Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- 400/added_tokens.json +16 -0
- 400/chat_template.jinja +7 -0
- 400/config.json +97 -0
- 400/generation_config.json +13 -0
- 400/merges.txt +0 -0
- 400/model.safetensors +3 -0
- 400/preprocessor_config.json +36 -0
- 400/special_tokens_map.json +31 -0
- 400/tokenizer.json +3 -0
- 400/tokenizer_config.json +144 -0
- 400/trainer_state.json +2834 -0
- 400/training_args.bin +3 -0
- 400/video_preprocessor_config.json +86 -0
- 400/vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* 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 |
200/tokenizer.json 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 |
200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
400/added_tokens.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|box_end|>": 151649,
|
| 3 |
+
"<|box_start|>": 151648,
|
| 4 |
+
"<|endoftext|>": 151643,
|
| 5 |
+
"<|im_end|>": 151645,
|
| 6 |
+
"<|im_start|>": 151644,
|
| 7 |
+
"<|image_pad|>": 151655,
|
| 8 |
+
"<|object_ref_end|>": 151647,
|
| 9 |
+
"<|object_ref_start|>": 151646,
|
| 10 |
+
"<|quad_end|>": 151651,
|
| 11 |
+
"<|quad_start|>": 151650,
|
| 12 |
+
"<|video_pad|>": 151656,
|
| 13 |
+
"<|vision_end|>": 151653,
|
| 14 |
+
"<|vision_pad|>": 151654,
|
| 15 |
+
"<|vision_start|>": 151652
|
| 16 |
+
}
|
400/chat_template.jinja
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system
|
| 2 |
+
You are a helpful assistant.<|im_end|>
|
| 3 |
+
{% endif %}<|im_start|>{{ message['role'] }}
|
| 4 |
+
{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
|
| 5 |
+
{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>
|
| 6 |
+
{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
|
| 7 |
+
{% endif %}
|
400/config.json
ADDED
|
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen2VLForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 151643,
|
| 7 |
+
"eos_token_id": 151645,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 1536,
|
| 10 |
+
"image_token_id": 151655,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 28,
|
| 15 |
+
"model_type": "qwen2_vl",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_scaling": {
|
| 21 |
+
"mrope_section": [
|
| 22 |
+
16,
|
| 23 |
+
24,
|
| 24 |
+
24
|
| 25 |
+
],
|
| 26 |
+
"rope_type": "default",
|
| 27 |
+
"type": "default"
|
| 28 |
+
},
|
| 29 |
+
"rope_theta": 1000000.0,
|
| 30 |
+
"sliding_window": 32768,
|
| 31 |
+
"text_config": {
|
| 32 |
+
"architectures": [
|
| 33 |
+
"Qwen2VLForConditionalGeneration"
|
| 34 |
+
],
|
| 35 |
+
"attention_dropout": 0.0,
|
| 36 |
+
"bos_token_id": 151643,
|
| 37 |
+
"eos_token_id": 151645,
|
| 38 |
+
"hidden_act": "silu",
|
| 39 |
+
"hidden_size": 1536,
|
| 40 |
+
"image_token_id": null,
|
| 41 |
+
"initializer_range": 0.02,
|
| 42 |
+
"intermediate_size": 8960,
|
| 43 |
+
"max_position_embeddings": 32768,
|
| 44 |
+
"max_window_layers": 28,
|
| 45 |
+
"model_type": "qwen2_vl_text",
|
| 46 |
+
"num_attention_heads": 12,
|
| 47 |
+
"num_hidden_layers": 28,
|
| 48 |
+
"num_key_value_heads": 2,
|
| 49 |
+
"rms_norm_eps": 1e-06,
|
| 50 |
+
"rope_scaling": {
|
| 51 |
+
"mrope_section": [
|
| 52 |
+
16,
|
| 53 |
+
24,
|
| 54 |
+
24
|
| 55 |
+
],
|
| 56 |
+
"rope_type": "default",
|
| 57 |
+
"type": "default"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 1000000.0,
|
| 60 |
+
"sliding_window": 32768,
|
| 61 |
+
"tie_word_embeddings": true,
|
| 62 |
+
"torch_dtype": "bfloat16",
|
| 63 |
+
"use_cache": false,
|
| 64 |
+
"use_sliding_window": false,
|
| 65 |
+
"video_token_id": null,
|
| 66 |
+
"vision_end_token_id": 151653,
|
| 67 |
+
"vision_start_token_id": 151652,
|
| 68 |
+
"vision_token_id": 151654,
|
| 69 |
+
"vocab_size": 151936
|
| 70 |
+
},
|
| 71 |
+
"torch_dtype": "bfloat16",
|
| 72 |
+
"transformers_version": "4.52.4",
|
| 73 |
+
"use_cache": false,
|
| 74 |
+
"use_sliding_window": false,
|
| 75 |
+
"video_token_id": 151656,
|
| 76 |
+
"vision_config": {
|
| 77 |
+
"depth": 32,
|
| 78 |
+
"embed_dim": 1280,
|
| 79 |
+
"hidden_act": "quick_gelu",
|
| 80 |
+
"hidden_size": 1536,
|
| 81 |
+
"in_channels": 3,
|
| 82 |
+
"in_chans": 3,
|
| 83 |
+
"initializer_range": 0.02,
|
| 84 |
+
"mlp_ratio": 4,
|
| 85 |
+
"model_type": "qwen2_vl",
|
| 86 |
+
"num_heads": 16,
|
| 87 |
+
"patch_size": 14,
|
| 88 |
+
"spatial_merge_size": 2,
|
| 89 |
+
"spatial_patch_size": 14,
|
| 90 |
+
"temporal_patch_size": 2,
|
| 91 |
+
"torch_dtype": "bfloat16"
|
| 92 |
+
},
|
| 93 |
+
"vision_end_token_id": 151653,
|
| 94 |
+
"vision_start_token_id": 151652,
|
| 95 |
+
"vision_token_id": 151654,
|
| 96 |
+
"vocab_size": 151936
|
| 97 |
+
}
|
400/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"temperature": 0.01,
|
| 10 |
+
"top_k": 1,
|
| 11 |
+
"top_p": 0.001,
|
| 12 |
+
"transformers_version": "4.52.4"
|
| 13 |
+
}
|
400/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
400/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94fed97cd1772a135b001e1de70f88c1b19c9f5870754587f159f81eae8ed704
|
| 3 |
+
size 4884798456
|
400/preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": null,
|
| 3 |
+
"data_format": "channels_first",
|
| 4 |
+
"default_to_square": true,
|
| 5 |
+
"device": null,
|
| 6 |
+
"do_center_crop": null,
|
| 7 |
+
"do_convert_rgb": true,
|
| 8 |
+
"do_normalize": true,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"image_mean": [
|
| 12 |
+
0.48145466,
|
| 13 |
+
0.4578275,
|
| 14 |
+
0.40821073
|
| 15 |
+
],
|
| 16 |
+
"image_processor_type": "Qwen2VLImageProcessorFast",
|
| 17 |
+
"image_std": [
|
| 18 |
+
0.26862954,
|
| 19 |
+
0.26130258,
|
| 20 |
+
0.27577711
|
| 21 |
+
],
|
| 22 |
+
"input_data_format": null,
|
| 23 |
+
"max_pixels": 12845056,
|
| 24 |
+
"merge_size": 2,
|
| 25 |
+
"min_pixels": 3136,
|
| 26 |
+
"patch_size": 14,
|
| 27 |
+
"processor_class": "Qwen2VLProcessor",
|
| 28 |
+
"resample": 3,
|
| 29 |
+
"rescale_factor": 0.00392156862745098,
|
| 30 |
+
"return_tensors": null,
|
| 31 |
+
"size": {
|
| 32 |
+
"longest_edge": 12845056,
|
| 33 |
+
"shortest_edge": 3136
|
| 34 |
+
},
|
| 35 |
+
"temporal_patch_size": 2
|
| 36 |
+
}
|
400/special_tokens_map.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|im_end|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": {
|
| 25 |
+
"content": "<|endoftext|>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
}
|
| 31 |
+
}
|
400/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
|
| 3 |
+
size 11420371
|
400/tokenizer_config.json
ADDED
|
@@ -0,0 +1,144 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"151646": {
|
| 29 |
+
"content": "<|object_ref_start|>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"151647": {
|
| 37 |
+
"content": "<|object_ref_end|>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"151648": {
|
| 45 |
+
"content": "<|box_start|>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"151649": {
|
| 53 |
+
"content": "<|box_end|>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"151650": {
|
| 61 |
+
"content": "<|quad_start|>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"151651": {
|
| 69 |
+
"content": "<|quad_end|>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"151652": {
|
| 77 |
+
"content": "<|vision_start|>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"151653": {
|
| 85 |
+
"content": "<|vision_end|>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"151654": {
|
| 93 |
+
"content": "<|vision_pad|>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"151655": {
|
| 101 |
+
"content": "<|image_pad|>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"151656": {
|
| 109 |
+
"content": "<|video_pad|>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
}
|
| 116 |
+
},
|
| 117 |
+
"additional_special_tokens": [
|
| 118 |
+
"<|im_start|>",
|
| 119 |
+
"<|im_end|>",
|
| 120 |
+
"<|object_ref_start|>",
|
| 121 |
+
"<|object_ref_end|>",
|
| 122 |
+
"<|box_start|>",
|
| 123 |
+
"<|box_end|>",
|
| 124 |
+
"<|quad_start|>",
|
| 125 |
+
"<|quad_end|>",
|
| 126 |
+
"<|vision_start|>",
|
| 127 |
+
"<|vision_end|>",
|
| 128 |
+
"<|vision_pad|>",
|
| 129 |
+
"<|image_pad|>",
|
| 130 |
+
"<|video_pad|>"
|
| 131 |
+
],
|
| 132 |
+
"bos_token": null,
|
| 133 |
+
"clean_up_tokenization_spaces": false,
|
| 134 |
+
"eos_token": "<|im_end|>",
|
| 135 |
+
"errors": "replace",
|
| 136 |
+
"extra_special_tokens": {},
|
| 137 |
+
"model_max_length": 32768,
|
| 138 |
+
"pad_token": "<|endoftext|>",
|
| 139 |
+
"padding_side": "right",
|
| 140 |
+
"processor_class": "Qwen2VLProcessor",
|
| 141 |
+
"split_special_tokens": false,
|
| 142 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 143 |
+
"unk_token": null
|
| 144 |
+
}
|
400/trainer_state.json
ADDED
|
@@ -0,0 +1,2834 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.704,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 400,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.004266666666666667,
|
| 14 |
+
"grad_norm": 22.56278419494629,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 1.003,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008533333333333334,
|
| 21 |
+
"grad_norm": 22.34969711303711,
|
| 22 |
+
"learning_rate": 4.2553191489361707e-08,
|
| 23 |
+
"loss": 1.0007,
|
| 24 |
+
"step": 2
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0128,
|
| 28 |
+
"grad_norm": 22.497220993041992,
|
| 29 |
+
"learning_rate": 8.510638297872341e-08,
|
| 30 |
+
"loss": 0.9989,
|
| 31 |
+
"step": 3
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.017066666666666667,
|
| 35 |
+
"grad_norm": 22.301002502441406,
|
| 36 |
+
"learning_rate": 1.276595744680851e-07,
|
| 37 |
+
"loss": 0.9972,
|
| 38 |
+
"step": 4
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.021333333333333333,
|
| 42 |
+
"grad_norm": 22.443401336669922,
|
| 43 |
+
"learning_rate": 1.7021276595744683e-07,
|
| 44 |
+
"loss": 0.9995,
|
| 45 |
+
"step": 5
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0256,
|
| 49 |
+
"grad_norm": 22.423526763916016,
|
| 50 |
+
"learning_rate": 2.1276595744680852e-07,
|
| 51 |
+
"loss": 0.9983,
|
| 52 |
+
"step": 6
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.029866666666666666,
|
| 56 |
+
"grad_norm": 22.28087043762207,
|
| 57 |
+
"learning_rate": 2.553191489361702e-07,
|
| 58 |
+
"loss": 0.9965,
|
| 59 |
+
"step": 7
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.034133333333333335,
|
| 63 |
+
"grad_norm": 22.648902893066406,
|
| 64 |
+
"learning_rate": 2.9787234042553196e-07,
|
| 65 |
+
"loss": 0.9996,
|
| 66 |
+
"step": 8
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0384,
|
| 70 |
+
"grad_norm": 21.830528259277344,
|
| 71 |
+
"learning_rate": 3.4042553191489365e-07,
|
| 72 |
+
"loss": 0.9895,
|
| 73 |
+
"step": 9
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.042666666666666665,
|
| 77 |
+
"grad_norm": 22.060640335083008,
|
| 78 |
+
"learning_rate": 3.8297872340425535e-07,
|
| 79 |
+
"loss": 0.9927,
|
| 80 |
+
"step": 10
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.046933333333333334,
|
| 84 |
+
"grad_norm": 20.831892013549805,
|
| 85 |
+
"learning_rate": 4.2553191489361704e-07,
|
| 86 |
+
"loss": 0.9694,
|
| 87 |
+
"step": 11
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0512,
|
| 91 |
+
"grad_norm": 20.609140396118164,
|
| 92 |
+
"learning_rate": 4.6808510638297873e-07,
|
| 93 |
+
"loss": 0.9665,
|
| 94 |
+
"step": 12
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.055466666666666664,
|
| 98 |
+
"grad_norm": 20.27930450439453,
|
| 99 |
+
"learning_rate": 5.106382978723404e-07,
|
| 100 |
+
"loss": 0.9665,
|
| 101 |
+
"step": 13
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.05973333333333333,
|
| 105 |
+
"grad_norm": 20.32429313659668,
|
| 106 |
+
"learning_rate": 5.531914893617021e-07,
|
| 107 |
+
"loss": 0.9611,
|
| 108 |
+
"step": 14
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.064,
|
| 112 |
+
"grad_norm": 16.961851119995117,
|
| 113 |
+
"learning_rate": 5.957446808510639e-07,
|
| 114 |
+
"loss": 0.8898,
|
| 115 |
+
"step": 15
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06826666666666667,
|
| 119 |
+
"grad_norm": 16.51335906982422,
|
| 120 |
+
"learning_rate": 6.382978723404255e-07,
|
| 121 |
+
"loss": 0.8859,
|
| 122 |
+
"step": 16
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.07253333333333334,
|
| 126 |
+
"grad_norm": 16.328777313232422,
|
| 127 |
+
"learning_rate": 6.808510638297873e-07,
|
| 128 |
+
"loss": 0.8805,
|
| 129 |
+
"step": 17
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0768,
|
| 133 |
+
