Upload folder using huggingface_hub
Browse files- .gitattributes +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/adapter_config.json +48 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/adapter_model.safetensors +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/tokenizer.json +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/tokenizer_config.json +30 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/trainer_state.json +734 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/adapter_config.json +48 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/adapter_model.safetensors +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/tokenizer.json +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/tokenizer_config.json +30 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/trainer_state.json +755 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/adapter_config.json +48 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/adapter_model.safetensors +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/base_model_weights.safetensors +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/tokenizer.json +3 -0
- results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/tokenizer_config.json +30 -0
- results_code/qwen3/pissa_r32/checkpoint-final/train_metrics.json +429 -0
.gitattributes
CHANGED
|
@@ -202,3 +202,6 @@ results_code/qwen3/pissa_r8/checkpoint-final/checkpoint-final/tokenizer.json fil
|
|
| 202 |
results_code/qwen3/pissa_r16/checkpoint-final/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 203 |
results_code/qwen3/pissa_r16/checkpoint-final/checkpoint-1038/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 204 |
results_code/qwen3/pissa_r16/checkpoint-final/checkpoint-final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
results_code/qwen3/pissa_r16/checkpoint-final/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 203 |
results_code/qwen3/pissa_r16/checkpoint-final/checkpoint-1038/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 204 |
results_code/qwen3/pissa_r16/checkpoint-final/checkpoint-final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 205 |
+
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 206 |
+
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 207 |
+
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/adapter_config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "/home/zhl923/qwen3-8b",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": "pissa",
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 32,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.0,
|
| 22 |
+
"lora_ga_config": null,
|
| 23 |
+
"megatron_config": null,
|
| 24 |
+
"megatron_core": "megatron.core",
|
| 25 |
+
"modules_to_save": null,
|
| 26 |
+
"peft_type": "LORA",
|
| 27 |
+
"peft_version": "0.19.1",
|
| 28 |
+
"qalora_group_size": 16,
|
| 29 |
+
"r": 32,
|
| 30 |
+
"rank_pattern": {},
|
| 31 |
+
"revision": null,
|
| 32 |
+
"target_modules": [
|
| 33 |
+
"gate_proj",
|
| 34 |
+
"v_proj",
|
| 35 |
+
"up_proj",
|
| 36 |
+
"o_proj",
|
| 37 |
+
"k_proj",
|
| 38 |
+
"down_proj",
|
| 39 |
+
"q_proj"
|
| 40 |
+
],
|
| 41 |
+
"target_parameters": null,
|
| 42 |
+
"task_type": "CAUSAL_LM",
|
| 43 |
+
"trainable_token_indices": null,
|
| 44 |
+
"use_bdlora": null,
|
| 45 |
+
"use_dora": false,
|
| 46 |
+
"use_qalora": false,
|
| 47 |
+
"use_rslora": false
|
| 48 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69e014df172812382656c9178ab5479a9b6311024f930447e439f738b2cb1fda
|
| 3 |
+
size 349243752
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
| 3 |
+
size 11422650
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": true,
|
| 24 |
+
"local_files_only": false,
|
| 25 |
+
"model_max_length": 131072,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,734 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9640877319836105,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.009640877319836105,
|
| 14 |
+
"grad_norm": 0.9736032485961914,
|
| 15 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 16 |
+
"loss": 0.8707587242126464,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01928175463967221,
|
| 21 |
+
"grad_norm": 0.5271208882331848,
|
| 22 |
+
"learning_rate": 0.00011875,
|
| 23 |
+
"loss": 0.7253766536712647,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.028922631959508314,
|
| 28 |
+
"grad_norm": 0.5695979595184326,
|
| 29 |
+
"learning_rate": 0.00018125000000000001,
|
| 30 |
+
"loss": 0.7168488502502441,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03856350927934442,
|
| 35 |
+
"grad_norm": 0.469108521938324,
|
| 36 |
+
"learning_rate": 0.00019997610799626618,
|
| 37 |
+
"loss": 0.6930637836456299,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.048204386599180526,
|
| 42 |
+
"grad_norm": 0.41781583428382874,
|
| 43 |
+
"learning_rate": 0.0001998591134197538,
|
| 44 |
+
"loss": 0.7164297580718995,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05784526391901663,
|
| 49 |
+
"grad_norm": 0.4602639675140381,
|
| 50 |
+
"learning_rate": 0.0001996447418870803,
|
| 51 |
+
"loss": 0.7047216415405273,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06748614123885273,
|
| 56 |
+
"grad_norm": 0.3851037919521332,
|
| 57 |
+
"learning_rate": 0.00019933320244123256,
|
| 58 |
+
"loss": 0.7204823017120361,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07712701855868884,
|
| 63 |
+
"grad_norm": 0.40421193838119507,
|
| 64 |
+
"learning_rate": 0.0001989247988778475,
|
| 65 |
+
"loss": 0.6821190357208252,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08676789587852494,
|
| 70 |
+
"grad_norm": 0.4095012843608856,
|
| 71 |
+
"learning_rate": 0.00019841992944896766,
|
| 72 |
+
"loss": 0.6873124599456787,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.09640877319836105,
|
| 77 |
+
"grad_norm": 0.4433162808418274,
|
| 78 |
+
"learning_rate": 0.00019781908647468845,
|
| 79 |
+
"loss": 0.6788602828979492,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.10604965051819716,
|
| 84 |
+
"grad_norm": 0.42396649718284607,
|
| 85 |
+
"learning_rate": 0.00019712285586307536,
|
| 86 |
+
"loss": 0.6963721752166748,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.11569052783803326,
|
| 91 |
+
"grad_norm": 0.4311012327671051,
|
| 92 |
+
"learning_rate": 0.00019633191653881959,
|
| 93 |
+
"loss": 0.6516051769256592,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.12533140515786936,
|
| 98 |
+
"grad_norm": 0.5481902956962585,
|
| 99 |
+
"learning_rate": 0.0001954470397811892,
|
| 100 |
+
"loss": 0.7184146404266357,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.13497228247770546,
|
| 105 |
+
"grad_norm": 0.6648269891738892,
|
| 106 |
+
"learning_rate": 0.0001944690884719211,
|
| 107 |
+
"loss": 0.6581430912017823,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14461315979754158,
|
| 112 |
+
"grad_norm": 0.4471873641014099,
|
| 113 |
+
"learning_rate": 0.0001933990162537877,
|
| 114 |
+
"loss": 0.6832843780517578,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.15425403711737767,
|
| 119 |
+
"grad_norm": 0.4480057656764984,
|
| 120 |
+
"learning_rate": 0.00019223786660065836,
|
| 121 |
+
"loss": 0.6737677097320557,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1638949144372138,
|
| 126 |
+
"grad_norm": 0.3746876120567322,
|
| 127 |
+
"learning_rate": 0.00019098677179996284,
|
| 128 |
+
"loss": 0.693084716796875,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1735357917570499,
|
| 133 |
+
"grad_norm": 0.39373138546943665,
|
| 134 |
+
"learning_rate": 0.0001896469518485486,
|
| 135 |
+
"loss": 0.6571917057037353,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.18317666907688598,
