-07, + "loss": 0.3245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2218 + }, + { + "epoch": 2.7361282367447597, + "grad_norm": 2.668887841892006, + "learning_rate": 3.990148472162414e-07, + "loss": 0.3048, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2219 + }, + { + "epoch": 2.7373612823674476, + "grad_norm": 3.1879390801050485, + "learning_rate": 3.953363089802853e-07, + "loss": 0.4051, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2220 + }, + { + "epoch": 2.7385943279901355, + "grad_norm": 2.7902737194957727, + "learning_rate": 3.916744637199998e-07, + "loss": 0.3309, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2221 + }, + { + "epoch": 2.739827373612824, + "grad_norm": 3.2040231447624055, + "learning_rate": 3.880293177996941e-07, + "loss": 0.378, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2222 + }, + { + "epoch": 2.7410604192355117, + "grad_norm": 2.9508071245897964, + "learning_rate": 3.844008775546615e-07, + "loss": 0.3066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2223 + }, + { + "epoch": 2.7422934648582, + "grad_norm": 2.9813075667563194, + "learning_rate": 3.807891492911586e-07, + "loss": 0.4095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2224 + }, + { + "epoch": 2.743526510480888, + "grad_norm": 2.7175907204726895, + "learning_rate": 3.7719413928639426e-07, + "loss": 0.3196, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2225 + }, + { + "epoch": 2.744759556103576, + "grad_norm": 2.872884129460395, + "learning_rate": 3.736158537885215e-07, + "loss": 0.3266, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2226 + }, + { + "epoch": 2.7459926017262637, + "grad_norm": 2.8178474009372367, + "learning_rate": 3.700542990166256e-07, + "loss": 0.2714, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2227 + }, + { + "epoch": 2.747225647348952, + "grad_norm": 2.8963336563742867, + "learning_rate": 3.665094811607184e-07, + "loss": 0.3066, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2228 + }, + { + "epoch": 2.74845869297164, + "grad_norm": 2.4948640882118287, + "learning_rate": 3.629814063817161e-07, + "loss": 0.3546, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2229 + }, + { + "epoch": 2.7496917385943282, + "grad_norm": 3.1010149751360507, + "learning_rate": 3.594700808114393e-07, + "loss": 0.3009, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2230 + }, + { + "epoch": 2.750924784217016, + "grad_norm": 2.718358950761102, + "learning_rate": 3.559755105525986e-07, + "loss": 0.3251, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2231 + }, + { + "epoch": 2.752157829839704, + "grad_norm": 2.8454348321878014, + "learning_rate": 3.5249770167878117e-07, + "loss": 0.3838, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2232 + }, + { + "epoch": 2.753390875462392, + "grad_norm": 2.7730549519374863, + "learning_rate": 3.490366602344453e-07, + "loss": 0.305, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2233 + }, + { + "epoch": 2.7546239210850803, + "grad_norm": 2.3291745633960277, + "learning_rate": 3.455923922349047e-07, + "loss": 0.2245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2234 + }, + { + "epoch": 2.755856966707768, + "grad_norm": 2.5747655685727455, + "learning_rate": 3.421649036663244e-07, + "loss": 0.2972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2235 + }, + { + "epoch": 2.7570900123304565, + "grad_norm": 2.9115771640323764, + "learning_rate": 3.3875420048570254e-07, + "loss": 0.2952, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2236 + }, + { + "epoch": 2.7583230579531444, + "grad_norm": 2.7112719680424355, + "learning_rate": 3.3536028862086736e-07, + "loss": 0.363, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2237 + }, + { + "epoch": 2.7595561035758323, + "grad_norm": 2.97836327107677, + "learning_rate": 3.3198317397046376e-07, + "loss": 0.3279, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2238 + }, + { + "epoch": 2.76078914919852, + "grad_norm": 2.4918802800025723, + "learning_rate": 3.2862286240393894e-07, + "loss": 0.3013, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2239 + }, + { + "epoch": 2.7620221948212085, + "grad_norm": 2.8237580588547435, + "learning_rate": 3.2527935976154113e-07, + "loss": 0.3396, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2240 + }, + { + "epoch": 2.7632552404438964, + "grad_norm": 2.343266269407276, + "learning_rate": 3.219526718543009e-07, + "loss": 0.2082, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2241 + }, + { + "epoch": 2.7644882860665847, + "grad_norm": 2.782536401131464, + "learning_rate": 3.1864280446402774e-07, + "loss": 0.362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2242 + }, + { + "epoch": 2.7657213316892726, + "grad_norm": 2.7374427372779664, + "learning_rate": 3.1534976334329225e-07, + "loss": 0.4243, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2243 + }, + { + "epoch": 2.7669543773119605, + "grad_norm": 2.671382874586831, + "learning_rate": 3.120735542154285e-07, + "loss": 0.2668, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2244 + }, + { + "epoch": 2.7681874229346484, + "grad_norm": 2.1985023090373628, + "learning_rate": 3.0881418277450835e-07, + "loss": 0.2039, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2245 + }, + { + "epoch": 2.7694204685573367, + "grad_norm": 3.0341985461961216, + "learning_rate": 3.055716546853438e-07, + "loss": 