Upload OneReason v2 SFT model
Browse files- .gitattributes +1 -0
- README.md +60 -0
- all_results.json +8 -0
- chat_template.jinja +89 -0
- config.json +63 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +15 -0
- train_results.json +8 -0
- trainer_log.jsonl +133 -0
- trainer_state.json +967 -0
- training_args.bin +3 -0
- training_loss.png +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: other
|
| 4 |
+
tags:
|
| 5 |
+
- llama-factory
|
| 6 |
+
- full
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: sft-ds2-v2
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# sft-ds2-v2
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of a local OneReason-0.8B pretraining checkpoint on the onereason_sft and text2sid_eval_aligned_preview datasets.
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 1e-05
|
| 38 |
+
- train_batch_size: 1
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 8
|
| 43 |
+
- gradient_accumulation_steps: 8
|
| 44 |
+
- total_train_batch_size: 64
|
| 45 |
+
- total_eval_batch_size: 64
|
| 46 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_steps: 0.03
|
| 49 |
+
- num_epochs: 1.0
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- Transformers 5.3.0
|
| 58 |
+
- Pytorch 2.4.1
|
| 59 |
+
- Datasets 4.0.0
|
| 60 |
+
- Tokenizers 0.22.2
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.9898718236966912e+17,
|
| 4 |
+
"train_loss": 1.3105711570705276,
|
| 5 |
+
"train_runtime": 2573.9711,
|
| 6 |
+
"train_samples_per_second": 16.504,
|
| 7 |
+
"train_steps_per_second": 0.258
|
| 8 |
+
}
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n' }}
|
| 86 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 87 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention"
|
| 44 |
+
],
|
| 45 |
+
"max_position_embeddings": 40960,
|
| 46 |
+
"max_window_layers": 28,
|
| 47 |
+
"model_type": "qwen3",
|
| 48 |
+
"num_attention_heads": 16,
|
| 49 |
+
"num_hidden_layers": 28,
|
| 50 |
+
"num_key_value_heads": 8,
|
| 51 |
+
"pad_token_id": 151643,
|
| 52 |
+
"rms_norm_eps": 1e-06,
|
| 53 |
+
"rope_parameters": {
|
| 54 |
+
"rope_theta": 1000000,
|
| 55 |
+
"rope_type": "default"
|
| 56 |
+
},
|
| 57 |
+
"sliding_window": null,
|
| 58 |
+
"tie_word_embeddings": false,
|
| 59 |
+
"transformers_version": "5.3.0",
|
| 60 |
+
"use_cache": false,
|
| 61 |
+
"use_sliding_window": false,
|
| 62 |
+
"vocab_size": 176253
|
| 63 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"temperature": 0.6,
|
| 9 |
+
"top_k": 20,
|
| 10 |
+
"top_p": 0.95,
|
| 11 |
+
"transformers_version": "5.3.0"
|
| 12 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19bb56ba600bf8da35daaf99e5d4dde25cdc61ff47d1bef52a40dc0bdb4881f1
|
| 3 |
+
size 1602902832
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd4d15f596979aecbc11ba12668cb21c9fb8452ce1235cd9d7878cc950170421
|
| 3 |
+
size 16016554
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"is_local": true,
|
| 9 |
+
"model_max_length": 131072,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"padding_side": "right",
|
| 12 |
+
"split_special_tokens": false,
|
| 13 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 14 |
+
"unk_token": null
|
| 15 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.9898718236966912e+17,
|
| 4 |
+
"train_loss": 1.3105711570705276,
|
| 5 |
+
"train_runtime": 2573.9711,
|
| 6 |
+
"train_samples_per_second": 16.504,
|
| 7 |
+
"train_steps_per_second": 0.258
|
| 8 |
+
}
|
trainer_log.jsonl
ADDED
|
@@ -0,0 +1,133 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"current_steps": 5, "total_steps": 664, "loss": 3.0042816162109376, "lr": 2.0000000000000003e-06, "epoch": 0.007532956685499058, "percentage": 0.75, "elapsed_time": "0:00:19", "remaining_time": "0:42:00"}
|
| 2 |
+
{"current_steps": 10, "total_steps": 664, "loss": 2.759540557861328, "lr": 4.5e-06, "epoch": 0.015065913370998116, "percentage": 1.51, "elapsed_time": "0:00:38", "remaining_time": "0:42:28"}
|
| 3 |
+
{"current_steps": 15, "total_steps": 664, "loss": 2.400727462768555, "lr": 7e-06, "epoch": 0.022598870056497175, "percentage": 2.26, "elapsed_time": "0:00:59", "remaining_time": "0:43:01"}
|
| 4 |
+
{"current_steps": 20, "total_steps": 664, "loss": 2.1121063232421875, "lr": 9.5e-06, "epoch": 0.030131826741996232, "percentage": 3.01, "elapsed_time": "0:01:19", "remaining_time": "0:42:39"}
|
| 5 |
+
{"current_steps": 25, "total_steps": 664, "loss": 1.910271453857422, "lr": 9.999048137490364e-06, "epoch": 0.03766478342749529, "percentage": 3.77, "elapsed_time": "0:01:38", "remaining_time": "0:41:58"}
|
| 6 |
+
{"current_steps": 30, "total_steps": 664, "loss": 1.806014633178711, "lr": 9.995181817171016e-06, "epoch": 0.04519774011299435, "percentage": 4.52, "elapsed_time": "0:01:56", "remaining_time": "0:40:54"}
|
| 7 |
+
{"current_steps": 35, "total_steps": 664, "loss": 1.7168695449829101, "lr": 9.988343845952697e-06, "epoch": 0.05273069679849341, "percentage": 5.27, "elapsed_time": "0:02:14", "remaining_time": "0:40:26"}
|
| 8 |
+
{"current_steps": 40, "total_steps": 664, "loss": 1.6618186950683593, "lr": 9.978538291768083e-06, "epoch": 0.060263653483992465, "percentage": 6.02, "elapsed_time": "0:02:33", "remaining_time": "0:39:52"}
|
| 9 |
+
{"current_steps": 45, "total_steps": 664, "loss": 1.6461427688598633, "lr": 9.96577098797519e-06, "epoch": 0.06779661016949153, "percentage": 6.78, "elapsed_time": "0:02:54", "remaining_time": "0:39:55"}
|
| 10 |
+
{"current_steps": 50, "total_steps": 664, "loss": 1.5949559211730957, "lr": 9.950049529887098e-06, "epoch": 0.07532956685499058, "percentage": 7.53, "elapsed_time": "0:03:14", "remaining_time": "0:39:46"}
|
| 11 |
+
{"current_steps": 55, "total_steps": 664, "loss": 1.5772063255310058, "lr": 9.931383270253475e-06, "epoch": 0.08286252354048965, "percentage": 8.28, "elapsed_time": "0:03:33", "remaining_time": "0:39:27"}
|
| 12 |
+
{"current_steps": 60, "total_steps": 664, "loss": 1.5475144386291504, "lr": 9.90978331369658e-06, "epoch": 0.0903954802259887, "percentage": 9.04, "elapsed_time": "0:03:52", "remaining_time": "0:39:04"}
|
| 13 |
+
{"current_steps": 65, "total_steps": 664, "loss": 1.5141769409179688, "lr": 9.885262510105102e-06, "epoch": 0.09792843691148775, "percentage": 9.79, "elapsed_time": "0:04:12", "remaining_time": "0:38:47"}
|
| 14 |
+
{"current_steps": 70, "total_steps": 664, "loss": 1.5146191596984864, "lr": 9.857835446989708e-06, "epoch": 0.10546139359698682, "percentage": 10.54, "elapsed_time": "0:04:29", "remaining_time": "0:38:07"}
|
| 15 |
+
{"current_steps": 75, "total_steps": 664, "loss": 1.4855974197387696, "lr": 9.827518440804882e-06, "epoch": 0.11299435028248588, "percentage": 11.3, "elapsed_time": "0:04:50", "remaining_time": "0:38:03"}
|
| 16 |
+
{"current_steps": 80, "total_steps": 664, "loss": 1.4978950500488282, "lr": 9.794329527242209e-06, "epoch": 0.12052730696798493, "percentage": 12.05, "elapsed_time": "0:05:13", "remaining_time": "0:38:05"}
|
| 17 |
+
{"current_steps": 85, "total_steps": 664, "loss": 1.4637947082519531, "lr": 9.75828845050089e-06, "epoch": 0.128060263653484, "percentage": 12.8, "elapsed_time": "0:05:32", "remaining_time": "0:37:43"}
|
| 18 |
+
{"current_steps": 90, "total_steps": 664, "loss": 1.4721240043640136, "lr": 9.719416651541839e-06, "epoch": 0.13559322033898305, "percentage": 13.55, "elapsed_time": "0:05:52", "remaining_time": "0:37:30"}
|
| 19 |
+
{"current_steps": 95, "total_steps": 664, "loss": 1.444606113433838, "lr": 9.677737255332381e-06, "epoch": 0.1431261770244821, "percentage": 14.31, "elapsed_time": "0:06:13", "remaining_time": "0:37:14"}
|
| 20 |
+