|
| 140 |
+
"grad_norm": 0.36390042304992676,
|
| 141 |
+
"learning_rate": 0.0001882197132630089,
|
| 142 |
+
"loss": 0.6517327785491943,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1928175463967221,
|
| 147 |
+
"grad_norm": 0.373504102230072,
|
| 148 |
+
"learning_rate": 0.00018670644780564177,
|
| 149 |
+
"loss": 0.6471811771392822,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2024584237165582,
|
| 154 |
+
"grad_norm": 0.4035699963569641,
|
| 155 |
+
"learning_rate": 0.00018510863112728226,
|
| 156 |
+
"loss": 0.6675207614898682,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.21209930103639432,
|
| 161 |
+
"grad_norm": 0.4150988757610321,
|
| 162 |
+
"learning_rate": 0.0001834278213283312,
|
| 163 |
+
"loss": 0.6660717964172364,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.22174017835623042,
|
| 168 |
+
"grad_norm": 0.36883264780044556,
|
| 169 |
+
"learning_rate": 0.0001816656574393837,
|
| 170 |
+
"loss": 0.6911928176879882,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2313810556760665,
|
| 175 |
+
"grad_norm": 0.3509788513183594,
|
| 176 |
+
"learning_rate": 0.00017982385782293968,
|
| 177 |
+
"loss": 0.6443495273590087,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.24102193299590263,
|
| 182 |
+
"grad_norm": 0.40021684765815735,
|
| 183 |
+
"learning_rate": 0.00017790421849775358,
|
| 184 |
+
"loss": 0.6613352775573731,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2506628103157387,
|
| 189 |
+
"grad_norm": 0.39133599400520325,
|
| 190 |
+
"learning_rate": 0.0001759086113874587,
|
| 191 |
+
"loss": 0.6512939929962158,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2603036876355748,
|
| 196 |
+
"grad_norm": 0.39297354221343994,
|
| 197 |
+
"learning_rate": 0.0001738389824951733,
|
| 198 |
+
"loss": 0.6607202053070068,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2699445649554109,
|
| 203 |
+
"grad_norm": 0.34986260533332825,
|
| 204 |
+
"learning_rate": 0.0001716973500058682,
|
| 205 |
+
"loss": 0.6551290035247803,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.27958544227524706,
|
| 210 |
+
"grad_norm": 0.350807249546051,
|
| 211 |
+
"learning_rate": 0.00016948580231834737,
|
| 212 |
+
"loss": 0.6323615550994873,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.28922631959508316,
|
| 217 |
+
"grad_norm": 0.38723087310791016,
|
| 218 |
+
"learning_rate": 0.00016720649600875936,
|
| 219 |
+
"loss": 0.6942133903503418,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.29886719691491925,
|
| 224 |
+
"grad_norm": 0.3664536476135254,
|
| 225 |
+
"learning_rate": 0.00016486165372762635,
|
| 226 |
+
"loss": 0.6656977653503418,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.30850807423475535,
|
| 231 |
+
"grad_norm": 0.40660181641578674,
|
| 232 |
+
"learning_rate": 0.00016245356203244108,
|
| 233 |
+
"loss": 0.6788680076599121,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.31814895155459144,
|
| 238 |
+
"grad_norm": 0.3671169579029083,
|
| 239 |
+
"learning_rate": 0.00015998456915794558,
|
| 240 |
+
"loss": 0.6604236125946045,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3277898288744276,
|
| 245 |
+
"grad_norm": 0.33644863963127136,
|
| 246 |
+
"learning_rate": 0.00015745708272626533,
|
| 247 |
+
"loss": 0.6478409767150879,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3374307061942637,
|
| 252 |
+
"grad_norm": 0.3842611610889435,
|
| 253 |
+
"learning_rate": 0.00015487356739913249,
|
| 254 |
+
"loss": 0.6721868991851807,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3470715835140998,
|
| 259 |
+
"grad_norm": 0.330951988697052,
|
| 260 |
+
"learning_rate": 0.00015223654247448732,
|
| 261 |
+
"loss": 0.6553137302398682,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3567124608339359,
|
| 266 |
+
"grad_norm": 0.3382876217365265,
|
| 267 |
+
"learning_rate": 0.0001495485794298013,
|
| 268 |
+
"loss": 0.6464945793151855,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.36635333815377197,
|
| 273 |
+
"grad_norm": 0.35035449266433716,
|
| 274 |
+
"learning_rate": 0.00014681229941451794,
|
| 275 |
+
"loss": 0.6719611644744873,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3759942154736081,
|
| 280 |
+
"grad_norm": 0.37467074394226074,
|
| 281 |
+
"learning_rate": 0.000144030370694056,
|
| 282 |
+
"loss": 0.6390235900878907,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3856350927934442,
|
| 287 |
+
"grad_norm": 0.3584175705909729,
|
| 288 |
+
"learning_rate": 0.0001412055060478683,
|
| 289 |
+
"loss": 0.6818144798278809,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3952759701132803,
|
| 294 |
+
"grad_norm": 0.39478713274002075,
|
| 295 |
+
"learning_rate": 0.0001383404601240925,
|
| 296 |
+
"loss": 0.6631309509277343,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4049168474331164,
|
| 301 |
+
"grad_norm": 0.35264909267425537,
|
| 302 |
+
"learning_rate": 0.00013543802675337421,
|
| 303 |
+
"loss": 0.6583386421203613,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4145577247529525,
|
| 308 |
+
"grad_norm": 0.36185845732688904,
|
| 309 |
+
"learning_rate": 0.0001325010362244816,
|
| 310 |
+
"loss": 0.6544721126556396,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.42419860207278864,
|
| 315 |
+
"grad_norm": 0.3521538972854614,
|
| 316 |
+
"learning_rate": 0.00012953235252436763,
|
| 317 |
+
"loss": 0.6781279087066651,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.43383947939262474,
|
| 322 |
+
"grad_norm": 0.3444727659225464,
|
| 323 |
+
"learning_rate": 0.00012653487054537237,
|
| 324 |
+
"loss": 0.6425844669342041,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.44348035671246083,
|
| 329 |
+
"grad_norm": 0.3636373281478882,
|
| 330 |
+
"learning_rate": 0.00012351151326228768,
|
| 331 |
+
"loss": 0.6464450359344482,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4531212340322969,
|
| 336 |
+
"grad_norm": 0.3635808825492859,
|
| 337 |
+
"learning_rate": 0.0001204652288820376,
|
| 338 |
+
"loss": 0.6836107730865478,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.462762111352133,
|
| 343 |
+
"grad_norm": 0.32124996185302734,
|
| 344 |
+
"learning_rate": 0.00011739898796875383,
|
| 345 |
+
"loss": 0.6507717609405518,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.47240298867196917,
|
| 350 |
+
"grad_norm": 0.35888606309890747,
|
| 351 |
+
"learning_rate": 0.00011431578054704977,
|
| 352 |
+
"loss": 0.6604080200195312,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.48204386599180526,
|
| 357 |
+
"grad_norm": 0.35348206758499146,
|
| 358 |
+
"learning_rate": 0.00011121861318631762,
|
| 359 |
+
"loss": 0.6555716514587402,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.49168474331164136,
|
| 364 |
+
"grad_norm": 0.3726312518119812,
|
| 365 |
+
"learning_rate": 0.00010811050606889199,
|
| 366 |
+
"loss": 0.6564616203308106,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5013256206314775,
|
| 371 |
+
"grad_norm": 0.35936763882637024,
|
| 372 |
+
"learning_rate": 0.00010499449004493928,
|
| 373 |
+
"loss": 0.6298168182373047,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5109664979513135,
|
| 378 |
+
"grad_norm": 0.3345901072025299,
|
| 379 |
+
"learning_rate": 0.00010187360367694394,
|
| 380 |
+
"loss": 0.6319252490997315,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5206073752711496,
|
| 385 |
+
"grad_norm": 0.39399752020835876,
|
| 386 |
+
"learning_rate": 9.875089027667408e-05,
|
| 387 |
+
"loss": 0.6732240676879883,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5302482525909857,
|
| 392 |
+
"grad_norm": 0.3447263538837433,
|
| 393 |
+
"learning_rate": 9.562939493751663e-05,