0.3804, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2246 + }, + { + "epoch": 2.7706535141800246, + "grad_norm": 2.9462666138738096, + "learning_rate": 3.0234597558347347e-07, + "loss": 0.3254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2247 + }, + { + "epoch": 2.771886559802713, + "grad_norm": 2.7858704895131208, + "learning_rate": 2.991371510751506e-07, + "loss": 0.2527, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2248 + }, + { + "epoch": 2.773119605425401, + "grad_norm": 3.1931336874353553, + "learning_rate": 2.959451867373353e-07, + "loss": 0.4064, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2249 + }, + { + "epoch": 2.7743526510480887, + "grad_norm": 3.100258436417072, + "learning_rate": 2.927700881176854e-07, + "loss": 0.3543, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2250 + }, + { + "epoch": 2.7755856966707766, + "grad_norm": 3.0624613150847457, + "learning_rate": 2.896118607345444e-07, + "loss": 0.4073, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2251 + }, + { + "epoch": 2.776818742293465, + "grad_norm": 3.1757022662898478, + "learning_rate": 2.8647051007693716e-07, + "loss": 0.4605, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2252 + }, + { + "epoch": 2.778051787916153, + "grad_norm": 3.0776734195635247, + "learning_rate": 2.833460416045508e-07, + "loss": 0.3709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2253 + }, + { + "epoch": 2.779284833538841, + "grad_norm": 2.6641081093740175, + "learning_rate": 2.802384607477371e-07, + "loss": 0.3019, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2254 + }, + { + "epoch": 2.780517879161529, + "grad_norm": 3.1682517526338185, + "learning_rate": 2.771477729074923e-07, + "loss": 0.4295, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2255 + }, + { + "epoch": 2.781750924784217, + "grad_norm": 2.8790887424320295, + "learning_rate": 2.740739834554562e-07, + "loss": 0.3337, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2256 + }, + { + "epoch": 2.782983970406905, + "grad_norm": 3.1308022012414076, + "learning_rate": 2.710170977338955e-07, + "loss": 0.2957, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2257 + }, + { + "epoch": 2.784217016029593, + "grad_norm": 2.9616869716016914, + "learning_rate": 2.679771210557003e-07, + "loss": 0.3355, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2258 + }, + { + "epoch": 2.785450061652281, + "grad_norm": 2.1765488671337363, + "learning_rate": 2.649540587043742e-07, + "loss": 0.1907, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2259 + }, + { + "epoch": 2.7866831072749694, + "grad_norm": 3.3985949715424346, + "learning_rate": 2.619479159340199e-07, + "loss": 0.4913, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2260 + }, + { + "epoch": 2.7879161528976573, + "grad_norm": 2.804988011712038, + "learning_rate": 2.5895869796933706e-07, + "loss": 0.286, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2261 + }, + { + "epoch": 2.789149198520345, + "grad_norm": 2.8392706912170484, + "learning_rate": 2.559864100056042e-07, + "loss": 0.3134, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2262 + }, + { + "epoch": 2.790382244143033, + "grad_norm": 2.561774901582968, + "learning_rate": 2.530310572086836e-07, + "loss": 0.2028, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2263 + }, + { + "epoch": 2.7916152897657214, + "grad_norm": 2.664916628749395, + "learning_rate": 2.5009264471500094e-07, + "loss": 0.2946, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2264 + }, + { + "epoch": 2.7928483353884093, + "grad_norm": 2.6490387445909898, + "learning_rate": 2.471711776315344e-07, + "loss": 0.342, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2265 + }, + { + "epoch": 2.7940813810110976, + "grad_norm": 3.2893340874143586, + "learning_rate": 2.4426666103581797e-07, + "loss": 0.4316, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2266 + }, + { + "epoch": 2.7953144266337855, + "grad_norm": 3.0630439813594954, + "learning_rate": 2.4137909997592137e-07, + "loss": 0.4149, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2267 + }, + { + "epoch": 2.7965474722564734, + "grad_norm": 2.7247402896097634, + "learning_rate": 2.3850849947044897e-07, + "loss": 0.3485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2268 + }, + { + "epoch": 2.7977805178791613, + "grad_norm": 2.6752413401294897, + "learning_rate": 2.3565486450852214e-07, + "loss": 0.2184, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2269 + }, + { + "epoch": 2.7990135635018496, + "grad_norm": 2.776357831139634, + "learning_rate": 2.328182000497814e-07, + "loss": 0.2473, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2270 + }, + { + "epoch": 2.8002466091245375, + "grad_norm": 3.1794260662999485, + "learning_rate": 2.2999851102437077e-07, + "loss": 0.4223, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2271 + }, + { + "epoch": 2.801479654747226, + "grad_norm": 3.36115472174311, + "learning_rate": 2.2719580233292904e-07, + "loss": 0.5301, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2272 + }, + { + "epoch": 2.8027127003699137, + "grad_norm": 2.9493395733276957, + "learning_rate": 2.2441007884658526e-07, + "loss": 0.2911, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2273 + }, + { + "epoch": 2.8039457459926016, + "grad_norm": 3.1788612856525553, + "learning_rate": 2.216413454069455e-07, + "loss": 