{"current_steps": 100, "total_steps": 664, "loss": 1.421457862854004, "lr": 9.63327505708914e-06, "epoch": 0.15065913370998116, "percentage": 15.06, "elapsed_time": "0:06:30", "remaining_time": "0:36:43"}
|
| 21 |
+
{"current_steps": 105, "total_steps": 664, "loss": 1.4258420944213868, "lr": 9.586056507527266e-06, "epoch": 0.15819209039548024, "percentage": 15.81, "elapsed_time": "0:06:52", "remaining_time": "0:36:34"}
|
| 22 |
+
{"current_steps": 110, "total_steps": 664, "loss": 1.405913734436035, "lr": 9.536109697124822e-06, "epoch": 0.1657250470809793, "percentage": 16.57, "elapsed_time": "0:07:09", "remaining_time": "0:36:04"}
|
| 23 |
+
{"current_steps": 115, "total_steps": 664, "loss": 1.4174596786499023, "lr": 9.483464339411658e-06, "epoch": 0.17325800376647835, "percentage": 17.32, "elapsed_time": "0:07:31", "remaining_time": "0:35:56"}
|
| 24 |
+
{"current_steps": 120, "total_steps": 664, "loss": 1.4283493995666503, "lr": 9.42815175329273e-06, "epoch": 0.1807909604519774, "percentage": 18.07, "elapsed_time": "0:07:53", "remaining_time": "0:35:46"}
|
| 25 |
+
{"current_steps": 125, "total_steps": 664, "loss": 1.384162712097168, "lr": 9.370204844416381e-06, "epoch": 0.18832391713747645, "percentage": 18.83, "elapsed_time": "0:08:12", "remaining_time": "0:35:23"}
|
| 26 |
+
{"current_steps": 130, "total_steps": 664, "loss": 1.4157112121582032, "lr": 9.309658085598646e-06, "epoch": 0.1958568738229755, "percentage": 19.58, "elapsed_time": "0:08:34", "remaining_time": "0:35:11"}
|
| 27 |
+
{"current_steps": 135, "total_steps": 664, "loss": 1.3878924369812011, "lr": 9.24654749631526e-06, "epoch": 0.2033898305084746, "percentage": 20.33, "elapsed_time": "0:08:54", "remaining_time": "0:34:55"}
|
| 28 |
+
{"current_steps": 140, "total_steps": 664, "loss": 1.3945151329040528, "lr": 9.180910621273555e-06, "epoch": 0.21092278719397364, "percentage": 21.08, "elapsed_time": "0:09:11", "remaining_time": "0:34:25"}
|
| 29 |
+
{"current_steps": 145, "total_steps": 664, "loss": 1.3651309967041017, "lr": 9.112786508076972e-06, "epoch": 0.2184557438794727, "percentage": 21.84, "elapsed_time": "0:09:31", "remaining_time": "0:34:04"}
|
| 30 |
+
{"current_steps": 150, "total_steps": 664, "loss": 1.3733718872070313, "lr": 9.042215683995516e-06, "epoch": 0.22598870056497175, "percentage": 22.59, "elapsed_time": "0:09:47", "remaining_time": "0:33:31"}
|
| 31 |
+
{"current_steps": 155, "total_steps": 664, "loss": 1.3404953002929687, "lr": 8.96924013185594e-06, "epoch": 0.2335216572504708, "percentage": 23.34, "elapsed_time": "0:10:06", "remaining_time": "0:33:13"}
|
| 32 |
+
{"current_steps": 160, "total_steps": 664, "loss": 1.36053409576416, "lr": 8.893903265066011e-06, "epoch": 0.24105461393596986, "percentage": 24.1, "elapsed_time": "0:10:23", "remaining_time": "0:32:43"}
|
| 33 |
+
{"current_steps": 165, "total_steps": 664, "loss": 1.3736638069152831, "lr": 8.816249901787736e-06, "epoch": 0.24858757062146894, "percentage": 24.85, "elapsed_time": "0:10:40", "remaining_time": "0:32:16"}
|
| 34 |
+
{"current_steps": 170, "total_steps": 664, "loss": 1.301325225830078, "lr": 8.736326238274873e-06, "epoch": 0.256120527306968, "percentage": 25.6, "elapsed_time": "0:11:00", "remaining_time": "0:31:59"}
|
| 35 |
+
{"current_steps": 175, "total_steps": 664, "loss": 1.317108726501465, "lr": 8.65417982139062e-06, "epoch": 0.263653483992467, "percentage": 26.36, "elapsed_time": "0:11:23", "remaining_time": "0:31:50"}
|
| 36 |
+
{"current_steps": 180, "total_steps": 664, "loss": 1.331227684020996, "lr": 8.569859520321835e-06, "epoch": 0.2711864406779661, "percentage": 27.11, "elapsed_time": "0:11:42", "remaining_time": "0:31:29"}
|
| 37 |
+
{"current_steps": 185, "total_steps": 664, "loss": 1.360618782043457, "lr": 8.483415497506567e-06, "epoch": 0.2787193973634652, "percentage": 27.86, "elapsed_time": "0:12:02", "remaining_time": "0:31:10"}
|
| 38 |
+
{"current_steps": 190, "total_steps": 664, "loss": 1.3096031188964843, "lr": 8.394899178792262e-06, "epoch": 0.2862523540489642, "percentage": 28.61, "elapsed_time": "0:12:21", "remaining_time": "0:30:48"}
|
| 39 |
+
{"current_steps": 195, "total_steps": 664, "loss": 1.3265652656555176, "lr": 8.304363222842358e-06, "epoch": 0.2937853107344633, "percentage": 29.37, "elapsed_time": "0:12:40", "remaining_time": "0:30:28"}
|
| 40 |
+
{"current_steps": 200, "total_steps": 664, "loss": 1.3142696380615235, "lr": 8.211861489809441e-06, "epoch": 0.3013182674199623, "percentage": 30.12, "elapsed_time": "0:13:00", "remaining_time": "0:30:10"}
|
| 41 |
+
{"current_steps": 205, "total_steps": 664, "loss": 1.3162962913513183, "lr": 8.117449009293668e-06, "epoch": 0.3088512241054614, "percentage": 30.87, "elapsed_time": "0:13:17", "remaining_time": "0:29:45"}
|
| 42 |
+
{"current_steps": 210, "total_steps": 664, "loss": 1.305775260925293, "lr": 8.021181947605474e-06, "epoch": 0.3163841807909605, "percentage": 31.63, "elapsed_time": "0:13:40", "remaining_time": "0:29:34"}
|
| 43 |
+
{"current_steps": 215, "total_steps": 664, "loss": 1.3017974853515626, "lr": 7.923117574352024e-06, "epoch": 0.3239171374764595, "percentage": 32.38, "elapsed_time": "0:14:00", "remaining_time": "0:29:15"}
|
| 44 |
+
{"current_steps": 220, "total_steps": 664, "loss": 1.3172372817993163, "lr": 7.82331422836734e-06, "epoch": 0.3314500941619586, "percentage": 33.13, "elapsed_time": "0:14:19", "remaining_time": "0:28:54"}
|
| 45 |
+
{"current_steps": 225, "total_steps": 664, "loss": 1.2993927001953125, "lr": 7.721831283006323e-06, "epoch": 0.3389830508474576, "percentage": 33.89, "elapsed_time": "0:14:38", "remaining_time": "0:28:33"}
|
| 46 |
+
{"current_steps": 230, "total_steps": 664, "loss": 1.3186939239501954, "lr": 7.618729110823334e-06, "epoch": 0.3465160075329567, "percentage": 34.64, "elapsed_time": "0:14:59", "remaining_time": "0:28:16"}
|
| 47 |
+
{"current_steps": 235, "total_steps": 664, "loss": 1.2768539428710937, "lr": 7.51406904765636e-06, "epoch": 0.3540489642184557, "percentage": 35.39, "elapsed_time": "0:15:20", "remaining_time": "0:27:59"}
|
| 48 |
+
{"current_steps": 240, "total_steps": 664, "loss": 1.290719223022461, "lr": 7.40791335613811e-06, "epoch": 0.3615819209039548, "percentage": 36.14, "elapsed_time": "0:15:40", "remaining_time": "0:27:40"}
|
| 49 |
+
{"current_steps": 245, "total_steps": 664, "loss": 1.2698402404785156, "lr": 7.300325188655762e-06, "epoch": 0.3691148775894539, "percentage": 36.9, "elapsed_time": "0:16:00", "remaining_time": "0:27:21"}
|
| 50 |
+
{"current_steps": 250, "total_steps": 664, "loss": 1.2782094955444336, "lr": 7.191368549781382e-06, "epoch": 0.3766478342749529, "percentage": 37.65, "elapsed_time": "0:16:20", "remaining_time": "0:27:03"}
|
| 51 |
+
{"current_steps": 255, "total_steps": 664, "loss": 1.288484764099121, "lr": 7.0811082581953935e-06, "epoch": 0.384180790960452, "percentage": 38.4, "elapsed_time": "0:16:38", "remaining_time": "0:26:42"}
|
| 52 |
+
{"current_steps": 260, "total_steps": 664, "loss": 1.296529769897461, "lr": 6.9696099081257176e-06, "epoch": 0.391713747645951, "percentage": 39.16, "elapsed_time": "0:16:58", "remaining_time": "0:26:21"}
|
| 53 |
+
{"current_steps": 265, "total_steps": 664, "loss": 1.2570140838623047, "lr": 6.8569398303255345e-06, "epoch": 0.3992467043314501, "percentage": 39.91, "elapsed_time": "0:17:17", "remaining_time": "0:26:02"}
|
| 54 |
+
{"current_steps": 270, "total_steps": 664, "loss": 1.288723087310791, "lr": 6.743165052612899e-06, "epoch": 0.4067796610169492, "percentage": 40.66, "elapsed_time": "0:17:36", "remaining_time": "0:25:41"}
|
| 55 |
+