|
| 394 |
+
"loss": 0.6434835433959961,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5398891299108218,
|
| 399 |
+
"grad_norm": 0.45187753438949585,
|
| 400 |
+
"learning_rate": 9.251216156507397e-05,
|
| 401 |
+
"loss": 0.6498741149902344,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.549530007230658,
|
| 406 |
+
"grad_norm": 0.3480952978134155,
|
| 407 |
+
"learning_rate": 8.940222990891989e-05,
|
| 408 |
+
"loss": 0.6715107917785644,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5591708845504941,
|
| 413 |
+
"grad_norm": 0.3822426199913025,
|
| 414 |
+
"learning_rate": 8.63026325984076e-05,
|
| 415 |
+
"loss": 0.6553134441375732,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5688117618703302,
|
| 420 |
+
"grad_norm": 0.3563183844089508,
|
| 421 |
+
"learning_rate": 8.321639218542137e-05,
|
| 422 |
+
"loss": 0.6121355056762695,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5784526391901663,
|
| 427 |
+
"grad_norm": 0.3752917945384979,
|
| 428 |
+
"learning_rate": 8.014651819695478e-05,
|
| 429 |
+
"loss": 0.6503292083740234,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5880935165100024,
|
| 434 |
+
"grad_norm": 0.3247714638710022,
|
| 435 |
+
"learning_rate": 7.709600420039053e-05,
|
| 436 |
+
"loss": 0.6336544513702392,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5977343938298385,
|
| 441 |
+
"grad_norm": 0.36859628558158875,
|
| 442 |
+
"learning_rate": 7.406782488434285e-05,
|
| 443 |
+
"loss": 0.6424888610839844,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6073752711496746,
|
| 448 |
+
"grad_norm": 0.34594741463661194,
|
| 449 |
+
"learning_rate": 7.106493315790938e-05,
|
| 450 |
+
"loss": 0.6859614849090576,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6170161484695107,
|
| 455 |
+
"grad_norm": 0.3520846962928772,
|
| 456 |
+
"learning_rate": 6.809025727116153e-05,
|
| 457 |
+
"loss": 0.6500971794128418,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6266570257893468,
|
| 462 |
+
"grad_norm": 0.36558225750923157,
|
| 463 |
+
"learning_rate": 6.514669795968067e-05,
|
| 464 |
+
"loss": 0.6580742359161377,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6362979031091829,
|
| 469 |
+
"grad_norm": 0.3209807276725769,
|
| 470 |
+
"learning_rate": 6.223712561592486e-05,
|
| 471 |
+
"loss": 0.630165958404541,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6459387804290191,
|
| 476 |
+
"grad_norm": 0.34253862500190735,
|
| 477 |
+
"learning_rate": 5.936437749018466e-05,
|
| 478 |
+
"loss": 0.6434316635131836,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6555796577488552,
|
| 483 |
+
"grad_norm": 0.34325432777404785,
|
| 484 |
+
"learning_rate": 5.6531254923857236e-05,
|
| 485 |
+
"loss": 0.6330021381378174,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6652205350686913,
|
| 490 |
+
"grad_norm": 0.384233295917511,
|
| 491 |
+
"learning_rate": 5.374052061773662e-05,
|
| 492 |
+
"loss": 0.634852409362793,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6748614123885274,
|
| 497 |
+
"grad_norm": 0.3439292311668396,
|
| 498 |
+
"learning_rate": 5.099489593798441e-05,
|
| 499 |
+
"loss": 0.6272209167480469,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6845022897083635,
|
| 504 |
+
"grad_norm": 0.3354477882385254,
|
| 505 |
+
"learning_rate": 4.829705826240736e-05,
|
| 506 |
+
"loss": 0.655023193359375,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6941431670281996,
|
| 511 |
+
"grad_norm": 0.34607255458831787,
|
| 512 |
+
"learning_rate": 4.564963836963011e-05,
|
| 513 |
+
"loss": 0.652175760269165,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7037840443480357,
|
| 518 |
+
"grad_norm": 0.30681824684143066,
|
| 519 |
+
"learning_rate": 4.305521787370891e-05,
|
| 520 |
+
"loss": 0.6359724044799805,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7134249216678717,
|
| 525 |
+
"grad_norm": 0.34687328338623047,
|
| 526 |
+
"learning_rate": 4.05163267066874e-05,
|
| 527 |
+
"loss": 0.6496444702148437,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7230657989877078,
|
| 532 |
+
"grad_norm": 0.3489939272403717,
|
| 533 |
+
"learning_rate": 3.8035440651550246e-05,
|
| 534 |
+
"loss": 0.6331109523773193,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.7327066763075439,
|
| 539 |
+
"grad_norm": 0.31731027364730835,
|
| 540 |
+
"learning_rate": 3.561497892797965e-05,
|
| 541 |
+
"loss": 0.6367918014526367,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7423475536273801,
|
| 546 |
+
"grad_norm": 0.3635317087173462,
|
| 547 |
+
"learning_rate": 3.3257301833269204e-05,
|
| 548 |
+
"loss": 0.6265190601348877,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.7519884309472162,
|
| 553 |
+
"grad_norm": 0.394301176071167,
|
| 554 |
+
"learning_rate": 3.0964708440695535e-05,
|
| 555 |
+
"loss": 0.6516129016876221,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7616293082670523,
|
| 560 |
+
"grad_norm": 0.338499516248703,
|
| 561 |
+
"learning_rate": 2.8739434357592276e-05,
|
| 562 |
+
"loss": 0.644101905822754,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7712701855868884,
|
| 567 |
+
"grad_norm": 0.34166204929351807,
|
| 568 |
+
"learning_rate": 2.6583649545312085e-05,
|
| 569 |
+
"loss": 0.6312844753265381,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7809110629067245,
|
| 574 |
+
"grad_norm": 0.34724798798561096,
|
| 575 |
+
"learning_rate": 2.449945620320333e-05,
|
| 576 |
+
"loss": 0.6377103328704834,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7905519402265606,
|
| 581 |
+
"grad_norm": 0.3760109543800354,
|
| 582 |
+
"learning_rate": 2.2488886718664148e-05,
|
| 583 |
+
"loss": 0.6174637794494628,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8001928175463967,
|
| 588 |
+
"grad_norm": 0.34155479073524475,
|
| 589 |
+
"learning_rate": 2.05539016852734e-05,
|
| 590 |
+
"loss": 0.6319977283477783,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.8098336948662328,
|
| 595 |
+
"grad_norm": 0.3997814953327179,
|
| 596 |
+
"learning_rate": 1.869638799093052e-05,
|
| 597 |
+
"loss": 0.6516120433807373,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8194745721860689,
|
| 602 |
+
"grad_norm": 0.33936548233032227,
|
| 603 |
+
"learning_rate": 1.6918156977869535e-05,
|
| 604 |
+
"loss": 0.6446984767913818,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.829115449505905,
|
| 609 |
+
"grad_norm": 0.3098904490470886,
|
| 610 |
+
"learning_rate": 1.5220942676340365e-05,
|
| 611 |
+
"loss": 0.6403415203094482,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.8387563268257412,
|
| 616 |
+
"grad_norm": 0.32174718379974365,
|
| 617 |
+
"learning_rate": 1.3606400113680806e-05,
|
| 618 |
+
"loss": 0.6233531951904296,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.8483972041455773,
|
| 623 |
+
"grad_norm": 0.34311968088150024,
|
| 624 |
+
"learning_rate": 1.2076103700427432e-05,
|
| 625 |
+
"loss": 0.6372226715087891,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8580380814654134,
|
| 630 |
+
"grad_norm": 0.3492196500301361,
|
| 631 |
+
"learning_rate": 1.0631545695039657e-05,
|
| 632 |
+
"loss": 0.6339581966400146,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8676789587852495,
|
| 637 |
+
"grad_norm": 0.36382928490638733,
|
| 638 |
+
"learning_rate": 9.274134748733643e-06,
|
| 639 |
+
"loss": 0.6436203956604004,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8773198361050856,
|
| 644 |
+
"grad_norm": 0.3259238302707672,
|
| 645 |
+
"learning_rate": 8.00519453184534e-06,
|
| 646 |
+
"loss": 0.6310982704162598,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8869607134249217,
|
| 651 |
+
"grad_norm": 0.32131025195121765,
|