"grad_norm": 15.942858695983887,
|
| 134 |
+
"learning_rate": 7.234042553191489e-07,
|
| 135 |
+
"loss": 0.8662,
|
| 136 |
+
"step": 18
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.08106666666666666,
|
| 140 |
+
"grad_norm": 15.772059440612793,
|
| 141 |
+
"learning_rate": 7.659574468085107e-07,
|
| 142 |
+
"loss": 0.8618,
|
| 143 |
+
"step": 19
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.08533333333333333,
|
| 147 |
+
"grad_norm": 15.53687858581543,
|
| 148 |
+
"learning_rate": 8.085106382978725e-07,
|
| 149 |
+
"loss": 0.8506,
|
| 150 |
+
"step": 20
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.0896,
|
| 154 |
+
"grad_norm": 10.289249420166016,
|
| 155 |
+
"learning_rate": 8.510638297872341e-07,
|
| 156 |
+
"loss": 0.6896,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.09386666666666667,
|
| 161 |
+
"grad_norm": 10.452590942382812,
|
| 162 |
+
"learning_rate": 8.936170212765959e-07,
|
| 163 |
+
"loss": 0.6914,
|
| 164 |
+
"step": 22
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.09813333333333334,
|
| 168 |
+
"grad_norm": 10.317420959472656,
|
| 169 |
+
"learning_rate": 9.361702127659575e-07,
|
| 170 |
+
"loss": 0.6787,
|
| 171 |
+
"step": 23
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.1024,
|
| 175 |
+
"grad_norm": 9.890551567077637,
|
| 176 |
+
"learning_rate": 9.787234042553193e-07,
|
| 177 |
+
"loss": 0.6696,
|
| 178 |
+
"step": 24
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.10666666666666667,
|
| 182 |
+
"grad_norm": 9.151979446411133,
|
| 183 |
+
"learning_rate": 1.0212765957446809e-06,
|
| 184 |
+
"loss": 0.6481,
|
| 185 |
+
"step": 25
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.11093333333333333,
|
| 189 |
+
"grad_norm": 8.83176040649414,
|
| 190 |
+
"learning_rate": 1.0638297872340427e-06,
|
| 191 |
+
"loss": 0.6352,
|
| 192 |
+
"step": 26
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.1152,
|
| 196 |
+
"grad_norm": 8.056096076965332,
|
| 197 |
+
"learning_rate": 1.1063829787234042e-06,
|
| 198 |
+
"loss": 0.6139,
|
| 199 |
+
"step": 27
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.11946666666666667,
|
| 203 |
+
"grad_norm": 8.634873390197754,
|
| 204 |
+
"learning_rate": 1.148936170212766e-06,
|
| 205 |
+
"loss": 0.5139,
|
| 206 |
+
"step": 28
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.12373333333333333,
|
| 210 |
+
"grad_norm": 6.985842704772949,
|
| 211 |
+
"learning_rate": 1.1914893617021278e-06,
|
| 212 |
+
"loss": 0.4903,
|
| 213 |
+
"step": 29
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.128,
|
| 217 |
+
"grad_norm": 6.07671594619751,
|
| 218 |
+
"learning_rate": 1.2340425531914894e-06,
|
| 219 |
+
"loss": 0.4741,
|
| 220 |
+
"step": 30
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.13226666666666667,
|
| 224 |
+
"grad_norm": 5.922327995300293,
|
| 225 |
+
"learning_rate": 1.276595744680851e-06,
|
| 226 |
+
"loss": 0.456,
|
| 227 |
+
"step": 31
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.13653333333333334,
|
| 231 |
+
"grad_norm": 5.909642696380615,
|
| 232 |
+
"learning_rate": 1.3191489361702128e-06,
|
| 233 |
+
"loss": 0.4431,
|
| 234 |
+
"step": 32
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1408,
|
| 238 |
+
"grad_norm": 5.159115314483643,
|
| 239 |
+
"learning_rate": 1.3617021276595746e-06,
|
| 240 |
+
"loss": 0.4265,
|
| 241 |
+
"step": 33
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.14506666666666668,
|
| 245 |
+
"grad_norm": 4.204401016235352,
|
| 246 |
+
"learning_rate": 1.4042553191489364e-06,
|
| 247 |
+
"loss": 0.408,
|
| 248 |
+
"step": 34
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.14933333333333335,
|
| 252 |
+
"grad_norm": 3.8235232830047607,
|
| 253 |
+
"learning_rate": 1.4468085106382978e-06,
|
| 254 |
+
"loss": 0.3929,
|
| 255 |
+
"step": 35
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.1536,
|
| 259 |
+
"grad_norm": 3.8501057624816895,
|
| 260 |
+
"learning_rate": 1.4893617021276596e-06,
|
| 261 |
+
"loss": 0.3837,
|
| 262 |
+
"step": 36
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.15786666666666666,
|
| 266 |
+
"grad_norm": 4.0237860679626465,
|
| 267 |
+
"learning_rate": 1.5319148936170214e-06,
|
| 268 |
+
"loss": 0.3512,
|
| 269 |
+
"step": 37
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.16213333333333332,
|
| 273 |
+
"grad_norm": 4.531589031219482,
|
| 274 |
+
"learning_rate": 1.5744680851063832e-06,
|
| 275 |
+
"loss": 0.3299,
|
| 276 |
+
"step": 38
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.1664,
|
| 280 |
+
"grad_norm": 4.304875373840332,
|
| 281 |
+
"learning_rate": 1.617021276595745e-06,
|
| 282 |
+
"loss": 0.3179,
|
| 283 |
+
"step": 39
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.17066666666666666,
|
| 287 |
+
"grad_norm": 3.1829235553741455,
|
| 288 |
+
"learning_rate": 1.6595744680851064e-06,
|
| 289 |
+
"loss": 0.3087,
|
| 290 |
+
"step": 40
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.17493333333333333,
|
| 294 |
+
"grad_norm": 2.890695810317993,
|
| 295 |
+
"learning_rate": 1.7021276595744682e-06,
|
| 296 |
+
"loss": 0.3009,
|
| 297 |
+
"step": 41
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.1792,
|
| 301 |
+
"grad_norm": 3.62577223777771,
|
| 302 |
+
"learning_rate": 1.74468085106383e-06,
|
| 303 |
+
"loss": 0.2926,
|
| 304 |
+
"step": 42
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.18346666666666667,
|
| 308 |
+
"grad_norm": 3.2570207118988037,
|
| 309 |
+
"learning_rate": 1.7872340425531918e-06,
|
| 310 |
+
"loss": 0.2857,
|
| 311 |
+
"step": 43
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.18773333333333334,
|
| 315 |
+
"grad_norm": 2.0917487144470215,
|
| 316 |
+
"learning_rate": 1.8297872340425531e-06,
|
| 317 |
+
"loss": 0.2734,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.192,
|
| 322 |
+
"grad_norm": 2.1430346965789795,
|
| 323 |
+
"learning_rate": 1.872340425531915e-06,
|
| 324 |
+
"loss": 0.2669,
|
| 325 |
+
"step": 45
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.19626666666666667,
|
| 329 |
+
"grad_norm": 2.666980266571045,
|
| 330 |
+
"learning_rate": 1.9148936170212767e-06,
|
| 331 |
+
"loss": 0.2582,
|
| 332 |
+
"step": 46
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.20053333333333334,
|
| 336 |
+
"grad_norm": 1.869988203048706,
|
| 337 |
+
"learning_rate": 1.9574468085106385e-06,
|
| 338 |
+
"loss": 0.2517,
|
| 339 |
+
"step": 47
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.2048,
|
| 343 |
+
"grad_norm": 2.069547653198242,
|
| 344 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 345 |
+
"loss": 0.2445,
|
| 346 |
+
"step": 48
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.20906666666666668,
|
| 350 |
+
"grad_norm": 2.0606069564819336,
|
| 351 |
+
"learning_rate": 2.0425531914893617e-06,
|
| 352 |
+
"loss": 0.238,
|
| 353 |
+
"step": 49
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.21333333333333335,
|
| 357 |
+
"grad_norm": 1.706727147102356,
|
| 358 |
+
"learning_rate": 2.0851063829787235e-06,
|
| 359 |
+
"loss": 0.2335,
|
| 360 |
+
"step": 50
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.2176,
|
| 364 |
+
"grad_norm": 2.278432846069336,
|
| 365 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 366 |
+
"loss": 0.2308,
|
| 367 |
+
"step": 51
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.22186666666666666,
|
| 371 |
+
"grad_norm": 1.8908629417419434,
|
| 372 |
+
"learning_rate": 2.170212765957447e-06,
|
| 373 |
+
"loss": 0.2262,
|
| 374 |
+
"step": 52
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.22613333333333333,
|
| 378 |
+
"grad_norm": 1.949669599533081,
|
| 379 |
+
"learning_rate": 2.2127659574468085e-06,
|
| 380 |
+
"loss": 0.219,
|
| 381 |
+
"step": 53
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.2304,
|
| 385 |
+
"grad_norm": 1.920652151107788,
|
| 386 |
+
"learning_rate": 2.2553191489361703e-06,
|
| 387 |
+
"loss": 0.2162,
|
| 388 |
+
"step": 54
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.23466666666666666,
|
| 392 |
+
"grad_norm": 2.028974771499634,
|
| 393 |
+
"learning_rate": 2.297872340425532e-06,
|
| 394 |
+
"loss": 0.2121,
|
| 395 |
+
"step": 55
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.23893333333333333,
|
| 399 |
+
"grad_norm": 1.6449410915374756,
|
| 400 |
+
"learning_rate": 2.340425531914894e-06,
|
| 401 |
+
"loss": 0.2078,
|
| 402 |
+
"step": 56
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2432,
|
| 406 |
+
"grad_norm": 1.5636619329452515,
|
| 407 |
+
"learning_rate": 2.3829787234042557e-06,
|
| 408 |
+
"loss": 0.2003,
|
| 409 |
+
"step": 57
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.24746666666666667,
|
| 413 |
+
"grad_norm": 1.9496514797210693,
|
| 414 |
+
"learning_rate": 2.425531914893617e-06,
|
| 415 |
+
"loss": 0.1962,
|
| 416 |
+
"step": 58
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2517333333333333,
|
| 420 |
+
"grad_norm": 1.6503078937530518,
|
| 421 |
+
"learning_rate": 2.468085106382979e-06,
|
| 422 |
+
"loss": 0.1907,
|
| 423 |
+
"step": 59
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.256,
|
| 427 |
+
"grad_norm": 1.2683771848678589,
|
| 428 |
+
"learning_rate": 2.5106382978723402e-06,
|
| 429 |
+
"loss": 0.1878,
|
| 430 |
+
"step": 60
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.26026666666666665,
|
| 434 |
+
"grad_norm": 2.003361701965332,
|
| 435 |
+
"learning_rate": 2.553191489361702e-06,
|
| 436 |
+
"loss": 0.1827,
|
| 437 |
+
"step": 61
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.26453333333333334,
|
| 441 |
+
"grad_norm": 1.4562724828720093,
|
| 442 |
+
"learning_rate": 2.595744680851064e-06,
|
| 443 |
+
"loss": 0.1817,
|
| 444 |
+
"step": 62
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.2688,
|
| 448 |
+
"grad_norm": 2.4795641899108887,
|
| 449 |
+
"learning_rate": 2.6382978723404256e-06,
|
| 450 |
+
"loss": 0.1787,
|
| 451 |
+
"step": 63
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.2730666666666667,
|
| 455 |
+
"grad_norm": 2.1848199367523193,
|
| 456 |
+
"learning_rate": 2.6808510638297874e-06,
|
| 457 |
+
"loss": 0.1759,
|
| 458 |
+
"step": 64
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.2773333333333333,
|
| 462 |
+
"grad_norm": 1.6162073612213135,
|
| 463 |
+
"learning_rate": 2.7234042553191492e-06,
|
| 464 |
+
"loss": 0.1725,
|
| 465 |
+
"step": 65
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.2816,
|
| 469 |
+
"grad_norm": 3.0534298419952393,
|
| 470 |
+
"learning_rate": 2.765957446808511e-06,
|
| 471 |
+
"loss": 0.169,
|
| 472 |
+
"step": 66
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.28586666666666666,
|
| 476 |
+
"grad_norm": 1.6378310918807983,
|
| 477 |
+
"learning_rate": 2.808510638297873e-06,
|
| 478 |
+
"loss": 0.1662,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.29013333333333335,
|
| 483 |
+
"grad_norm": 3.190389394760132,
|
| 484 |
+
"learning_rate": 2.8510638297872346e-06,
|
| 485 |
+
"loss": 0.1638,
|
| 486 |
+
"step": 68
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.2944,
|
| 490 |
+
"grad_norm": 2.222900152206421,
|
| 491 |
+
"learning_rate": 2.8936170212765956e-06,
|
| 492 |
+
"loss": 0.1643,
|
| 493 |
+
"step": 69
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.2986666666666667,
|
| 497 |
+
"grad_norm": 2.720613956451416,
|
| 498 |
+
"learning_rate": 2.9361702127659574e-06,
|
| 499 |
+
"loss": 0.16,
|
| 500 |
+
"step": 70
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.30293333333333333,
|
| 504 |
+
"grad_norm": 2.4706056118011475,
|
| 505 |
+
"learning_rate": 2.978723404255319e-06,
|
| 506 |
+
"loss": 0.1595,
|
| 507 |
+
"step": 71
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.3072,
|
| 511 |
+
"grad_norm": 1.7160310745239258,
|
| 512 |
+
"learning_rate": 3.021276595744681e-06,
|
| 513 |
+
"loss": 0.1579,
|
| 514 |
+
"step": 72
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.31146666666666667,
|
| 518 |
+
"grad_norm": 3.9026131629943848,
|
| 519 |
+
"learning_rate": 3.0638297872340428e-06,
|
| 520 |
+
"loss": 0.1548,
|
| 521 |
+
"step": 73
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.3157333333333333,
|
| 525 |
+
"grad_norm": 2.223243236541748,
|
| 526 |
+
"learning_rate": 3.1063829787234046e-06,
|
| 527 |
+
"loss": 0.1571,
|
| 528 |
+
"step": 74
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.32,
|
| 532 |
+
"grad_norm": 3.901400327682495,
|
| 533 |
+
"learning_rate": 3.1489361702127664e-06,
|
| 534 |
+
"loss": 0.1532,
|
| 535 |
+
"step": 75
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.32426666666666665,