{"current_steps": 275, "total_steps": 664, "loss": 1.265315628051758, "lr": 6.62835325999565e-06, "epoch": 0.4143126177024482, "percentage": 41.42, "elapsed_time": "0:17:59", "remaining_time": "0:25:26"}
|
| 56 |
+
{"current_steps": 280, "total_steps": 664, "loss": 1.2478508949279785, "lr": 6.51257275440538e-06, "epoch": 0.4218455743879473, "percentage": 42.17, "elapsed_time": "0:18:18", "remaining_time": "0:25:06"}
|
| 57 |
+
{"current_steps": 285, "total_steps": 664, "loss": 1.243788719177246, "lr": 6.395892414064363e-06, "epoch": 0.4293785310734463, "percentage": 42.92, "elapsed_time": "0:18:36", "remaining_time": "0:24:45"}
|
| 58 |
+
{"current_steps": 290, "total_steps": 664, "loss": 1.2769552230834962, "lr": 6.2783816525096765e-06, "epoch": 0.4369114877589454, "percentage": 43.67, "elapsed_time": "0:18:54", "remaining_time": "0:24:22"}
|
| 59 |
+
{"current_steps": 295, "total_steps": 664, "loss": 1.2421723365783692, "lr": 6.16011037729884e-06, "epoch": 0.4444444444444444, "percentage": 44.43, "elapsed_time": "0:19:12", "remaining_time": "0:24:02"}
|
| 60 |
+
{"current_steps": 300, "total_steps": 664, "loss": 1.262312126159668, "lr": 6.041148948421575e-06, "epoch": 0.4519774011299435, "percentage": 45.18, "elapsed_time": "0:19:33", "remaining_time": "0:23:43"}
|
| 61 |
+
{"current_steps": 305, "total_steps": 664, "loss": 1.244283676147461, "lr": 5.9215681364423975e-06, "epoch": 0.4595103578154426, "percentage": 45.93, "elapsed_time": "0:19:54", "remaining_time": "0:23:25"}
|
| 62 |
+
{"current_steps": 310, "total_steps": 664, "loss": 1.2538028717041017, "lr": 5.801439080398968e-06, "epoch": 0.4670433145009416, "percentage": 46.69, "elapsed_time": "0:20:11", "remaining_time": "0:23:03"}
|
| 63 |
+
{"current_steps": 315, "total_steps": 664, "loss": 1.2467006683349608, "lr": 5.680833245481234e-06, "epoch": 0.4745762711864407, "percentage": 47.44, "elapsed_time": "0:20:30", "remaining_time": "0:22:43"}
|
| 64 |
+
{"current_steps": 320, "total_steps": 664, "loss": 1.2493385314941405, "lr": 5.559822380516539e-06, "epoch": 0.4821092278719397, "percentage": 48.19, "elapsed_time": "0:20:50", "remaining_time": "0:22:24"}
|
| 65 |
+
{"current_steps": 325, "total_steps": 664, "loss": 1.2275656700134276, "lr": 5.438478475286003e-06, "epoch": 0.4896421845574388, "percentage": 48.95, "elapsed_time": "0:21:06", "remaining_time": "0:22:00"}
|
| 66 |
+
{"current_steps": 330, "total_steps": 664, "loss": 1.201064682006836, "lr": 5.316873717697545e-06, "epoch": 0.4971751412429379, "percentage": 49.7, "elapsed_time": "0:21:29", "remaining_time": "0:21:44"}
|
| 67 |
+
{"current_steps": 335, "total_steps": 664, "loss": 1.2228626251220702, "lr": 5.19508045084105e-06, "epoch": 0.504708097928437, "percentage": 50.45, "elapsed_time": "0:21:45", "remaining_time": "0:21:22"}
|
| 68 |
+
{"current_steps": 340, "total_steps": 664, "loss": 1.1998630523681642, "lr": 5.073171129951223e-06, "epoch": 0.512241054613936, "percentage": 51.2, "elapsed_time": "0:22:03", "remaining_time": "0:21:01"}
|
| 69 |
+
{"current_steps": 345, "total_steps": 664, "loss": 1.2172012329101562, "lr": 4.95121827930371e-06, "epoch": 0.519774011299435, "percentage": 51.96, "elapsed_time": "0:22:26", "remaining_time": "0:20:44"}
|
| 70 |
+
{"current_steps": 350, "total_steps": 664, "loss": 1.2077813148498535, "lr": 4.829294449070161e-06, "epoch": 0.527306967984934, "percentage": 52.71, "elapsed_time": "0:22:43", "remaining_time": "0:20:22"}
|
| 71 |
+
{"current_steps": 355, "total_steps": 664, "loss": 1.227747344970703, "lr": 4.707472172157882e-06, "epoch": 0.5348399246704332, "percentage": 53.46, "elapsed_time": "0:22:57", "remaining_time": "0:19:59"}
|
| 72 |
+
{"current_steps": 360, "total_steps": 664, "loss": 1.2278255462646483, "lr": 4.585823921059757e-06, "epoch": 0.5423728813559322, "percentage": 54.22, "elapsed_time": "0:23:17", "remaining_time": "0:19:39"}
|
| 73 |
+
{"current_steps": 365, "total_steps": 664, "loss": 1.2057710647583009, "lr": 4.464422064740114e-06, "epoch": 0.5499058380414312, "percentage": 54.97, "elapsed_time": "0:23:31", "remaining_time": "0:19:16"}
|
| 74 |
+
{"current_steps": 370, "total_steps": 664, "loss": 1.220183753967285, "lr": 4.343338825582173e-06, "epoch": 0.5574387947269304, "percentage": 55.72, "elapsed_time": "0:23:51", "remaining_time": "0:18:57"}
|
| 75 |
+
{"current_steps": 375, "total_steps": 664, "loss": 1.2171411514282227, "lr": 4.222646236422711e-06, "epoch": 0.5649717514124294, "percentage": 56.48, "elapsed_time": "0:24:07", "remaining_time": "0:18:35"}
|
| 76 |
+
{"current_steps": 380, "total_steps": 664, "loss": 1.1992236137390138, "lr": 4.102416097699471e-06, "epoch": 0.5725047080979284, "percentage": 57.23, "elapsed_time": "0:24:26", "remaining_time": "0:18:15"}
|
| 77 |
+
{"current_steps": 385, "total_steps": 664, "loss": 1.184898567199707, "lr": 3.982719934736832e-06, "epoch": 0.5800376647834274, "percentage": 57.98, "elapsed_time": "0:24:45", "remaining_time": "0:17:56"}
|
| 78 |
+
{"current_steps": 390, "total_steps": 664, "loss": 1.1996670722961427, "lr": 3.8636289551951545e-06, "epoch": 0.5875706214689266, "percentage": 58.73, "elapsed_time": "0:25:03", "remaining_time": "0:17:36"}
|
| 79 |
+
{"current_steps": 395, "total_steps": 664, "loss": 1.2165474891662598, "lr": 3.7452140067090936e-06, "epoch": 0.5951035781544256, "percentage": 59.49, "elapsed_time": "0:25:24", "remaining_time": "0:17:18"}
|
| 80 |
+
{"current_steps": 400, "total_steps": 664, "loss": 1.1614250183105468, "lr": 3.6275455347400952e-06, "epoch": 0.6026365348399246, "percentage": 60.24, "elapsed_time": "0:25:42", "remaining_time": "0:16:58"}
|
| 81 |
+
{"current_steps": 405, "total_steps": 664, "loss": 1.182399368286133, "lr": 3.510693540668151e-06, "epoch": 0.6101694915254238, "percentage": 60.99, "elapsed_time": "0:26:03", "remaining_time": "0:16:39"}
|
| 82 |
+
{"current_steps": 410, "total_steps": 664, "loss": 1.1581405639648437, "lr": 3.394727540147732e-06, "epoch": 0.6177024482109228, "percentage": 61.75, "elapsed_time": "0:26:23", "remaining_time": "0:16:20"}
|
| 83 |
+
{"current_steps": 415, "total_steps": 664, "loss": 1.2053539276123046, "lr": 3.2797165217526915e-06, "epoch": 0.6252354048964218, "percentage": 62.5, "elapsed_time": "0:26:44", "remaining_time": "0:16:02"}
|
| 84 |
+
{"current_steps": 420, "total_steps": 664, "loss": 1.172200870513916, "lr": 3.1657289059347184e-06, "epoch": 0.632768361581921, "percentage": 63.25, "elapsed_time": "0:27:04", "remaining_time": "0:15:44"}
|
| 85 |
+
{"current_steps": 425, "total_steps": 664, "loss": 1.1764010429382323, "lr": 3.0528325043197786e-06, "epoch": 0.64030131826742, "percentage": 64.01, "elapsed_time": "0:27:21", "remaining_time": "0:15:23"}
|
| 86 |
+
{"current_steps": 430, "total_steps": 664, "loss": 1.192509651184082, "lr": 2.941094479366744e-06, "epoch": 0.647834274952919, "percentage": 64.76, "elapsed_time": "0:27:42", "remaining_time": "0:15:04"}
|
| 87 |
+
{"current_steps": 435, "total_steps": 664, "loss": 1.1889172554016114, "lr": 2.83058130441221e-06, "epoch": 0.655367231638418, "percentage": 65.51, "elapsed_time": "0:28:02", "remaining_time": "0:14:45"}
|
| 88 |
+
{"current_steps": 440, "total_steps": 664, "loss": 1.1603904724121095, "lr": 2.721358724125287e-06, "epoch": 0.6629001883239172, "percentage": 66.27, "elapsed_time": "0:28:24", "remaining_time": "0:14:27"}
|
| 89 |
+
{"current_steps": 445, "total_steps": 664, "loss": 1.1712430953979491, "lr": 2.613491715395861e-06, "epoch": 0.6704331450094162, "percentage": 67.02, "elapsed_time": "0:28:45", "remaining_time": "0:14:09"}
|
| 90 |
+