| 652 |
+
"learning_rate": 6.825962443062173e-06,
|
| 653 |
+
"loss": 0.620008897781372,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8966015907447578,
|
| 658 |
+
"grad_norm": 0.36284807324409485,
|
| 659 |
+
"learning_rate": 5.73758840278179e-06,
|
| 660 |
+
"loss": 0.620122766494751,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9062424680645939,
|
| 665 |
+
"grad_norm": 0.33615875244140625,
|
| 666 |
+
"learning_rate": 4.741133731774783e-06,
|
| 667 |
+
"loss": 0.640928316116333,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.91588334538443,
|
| 672 |
+
"grad_norm": 0.38269132375717163,
|
| 673 |
+
"learning_rate": 3.837570116244815e-06,
|
| 674 |
+
"loss": 0.6369370937347412,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.925524222704266,
|
| 679 |
+
"grad_norm": 0.36093825101852417,
|
| 680 |
+
"learning_rate": 3.0277786602950776e-06,
|
| 681 |
+
"loss": 0.6047839641571044,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.9351651000241022,
|
| 686 |
+
"grad_norm": 0.36792394518852234,
|
| 687 |
+
"learning_rate": 2.3125490267255414e-06,
|
| 688 |
+
"loss": 0.6083663463592529,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9448059773439383,
|
| 693 |
+
"grad_norm": 0.3627108335494995,
|
| 694 |
+
"learning_rate": 1.6925786669984566e-06,
|
| 695 |
+
"loss": 0.6259819030761719,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9544468546637744,
|
| 700 |
+
"grad_norm": 0.35401254892349243,
|
| 701 |
+
"learning_rate": 1.1684721411231558e-06,
|
| 702 |
+
"loss": 0.6372841835021973,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9640877319836105,
|
| 707 |
+
"grad_norm": 0.34662875533103943,
|
| 708 |
+
"learning_rate": 7.407405281233182e-07,
|
| 709 |
+
"loss": 0.6086503505706787,
|
| 710 |
+
"step": 1000
|
| 711 |
+
}
|
| 712 |
+
],
|
| 713 |
+
"logging_steps": 10,
|
| 714 |
+
"max_steps": 1038,
|
| 715 |
+
"num_input_tokens_seen": 0,
|
| 716 |
+
"num_train_epochs": 1,
|
| 717 |
+
"save_steps": 500,
|
| 718 |
+
"stateful_callbacks": {
|
| 719 |
+
"TrainerControl": {
|
| 720 |
+
"args": {
|
| 721 |
+
"should_epoch_stop": false,
|
| 722 |
+
"should_evaluate": false,
|
| 723 |
+
"should_log": false,
|
| 724 |
+
"should_save": true,
|
| 725 |
+
"should_training_stop": false
|
| 726 |
+
},
|
| 727 |
+
"attributes": {}
|
| 728 |
+
}
|
| 729 |
+
},
|
| 730 |
+
"total_flos": 2.5526074094483866e+18,
|
| 731 |
+
"train_batch_size": 4,
|
| 732 |
+
"trial_name": null,
|
| 733 |
+
"trial_params": null
|
| 734 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/adapter_config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "/home/zhl923/qwen3-8b",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": "pissa",
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 32,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.0,
|
| 22 |
+
"lora_ga_config": null,
|
| 23 |
+
"megatron_config": null,
|
| 24 |
+
"megatron_core": "megatron.core",
|
| 25 |
+
"modules_to_save": null,
|
| 26 |
+
"peft_type": "LORA",
|
| 27 |
+
"peft_version": "0.19.1",
|
| 28 |
+
"qalora_group_size": 16,
|
| 29 |
+
"r": 32,
|
| 30 |
+
"rank_pattern": {},
|
| 31 |
+
"revision": null,
|
| 32 |
+
"target_modules": [
|
| 33 |
+
"gate_proj",
|
| 34 |
+
"v_proj",
|
| 35 |
+
"up_proj",
|
| 36 |
+
"o_proj",
|
| 37 |
+
"k_proj",
|
| 38 |
+
"down_proj",
|
| 39 |
+
"q_proj"
|
| 40 |
+
],
|
| 41 |
+
"target_parameters": null,
|
| 42 |
+
"task_type": "CAUSAL_LM",
|
| 43 |
+
"trainable_token_indices": null,
|
| 44 |
+
"use_bdlora": null,
|
| 45 |
+
"use_dora": false,
|
| 46 |
+
"use_qalora": false,
|
| 47 |
+
"use_rslora": false
|
| 48 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a23cdeac0ed65f2618f2fc0e44576548db33989a64135cdfe4314b8068cb8e0e
|
| 3 |
+
size 349243752
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
| 3 |
+
size 11422650
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": true,
|
| 24 |
+
"local_files_only": false,
|
| 25 |
+
"model_max_length": 131072,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-1038/trainer_state.json
ADDED
|
@@ -0,0 +1,755 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1038,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.009640877319836105,
|
| 14 |
+
"grad_norm": 0.9736032485961914,
|
| 15 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 16 |
+
"loss": 0.8707587242126464,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01928175463967221,
|
| 21 |
+
"grad_norm": 0.5271208882331848,
|
| 22 |
+
"learning_rate": 0.00011875,
|
| 23 |
+
"loss": 0.7253766536712647,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.028922631959508314,
|
| 28 |
+
"grad_norm": 0.5695979595184326,
|
| 29 |
+
"learning_rate": 0.00018125000000000001,
|
| 30 |
+
"loss": 0.7168488502502441,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03856350927934442,
|
| 35 |
+
"grad_norm": 0.469108521938324,
|
| 36 |
+
"learning_rate": 0.00019997610799626618,
|
| 37 |
+
"loss": 0.6930637836456299,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.048204386599180526,
|
| 42 |
+
"grad_norm": 0.41781583428382874,
|
| 43 |
+
"learning_rate": 0.0001998591134197538,
|
| 44 |
+
"loss": 0.7164297580718995,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05784526391901663,
|
| 49 |
+
"grad_norm": 0.4602639675140381,
|
| 50 |
+
"learning_rate": 0.0001996447418870803,
|
| 51 |
+
"loss": 0.7047216415405273,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06748614123885273,
|
| 56 |
+
"grad_norm": 0.3851037919521332,
|
| 57 |
+
"learning_rate": 0.00019933320244123256,
|
| 58 |
+
"loss": 0.7204823017120361,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07712701855868884,
|
| 63 |
+
"grad_norm": 0.40421193838119507,
|
| 64 |
+
"learning_rate": 0.0001989247988778475,
|
| 65 |
+
"loss": 0.6821190357208252,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08676789587852494,
|
| 70 |
+
"grad_norm": 0.4095012843608856,
|
| 71 |
+
"learning_rate": 0.00019841992944896766,
|
| 72 |
+
"loss": 0.6873124599456787,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.09640877319836105,
|
| 77 |
+
"grad_norm": 0.4433162808418274,
|
| 78 |
+
"learning_rate": 0.00019781908647468845,
|
| 79 |
+
"loss": 0.6788602828979492,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.10604965051819716,
|
| 84 |
+
"grad_norm": 0.42396649718284607,
|
| 85 |
+
"learning_rate": 0.00019712285586307536,
|
| 86 |
+
"loss": 0.6963721752166748,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.11569052783803326,
|
| 91 |
+
"grad_norm": 0.4311012327671051,
|
| 92 |
+
"learning_rate": 0.00019633191653881959,
|
| 93 |
+
"loss": 0.6516051769256592,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.12533140515786936,
|
| 98 |
+
"grad_norm": 0.5481902956962585,
|
| 99 |
+
"learning_rate": 0.0001954470397811892,
|
| 100 |
+
"loss": 0.7184146404266357,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.13497228247770546,
|
| 105 |
+
"grad_norm": 0.6648269891738892,
|
| 106 |
+
"learning_rate": 0.0001944690884719211,
|
| 107 |
+
"loss": 0.6581430912017823,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14461315979754158,
|
| 112 |
+
"grad_norm": 0.4471873641014099,
|
| 113 |
+
"learning_rate": 0.0001933990162537877,
|
| 114 |
+
"loss": 0.6832843780517578,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.15425403711737767,
|
| 119 |
+
"grad_norm": 0.4480057656764984,
|
| 120 |
+
"learning_rate": 0.00019223786660065836,
|
| 121 |
+
"loss": 0.6737677097320557,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1638949144372138,
|
| 126 |
+
"grad_norm": 0.3746876120567322,
|
| 127 |
+
"learning_rate": 0.00019098677179996284,
|
| 128 |
+