0.3463, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2274 + }, + { + "epoch": 2.8051787916152895, + "grad_norm": 2.5433045645398535, + "learning_rate": 2.188896068260915e-07, + "loss": 0.2945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2275 + }, + { + "epoch": 2.806411837237978, + "grad_norm": 2.306758823846565, + "learning_rate": 2.1615486788656436e-07, + "loss": 0.2134, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2276 + }, + { + "epoch": 2.8076448828606657, + "grad_norm": 2.7786817694469077, + "learning_rate": 2.1343713334135763e-07, + "loss": 0.3362, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2277 + }, + { + "epoch": 2.808877928483354, + "grad_norm": 2.709760199279053, + "learning_rate": 2.1073640791391624e-07, + "loss": 0.3176, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2278 + }, + { + "epoch": 2.810110974106042, + "grad_norm": 2.8890364822488586, + "learning_rate": 2.0805269629812108e-07, + "loss": 0.3462, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2279 + }, + { + "epoch": 2.81134401972873, + "grad_norm": 2.528190834681343, + "learning_rate": 2.0538600315828104e-07, + "loss": 0.2159, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2280 + }, + { + "epoch": 2.812577065351418, + "grad_norm": 2.7570828595992483, + "learning_rate": 2.0273633312912878e-07, + "loss": 0.3532, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2281 + }, + { + "epoch": 2.813810110974106, + "grad_norm": 2.769256431070824, + "learning_rate": 2.0010369081581048e-07, + "loss": 0.2816, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2282 + }, + { + "epoch": 2.815043156596794, + "grad_norm": 3.253599717767732, + "learning_rate": 1.9748808079387726e-07, + "loss": 0.4642, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2283 + }, + { + "epoch": 2.8162762022194823, + "grad_norm": 2.9386468420774947, + "learning_rate": 1.9488950760927938e-07, + "loss": 0.3552, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2284 + }, + { + "epoch": 2.81750924784217, + "grad_norm": 2.9829829993305297, + "learning_rate": 1.9230797577835524e-07, + "loss": 0.3172, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2285 + }, + { + "epoch": 2.818742293464858, + "grad_norm": 2.541262386497395, + "learning_rate": 1.8974348978782474e-07, + "loss": 0.2113, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2286 + }, + { + "epoch": 2.8199753390875464, + "grad_norm": 2.703920047303226, + "learning_rate": 1.871960540947859e-07, + "loss": 0.28, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2287 + }, + { + "epoch": 2.8212083847102343, + "grad_norm": 2.650165079079577, + "learning_rate": 1.8466567312670048e-07, + "loss": 0.3356, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2288 + }, + { + "epoch": 2.822441430332922, + "grad_norm": 2.533426652228835, + "learning_rate": 1.821523512813861e-07, + "loss": 0.3212, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2289 + }, + { + "epoch": 2.8236744759556105, + "grad_norm": 2.5878782348414227, + "learning_rate": 1.7965609292701747e-07, + "loss": 0.2992, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2290 + }, + { + "epoch": 2.8249075215782984, + "grad_norm": 2.878804739891148, + "learning_rate": 1.771769024021086e-07, + "loss": 0.2959, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2291 + }, + { + "epoch": 2.8261405672009863, + "grad_norm": 2.428309506253573, + "learning_rate": 1.7471478401551057e-07, + "loss": 0.2658, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2292 + }, + { + "epoch": 2.8273736128236746, + "grad_norm": 3.1123953928618895, + "learning_rate": 1.7226974204640257e-07, + "loss": 0.4519, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2293 + }, + { + "epoch": 2.8286066584463625, + "grad_norm": 2.2770691838923485, + "learning_rate": 1.6984178074428758e-07, + "loss": 0.2459, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2294 + }, + { + "epoch": 2.829839704069051, + "grad_norm": 3.2472282374199275, + "learning_rate": 1.6743090432897792e-07, + "loss": 0.4179, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2295 + }, + { + "epoch": 2.8310727496917387, + "grad_norm": 2.896901413820259, + "learning_rate": 1.650371169905951e-07, + "loss": 0.3518, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2296 + }, + { + "epoch": 2.8323057953144266, + "grad_norm": 2.9026279723761967, + "learning_rate": 1.626604228895601e-07, + "loss": 0.3308, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2297 + }, + { + "epoch": 2.8335388409371145, + "grad_norm": 2.378847822286896, + "learning_rate": 1.6030082615658304e-07, + "loss": 0.2818, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2298 + }, + { + "epoch": 2.834771886559803, + "grad_norm": 2.833425799780848, + "learning_rate": 1.5795833089266134e-07, + "loss": 0.3994, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2299 + }, + { + "epoch": 2.8360049321824907, + "grad_norm": 3.1189414053204336, + "learning_rate": 1.5563294116906825e-07, + "loss": 0.4709, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2300 + }, + { + "epoch": 2.837237977805179, + "grad_norm": 2.707296626328123, + "learning_rate": 1.5332466102734867e-07, + "loss": 0.2953, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2301 + }, + { + "epoch": 2.838471023427867, + "grad_norm": 2.516960779402883, + "learning_rate": 1.5103349447930904e-07, + "loss": 0.245, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2302 + }, + { + "epoch": 2.839704069050555, + "grad_norm": 2.958582681150781, + "learning_rate": 1.4875944550701514e-07, + "loss": 0.3958, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2303 + }, + { + "epoch": 2.8409371146732427, + "grad_norm": 2.7254557023940844, + "learning_rate": 1.4650251806277993e-07, + "loss": 0.2945, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2304 + }, + { + "epoch": 2.842170160295931, + "grad_norm": 3.082043876526701, + "learning_rate": 1.4426271606916008e-07, + "loss": 0.35, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2305 + }, + { + "epoch": 2.843403205918619, + "grad_norm": 2.5067497120435225, + "learning_rate": 1.420400434189484e-07, + "loss": 0.2294, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2306 + }, + { + "epoch": 2.8446362515413073, + "grad_norm": 2.7651441031561603, + "learning_rate": 1.3983450397516474e-07, + "loss": 0.3267, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2307 + }, + { + "epoch": 2.845869297163995, + "grad_norm": 2.828803793300329, + "learning_rate": 1.3764610157105395e-07, + "loss": 0.2936, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2308 + }, + { + "epoch": 2.847102342786683, + "grad_norm": 3.2049883954799063, + "learning_rate": 1.354748400100736e-07, + "loss": 0.5547, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2309 + }, + { + "epoch": 2.848335388409371, + "grad_norm": 2.9233069105494747, + "learning_rate": 1.3332072306589393e-07, + "loss": 0.3569, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2310 + }, + { + "epoch": 2.8495684340320593, + "grad_norm": 2.814647176919941, + "learning_rate": 1.3118375448238685e-07, + "loss": 0.4375, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2311 + }, + { + "epoch": 2.850801479654747, + "grad_norm": 2.5122826761547006, + "learning_rate": 1.2906393797361695e-07, + "loss": 0.2806, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2312 + }, + { + "epoch": 2.8520345252774355, + "grad_norm": 2.532670655275563, + "learning_rate": 1.2696127722384266e-07, + "loss": 0.3203, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2313 + }, + { + "epoch": 2.8532675709001234, + "grad_norm": 2.7563586812416765, + "learning_rate": 1.2487577588750187e-07, + "loss": 0.3151, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2314 + }, + { + "epoch": 2.8545006165228113, + "grad_norm": 2.818501024370185, + "learning_rate": 1.22807437589213e-07, + "loss": 0.254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2315 + }, + { + "epoch": 2.855733662145499, + "grad_norm": 2.270871205989627, + "learning_rate": 1.2075626592376154e-07, + "loss": 0.2095, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2316 + }, + { + "epoch": 2.8569667077681875, + "grad_norm": 2.7330456862949304, + "learning_rate": 1.1872226445610035e-07, + "loss": 0.3015, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2317 + }, + { + "epoch": 2.8581997533908754, + "grad_norm": 2.2997809558601743, + "learning_rate": 1.1670543672133827e-07, + "loss": 0.1677, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2318 + }, + { + "epoch": 2.8594327990135637, + "grad_norm": 2.2344707564071755, + "learning_rate": 1.1470578622473583e-07, + "loss": 0.1507, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2319 + }, + { + "epoch": 2.8606658446362516, + "grad_norm": 2.4848246222561143, + "learning_rate": 1.1272331644170187e-07, + "loss": 0.2855, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2320 + }, + { + "epoch": 2.8618988902589395, + "grad_norm": 3.108439216735305, + "learning_rate": 1.1075803081778025e-07, + "loss": 0.4166, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2321 + }, + { + "epoch": 2.8631319358816274, + "grad_norm": 3.303321471795626, + "learning_rate": 1.0880993276865648e-07, + "loss": 0.4139, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2322 + }, + { + "epoch": 2.8643649815043157, + "grad_norm": 3.3555172011940124, + "learning_rate": 1.0687902568013553e-07, + "loss": 0.4597, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2323 + }, + { + "epoch": 2.8655980271270036, + "grad_norm": 3.0011825883832692, + "learning_rate": 1.0496531290814849e-07, + "loss": 0.3156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2324 + }, + { + "epoch": 2.866831072749692, + "grad_norm": 3.0113807671712842, + "learning_rate": 1.0306879777874257e-07, + "loss": 0.3843, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2325 + }, + { + "epoch": 2.86806411837238, + "grad_norm": 2.684762978199241, + "learning_rate": 1.0118948358807557e-07, + "loss": 0.3367, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2326 + }, + { + "epoch": 2.8692971639950677, + "grad_norm": 2.8945017983078, + "learning_rate": 9.932737360240919e-08, + "loss": 0.4249, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2327 + }, + { + "epoch": 2.8705302096177556, + "grad_norm": 2.613387198017081, + "learning_rate": 9.74824710581035e-08, + "loss": 0.3233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2328 + }, + { + "epoch": 2.871763255240444, + "grad_norm": 3.2634500998065588, + "learning_rate": 9.565477916161248e-08, + "loss": 0.3712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2329 + }, + { + "epoch": 2.872996300863132, + "grad_norm": 2.898776109691394, + "learning_rate": 9.38443010894785e-08, + "loss": 0.3776, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2330 + }, + { + "epoch": 2.87422934648582, + "grad_norm": 2.5170635355271123, + "learning_rate": 9.205103998832566e-08, + "loss": 0.2385, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2331 + }, + { + "epoch": 2.875462392108508, + "grad_norm": 2.814264097346992, + "learning_rate": 9.027499897485526e-08, + "loss": 0.3007, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2332 + }, + { + "epoch": 2.876695437731196, + "grad_norm": 2.9787431897490397, + "learning_rate": 8.851618113583815e-08, + "loss": 0.3657, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2333 + }, + { + "epoch": 2.877928483353884, + "grad_norm": 2.9632857053233606, + "learning_rate": 8.677458952811468e-08, + "loss": 0.3024, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2334 + }, + { + "epoch": 2.879161528976572, + "grad_norm": 2.688222043176526, + "learning_rate": 8.505022717858247e-08, + "loss": 0.343, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2335 + }, + { + "epoch": 2.88039457459926, + "grad_norm": 2.7859812858572686, + "learning_rate": 8.33430970841953e-08, + "loss": 0.2982, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2336 + }, + { + "epoch": 2.8816276202219484, + "grad_norm": 2.653789668740422, + "learning_rate": 8.165320221195872e-08, + "loss": 0.2534, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2337 + }, + { + "epoch": 2.8828606658446363, + "grad_norm": 2.7169769335024045, + "learning_rate": 7.99805454989222e-08, + "loss": 0.3009, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2338 + }, + { + "epoch": 2.884093711467324, + "grad_norm": 3.155573404530684, + "learning_rate": 7.832512985217588e-08, + "loss": 0.3467, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2339 + }, + { + "epoch": 2.885326757090012, + "grad_norm": 2.32004368047518, + "learning_rate": 7.668695814884165e-08, + "loss": 0.1593, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2340 + }, + { + "epoch": 2.8865598027127004, + "grad_norm": 2.8891918362696085, + "learning_rate": 7.506603323607752e-08, + "loss": 0.3335, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2341 + }, + { + "epoch": 2.8877928483353883, + "grad_norm": 2.615539876408495, + "learning_rate": 7.346235793106115e-08, + "loss": 0.3254, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2342 + }, + { + "epoch": 2.8890258939580766, + "grad_norm": 2.638718119758046, + "learning_rate": 7.187593502099078e-08, + "loss": 0.3052, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2343 + }, + { + "epoch": 2.8902589395807645, + "grad_norm": 2.6874437167086596, + "learning_rate": 7.030676726308417e-08, + "loss": 0.2972, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2344 + }, + { + "epoch": 2.8914919852034524, + "grad_norm": 2.3740841845340923, + "learning_rate": 6.8754857384562e-08, + "loss": 0.233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2345 + }, + { + "epoch": 2.8927250308261403, + "grad_norm": 2.846355262871058, + "learning_rate": 6.722020808265894e-08, + "loss": 0.3693, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2346 + }, + { + "epoch": 2.8939580764488286, + "grad_norm": 2.4693519695538804, + "learning_rate": 6.570282202460476e-08, + "loss": 0.2799, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2347 + }, + { + "epoch": 2.8951911220715165, + "grad_norm": 2.968281861478461, + "learning_rate": 6.420270184762878e-08, + "loss": 0.2877, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2348 + }, + { + "epoch": 2.896424167694205, + "grad_norm": 2.903754378430891, + "learning_rate": 6.271985015894766e-08, + "loss": 0.3435, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2349 + }, + { + "epoch": 2.8976572133168927, + "grad_norm": 2.913197043007887, + "learning_rate": 6.125426953577318e-08, + "loss": 0.3928, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2350 + }, + { + "epoch": 2.8988902589395806, + "grad_norm": 2.457268308542506, + "learning_rate": 5.980596252529114e-08, + "loss": 0.2408, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2351 + }, + { + "epoch": 2.900123304562269, + "grad_norm": 2.414287152689473, + "learning_rate": 5.837493164467134e-08, + "loss": 0.2263, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2352 + }, + { + "epoch": 2.901356350184957, + "grad_norm": 3.2357207733394002, + "learning_rate": 5.6961179381056496e-08, + "loss": 0.5101, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2353 + }, + { + "epoch": 2.9025893958076447, + "grad_norm": 2.4452525574012713, + "learning_rate": 5.556470819155779e-08, + "loss": 0.2686, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2354 + }, + { + "epoch": 2.903822441430333, + "grad_norm": 2.682181585478765, + "learning_rate": 5.418552050325154e-08, + "loss": 0.282, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2355 + }, + { + "epoch": 2.905055487053021, + "grad_norm": 2.8764035202900393, + "learning_rate": 5.282361871317698e-08, + "loss": 0.4184, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2356 + }, + { + "epoch": 2.906288532675709, + "grad_norm": 3.086081235268078, + "learning_rate": 5.147900518832849e-08, + "loss": 0.4175, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2357 + }, + { + "epoch": 2.907521578298397, + "grad_norm": 2.7441802713817545, + "learning_rate": 5.01516822656567e-08, + "loss": 0.3591, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2358 + }, + { + "epoch": 2.908754623921085, + "grad_norm": 