{"current_steps": 450, "total_steps": 664, "loss": 1.1785946846008302, "lr": 2.507044448679621e-06, "epoch": 0.6779661016949152, "percentage": 67.77, "elapsed_time": "0:29:02", "remaining_time": "0:13:48"}
|
| 91 |
+
{"current_steps": 455, "total_steps": 664, "loss": 1.1463645935058593, "lr": 2.4020802498228333e-06, "epoch": 0.6854990583804144, "percentage": 68.52, "elapsed_time": "0:29:22", "remaining_time": "0:13:29"}
|
| 92 |
+
{"current_steps": 460, "total_steps": 664, "loss": 1.1732158660888672, "lr": 2.2986615623895774e-06, "epoch": 0.6930320150659134, "percentage": 69.28, "elapsed_time": "0:29:42", "remaining_time": "0:13:10"}
|
| 93 |
+
{"current_steps": 465, "total_steps": 664, "loss": 1.136567974090576, "lr": 2.196849910513858e-06, "epoch": 0.7005649717514124, "percentage": 70.03, "elapsed_time": "0:29:58", "remaining_time": "0:12:49"}
|
| 94 |
+
{"current_steps": 470, "total_steps": 664, "loss": 1.1899651527404784, "lr": 2.096705862298676e-06, "epoch": 0.7080979284369114, "percentage": 70.78, "elapsed_time": "0:30:18", "remaining_time": "0:12:30"}
|
| 95 |
+
{"current_steps": 475, "total_steps": 664, "loss": 1.1970231056213378, "lr": 1.9982889937838635e-06, "epoch": 0.7156308851224106, "percentage": 71.54, "elapsed_time": "0:30:37", "remaining_time": "0:12:11"}
|
| 96 |
+
{"current_steps": 480, "total_steps": 664, "loss": 1.187626838684082, "lr": 1.901657853504088e-06, "epoch": 0.7231638418079096, "percentage": 72.29, "elapsed_time": "0:30:55", "remaining_time": "0:11:51"}
|
| 97 |
+
{"current_steps": 485, "total_steps": 664, "loss": 1.1518936157226562, "lr": 1.8068699276581286e-06, "epoch": 0.7306967984934086, "percentage": 73.04, "elapsed_time": "0:31:17", "remaining_time": "0:11:32"}
|
| 98 |
+
{"current_steps": 490, "total_steps": 664, "loss": 1.169551467895508, "lr": 1.7139816059101372e-06, "epoch": 0.7382297551789078, "percentage": 73.8, "elapsed_time": "0:31:34", "remaining_time": "0:11:12"}
|
| 99 |
+
{"current_steps": 495, "total_steps": 664, "loss": 1.1763624191284179, "lr": 1.6230481478432364e-06, "epoch": 0.7457627118644068, "percentage": 74.55, "elapsed_time": "0:31:55", "remaining_time": "0:10:53"}
|
| 100 |
+
{"current_steps": 500, "total_steps": 664, "loss": 1.1237783432006836, "lr": 1.5341236500853963e-06, "epoch": 0.7532956685499058, "percentage": 75.3, "elapsed_time": "0:32:16", "remaining_time": "0:10:35"}
|
| 101 |
+
{"current_steps": 505, "total_steps": 664, "loss": 1.1419706344604492, "lr": 1.447261014127167e-06, "epoch": 0.7608286252354048, "percentage": 76.05, "elapsed_time": "0:32:38", "remaining_time": "0:10:16"}
|
| 102 |
+
{"current_steps": 510, "total_steps": 664, "loss": 1.1483024597167968, "lr": 1.3625119148504013e-06, "epoch": 0.768361581920904, "percentage": 76.81, "elapsed_time": "0:32:54", "remaining_time": "0:09:56"}
|
| 103 |
+
{"current_steps": 515, "total_steps": 664, "loss": 1.1998208999633788, "lr": 1.2799267697866868e-06, "epoch": 0.775894538606403, "percentage": 77.56, "elapsed_time": "0:33:12", "remaining_time": "0:09:36"}
|
| 104 |
+
{"current_steps": 520, "total_steps": 664, "loss": 1.1890599250793457, "lr": 1.1995547091237814e-06, "epoch": 0.783427495291902, "percentage": 78.31, "elapsed_time": "0:33:31", "remaining_time": "0:09:17"}
|
| 105 |
+
{"current_steps": 525, "total_steps": 664, "loss": 1.1614674568176269, "lr": 1.1214435464779006e-06, "epoch": 0.7909604519774012, "percentage": 79.07, "elapsed_time": "0:33:49", "remaining_time": "0:08:57"}
|
| 106 |
+
{"current_steps": 530, "total_steps": 664, "loss": 1.1164636611938477, "lr": 1.045639750449225e-06, "epoch": 0.7984934086629002, "percentage": 79.82, "elapsed_time": "0:34:05", "remaining_time": "0:08:37"}
|
| 107 |
+
{"current_steps": 535, "total_steps": 664, "loss": 1.1408018112182616, "lr": 9.721884169775735e-07, "epoch": 0.8060263653483992, "percentage": 80.57, "elapsed_time": "0:34:21", "remaining_time": "0:08:17"}
|
| 108 |
+
{"current_steps": 540, "total_steps": 664, "loss": 1.1225814819335938, "lr": 9.011332425146734e-07, "epoch": 0.8135593220338984, "percentage": 81.33, "elapsed_time": "0:34:38", "remaining_time": "0:07:57"}
|
| 109 |
+
{"current_steps": 545, "total_steps": 664, "loss": 1.1628660202026366, "lr": 8.325164980289896e-07, "epoch": 0.8210922787193974, "percentage": 82.08, "elapsed_time": "0:34:56", "remaining_time": "0:07:37"}
|
| 110 |
+
{"current_steps": 550, "total_steps": 664, "loss": 1.1538756370544434, "lr": 7.663790038585794e-07, "epoch": 0.8286252354048964, "percentage": 82.83, "elapsed_time": "0:35:13", "remaining_time": "0:07:18"}
|
| 111 |
+
{"current_steps": 555, "total_steps": 664, "loss": 1.1323760986328124, "lr": 7.027601054269329e-07, "epoch": 0.8361581920903954, "percentage": 83.58, "elapsed_time": "0:35:34", "remaining_time": "0:06:59"}
|
| 112 |
+
{"current_steps": 560, "total_steps": 664, "loss": 1.1465997695922852, "lr": 6.416976498362432e-07, "epoch": 0.8436911487758946, "percentage": 84.34, "elapsed_time": "0:35:54", "remaining_time": "0:06:40"}
|
| 113 |
+
{"current_steps": 565, "total_steps": 664, "loss": 1.1509078025817872, "lr": 5.832279633520411e-07, "epoch": 0.8512241054613936, "percentage": 85.09, "elapsed_time": "0:36:15", "remaining_time": "0:06:21"}
|
| 114 |
+
{"current_steps": 570, "total_steps": 664, "loss": 1.1476680755615234, "lr": 5.273858297925649e-07, "epoch": 0.8587570621468926, "percentage": 85.84, "elapsed_time": "0:36:35", "remaining_time": "0:06:02"}
|
| 115 |
+
{"current_steps": 575, "total_steps": 664, "loss": 1.1520160675048827, "lr": 4.7420446983575286e-07, "epoch": 0.8662900188323918, "percentage": 86.6, "elapsed_time": "0:36:52", "remaining_time": "0:05:42"}
|
| 116 |
+
{"current_steps": 580, "total_steps": 664, "loss": 1.1053319931030274, "lr": 4.2371552125614614e-07, "epoch": 0.8738229755178908, "percentage": 87.35, "elapsed_time": "0:37:10", "remaining_time": "0:05:23"}
|
| 117 |
+
{"current_steps": 585, "total_steps": 664, "loss": 1.1290611267089843, "lr": 3.759490201034677e-07, "epoch": 0.8813559322033898, "percentage": 88.1, "elapsed_time": "0:37:31", "remaining_time": "0:05:04"}
|
| 118 |
+
{"current_steps": 590, "total_steps": 664, "loss": 1.1444223403930665, "lr": 3.309333828340722e-07, "epoch": 0.8888888888888888, "percentage": 88.86, "elapsed_time": "0:37:52", "remaining_time": "0:04:45"}
|
| 119 |
+
{"current_steps": 595, "total_steps": 664, "loss": 1.1631834030151367, "lr": 2.88695389405898e-07, "epoch": 0.896421845574388, "percentage": 89.61, "elapsed_time": "0:38:15", "remaining_time": "0:04:26"}
|
| 120 |
+
{"current_steps": 600, "total_steps": 664, "loss": 1.139974021911621, "lr": 2.49260167346978e-07, "epoch": 0.903954802259887, "percentage": 90.36, "elapsed_time": "0:38:33", "remaining_time": "0:04:06"}
|
| 121 |
+
{"current_steps": 605, "total_steps": 664, "loss": 1.1391963958740234, "lr": 2.1265117680698665e-07, "epoch": 0.911487758945386, "percentage": 91.11, "elapsed_time": "0:38:52", "remaining_time": "0:03:47"}
|
| 122 |
+
{"current_steps": 610, "total_steps": 664, "loss": 1.1376457214355469, "lr": 1.7889019660071728e-07, "epoch": 0.9190207156308852, "percentage": 91.87, "elapsed_time": "0:39:12", "remaining_time": "0:03:28"}
|
| 123 |
+
{"current_steps": 615, "total_steps": 664, "loss": 1.1459515571594239, "lr": 1.479973112517863e-07, "epoch": 0.9265536723163842, "percentage": 92.62, "elapsed_time": "0:39:29", "remaining_time": "0:03:08"}
|
| 124 |
+
{"current_steps": 620, "total_steps": 664, "loss": 1.1480701446533204, "lr": 1.1999089904428385e-07, "epoch": 0.9340866290018832, "percentage": 93.37, "elapsed_time": "0:39:51", "remaining_time": "0:02:49"}
|
| 125 |
+