|
| 539 |
+
"grad_norm": 3.2121057510375977,
|
| 540 |
+
"learning_rate": 3.191489361702128e-06,
|
| 541 |
+
"loss": 0.1535,
|
| 542 |
+
"step": 76
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.32853333333333334,
|
| 546 |
+
"grad_norm": 2.6652214527130127,
|
| 547 |
+
"learning_rate": 3.23404255319149e-06,
|
| 548 |
+
"loss": 0.1513,
|
| 549 |
+
"step": 77
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3328,
|
| 553 |
+
"grad_norm": 5.308091640472412,
|
| 554 |
+
"learning_rate": 3.276595744680851e-06,
|
| 555 |
+
"loss": 0.1501,
|
| 556 |
+
"step": 78
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3370666666666667,
|
| 560 |
+
"grad_norm": 4.369630813598633,
|
| 561 |
+
"learning_rate": 3.3191489361702127e-06,
|
| 562 |
+
"loss": 0.1505,
|
| 563 |
+
"step": 79
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.3413333333333333,
|
| 567 |
+
"grad_norm": 4.927572727203369,
|
| 568 |
+
"learning_rate": 3.3617021276595745e-06,
|
| 569 |
+
"loss": 0.1486,
|
| 570 |
+
"step": 80
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.3456,
|
| 574 |
+
"grad_norm": 3.150054931640625,
|
| 575 |
+
"learning_rate": 3.4042553191489363e-06,
|
| 576 |
+
"loss": 0.1488,
|
| 577 |
+
"step": 81
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.34986666666666666,
|
| 581 |
+
"grad_norm": 6.687278747558594,
|
| 582 |
+
"learning_rate": 3.446808510638298e-06,
|
| 583 |
+
"loss": 0.1467,
|
| 584 |
+
"step": 82
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.35413333333333336,
|
| 588 |
+
"grad_norm": 5.45517635345459,
|
| 589 |
+
"learning_rate": 3.48936170212766e-06,
|
| 590 |
+
"loss": 0.1468,
|
| 591 |
+
"step": 83
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3584,
|
| 595 |
+
"grad_norm": 5.900143623352051,
|
| 596 |
+
"learning_rate": 3.5319148936170217e-06,
|
| 597 |
+
"loss": 0.1457,
|
| 598 |
+
"step": 84
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3626666666666667,
|
| 602 |
+
"grad_norm": 6.59785795211792,
|
| 603 |
+
"learning_rate": 3.5744680851063835e-06,
|
| 604 |
+
"loss": 0.1479,
|
| 605 |
+
"step": 85
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.36693333333333333,
|
| 609 |
+
"grad_norm": 3.661660671234131,
|
| 610 |
+
"learning_rate": 3.6170212765957453e-06,
|
| 611 |
+
"loss": 0.1436,
|
| 612 |
+
"step": 86
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3712,
|
| 616 |
+
"grad_norm": 4.646217346191406,
|
| 617 |
+
"learning_rate": 3.6595744680851063e-06,
|
| 618 |
+
"loss": 0.1431,
|
| 619 |
+
"step": 87
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.37546666666666667,
|
| 623 |
+
"grad_norm": 3.3780157566070557,
|
| 624 |
+
"learning_rate": 3.702127659574468e-06,
|
| 625 |
+
"loss": 0.1426,
|
| 626 |
+
"step": 88
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.3797333333333333,
|
| 630 |
+
"grad_norm": 4.596078395843506,
|
| 631 |
+
"learning_rate": 3.74468085106383e-06,
|
| 632 |
+
"loss": 0.1422,
|
| 633 |
+
"step": 89
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.384,
|
| 637 |
+
"grad_norm": 3.384758949279785,
|
| 638 |
+
"learning_rate": 3.7872340425531917e-06,
|
| 639 |
+
"loss": 0.1399,
|
| 640 |
+
"step": 90
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.38826666666666665,
|
| 644 |
+
"grad_norm": 6.492227077484131,
|
| 645 |
+
"learning_rate": 3.8297872340425535e-06,
|
| 646 |
+
"loss": 0.1408,
|
| 647 |
+
"step": 91
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.39253333333333335,
|
| 651 |
+
"grad_norm": 4.704247951507568,
|
| 652 |
+
"learning_rate": 3.872340425531915e-06,
|
| 653 |
+
"loss": 0.1392,
|
| 654 |
+
"step": 92
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3968,
|
| 658 |
+
"grad_norm": 7.741692543029785,
|
| 659 |
+
"learning_rate": 3.914893617021277e-06,
|
| 660 |
+
"loss": 0.1413,
|
| 661 |
+
"step": 93
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.4010666666666667,
|
| 665 |
+
"grad_norm": 8.09485149383545,
|
| 666 |
+
"learning_rate": 3.957446808510639e-06,
|
| 667 |
+
"loss": 0.1433,
|
| 668 |
+
"step": 94
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.4053333333333333,
|
| 672 |
+
"grad_norm": 2.9961678981781006,
|
| 673 |
+
"learning_rate": 4.000000000000001e-06,
|
| 674 |
+
"loss": 0.1381,
|
| 675 |
+
"step": 95
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.4096,
|
| 679 |
+
"grad_norm": 4.540401935577393,
|
| 680 |
+
"learning_rate": 4.042553191489362e-06,
|
| 681 |
+
"loss": 0.1376,
|
| 682 |
+
"step": 96
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.41386666666666666,
|
| 686 |
+
"grad_norm": 4.068235397338867,
|
| 687 |
+
"learning_rate": 4.085106382978723e-06,
|
| 688 |
+
"loss": 0.1369,
|
| 689 |
+
"step": 97
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.41813333333333336,
|
| 693 |
+
"grad_norm": 3.181708812713623,
|
| 694 |
+
"learning_rate": 4.127659574468085e-06,
|
| 695 |
+
"loss": 0.1352,
|
| 696 |
+
"step": 98
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.4224,
|
| 700 |
+
"grad_norm": 4.561326026916504,
|
| 701 |
+
"learning_rate": 4.170212765957447e-06,
|
| 702 |
+
"loss": 0.1358,
|
| 703 |
+
"step": 99
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4266666666666667,
|
| 707 |
+
"grad_norm": 3.7372615337371826,
|
| 708 |
+
"learning_rate": 4.212765957446809e-06,
|
| 709 |
+
"loss": 0.1338,
|
| 710 |
+
"step": 100
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.43093333333333333,
|
| 714 |
+
"grad_norm": 6.206492900848389,
|
| 715 |
+
"learning_rate": 4.255319148936171e-06,
|
| 716 |
+
"loss": 0.1376,
|
| 717 |
+
"step": 101
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4352,
|
| 721 |
+
"grad_norm": 4.748600006103516,
|
| 722 |
+
"learning_rate": 4.297872340425532e-06,
|
| 723 |
+
"loss": 0.1353,
|
| 724 |
+
"step": 102
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.43946666666666667,
|
| 728 |
+
"grad_norm": 8.013435363769531,
|
| 729 |
+
"learning_rate": 4.340425531914894e-06,
|
| 730 |
+
"loss": 0.136,
|
| 731 |
+
"step": 103
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4437333333333333,
|
| 735 |
+
"grad_norm": 7.1991190910339355,
|
| 736 |
+
"learning_rate": 4.382978723404256e-06,
|
| 737 |
+
"loss": 0.137,
|
| 738 |
+
"step": 104
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.448,
|
| 742 |
+
"grad_norm": 4.625593185424805,
|
| 743 |
+
"learning_rate": 4.425531914893617e-06,
|
| 744 |
+
"loss": 0.1319,
|
| 745 |
+
"step": 105
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.45226666666666665,
|
| 749 |
+
"grad_norm": 4.910801887512207,
|
| 750 |
+
"learning_rate": 4.468085106382979e-06,
|
| 751 |
+
"loss": 0.134,
|
| 752 |
+
"step": 106
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.45653333333333335,
|
| 756 |
+
"grad_norm": 6.377923488616943,
|
| 757 |
+
"learning_rate": 4.5106382978723406e-06,
|
| 758 |
+
"loss": 0.1342,
|
| 759 |
+
"step": 107
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.4608,
|
| 763 |
+
"grad_norm": 5.0228986740112305,
|
| 764 |
+
"learning_rate": 4.553191489361702e-06,
|
| 765 |
+
"loss": 0.1343,
|
| 766 |
+
"step": 108
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4650666666666667,
|
| 770 |
+
"grad_norm": 6.513828754425049,
|
| 771 |
+
"learning_rate": 4.595744680851064e-06,
|
| 772 |
+
"loss": 0.1308,
|
| 773 |
+
"step": 109
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4693333333333333,
|
| 777 |
+
"grad_norm": 5.182156085968018,
|
| 778 |
+
"learning_rate": 4.638297872340426e-06,
|
| 779 |
+
"loss": 0.1324,
|
| 780 |
+
"step": 110
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4736,
|
| 784 |
+
"grad_norm": 6.711518287658691,
|
| 785 |
+
"learning_rate": 4.680851063829788e-06,
|
| 786 |
+
"loss": 0.1327,
|
| 787 |
+
"step": 111
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.47786666666666666,
|
| 791 |
+
"grad_norm": 5.540736198425293,
|
| 792 |
+
"learning_rate": 4.7234042553191496e-06,
|
| 793 |
+
"loss": 0.1316,
|
| 794 |
+
"step": 112
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.48213333333333336,
|
| 798 |
+
"grad_norm": 5.925957202911377,
|
| 799 |
+
"learning_rate": 4.765957446808511e-06,
|
| 800 |
+
"loss": 0.1302,
|
| 801 |
+
"step": 113
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4864,
|
| 805 |
+
"grad_norm": 5.579200744628906,
|
| 806 |
+
"learning_rate": 4.808510638297872e-06,
|
| 807 |
+
"loss": 0.131,
|
| 808 |
+
"step": 114
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.49066666666666664,
|
| 812 |
+
"grad_norm": 4.405421733856201,
|
| 813 |
+
"learning_rate": 4.851063829787234e-06,
|
| 814 |
+
"loss": 0.1296,
|
| 815 |
+
"step": 115
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.49493333333333334,
|
| 819 |
+
"grad_norm": 5.265398025512695,
|
| 820 |
+
"learning_rate": 4.893617021276596e-06,
|
| 821 |
+
"loss": 0.1291,
|
| 822 |
+
"step": 116
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.4992,
|
| 826 |
+
"grad_norm": 3.860846519470215,
|
| 827 |
+
"learning_rate": 4.936170212765958e-06,
|
| 828 |
+
"loss": 0.1303,
|
| 829 |
+
"step": 117
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.5034666666666666,
|
| 833 |
+
"grad_norm": 2.226341724395752,
|
| 834 |
+
"learning_rate": 4.9787234042553195e-06,
|
| 835 |
+
"loss": 0.1256,
|
| 836 |
+
"step": 118
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.5077333333333334,
|
| 840 |
+
"grad_norm": 8.795147895812988,
|
| 841 |
+
"learning_rate": 5.0212765957446805e-06,
|
| 842 |
+
"loss": 0.1305,
|
| 843 |
+
"step": 119
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.512,
|
| 847 |
+
"grad_norm": 4.675596714019775,
|
| 848 |
+
"learning_rate": 5.063829787234042e-06,
|
| 849 |
+
"loss": 0.1295,
|
| 850 |
+
"step": 120
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.5162666666666667,
|
| 854 |
+
"grad_norm": 11.608587265014648,
|
| 855 |
+
"learning_rate": 5.106382978723404e-06,
|
| 856 |
+
"loss": 0.1315,
|
| 857 |
+
"step": 121
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.5205333333333333,
|
| 861 |
+
"grad_norm": 11.548611640930176,
|
| 862 |
+
"learning_rate": 5.148936170212766e-06,
|
| 863 |
+
"loss": 0.1354,
|
| 864 |
+
"step": 122
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5248,
|
| 868 |
+
"grad_norm": 4.825727462768555,
|
| 869 |
+
"learning_rate": 5.191489361702128e-06,
|
| 870 |
+
"loss": 0.1299,
|
| 871 |
+
"step": 123
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5290666666666667,
|
| 875 |
+
"grad_norm": 7.2229413986206055,
|
| 876 |
+
"learning_rate": 5.2340425531914895e-06,
|
| 877 |
+
"loss": 0.1306,
|
| 878 |
+
"step": 124
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5333333333333333,
|
| 882 |
+
"grad_norm": 6.224844932556152,
|
| 883 |
+
"learning_rate": 5.276595744680851e-06,
|
| 884 |
+
"loss": 0.1332,
|
| 885 |
+
"step": 125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5376,
|
| 889 |
+
"grad_norm": 5.247742652893066,
|
| 890 |
+
"learning_rate": 5.319148936170213e-06,
|
| 891 |
+
"loss": 0.1277,
|
| 892 |
+
"step": 126
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.5418666666666667,
|
| 896 |
+
"grad_norm": 7.463538646697998,
|
| 897 |
+
"learning_rate": 5.361702127659575e-06,
|
| 898 |
+
"loss": 0.1313,
|
| 899 |
+
"step": 127
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5461333333333334,
|
| 903 |
+
"grad_norm": 5.209482192993164,
|
| 904 |
+
"learning_rate": 5.404255319148937e-06,
|
| 905 |
+
"loss": 0.1304,
|
| 906 |
+
"step": 128
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5504,
|
| 910 |
+
"grad_norm": 7.131233215332031,
|
| 911 |
+
"learning_rate": 5.4468085106382985e-06,
|
| 912 |
+
"loss": 0.1283,
|
| 913 |
+
"step": 129
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5546666666666666,
|
| 917 |
+
"grad_norm": 5.096197128295898,
|
| 918 |
+
"learning_rate": 5.48936170212766e-06,
|
| 919 |
+
"loss": 0.1314,
|
| 920 |
+
"step": 130
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5589333333333333,
|
| 924 |
+
"grad_norm": 8.022686004638672,
|
| 925 |
+
"learning_rate": 5.531914893617022e-06,
|
| 926 |
+
"loss": 0.1287,
|
| 927 |
+
"step": 131
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5632,
|
| 931 |
+
"grad_norm": 5.0473175048828125,
|
| 932 |
+
"learning_rate": 5.574468085106384e-06,
|
| 933 |
+
"loss": 0.129,
|
| 934 |
+
"step": 132
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5674666666666667,
|
| 938 |
+
"grad_norm": 8.857738494873047,
|
| 939 |
+
"learning_rate": 5.617021276595746e-06,
|
| 940 |
+
