upload checkpoint-330
Browse files- .gitattributes +1 -0
- adapter_config.json +51 -0
- adapter_model.safetensors +3 -0
- chat_template.jinja +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +31 -0
- trainer_state.json +1684 -0
- training_args.bin +3 -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
|
adapter_config.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "/home/aadi/.cache/huggingface/hub/models--Qwen--Qwen3.5-9B/snapshots/c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 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": true,
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 64,
|
| 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 |
+
"up_proj",
|
| 34 |
+
"out_proj",
|
| 35 |
+
"o_proj",
|
| 36 |
+
"k_proj",
|
| 37 |
+
"gate_proj",
|
| 38 |
+
"in_proj_ba",
|
| 39 |
+
"down_proj",
|
| 40 |
+
"v_proj",
|
| 41 |
+
"q_proj",
|
| 42 |
+
"in_proj_qkvz"
|
| 43 |
+
],
|
| 44 |
+
"target_parameters": null,
|
| 45 |
+
"task_type": "CAUSAL_LM",
|
| 46 |
+
"trainable_token_indices": null,
|
| 47 |
+
"use_bdlora": null,
|
| 48 |
+
"use_dora": false,
|
| 49 |
+
"use_qalora": false,
|
| 50 |
+
"use_rslora": false
|
| 51 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd926d122fbe468c87395ac569b706160b63a352b901fe42f6be25a1ae2452c0
|
| 3 |
+
size 257990608
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- for message in messages %}{{- '<|im_start|>' + message['role'] + '
|
| 2 |
+
' }}{%- if message['role'] == 'assistant' %}{% generation %}{{ message['content'] }}{% endgeneration %}{%- else %}{{ message['content'] }}{%- endif %}{{- '<|im_end|>
|
| 3 |
+
' }}{%- endfor %}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
|
| 3 |
+
size 19989343
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"model_max_length": 262144,
|
| 14 |
+
"model_specific_special_tokens": {
|
| 15 |
+
"audio_bos_token": "<|audio_start|>",
|
| 16 |
+
"audio_eos_token": "<|audio_end|>",
|
| 17 |
+
"audio_token": "<|audio_pad|>",
|
| 18 |
+
"image_token": "<|image_pad|>",
|
| 19 |
+
"video_token": "<|video_pad|>",
|
| 20 |
+
"vision_bos_token": "<|vision_start|>",
|
| 21 |
+
"vision_eos_token": "<|vision_end|>"
|
| 22 |
+
},
|
| 23 |
+
"pad_token": "<|endoftext|>",
|
| 24 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 25 |
+
"split_special_tokens": false,
|
| 26 |
+
"tokenizer_class": "TokenizersBackend",
|
| 27 |
+
"unk_token": null,
|
| 28 |
+
"video_token": "<|video_pad|>",
|
| 29 |
+
"vision_bos_token": "<|vision_start|>",
|
| 30 |
+
"vision_eos_token": "<|vision_end|>"
|
| 31 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1684 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 330,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.1453630048781633,
|
| 14 |
+
"epoch": 0.006066350710900474,
|
| 15 |
+
"grad_norm": 0.14734439551830292,
|
| 16 |
+
"learning_rate": 6.060606060606061e-06,
|
| 17 |
+
"loss": 1.0426826477050781,
|
| 18 |
+
"mean_token_accuracy": 0.6963728871196508,
|
| 19 |
+
"num_tokens": 86005.0,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.1360931787639856,
|
| 24 |
+
"epoch": 0.012132701421800948,
|
| 25 |
+
"grad_norm": 0.11603065580129623,
|
| 26 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 27 |
+
"loss": 0.9370957612991333,
|
| 28 |
+
"mean_token_accuracy": 0.7134113218635321,
|
| 29 |
+
"num_tokens": 179122.0,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.1894744168967009,
|
| 34 |
+
"epoch": 0.018199052132701423,
|
| 35 |
+
"grad_norm": 0.09422004967927933,
|
| 36 |
+
"learning_rate": 3.0303030303030306e-05,
|
| 37 |
+
"loss": 0.9142184257507324,
|
| 38 |
+
"mean_token_accuracy": 0.7363827731460333,
|
| 39 |
+
"num_tokens": 260270.0,
|
| 40 |
+
"step": 6
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.2145136073231697,
|
| 44 |
+
"epoch": 0.024265402843601895,
|
| 45 |
+
"grad_norm": 0.06607136130332947,
|
| 46 |
+
"learning_rate": 4.242424242424243e-05,
|
| 47 |
+
"loss": 0.7572961449623108,
|
| 48 |
+
"mean_token_accuracy": 0.7646264806389809,
|
| 49 |
+
"num_tokens": 337790.0,
|
| 50 |
+
"step": 8
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.2297200430184603,
|
| 54 |
+
"epoch": 0.03033175355450237,
|
| 55 |
+
"grad_norm": 0.044201456010341644,
|
| 56 |
+
"learning_rate": 5.4545454545454546e-05,
|
| 57 |
+
"loss": 0.7195202112197876,
|
| 58 |
+
"mean_token_accuracy": 0.7812480926513672,
|
| 59 |
+
"num_tokens": 415482.0,
|
| 60 |
+
"step": 10
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 1.188864953815937,
|
| 64 |
+
"epoch": 0.036398104265402846,
|
| 65 |
+
"grad_norm": 0.055669352412223816,
|
| 66 |
+
"learning_rate": 6.666666666666667e-05,
|
| 67 |
+
"loss": 0.6629209518432617,
|
| 68 |
+
"mean_token_accuracy": 0.7944901511073112,
|
| 69 |
+
"num_tokens": 500390.0,
|
| 70 |
+
"step": 12
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 1.1908684950321913,
|
| 74 |
+
"epoch": 0.042464454976303315,
|
| 75 |
+
"grad_norm": 0.06397577375173569,
|
| 76 |
+
"learning_rate": 7.878787878787879e-05,
|
| 77 |
+
"loss": 0.7490695118904114,
|
| 78 |
+
"mean_token_accuracy": 0.7825453281402588,
|
| 79 |
+
"num_tokens": 581958.0,
|
| 80 |
+
"step": 14
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 1.1832681559026241,
|
| 84 |
+
"epoch": 0.04853080568720379,
|
| 85 |
+
"grad_norm": 0.04919273033738136,
|
| 86 |
+
"learning_rate": 9.090909090909092e-05,
|
| 87 |
+
"loss": 0.6293701529502869,
|
| 88 |
+
"mean_token_accuracy": 0.794433705508709,
|
| 89 |
+
"num_tokens": 667139.0,
|
| 90 |
+
"step": 16
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.1474704146385193,
|
| 94 |
+
"epoch": 0.054597156398104266,
|
| 95 |
+
"grad_norm": 0.04964882507920265,
|
| 96 |
+
"learning_rate": 0.00010303030303030303,
|
| 97 |
+
"loss": 0.5306020975112915,
|
| 98 |
+
"mean_token_accuracy": 0.8226693235337734,
|
| 99 |
+
"num_tokens": 760116.0,
|
| 100 |
+
"step": 18
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.1298045758157969,
|
| 104 |
+
"epoch": 0.06066350710900474,
|
| 105 |
+
"grad_norm": 0.05897225812077522,
|
| 106 |
+
"learning_rate": 0.00011515151515151516,
|
| 107 |
+
"loss": 0.608231246471405,
|
| 108 |
+
"mean_token_accuracy": 0.8276951387524605,
|
| 109 |
+
"num_tokens": 846847.0,
|
| 110 |
+
"step": 20
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"entropy": 1.1703982092440128,
|
| 114 |
+
"epoch": 0.06672985781990522,
|
| 115 |
+
"grad_norm": 0.04560555890202522,
|
| 116 |
+
"learning_rate": 0.00012727272727272728,
|
| 117 |
+
"loss": 0.6106629967689514,
|
| 118 |
+
"mean_token_accuracy": 0.8152587823569775,
|
| 119 |
+
"num_tokens": 929643.0,
|
| 120 |
+
"step": 22
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"entropy": 1.1814553122967482,
|
| 124 |
+
"epoch": 0.07279620853080569,
|
| 125 |
+
"grad_norm": 0.05013424530625343,
|
| 126 |
+
"learning_rate": 0.0001393939393939394,
|
| 127 |
+
"loss": 0.6171225309371948,
|
| 128 |
+
"mean_token_accuracy": 0.8334241211414337,
|
| 129 |
+
"num_tokens": 1009374.0,
|
| 130 |
+
"step": 24
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"entropy": 1.1661318112164736,
|
| 134 |
+
"epoch": 0.07886255924170617,
|
| 135 |
+
"grad_norm": 0.060947954654693604,
|
| 136 |
+
"learning_rate": 0.00015151515151515152,
|
| 137 |
+
"loss": 0.5495172739028931,
|
| 138 |
+
"mean_token_accuracy": 0.8268315587192774,
|
| 139 |
+
"num_tokens": 1092017.0,
|
| 140 |
+
"step": 26
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"entropy": 1.2133738622069359,
|
| 144 |
+
"epoch": 0.08492890995260663,
|
| 145 |
+
"grad_norm": 0.0371452271938324,
|
| 146 |
+
"learning_rate": 0.00016363636363636366,
|
| 147 |
+
"loss": 0.4809308648109436,
|
| 148 |
+
"mean_token_accuracy": 0.8424171935766935,
|
| 149 |
+
"num_tokens": 1169179.0,
|
| 150 |
+
"step": 28
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"entropy": 1.159761905670166,
|
| 154 |
+
"epoch": 0.0909952606635071,
|
| 155 |
+
"grad_norm": 0.050072550773620605,
|
| 156 |
+
"learning_rate": 0.00017575757575757578,
|
| 157 |
+
"loss": 0.5333450436592102,
|
| 158 |
+
"mean_token_accuracy": 0.8286342993378639,
|
| 159 |
+
"num_tokens": 1250393.0,
|
| 160 |
+
"step": 30
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"entropy": 1.1467574667185545,
|
| 164 |
+
"epoch": 0.09706161137440758,
|
| 165 |
+
"grad_norm": 0.048133622854948044,