{"current_steps": 625, "total_steps": 664, "loss": 1.1705062866210938, "lr": 9.488762108946891e-08, "epoch": 0.9416195856873822, "percentage": 94.13, "elapsed_time": "0:40:10", "remaining_time": "0:02:30"}
|
| 126 |
+
{"current_steps": 630, "total_steps": 664, "loss": 1.1306757926940918, "lr": 7.270241141401568e-08, "epoch": 0.9491525423728814, "percentage": 94.88, "elapsed_time": "0:40:31", "remaining_time": "0:02:11"}
|
| 127 |
+
{"current_steps": 635, "total_steps": 664, "loss": 1.1814102172851562, "lr": 5.34484680757108e-08, "epoch": 0.9566854990583804, "percentage": 95.63, "elapsed_time": "0:40:48", "remaining_time": "0:01:51"}
|
| 128 |
+
{"current_steps": 640, "total_steps": 664, "loss": 1.1623525619506836, "lr": 3.713724531188234e-08, "epoch": 0.9642184557438794, "percentage": 96.39, "elapsed_time": "0:41:08", "remaining_time": "0:01:32"}
|
| 129 |
+
{"current_steps": 645, "total_steps": 664, "loss": 1.1566267013549805, "lr": 2.3778446725233328e-08, "epoch": 0.9717514124293786, "percentage": 97.14, "elapsed_time": "0:41:26", "remaining_time": "0:01:13"}
|
| 130 |
+
{"current_steps": 650, "total_steps": 664, "loss": 1.1434457778930665, "lr": 1.338001951113943e-08, "epoch": 0.9792843691148776, "percentage": 97.89, "elapsed_time": "0:41:48", "remaining_time": "0:00:54"}
|
| 131 |
+
{"current_steps": 655, "total_steps": 664, "loss": 1.129819679260254, "lr": 5.9481497298291556e-09, "epoch": 0.9868173258003766, "percentage": 98.64, "elapsed_time": "0:42:08", "remaining_time": "0:00:34"}
|
| 132 |
+
{"current_steps": 660, "total_steps": 664, "loss": 1.1079357147216797, "lr": 1.4872586262792975e-09, "epoch": 0.9943502824858758, "percentage": 99.4, "elapsed_time": "0:42:23", "remaining_time": "0:00:15"}
|
| 133 |
+
{"current_steps": 664, "total_steps": 664, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:42:53", "remaining_time": "0:00:00"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,967 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 664,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007532956685499058,
|
| 14 |
+
"grad_norm": 45.83114242553711,
|
| 15 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 16 |
+
"loss": 3.0042816162109376,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.015065913370998116,
|
| 21 |
+
"grad_norm": 7.077414035797119,
|
| 22 |
+
"learning_rate": 4.5e-06,
|
| 23 |
+
"loss": 2.759540557861328,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.022598870056497175,
|
| 28 |
+
"grad_norm": 8.423027992248535,
|
| 29 |
+
"learning_rate": 7e-06,
|
| 30 |
+
"loss": 2.400727462768555,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.030131826741996232,
|
| 35 |
+
"grad_norm": 4.605725288391113,
|
| 36 |
+
"learning_rate": 9.5e-06,
|
| 37 |
+
"loss": 2.1121063232421875,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03766478342749529,
|
| 42 |
+
"grad_norm": 3.3838229179382324,
|
| 43 |
+
"learning_rate": 9.999048137490364e-06,
|
| 44 |
+
"loss": 1.910271453857422,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04519774011299435,
|
| 49 |
+
"grad_norm": 2.8471519947052,
|
| 50 |
+
"learning_rate": 9.995181817171016e-06,
|
| 51 |
+
"loss": 1.806014633178711,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05273069679849341,
|
| 56 |
+
"grad_norm": 2.7457709312438965,
|
| 57 |
+
"learning_rate": 9.988343845952697e-06,
|
| 58 |
+
"loss": 1.7168695449829101,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.060263653483992465,
|
| 63 |
+
"grad_norm": 2.497147560119629,
|
| 64 |
+
"learning_rate": 9.978538291768083e-06,
|
| 65 |
+
"loss": 1.6618186950683593,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.06779661016949153,
|
| 70 |
+
"grad_norm": 2.694253444671631,
|
| 71 |
+
"learning_rate": 9.96577098797519e-06,
|
| 72 |
+
"loss": 1.6461427688598633,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07532956685499058,
|
| 77 |
+
"grad_norm": 2.6964664459228516,
|
| 78 |
+
"learning_rate": 9.950049529887098e-06,
|
| 79 |
+
"loss": 1.5949559211730957,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08286252354048965,
|
| 84 |
+
"grad_norm": 3.099248170852661,
|
| 85 |
+
"learning_rate": 9.931383270253475e-06,
|
| 86 |
+
"loss": 1.5772063255310058,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0903954802259887,
|
| 91 |
+
"grad_norm": 2.4634487628936768,
|
| 92 |
+
"learning_rate": 9.90978331369658e-06,
|
| 93 |
+
"loss": 1.5475144386291504,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.09792843691148775,
|
| 98 |
+
"grad_norm": 2.647801637649536,
|
| 99 |
+
"learning_rate": 9.885262510105102e-06,
|
| 100 |
+
"loss": 1.5141769409179688,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.10546139359698682,
|
| 105 |
+
"grad_norm": 2.6466736793518066,
|
| 106 |
+
"learning_rate": 9.857835446989708e-06,
|
| 107 |
+
"loss": 1.5146191596984864,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11299435028248588,
|
| 112 |
+
"grad_norm": 2.750391721725464,
|
| 113 |
+
"learning_rate": 9.827518440804882e-06,
|
| 114 |
+
"loss": 1.4855974197387696,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12052730696798493,
|
| 119 |
+
"grad_norm": 2.5691773891448975,
|
| 120 |
+
"learning_rate": 9.794329527242209e-06,
|
| 121 |
+
"loss": 1.4978950500488282,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.128060263653484,
|
| 126 |
+
"grad_norm": 2.5253233909606934,
|
| 127 |
+
"learning_rate": 9.75828845050089e-06,
|
| 128 |
+
"loss": 1.4637947082519531,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.13559322033898305,
|
| 133 |
+
"grad_norm": 2.7029972076416016,
|
| 134 |
+
"learning_rate": 9.719416651541839e-06,
|
| 135 |
+
"loss": 1.4721240043640136,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1431261770244821,
|
| 140 |
+
"grad_norm": 2.485734701156616,
|
| 141 |
+
"learning_rate": 9.677737255332381e-06,
|
| 142 |
+
"loss": 1.444606113433838,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.15065913370998116,
|
| 147 |
+
"grad_norm": 2.5694262981414795,
|
| 148 |
+
"learning_rate": 9.63327505708914e-06,
|
| 149 |
+
"loss": 1.421457862854004,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.15819209039548024,
|
| 154 |
+
"grad_norm": 2.711113452911377,
|
| 155 |
+
"learning_rate": 9.586056507527266e-06,
|
| 156 |
+
"loss": 1.4258420944213868,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1657250470809793,
|
| 161 |
+
"grad_norm": 2.6699893474578857,
|
| 162 |
+
"learning_rate": 9.536109697124822e-06,
|
| 163 |
+
"loss": 1.405913734436035,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.17325800376647835,
|
| 168 |
+
"grad_norm": 2.531147003173828,
|
| 169 |
+
"learning_rate": 9.483464339411658e-06,
|
| 170 |
+
"loss": 1.4174596786499023,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.1807909604519774,
|
| 175 |
+
"grad_norm": 2.605435609817505,
|
| 176 |
+
"learning_rate": 9.42815175329273e-06,
|
| 177 |
+
"loss": 1.4283493995666503,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.18832391713747645,
|
| 182 |
+
"grad_norm": 2.8733339309692383,
|
| 183 |
+
"learning_rate": 9.370204844416381e-06,
|
| 184 |
+
"loss": 1.384162712097168,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1958568738229755,
|
| 189 |
+
"grad_norm": 2.715965747833252,
|
| 190 |
+
"learning_rate": 9.309658085598646e-06,
|
| 191 |
+
"loss": 1.4157112121582032,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2033898305084746,
|
| 196 |
+
"grad_norm": 2.880955934524536,
|
| 197 |
+
"learning_rate": 9.24654749631526e-06,
|
| 198 |
+
"loss": 1.3878924369812011,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.21092278719397364,
|
| 203 |
+
"grad_norm": 2.909132242202759,