"loss": 0.1288,
|
| 941 |
+
"step": 133
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5717333333333333,
|
| 945 |
+
"grad_norm": 5.748481750488281,
|
| 946 |
+
"learning_rate": 5.6595744680851075e-06,
|
| 947 |
+
"loss": 0.1298,
|
| 948 |
+
"step": 134
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.576,
|
| 952 |
+
"grad_norm": 9.72546100616455,
|
| 953 |
+
"learning_rate": 5.702127659574469e-06,
|
| 954 |
+
"loss": 0.1283,
|
| 955 |
+
"step": 135
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5802666666666667,
|
| 959 |
+
"grad_norm": 7.091835975646973,
|
| 960 |
+
"learning_rate": 5.744680851063831e-06,
|
| 961 |
+
"loss": 0.1286,
|
| 962 |
+
"step": 136
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5845333333333333,
|
| 966 |
+
"grad_norm": 10.19481372833252,
|
| 967 |
+
"learning_rate": 5.787234042553191e-06,
|
| 968 |
+
"loss": 0.1311,
|
| 969 |
+
"step": 137
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5888,
|
| 973 |
+
"grad_norm": 9.936722755432129,
|
| 974 |
+
"learning_rate": 5.829787234042553e-06,
|
| 975 |
+
"loss": 0.1273,
|
| 976 |
+
"step": 138
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5930666666666666,
|
| 980 |
+
"grad_norm": 5.595465183258057,
|
| 981 |
+
"learning_rate": 5.872340425531915e-06,
|
| 982 |
+
"loss": 0.1242,
|
| 983 |
+
"step": 139
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5973333333333334,
|
| 987 |
+
"grad_norm": 5.289859771728516,
|
| 988 |
+
"learning_rate": 5.9148936170212766e-06,
|
| 989 |
+
"loss": 0.1242,
|
| 990 |
+
"step": 140
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.6016,
|
| 994 |
+
"grad_norm": 8.406579971313477,
|
| 995 |
+
"learning_rate": 5.957446808510638e-06,
|
| 996 |
+
"loss": 0.1251,
|
| 997 |
+
"step": 141
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.6058666666666667,
|
| 1001 |
+
"grad_norm": 6.704819679260254,
|
| 1002 |
+
"learning_rate": 6e-06,
|
| 1003 |
+
"loss": 0.1255,
|
| 1004 |
+
"step": 142
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.6101333333333333,
|
| 1008 |
+
"grad_norm": 9.0625,
|
| 1009 |
+
"learning_rate": 6.042553191489362e-06,
|
| 1010 |
+
"loss": 0.1255,
|
| 1011 |
+
"step": 143
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.6144,
|
| 1015 |
+
"grad_norm": 8.436838150024414,
|
| 1016 |
+
"learning_rate": 6.085106382978724e-06,
|
| 1017 |
+
"loss": 0.1247,
|
| 1018 |
+
"step": 144
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.6186666666666667,
|
| 1022 |
+
"grad_norm": 7.0138630867004395,
|
| 1023 |
+
"learning_rate": 6.1276595744680855e-06,
|
| 1024 |
+
"loss": 0.1268,
|
| 1025 |
+
"step": 145
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.6229333333333333,
|
| 1029 |
+
"grad_norm": 6.2439117431640625,
|
| 1030 |
+
"learning_rate": 6.170212765957447e-06,
|
| 1031 |
+
"loss": 0.1231,
|
| 1032 |
+
"step": 146
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.6272,
|
| 1036 |
+
"grad_norm": 8.84973430633545,
|
| 1037 |
+
"learning_rate": 6.212765957446809e-06,
|
| 1038 |
+
"loss": 0.1255,
|
| 1039 |
+
"step": 147
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.6314666666666666,
|
| 1043 |
+
"grad_norm": 7.984808921813965,
|
| 1044 |
+
"learning_rate": 6.255319148936171e-06,
|
| 1045 |
+
"loss": 0.1238,
|
| 1046 |
+
"step": 148
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.6357333333333334,
|
| 1050 |
+
"grad_norm": 6.803660869598389,
|
| 1051 |
+
"learning_rate": 6.297872340425533e-06,
|
| 1052 |
+
"loss": 0.1226,
|
| 1053 |
+
"step": 149
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.64,
|
| 1057 |
+
"grad_norm": 5.848384380340576,
|
| 1058 |
+
"learning_rate": 6.3404255319148945e-06,
|
| 1059 |
+
"loss": 0.1224,
|
| 1060 |
+
"step": 150
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6442666666666667,
|
| 1064 |
+
"grad_norm": 8.635106086730957,
|
| 1065 |
+
"learning_rate": 6.382978723404256e-06,
|
| 1066 |
+
"loss": 0.1232,
|
| 1067 |
+
"step": 151
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.6485333333333333,
|
| 1071 |
+
"grad_norm": 7.334502220153809,
|
| 1072 |
+
"learning_rate": 6.425531914893618e-06,
|
| 1073 |
+
"loss": 0.1235,
|
| 1074 |
+
"step": 152
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6528,
|
| 1078 |
+
"grad_norm": 7.256570816040039,
|
| 1079 |
+
"learning_rate": 6.46808510638298e-06,
|
| 1080 |
+
"loss": 0.1235,
|
| 1081 |
+
"step": 153
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6570666666666667,
|
| 1085 |
+
"grad_norm": 6.407953262329102,
|
| 1086 |
+
"learning_rate": 6.510638297872342e-06,
|
| 1087 |
+
"loss": 0.122,
|
| 1088 |
+
"step": 154
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6613333333333333,
|
| 1092 |
+
"grad_norm": 8.223788261413574,
|
| 1093 |
+
"learning_rate": 6.553191489361702e-06,
|
| 1094 |
+
"loss": 0.124,
|
| 1095 |
+
"step": 155
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6656,
|
| 1099 |
+
"grad_norm": 6.739434242248535,
|
| 1100 |
+
"learning_rate": 6.595744680851064e-06,
|
| 1101 |
+
"loss": 0.123,
|
| 1102 |
+
"step": 156
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6698666666666667,
|
| 1106 |
+
"grad_norm": 7.93510103225708,
|
| 1107 |
+
"learning_rate": 6.6382978723404254e-06,
|
| 1108 |
+
"loss": 0.123,
|
| 1109 |
+
"step": 157
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6741333333333334,
|
| 1113 |
+
"grad_norm": 7.154916763305664,
|
| 1114 |
+
"learning_rate": 6.680851063829787e-06,
|
| 1115 |
+
"loss": 0.1221,
|
| 1116 |
+
"step": 158
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6784,
|
| 1120 |
+
"grad_norm": 6.691060543060303,
|
| 1121 |
+
"learning_rate": 6.723404255319149e-06,
|
| 1122 |
+
"loss": 0.1225,
|
| 1123 |
+
"step": 159
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6826666666666666,
|
| 1127 |
+
"grad_norm": 5.641987323760986,
|
| 1128 |
+
"learning_rate": 6.765957446808511e-06,
|
| 1129 |
+
"loss": 0.1208,
|
| 1130 |
+
"step": 160
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6869333333333333,
|
| 1134 |
+
"grad_norm": 7.909478187561035,
|
| 1135 |
+
"learning_rate": 6.808510638297873e-06,
|
| 1136 |
+
"loss": 0.1229,
|
| 1137 |
+
"step": 161
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6912,
|
| 1141 |
+
"grad_norm": 6.212748050689697,
|
| 1142 |
+
"learning_rate": 6.8510638297872344e-06,
|
| 1143 |
+
"loss": 0.1201,
|
| 1144 |
+
"step": 162
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6954666666666667,
|
| 1148 |
+
"grad_norm": 7.641393661499023,
|
| 1149 |
+
"learning_rate": 6.893617021276596e-06,
|
| 1150 |
+
"loss": 0.1215,
|
| 1151 |
+
"step": 163
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6997333333333333,
|
| 1155 |
+
"grad_norm": 6.818618297576904,
|
| 1156 |
+
"learning_rate": 6.936170212765958e-06,
|
| 1157 |
+
"loss": 0.1212,
|
| 1158 |
+
"step": 164
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.704,
|
| 1162 |
+
"grad_norm": 6.080795764923096,
|
| 1163 |
+
"learning_rate": 6.97872340425532e-06,
|
| 1164 |
+
"loss": 0.1204,
|
| 1165 |
+
"step": 165
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.7082666666666667,
|
| 1169 |
+
"grad_norm": 5.294103145599365,
|
| 1170 |
+
"learning_rate": 7.021276595744682e-06,
|
| 1171 |
+
"loss": 0.12,
|
| 1172 |
+
"step": 166
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.7125333333333334,
|
| 1176 |
+
"grad_norm": 7.565417289733887,
|
| 1177 |
+
"learning_rate": 7.0638297872340434e-06,
|
| 1178 |
+
"loss": 0.1209,
|
| 1179 |
+
"step": 167
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.7168,
|
| 1183 |
+
"grad_norm": 6.280045509338379,
|
| 1184 |
+
"learning_rate": 7.106382978723405e-06,
|
| 1185 |
+
"loss": 0.1212,
|
| 1186 |
+
"step": 168
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.7210666666666666,
|
| 1190 |
+
"grad_norm": 6.898222923278809,
|
| 1191 |
+
"learning_rate": 7.148936170212767e-06,
|
| 1192 |
+
"loss": 0.1206,
|
| 1193 |
+
"step": 169
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.7253333333333334,
|
| 1197 |
+
"grad_norm": 6.254840850830078,
|
| 1198 |
+
"learning_rate": 7.191489361702129e-06,
|
| 1199 |
+
"loss": 0.1203,
|
| 1200 |
+
"step": 170
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.7296,
|
| 1204 |
+
"grad_norm": 6.527096748352051,
|
| 1205 |
+
"learning_rate": 7.234042553191491e-06,
|
| 1206 |
+
"loss": 0.122,
|
| 1207 |
+
"step": 171
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.7338666666666667,
|
| 1211 |
+
"grad_norm": 5.724420070648193,
|
| 1212 |
+
"learning_rate": 7.2765957446808524e-06,
|
| 1213 |
+
"loss": 0.1207,
|
| 1214 |
+
"step": 172
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7381333333333333,
|
| 1218 |
+
"grad_norm": 6.684255123138428,
|
| 1219 |
+
"learning_rate": 7.3191489361702125e-06,
|
| 1220 |
+
"loss": 0.1219,
|
| 1221 |
+
"step": 173
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7424,
|
| 1225 |
+
"grad_norm": 6.503657341003418,
|
| 1226 |
+
"learning_rate": 7.361702127659574e-06,
|
| 1227 |
+
"loss": 0.1178,
|
| 1228 |
+
"step": 174
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.7466666666666667,
|
| 1232 |
+
"grad_norm": 5.723449230194092,
|
| 1233 |
+
"learning_rate": 7.404255319148936e-06,
|
| 1234 |
+
"loss": 0.1199,
|
| 1235 |
+
"step": 175
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7509333333333333,
|
| 1239 |
+
"grad_norm": 4.989058017730713,
|
| 1240 |
+
"learning_rate": 7.446808510638298e-06,
|
| 1241 |
+
"loss": 0.1181,
|
| 1242 |
+
"step": 176
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7552,
|
| 1246 |
+
"grad_norm": 7.475578784942627,
|
| 1247 |
+
"learning_rate": 7.48936170212766e-06,
|
| 1248 |
+
"loss": 0.1221,
|
| 1249 |
+
"step": 177
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7594666666666666,
|
| 1253 |
+
"grad_norm": 6.384317398071289,
|
| 1254 |
+
"learning_rate": 7.5319148936170215e-06,
|
| 1255 |
+
"loss": 0.1193,
|
| 1256 |
+
"step": 178
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7637333333333334,
|
| 1260 |
+
"grad_norm": 5.993687629699707,
|
| 1261 |
+
"learning_rate": 7.574468085106383e-06,
|
| 1262 |
+
"loss": 0.1197,
|
| 1263 |
+
"step": 179
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.768,
|
| 1267 |
+
"grad_norm": 5.651639938354492,
|
| 1268 |
+
"learning_rate": 7.617021276595745e-06,
|
| 1269 |
+
"loss": 0.118,
|
| 1270 |
+
"step": 180
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.7722666666666667,
|
| 1274 |
+
"grad_norm": 5.343756675720215,
|
| 1275 |
+
"learning_rate": 7.659574468085107e-06,
|
| 1276 |
+
"loss": 0.1191,
|
| 1277 |
+
"step": 181
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.7765333333333333,
|
| 1281 |
+
"grad_norm": 3.9733238220214844,
|
| 1282 |
+
"learning_rate": 7.702127659574469e-06,
|
| 1283 |
+
"loss": 0.117,
|
| 1284 |
+
"step": 182
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7808,
|
| 1288 |
+
"grad_norm": 8.402620315551758,
|
| 1289 |
+
"learning_rate": 7.74468085106383e-06,
|
| 1290 |
+
"loss": 0.1235,
|
| 1291 |
+
"step": 183
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7850666666666667,
|
| 1295 |
+
"grad_norm": 8.289787292480469,
|
| 1296 |
+
"learning_rate": 7.787234042553192e-06,
|
| 1297 |
+
"loss": 0.121,
|
| 1298 |
+
"step": 184
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7893333333333333,
|
| 1302 |
+
"grad_norm": 3.6568987369537354,
|
| 1303 |
+
"learning_rate": 7.829787234042554e-06,
|
| 1304 |
+
"loss": 0.1174,
|
| 1305 |
+
"step": 185
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7936,
|
| 1309 |
+
"grad_norm": 11.13115119934082,
|
| 1310 |
+
"learning_rate": 7.872340425531916e-06,
|
| 1311 |
+
"loss": 0.1257,
|
| 1312 |
+
"step": 186
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7978666666666666,
|
| 1316 |
+
"grad_norm": 5.116509914398193,
|
| 1317 |
+
"learning_rate": 7.914893617021278e-06,
|
| 1318 |
+
"loss": 0.1195,
|
| 1319 |
+
"step": 187
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.8021333333333334,
|
| 1323 |
+
"grad_norm": 15.565638542175293,
|
| 1324 |
+
"learning_rate": 7.95744680851064e-06,
|
| 1325 |
+
"loss": 0.1417,
|
| 1326 |
+
"step": 188
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.8064,
|
| 1330 |
+
"grad_norm": 16.17276954650879,
|
| 1331 |
+
"learning_rate": 8.000000000000001e-06,
|
| 1332 |
+
"loss": 0.152,
|
| 1333 |
+
"step": 189
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.8106666666666666,
|
| 1337 |
+
"grad_norm": 6.167568206787109,
|
| 1338 |
+
"learning_rate": 8.042553191489363e-06,
|
| 1339 |
+