"loss": 0.693084716796875,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1735357917570499,
|
| 133 |
+
"grad_norm": 0.39373138546943665,
|
| 134 |
+
"learning_rate": 0.0001896469518485486,
|
| 135 |
+
"loss": 0.6571917057037353,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.18317666907688598,
|
| 140 |
+
"grad_norm": 0.36390042304992676,
|
| 141 |
+
"learning_rate": 0.0001882197132630089,
|
| 142 |
+
"loss": 0.6517327785491943,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1928175463967221,
|
| 147 |
+
"grad_norm": 0.373504102230072,
|
| 148 |
+
"learning_rate": 0.00018670644780564177,
|
| 149 |
+
"loss": 0.6471811771392822,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2024584237165582,
|
| 154 |
+
"grad_norm": 0.4035699963569641,
|
| 155 |
+
"learning_rate": 0.00018510863112728226,
|
| 156 |
+
"loss": 0.6675207614898682,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.21209930103639432,
|
| 161 |
+
"grad_norm": 0.4150988757610321,
|
| 162 |
+
"learning_rate": 0.0001834278213283312,
|
| 163 |
+
"loss": 0.6660717964172364,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.22174017835623042,
|
| 168 |
+
"grad_norm": 0.36883264780044556,
|
| 169 |
+
"learning_rate": 0.0001816656574393837,
|
| 170 |
+
"loss": 0.6911928176879882,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2313810556760665,
|
| 175 |
+
"grad_norm": 0.3509788513183594,
|
| 176 |
+
"learning_rate": 0.00017982385782293968,
|
| 177 |
+
"loss": 0.6443495273590087,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.24102193299590263,
|
| 182 |
+
"grad_norm": 0.40021684765815735,
|
| 183 |
+
"learning_rate": 0.00017790421849775358,
|
| 184 |
+
"loss": 0.6613352775573731,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2506628103157387,
|
| 189 |
+
"grad_norm": 0.39133599400520325,
|
| 190 |
+
"learning_rate": 0.0001759086113874587,
|
| 191 |
+
"loss": 0.6512939929962158,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2603036876355748,
|
| 196 |
+
"grad_norm": 0.39297354221343994,
|
| 197 |
+
"learning_rate": 0.0001738389824951733,
|
| 198 |
+
"loss": 0.6607202053070068,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2699445649554109,
|
| 203 |
+
"grad_norm": 0.34986260533332825,
|
| 204 |
+
"learning_rate": 0.0001716973500058682,
|
| 205 |
+
"loss": 0.6551290035247803,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.27958544227524706,
|
| 210 |
+
"grad_norm": 0.350807249546051,
|
| 211 |
+
"learning_rate": 0.00016948580231834737,
|
| 212 |
+
"loss": 0.6323615550994873,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.28922631959508316,
|
| 217 |
+
"grad_norm": 0.38723087310791016,
|
| 218 |
+
"learning_rate": 0.00016720649600875936,
|
| 219 |
+
"loss": 0.6942133903503418,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.29886719691491925,
|
| 224 |
+
"grad_norm": 0.3664536476135254,
|
| 225 |
+
"learning_rate": 0.00016486165372762635,
|
| 226 |
+
"loss": 0.6656977653503418,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.30850807423475535,
|
| 231 |
+
"grad_norm": 0.40660181641578674,
|
| 232 |
+
"learning_rate": 0.00016245356203244108,
|
| 233 |
+
"loss": 0.6788680076599121,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.31814895155459144,
|
| 238 |
+
"grad_norm": 0.3671169579029083,
|
| 239 |
+
"learning_rate": 0.00015998456915794558,
|
| 240 |
+
"loss": 0.6604236125946045,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3277898288744276,
|
| 245 |
+
"grad_norm": 0.33644863963127136,
|
| 246 |
+
"learning_rate": 0.00015745708272626533,
|
| 247 |
+
"loss": 0.6478409767150879,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3374307061942637,
|
| 252 |
+
"grad_norm": 0.3842611610889435,
|
| 253 |
+
"learning_rate": 0.00015487356739913249,
|
| 254 |
+
"loss": 0.6721868991851807,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3470715835140998,
|
| 259 |
+
"grad_norm": 0.330951988697052,
|
| 260 |
+
"learning_rate": 0.00015223654247448732,
|
| 261 |
+
"loss": 0.6553137302398682,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3567124608339359,
|
| 266 |
+
"grad_norm": 0.3382876217365265,
|
| 267 |
+
"learning_rate": 0.0001495485794298013,
|
| 268 |
+
"loss": 0.6464945793151855,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.36635333815377197,
|
| 273 |
+
"grad_norm": 0.35035449266433716,
|
| 274 |
+
"learning_rate": 0.00014681229941451794,
|
| 275 |
+
"loss": 0.6719611644744873,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3759942154736081,
|
| 280 |
+
"grad_norm": 0.37467074394226074,
|
| 281 |
+
"learning_rate": 0.000144030370694056,
|
| 282 |
+
"loss": 0.6390235900878907,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3856350927934442,
|
| 287 |
+
"grad_norm": 0.3584175705909729,
|
| 288 |
+
"learning_rate": 0.0001412055060478683,
|
| 289 |
+
"loss": 0.6818144798278809,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3952759701132803,
|
| 294 |
+
"grad_norm": 0.39478713274002075,
|
| 295 |
+
"learning_rate": 0.0001383404601240925,
|
| 296 |
+
"loss": 0.6631309509277343,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4049168474331164,
|
| 301 |
+
"grad_norm": 0.35264909267425537,
|
| 302 |
+
"learning_rate": 0.00013543802675337421,
|
| 303 |
+
"loss": 0.6583386421203613,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4145577247529525,
|
| 308 |
+
"grad_norm": 0.36185845732688904,
|
| 309 |
+
"learning_rate": 0.0001325010362244816,
|
| 310 |
+
"loss": 0.6544721126556396,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.42419860207278864,
|
| 315 |
+
"grad_norm": 0.3521538972854614,
|
| 316 |
+
"learning_rate": 0.00012953235252436763,
|
| 317 |
+
"loss": 0.6781279087066651,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.43383947939262474,
|
| 322 |
+
"grad_norm": 0.3444727659225464,
|
| 323 |
+
"learning_rate": 0.00012653487054537237,
|
| 324 |
+
"loss": 0.6425844669342041,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.44348035671246083,
|
| 329 |
+
"grad_norm": 0.3636373281478882,
|
| 330 |
+
"learning_rate": 0.00012351151326228768,
|
| 331 |
+
"loss": 0.6464450359344482,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4531212340322969,
|
| 336 |
+
"grad_norm": 0.3635808825492859,
|
| 337 |
+
"learning_rate": 0.0001204652288820376,
|
| 338 |
+
"loss": 0.6836107730865478,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.462762111352133,
|
| 343 |
+
"grad_norm": 0.32124996185302734,
|
| 344 |
+
"learning_rate": 0.00011739898796875383,
|
| 345 |
+
"loss": 0.6507717609405518,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.47240298867196917,
|
| 350 |
+
"grad_norm": 0.35888606309890747,
|
| 351 |
+
"learning_rate": 0.00011431578054704977,
|
| 352 |
+
"loss": 0.6604080200195312,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.48204386599180526,
|
| 357 |
+
"grad_norm": 0.35348206758499146,
|
| 358 |
+
"learning_rate": 0.00011121861318631762,
|
| 359 |
+
"loss": 0.6555716514587402,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.49168474331164136,
|
| 364 |
+
"grad_norm": 0.3726312518119812,
|
| 365 |
+
"learning_rate": 0.00010811050606889199,
|
| 366 |
+
"loss": 0.6564616203308106,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5013256206314775,
|
| 371 |
+
"grad_norm": 0.35936763882637024,
|
| 372 |
+
"learning_rate": 0.00010499449004493928,
|
| 373 |
+
"loss": 0.6298168182373047,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5109664979513135,
|
| 378 |
+
"grad_norm": 0.3345901072025299,
|
| 379 |
+
"learning_rate": 0.00010187360367694394,
|
| 380 |
+
"loss": 0.6319252490997315,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5206073752711496,
|
| 385 |
+
"grad_norm": 0.39399752020835876,
|
| 386 |
+