|
| 166 |
+
"learning_rate": 0.0001878787878787879,
|
| 167 |
+
"loss": 0.5849621295928955,
|
| 168 |
+
"mean_token_accuracy": 0.8169677555561066,
|
| 169 |
+
"num_tokens": 1333954.0,
|
| 170 |
+
"step": 32
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"entropy": 1.1128078736364841,
|
| 174 |
+
"epoch": 0.10312796208530806,
|
| 175 |
+
"grad_norm": 0.04460649564862251,
|
| 176 |
+
"learning_rate": 0.0002,
|
| 177 |
+
"loss": 0.6192362904548645,
|
| 178 |
+
"mean_token_accuracy": 0.8245043996721506,
|
| 179 |
+
"num_tokens": 1423831.0,
|
| 180 |
+
"step": 34
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"entropy": 1.0983344987034798,
|
| 184 |
+
"epoch": 0.10919431279620853,
|
| 185 |
+
"grad_norm": 0.034185897558927536,
|
| 186 |
+
"learning_rate": 0.00019999497899070873,
|
| 187 |
+
"loss": 0.4764864146709442,
|
| 188 |
+
"mean_token_accuracy": 0.8249986916780472,
|
| 189 |
+
"num_tokens": 1511985.0,
|
| 190 |
+
"step": 36
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"entropy": 1.1433169227093458,
|
| 194 |
+
"epoch": 0.115260663507109,
|
| 195 |
+
"grad_norm": 0.045258454978466034,
|
| 196 |
+
"learning_rate": 0.00019997991646704551,
|
| 197 |
+
"loss": 0.6904657483100891,
|
| 198 |
+
"mean_token_accuracy": 0.7990726083517075,
|
| 199 |
+
"num_tokens": 1592395.0,
|
| 200 |
+
"step": 38
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"entropy": 1.125565581023693,
|
| 204 |
+
"epoch": 0.12132701421800948,
|
| 205 |
+
"grad_norm": 0.041661474853754044,
|
| 206 |
+
"learning_rate": 0.00019995481394159188,
|
| 207 |
+
"loss": 0.46054211258888245,
|
| 208 |
+
"mean_token_accuracy": 0.8317109365016222,
|
| 209 |
+
"num_tokens": 1680874.0,
|
| 210 |
+
"step": 40
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"entropy": 1.113579124212265,
|
| 214 |
+
"epoch": 0.12739336492890996,
|
| 215 |
+
"grad_norm": 0.038148220628499985,
|
| 216 |
+
"learning_rate": 0.000199919673935148,
|
| 217 |
+
"loss": 0.46925485134124756,
|
| 218 |
+
"mean_token_accuracy": 0.8534338232129812,
|
| 219 |
+
"num_tokens": 1766235.0,
|
| 220 |
+
"step": 42
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"entropy": 1.1232952550053596,
|
| 224 |
+
"epoch": 0.13345971563981043,
|
| 225 |
+
"grad_norm": 0.033694252371788025,
|
| 226 |
+
"learning_rate": 0.0001998744999764799,
|
| 227 |
+
"loss": 0.44831910729408264,
|
| 228 |
+
"mean_token_accuracy": 0.8348242975771427,
|
| 229 |
+
"num_tokens": 1851601.0,
|
| 230 |
+
"step": 44
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"entropy": 1.1306548938155174,
|
| 234 |
+
"epoch": 0.1395260663507109,
|
| 235 |
+
"grad_norm": 0.03199618309736252,
|
| 236 |
+
"learning_rate": 0.00019981929660196492,
|
| 237 |
+
"loss": 0.40715843439102173,
|
| 238 |
+
"mean_token_accuracy": 0.8530372772365808,
|
| 239 |
+
"num_tokens": 1935320.0,
|
| 240 |
+
"step": 46
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"entropy": 1.1122788116335869,
|
| 244 |
+
"epoch": 0.14559241706161138,
|
| 245 |
+
"grad_norm": 0.04616051912307739,
|
| 246 |
+
"learning_rate": 0.00019975406935513615,
|
| 247 |
+
"loss": 0.597642183303833,
|
| 248 |
+
"mean_token_accuracy": 0.8134283572435379,
|
| 249 |
+
"num_tokens": 2020485.0,
|
| 250 |
+
"step": 48
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"entropy": 1.0579780209809542,
|
| 254 |
+
"epoch": 0.15165876777251186,
|
| 255 |
+
"grad_norm": 0.030796177685260773,
|
| 256 |
+
"learning_rate": 0.00019967882478612583,
|
| 257 |
+
"loss": 0.4499819874763489,
|
| 258 |
+
"mean_token_accuracy": 0.856515796855092,
|
| 259 |
+
"num_tokens": 2106486.0,
|
| 260 |
+
"step": 50
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"entropy": 1.0973982904106379,
|
| 264 |
+
"epoch": 0.15772511848341234,
|
| 265 |
+
"grad_norm": 0.039019569754600525,
|
| 266 |
+
"learning_rate": 0.00019959357045100764,
|
| 267 |
+
"loss": 0.4998232424259186,
|
| 268 |
+
"mean_token_accuracy": 0.8463094253093004,
|
| 269 |
+
"num_tokens": 2192072.0,
|
| 270 |
+
"step": 52
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"entropy": 1.1169700846076012,
|
| 274 |
+
"epoch": 0.16379146919431278,
|
| 275 |
+
"grad_norm": 0.037657231092453,
|
| 276 |
+
"learning_rate": 0.00019949831491103764,
|
| 277 |
+
"loss": 0.5231456756591797,
|
| 278 |
+
"mean_token_accuracy": 0.8482763543725014,
|
| 279 |
+
"num_tokens": 2270868.0,
|
| 280 |
+
"step": 54
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"entropy": 1.1035414598882198,
|
| 284 |
+
"epoch": 0.16985781990521326,
|
| 285 |
+
"grad_norm": 0.03033546172082424,
|
| 286 |
+
"learning_rate": 0.00019939306773179497,
|
| 287 |
+
"loss": 0.39799362421035767,
|
| 288 |
+
"mean_token_accuracy": 0.8570526372641325,
|
| 289 |
+
"num_tokens": 2352196.0,
|
| 290 |
+
"step": 56
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"entropy": 1.0822282023727894,
|
| 294 |
+
"epoch": 0.17592417061611373,
|
| 295 |
+
"grad_norm": 0.04231107234954834,
|
| 296 |
+
"learning_rate": 0.00019927783948222084,
|
| 297 |
+
"loss": 0.5166806578636169,
|
| 298 |
+
"mean_token_accuracy": 0.8385925889015198,
|
| 299 |
+
"num_tokens": 2436039.0,
|
| 300 |
+
"step": 58
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"entropy": 1.063869059085846,
|
| 304 |
+
"epoch": 0.1819905213270142,
|
| 305 |
+
"grad_norm": 0.03918987140059471,
|
| 306 |
+
"learning_rate": 0.0001991526417335575,
|
| 307 |
+
"loss": 0.5763710141181946,
|
| 308 |
+
"mean_token_accuracy": 0.8299813587218523,
|
| 309 |
+
"num_tokens": 2529311.0,
|
| 310 |
+
"step": 60
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"entropy": 1.063054159283638,
|
| 314 |
+
"epoch": 0.18805687203791469,
|
| 315 |
+
"grad_norm": 0.030817104503512383,
|
| 316 |
+
"learning_rate": 0.0001990174870581862,
|
| 317 |
+
"loss": 0.41522204875946045,
|
| 318 |
+
"mean_token_accuracy": 0.8667358681559563,
|
| 319 |
+
"num_tokens": 2617613.0,
|
| 320 |
+
"step": 62
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"entropy": 1.0713623315095901,
|
| 324 |
+
"epoch": 0.19412322274881516,
|
| 325 |
+
"grad_norm": 0.05068995803594589,
|
| 326 |
+
"learning_rate": 0.00019887238902836448,
|
| 327 |
+
"loss": 0.541697084903717,
|
| 328 |
+
"mean_token_accuracy": 0.8430546559393406,
|
| 329 |
+
"num_tokens": 2704701.0,
|
| 330 |
+
"step": 64
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"entropy": 1.0977294705808163,
|
| 334 |
+
"epoch": 0.20018957345971564,
|
| 335 |
+
"grad_norm": 0.03129807487130165,
|
| 336 |
+
"learning_rate": 0.00019871736221486346,
|
| 337 |
+
"loss": 0.44492095708847046,
|
| 338 |
+
"mean_token_accuracy": 0.8514020573347807,
|
| 339 |
+
"num_tokens": 2788486.0,
|
| 340 |
+
"step": 66
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"entropy": 1.0614943969994783,
|
| 344 |
+
"epoch": 0.2062559241706161,
|
| 345 |
+
"grad_norm": 0.036761052906513214,
|
| 346 |
+
"learning_rate": 0.00019855242218550466,
|
| 347 |
+
"loss": 0.43396472930908203,
|
| 348 |
+
"mean_token_accuracy": 0.8538590967655182,
|
| 349 |
+
"num_tokens": 2873289.0,
|
| 350 |
+
"step": 68
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"entropy": 1.1275603212416172,
|
| 354 |
+
"epoch": 0.2123222748815166,
|
| 355 |
+
"grad_norm": 0.03448108583688736,
|
| 356 |
+
"learning_rate": 0.00019837758550359636,
|
| 357 |
+
"loss": 0.4181627333164215,
|
| 358 |
+
"mean_token_accuracy": 0.8666124399751425,
|
| 359 |
+
"num_tokens": 2948652.0,
|
| 360 |
+
"step": 70
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"entropy": 1.0930265486240387,
|
| 364 |
+
"epoch": 0.21838862559241706,
|
| 365 |
+
"grad_norm": 0.028693048283457756,
|
| 366 |
+
"learning_rate": 0.00019819286972627066,
|
| 367 |
+
"loss": 0.4703572392463684,
|
| 368 |
+
"mean_token_accuracy": 0.8563728202134371,
|
| 369 |
+
"num_tokens": 3029064.0,
|
| 370 |
+
"step": 72
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"entropy": 1.1084971856325865,
|
| 374 |
+
"epoch": 0.22445497630331754,
|
| 375 |
+
"grad_norm": 0.0367777943611145,
|
| 376 |
+
"learning_rate": 0.00019799829340272034,
|
| 377 |
+
"loss": 0.41802218556404114,
|
| 378 |
+
"mean_token_accuracy": 0.8781944885849953,
|
| 379 |
+
"num_tokens": 3107187.0,
|
| 380 |
+
"step": 74
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"entropy": 1.0890926662832499,
|
| 384 |
+