"loss": 0.1233,
|
| 1340 |
+
"step": 190
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.8149333333333333,
|
| 1344 |
+
"grad_norm": 12.830926895141602,
|
| 1345 |
+
"learning_rate": 8.085106382978723e-06,
|
| 1346 |
+
"loss": 0.1399,
|
| 1347 |
+
"step": 191
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.8192,
|
| 1351 |
+
"grad_norm": 12.8762845993042,
|
| 1352 |
+
"learning_rate": 8.127659574468085e-06,
|
| 1353 |
+
"loss": 0.1521,
|
| 1354 |
+
"step": 192
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.8234666666666667,
|
| 1358 |
+
"grad_norm": 8.177667617797852,
|
| 1359 |
+
"learning_rate": 8.170212765957447e-06,
|
| 1360 |
+
"loss": 0.1298,
|
| 1361 |
+
"step": 193
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.8277333333333333,
|
| 1365 |
+
"grad_norm": 6.849120616912842,
|
| 1366 |
+
"learning_rate": 8.212765957446809e-06,
|
| 1367 |
+
"loss": 0.1287,
|
| 1368 |
+
"step": 194
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.832,
|
| 1372 |
+
"grad_norm": 7.727602481842041,
|
| 1373 |
+
"learning_rate": 8.25531914893617e-06,
|
| 1374 |
+
"loss": 0.1332,
|
| 1375 |
+
"step": 195
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.8362666666666667,
|
| 1379 |
+
"grad_norm": 5.658266067504883,
|
| 1380 |
+
"learning_rate": 8.297872340425532e-06,
|
| 1381 |
+
"loss": 0.1278,
|
| 1382 |
+
"step": 196
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.8405333333333334,
|
| 1386 |
+
"grad_norm": 5.271341800689697,
|
| 1387 |
+
"learning_rate": 8.340425531914894e-06,
|
| 1388 |
+
"loss": 0.1254,
|
| 1389 |
+
"step": 197
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.8448,
|
| 1393 |
+
"grad_norm": 5.509979248046875,
|
| 1394 |
+
"learning_rate": 8.382978723404256e-06,
|
| 1395 |
+
"loss": 0.1283,
|
| 1396 |
+
"step": 198
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.8490666666666666,
|
| 1400 |
+
"grad_norm": 4.625118732452393,
|
| 1401 |
+
"learning_rate": 8.425531914893618e-06,
|
| 1402 |
+
"loss": 0.1226,
|
| 1403 |
+
"step": 199
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.8533333333333334,
|
| 1407 |
+
"grad_norm": 5.153396129608154,
|
| 1408 |
+
"learning_rate": 8.46808510638298e-06,
|
| 1409 |
+
"loss": 0.1258,
|
| 1410 |
+
"step": 200
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.8576,
|
| 1414 |
+
"grad_norm": 4.286827564239502,
|
| 1415 |
+
"learning_rate": 8.510638297872341e-06,
|
| 1416 |
+
"loss": 0.1223,
|
| 1417 |
+
"step": 201
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8618666666666667,
|
| 1421 |
+
"grad_norm": 5.461269855499268,
|
| 1422 |
+
"learning_rate": 8.553191489361703e-06,
|
| 1423 |
+
"loss": 0.1255,
|
| 1424 |
+
"step": 202
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.8661333333333333,
|
| 1428 |
+
"grad_norm": 3.8906216621398926,
|
| 1429 |
+
"learning_rate": 8.595744680851065e-06,
|
| 1430 |
+
"loss": 0.1216,
|
| 1431 |
+
"step": 203
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.8704,
|
| 1435 |
+
"grad_norm": 5.908001899719238,
|
| 1436 |
+
"learning_rate": 8.638297872340427e-06,
|
| 1437 |
+
"loss": 0.1266,
|
| 1438 |
+
"step": 204
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.8746666666666667,
|
| 1442 |
+
"grad_norm": 3.639004945755005,
|
| 1443 |
+
"learning_rate": 8.680851063829788e-06,
|
| 1444 |
+
"loss": 0.1227,
|
| 1445 |
+
"step": 205
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.8789333333333333,
|
| 1449 |
+
"grad_norm": 7.122750282287598,
|
| 1450 |
+
"learning_rate": 8.72340425531915e-06,
|
| 1451 |
+
"loss": 0.1322,
|
| 1452 |
+
"step": 206
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.8832,
|
| 1456 |
+
"grad_norm": 5.722088813781738,
|
| 1457 |
+
"learning_rate": 8.765957446808512e-06,
|
| 1458 |
+
"loss": 0.1298,
|
| 1459 |
+
"step": 207
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8874666666666666,
|
| 1463 |
+
"grad_norm": 5.114867210388184,
|
| 1464 |
+
"learning_rate": 8.808510638297874e-06,
|
| 1465 |
+
"loss": 0.1245,
|
| 1466 |
+
"step": 208
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.8917333333333334,
|
| 1470 |
+
"grad_norm": 5.530348300933838,
|
| 1471 |
+
"learning_rate": 8.851063829787234e-06,
|
| 1472 |
+
"loss": 0.126,
|
| 1473 |
+
"step": 209
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.896,
|
| 1477 |
+
"grad_norm": 3.323356866836548,
|
| 1478 |
+
"learning_rate": 8.893617021276596e-06,
|
| 1479 |
+
"loss": 0.1221,
|
| 1480 |
+
"step": 210
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.9002666666666667,
|
| 1484 |
+
"grad_norm": 5.217687606811523,
|
| 1485 |
+
"learning_rate": 8.936170212765958e-06,
|
| 1486 |
+
"loss": 0.1234,
|
| 1487 |
+
"step": 211
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.9045333333333333,
|
| 1491 |
+
"grad_norm": 2.8208258152008057,
|
| 1492 |
+
"learning_rate": 8.97872340425532e-06,
|
| 1493 |
+
"loss": 0.1226,
|
| 1494 |
+
"step": 212
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.9088,
|
| 1498 |
+
"grad_norm": 6.277392864227295,
|
| 1499 |
+
"learning_rate": 9.021276595744681e-06,
|
| 1500 |
+
"loss": 0.1306,
|
| 1501 |
+
"step": 213
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.9130666666666667,
|
| 1505 |
+
"grad_norm": 4.85951042175293,
|
| 1506 |
+
"learning_rate": 9.063829787234043e-06,
|
| 1507 |
+
"loss": 0.1375,
|
| 1508 |
+
"step": 214
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.9173333333333333,
|
| 1512 |
+
"grad_norm": 5.551103115081787,
|
| 1513 |
+
"learning_rate": 9.106382978723405e-06,
|
| 1514 |
+
"loss": 0.1224,
|
| 1515 |
+
"step": 215
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.9216,
|
| 1519 |
+
"grad_norm": 4.771853923797607,
|
| 1520 |
+
"learning_rate": 9.148936170212767e-06,
|
| 1521 |
+
"loss": 0.1347,
|
| 1522 |
+
"step": 216
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.9258666666666666,
|
| 1526 |
+
"grad_norm": 4.560388088226318,
|
| 1527 |
+
"learning_rate": 9.191489361702128e-06,
|
| 1528 |
+
"loss": 0.1224,
|
| 1529 |
+
"step": 217
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.9301333333333334,
|
| 1533 |
+
"grad_norm": 3.6074557304382324,
|
| 1534 |
+
"learning_rate": 9.23404255319149e-06,
|
| 1535 |
+
"loss": 0.1249,
|
| 1536 |
+
"step": 218
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.9344,
|
| 1540 |
+
"grad_norm": 4.925154685974121,
|
| 1541 |
+
"learning_rate": 9.276595744680852e-06,
|
| 1542 |
+
"loss": 0.1256,
|
| 1543 |
+
"step": 219
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.9386666666666666,
|
| 1547 |
+
"grad_norm": 3.3502466678619385,
|
| 1548 |
+
"learning_rate": 9.319148936170214e-06,
|
| 1549 |
+
"loss": 0.1221,
|
| 1550 |
+
"step": 220
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.9429333333333333,
|
| 1554 |
+
"grad_norm": 5.651970386505127,
|
| 1555 |
+
"learning_rate": 9.361702127659576e-06,
|
| 1556 |
+
"loss": 0.1258,
|
| 1557 |
+
"step": 221
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.9472,
|
| 1561 |
+
"grad_norm": 5.056815147399902,
|
| 1562 |
+
"learning_rate": 9.404255319148937e-06,
|
| 1563 |
+
"loss": 0.1293,
|
| 1564 |
+
"step": 222
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.9514666666666667,
|
| 1568 |
+
"grad_norm": 4.17104959487915,
|
| 1569 |
+
"learning_rate": 9.446808510638299e-06,
|
| 1570 |
+
"loss": 0.1197,
|
| 1571 |
+
"step": 223
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.9557333333333333,
|
| 1575 |
+
"grad_norm": 4.450770378112793,
|
| 1576 |
+
"learning_rate": 9.489361702127661e-06,
|
| 1577 |
+
"loss": 0.1237,
|
| 1578 |
+
"step": 224
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.96,
|
| 1582 |
+
"grad_norm": 4.091788291931152,
|
| 1583 |
+
"learning_rate": 9.531914893617023e-06,
|
| 1584 |
+
"loss": 0.1211,
|
| 1585 |
+
"step": 225
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.9642666666666667,
|
| 1589 |
+
"grad_norm": 3.8398008346557617,
|
| 1590 |
+
"learning_rate": 9.574468085106385e-06,
|
| 1591 |
+
"loss": 0.1214,
|
| 1592 |
+
"step": 226
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.9685333333333334,
|
| 1596 |
+
"grad_norm": 3.1588594913482666,
|
| 1597 |
+
"learning_rate": 9.617021276595745e-06,
|
| 1598 |
+
"loss": 0.1185,
|
| 1599 |
+
"step": 227
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.9728,
|
| 1603 |
+
"grad_norm": 3.5648460388183594,
|
| 1604 |
+
"learning_rate": 9.659574468085106e-06,
|
| 1605 |
+
"loss": 0.1174,
|
| 1606 |
+
"step": 228
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.9770666666666666,
|
| 1610 |
+
"grad_norm": 5.8639235496521,
|
| 1611 |
+
"learning_rate": 9.702127659574468e-06,
|
| 1612 |
+
"loss": 0.1249,
|
| 1613 |
+
"step": 229
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.9813333333333333,
|
| 1617 |
+
"grad_norm": 2.6490297317504883,
|
| 1618 |
+
"learning_rate": 9.74468085106383e-06,
|
| 1619 |
+
"loss": 0.1178,
|
| 1620 |
+
"step": 230
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.9856,
|
| 1624 |
+
"grad_norm": 6.386547088623047,
|
| 1625 |
+
"learning_rate": 9.787234042553192e-06,
|
| 1626 |
+
"loss": 0.1357,
|
| 1627 |
+
"step": 231
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.9898666666666667,
|
| 1631 |
+
"grad_norm": 5.654618740081787,
|
| 1632 |
+
"learning_rate": 9.829787234042554e-06,
|
| 1633 |
+
"loss": 0.1374,
|
| 1634 |
+
"step": 232
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.9941333333333333,
|
| 1638 |
+
"grad_norm": 4.289252758026123,
|
| 1639 |
+
"learning_rate": 9.872340425531915e-06,
|
| 1640 |
+
"loss": 0.1197,
|
| 1641 |
+
"step": 233
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9984,
|
| 1645 |
+
"grad_norm": 7.1161723136901855,
|
| 1646 |
+
"learning_rate": 9.914893617021277e-06,
|
| 1647 |
+
"loss": 0.1322,
|
| 1648 |
+
"step": 234
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.0,
|
| 1652 |
+
"grad_norm": 7.1161723136901855,
|
| 1653 |
+
"learning_rate": 9.957446808510639e-06,
|
| 1654 |
+
"loss": 0.0488,
|
| 1655 |
+
"step": 235
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.0042666666666666,
|
| 1659 |
+
"grad_norm": 5.935182571411133,
|
| 1660 |
+
"learning_rate": 1e-05,
|
| 1661 |
+
"loss": 0.1287,
|
| 1662 |
+
"step": 236
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.0085333333333333,
|
| 1666 |
+
"grad_norm": 5.292780876159668,
|
| 1667 |
+
"learning_rate": 9.999994484067655e-06,
|
| 1668 |
+
"loss": 0.1246,
|
| 1669 |
+
"step": 237
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.0128,
|
| 1673 |
+
"grad_norm": 5.135565757751465,
|
| 1674 |
+
"learning_rate": 9.999977936282789e-06,
|
| 1675 |
+
"loss": 0.1253,
|
| 1676 |
+
"step": 238
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.0170666666666666,
|
| 1680 |
+
"grad_norm": 3.817563533782959,
|
| 1681 |
+
"learning_rate": 9.999950356681914e-06,
|
| 1682 |
+
"loss": 0.1245,
|
| 1683 |
+
"step": 239
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.0213333333333334,
|
| 1687 |
+
"grad_norm": 5.6104278564453125,
|
| 1688 |
+
"learning_rate": 9.999911745325877e-06,
|
| 1689 |
+
"loss": 0.1267,
|
| 1690 |
+
"step": 240
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.0256,
|
| 1694 |
+
"grad_norm": 3.947054624557495,
|
| 1695 |
+
"learning_rate": 9.999862102299874e-06,
|
| 1696 |
+
"loss": 0.1252,
|
| 1697 |
+
"step": 241
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.0298666666666667,
|
| 1701 |
+
"grad_norm": 6.3606343269348145,
|
| 1702 |
+
"learning_rate": 9.999801427713432e-06,
|
| 1703 |
+
"loss": 0.1268,
|
| 1704 |
+
"step": 242
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.0341333333333333,
|
| 1708 |
+
"grad_norm": 4.345498561859131,
|
| 1709 |
+
"learning_rate": 9.999729721700426e-06,
|
| 1710 |
+
"loss": 0.128,
|
| 1711 |
+
"step": 243
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.0384,
|
| 1715 |
+
"grad_norm": 7.323845863342285,
|
| 1716 |
+
"learning_rate": 9.999646984419061e-06,
|
| 1717 |
+
"loss": 0.1287,
|
| 1718 |
+
"step": 244
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.0426666666666666,
|
| 1722 |
+
"grad_norm": 5.8514814376831055,
|
| 1723 |
+
"learning_rate": 9.999553216051892e-06,
|
| 1724 |
+
"loss": 0.1331,
|
| 1725 |
+
"step": 245
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.0469333333333333,
|
| 1729 |
+
"grad_norm": 5.87026834487915,
|
| 1730 |
+
"learning_rate": 9.999448416805802e-06,
|
| 1731 |
+
"loss": 0.125,
|
| 1732 |
+
"step": 246
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.0512,
|
| 1736 |
+
"grad_norm": 5.522796154022217,