|
| 204 |
+
"learning_rate": 9.180910621273555e-06,
|
| 205 |
+
"loss": 1.3945151329040528,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2184557438794727,
|
| 210 |
+
"grad_norm": 2.7461202144622803,
|
| 211 |
+
"learning_rate": 9.112786508076972e-06,
|
| 212 |
+
"loss": 1.3651309967041017,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.22598870056497175,
|
| 217 |
+
"grad_norm": 2.803957223892212,
|
| 218 |
+
"learning_rate": 9.042215683995516e-06,
|
| 219 |
+
"loss": 1.3733718872070313,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2335216572504708,
|
| 224 |
+
"grad_norm": 2.626370668411255,
|
| 225 |
+
"learning_rate": 8.96924013185594e-06,
|
| 226 |
+
"loss": 1.3404953002929687,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.24105461393596986,
|
| 231 |
+
"grad_norm": 2.6976120471954346,
|
| 232 |
+
"learning_rate": 8.893903265066011e-06,
|
| 233 |
+
"loss": 1.36053409576416,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.24858757062146894,
|
| 238 |
+
"grad_norm": 2.655586004257202,
|
| 239 |
+
"learning_rate": 8.816249901787736e-06,
|
| 240 |
+
"loss": 1.3736638069152831,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.256120527306968,
|
| 245 |
+
"grad_norm": 2.355863571166992,
|
| 246 |
+
"learning_rate": 8.736326238274873e-06,
|
| 247 |
+
"loss": 1.301325225830078,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.263653483992467,
|
| 252 |
+
"grad_norm": 2.7942354679107666,
|
| 253 |
+
"learning_rate": 8.65417982139062e-06,
|
| 254 |
+
"loss": 1.317108726501465,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2711864406779661,
|
| 259 |
+
"grad_norm": 2.784876823425293,
|
| 260 |
+
"learning_rate": 8.569859520321835e-06,
|
| 261 |
+
"loss": 1.331227684020996,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2787193973634652,
|
| 266 |
+
"grad_norm": 2.847752809524536,
|
| 267 |
+
"learning_rate": 8.483415497506567e-06,
|
| 268 |
+
"loss": 1.360618782043457,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2862523540489642,
|
| 273 |
+
"grad_norm": 2.705840826034546,
|
| 274 |
+
"learning_rate": 8.394899178792262e-06,
|
| 275 |
+
"loss": 1.3096031188964843,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.2937853107344633,
|
| 280 |
+
"grad_norm": 2.3822481632232666,
|
| 281 |
+
"learning_rate": 8.304363222842358e-06,
|
| 282 |
+
"loss": 1.3265652656555176,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3013182674199623,
|
| 287 |
+
"grad_norm": 2.9162850379943848,
|
| 288 |
+
"learning_rate": 8.211861489809441e-06,
|
| 289 |
+
"loss": 1.3142696380615235,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3088512241054614,
|
| 294 |
+
"grad_norm": 2.625302791595459,
|
| 295 |
+
"learning_rate": 8.117449009293668e-06,
|
| 296 |
+
"loss": 1.3162962913513183,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3163841807909605,
|
| 301 |
+
"grad_norm": 2.6135404109954834,
|
| 302 |
+
"learning_rate": 8.021181947605474e-06,
|
| 303 |
+
"loss": 1.305775260925293,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3239171374764595,
|
| 308 |
+
"grad_norm": 2.6781067848205566,
|
| 309 |
+
"learning_rate": 7.923117574352024e-06,
|
| 310 |
+
"loss": 1.3017974853515626,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3314500941619586,
|
| 315 |
+
"grad_norm": 2.6843442916870117,
|
| 316 |
+
"learning_rate": 7.82331422836734e-06,
|
| 317 |
+
"loss": 1.3172372817993163,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3389830508474576,
|
| 322 |
+
"grad_norm": 2.6892807483673096,
|
| 323 |
+
"learning_rate": 7.721831283006323e-06,
|
| 324 |
+
"loss": 1.2993927001953125,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3465160075329567,
|
| 329 |
+
"grad_norm": 2.536848306655884,
|
| 330 |
+
"learning_rate": 7.618729110823334e-06,
|
| 331 |
+
"loss": 1.3186939239501954,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3540489642184557,
|
| 336 |
+
"grad_norm": 2.997406005859375,
|
| 337 |
+
"learning_rate": 7.51406904765636e-06,
|
| 338 |
+
"loss": 1.2768539428710937,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.3615819209039548,
|
| 343 |
+
"grad_norm": 2.635805606842041,
|
| 344 |
+
"learning_rate": 7.40791335613811e-06,
|
| 345 |
+
"loss": 1.290719223022461,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3691148775894539,
|
| 350 |
+
"grad_norm": 2.787654161453247,
|
| 351 |
+
"learning_rate": 7.300325188655762e-06,
|
| 352 |
+
"loss": 1.2698402404785156,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.3766478342749529,
|
| 357 |
+
"grad_norm": 2.8937535285949707,
|
| 358 |
+
"learning_rate": 7.191368549781382e-06,
|
| 359 |
+
"loss": 1.2782094955444336,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.384180790960452,
|
| 364 |
+
"grad_norm": 2.4497976303100586,
|
| 365 |
+
"learning_rate": 7.0811082581953935e-06,
|
| 366 |
+
"loss": 1.288484764099121,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.391713747645951,
|
| 371 |
+
"grad_norm": 2.651625394821167,
|
| 372 |
+
"learning_rate": 6.9696099081257176e-06,
|
| 373 |
+
"loss": 1.296529769897461,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.3992467043314501,
|
| 378 |
+
"grad_norm": 2.577359199523926,
|
| 379 |
+
"learning_rate": 6.8569398303255345e-06,
|
| 380 |
+
"loss": 1.2570140838623047,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4067796610169492,
|
| 385 |
+
"grad_norm": 2.638181209564209,
|
| 386 |
+
"learning_rate": 6.743165052612899e-06,
|
| 387 |
+
"loss": 1.288723087310791,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4143126177024482,
|
| 392 |
+
"grad_norm": 2.904635190963745,
|
| 393 |
+
"learning_rate": 6.62835325999565e-06,
|
| 394 |
+
"loss": 1.265315628051758,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4218455743879473,
|
| 399 |
+
"grad_norm": 2.7123067378997803,
|
| 400 |
+
"learning_rate": 6.51257275440538e-06,
|
| 401 |
+
"loss": 1.2478508949279785,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4293785310734463,
|
| 406 |
+
"grad_norm": 2.958916425704956,
|
| 407 |
+
"learning_rate": 6.395892414064363e-06,
|
| 408 |
+
"loss": 1.243788719177246,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4369114877589454,
|
| 413 |
+
"grad_norm": 3.0284829139709473,
|
| 414 |
+
"learning_rate": 6.2783816525096765e-06,
|
| 415 |
+
"loss": 1.2769552230834962,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.4444444444444444,
|
| 420 |
+
"grad_norm": 2.736923933029175,
|
| 421 |
+
"learning_rate": 6.16011037729884e-06,
|
| 422 |
+
"loss": 1.2421723365783692,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.4519774011299435,
|
| 427 |
+
"grad_norm": 2.8373374938964844,
|
| 428 |
+
"learning_rate": 6.041148948421575e-06,
|
| 429 |
+
"loss": 1.262312126159668,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4595103578154426,
|
| 434 |
+
"grad_norm": 2.320935010910034,
|
| 435 |
+
"learning_rate": 5.9215681364423975e-06,
|
| 436 |
+
"loss": 1.244283676147461,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.4670433145009416,
|
| 441 |
+
"grad_norm": 2.490412473678589,
|
| 442 |
+
"learning_rate": 5.801439080398968e-06,
|
| 443 |
+
"loss": 1.2538028717041017,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.4745762711864407,
|
| 448 |
+
"grad_norm": 3.083692789077759,
|
| 449 |
+
"learning_rate": 5.680833245481234e-06,
|
| 450 |
+
"loss": 1.2467006683349608,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.4821092278719397,
|
| 455 |
+
"grad_norm": 2.6125857830047607,
|
| 456 |
+
"learning_rate": 5.559822380516539e-06,
|
| 457 |
+
"loss": 1.2493385314941405,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.4896421845574388,