"epoch": 0.230521327014218,
|
| 385 |
+
"grad_norm": 0.029872339218854904,
|
| 386 |
+
"learning_rate": 0.00019779387607233586,
|
| 387 |
+
"loss": 0.43851789832115173,
|
| 388 |
+
"mean_token_accuracy": 0.8546587247401476,
|
| 389 |
+
"num_tokens": 3195901.0,
|
| 390 |
+
"step": 76
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"entropy": 1.1280810628086329,
|
| 394 |
+
"epoch": 0.2365876777251185,
|
| 395 |
+
"grad_norm": 0.03960474207997322,
|
| 396 |
+
"learning_rate": 0.00019757963826274357,
|
| 397 |
+
"loss": 0.42227938771247864,
|
| 398 |
+
"mean_token_accuracy": 0.8712473902851343,
|
| 399 |
+
"num_tokens": 3274625.0,
|
| 400 |
+
"step": 78
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"entropy": 1.114132758229971,
|
| 404 |
+
"epoch": 0.24265402843601896,
|
| 405 |
+
"grad_norm": 0.035365257412195206,
|
| 406 |
+
"learning_rate": 0.00019735560148774413,
|
| 407 |
+
"loss": 0.4249793291091919,
|
| 408 |
+
"mean_token_accuracy": 0.88199676387012,
|
| 409 |
+
"num_tokens": 3358849.0,
|
| 410 |
+
"step": 80
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"entropy": 1.1247058976441622,
|
| 414 |
+
"epoch": 0.24872037914691944,
|
| 415 |
+
"grad_norm": 0.02779199369251728,
|
| 416 |
+
"learning_rate": 0.00019712178824515212,
|
| 417 |
+
"loss": 0.38992127776145935,
|
| 418 |
+
"mean_token_accuracy": 0.8719186093658209,
|
| 419 |
+
"num_tokens": 3442378.0,
|
| 420 |
+
"step": 82
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"entropy": 1.0901147946715355,
|
| 424 |
+
"epoch": 0.2547867298578199,
|
| 425 |
+
"grad_norm": 0.031184526160359383,
|
| 426 |
+
"learning_rate": 0.00019687822201453674,
|
| 427 |
+
"loss": 0.47365424036979675,
|
| 428 |
+
"mean_token_accuracy": 0.852692287415266,
|
| 429 |
+
"num_tokens": 3524832.0,
|
| 430 |
+
"step": 84
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"entropy": 1.0656033772975206,
|
| 434 |
+
"epoch": 0.26085308056872036,
|
| 435 |
+
"grad_norm": 0.03132278472185135,
|
| 436 |
+
"learning_rate": 0.0001966249272548642,
|
| 437 |
+
"loss": 0.33241140842437744,
|
| 438 |
+
"mean_token_accuracy": 0.8878429923206568,
|
| 439 |
+
"num_tokens": 3617601.0,
|
| 440 |
+
"step": 86
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"entropy": 1.0888057127594948,
|
| 444 |
+
"epoch": 0.26691943127962087,
|
| 445 |
+
"grad_norm": 0.031927064061164856,
|
| 446 |
+
"learning_rate": 0.00019636192940204134,
|
| 447 |
+
"loss": 0.4007760286331177,
|
| 448 |
+
"mean_token_accuracy": 0.8706978503614664,
|
| 449 |
+
"num_tokens": 3704884.0,
|
| 450 |
+
"step": 88
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"entropy": 1.1373948883265257,
|
| 454 |
+
"epoch": 0.2729857819905213,
|
| 455 |
+
"grad_norm": 0.031416162848472595,
|
| 456 |
+
"learning_rate": 0.00019608925486636137,
|
| 457 |
+
"loss": 0.4319029450416565,
|
| 458 |
+
"mean_token_accuracy": 0.8511507343500853,
|
| 459 |
+
"num_tokens": 3787613.0,
|
| 460 |
+
"step": 90
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"entropy": 1.2081235498189926,
|
| 464 |
+
"epoch": 0.2790521327014218,
|
| 465 |
+
"grad_norm": 0.03321005031466484,
|
| 466 |
+
"learning_rate": 0.00019580693102985185,
|
| 467 |
+
"loss": 0.4001643657684326,
|
| 468 |
+
"mean_token_accuracy": 0.8825173135846853,
|
| 469 |
+
"num_tokens": 3858798.0,
|
| 470 |
+
"step": 92
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"entropy": 1.1129657309502363,
|
| 474 |
+
"epoch": 0.28511848341232227,
|
| 475 |
+
"grad_norm": 0.03393163904547691,
|
| 476 |
+
"learning_rate": 0.00019551498624352496,
|
| 477 |
+
"loss": 0.5376895070075989,
|
| 478 |
+
"mean_token_accuracy": 0.8365297745913267,
|
| 479 |
+
"num_tokens": 3945038.0,
|
| 480 |
+
"step": 94
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"entropy": 1.1166121661663055,
|
| 484 |
+
"epoch": 0.29118483412322277,
|
| 485 |
+
"grad_norm": 0.03640655428171158,
|
| 486 |
+
"learning_rate": 0.0001952134498245303,
|
| 487 |
+
"loss": 0.4394976496696472,
|
| 488 |
+
"mean_token_accuracy": 0.8726736102253199,
|
| 489 |
+
"num_tokens": 4031223.0,
|
| 490 |
+
"step": 96
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"entropy": 1.0974710527807474,
|
| 494 |
+
"epoch": 0.2972511848341232,
|
| 495 |
+
"grad_norm": 0.027308620512485504,
|
| 496 |
+
"learning_rate": 0.00019490235205321113,
|
| 497 |
+
"loss": 0.38598430156707764,
|
| 498 |
+
"mean_token_accuracy": 0.8961432129144669,
|
| 499 |
+
"num_tokens": 4115115.0,
|
| 500 |
+
"step": 98
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"entropy": 1.0804498698562384,
|
| 504 |
+
"epoch": 0.3033175355450237,
|
| 505 |
+
"grad_norm": 0.027602896094322205,
|
| 506 |
+
"learning_rate": 0.00019458172417006347,
|
| 507 |
+
"loss": 0.3249405026435852,
|
| 508 |
+
"mean_token_accuracy": 0.8995124101638794,
|
| 509 |
+
"num_tokens": 4197509.0,
|
| 510 |
+
"step": 100
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"entropy": 1.087203273549676,
|
| 514 |
+
"epoch": 0.30938388625592417,
|
| 515 |
+
"grad_norm": 0.03812822327017784,
|
| 516 |
+
"learning_rate": 0.00019425159837259894,
|
| 517 |
+
"loss": 0.4760231673717499,
|
| 518 |
+
"mean_token_accuracy": 0.859690461307764,
|
| 519 |
+
"num_tokens": 4284514.0,
|
| 520 |
+
"step": 102
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"entropy": 1.13011266104877,
|
| 524 |
+
"epoch": 0.31545023696682467,
|
| 525 |
+
"grad_norm": 0.03254610672593117,
|
| 526 |
+
"learning_rate": 0.00019391200781211144,
|
| 527 |
+
"loss": 0.3947427570819855,
|
| 528 |
+
"mean_token_accuracy": 0.8607379104942083,
|
| 529 |
+
"num_tokens": 4368568.0,
|
| 530 |
+
"step": 104
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"entropy": 1.1309633571654558,
|
| 534 |
+
"epoch": 0.3215165876777251,
|
| 535 |
+
"grad_norm": 0.03166705742478371,
|
| 536 |
+
"learning_rate": 0.00019356298659034817,
|
| 537 |
+
"loss": 0.3969774842262268,
|
| 538 |
+
"mean_token_accuracy": 0.8869058713316917,
|
| 539 |
+
"num_tokens": 4453471.0,
|
| 540 |
+
"step": 106
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"entropy": 1.1543912217020988,
|
| 544 |
+
"epoch": 0.32758293838862557,
|
| 545 |
+
"grad_norm": 0.030770013108849525,
|
| 546 |
+
"learning_rate": 0.0001932045697560851,
|
| 547 |
+
"loss": 0.4325692355632782,
|
| 548 |
+
"mean_token_accuracy": 0.884726457297802,
|
| 549 |
+
"num_tokens": 4531931.0,
|
| 550 |
+
"step": 108
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"entropy": 1.08841666392982,
|
| 554 |
+
"epoch": 0.33364928909952607,
|
| 555 |
+
"grad_norm": 0.02102690003812313,
|
| 556 |
+
"learning_rate": 0.00019283679330160726,
|
| 557 |
+
"loss": 0.3601663410663605,
|
| 558 |
+
"mean_token_accuracy": 0.8804576676338911,
|
| 559 |
+
"num_tokens": 4622331.0,
|
| 560 |
+
"step": 110
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"entropy": 1.0651407334953547,
|
| 564 |
+
"epoch": 0.3397156398104265,
|
| 565 |
+
"grad_norm": 0.029801692813634872,
|
| 566 |
+
"learning_rate": 0.00019245969415909465,
|
| 567 |
+
"loss": 0.3369869887828827,
|
| 568 |
+
"mean_token_accuracy": 0.8879600297659636,
|
| 569 |
+
"num_tokens": 4707482.0,
|
| 570 |
+
"step": 112
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"entropy": 1.0574357081204653,
|
| 574 |
+
"epoch": 0.345781990521327,
|
| 575 |
+
"grad_norm": 0.02319478802382946,
|
| 576 |
+
"learning_rate": 0.00019207331019691312,
|
| 577 |
+
"loss": 0.40834516286849976,
|
| 578 |
+
"mean_token_accuracy": 0.8533938657492399,
|
| 579 |
+
"num_tokens": 4795769.0,
|
| 580 |
+
"step": 114
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"entropy": 1.0692832916975021,
|
| 584 |
+
"epoch": 0.35184834123222747,
|
| 585 |
+
"grad_norm": 0.029467541724443436,
|
| 586 |
+
"learning_rate": 0.00019167768021581208,
|
| 587 |
+
"loss": 0.3681870102882385,
|
| 588 |
+
"mean_token_accuracy": 0.8964284509420395,
|
| 589 |
+
"num_tokens": 4878294.0,
|
| 590 |
+
"step": 116
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"entropy": 1.0285058580338955,
|
| 594 |
+
"epoch": 0.35791469194312797,
|
| 595 |
+
"grad_norm": 0.024983646348118782,
|
| 596 |
+
"learning_rate": 0.0001912728439450276,
|
| 597 |
+
"loss": 0.3748064637184143,
|
| 598 |
+
"mean_token_accuracy": 0.8837519232183695,
|
| 599 |
+
"num_tokens": 4968400.0,
|
| 600 |
+
"step": 118