3.134231898861002, + "learning_rate": 4.884165225205517e-08, + "loss": 0.4485, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2359 + }, + { + "epoch": 2.909987669543773, + "grad_norm": 2.7126388402681103, + "learning_rate": 4.754891742436818e-08, + "loss": 0.3375, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2360 + }, + { + "epoch": 2.9112207151664613, + "grad_norm": 2.803937797347242, + "learning_rate": 4.627348002937626e-08, + "loss": 0.3179, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2361 + }, + { + "epoch": 2.912453760789149, + "grad_norm": 2.6367616772488938, + "learning_rate": 4.5015342283800665e-08, + "loss": 0.2715, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2362 + }, + { + "epoch": 2.913686806411837, + "grad_norm": 2.7735114367169866, + "learning_rate": 4.3774506374293365e-08, + "loss": 0.3685, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2363 + }, + { + "epoch": 2.9149198520345254, + "grad_norm": 2.8258509346206315, + "learning_rate": 4.255097445743595e-08, + "loss": 0.3056, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2364 + }, + { + "epoch": 2.9161528976572133, + "grad_norm": 3.231490891731339, + "learning_rate": 4.134474865973626e-08, + "loss": 0.3529, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.31, + "memory/max_mem_allocated(gib)": 18.33, + "step": 2365 + }, + { + "epoch": 2.917385943279901, + "grad_norm": 2.4747049805011265, + "learning_rate": 4.015583107762289e-08, + "loss": 0.2746, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2366 + }, + { + "epoch": 2.9186189889025895, + "grad_norm": 2.6336400842808856, + "learning_rate": 3.8984223777444044e-08, + "loss": 0.2888, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2367 + }, + { + "epoch": 2.9198520345252774, + "grad_norm": 2.3335120551513544, + "learning_rate": 3.782992879546088e-08, + "loss": 0.2153, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2368 + }, + { + "epoch": 2.9210850801479653, + "grad_norm": 2.561101935591807, + "learning_rate": 3.669294813784752e-08, + "loss": 0.3228, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2369 + }, + { + "epoch": 2.9223181257706536, + "grad_norm": 2.721454128282675, + "learning_rate": 3.5573283780684365e-08, + "loss": 0.3156, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2370 + }, + { + "epoch": 2.9235511713933415, + "grad_norm": 2.801677495179074, + "learning_rate": 3.4470937669955907e-08, + "loss": 0.2399, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2371 + }, + { + "epoch": 2.92478421701603, + "grad_norm": 2.537502571513842, + "learning_rate": 3.338591172154737e-08, + "loss": 0.292, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2372 + }, + { + "epoch": 2.9260172626387178, + "grad_norm": 2.9400079104711856, + "learning_rate": 3.231820782124251e-08, + "loss": 0.248, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2373 + }, + { + "epoch": 2.9272503082614056, + "grad_norm": 2.629520993372226, + "learning_rate": 3.1267827824719155e-08, + "loss": 0.2544, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2374 + }, + { + "epoch": 2.9284833538840935, + "grad_norm": 3.4111870460219325, + "learning_rate": 3.023477355754478e-08, + "loss": 0.4689, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2375 + }, + { + "epoch": 2.929716399506782, + "grad_norm": 2.778777508894336, + "learning_rate": 2.921904681517318e-08, + "loss": 0.3413, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2376 + }, + { + "epoch": 2.9309494451294698, + "grad_norm": 2.7204460257094376, + "learning_rate": 2.8220649362948882e-08, + "loss": 0.3673, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2377 + }, + { + "epoch": 2.932182490752158, + "grad_norm": 2.429633881666491, + "learning_rate": 2.7239582936090524e-08, + "loss": 0.2308, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2378 + }, + { + "epoch": 2.933415536374846, + "grad_norm": 2.691769683391875, + "learning_rate": 2.6275849239700836e-08, + "loss": 0.3466, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2379 + }, + { + "epoch": 2.934648581997534, + "grad_norm": 3.090229063494939, + "learning_rate": 2.5329449948754413e-08, + "loss": 0.3933, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2380 + }, + { + "epoch": 2.9358816276202218, + "grad_norm": 2.988766697433382, + "learning_rate": 2.4400386708101075e-08, + "loss": 0.4223, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2381 + }, + { + "epoch": 2.93711467324291, + "grad_norm": 2.9360632136985227, + "learning_rate": 2.3488661132459178e-08, + "loss": 0.3542, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2382 + }, + { + "epoch": 2.938347718865598, + "grad_norm": 2.835456114817134, + "learning_rate": 2.2594274806415628e-08, + "loss": 0.4228, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2383 + }, + { + "epoch": 2.9395807644882863, + "grad_norm": 2.6394578394291695, + "learning_rate": 2.1717229284418108e-08, + "loss": 0.2947, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2384 + }, + { + "epoch": 2.940813810110974, + "grad_norm": 3.196900724472232, + "learning_rate": 2.0857526090780622e-08, + "loss": 0.4001, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2385 + }, + { + "epoch": 2.942046855733662, + "grad_norm": 2.949209733303208, + "learning_rate": 2.0015166719671297e-08, + "loss": 0.3159, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2386 + }, + { + "epoch": 2.94327990135635, + "grad_norm": 2.7708399142568063, + "learning_rate": 1.919015263511903e-08, + "loss": 0.3368, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2387 + }, + { + "epoch": 2.9445129469790383, + "grad_norm": 3.172762758546151, + "learning_rate": 1.8382485271002392e-08, + "loss": 0.4055, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2388 + }, + { + "epoch": 2.945745992601726, + "grad_norm": 2.546861952506637, + "learning_rate": 1.759216603105407e-08, + "loss": 0.2795, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2389 + }, + { + "epoch": 2.9469790382244145, + "grad_norm": 2.4738103501045794, + "learning_rate": 1.681919628885531e-08, + "loss": 0.2601, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2390 + }, + { + "epoch": 2.9482120838471024, + "grad_norm": 3.002915986037592, + "learning_rate": 1.6063577387831487e-08, + "loss": 0.4182, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2391 + }, + { + "epoch": 2.9494451294697903, + "grad_norm": 3.20095141477087, + "learning_rate": 1.532531064125542e-08, + "loss": 0.4848, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2392 + }, + { + "epoch": 2.950678175092478, + "grad_norm": 3.426449346036598, + "learning_rate": 1.4604397332239616e-08, + "loss": 0.4655, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2393 + }, + { + "epoch": 2.9519112207151665, + "grad_norm": 2.8200758190513575, + "learning_rate": 1.3900838713736264e-08, + "loss": 0.3676, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2394 + }, + { + "epoch": 2.9531442663378544, + "grad_norm": 2.7116624282843818, + "learning_rate": 1.321463600853612e-08, + "loss": 0.3205, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2395 + }, + { + "epoch": 2.9543773119605428, + "grad_norm": 2.9299214821926336, + "learning_rate": 1.2545790409264068e-08, + "loss": 0.3712, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2396 + }, + { + "epoch": 2.9556103575832307, + "grad_norm": 2.7430681245826425, + "learning_rate": 1.1894303078378023e-08, + "loss": 0.414, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2397 + }, + { + "epoch": 2.9568434032059185, + "grad_norm": 3.2851887590121573, + "learning_rate": 1.1260175148168907e-08, + "loss": 0.45, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2398 + }, + { + "epoch": 2.9580764488286064, + "grad_norm": 2.837570696404309, + "learning_rate": 1.0643407720754007e-08, + "loss": 0.2574, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2399 + }, + { + "epoch": 2.9593094944512948, + "grad_norm": 2.3458507278581515, + "learning_rate": 1.004400186808252e-08, + "loss": 0.2036, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2400 + }, + { + "epoch": 2.9605425400739827, + "grad_norm": 2.78865642239812, + "learning_rate": 9.46195863192334e-09, + "loss": 0.3839, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2401 + }, + { + "epoch": 2.961775585696671, + "grad_norm": 2.675429682222215, + "learning_rate": 8.89727902387283e-09, + "loss": 0.3357, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2402 + }, + { + "epoch": 2.963008631319359, + "grad_norm": 2.430708350126482, + "learning_rate": 8.349964025348156e-09, + "loss": 0.2806, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2403 + }, + { + "epoch": 2.9642416769420468, + "grad_norm": 2.5755964050751174, + "learning_rate": 7.8200145875873e-09, + "loss": 0.3307, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2404 + }, + { + "epoch": 2.9654747225647347, + "grad_norm": 2.9696820033655027, + "learning_rate": 7.307431631645712e-09, + "loss": 0.3355, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2405 + }, + { + "epoch": 2.966707768187423, + "grad_norm": 2.703895979459011, + "learning_rate": 6.8122160483974354e-09, + "loss": 0.301, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2406 + }, + { + "epoch": 2.967940813810111, + "grad_norm": 2.3260593125922457, + "learning_rate": 6.334368698529547e-09, + "loss": 0.2921, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2407 + }, + { + "epoch": 2.969173859432799, + "grad_norm": 2.526282146776256, + "learning_rate": 5.873890412543271e-09, + "loss": 0.2595, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2408 + }, + { + "epoch": 2.970406905055487, + "grad_norm": 3.0153815760665594, + "learning_rate": 5.430781990756195e-09, + "loss": 0.3429, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2409 + }, + { + "epoch": 2.971639950678175, + "grad_norm": 2.9474268605575165, + "learning_rate": 5.005044203292286e-09, + "loss": 0.2852, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2410 + }, + { + "epoch": 2.972872996300863, + "grad_norm": 2.940342921639072, + "learning_rate": 4.5966777900863235e-09, + "loss": 0.4145, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2411 + }, + { + "epoch": 2.974106041923551, + "grad_norm": 3.150391812677107, + "learning_rate": 4.205683460882792e-09, + "loss": 0.4555, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2412 + }, + { + "epoch": 2.975339087546239, + "grad_norm": 3.1251534564642793, + "learning_rate": 3.832061895232553e-09, + "loss": 0.4224, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2413 + }, + { + "epoch": 2.9765721331689274, + "grad_norm": 3.179886059439302, + "learning_rate": 3.4758137424928394e-09, + "loss": 0.3382, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2414 + }, + { + "epoch": 