|
| 1737 |
+
"learning_rate": 9.99933258691202e-06,
|
| 1738 |
+
"loss": 0.1237,
|
| 1739 |
+
"step": 247
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.0554666666666668,
|
| 1743 |
+
"grad_norm": 3.951136589050293,
|
| 1744 |
+
"learning_rate": 9.999205726626109e-06,
|
| 1745 |
+
"loss": 0.1233,
|
| 1746 |
+
"step": 248
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.0597333333333334,
|
| 1750 |
+
"grad_norm": 4.852754592895508,
|
| 1751 |
+
"learning_rate": 9.99906783622797e-06,
|
| 1752 |
+
"loss": 0.122,
|
| 1753 |
+
"step": 249
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.064,
|
| 1757 |
+
"grad_norm": 4.077630043029785,
|
| 1758 |
+
"learning_rate": 9.998918916021842e-06,
|
| 1759 |
+
"loss": 0.1207,
|
| 1760 |
+
"step": 250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.0682666666666667,
|
| 1764 |
+
"grad_norm": 4.453201770782471,
|
| 1765 |
+
"learning_rate": 9.998758966336296e-06,
|
| 1766 |
+
"loss": 0.12,
|
| 1767 |
+
"step": 251
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.0725333333333333,
|
| 1771 |
+
"grad_norm": 2.438328504562378,
|
| 1772 |
+
"learning_rate": 9.998587987524243e-06,
|
| 1773 |
+
"loss": 0.117,
|
| 1774 |
+
"step": 252
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.0768,
|
| 1778 |
+
"grad_norm": 3.7442538738250732,
|
| 1779 |
+
"learning_rate": 9.998405979962926e-06,
|
| 1780 |
+
"loss": 0.1187,
|
| 1781 |
+
"step": 253
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.0810666666666666,
|
| 1785 |
+
"grad_norm": 3.887525796890259,
|
| 1786 |
+
"learning_rate": 9.99821294405392e-06,
|
| 1787 |
+
"loss": 0.1184,
|
| 1788 |
+
"step": 254
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.0853333333333333,
|
| 1792 |
+
"grad_norm": 6.806149005889893,
|
| 1793 |
+
"learning_rate": 9.998008880223134e-06,
|
| 1794 |
+
"loss": 0.1204,
|
| 1795 |
+
"step": 255
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.0896,
|
| 1799 |
+
"grad_norm": 3.5106143951416016,
|
| 1800 |
+
"learning_rate": 9.99779378892081e-06,
|
| 1801 |
+
"loss": 0.1215,
|
| 1802 |
+
"step": 256
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.0938666666666668,
|
| 1806 |
+
"grad_norm": 7.366903305053711,
|
| 1807 |
+
"learning_rate": 9.997567670621521e-06,
|
| 1808 |
+
"loss": 0.1273,
|
| 1809 |
+
"step": 257
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.0981333333333334,
|
| 1813 |
+
"grad_norm": 5.446555137634277,
|
| 1814 |
+
"learning_rate": 9.997330525824168e-06,
|
| 1815 |
+
"loss": 0.1254,
|
| 1816 |
+
"step": 258
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.1024,
|
| 1820 |
+
"grad_norm": 5.372413158416748,
|
| 1821 |
+
"learning_rate": 9.997082355051976e-06,
|
| 1822 |
+
"loss": 0.1202,
|
| 1823 |
+
"step": 259
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.1066666666666667,
|
| 1827 |
+
"grad_norm": 6.54384183883667,
|
| 1828 |
+
"learning_rate": 9.996823158852508e-06,
|
| 1829 |
+
"loss": 0.1248,
|
| 1830 |
+
"step": 260
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.1109333333333333,
|
| 1834 |
+
"grad_norm": 4.519038200378418,
|
| 1835 |
+
"learning_rate": 9.996552937797646e-06,
|
| 1836 |
+
"loss": 0.1218,
|
| 1837 |
+
"step": 261
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.1152,
|
| 1841 |
+
"grad_norm": 6.259124279022217,
|
| 1842 |
+
"learning_rate": 9.996271692483597e-06,
|
| 1843 |
+
"loss": 0.1217,
|
| 1844 |
+
"step": 262
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.1194666666666666,
|
| 1848 |
+
"grad_norm": 4.168610572814941,
|
| 1849 |
+
"learning_rate": 9.995979423530893e-06,
|
| 1850 |
+
"loss": 0.1215,
|
| 1851 |
+
"step": 263
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.1237333333333333,
|
| 1855 |
+
"grad_norm": 3.8914642333984375,
|
| 1856 |
+
"learning_rate": 9.995676131584389e-06,
|
| 1857 |
+
"loss": 0.1174,
|
| 1858 |
+
"step": 264
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.1280000000000001,
|
| 1862 |
+
"grad_norm": 5.742394924163818,
|
| 1863 |
+
"learning_rate": 9.995361817313262e-06,
|
| 1864 |
+
"loss": 0.1226,
|
| 1865 |
+
"step": 265
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.1322666666666668,
|
| 1869 |
+
"grad_norm": 4.259416580200195,
|
| 1870 |
+
"learning_rate": 9.995036481411005e-06,
|
| 1871 |
+
"loss": 0.1196,
|
| 1872 |
+
"step": 266
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.1365333333333334,
|
| 1876 |
+
"grad_norm": 5.219086647033691,
|
| 1877 |
+
"learning_rate": 9.994700124595429e-06,
|
| 1878 |
+
"loss": 0.1192,
|
| 1879 |
+
"step": 267
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.1408,
|
| 1883 |
+
"grad_norm": 5.503093719482422,
|
| 1884 |
+
"learning_rate": 9.994352747608663e-06,
|
| 1885 |
+
"loss": 0.1215,
|
| 1886 |
+
"step": 268
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.1450666666666667,
|
| 1890 |
+
"grad_norm": 4.057020664215088,
|
| 1891 |
+
"learning_rate": 9.993994351217153e-06,
|
| 1892 |
+
"loss": 0.1162,
|
| 1893 |
+
"step": 269
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.1493333333333333,
|
| 1897 |
+
"grad_norm": 6.2084269523620605,
|
| 1898 |
+
"learning_rate": 9.993624936211652e-06,
|
| 1899 |
+
"loss": 0.1181,
|
| 1900 |
+
"step": 270
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.1536,
|
| 1904 |
+
"grad_norm": 3.86226487159729,
|
| 1905 |
+
"learning_rate": 9.993244503407227e-06,
|
| 1906 |
+
"loss": 0.1197,
|
| 1907 |
+
"step": 271
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.1578666666666666,
|
| 1911 |
+
"grad_norm": 5.438651084899902,
|
| 1912 |
+
"learning_rate": 9.992853053643257e-06,
|
| 1913 |
+
"loss": 0.1181,
|
| 1914 |
+
"step": 272
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.1621333333333332,
|
| 1918 |
+
"grad_norm": 3.089888095855713,
|
| 1919 |
+
"learning_rate": 9.992450587783426e-06,
|
| 1920 |
+
"loss": 0.1197,
|
| 1921 |
+
"step": 273
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.1663999999999999,
|
| 1925 |
+
"grad_norm": 6.028289794921875,
|
| 1926 |
+
"learning_rate": 9.992037106715722e-06,
|
| 1927 |
+
"loss": 0.121,
|
| 1928 |
+
"step": 274
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.1706666666666667,
|
| 1932 |
+
"grad_norm": 3.518129587173462,
|
| 1933 |
+
"learning_rate": 9.991612611352439e-06,
|
| 1934 |
+
"loss": 0.1209,
|
| 1935 |
+
"step": 275
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.1749333333333334,
|
| 1939 |
+
"grad_norm": 5.490471839904785,
|
| 1940 |
+
"learning_rate": 9.991177102630173e-06,
|
| 1941 |
+
"loss": 0.122,
|
| 1942 |
+
"step": 276
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.1792,
|
| 1946 |
+
"grad_norm": 4.619834899902344,
|
| 1947 |
+
"learning_rate": 9.990730581509817e-06,
|
| 1948 |
+
"loss": 0.1232,
|
| 1949 |
+
"step": 277
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.1834666666666667,
|
| 1953 |
+
"grad_norm": 3.749178886413574,
|
| 1954 |
+
"learning_rate": 9.990273048976566e-06,
|
| 1955 |
+
"loss": 0.1188,
|
| 1956 |
+
"step": 278
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.1877333333333333,
|
| 1960 |
+
"grad_norm": 5.667284965515137,
|
| 1961 |
+
"learning_rate": 9.989804506039905e-06,
|
| 1962 |
+
"loss": 0.1236,
|
| 1963 |
+
"step": 279
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.192,
|
| 1967 |
+
"grad_norm": 3.1727864742279053,
|
| 1968 |
+
"learning_rate": 9.989324953733615e-06,
|
| 1969 |
+
"loss": 0.1229,
|
| 1970 |
+
"step": 280
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.1962666666666666,
|
| 1974 |
+
"grad_norm": 5.744826793670654,
|
| 1975 |
+
"learning_rate": 9.988834393115768e-06,
|
| 1976 |
+
"loss": 0.1255,
|
| 1977 |
+
"step": 281
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.2005333333333335,
|
| 1981 |
+
"grad_norm": 4.079339504241943,
|
| 1982 |
+
"learning_rate": 9.98833282526872e-06,
|
| 1983 |
+
"loss": 0.1287,
|
| 1984 |
+
"step": 282
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.2048,
|
| 1988 |
+
"grad_norm": 5.137429237365723,
|
| 1989 |
+
"learning_rate": 9.987820251299121e-06,
|
| 1990 |
+
"loss": 0.1243,
|
| 1991 |
+
"step": 283
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.2090666666666667,
|
| 1995 |
+
"grad_norm": 4.121692657470703,
|
| 1996 |
+
"learning_rate": 9.9872966723379e-06,
|
| 1997 |
+
"loss": 0.1231,
|
| 1998 |
+
"step": 284
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.2133333333333334,
|
| 2002 |
+
"grad_norm": 4.33402156829834,
|
| 2003 |
+
"learning_rate": 9.986762089540268e-06,
|
| 2004 |
+
"loss": 0.1233,
|
| 2005 |
+
"step": 285
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.2176,
|
| 2009 |
+
"grad_norm": 4.249146461486816,
|
| 2010 |
+
"learning_rate": 9.986216504085709e-06,
|
| 2011 |
+
"loss": 0.12,
|
| 2012 |
+
"step": 286
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.2218666666666667,
|
| 2016 |
+
"grad_norm": 4.126492500305176,
|
| 2017 |
+
"learning_rate": 9.985659917177991e-06,
|
| 2018 |
+
"loss": 0.1198,
|
| 2019 |
+
"step": 287
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.2261333333333333,
|
| 2023 |
+
"grad_norm": 4.10832405090332,
|
| 2024 |
+
"learning_rate": 9.985092330045156e-06,
|
| 2025 |
+
"loss": 0.1185,
|
| 2026 |
+
"step": 288
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.2304,
|
| 2030 |
+
"grad_norm": 3.8791186809539795,
|
| 2031 |
+
"learning_rate": 9.98451374393951e-06,
|
| 2032 |
+
"loss": 0.1183,
|
| 2033 |
+
"step": 289
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.2346666666666666,
|
| 2037 |
+
"grad_norm": 4.378361701965332,
|
| 2038 |
+
"learning_rate": 9.983924160137627e-06,
|
| 2039 |
+
"loss": 0.1174,
|
| 2040 |
+
"step": 290
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.2389333333333332,
|
| 2044 |
+
"grad_norm": 2.184255361557007,
|
| 2045 |
+
"learning_rate": 9.983323579940351e-06,
|
| 2046 |
+
"loss": 0.1191,
|
| 2047 |
+
"step": 291
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.2432,
|
| 2051 |
+
"grad_norm": 5.996938228607178,
|
| 2052 |
+
"learning_rate": 9.982712004672787e-06,
|
| 2053 |
+
"loss": 0.1297,
|
| 2054 |
+
"step": 292
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.2474666666666667,
|
| 2058 |
+
"grad_norm": 5.170231819152832,
|
| 2059 |
+
"learning_rate": 9.982089435684297e-06,
|
| 2060 |
+
"loss": 0.1272,
|
| 2061 |
+
"step": 293
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.2517333333333334,
|
| 2065 |
+
"grad_norm": 3.8409669399261475,
|
| 2066 |
+
"learning_rate": 9.9814558743485e-06,
|
| 2067 |
+
"loss": 0.1191,
|
| 2068 |
+
"step": 294
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.256,
|
| 2072 |
+
"grad_norm": 4.199260234832764,
|
| 2073 |
+
"learning_rate": 9.980811322063268e-06,
|
| 2074 |
+
"loss": 0.1201,
|
| 2075 |
+
"step": 295
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.2602666666666666,
|
| 2079 |
+
"grad_norm": 2.8783466815948486,
|
| 2080 |
+
"learning_rate": 9.980155780250728e-06,
|
| 2081 |
+
"loss": 0.1196,
|
| 2082 |
+
"step": 296
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.2645333333333333,
|
| 2086 |
+
"grad_norm": 5.591996192932129,
|
| 2087 |
+
"learning_rate": 9.979489250357243e-06,
|
| 2088 |
+
"loss": 0.1234,
|
| 2089 |
+
"step": 297
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.2688,
|
| 2093 |
+
"grad_norm": 3.9676425457000732,
|
| 2094 |
+
"learning_rate": 9.978811733853433e-06,
|
| 2095 |
+
"loss": 0.1242,
|
| 2096 |
+
"step": 298
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.2730666666666668,
|
| 2100 |
+
"grad_norm": 5.0102057456970215,
|
| 2101 |
+
"learning_rate": 9.978123232234147e-06,
|
| 2102 |
+
"loss": 0.1207,
|
| 2103 |
+
"step": 299
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.2773333333333334,
|
| 2107 |
+
"grad_norm": 4.265115737915039,
|
| 2108 |
+
"learning_rate": 9.97742374701848e-06,
|
| 2109 |
+
"loss": 0.1209,
|
| 2110 |
+
"step": 300
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.2816,
|
| 2114 |
+
"grad_norm": 6.160897254943848,
|
| 2115 |
+
"learning_rate": 9.976713279749754e-06,
|
| 2116 |
+
"loss": 0.1254,
|
| 2117 |
+
"step": 301
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.2858666666666667,
|
| 2121 |
+
"grad_norm": 4.771708965301514,
|
| 2122 |
+
"learning_rate": 9.975991831995528e-06,
|
| 2123 |
+
"loss": 0.1237,
|
| 2124 |
+
"step": 302
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.2901333333333334,
|
| 2128 |
+
"grad_norm": 6.948702335357666,
|
| 2129 |
+
"learning_rate": 9.975259405347583e-06,
|
| 2130 |
+
"loss": 0.1192,
|
| 2131 |
+