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"entropy": 1.1153130996972322,
|
| 604 |
+
"epoch": 0.3639810426540284,
|
| 605 |
+
"grad_norm": 0.023784542456269264,
|
| 606 |
+
"learning_rate": 0.00019085884203829342,
|
| 607 |
+
"loss": 0.32293763756752014,
|
| 608 |
+
"mean_token_accuracy": 0.8811638858169317,
|
| 609 |
+
"num_tokens": 5048307.0,
|
| 610 |
+
"step": 120
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"entropy": 1.0716180335730314,
|
| 614 |
+
"epoch": 0.3700473933649289,
|
| 615 |
+
"grad_norm": 0.02309606969356537,
|
| 616 |
+
"learning_rate": 0.00019043571606975777,
|
| 617 |
+
"loss": 0.2915813624858856,
|
| 618 |
+
"mean_token_accuracy": 0.9026055429130793,
|
| 619 |
+
"num_tokens": 5134389.0,
|
| 620 |
+
"step": 122
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 1.0711108148097992,
|
| 624 |
+
"epoch": 0.37611374407582937,
|
| 625 |
+
"grad_norm": 0.03030751645565033,
|
| 626 |
+
"learning_rate": 0.00019000350852980909,
|
| 627 |
+
"loss": 0.3965684473514557,
|
| 628 |
+
"mean_token_accuracy": 0.8851604331284761,
|
| 629 |
+
"num_tokens": 5222669.0,
|
| 630 |
+
"step": 124
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 1.1374715939164162,
|
| 634 |
+
"epoch": 0.3821800947867299,
|
| 635 |
+
"grad_norm": 0.038585301488637924,
|
| 636 |
+
"learning_rate": 0.0001895622628208089,
|
| 637 |
+
"loss": 0.4160141348838806,
|
| 638 |
+
"mean_token_accuracy": 0.8680315278470516,
|
| 639 |
+
"num_tokens": 5298893.0,
|
| 640 |
+
"step": 126
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 1.066998591646552,
|
| 644 |
+
"epoch": 0.3882464454976303,
|
| 645 |
+
"grad_norm": 0.027405690401792526,
|
| 646 |
+
"learning_rate": 0.00018911202325273325,
|
| 647 |
+
"loss": 0.32344281673431396,
|
| 648 |
+
"mean_token_accuracy": 0.894629979506135,
|
| 649 |
+
"num_tokens": 5385393.0,
|
| 650 |
+
"step": 128
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 1.1246034801006317,
|
| 654 |
+
"epoch": 0.3943127962085308,
|
| 655 |
+
"grad_norm": 0.02833821251988411,
|
| 656 |
+
"learning_rate": 0.00018865283503872324,
|
| 657 |
+
"loss": 0.4167165756225586,
|
| 658 |
+
"mean_token_accuracy": 0.8629612661898136,
|
| 659 |
+
"num_tokens": 5469298.0,
|
| 660 |
+
"step": 130
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 1.1521866731345654,
|
| 664 |
+
"epoch": 0.4003791469194313,
|
| 665 |
+
"grad_norm": 0.02951820380985737,
|
| 666 |
+
"learning_rate": 0.00018818474429054463,
|
| 667 |
+
"loss": 0.37589696049690247,
|
| 668 |
+
"mean_token_accuracy": 0.8920078761875629,
|
| 669 |
+
"num_tokens": 5543910.0,
|
| 670 |
+
"step": 132
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 1.0356314852833748,
|
| 674 |
+
"epoch": 0.4064454976303318,
|
| 675 |
+
"grad_norm": 0.01839367486536503,
|
| 676 |
+
"learning_rate": 0.0001877077980139574,
|
| 677 |
+
"loss": 0.28368115425109863,
|
| 678 |
+
"mean_token_accuracy": 0.8818562850356102,
|
| 679 |
+
"num_tokens": 5647162.0,
|
| 680 |
+
"step": 134
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 1.064600083976984,
|
| 684 |
+
"epoch": 0.4125118483412322,
|
| 685 |
+
"grad_norm": 0.027259480208158493,
|
| 686 |
+
"learning_rate": 0.0001872220441039952,
|
| 687 |
+
"loss": 0.4667126536369324,
|
| 688 |
+
"mean_token_accuracy": 0.8528099935501814,
|
| 689 |
+
"num_tokens": 5738448.0,
|
| 690 |
+
"step": 136
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 1.0653144717216492,
|
| 694 |
+
"epoch": 0.4185781990521327,
|
| 695 |
+
"grad_norm": 0.044929664582014084,
|
| 696 |
+
"learning_rate": 0.00018672753134015596,
|
| 697 |
+
"loss": 0.37731876969337463,
|
| 698 |
+
"mean_token_accuracy": 0.8820401430130005,
|
| 699 |
+
"num_tokens": 5829245.0,
|
| 700 |
+
"step": 138
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 1.1369449868798256,
|
| 704 |
+
"epoch": 0.4246445497630332,
|
| 705 |
+
"grad_norm": 0.028195707127451897,
|
| 706 |
+
"learning_rate": 0.00018622430938150338,
|
| 707 |
+
"loss": 0.3843649923801422,
|
| 708 |
+
"mean_token_accuracy": 0.8765210602432489,
|
| 709 |
+
"num_tokens": 5905331.0,
|
| 710 |
+
"step": 140
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 1.1133542396128178,
|
| 714 |
+
"epoch": 0.4307109004739336,
|
| 715 |
+
"grad_norm": 0.0323888398706913,
|
| 716 |
+
"learning_rate": 0.00018571242876167996,
|
| 717 |
+
"loss": 0.40193086862564087,
|
| 718 |
+
"mean_token_accuracy": 0.8878476917743683,
|
| 719 |
+
"num_tokens": 5986315.0,
|
| 720 |
+
"step": 142
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"entropy": 1.0604677107185125,
|
| 724 |
+
"epoch": 0.4367772511848341,
|
| 725 |
+
"grad_norm": 0.033294927328825,
|
| 726 |
+
"learning_rate": 0.00018519194088383273,
|
| 727 |
+
"loss": 0.479554682970047,
|
| 728 |
+
"mean_token_accuracy": 0.8688712641596794,
|
| 729 |
+
"num_tokens": 6079217.0,
|
| 730 |
+
"step": 144
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 1.0793759282678366,
|
| 734 |
+
"epoch": 0.4428436018957346,
|
| 735 |
+
"grad_norm": 0.028335442766547203,
|
| 736 |
+
"learning_rate": 0.00018466289801545105,
|
| 737 |
+
"loss": 0.3876863121986389,
|
| 738 |
+
"mean_token_accuracy": 0.8869561217725277,
|
| 739 |
+
"num_tokens": 6162765.0,
|
| 740 |
+
"step": 146
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 1.0978383664041758,
|
| 744 |
+
"epoch": 0.4489099526066351,
|
| 745 |
+
"grad_norm": 0.024287773296236992,
|
| 746 |
+
"learning_rate": 0.00018412535328311814,
|
| 747 |
+
"loss": 0.37590470910072327,
|
| 748 |
+
"mean_token_accuracy": 0.8878515716642141,
|
| 749 |
+
"num_tokens": 6243547.0,
|
| 750 |
+
"step": 148
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 1.0713757798075676,
|
| 754 |
+
"epoch": 0.4549763033175355,
|
| 755 |
+
"grad_norm": 0.02222955971956253,
|
| 756 |
+
"learning_rate": 0.00018357936066717585,
|
| 757 |
+
"loss": 0.40934571623802185,
|
| 758 |
+
"mean_token_accuracy": 0.8742999695241451,
|
| 759 |
+
"num_tokens": 6326139.0,
|
| 760 |
+
"step": 150
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 1.0939715839922428,
|
| 764 |
+
"epoch": 0.461042654028436,
|
| 765 |
+
"grad_norm": 0.036705683916807175,
|
| 766 |
+
"learning_rate": 0.00018302497499630415,
|
| 767 |
+
"loss": 0.4863278269767761,
|
| 768 |
+
"mean_token_accuracy": 0.8565606363117695,
|
| 769 |
+
"num_tokens": 6408101.0,
|
| 770 |
+
"step": 152
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 1.052231615409255,
|
| 774 |
+
"epoch": 0.4671090047393365,
|
| 775 |
+
"grad_norm": 0.03086651675403118,
|
| 776 |
+
"learning_rate": 0.00018246225194201517,
|
| 777 |
+
"loss": 0.48710301518440247,
|
| 778 |
+
"mean_token_accuracy": 0.8606483079493046,
|
| 779 |
+
"num_tokens": 6496996.0,
|
| 780 |
+
"step": 154
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 1.0684311669319868,
|
| 784 |
+
"epoch": 0.473175355450237,
|
| 785 |
+
"grad_norm": 0.032397400587797165,
|
| 786 |
+
"learning_rate": 0.00018189124801306253,
|
| 787 |
+
"loss": 0.45748335123062134,
|
| 788 |
+
"mean_token_accuracy": 0.8623519483953714,
|
| 789 |
+
"num_tokens": 6583609.0,
|
| 790 |
+
"step": 156
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 1.0559804383665323,
|
| 794 |
+
"epoch": 0.4792417061611374,
|
| 795 |
+
"grad_norm": 0.022603873163461685,
|
| 796 |
+
"learning_rate": 0.00018131202054976687,
|
| 797 |
+
"loss": 0.3581934869289398,
|
| 798 |
+
"mean_token_accuracy": 0.8731643594801426,
|
| 799 |
+
"num_tokens": 6668491.0,
|
| 800 |
+
"step": 158
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.9723871815949678,
|
| 804 |
+
"epoch": 0.48530805687203793,
|
| 805 |
+
"grad_norm": 0.02817658893764019,
|
| 806 |
+
"learning_rate": 0.0001807246277182578,
|
| 807 |
+
"loss": 0.3579464256763458,
|
| 808 |
+
"mean_token_accuracy": 0.893825015053153,
|
| 809 |
+
"num_tokens": 6764577.0,
|
| 810 |
+
"step": 160
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 1.0638125743716955,
|
| 814 |
+
"epoch": 0.4913744075829384,
|
| 815 |
+
"grad_norm": 0.025951461866497993,
|
| 816 |
+
"learning_rate": 0.00018012912850463248,
|
| 817 |
+
"loss": 0.42468294501304626,
|
| 818 |
+
"mean_token_accuracy": 0.8819459583610296,
|
| 819 |
+
"num_tokens": 6842543.0,
|
| 820 |
+
"step": 162
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 1.0202238596975803,
|
| 824 |
+
"epoch": 0.4974407582938389,
|
| 825 |
+