2.9778051787916153, + "grad_norm": 3.042424095220823, + "learning_rate": 3.1369396218250414e-09, + "loss": 0.4189, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2415 + }, + { + "epoch": 2.979038224414303, + "grad_norm": 2.5441930572060314, + "learning_rate": 2.815440122195812e-09, + "loss": 0.2674, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2416 + }, + { + "epoch": 2.980271270036991, + "grad_norm": 2.1545344306090257, + "learning_rate": 2.5113158023737373e-09, + "loss": 0.176, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2417 + }, + { + "epoch": 2.9815043156596794, + "grad_norm": 2.5899779165965144, + "learning_rate": 2.224567190927118e-09, + "loss": 0.2277, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2418 + }, + { + "epoch": 2.9827373612823673, + "grad_norm": 2.561440540818118, + "learning_rate": 1.9551947862295195e-09, + "loss": 0.3699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2419 + }, + { + "epoch": 2.9839704069050557, + "grad_norm": 2.53172176100944, + "learning_rate": 1.7031990564508881e-09, + "loss": 0.3247, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2420 + }, + { + "epoch": 2.9852034525277436, + "grad_norm": 2.7497754746493692, + "learning_rate": 1.468580439560885e-09, + "loss": 0.3508, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2421 + }, + { + "epoch": 2.9864364981504314, + "grad_norm": 3.094401340403718, + "learning_rate": 1.2513393433311038e-09, + "loss": 0.3915, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2422 + }, + { + "epoch": 2.9876695437731193, + "grad_norm": 3.163763791922283, + "learning_rate": 1.051476145325081e-09, + "loss": 0.3806, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2423 + }, + { + "epoch": 2.9889025893958077, + "grad_norm": 2.945945250516526, + "learning_rate": 8.689911929082861e-10, + "loss": 0.339, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2424 + }, + { + "epoch": 2.9901356350184956, + "grad_norm": 2.8246910666829224, + "learning_rate": 7.038848032403511e-10, + "loss": 0.3233, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2425 + }, + { + "epoch": 2.991368680641184, + "grad_norm": 2.8381854402881173, + "learning_rate": 5.561572632772905e-10, + "loss": 0.3565, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2426 + }, + { + "epoch": 2.9926017262638718, + "grad_norm": 2.775613660667215, + "learning_rate": 4.2580882977039106e-10, + "loss": 0.3578, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2427 + }, + { + "epoch": 2.9938347718865597, + "grad_norm": 2.7783113806176623, + "learning_rate": 3.128397292673224e-10, + "loss": 0.1957, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2428 + }, + { + "epoch": 2.995067817509248, + "grad_norm": 2.7434381709637625, + "learning_rate": 2.1725015810880602e-10, + "loss": 0.2832, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2429 + }, + { + "epoch": 2.996300863131936, + "grad_norm": 2.536468009270092, + "learning_rate": 1.3904028242972546e-10, + "loss": 0.3078, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2430 + }, + { + "epoch": 2.9975339087546238, + "grad_norm": 2.636885958956659, + "learning_rate": 7.821023815912653e-11, + "loss": 0.2699, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2431 + }, + { + "epoch": 2.998766954377312, + "grad_norm": 2.740657558560787, + "learning_rate": 3.476013102021725e-11, + "loss": 0.3348, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2432 + }, + { + "epoch": 3.0, + "grad_norm": 3.1979379775555703, + "learning_rate": 8.69003653147793e-12, + "loss": 0.3939, + "memory/device_mem_reserved(gib)": 22.42, + "memory/max_mem_active(gib)": 19.33, + "memory/max_mem_allocated(gib)": 18.36, + "step": 2433 + } + ], + "logging_steps": 1, + "max_steps": 2433, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 406, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 3995768633425920.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-2433/training_args.bin b/checkpoint-2433/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7b675d785639ed4b99af767b6c87a1227c13ec1b --- /dev/null +++ b/checkpoint-2433/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e80ea310ffc6c44543f32aa82ecc00b1bf9d3e329881647443a895f041e17955 +size 9208 diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4c0f53b6c66a0b8a2a4eb621c949412e2a7f353a --- /dev/null +++ b/config.json @@ -0,0 +1,35 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.2", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b513e54e3195b917260c9a8a04c9f3683f19de35 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.55.2" +} diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2de9c89cd2fed0f3ae65ea5c2bafc2c1da5de87b --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d00ab5dcfd3b949feab0bfb18da3457ad431f69815f9971015d3a83d6d7ffb2b +size 1168138808 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..58a7ef40d52ab3340cf131cf33f8923f5144aaf9 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3c1d04911c269b925af977a3151c9704e990e4d0 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..c68051fe3c4d23234a59316bc52d21f6e3a4182c --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|finetune_right_pad_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7b675d785639ed4b99af767b6c87a1227c13ec1b --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e80ea310ffc6c44543f32aa82ecc00b1bf9d3e329881647443a895f041e17955 +size 9208