|
| 462 |
+
"grad_norm": 2.736588478088379,
|
| 463 |
+
"learning_rate": 5.438478475286003e-06,
|
| 464 |
+
"loss": 1.2275656700134276,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.4971751412429379,
|
| 469 |
+
"grad_norm": 2.8685970306396484,
|
| 470 |
+
"learning_rate": 5.316873717697545e-06,
|
| 471 |
+
"loss": 1.201064682006836,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.504708097928437,
|
| 476 |
+
"grad_norm": 2.879923105239868,
|
| 477 |
+
"learning_rate": 5.19508045084105e-06,
|
| 478 |
+
"loss": 1.2228626251220702,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.512241054613936,
|
| 483 |
+
"grad_norm": 3.2230188846588135,
|
| 484 |
+
"learning_rate": 5.073171129951223e-06,
|
| 485 |
+
"loss": 1.1998630523681642,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.519774011299435,
|
| 490 |
+
"grad_norm": 2.8035128116607666,
|
| 491 |
+
"learning_rate": 4.95121827930371e-06,
|
| 492 |
+
"loss": 1.2172012329101562,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.527306967984934,
|
| 497 |
+
"grad_norm": 2.6279988288879395,
|
| 498 |
+
"learning_rate": 4.829294449070161e-06,
|
| 499 |
+
"loss": 1.2077813148498535,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5348399246704332,
|
| 504 |
+
"grad_norm": 2.972382068634033,
|
| 505 |
+
"learning_rate": 4.707472172157882e-06,
|
| 506 |
+
"loss": 1.227747344970703,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5423728813559322,
|
| 511 |
+
"grad_norm": 2.6692798137664795,
|
| 512 |
+
"learning_rate": 4.585823921059757e-06,
|
| 513 |
+
"loss": 1.2278255462646483,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5499058380414312,
|
| 518 |
+
"grad_norm": 2.732722043991089,
|
| 519 |
+
"learning_rate": 4.464422064740114e-06,
|
| 520 |
+
"loss": 1.2057710647583009,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.5574387947269304,
|
| 525 |
+
"grad_norm": 2.7024922370910645,
|
| 526 |
+
"learning_rate": 4.343338825582173e-06,
|
| 527 |
+
"loss": 1.220183753967285,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.5649717514124294,
|
| 532 |
+
"grad_norm": 3.1108171939849854,
|
| 533 |
+
"learning_rate": 4.222646236422711e-06,
|
| 534 |
+
"loss": 1.2171411514282227,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5725047080979284,
|
| 539 |
+
"grad_norm": 2.7390384674072266,
|
| 540 |
+
"learning_rate": 4.102416097699471e-06,
|
| 541 |
+
"loss": 1.1992236137390138,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.5800376647834274,
|
| 546 |
+
"grad_norm": 2.863008975982666,
|
| 547 |
+
"learning_rate": 3.982719934736832e-06,
|
| 548 |
+
"loss": 1.184898567199707,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.5875706214689266,
|
| 553 |
+
"grad_norm": 3.165865182876587,
|
| 554 |
+
"learning_rate": 3.8636289551951545e-06,
|
| 555 |
+
"loss": 1.1996670722961427,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.5951035781544256,
|
| 560 |
+
"grad_norm": 3.1176416873931885,
|
| 561 |
+
"learning_rate": 3.7452140067090936e-06,
|
| 562 |
+
"loss": 1.2165474891662598,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6026365348399246,
|
| 567 |
+
"grad_norm": 2.7519948482513428,
|
| 568 |
+
"learning_rate": 3.6275455347400952e-06,
|
| 569 |
+
"loss": 1.1614250183105468,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6101694915254238,
|
| 574 |
+
"grad_norm": 2.8069188594818115,
|
| 575 |
+
"learning_rate": 3.510693540668151e-06,
|
| 576 |
+
"loss": 1.182399368286133,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6177024482109228,
|
| 581 |
+
"grad_norm": 3.1121582984924316,
|
| 582 |
+
"learning_rate": 3.394727540147732e-06,
|
| 583 |
+
"loss": 1.1581405639648437,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6252354048964218,
|
| 588 |
+
"grad_norm": 2.7260525226593018,
|
| 589 |
+
"learning_rate": 3.2797165217526915e-06,
|
| 590 |
+
"loss": 1.2053539276123046,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.632768361581921,
|
| 595 |
+
"grad_norm": 2.942034959793091,
|
| 596 |
+
"learning_rate": 3.1657289059347184e-06,
|
| 597 |
+
"loss": 1.172200870513916,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.64030131826742,
|
| 602 |
+
"grad_norm": 3.039311647415161,
|
| 603 |
+
"learning_rate": 3.0528325043197786e-06,
|
| 604 |
+
"loss": 1.1764010429382323,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.647834274952919,
|
| 609 |
+
"grad_norm": 2.9293882846832275,
|
| 610 |
+
"learning_rate": 2.941094479366744e-06,
|
| 611 |
+
"loss": 1.192509651184082,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.655367231638418,
|
| 616 |
+
"grad_norm": 2.4857757091522217,
|
| 617 |
+
"learning_rate": 2.83058130441221e-06,
|
| 618 |
+
"loss": 1.1889172554016114,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6629001883239172,
|
| 623 |
+
"grad_norm": 2.9847218990325928,
|
| 624 |
+
"learning_rate": 2.721358724125287e-06,
|
| 625 |
+
"loss": 1.1603904724121095,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6704331450094162,
|
| 630 |
+
"grad_norm": 2.9232635498046875,
|
| 631 |
+
"learning_rate": 2.613491715395861e-06,
|
| 632 |
+
"loss": 1.1712430953979491,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.6779661016949152,
|
| 637 |
+
"grad_norm": 2.707500457763672,
|
| 638 |
+
"learning_rate": 2.507044448679621e-06,
|
| 639 |
+
"loss": 1.1785946846008302,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.6854990583804144,
|
| 644 |
+
"grad_norm": 2.8435614109039307,
|
| 645 |
+
"learning_rate": 2.4020802498228333e-06,
|
| 646 |
+
"loss": 1.1463645935058593,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6930320150659134,
|
| 651 |
+
"grad_norm": 2.75783634185791,
|
| 652 |
+
"learning_rate": 2.2986615623895774e-06,
|
| 653 |
+
"loss": 1.1732158660888672,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7005649717514124,
|
| 658 |
+
"grad_norm": 2.9693729877471924,
|
| 659 |
+
"learning_rate": 2.196849910513858e-06,
|
| 660 |
+
"loss": 1.136567974090576,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7080979284369114,
|
| 665 |
+
"grad_norm": 2.608729362487793,
|
| 666 |
+
"learning_rate": 2.096705862298676e-06,
|
| 667 |
+
"loss": 1.1899651527404784,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7156308851224106,
|
| 672 |
+
"grad_norm": 2.6576435565948486,
|
| 673 |
+
"learning_rate": 1.9982889937838635e-06,
|
| 674 |
+
"loss": 1.1970231056213378,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.7231638418079096,
|
| 679 |
+
"grad_norm": 2.5832364559173584,
|
| 680 |
+
"learning_rate": 1.901657853504088e-06,
|
| 681 |
+
"loss": 1.187626838684082,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.7306967984934086,
|
| 686 |
+
"grad_norm": 3.1484322547912598,
|
| 687 |
+
"learning_rate": 1.8068699276581286e-06,
|
| 688 |
+
"loss": 1.1518936157226562,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7382297551789078,
|
| 693 |
+
"grad_norm": 2.6287901401519775,
|
| 694 |
+
"learning_rate": 1.7139816059101372e-06,
|
| 695 |
+
"loss": 1.169551467895508,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7457627118644068,
|
| 700 |
+
"grad_norm": 2.8365728855133057,
|
| 701 |
+
"learning_rate": 1.6230481478432364e-06,
|
| 702 |
+
"loss": 1.1763624191284179,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7532956685499058,
|
| 707 |
+
"grad_norm": 3.0020880699157715,
|
| 708 |
+
"learning_rate": 1.5341236500853963e-06,
|
| 709 |
+
"loss": 1.1237783432006836,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7608286252354048,
|
| 714 |
+
"grad_norm": 2.755495071411133,
|
| 715 |
+
"learning_rate": 1.447261014127167e-06,
|
| 716 |
+