"grad_norm": 0.03039274737238884,
|
| 826 |
+
"learning_rate": 0.0001795255827090327,
|
| 827 |
+
"loss": 0.380432665348053,
|
| 828 |
+
"mean_token_accuracy": 0.8789206333458424,
|
| 829 |
+
"num_tokens": 6927268.0,
|
| 830 |
+
"step": 164
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"entropy": 0.9518362749367952,
|
| 834 |
+
"epoch": 0.5035071090047394,
|
| 835 |
+
"grad_norm": 0.03697855398058891,
|
| 836 |
+
"learning_rate": 0.00017891405093963938,
|
| 837 |
+
"loss": 0.4224241375923157,
|
| 838 |
+
"mean_token_accuracy": 0.8823022078722715,
|
| 839 |
+
"num_tokens": 7021488.0,
|
| 840 |
+
"step": 166
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"entropy": 1.007318314164877,
|
| 844 |
+
"epoch": 0.5095734597156398,
|
| 845 |
+
"grad_norm": 0.021177517250180244,
|
| 846 |
+
"learning_rate": 0.0001782945946065864,
|
| 847 |
+
"loss": 0.36274152994155884,
|
| 848 |
+
"mean_token_accuracy": 0.8785165976732969,
|
| 849 |
+
"num_tokens": 7106653.0,
|
| 850 |
+
"step": 168
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"entropy": 0.9980434011667967,
|
| 854 |
+
"epoch": 0.5156398104265403,
|
| 855 |
+
"grad_norm": 0.029106592759490013,
|
| 856 |
+
"learning_rate": 0.00017766727591579388,
|
| 857 |
+
"loss": 0.3869830071926117,
|
| 858 |
+
"mean_token_accuracy": 0.8878410588949919,
|
| 859 |
+
"num_tokens": 7200510.0,
|
| 860 |
+
"step": 170
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"entropy": 1.0423732530325651,
|
| 864 |
+
"epoch": 0.5217061611374407,
|
| 865 |
+
"grad_norm": 0.02796538546681404,
|
| 866 |
+
"learning_rate": 0.0001770321578627213,
|
| 867 |
+
"loss": 0.3639932870864868,
|
| 868 |
+
"mean_token_accuracy": 0.8863089494407177,
|
| 869 |
+
"num_tokens": 7284329.0,
|
| 870 |
+
"step": 172
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"entropy": 1.0723978746682405,
|
| 874 |
+
"epoch": 0.5277725118483413,
|
| 875 |
+
"grad_norm": 0.03264309838414192,
|
| 876 |
+
"learning_rate": 0.0001763893042260416,
|
| 877 |
+
"loss": 0.3936094641685486,
|
| 878 |
+
"mean_token_accuracy": 0.8959641829133034,
|
| 879 |
+
"num_tokens": 7358843.0,
|
| 880 |
+
"step": 174
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"entropy": 1.020014975219965,
|
| 884 |
+
"epoch": 0.5338388625592417,
|
| 885 |
+
"grad_norm": 0.03520318865776062,
|
| 886 |
+
"learning_rate": 0.00017573877956123638,
|
| 887 |
+
"loss": 0.36571091413497925,
|
| 888 |
+
"mean_token_accuracy": 0.8804904017597437,
|
| 889 |
+
"num_tokens": 7445719.0,
|
| 890 |
+
"step": 176
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"entropy": 1.0879373736679554,
|
| 894 |
+
"epoch": 0.5399052132701422,
|
| 895 |
+
"grad_norm": 0.024853892624378204,
|
| 896 |
+
"learning_rate": 0.00017508064919411344,
|
| 897 |
+
"loss": 0.3648242950439453,
|
| 898 |
+
"mean_token_accuracy": 0.8924454972147942,
|
| 899 |
+
"num_tokens": 7523168.0,
|
| 900 |
+
"step": 178
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"entropy": 1.0175370275974274,
|
| 904 |
+
"epoch": 0.5459715639810426,
|
| 905 |
+
"grad_norm": 0.022838793694972992,
|
| 906 |
+
"learning_rate": 0.00017441497921424649,
|
| 907 |
+
"loss": 0.3366798162460327,
|
| 908 |
+
"mean_token_accuracy": 0.8918642085045576,
|
| 909 |
+
"num_tokens": 7608384.0,
|
| 910 |
+
"step": 180
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"entropy": 1.044716577976942,
|
| 914 |
+
"epoch": 0.5520379146919431,
|
| 915 |
+
"grad_norm": 0.02847268059849739,
|
| 916 |
+
"learning_rate": 0.0001737418364683386,
|
| 917 |
+
"loss": 0.3668797016143799,
|
| 918 |
+
"mean_token_accuracy": 0.8863429147750139,
|
| 919 |
+
"num_tokens": 7694243.0,
|
| 920 |
+
"step": 182
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"entropy": 1.0394886285066605,
|
| 924 |
+
"epoch": 0.5581042654028436,
|
| 925 |
+
"grad_norm": 0.030523963272571564,
|
| 926 |
+
"learning_rate": 0.00017306128855350942,
|
| 927 |
+
"loss": 0.43834301829338074,
|
| 928 |
+
"mean_token_accuracy": 0.8756468072533607,
|
| 929 |
+
"num_tokens": 7779328.0,
|
| 930 |
+
"step": 184
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"entropy": 1.0138581153005362,
|
| 934 |
+
"epoch": 0.5641706161137441,
|
| 935 |
+
"grad_norm": 0.02064625732600689,
|
| 936 |
+
"learning_rate": 0.00017237340381050703,
|
| 937 |
+
"loss": 0.2971416711807251,
|
| 938 |
+
"mean_token_accuracy": 0.8839075043797493,
|
| 939 |
+
"num_tokens": 7867661.0,
|
| 940 |
+
"step": 186
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"entropy": 1.0050568506121635,
|
| 944 |
+
"epoch": 0.5702369668246445,
|
| 945 |
+
"grad_norm": 0.02642105333507061,
|
| 946 |
+
"learning_rate": 0.00017167825131684513,
|
| 947 |
+
"loss": 0.3613814413547516,
|
| 948 |
+
"mean_token_accuracy": 0.8899861499667168,
|
| 949 |
+
"num_tokens": 7954692.0,
|
| 950 |
+
"step": 188
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"entropy": 1.0598428342491388,
|
| 954 |
+
"epoch": 0.576303317535545,
|
| 955 |
+
"grad_norm": 0.029336227104067802,
|
| 956 |
+
"learning_rate": 0.00017097590087986633,
|
| 957 |
+
"loss": 0.3905699551105499,
|
| 958 |
+
"mean_token_accuracy": 0.8698614034801722,
|
| 959 |
+
"num_tokens": 8035095.0,
|
| 960 |
+
"step": 190
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"entropy": 1.085257513448596,
|
| 964 |
+
"epoch": 0.5823696682464455,
|
| 965 |
+
"grad_norm": 0.029184900224208832,
|
| 966 |
+
"learning_rate": 0.00017026642302973204,
|
| 967 |
+
"loss": 0.29052889347076416,
|
| 968 |
+
"mean_token_accuracy": 0.9125291369855404,
|
| 969 |
+
"num_tokens": 8110214.0,
|
| 970 |
+
"step": 192
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"entropy": 1.0109102819114923,
|
| 974 |
+
"epoch": 0.588436018957346,
|
| 975 |
+
"grad_norm": 0.02269585430622101,
|
| 976 |
+
"learning_rate": 0.00016954988901233975,
|
| 977 |
+
"loss": 0.31015294790267944,
|
| 978 |
+
"mean_token_accuracy": 0.8990396484732628,
|
| 979 |
+
"num_tokens": 8192843.0,
|
| 980 |
+
"step": 194
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"entropy": 1.0097952149808407,
|
| 984 |
+
"epoch": 0.5945023696682464,
|
| 985 |
+
"grad_norm": 0.0359276607632637,
|
| 986 |
+
"learning_rate": 0.00016882637078216868,
|
| 987 |
+
"loss": 0.3679681718349457,
|
| 988 |
+
"mean_token_accuracy": 0.8857240658253431,
|
| 989 |
+
"num_tokens": 8278978.0,
|
| 990 |
+
"step": 196
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"entropy": 1.0255634803324938,
|
| 994 |
+
"epoch": 0.6005687203791469,
|
| 995 |
+
"grad_norm": 0.02525828778743744,
|
| 996 |
+
"learning_rate": 0.00016809594099505395,
|
| 997 |
+
"loss": 0.3437214195728302,
|
| 998 |
+
"mean_token_accuracy": 0.8847600240260363,
|
| 999 |
+
"num_tokens": 8364237.0,
|
| 1000 |
+
"step": 198
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"entropy": 1.019953290000558,
|
| 1004 |
+
"epoch": 0.6066350710900474,
|
| 1005 |
+
"grad_norm": 0.02443118393421173,
|
| 1006 |
+
"learning_rate": 0.00016735867300089052,
|
| 1007 |
+
"loss": 0.31955114006996155,
|
| 1008 |
+
"mean_token_accuracy": 0.8951585032045841,
|
| 1009 |
+
"num_tokens": 8447472.0,
|
| 1010 |
+
"step": 200
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"entropy": 1.0121128801256418,
|
| 1014 |
+
"epoch": 0.6127014218009479,
|
| 1015 |
+
"grad_norm": 0.03079301491379738,
|
| 1016 |
+
"learning_rate": 0.00016661464083626734,
|
| 1017 |
+
"loss": 0.3247457444667816,
|
| 1018 |
+
"mean_token_accuracy": 0.9060451090335846,
|
| 1019 |
+
"num_tokens": 8532253.0,
|
| 1020 |
+
"step": 202
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"entropy": 1.0057248044759035,
|
| 1024 |
+
"epoch": 0.6187677725118483,
|
| 1025 |
+
"grad_norm": 0.023939011618494987,
|
| 1026 |
+
"learning_rate": 0.00016586391921703269,
|
| 1027 |
+
"loss": 0.31284603476524353,
|
| 1028 |
+
"mean_token_accuracy": 0.8891441822052002,
|
| 1029 |
+
"num_tokens": 8618958.0,
|
| 1030 |
+
"step": 204
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"entropy": 1.0255846306681633,
|
| 1034 |
+
"epoch": 0.6248341232227488,
|
| 1035 |
+
"grad_norm": 0.03631991520524025,
|
| 1036 |
+
"learning_rate": 0.00016510658353079103,
|
| 1037 |
+
"loss": 0.2918674647808075,
|
| 1038 |
+
"mean_token_accuracy": 0.9063687976449728,
|
| 1039 |
+
"num_tokens": 8704036.0,
|
| 1040 |
+
"step": 206
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"entropy": 0.9965859483927488,