"learning_rate": 9.875089027667408e-05,
|
| 387 |
+
"loss": 0.6732240676879883,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5302482525909857,
|
| 392 |
+
"grad_norm": 0.3447263538837433,
|
| 393 |
+
"learning_rate": 9.562939493751663e-05,
|
| 394 |
+
"loss": 0.6434835433959961,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5398891299108218,
|
| 399 |
+
"grad_norm": 0.45187753438949585,
|
| 400 |
+
"learning_rate": 9.251216156507397e-05,
|
| 401 |
+
"loss": 0.6498741149902344,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.549530007230658,
|
| 406 |
+
"grad_norm": 0.3480952978134155,
|
| 407 |
+
"learning_rate": 8.940222990891989e-05,
|
| 408 |
+
"loss": 0.6715107917785644,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5591708845504941,
|
| 413 |
+
"grad_norm": 0.3822426199913025,
|
| 414 |
+
"learning_rate": 8.63026325984076e-05,
|
| 415 |
+
"loss": 0.6553134441375732,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5688117618703302,
|
| 420 |
+
"grad_norm": 0.3563183844089508,
|
| 421 |
+
"learning_rate": 8.321639218542137e-05,
|
| 422 |
+
"loss": 0.6121355056762695,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5784526391901663,
|
| 427 |
+
"grad_norm": 0.3752917945384979,
|
| 428 |
+
"learning_rate": 8.014651819695478e-05,
|
| 429 |
+
"loss": 0.6503292083740234,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5880935165100024,
|
| 434 |
+
"grad_norm": 0.3247714638710022,
|
| 435 |
+
"learning_rate": 7.709600420039053e-05,
|
| 436 |
+
"loss": 0.6336544513702392,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5977343938298385,
|
| 441 |
+
"grad_norm": 0.36859628558158875,
|
| 442 |
+
"learning_rate": 7.406782488434285e-05,
|
| 443 |
+
"loss": 0.6424888610839844,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6073752711496746,
|
| 448 |
+
"grad_norm": 0.34594741463661194,
|
| 449 |
+
"learning_rate": 7.106493315790938e-05,
|
| 450 |
+
"loss": 0.6859614849090576,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6170161484695107,
|
| 455 |
+
"grad_norm": 0.3520846962928772,
|
| 456 |
+
"learning_rate": 6.809025727116153e-05,
|
| 457 |
+
"loss": 0.6500971794128418,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6266570257893468,
|
| 462 |
+
"grad_norm": 0.36558225750923157,
|
| 463 |
+
"learning_rate": 6.514669795968067e-05,
|
| 464 |
+
"loss": 0.6580742359161377,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6362979031091829,
|
| 469 |
+
"grad_norm": 0.3209807276725769,
|
| 470 |
+
"learning_rate": 6.223712561592486e-05,
|
| 471 |
+
"loss": 0.630165958404541,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6459387804290191,
|
| 476 |
+
"grad_norm": 0.34253862500190735,
|
| 477 |
+
"learning_rate": 5.936437749018466e-05,
|
| 478 |
+
"loss": 0.6434316635131836,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6555796577488552,
|
| 483 |
+
"grad_norm": 0.34325432777404785,
|
| 484 |
+
"learning_rate": 5.6531254923857236e-05,
|
| 485 |
+
"loss": 0.6330021381378174,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6652205350686913,
|
| 490 |
+
"grad_norm": 0.384233295917511,
|
| 491 |
+
"learning_rate": 5.374052061773662e-05,
|
| 492 |
+
"loss": 0.634852409362793,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6748614123885274,
|
| 497 |
+
"grad_norm": 0.3439292311668396,
|
| 498 |
+
"learning_rate": 5.099489593798441e-05,
|
| 499 |
+
"loss": 0.6272209167480469,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6845022897083635,
|
| 504 |
+
"grad_norm": 0.3354477882385254,
|
| 505 |
+
"learning_rate": 4.829705826240736e-05,
|
| 506 |
+
"loss": 0.655023193359375,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6941431670281996,
|
| 511 |
+
"grad_norm": 0.34607255458831787,
|
| 512 |
+
"learning_rate": 4.564963836963011e-05,
|
| 513 |
+
"loss": 0.652175760269165,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7037840443480357,
|
| 518 |
+
"grad_norm": 0.30681824684143066,
|
| 519 |
+
"learning_rate": 4.305521787370891e-05,
|
| 520 |
+
"loss": 0.6359724044799805,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7134249216678717,
|
| 525 |
+
"grad_norm": 0.34687328338623047,
|
| 526 |
+
"learning_rate": 4.05163267066874e-05,
|
| 527 |
+
"loss": 0.6496444702148437,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7230657989877078,
|
| 532 |
+
"grad_norm": 0.3489939272403717,
|
| 533 |
+
"learning_rate": 3.8035440651550246e-05,
|
| 534 |
+
"loss": 0.6331109523773193,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.7327066763075439,
|
| 539 |
+
"grad_norm": 0.31731027364730835,
|
| 540 |
+
"learning_rate": 3.561497892797965e-05,
|
| 541 |
+
"loss": 0.6367918014526367,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7423475536273801,
|
| 546 |
+
"grad_norm": 0.3635317087173462,
|
| 547 |
+
"learning_rate": 3.3257301833269204e-05,
|
| 548 |
+
"loss": 0.6265190601348877,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.7519884309472162,
|
| 553 |
+
"grad_norm": 0.394301176071167,
|
| 554 |
+
"learning_rate": 3.0964708440695535e-05,
|
| 555 |
+
"loss": 0.6516129016876221,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7616293082670523,
|
| 560 |
+
"grad_norm": 0.338499516248703,
|
| 561 |
+
"learning_rate": 2.8739434357592276e-05,
|
| 562 |
+
"loss": 0.644101905822754,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7712701855868884,
|
| 567 |
+
"grad_norm": 0.34166204929351807,
|
| 568 |
+
"learning_rate": 2.6583649545312085e-05,
|
| 569 |
+
"loss": 0.6312844753265381,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7809110629067245,
|
| 574 |
+
"grad_norm": 0.34724798798561096,
|
| 575 |
+
"learning_rate": 2.449945620320333e-05,
|
| 576 |
+
"loss": 0.6377103328704834,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7905519402265606,
|
| 581 |
+
"grad_norm": 0.3760109543800354,
|
| 582 |
+
"learning_rate": 2.2488886718664148e-05,
|
| 583 |
+
"loss": 0.6174637794494628,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8001928175463967,
|
| 588 |
+
"grad_norm": 0.34155479073524475,
|
| 589 |
+
"learning_rate": 2.05539016852734e-05,
|
| 590 |
+
"loss": 0.6319977283477783,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.8098336948662328,
|
| 595 |
+
"grad_norm": 0.3997814953327179,
|
| 596 |
+
"learning_rate": 1.869638799093052e-05,
|
| 597 |
+
"loss": 0.6516120433807373,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8194745721860689,
|
| 602 |
+
"grad_norm": 0.33936548233032227,
|
| 603 |
+
"learning_rate": 1.6918156977869535e-05,
|
| 604 |
+
"loss": 0.6446984767913818,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.829115449505905,
|
| 609 |
+
"grad_norm": 0.3098904490470886,
|
| 610 |
+
"learning_rate": 1.5220942676340365e-05,
|
| 611 |
+
"loss": 0.6403415203094482,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.8387563268257412,
|
| 616 |
+
"grad_norm": 0.32174718379974365,
|
| 617 |
+
"learning_rate": 1.3606400113680806e-05,
|
| 618 |
+
"loss": 0.6233531951904296,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.8483972041455773,
|
| 623 |
+
"grad_norm": 0.34311968088150024,
|
| 624 |
+
"learning_rate": 1.2076103700427432e-05,
|
| 625 |
+
"loss": 0.6372226715087891,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8580380814654134,
|
| 630 |
+
"grad_norm": 0.3492196500301361,
|
| 631 |
+
"learning_rate": 1.0631545695039657e-05,
|
| 632 |
+
"loss": 0.6339581966400146,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8676789587852495,
|
| 637 |
+
"grad_norm": 0.36382928490638733,
|
| 638 |
+
"learning_rate": 9.274134748733643e-06,
|
| 639 |
+
"loss": 0.6436203956604004,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8773198361050856,
|
| 644 |
+
"grad_norm": 0.3259238302707672,