"step": 303
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.2944,
|
| 2135 |
+
"grad_norm": 5.352514743804932,
|
| 2136 |
+
"learning_rate": 9.974516001421926e-06,
|
| 2137 |
+
"loss": 0.1215,
|
| 2138 |
+
"step": 304
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.2986666666666666,
|
| 2142 |
+
"grad_norm": 5.381694793701172,
|
| 2143 |
+
"learning_rate": 9.97376162185878e-06,
|
| 2144 |
+
"loss": 0.1187,
|
| 2145 |
+
"step": 305
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.3029333333333333,
|
| 2149 |
+
"grad_norm": 5.543086528778076,
|
| 2150 |
+
"learning_rate": 9.972996268322594e-06,
|
| 2151 |
+
"loss": 0.1153,
|
| 2152 |
+
"step": 306
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.3072,
|
| 2156 |
+
"grad_norm": 3.0159881114959717,
|
| 2157 |
+
"learning_rate": 9.972219942502017e-06,
|
| 2158 |
+
"loss": 0.1154,
|
| 2159 |
+
"step": 307
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.3114666666666666,
|
| 2163 |
+
"grad_norm": 5.667745590209961,
|
| 2164 |
+
"learning_rate": 9.971432646109919e-06,
|
| 2165 |
+
"loss": 0.117,
|
| 2166 |
+
"step": 308
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.3157333333333332,
|
| 2170 |
+
"grad_norm": 3.8900418281555176,
|
| 2171 |
+
"learning_rate": 9.970634380883364e-06,
|
| 2172 |
+
"loss": 0.1143,
|
| 2173 |
+
"step": 309
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.32,
|
| 2177 |
+
"grad_norm": 3.812140941619873,
|
| 2178 |
+
"learning_rate": 9.969825148583627e-06,
|
| 2179 |
+
"loss": 0.1135,
|
| 2180 |
+
"step": 310
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.3242666666666667,
|
| 2184 |
+
"grad_norm": 4.56724739074707,
|
| 2185 |
+
"learning_rate": 9.969004950996175e-06,
|
| 2186 |
+
"loss": 0.1133,
|
| 2187 |
+
"step": 311
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.3285333333333333,
|
| 2191 |
+
"grad_norm": 5.430902481079102,
|
| 2192 |
+
"learning_rate": 9.968173789930668e-06,
|
| 2193 |
+
"loss": 0.112,
|
| 2194 |
+
"step": 312
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.3328,
|
| 2198 |
+
"grad_norm": 3.464324712753296,
|
| 2199 |
+
"learning_rate": 9.96733166722096e-06,
|
| 2200 |
+
"loss": 0.1123,
|
| 2201 |
+
"step": 313
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.3370666666666666,
|
| 2205 |
+
"grad_norm": 4.844167709350586,
|
| 2206 |
+
"learning_rate": 9.966478584725085e-06,
|
| 2207 |
+
"loss": 0.1125,
|
| 2208 |
+
"step": 314
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.3413333333333333,
|
| 2212 |
+
"grad_norm": 4.899601459503174,
|
| 2213 |
+
"learning_rate": 9.965614544325264e-06,
|
| 2214 |
+
"loss": 0.1126,
|
| 2215 |
+
"step": 315
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 1.3456000000000001,
|
| 2219 |
+
"grad_norm": 4.0164079666137695,
|
| 2220 |
+
"learning_rate": 9.964739547927892e-06,
|
| 2221 |
+
"loss": 0.1117,
|
| 2222 |
+
"step": 316
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.3498666666666668,
|
| 2226 |
+
"grad_norm": 3.449429988861084,
|
| 2227 |
+
"learning_rate": 9.963853597463533e-06,
|
| 2228 |
+
"loss": 0.1101,
|
| 2229 |
+
"step": 317
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.3541333333333334,
|
| 2233 |
+
"grad_norm": 6.577143669128418,
|
| 2234 |
+
"learning_rate": 9.96295669488693e-06,
|
| 2235 |
+
"loss": 0.1135,
|
| 2236 |
+
"step": 318
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.3584,
|
| 2240 |
+
"grad_norm": 2.9539217948913574,
|
| 2241 |
+
"learning_rate": 9.96204884217698e-06,
|
| 2242 |
+
"loss": 0.1159,
|
| 2243 |
+
"step": 319
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.3626666666666667,
|
| 2247 |
+
"grad_norm": 6.194974899291992,
|
| 2248 |
+
"learning_rate": 9.961130041336748e-06,
|
| 2249 |
+
"loss": 0.1212,
|
| 2250 |
+
"step": 320
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.3669333333333333,
|
| 2254 |
+
"grad_norm": 4.237511157989502,
|
| 2255 |
+
"learning_rate": 9.96020029439345e-06,
|
| 2256 |
+
"loss": 0.1217,
|
| 2257 |
+
"step": 321
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.3712,
|
| 2261 |
+
"grad_norm": 5.736575603485107,
|
| 2262 |
+
"learning_rate": 9.959259603398454e-06,
|
| 2263 |
+
"loss": 0.1219,
|
| 2264 |
+
"step": 322
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.3754666666666666,
|
| 2268 |
+
"grad_norm": 4.737814426422119,
|
| 2269 |
+
"learning_rate": 9.958307970427276e-06,
|
| 2270 |
+
"loss": 0.1215,
|
| 2271 |
+
"step": 323
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.3797333333333333,
|
| 2275 |
+
"grad_norm": 5.155465602874756,
|
| 2276 |
+
"learning_rate": 9.957345397579573e-06,
|
| 2277 |
+
"loss": 0.1186,
|
| 2278 |
+
"step": 324
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.384,
|
| 2282 |
+
"grad_norm": 5.585412979125977,
|
| 2283 |
+
"learning_rate": 9.956371886979139e-06,
|
| 2284 |
+
"loss": 0.1217,
|
| 2285 |
+
"step": 325
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.3882666666666665,
|
| 2289 |
+
"grad_norm": 4.1086506843566895,
|
| 2290 |
+
"learning_rate": 9.955387440773902e-06,
|
| 2291 |
+
"loss": 0.1164,
|
| 2292 |
+
"step": 326
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.3925333333333334,
|
| 2296 |
+
"grad_norm": 3.985306978225708,
|
| 2297 |
+
"learning_rate": 9.954392061135916e-06,
|
| 2298 |
+
"loss": 0.1163,
|
| 2299 |
+
"step": 327
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.3968,
|
| 2303 |
+
"grad_norm": 4.945534706115723,
|
| 2304 |
+
"learning_rate": 9.953385750261363e-06,
|
| 2305 |
+
"loss": 0.1151,
|
| 2306 |
+
"step": 328
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.4010666666666667,
|
| 2310 |
+
"grad_norm": 2.9012649059295654,
|
| 2311 |
+
"learning_rate": 9.952368510370538e-06,
|
| 2312 |
+
"loss": 0.1141,
|
| 2313 |
+
"step": 329
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.4053333333333333,
|
| 2317 |
+
"grad_norm": 7.978700160980225,
|
| 2318 |
+
"learning_rate": 9.951340343707852e-06,
|
| 2319 |
+
"loss": 0.119,
|
| 2320 |
+
"step": 330
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.4096,
|
| 2324 |
+
"grad_norm": 5.617311954498291,
|
| 2325 |
+
"learning_rate": 9.950301252541824e-06,
|
| 2326 |
+
"loss": 0.1181,
|
| 2327 |
+
"step": 331
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.4138666666666666,
|
| 2331 |
+
"grad_norm": 7.305325031280518,
|
| 2332 |
+
"learning_rate": 9.949251239165076e-06,
|
| 2333 |
+
"loss": 0.1164,
|
| 2334 |
+
"step": 332
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.4181333333333335,
|
| 2338 |
+
"grad_norm": 6.810077667236328,
|
| 2339 |
+
"learning_rate": 9.94819030589433e-06,
|
| 2340 |
+
"loss": 0.1171,
|
| 2341 |
+
"step": 333
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.4224,
|
| 2345 |
+
"grad_norm": 4.517914772033691,
|
| 2346 |
+
"learning_rate": 9.9471184550704e-06,
|
| 2347 |
+
"loss": 0.1141,
|
| 2348 |
+
"step": 334
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.4266666666666667,
|
| 2352 |
+
"grad_norm": 7.977745056152344,
|
| 2353 |
+
"learning_rate": 9.946035689058189e-06,
|
| 2354 |
+
"loss": 0.1185,
|
| 2355 |
+
"step": 335
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 1.4309333333333334,
|
| 2359 |
+
"grad_norm": 5.102802753448486,
|
| 2360 |
+
"learning_rate": 9.944942010246681e-06,
|
| 2361 |
+
"loss": 0.1179,
|
| 2362 |
+
"step": 336
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.4352,
|
| 2366 |
+
"grad_norm": 8.595138549804688,
|
| 2367 |
+
"learning_rate": 9.943837421048942e-06,
|
| 2368 |
+
"loss": 0.1179,
|
| 2369 |
+
"step": 337
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.4394666666666667,
|
| 2373 |
+
"grad_norm": 6.408101558685303,
|
| 2374 |
+
"learning_rate": 9.942721923902105e-06,
|
| 2375 |
+
"loss": 0.1209,
|
| 2376 |
+
"step": 338
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.4437333333333333,
|
| 2380 |
+
"grad_norm": 7.459262847900391,
|
| 2381 |
+
"learning_rate": 9.941595521267377e-06,
|
| 2382 |
+
"loss": 0.1151,
|
| 2383 |
+
"step": 339
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.448,
|
| 2387 |
+
"grad_norm": 4.301973819732666,
|
| 2388 |
+
"learning_rate": 9.940458215630017e-06,
|
| 2389 |
+
"loss": 0.1146,
|
| 2390 |
+
"step": 340
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.4522666666666666,
|
| 2394 |
+
"grad_norm": 8.624367713928223,
|
| 2395 |
+
"learning_rate": 9.939310009499348e-06,
|
| 2396 |
+
"loss": 0.1188,
|
| 2397 |
+
"step": 341
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.4565333333333332,
|
| 2401 |
+
"grad_norm": 6.233131408691406,
|
| 2402 |
+
"learning_rate": 9.938150905408741e-06,
|
| 2403 |
+
"loss": 0.1186,
|
| 2404 |
+
"step": 342
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.4607999999999999,
|
| 2408 |
+
"grad_norm": 7.494032859802246,
|
| 2409 |
+
"learning_rate": 9.936980905915612e-06,
|
| 2410 |
+
"loss": 0.1173,
|
| 2411 |
+
"step": 343
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.4650666666666667,
|
| 2415 |
+
"grad_norm": 7.334600925445557,
|
| 2416 |
+
"learning_rate": 9.935800013601415e-06,
|
| 2417 |
+
"loss": 0.1163,
|
| 2418 |
+
"step": 344
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.4693333333333334,
|
| 2422 |
+
"grad_norm": 4.524255275726318,
|
| 2423 |
+
"learning_rate": 9.93460823107164e-06,
|
| 2424 |
+
"loss": 0.1134,
|
| 2425 |
+
"step": 345
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.4736,
|
| 2429 |
+
"grad_norm": 4.376001834869385,
|
| 2430 |
+
"learning_rate": 9.933405560955805e-06,
|
| 2431 |
+
"loss": 0.1119,
|
| 2432 |
+
"step": 346
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.4778666666666667,
|
| 2436 |
+
"grad_norm": 5.788750171661377,
|
| 2437 |
+
"learning_rate": 9.932192005907446e-06,
|
| 2438 |
+
"loss": 0.113,
|
| 2439 |
+
"step": 347
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.4821333333333333,
|
| 2443 |
+
"grad_norm": 3.970564603805542,
|
| 2444 |
+
"learning_rate": 9.930967568604119e-06,
|
| 2445 |
+
"loss": 0.1131,
|
| 2446 |
+
"step": 348
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.4864,
|
| 2450 |
+
"grad_norm": 7.438868045806885,
|
| 2451 |
+
"learning_rate": 9.929732251747389e-06,
|
| 2452 |
+
"loss": 0.1178,
|
| 2453 |
+
"step": 349
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 1.4906666666666666,
|
| 2457 |
+
"grad_norm": 6.751838207244873,
|
| 2458 |
+
"learning_rate": 9.928486058062828e-06,
|
| 2459 |
+
"loss": 0.117,
|
| 2460 |
+
"step": 350
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 1.4949333333333334,
|
| 2464 |
+
"grad_norm": 4.1995625495910645,
|
| 2465 |
+
"learning_rate": 9.9272289903e-06,
|
| 2466 |
+
"loss": 0.1124,
|
| 2467 |
+
"step": 351
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 1.4992,
|
| 2471 |
+
"grad_norm": 4.800899028778076,
|
| 2472 |
+
"learning_rate": 9.925961051232469e-06,
|
| 2473 |
+
"loss": 0.1129,
|
| 2474 |
+
"step": 352
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.5034666666666667,
|
| 2478 |
+
"grad_norm": 4.623241424560547,
|
| 2479 |
+
"learning_rate": 9.92468224365778e-06,
|
| 2480 |
+
"loss": 0.1117,
|
| 2481 |
+
"step": 353
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.5077333333333334,
|
| 2485 |
+
"grad_norm": 3.0356364250183105,
|
| 2486 |
+
"learning_rate": 9.923392570397459e-06,
|
| 2487 |
+
"loss": 0.1117,
|
| 2488 |
+
"step": 354
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.512,
|
| 2492 |
+
"grad_norm": 4.9218974113464355,
|
| 2493 |
+
"learning_rate": 9.922092034297007e-06,
|
| 2494 |
+
"loss": 0.1128,
|
| 2495 |
+
"step": 355
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 1.5162666666666667,
|
| 2499 |
+
"grad_norm": 3.4157235622406006,
|
| 2500 |
+
"learning_rate": 9.92078063822589e-06,
|
| 2501 |
+
"loss": 0.112,
|
| 2502 |
+
"step": 356
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 1.5205333333333333,
|
| 2506 |
+
"grad_norm": 5.657739162445068,
|
| 2507 |
+
"learning_rate": 9.91945838507754e-06,
|
| 2508 |
+
"loss": 0.1158,
|
| 2509 |
+
"step": 357
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 1.5248,
|
| 2513 |
+
"grad_norm": 5.096992492675781,
|
| 2514 |
+
"learning_rate": 9.918125277769336e-06,
|
| 2515 |
+
"loss": 0.1151,
|
| 2516 |
+
"step": 358
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.5290666666666666,
|
| 2520 |
+
"grad_norm": 4.734887599945068,
|
| 2521 |
+
"learning_rate": 9.916781319242616e-06,
|
| 2522 |
+
"loss": 0.1118,
|
| 2523 |
+
"step": 359
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 1.5333333333333332,
|
| 2527 |
+