|
| 1044 |
+
"epoch": 0.6309004739336493,
|
| 1045 |
+
"grad_norm": 0.026180190965533257,
|
| 1046 |
+
"learning_rate": 0.00016434270982933273,
|
| 1047 |
+
"loss": 0.3719344735145569,
|
| 1048 |
+
"mean_token_accuracy": 0.8785704597830772,
|
| 1049 |
+
"num_tokens": 8796466.0,
|
| 1050 |
+
"step": 208
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"entropy": 0.9797388929873705,
|
| 1054 |
+
"epoch": 0.6369668246445498,
|
| 1055 |
+
"grad_norm": 0.020046714693307877,
|
| 1056 |
+
"learning_rate": 0.00016357237482099684,
|
| 1057 |
+
"loss": 0.20497792959213257,
|
| 1058 |
+
"mean_token_accuracy": 0.9296829011291265,
|
| 1059 |
+
"num_tokens": 8883021.0,
|
| 1060 |
+
"step": 210
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"entropy": 1.0285145360976458,
|
| 1064 |
+
"epoch": 0.6430331753554502,
|
| 1065 |
+
"grad_norm": 0.022925999015569687,
|
| 1066 |
+
"learning_rate": 0.000162795655862968,
|
| 1067 |
+
"loss": 0.32777711749076843,
|
| 1068 |
+
"mean_token_accuracy": 0.8869332894682884,
|
| 1069 |
+
"num_tokens": 8965779.0,
|
| 1070 |
+
"step": 212
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"entropy": 1.0544075202196836,
|
| 1074 |
+
"epoch": 0.6490995260663507,
|
| 1075 |
+
"grad_norm": 0.027822060510516167,
|
| 1076 |
+
"learning_rate": 0.00016201263095350833,
|
| 1077 |
+
"loss": 0.42936456203460693,
|
| 1078 |
+
"mean_token_accuracy": 0.866716792806983,
|
| 1079 |
+
"num_tokens": 9044268.0,
|
| 1080 |
+
"step": 214
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"entropy": 1.0603154003620148,
|
| 1084 |
+
"epoch": 0.6551658767772511,
|
| 1085 |
+
"grad_norm": 0.027222227305173874,
|
| 1086 |
+
"learning_rate": 0.0001612233787241248,
|
| 1087 |
+
"loss": 0.30441197752952576,
|
| 1088 |
+
"mean_token_accuracy": 0.9163662530481815,
|
| 1089 |
+
"num_tokens": 9126333.0,
|
| 1090 |
+
"step": 216
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"entropy": 1.0139030050486326,
|
| 1094 |
+
"epoch": 0.6612322274881517,
|
| 1095 |
+
"grad_norm": 0.028546614572405815,
|
| 1096 |
+
"learning_rate": 0.0001604279784316729,
|
| 1097 |
+
"loss": 0.3144024908542633,
|
| 1098 |
+
"mean_token_accuracy": 0.8966110236942768,
|
| 1099 |
+
"num_tokens": 9212853.0,
|
| 1100 |
+
"step": 218
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"entropy": 1.0053436905145645,
|
| 1104 |
+
"epoch": 0.6672985781990521,
|
| 1105 |
+
"grad_norm": 0.017751583829522133,
|
| 1106 |
+
"learning_rate": 0.00015962650995039783,
|
| 1107 |
+
"loss": 0.27218693494796753,
|
| 1108 |
+
"mean_token_accuracy": 0.9059509225189686,
|
| 1109 |
+
"num_tokens": 9302183.0,
|
| 1110 |
+
"step": 220
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"entropy": 1.0631114523857832,
|
| 1114 |
+
"epoch": 0.6733649289099526,
|
| 1115 |
+
"grad_norm": 0.027248896658420563,
|
| 1116 |
+
"learning_rate": 0.00015881905376391336,
|
| 1117 |
+
"loss": 0.34936845302581787,
|
| 1118 |
+
"mean_token_accuracy": 0.9041569344699383,
|
| 1119 |
+
"num_tokens": 9378720.0,
|
| 1120 |
+
"step": 222
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"entropy": 0.9875895343720913,
|
| 1124 |
+
"epoch": 0.679431279620853,
|
| 1125 |
+
"grad_norm": 0.02259000390768051,
|
| 1126 |
+
"learning_rate": 0.00015800569095711982,
|
| 1127 |
+
"loss": 0.23766493797302246,
|
| 1128 |
+
"mean_token_accuracy": 0.9157550428062677,
|
| 1129 |
+
"num_tokens": 9466723.0,
|
| 1130 |
+
"step": 224
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"entropy": 1.031277822330594,
|
| 1134 |
+
"epoch": 0.6854976303317536,
|
| 1135 |
+
"grad_norm": 0.027090661227703094,
|
| 1136 |
+
"learning_rate": 0.00015718650320806142,
|
| 1137 |
+
"loss": 0.3199611008167267,
|
| 1138 |
+
"mean_token_accuracy": 0.9022912103682756,
|
| 1139 |
+
"num_tokens": 9549836.0,
|
| 1140 |
+
"step": 226
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"entropy": 0.9979973528534174,
|
| 1144 |
+
"epoch": 0.691563981042654,
|
| 1145 |
+
"grad_norm": 0.021849313750863075,
|
| 1146 |
+
"learning_rate": 0.00015636157277972414,
|
| 1147 |
+
"loss": 0.30752354860305786,
|
| 1148 |
+
"mean_token_accuracy": 0.9131694324314594,
|
| 1149 |
+
"num_tokens": 9635184.0,
|
| 1150 |
+
"step": 228
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"entropy": 1.0358617044985294,
|
| 1154 |
+
"epoch": 0.6976303317535545,
|
| 1155 |
+
"grad_norm": 0.02150997892022133,
|
| 1156 |
+
"learning_rate": 0.00015553098251177487,
|
| 1157 |
+
"loss": 0.342559814453125,
|
| 1158 |
+
"mean_token_accuracy": 0.8880700599402189,
|
| 1159 |
+
"num_tokens": 9719880.0,
|
| 1160 |
+
"step": 230
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"entropy": 1.0295154266059399,
|
| 1164 |
+
"epoch": 0.7036966824644549,
|
| 1165 |
+
"grad_norm": 0.026608098298311234,
|
| 1166 |
+
"learning_rate": 0.00015469481581224272,
|
| 1167 |
+
"loss": 0.39191675186157227,
|
| 1168 |
+
"mean_token_accuracy": 0.8837976977229118,
|
| 1169 |
+
"num_tokens": 9798693.0,
|
| 1170 |
+
"step": 232
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"entropy": 1.065580291673541,
|
| 1174 |
+
"epoch": 0.7097630331753555,
|
| 1175 |
+
"grad_norm": 0.024026455357670784,
|
| 1176 |
+
"learning_rate": 0.00015385315664914292,
|
| 1177 |
+
"loss": 0.2597208619117737,
|
| 1178 |
+
"mean_token_accuracy": 0.9143438749015331,
|
| 1179 |
+
"num_tokens": 9877375.0,
|
| 1180 |
+
"step": 234
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"entropy": 1.04672497138381,
|
| 1184 |
+
"epoch": 0.7158293838862559,
|
| 1185 |
+
"grad_norm": 0.022771285846829414,
|
| 1186 |
+
"learning_rate": 0.00015300608954204516,
|
| 1187 |
+
"loss": 0.31437191367149353,
|
| 1188 |
+
"mean_token_accuracy": 0.8901943564414978,
|
| 1189 |
+
"num_tokens": 9958267.0,
|
| 1190 |
+
"step": 236
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"entropy": 1.0483401622623205,
|
| 1194 |
+
"epoch": 0.7218957345971564,
|
| 1195 |
+
"grad_norm": 0.02012186124920845,
|
| 1196 |
+
"learning_rate": 0.00015215369955358566,
|
| 1197 |
+
"loss": 0.3091947138309479,
|
| 1198 |
+
"mean_token_accuracy": 0.8791900016367435,
|
| 1199 |
+
"num_tokens": 10040225.0,
|
| 1200 |
+
"step": 238
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"entropy": 1.009421955794096,
|
| 1204 |
+
"epoch": 0.7279620853080568,
|
| 1205 |
+
"grad_norm": 0.025728857144713402,
|
| 1206 |
+
"learning_rate": 0.0001512960722809255,
|
| 1207 |
+
"loss": 0.34165671467781067,
|
| 1208 |
+
"mean_token_accuracy": 0.8932214323431253,
|
| 1209 |
+
"num_tokens": 10132456.0,
|
| 1210 |
+
"step": 240
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"entropy": 1.0154480524361134,
|
| 1214 |
+
"epoch": 0.7340284360189574,
|
| 1215 |
+
"grad_norm": 0.033449891954660416,
|
| 1216 |
+
"learning_rate": 0.00015043329384715475,
|
| 1217 |
+
"loss": 0.42814844846725464,
|
| 1218 |
+
"mean_token_accuracy": 0.8720662202686071,
|
| 1219 |
+
"num_tokens": 10222309.0,
|
| 1220 |
+
"step": 242
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"entropy": 1.041564915329218,
|
| 1224 |
+
"epoch": 0.7400947867298578,
|
| 1225 |
+
"grad_norm": 0.03188253566622734,
|
| 1226 |
+
"learning_rate": 0.00014956545089264407,
|
| 1227 |
+
"loss": 0.38293400406837463,
|
| 1228 |
+
"mean_token_accuracy": 0.8816456682980061,
|
| 1229 |
+
"num_tokens": 10306169.0,
|
| 1230 |
+
"step": 244
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"entropy": 1.052070127800107,
|
| 1234 |
+
"epoch": 0.7461611374407583,
|
| 1235 |
+
"grad_norm": 0.021951910108327866,
|
| 1236 |
+
"learning_rate": 0.00014869263056634417,
|
| 1237 |
+
"loss": 0.35783806443214417,
|
| 1238 |
+
"mean_token_accuracy": 0.889878798276186,
|
| 1239 |
+
"num_tokens": 10392845.0,
|
| 1240 |
+
"step": 246
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"entropy": 1.0201445445418358,
|
| 1244 |
+
"epoch": 0.7522274881516587,
|
| 1245 |
+
"grad_norm": 0.025892196223139763,
|
| 1246 |
+
"learning_rate": 0.0001478149205170345,
|
| 1247 |
+
"loss": 0.351155549287796,
|
| 1248 |
+
"mean_token_accuracy": 0.889661330729723,
|
| 1249 |
+
"num_tokens": 10482376.0,
|
| 1250 |
+
"step": 248
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"entropy": 1.0708779022097588,
|
| 1254 |
+
"epoch": 0.7582938388625592,
|
| 1255 |
+
"grad_norm": 0.028000684455037117,
|
| 1256 |
+