|
| 645 |
+
"learning_rate": 8.00519453184534e-06,
|
| 646 |
+
"loss": 0.6310982704162598,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8869607134249217,
|
| 651 |
+
"grad_norm": 0.32131025195121765,
|
| 652 |
+
"learning_rate": 6.825962443062173e-06,
|
| 653 |
+
"loss": 0.620008897781372,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8966015907447578,
|
| 658 |
+
"grad_norm": 0.36284807324409485,
|
| 659 |
+
"learning_rate": 5.73758840278179e-06,
|
| 660 |
+
"loss": 0.620122766494751,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9062424680645939,
|
| 665 |
+
"grad_norm": 0.33615875244140625,
|
| 666 |
+
"learning_rate": 4.741133731774783e-06,
|
| 667 |
+
"loss": 0.640928316116333,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.91588334538443,
|
| 672 |
+
"grad_norm": 0.38269132375717163,
|
| 673 |
+
"learning_rate": 3.837570116244815e-06,
|
| 674 |
+
"loss": 0.6369370937347412,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.925524222704266,
|
| 679 |
+
"grad_norm": 0.36093825101852417,
|
| 680 |
+
"learning_rate": 3.0277786602950776e-06,
|
| 681 |
+
"loss": 0.6047839641571044,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.9351651000241022,
|
| 686 |
+
"grad_norm": 0.36792394518852234,
|
| 687 |
+
"learning_rate": 2.3125490267255414e-06,
|
| 688 |
+
"loss": 0.6083663463592529,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9448059773439383,
|
| 693 |
+
"grad_norm": 0.3627108335494995,
|
| 694 |
+
"learning_rate": 1.6925786669984566e-06,
|
| 695 |
+
"loss": 0.6259819030761719,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9544468546637744,
|
| 700 |
+
"grad_norm": 0.35401254892349243,
|
| 701 |
+
"learning_rate": 1.1684721411231558e-06,
|
| 702 |
+
"loss": 0.6372841835021973,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9640877319836105,
|
| 707 |
+
"grad_norm": 0.34662875533103943,
|
| 708 |
+
"learning_rate": 7.407405281233182e-07,
|
| 709 |
+
"loss": 0.6086503505706787,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9737286093034466,
|
| 714 |
+
"grad_norm": 0.30240076780319214,
|
| 715 |
+
"learning_rate": 4.0980092766169653e-07,
|
| 716 |
+
"loss": 0.6344860076904297,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9833694866232827,
|
| 721 |
+
"grad_norm": 0.3210108280181885,
|
| 722 |
+
"learning_rate": 1.7597605330805788e-07,
|
| 723 |
+
"loss": 0.6058479309082031,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.9930103639431188,
|
| 728 |
+
"grad_norm": 0.36996304988861084,
|
| 729 |
+
"learning_rate": 3.94939178471998e-08,
|
| 730 |
+
"loss": 0.6468423843383789,
|
| 731 |
+
"step": 1030
|
| 732 |
+
}
|
| 733 |
+
],
|
| 734 |
+
"logging_steps": 10,
|
| 735 |
+
"max_steps": 1038,
|
| 736 |
+
"num_input_tokens_seen": 0,
|
| 737 |
+
"num_train_epochs": 1,
|
| 738 |
+
"save_steps": 500,
|
| 739 |
+
"stateful_callbacks": {
|
| 740 |
+
"TrainerControl": {
|
| 741 |
+
"args": {
|
| 742 |
+
"should_epoch_stop": false,
|
| 743 |
+
"should_evaluate": false,
|
| 744 |
+
"should_log": false,
|
| 745 |
+
"should_save": true,
|
| 746 |
+
"should_training_stop": true
|
| 747 |
+
},
|
| 748 |
+
"attributes": {}
|
| 749 |
+
}
|
| 750 |
+
},
|
| 751 |
+
"total_flos": 2.6471972662744044e+18,
|
| 752 |
+
"train_batch_size": 4,
|
| 753 |
+
"trial_name": null,
|
| 754 |
+
"trial_params": null
|
| 755 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/adapter_config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "/home/zhl923/qwen3-8b",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": "pissa",
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 32,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.0,
|
| 22 |
+
"lora_ga_config": null,
|
| 23 |
+
"megatron_config": null,
|
| 24 |
+
"megatron_core": "megatron.core",
|
| 25 |
+
"modules_to_save": null,
|
| 26 |
+
"peft_type": "LORA",
|
| 27 |
+
"peft_version": "0.19.1",
|
| 28 |
+
"qalora_group_size": 16,
|
| 29 |
+
"r": 32,
|
| 30 |
+
"rank_pattern": {},
|
| 31 |
+
"revision": null,
|
| 32 |
+
"target_modules": [
|
| 33 |
+
"gate_proj",
|
| 34 |
+
"v_proj",
|
| 35 |
+
"up_proj",
|
| 36 |
+
"o_proj",
|
| 37 |
+
"k_proj",
|
| 38 |
+
"down_proj",
|
| 39 |
+
"q_proj"
|
| 40 |
+
],
|
| 41 |
+
"target_parameters": null,
|
| 42 |
+
"task_type": "CAUSAL_LM",
|
| 43 |
+
"trainable_token_indices": null,
|
| 44 |
+
"use_bdlora": null,
|
| 45 |
+
"use_dora": false,
|
| 46 |
+
"use_qalora": false,
|
| 47 |
+
"use_rslora": false
|
| 48 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a23cdeac0ed65f2618f2fc0e44576548db33989a64135cdfe4314b8068cb8e0e
|
| 3 |
+
size 349243752
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/base_model_weights.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88f40d465b6d385c4f0b51d16c828798a64b583930ae7789dd89afa7526d9a5c
|
| 3 |
+
size 16381517176
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
| 3 |
+
size 11422650
|
results_code/qwen3/pissa_r32/checkpoint-final/checkpoint-final/tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": true,
|
| 24 |
+
"local_files_only": false,
|
| 25 |
+
"model_max_length": 131072,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|
results_code/qwen3/pissa_r32/checkpoint-final/train_metrics.json
ADDED
|
@@ -0,0 +1,429 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model": "qwen3",
|
| 3 |
+
"method": "pissa",
|
| 4 |
+
"rank": 32,
|
| 5 |
+
"dataset": "m-a-p/Code-Feedback",
|
| 6 |
+
"n_samples": 66383,
|
| 7 |
+
"epochs": 1,
|
| 8 |
+
"total_params": 8278029312,
|
| 9 |
+
"trainable_params": 87293952,
|
| 10 |
+
"trainable_pct": 1.0545,
|
| 11 |
+
"init_time_sec": 445.52,
|
| 12 |
+
"train_time_sec": 29211.5,
|
| 13 |
+
"peak_gpu_mem_gb": 32.9,
|
| 14 |
+
"final_loss": 0.6468,
|
| 15 |
+
"loss_curve": [
|
| 16 |
+
{
|
| 17 |
+
"step": 10,
|
| 18 |
+
"loss": 0.8708
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"step": 20,
|
| 22 |
+
"loss": 0.7254
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"step": 30,
|
| 26 |
+
"loss": 0.7168
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"step": 40,
|
| 30 |
+
"loss": 0.6931
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"step": 50,
|
| 34 |
+
"loss": 0.7164
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"step": 60,
|
| 38 |
+
"loss": 0.7047
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"step": 70,
|
| 42 |
+
"loss": 0.7205
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"step": 80,
|
| 46 |
+
"loss": 0.6821
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"step": 90,
|
| 50 |
+
"loss": 0.6873
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"step": 100,
|
| 54 |
+
"loss": 0.6789
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"step": 110,
|
| 58 |
+
"loss": 0.6964
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"step": 120,
|
| 62 |
+
"loss": 0.6516
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"step": 130,
|
| 66 |
+
"loss": 0.7184
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"step": 140,
|
| 70 |
+
"loss": 0.6581
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"step": 150,
|
| 74 |
+
"loss": 0.6833
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"step": 160,
|
| 78 |
+
"loss": 0.6738
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"step": 170,
|
| 82 |
+
"loss": 0.6931
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"step": 180,
|
| 86 |
+
"loss": 0.6572
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"step": 190,
|
| 90 |
+
"loss": 0.6517
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"step": 200,
|
| 94 |
+
"loss": 0.6472
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"step": 210,
|
| 98 |
+
"loss": 0.6675