"grad_norm": 5.02221155166626,
|
| 2528 |
+
"learning_rate": 9.915426512462647e-06,
|
| 2529 |
+
"loss": 0.1114,
|
| 2530 |
+
"step": 360
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 1.5375999999999999,
|
| 2534 |
+
"grad_norm": 4.93018913269043,
|
| 2535 |
+
"learning_rate": 9.914060860418644e-06,
|
| 2536 |
+
"loss": 0.1118,
|
| 2537 |
+
"step": 361
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 1.5418666666666667,
|
| 2541 |
+
"grad_norm": 5.088624954223633,
|
| 2542 |
+
"learning_rate": 9.91268436612374e-06,
|
| 2543 |
+
"loss": 0.1105,
|
| 2544 |
+
"step": 362
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 1.5461333333333334,
|
| 2548 |
+
"grad_norm": 4.798462867736816,
|
| 2549 |
+
"learning_rate": 9.911297032614996e-06,
|
| 2550 |
+
"loss": 0.1095,
|
| 2551 |
+
"step": 363
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 1.5504,
|
| 2555 |
+
"grad_norm": 5.95226526260376,
|
| 2556 |
+
"learning_rate": 9.909898862953392e-06,
|
| 2557 |
+
"loss": 0.112,
|
| 2558 |
+
"step": 364
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 1.5546666666666666,
|
| 2562 |
+
"grad_norm": 6.883795738220215,
|
| 2563 |
+
"learning_rate": 9.908489860223805e-06,
|
| 2564 |
+
"loss": 0.1107,
|
| 2565 |
+
"step": 365
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 1.5589333333333333,
|
| 2569 |
+
"grad_norm": 3.146536111831665,
|
| 2570 |
+
"learning_rate": 9.907070027535022e-06,
|
| 2571 |
+
"loss": 0.108,
|
| 2572 |
+
"step": 366
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 1.5632000000000001,
|
| 2576 |
+
"grad_norm": 4.579812526702881,
|
| 2577 |
+
"learning_rate": 9.905639368019724e-06,
|
| 2578 |
+
"loss": 0.1104,
|
| 2579 |
+
"step": 367
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 1.5674666666666668,
|
| 2583 |
+
"grad_norm": 5.706833839416504,
|
| 2584 |
+
"learning_rate": 9.904197884834482e-06,
|
| 2585 |
+
"loss": 0.11,
|
| 2586 |
+
"step": 368
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 1.5717333333333334,
|
| 2590 |
+
"grad_norm": 2.8592660427093506,
|
| 2591 |
+
"learning_rate": 9.902745581159743e-06,
|
| 2592 |
+
"loss": 0.1159,
|
| 2593 |
+
"step": 369
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 1.576,
|
| 2597 |
+
"grad_norm": 12.39868450164795,
|
| 2598 |
+
"learning_rate": 9.901282460199831e-06,
|
| 2599 |
+
"loss": 0.1345,
|
| 2600 |
+
"step": 370
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 1.5802666666666667,
|
| 2604 |
+
"grad_norm": 11.114265441894531,
|
| 2605 |
+
"learning_rate": 9.899808525182935e-06,
|
| 2606 |
+
"loss": 0.1331,
|
| 2607 |
+
"step": 371
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 1.5845333333333333,
|
| 2611 |
+
"grad_norm": 9.193613052368164,
|
| 2612 |
+
"learning_rate": 9.898323779361108e-06,
|
| 2613 |
+
"loss": 0.1213,
|
| 2614 |
+
"step": 372
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 1.5888,
|
| 2618 |
+
"grad_norm": 9.077799797058105,
|
| 2619 |
+
"learning_rate": 9.89682822601025e-06,
|
| 2620 |
+
"loss": 0.1196,
|
| 2621 |
+
"step": 373
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 1.5930666666666666,
|
| 2625 |
+
"grad_norm": 9.783404350280762,
|
| 2626 |
+
"learning_rate": 9.895321868430113e-06,
|
| 2627 |
+
"loss": 0.1199,
|
| 2628 |
+
"step": 374
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 1.5973333333333333,
|
| 2632 |
+
"grad_norm": 4.7927141189575195,
|
| 2633 |
+
"learning_rate": 9.893804709944281e-06,
|
| 2634 |
+
"loss": 0.1153,
|
| 2635 |
+
"step": 375
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 1.6016,
|
| 2639 |
+
"grad_norm": 7.528342247009277,
|
| 2640 |
+
"learning_rate": 9.892276753900173e-06,
|
| 2641 |
+
"loss": 0.1193,
|
| 2642 |
+
"step": 376
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 1.6058666666666666,
|
| 2646 |
+
"grad_norm": 5.356241226196289,
|
| 2647 |
+
"learning_rate": 9.890738003669029e-06,
|
| 2648 |
+
"loss": 0.118,
|
| 2649 |
+
"step": 377
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 1.6101333333333332,
|
| 2653 |
+
"grad_norm": 4.04868745803833,
|
| 2654 |
+
"learning_rate": 9.889188462645905e-06,
|
| 2655 |
+
"loss": 0.116,
|
| 2656 |
+
"step": 378
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 1.6143999999999998,
|
| 2660 |
+
"grad_norm": 6.1326751708984375,
|
| 2661 |
+
"learning_rate": 9.887628134249669e-06,
|
| 2662 |
+
"loss": 0.1173,
|
| 2663 |
+
"step": 379
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 1.6186666666666667,
|
| 2667 |
+
"grad_norm": 4.443249702453613,
|
| 2668 |
+
"learning_rate": 9.886057021922984e-06,
|
| 2669 |
+
"loss": 0.116,
|
| 2670 |
+
"step": 380
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 1.6229333333333333,
|
| 2674 |
+
"grad_norm": 3.5723698139190674,
|
| 2675 |
+
"learning_rate": 9.884475129132312e-06,
|
| 2676 |
+
"loss": 0.1128,
|
| 2677 |
+
"step": 381
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 1.6272,
|
| 2681 |
+
"grad_norm": 6.835819244384766,
|
| 2682 |
+
"learning_rate": 9.882882459367898e-06,
|
| 2683 |
+
"loss": 0.1169,
|
| 2684 |
+
"step": 382
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.6314666666666666,
|
| 2688 |
+
"grad_norm": 5.1200852394104,
|
| 2689 |
+
"learning_rate": 9.881279016143766e-06,
|
| 2690 |
+
"loss": 0.1165,
|
| 2691 |
+
"step": 383
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 1.6357333333333335,
|
| 2695 |
+
"grad_norm": 4.170957565307617,
|
| 2696 |
+
"learning_rate": 9.879664802997708e-06,
|
| 2697 |
+
"loss": 0.1135,
|
| 2698 |
+
"step": 384
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 1.6400000000000001,
|
| 2702 |
+
"grad_norm": 9.073988914489746,
|
| 2703 |
+
"learning_rate": 9.878039823491282e-06,
|
| 2704 |
+
"loss": 0.1207,
|
| 2705 |
+
"step": 385
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 1.6442666666666668,
|
| 2709 |
+
"grad_norm": 5.946852684020996,
|
| 2710 |
+
"learning_rate": 9.876404081209796e-06,
|
| 2711 |
+
"loss": 0.1226,
|
| 2712 |
+
"step": 386
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 1.6485333333333334,
|
| 2716 |
+
"grad_norm": 10.303154945373535,
|
| 2717 |
+
"learning_rate": 9.87475757976231e-06,
|
| 2718 |
+
"loss": 0.1223,
|
| 2719 |
+
"step": 387
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 1.6528,
|
| 2723 |
+
"grad_norm": 9.911062240600586,
|
| 2724 |
+
"learning_rate": 9.873100322781622e-06,
|
| 2725 |
+
"loss": 0.1212,
|
| 2726 |
+
"step": 388
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 1.6570666666666667,
|
| 2730 |
+
"grad_norm": 4.888733863830566,
|
| 2731 |
+
"learning_rate": 9.871432313924255e-06,
|
| 2732 |
+
"loss": 0.115,
|
| 2733 |
+
"step": 389
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 1.6613333333333333,
|
| 2737 |
+
"grad_norm": 5.558224678039551,
|
| 2738 |
+
"learning_rate": 9.86975355687046e-06,
|
| 2739 |
+
"loss": 0.115,
|
| 2740 |
+
"step": 390
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 1.6656,
|
| 2744 |
+
"grad_norm": 5.1496429443359375,
|
| 2745 |
+
"learning_rate": 9.868064055324204e-06,
|
| 2746 |
+
"loss": 0.1142,
|
| 2747 |
+
"step": 391
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 1.6698666666666666,
|
| 2751 |
+
"grad_norm": 3.196533203125,
|
| 2752 |
+
"learning_rate": 9.866363813013153e-06,
|
| 2753 |
+
"loss": 0.1141,
|
| 2754 |
+
"step": 392
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 1.6741333333333333,
|
| 2758 |
+
"grad_norm": 5.469643592834473,
|
| 2759 |
+
"learning_rate": 9.864652833688678e-06,
|
| 2760 |
+
"loss": 0.1142,
|
| 2761 |
+
"step": 393
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 1.6784,
|
| 2765 |
+
"grad_norm": 3.354064702987671,
|
| 2766 |
+
"learning_rate": 9.862931121125837e-06,
|
| 2767 |
+
"loss": 0.1143,
|
| 2768 |
+
"step": 394
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 1.6826666666666665,
|
| 2772 |
+
"grad_norm": 5.456864356994629,
|
| 2773 |
+
"learning_rate": 9.86119867912337e-06,
|
| 2774 |
+
"loss": 0.1163,
|
| 2775 |
+
"step": 395
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 1.6869333333333332,
|
| 2779 |
+
"grad_norm": 4.276975154876709,
|
| 2780 |
+
"learning_rate": 9.859455511503691e-06,
|
| 2781 |
+
"loss": 0.1159,
|
| 2782 |
+
"step": 396
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 1.6912,
|
| 2786 |
+
"grad_norm": 5.062138080596924,
|
| 2787 |
+
"learning_rate": 9.857701622112877e-06,
|
| 2788 |
+
"loss": 0.1141,
|
| 2789 |
+
"step": 397
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 1.6954666666666667,
|
| 2793 |
+
"grad_norm": 4.016213417053223,
|
| 2794 |
+
"learning_rate": 9.85593701482066e-06,
|
| 2795 |
+
"loss": 0.1129,
|
| 2796 |
+
"step": 398
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 1.6997333333333333,
|
| 2800 |
+
"grad_norm": 4.987476825714111,
|
| 2801 |
+
"learning_rate": 9.854161693520426e-06,
|
| 2802 |
+
"loss": 0.1142,
|
| 2803 |
+
"step": 399
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 1.704,
|
| 2807 |
+
"grad_norm": 3.9129505157470703,
|
| 2808 |
+
"learning_rate": 9.852375662129193e-06,
|
| 2809 |
+
"loss": 0.1137,
|
| 2810 |
+
"step": 400
|
| 2811 |
+
}
|
| 2812 |
+
],
|
| 2813 |
+
"logging_steps": 1,
|
| 2814 |
+
"max_steps": 2350,
|
| 2815 |
+
"num_input_tokens_seen": 0,
|
| 2816 |
+
"num_train_epochs": 10,
|
| 2817 |
+
"save_steps": 100,
|
| 2818 |
+
"stateful_callbacks": {
|
| 2819 |
+
"TrainerControl": {
|
| 2820 |
+
"args": {
|
| 2821 |
+
"should_epoch_stop": false,
|
| 2822 |
+
"should_evaluate": false,
|
| 2823 |
+
"should_log": false,
|
| 2824 |
+
"should_save": true,
|
| 2825 |
+
"should_training_stop": false
|
| 2826 |
+
},
|
| 2827 |
+
"attributes": {}
|
| 2828 |
+
}
|
| 2829 |
+
},
|
| 2830 |
+
"total_flos": 1.0522802641229054e+19,
|
| 2831 |
+
"train_batch_size": 8,
|
| 2832 |
+
"trial_name": null,
|
| 2833 |
+
"trial_params": null
|
| 2834 |
+
}
|
400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d76a75bebf3491ba6c0223c474ab435c114b060370b046e532bc0adfccfd23c6
|
| 3 |
+
size 7761
|
400/video_preprocessor_config.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_kwargs_names": [
|
| 3 |
+
"do_convert_rgb",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"size_divisor",
|
| 7 |
+
"default_to_square",
|
| 8 |
+
"resample",
|
| 9 |
+
"do_rescale",
|
| 10 |
+
"rescale_factor",
|
| 11 |
+
"do_normalize",
|
| 12 |
+
"image_mean",
|
| 13 |
+
"image_std",
|
| 14 |
+
"do_pad",
|
| 15 |
+
"do_center_crop",
|
| 16 |
+
"crop_size",
|
| 17 |
+
"data_format",
|
| 18 |
+
"input_data_format",
|
| 19 |
+
"device",
|
| 20 |
+
"min_pixels",
|
| 21 |
+
"max_pixels",
|
| 22 |
+
"patch_size",
|
| 23 |
+
"temporal_patch_size",
|
| 24 |
+
"merge_size"
|
| 25 |
+
],
|
| 26 |
+
"crop_size": null,
|
| 27 |
+
"data_format": "channels_first",
|
| 28 |
+
"default_to_square": true,
|
| 29 |
+
"device": null,
|
| 30 |
+
"do_center_crop": null,
|
| 31 |
+
"do_convert_rgb": true,
|
| 32 |
+
"do_normalize": true,
|
| 33 |
+
"do_pad": null,
|
| 34 |
+
"do_rescale": true,
|
| 35 |
+
"do_resize": true,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.48145466,
|
| 38 |
+
0.4578275,
|
| 39 |
+
0.40821073
|
| 40 |
+
],
|
| 41 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 42 |
+
"image_std": [
|
| 43 |
+
0.26862954,
|
| 44 |
+
0.26130258,
|
| 45 |
+
0.27577711
|
| 46 |
+
],
|
| 47 |
+
"input_data_format": null,
|
| 48 |
+
"max_pixels": 12845056,
|
| 49 |
+
"merge_size": 2,
|
| 50 |
+
"min_pixels": 3136,
|
| 51 |
+
"model_valid_processing_keys": [
|
| 52 |
+
"do_convert_rgb",
|
| 53 |
+
"do_resize",
|
| 54 |
+
"size",
|
| 55 |
+
"size_divisor",
|
| 56 |
+
"default_to_square",
|
| 57 |
+
"resample",
|
| 58 |
+
"do_rescale",
|
| 59 |
+
"rescale_factor",
|
| 60 |
+
"do_normalize",
|
| 61 |
+
"image_mean",
|
| 62 |
+
"image_std",
|
| 63 |
+
"do_pad",
|
| 64 |
+
"do_center_crop",
|
| 65 |
+
"crop_size",
|
| 66 |
+
"data_format",
|
| 67 |
+
"input_data_format",
|
| 68 |
+
"device",
|
| 69 |
+
"min_pixels",
|
| 70 |
+
"max_pixels",
|
| 71 |
+
"patch_size",
|
| 72 |
+
"temporal_patch_size",
|
| 73 |
+
"merge_size"
|
| 74 |
+
],
|
| 75 |
+
"patch_size": 14,
|
| 76 |
+
"processor_class": "Qwen2VLProcessor",
|
| 77 |
+
"resample": 3,
|
| 78 |
+
"rescale_factor": 0.00392156862745098,
|
| 79 |
+
"size": {
|
| 80 |
+
"longest_edge": 12845056,
|
| 81 |
+
"shortest_edge": 3136
|
| 82 |
+
},
|
| 83 |
+
"size_divisor": null,
|
| 84 |
+
"temporal_patch_size": 2,
|
| 85 |
+
"video_processor_type": "Qwen2VLVideoProcessor"
|
| 86 |
+
}
|
400/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|