"learning_rate": 0.0001469324088845212,
|
| 1257 |
+
"loss": 0.3510592579841614,
|
| 1258 |
+
"mean_token_accuracy": 0.8943411149084568,
|
| 1259 |
+
"num_tokens": 10561626.0,
|
| 1260 |
+
"step": 250
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"entropy": 1.0424304846674204,
|
| 1264 |
+
"epoch": 0.7643601895734597,
|
| 1265 |
+
"grad_norm": 0.02899564430117607,
|
| 1266 |
+
"learning_rate": 0.00014604518429078652,
|
| 1267 |
+
"loss": 0.42678025364875793,
|
| 1268 |
+
"mean_token_accuracy": 0.8730289321392775,
|
| 1269 |
+
"num_tokens": 10648330.0,
|
| 1270 |
+
"step": 252
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"entropy": 1.0204415209591389,
|
| 1274 |
+
"epoch": 0.7704265402843602,
|
| 1275 |
+
"grad_norm": 0.024027369916439056,
|
| 1276 |
+
"learning_rate": 0.00014515333583108896,
|
| 1277 |
+
"loss": 0.31847134232521057,
|
| 1278 |
+
"mean_token_accuracy": 0.9021575674414635,
|
| 1279 |
+
"num_tokens": 10733363.0,
|
| 1280 |
+
"step": 254
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"entropy": 1.0702068824321032,
|
| 1284 |
+
"epoch": 0.7764928909952606,
|
| 1285 |
+
"grad_norm": 0.025766585022211075,
|
| 1286 |
+
"learning_rate": 0.00014425695306501658,
|
| 1287 |
+
"loss": 0.3848719000816345,
|
| 1288 |
+
"mean_token_accuracy": 0.8851827494800091,
|
| 1289 |
+
"num_tokens": 10809297.0,
|
| 1290 |
+
"step": 256
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"entropy": 1.0529421418905258,
|
| 1294 |
+
"epoch": 0.7825592417061611,
|
| 1295 |
+
"grad_norm": 0.026501361280679703,
|
| 1296 |
+
"learning_rate": 0.0001433561260074933,
|
| 1297 |
+
"loss": 0.35802873969078064,
|
| 1298 |
+
"mean_token_accuracy": 0.8852917850017548,
|
| 1299 |
+
"num_tokens": 10883618.0,
|
| 1300 |
+
"step": 258
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"entropy": 0.9485348463058472,
|
| 1304 |
+
"epoch": 0.7886255924170616,
|
| 1305 |
+
"grad_norm": 0.01836264319717884,
|
| 1306 |
+
"learning_rate": 0.00014245094511973968,
|
| 1307 |
+
"loss": 0.3131236135959625,
|
| 1308 |
+
"mean_token_accuracy": 0.8956947717815638,
|
| 1309 |
+
"num_tokens": 10980093.0,
|
| 1310 |
+
"step": 260
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"entropy": 0.9579282719641924,
|
| 1314 |
+
"epoch": 0.7946919431279621,
|
| 1315 |
+
"grad_norm": 0.022198352962732315,
|
| 1316 |
+
"learning_rate": 0.00014154150130018866,
|
| 1317 |
+
"loss": 0.37757307291030884,
|
| 1318 |
+
"mean_token_accuracy": 0.8951197322458029,
|
| 1319 |
+
"num_tokens": 11069570.0,
|
| 1320 |
+
"step": 262
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"entropy": 1.0135638751089573,
|
| 1324 |
+
"epoch": 0.8007582938388625,
|
| 1325 |
+
"grad_norm": 0.027133887633681297,
|
| 1326 |
+
"learning_rate": 0.00014062788587535757,
|
| 1327 |
+
"loss": 0.3821750283241272,
|
| 1328 |
+
"mean_token_accuracy": 0.8956535998731852,
|
| 1329 |
+
"num_tokens": 11149280.0,
|
| 1330 |
+
"step": 264
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"entropy": 1.0239354390650988,
|
| 1334 |
+
"epoch": 0.806824644549763,
|
| 1335 |
+
"grad_norm": 0.025353292003273964,
|
| 1336 |
+
"learning_rate": 0.0001397101905906772,
|
| 1337 |
+
"loss": 0.42222288250923157,
|
| 1338 |
+
"mean_token_accuracy": 0.8735412396490574,
|
| 1339 |
+
"num_tokens": 11227972.0,
|
| 1340 |
+
"step": 266
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"entropy": 1.0463627055287361,
|
| 1344 |
+
"epoch": 0.8128909952606636,
|
| 1345 |
+
"grad_norm": 0.02467101253569126,
|
| 1346 |
+
"learning_rate": 0.0001387885076012785,
|
| 1347 |
+
"loss": 0.41264429688453674,
|
| 1348 |
+
"mean_token_accuracy": 0.8757933098822832,
|
| 1349 |
+
"num_tokens": 11307711.0,
|
| 1350 |
+
"step": 268
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"entropy": 1.0219852272421122,
|
| 1354 |
+
"epoch": 0.818957345971564,
|
| 1355 |
+
"grad_norm": 0.026180898770689964,
|
| 1356 |
+
"learning_rate": 0.0001378629294627386,
|
| 1357 |
+
"loss": 0.3495751619338989,
|
| 1358 |
+
"mean_token_accuracy": 0.8855882193893194,
|
| 1359 |
+
"num_tokens": 11387909.0,
|
| 1360 |
+
"step": 270
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"entropy": 1.0157007835805416,
|
| 1364 |
+
"epoch": 0.8250236966824644,
|
| 1365 |
+
"grad_norm": 0.022954991087317467,
|
| 1366 |
+
"learning_rate": 0.0001369335491217861,
|
| 1367 |
+
"loss": 0.2733880877494812,
|
| 1368 |
+
"mean_token_accuracy": 0.9124114271253347,
|
| 1369 |
+
"num_tokens": 11468691.0,
|
| 1370 |
+
"step": 272
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"entropy": 1.0193607807159424,
|
| 1374 |
+
"epoch": 0.8310900473933649,
|
| 1375 |
+
"grad_norm": 0.022030817344784737,
|
| 1376 |
+
"learning_rate": 0.00013600045990696762,
|
| 1377 |
+
"loss": 0.32574281096458435,
|
| 1378 |
+
"mean_token_accuracy": 0.9108263738453388,
|
| 1379 |
+
"num_tokens": 11549322.0,
|
| 1380 |
+
"step": 274
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"entropy": 0.9909802060574293,
|
| 1384 |
+
"epoch": 0.8371563981042655,
|
| 1385 |
+
"grad_norm": 0.0237803366035223,
|
| 1386 |
+
"learning_rate": 0.00013506375551927547,
|
| 1387 |
+
"loss": 0.3146073818206787,
|
| 1388 |
+
"mean_token_accuracy": 0.8914975486695766,
|
| 1389 |
+
"num_tokens": 11639046.0,
|
| 1390 |
+
"step": 276
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"entropy": 0.9957366213202477,
|
| 1394 |
+
"epoch": 0.8432227488151659,
|
| 1395 |
+
"grad_norm": 0.020469270646572113,
|
| 1396 |
+
"learning_rate": 0.00013412353002273828,
|
| 1397 |
+
"loss": 0.2796984910964966,
|
| 1398 |
+
"mean_token_accuracy": 0.8986608982086182,
|
| 1399 |
+
"num_tokens": 11725997.0,
|
| 1400 |
+
"step": 278
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"entropy": 1.015178693458438,
|
| 1404 |
+
"epoch": 0.8492890995260663,
|
| 1405 |
+
"grad_norm": 0.0252392515540123,
|
| 1406 |
+
"learning_rate": 0.0001331798778349752,
|
| 1407 |
+
"loss": 0.3755182921886444,
|
| 1408 |
+
"mean_token_accuracy": 0.8900849334895611,
|
| 1409 |
+
"num_tokens": 11803811.0,
|
| 1410 |
+
"step": 280
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"entropy": 1.0187121462076902,
|
| 1414 |
+
"epoch": 0.8553554502369668,
|
| 1415 |
+
"grad_norm": 0.025252049788832664,
|
| 1416 |
+
"learning_rate": 0.00013223289371771426,
|
| 1417 |
+
"loss": 0.4176623225212097,
|
| 1418 |
+
"mean_token_accuracy": 0.8798485025763512,
|
| 1419 |
+
"num_tokens": 11885708.0,
|
| 1420 |
+
"step": 282
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"entropy": 1.0211249105632305,
|
| 1424 |
+
"epoch": 0.8614218009478672,
|
| 1425 |
+
"grad_norm": 0.02209135890007019,
|
| 1426 |
+
"learning_rate": 0.00013128267276727647,
|
| 1427 |
+
"loss": 0.3581184148788452,
|
| 1428 |
+
"mean_token_accuracy": 0.8814307861030102,
|
| 1429 |
+
"num_tokens": 11972596.0,
|
| 1430 |
+
"step": 284
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"entropy": 1.0198405291885138,
|
| 1434 |
+
"epoch": 0.8674881516587678,
|
| 1435 |
+
"grad_norm": 0.01992698200047016,
|
| 1436 |
+
"learning_rate": 0.00013032931040502627,
|
| 1437 |
+
"loss": 0.26552486419677734,
|
| 1438 |
+
"mean_token_accuracy": 0.896810881793499,
|
| 1439 |
+
"num_tokens": 12054259.0,
|
| 1440 |
+
"step": 286
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"entropy": 0.9992888607084751,
|
| 1444 |
+
"epoch": 0.8735545023696683,
|
| 1445 |
+
"grad_norm": 0.024810057133436203,
|
| 1446 |
+
"learning_rate": 0.0001293729023677892,
|
| 1447 |
+
"loss": 0.35126397013664246,
|
| 1448 |
+
"mean_token_accuracy": 0.8888840395957232,
|
| 1449 |
+
"num_tokens": 12139627.0,
|
| 1450 |
+
"step": 288
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"entropy": 1.039975594729185,
|
| 1454 |
+
"epoch": 0.8796208530805687,
|
| 1455 |
+
"grad_norm": 0.028860824182629585,
|
| 1456 |
+
"learning_rate": 0.00012841354469823814,
|
| 1457 |
+
"loss": 0.32350558042526245,
|
| 1458 |
+
"mean_token_accuracy": 0.9048479869961739,
|
| 1459 |
+
"num_tokens": 12213661.0,
|
| 1460 |
+
"step": 290
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"entropy": 0.9948016479611397,
|
| 1464 |
+
"epoch": 0.8856872037914691,
|
| 1465 |
+
"grad_norm": 0.01813443750143051,
|
| 1466 |
+
"learning_rate": 0.00012745133373524853,
|
| 1467 |
+
"loss": 0.3351580500602722,
|
| 1468 |
+
"mean_token_accuracy": 0.9004221297800541,
|