"loss": 1.1419706344604492,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.768361581920904,
|
| 721 |
+
"grad_norm": 2.8915786743164062,
|
| 722 |
+
"learning_rate": 1.3625119148504013e-06,
|
| 723 |
+
"loss": 1.1483024597167968,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.775894538606403,
|
| 728 |
+
"grad_norm": 2.9718737602233887,
|
| 729 |
+
"learning_rate": 1.2799267697866868e-06,
|
| 730 |
+
"loss": 1.1998208999633788,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.783427495291902,
|
| 735 |
+
"grad_norm": 2.6708669662475586,
|
| 736 |
+
"learning_rate": 1.1995547091237814e-06,
|
| 737 |
+
"loss": 1.1890599250793457,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.7909604519774012,
|
| 742 |
+
"grad_norm": 2.505589008331299,
|
| 743 |
+
"learning_rate": 1.1214435464779006e-06,
|
| 744 |
+
"loss": 1.1614674568176269,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.7984934086629002,
|
| 749 |
+
"grad_norm": 3.0118825435638428,
|
| 750 |
+
"learning_rate": 1.045639750449225e-06,
|
| 751 |
+
"loss": 1.1164636611938477,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8060263653483992,
|
| 756 |
+
"grad_norm": 2.741147518157959,
|
| 757 |
+
"learning_rate": 9.721884169775735e-07,
|
| 758 |
+
"loss": 1.1408018112182616,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.8135593220338984,
|
| 763 |
+
"grad_norm": 2.805220127105713,
|
| 764 |
+
"learning_rate": 9.011332425146734e-07,
|
| 765 |
+
"loss": 1.1225814819335938,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8210922787193974,
|
| 770 |
+
"grad_norm": 2.7636218070983887,
|
| 771 |
+
"learning_rate": 8.325164980289896e-07,
|
| 772 |
+
"loss": 1.1628660202026366,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.8286252354048964,
|
| 777 |
+
"grad_norm": 2.942866325378418,
|
| 778 |
+
"learning_rate": 7.663790038585794e-07,
|
| 779 |
+
"loss": 1.1538756370544434,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8361581920903954,
|
| 784 |
+
"grad_norm": 3.368988513946533,
|
| 785 |
+
"learning_rate": 7.027601054269329e-07,
|
| 786 |
+
"loss": 1.1323760986328124,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.8436911487758946,
|
| 791 |
+
"grad_norm": 2.687249183654785,
|
| 792 |
+
"learning_rate": 6.416976498362432e-07,
|
| 793 |
+
"loss": 1.1465997695922852,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8512241054613936,
|
| 798 |
+
"grad_norm": 2.8266866207122803,
|
| 799 |
+
"learning_rate": 5.832279633520411e-07,
|
| 800 |
+
"loss": 1.1509078025817872,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8587570621468926,
|
| 805 |
+
"grad_norm": 2.807955503463745,
|
| 806 |
+
"learning_rate": 5.273858297925649e-07,
|
| 807 |
+
"loss": 1.1476680755615234,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8662900188323918,
|
| 812 |
+
"grad_norm": 2.710559368133545,
|
| 813 |
+
"learning_rate": 4.7420446983575286e-07,
|
| 814 |
+
"loss": 1.1520160675048827,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.8738229755178908,
|
| 819 |
+
"grad_norm": 2.6868326663970947,
|
| 820 |
+
"learning_rate": 4.2371552125614614e-07,
|
| 821 |
+
"loss": 1.1053319931030274,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.8813559322033898,
|
| 826 |
+
"grad_norm": 2.8046789169311523,
|
| 827 |
+
"learning_rate": 3.759490201034677e-07,
|
| 828 |
+
"loss": 1.1290611267089843,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.8888888888888888,
|
| 833 |
+
"grad_norm": 2.7284066677093506,
|
| 834 |
+
"learning_rate": 3.309333828340722e-07,
|
| 835 |
+
"loss": 1.1444223403930665,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.896421845574388,
|
| 840 |
+
"grad_norm": 2.928938865661621,
|
| 841 |
+
"learning_rate": 2.88695389405898e-07,
|
| 842 |
+
"loss": 1.1631834030151367,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.903954802259887,
|
| 847 |
+
"grad_norm": 2.814770460128784,
|
| 848 |
+
"learning_rate": 2.49260167346978e-07,
|
| 849 |
+
"loss": 1.139974021911621,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.911487758945386,
|
| 854 |
+
"grad_norm": 3.0456888675689697,
|
| 855 |
+
"learning_rate": 2.1265117680698665e-07,
|
| 856 |
+
"loss": 1.1391963958740234,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.9190207156308852,
|
| 861 |
+
"grad_norm": 2.6722140312194824,
|
| 862 |
+
"learning_rate": 1.7889019660071728e-07,
|
| 863 |
+
"loss": 1.1376457214355469,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9265536723163842,
|
| 868 |
+
"grad_norm": 2.5319581031799316,
|
| 869 |
+
"learning_rate": 1.479973112517863e-07,
|
| 870 |
+
"loss": 1.1459515571594239,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.9340866290018832,
|
| 875 |
+
"grad_norm": 2.802717685699463,
|
| 876 |
+
"learning_rate": 1.1999089904428385e-07,
|
| 877 |
+
"loss": 1.1480701446533204,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9416195856873822,
|
| 882 |
+
"grad_norm": 3.0073602199554443,
|
| 883 |
+
"learning_rate": 9.488762108946891e-08,
|
| 884 |
+
"loss": 1.1705062866210938,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9491525423728814,
|
| 889 |
+
"grad_norm": 2.6973114013671875,
|
| 890 |
+
"learning_rate": 7.270241141401568e-08,
|
| 891 |
+
"loss": 1.1306757926940918,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.9566854990583804,
|
| 896 |
+
"grad_norm": 2.5668089389801025,
|
| 897 |
+
"learning_rate": 5.34484680757108e-08,
|
| 898 |
+
"loss": 1.1814102172851562,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9642184557438794,
|
| 903 |
+
"grad_norm": 2.92275333404541,
|
| 904 |
+
"learning_rate": 3.713724531188234e-08,
|
| 905 |
+
"loss": 1.1623525619506836,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.9717514124293786,
|
| 910 |
+
"grad_norm": 2.826354742050171,
|
| 911 |
+
"learning_rate": 2.3778446725233328e-08,
|
| 912 |
+
"loss": 1.1566267013549805,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.9792843691148776,
|
| 917 |
+
"grad_norm": 2.8368582725524902,
|
| 918 |
+
"learning_rate": 1.338001951113943e-08,
|
| 919 |
+
"loss": 1.1434457778930665,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9868173258003766,
|
| 924 |
+
"grad_norm": 2.8276095390319824,
|
| 925 |
+
"learning_rate": 5.9481497298291556e-09,
|
| 926 |
+
"loss": 1.129819679260254,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.9943502824858758,
|
| 931 |
+
"grad_norm": 2.8703908920288086,
|
| 932 |
+
"learning_rate": 1.4872586262792975e-09,
|
| 933 |
+
"loss": 1.1079357147216797,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0,
|
| 938 |
+
"step": 664,
|
| 939 |
+
"total_flos": 1.9898718236966912e+17,
|
| 940 |
+
"train_loss": 1.3105711570705276,
|
| 941 |
+
"train_runtime": 2573.9711,
|
| 942 |
+
"train_samples_per_second": 16.504,
|
| 943 |
+
"train_steps_per_second": 0.258
|
| 944 |
+
}
|
| 945 |
+
],
|
| 946 |
+
"logging_steps": 5,
|
| 947 |
+
"max_steps": 664,
|
| 948 |
+
"num_input_tokens_seen": 0,
|
| 949 |
+
"num_train_epochs": 1,
|
| 950 |
+
"save_steps": 10000,
|
| 951 |
+
"stateful_callbacks": {
|
| 952 |
+
"TrainerControl": {
|
| 953 |
+
"args": {
|
| 954 |
+
"should_epoch_stop": false,
|
| 955 |
+
"should_evaluate": false,
|
| 956 |
+
"should_log": false,
|
| 957 |
+
"should_save": true,
|
| 958 |
+
"should_training_stop": true
|
| 959 |
+
},
|
| 960 |
+
"attributes": {}
|
| 961 |
+
}
|
| 962 |
+
},
|
| 963 |
+
"total_flos": 1.9898718236966912e+17,
|
| 964 |
+
"train_batch_size": 1,
|
| 965 |
+
"trial_name": null,
|
| 966 |
+
"trial_params": null
|
| 967 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58c68276193ae906c275df613f40ecac1b338ed6b65a8688cc41bb8e4cedec22
|
| 3 |
+
size 7224
|
training_loss.png
ADDED
|