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"step": 220,
|
| 102 |
+
"loss": 0.6661
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"step": 230,
|
| 106 |
+
"loss": 0.6912
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"step": 240,
|
| 110 |
+
"loss": 0.6443
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"step": 250,
|
| 114 |
+
"loss": 0.6613
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"step": 260,
|
| 118 |
+
"loss": 0.6513
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"step": 270,
|
| 122 |
+
"loss": 0.6607
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"step": 280,
|
| 126 |
+
"loss": 0.6551
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"step": 290,
|
| 130 |
+
"loss": 0.6324
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"step": 300,
|
| 134 |
+
"loss": 0.6942
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"step": 310,
|
| 138 |
+
"loss": 0.6657
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"step": 320,
|
| 142 |
+
"loss": 0.6789
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"step": 330,
|
| 146 |
+
"loss": 0.6604
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"step": 340,
|
| 150 |
+
"loss": 0.6478
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"step": 350,
|
| 154 |
+
"loss": 0.6722
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"step": 360,
|
| 158 |
+
"loss": 0.6553
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"step": 370,
|
| 162 |
+
"loss": 0.6465
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"step": 380,
|
| 166 |
+
"loss": 0.672
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"step": 390,
|
| 170 |
+
"loss": 0.639
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"step": 400,
|
| 174 |
+
"loss": 0.6818
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"step": 410,
|
| 178 |
+
"loss": 0.6631
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"step": 420,
|
| 182 |
+
"loss": 0.6583
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"step": 430,
|
| 186 |
+
"loss": 0.6545
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"step": 440,
|
| 190 |
+
"loss": 0.6781
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"step": 450,
|
| 194 |
+
"loss": 0.6426
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"step": 460,
|
| 198 |
+
"loss": 0.6464
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"step": 470,
|
| 202 |
+
"loss": 0.6836
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"step": 480,
|
| 206 |
+
"loss": 0.6508
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"step": 490,
|
| 210 |
+
"loss": 0.6604
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"step": 500,
|
| 214 |
+
"loss": 0.6556
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"step": 510,
|
| 218 |
+
"loss": 0.6565
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"step": 520,
|
| 222 |
+
"loss": 0.6298
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"step": 530,
|
| 226 |
+
"loss": 0.6319
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"step": 540,
|
| 230 |
+
"loss": 0.6732
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"step": 550,
|
| 234 |
+
"loss": 0.6435
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"step": 560,
|
| 238 |
+
"loss": 0.6499
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"step": 570,
|
| 242 |
+
"loss": 0.6715
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"step": 580,
|
| 246 |
+
"loss": 0.6553
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"step": 590,
|
| 250 |
+
"loss": 0.6121
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"step": 600,
|
| 254 |
+
"loss": 0.6503
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"step": 610,
|
| 258 |
+
"loss": 0.6337
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"step": 620,
|
| 262 |
+
"loss": 0.6425
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"step": 630,
|
| 266 |
+
"loss": 0.686
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"step": 640,
|
| 270 |
+
"loss": 0.6501
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"step": 650,
|
| 274 |
+
"loss": 0.6581
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"step": 660,
|
| 278 |
+
"loss": 0.6302
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"step": 670,
|
| 282 |
+
"loss": 0.6434
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"step": 680,
|
| 286 |
+
"loss": 0.633
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"step": 690,
|
| 290 |
+
"loss": 0.6349
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"step": 700,
|
| 294 |
+
"loss": 0.6272
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"step": 710,
|
| 298 |
+
"loss": 0.655
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"step": 720,
|
| 302 |
+
"loss": 0.6522
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"step": 730,
|
| 306 |
+
"loss": 0.636
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"step": 740,
|
| 310 |
+
"loss": 0.6496
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"step": 750,
|
| 314 |
+
"loss": 0.6331
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"step": 760,
|
| 318 |
+
"loss": 0.6368
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"step": 770,
|
| 322 |
+
"loss": 0.6265
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"step": 780,
|
| 326 |
+
"loss": 0.6516
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"step": 790,
|
| 330 |
+
"loss": 0.6441
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"step": 800,
|
| 334 |
+
"loss": 0.6313
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"step": 810,
|
| 338 |
+
"loss": 0.6377
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"step": 820,
|
| 342 |
+
"loss": 0.6175
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"step": 830,
|
| 346 |
+
"loss": 0.632
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"step": 840,
|
| 350 |
+
"loss": 0.6516
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"step": 850,
|
| 354 |
+
"loss": 0.6447
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"step": 860,
|
| 358 |
+
"loss": 0.6403
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"step": 870,
|
| 362 |
+
"loss": 0.6234
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"step": 880,
|
| 366 |
+
"loss": 0.6372
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"step": 890,
|
| 370 |
+
"loss": 0.634
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"step": 900,
|
| 374 |
+
"loss": 0.6436
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"step": 910,
|
| 378 |
+
"loss": 0.6311
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"step": 920,
|
| 382 |
+
"loss": 0.62
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"step": 930,
|
| 386 |
+
"loss": 0.6201
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"step": 940,
|
| 390 |
+
"loss": 0.6409
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"step": 950,
|
| 394 |
+
"loss": 0.6369
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"step": 960,
|
| 398 |
+
"loss": 0.6048
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"step": 970,
|
| 402 |
+
"loss": 0.6084
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"step": 980,
|
| 406 |
+
"loss": 0.626
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"step": 990,
|
| 410 |
+
"loss": 0.6373
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"step": 1000,
|
| 414 |
+
"loss": 0.6087
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"step": 1010,
|
| 418 |
+
"loss": 0.6345
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"step": 1020,
|
| 422 |
+
"loss": 0.6058
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"step": 1030,
|
| 426 |
+
"loss": 0.6468
|
| 427 |
+
}
|
| 428 |
+
]
|
| 429 |
+
}
|