| 1469 |
+
"num_tokens": 12307248.0,
|
| 1470 |
+
"step": 292
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"entropy": 0.9768823459744453,
|
| 1474 |
+
"epoch": 0.8917535545023697,
|
| 1475 |
+
"grad_norm": 0.023104378953576088,
|
| 1476 |
+
"learning_rate": 0.000126486366104224,
|
| 1477 |
+
"loss": 0.31333544850349426,
|
| 1478 |
+
"mean_token_accuracy": 0.8828854896128178,
|
| 1479 |
+
"num_tokens": 12402310.0,
|
| 1480 |
+
"step": 294
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"entropy": 1.0372453946620226,
|
| 1484 |
+
"epoch": 0.8978199052132702,
|
| 1485 |
+
"grad_norm": 0.020197933539748192,
|
| 1486 |
+
"learning_rate": 0.0001255187387073935,
|
| 1487 |
+
"loss": 0.31656911969184875,
|
| 1488 |
+
"mean_token_accuracy": 0.9050396885722876,
|
| 1489 |
+
"num_tokens": 12485908.0,
|
| 1490 |
+
"step": 296
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"entropy": 0.9689353574067354,
|
| 1494 |
+
"epoch": 0.9038862559241706,
|
| 1495 |
+
"grad_norm": 0.029713748022913933,
|
| 1496 |
+
"learning_rate": 0.00012454854871407994,
|
| 1497 |
+
"loss": 0.3086496889591217,
|
| 1498 |
+
"mean_token_accuracy": 0.9022549409419298,
|
| 1499 |
+
"num_tokens": 12581306.0,
|
| 1500 |
+
"step": 298
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"entropy": 1.0563449021428823,
|
| 1504 |
+
"epoch": 0.909952606635071,
|
| 1505 |
+
"grad_norm": 0.022666629403829575,
|
| 1506 |
+
"learning_rate": 0.00012357589355094275,
|
| 1507 |
+
"loss": 0.3841848373413086,
|
| 1508 |
+
"mean_token_accuracy": 0.8740956950932741,
|
| 1509 |
+
"num_tokens": 12661727.0,
|
| 1510 |
+
"step": 300
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"entropy": 0.9897515531629324,
|
| 1514 |
+
"epoch": 0.9160189573459716,
|
| 1515 |
+
"grad_norm": 0.014572671614587307,
|
| 1516 |
+
"learning_rate": 0.00012260087089219413,
|
| 1517 |
+
"loss": 0.3274422585964203,
|
| 1518 |
+
"mean_token_accuracy": 0.8824338726699352,
|
| 1519 |
+
"num_tokens": 12748980.0,
|
| 1520 |
+
"step": 302
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"entropy": 0.9907026756554842,
|
| 1524 |
+
"epoch": 0.922085308056872,
|
| 1525 |
+
"grad_norm": 0.023578399792313576,
|
| 1526 |
+
"learning_rate": 0.00012162357864979072,
|
| 1527 |
+
"loss": 0.3978373110294342,
|
| 1528 |
+
"mean_token_accuracy": 0.8681950401514769,
|
| 1529 |
+
"num_tokens": 12835510.0,
|
| 1530 |
+
"step": 304
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"entropy": 1.001186614856124,
|
| 1534 |
+
"epoch": 0.9281516587677725,
|
| 1535 |
+
"grad_norm": 0.01945630833506584,
|
| 1536 |
+
"learning_rate": 0.00012064411496360109,
|
| 1537 |
+
"loss": 0.2402154505252838,
|
| 1538 |
+
"mean_token_accuracy": 0.9270794112235308,
|
| 1539 |
+
"num_tokens": 12916785.0,
|
| 1540 |
+
"step": 306
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"entropy": 1.0183254703879356,
|
| 1544 |
+
"epoch": 0.934218009478673,
|
| 1545 |
+
"grad_norm": 0.023567359894514084,
|
| 1546 |
+
"learning_rate": 0.00011966257819155062,
|
| 1547 |
+
"loss": 0.31015732884407043,
|
| 1548 |
+
"mean_token_accuracy": 0.9106874484568834,
|
| 1549 |
+
"num_tokens": 12994002.0,
|
| 1550 |
+
"step": 308
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"entropy": 1.0168362595140934,
|
| 1554 |
+
"epoch": 0.9402843601895735,
|
| 1555 |
+
"grad_norm": 0.02697554975748062,
|
| 1556 |
+
"learning_rate": 0.00011867906689974428,
|
| 1557 |
+
"loss": 0.32972174882888794,
|
| 1558 |
+
"mean_token_accuracy": 0.8983877934515476,
|
| 1559 |
+
"num_tokens": 13076689.0,
|
| 1560 |
+
"step": 310
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"entropy": 0.994679432362318,
|
| 1564 |
+
"epoch": 0.946350710900474,
|
| 1565 |
+
"grad_norm": 0.027006421238183975,
|
| 1566 |
+
"learning_rate": 0.00011769367985256887,
|
| 1567 |
+
"loss": 0.3707253038883209,
|
| 1568 |
+
"mean_token_accuracy": 0.8925164118409157,
|
| 1569 |
+
"num_tokens": 13161768.0,
|
| 1570 |
+
"step": 312
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"entropy": 0.9879488591104746,
|
| 1574 |
+
"epoch": 0.9524170616113744,
|
| 1575 |
+
"grad_norm": 0.02476397529244423,
|
| 1576 |
+
"learning_rate": 0.00011670651600277472,
|
| 1577 |
+
"loss": 0.40560826659202576,
|
| 1578 |
+
"mean_token_accuracy": 0.878234988078475,
|
| 1579 |
+
"num_tokens": 13252577.0,
|
| 1580 |
+
"step": 314
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"entropy": 1.0281283929944038,
|
| 1584 |
+
"epoch": 0.9584834123222749,
|
| 1585 |
+
"grad_norm": 0.021510828286409378,
|
| 1586 |
+
"learning_rate": 0.00011571767448153901,
|
| 1587 |
+
"loss": 0.3084154725074768,
|
| 1588 |
+
"mean_token_accuracy": 0.9019981864839792,
|
| 1589 |
+
"num_tokens": 13329272.0,
|
| 1590 |
+
"step": 316
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"entropy": 1.0063051134347916,
|
| 1594 |
+
"epoch": 0.9645497630331753,
|
| 1595 |
+
"grad_norm": 0.025342386215925217,
|
| 1596 |
+
"learning_rate": 0.00011472725458851116,
|
| 1597 |
+
"loss": 0.33964166045188904,
|
| 1598 |
+
"mean_token_accuracy": 0.8958082254976034,
|
| 1599 |
+
"num_tokens": 13416566.0,
|
| 1600 |
+
"step": 318
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"entropy": 0.9724898487329483,
|
| 1604 |
+
"epoch": 0.9706161137440759,
|
| 1605 |
+
"grad_norm": 0.02228512242436409,
|
| 1606 |
+
"learning_rate": 0.00011373535578184082,
|
| 1607 |
+
"loss": 0.3146652281284332,
|
| 1608 |
+
"mean_token_accuracy": 0.8933350183069706,
|
| 1609 |
+
"num_tokens": 13502738.0,
|
| 1610 |
+
"step": 320
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"entropy": 0.9833645019680262,
|
| 1614 |
+
"epoch": 0.9766824644549763,
|
| 1615 |
+
"grad_norm": 0.028228461742401123,
|
| 1616 |
+
"learning_rate": 0.0001127420776681905,
|
| 1617 |
+
"loss": 0.42645928263664246,
|
| 1618 |
+
"mean_token_accuracy": 0.8687106799334288,
|
| 1619 |
+
"num_tokens": 13592432.0,
|
| 1620 |
+
"step": 322
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"entropy": 1.008123904466629,
|
| 1624 |
+
"epoch": 0.9827488151658768,
|
| 1625 |
+
"grad_norm": 0.02406039461493492,
|
| 1626 |
+
"learning_rate": 0.0001117475199927329,
|
| 1627 |
+
"loss": 0.34203004837036133,
|
| 1628 |
+
"mean_token_accuracy": 0.8984533343464136,
|
| 1629 |
+
"num_tokens": 13672857.0,
|
| 1630 |
+
"step": 324
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"entropy": 1.0233813561499119,
|
| 1634 |
+
"epoch": 0.9888151658767772,
|
| 1635 |
+
"grad_norm": 0.019793907180428505,
|
| 1636 |
+
"learning_rate": 0.00011075178262913466,
|
| 1637 |
+
"loss": 0.2996736764907837,
|
| 1638 |
+
"mean_token_accuracy": 0.9128099251538515,
|
| 1639 |
+
"num_tokens": 13750070.0,
|
| 1640 |
+
"step": 326
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"entropy": 1.034196274355054,
|
| 1644 |
+
"epoch": 0.9948815165876778,
|
| 1645 |
+
"grad_norm": 0.019547009840607643,
|
| 1646 |
+
"learning_rate": 0.00010975496556952682,
|
| 1647 |
+
"loss": 0.2335723489522934,
|
| 1648 |
+
"mean_token_accuracy": 0.9135784525424242,
|
| 1649 |
+
"num_tokens": 13826736.0,
|
| 1650 |
+
"step": 328
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"entropy": 1.0028293905434784,
|
| 1654 |
+
"epoch": 1.0,
|
| 1655 |
+
"grad_norm": 0.036484770476818085,
|
| 1656 |
+
"learning_rate": 0.00010875716891446381,
|
| 1657 |
+
"loss": 0.29466891288757324,
|
| 1658 |
+
"mean_token_accuracy": 0.8920988021073518,
|
| 1659 |
+
"num_tokens": 13902842.0,
|
| 1660 |
+
"step": 330
|
| 1661 |
+
}
|
| 1662 |
+
],
|
| 1663 |
+
"logging_steps": 2,
|
| 1664 |
+
"max_steps": 660,
|
| 1665 |
+
"num_input_tokens_seen": 0,
|
| 1666 |
+
"num_train_epochs": 2,
|
| 1667 |
+
"save_steps": 500,
|
| 1668 |
+
"stateful_callbacks": {
|
| 1669 |
+
"TrainerControl": {
|
| 1670 |
+
"args": {
|
| 1671 |
+
"should_epoch_stop": false,
|
| 1672 |
+
"should_evaluate": false,
|
| 1673 |
+
"should_log": false,
|
| 1674 |
+
"should_save": true,
|
| 1675 |
+
"should_training_stop": false
|
| 1676 |
+
},
|
| 1677 |
+
"attributes": {}
|
| 1678 |
+
}
|
| 1679 |
+
},
|
| 1680 |
+
"total_flos": 6.674341781155983e+17,
|
| 1681 |
+
"train_batch_size": 1,
|
| 1682 |
+
"trial_name": null,
|
| 1683 |
+
"trial_params": null
|
| 1684 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a36a3d05537a890e2fc95995cafa86538b19318f2c774439e46c38b77595361
|
| 3 |
+
size 5649
|