Upload 15 files
Browse files- added_tokens.json +6 -0
- config.json +199 -0
- generation_config.json +15 -0
- merges.txt +0 -0
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- model.safetensors.index.json +0 -0
- special_tokens_map.json +20 -0
- tokenizer.json +0 -0
- tokenizer_config.json +53 -0
- trainer_state.json +1883 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image>": 151646,
|
| 3 |
+
"<|endoftext|>": 151643,
|
| 4 |
+
"<|im_end|>": 151645,
|
| 5 |
+
"<|im_start|>": 151644
|
| 6 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,199 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/workspace/LLaVA-NeXT-main/llava-ov-chat-qwen2-stage1-stage2-UGC-full",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaQwenForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 3584,
|
| 11 |
+
"image_aspect_ratio": "anyres_max_9",
|
| 12 |
+
"image_crop_resolution": null,
|
| 13 |
+
"image_grid_pinpoints": [
|
| 14 |
+
[
|
| 15 |
+
384,
|
| 16 |
+
384
|
| 17 |
+
],
|
| 18 |
+
[
|
| 19 |
+
384,
|
| 20 |
+
768
|
| 21 |
+
],
|
| 22 |
+
[
|
| 23 |
+
384,
|
| 24 |
+
1152
|
| 25 |
+
],
|
| 26 |
+
[
|
| 27 |
+
384,
|
| 28 |
+
1536
|
| 29 |
+
],
|
| 30 |
+
[
|
| 31 |
+
384,
|
| 32 |
+
1920
|
| 33 |
+
],
|
| 34 |
+
[
|
| 35 |
+
384,
|
| 36 |
+
2304
|
| 37 |
+
],
|
| 38 |
+
[
|
| 39 |
+
768,
|
| 40 |
+
384
|
| 41 |
+
],
|
| 42 |
+
[
|
| 43 |
+
768,
|
| 44 |
+
768
|
| 45 |
+
],
|
| 46 |
+
[
|
| 47 |
+
768,
|
| 48 |
+
1152
|
| 49 |
+
],
|
| 50 |
+
[
|
| 51 |
+
768,
|
| 52 |
+
1536
|
| 53 |
+
],
|
| 54 |
+
[
|
| 55 |
+
768,
|
| 56 |
+
1920
|
| 57 |
+
],
|
| 58 |
+
[
|
| 59 |
+
768,
|
| 60 |
+
2304
|
| 61 |
+
],
|
| 62 |
+
[
|
| 63 |
+
1152,
|
| 64 |
+
384
|
| 65 |
+
],
|
| 66 |
+
[
|
| 67 |
+
1152,
|
| 68 |
+
768
|
| 69 |
+
],
|
| 70 |
+
[
|
| 71 |
+
1152,
|
| 72 |
+
1152
|
| 73 |
+
],
|
| 74 |
+
[
|
| 75 |
+
1152,
|
| 76 |
+
1536
|
| 77 |
+
],
|
| 78 |
+
[
|
| 79 |
+
1152,
|
| 80 |
+
1920
|
| 81 |
+
],
|
| 82 |
+
[
|
| 83 |
+
1152,
|
| 84 |
+
2304
|
| 85 |
+
],
|
| 86 |
+
[
|
| 87 |
+
1536,
|
| 88 |
+
384
|
| 89 |
+
],
|
| 90 |
+
[
|
| 91 |
+
1536,
|
| 92 |
+
768
|
| 93 |
+
],
|
| 94 |
+
[
|
| 95 |
+
1536,
|
| 96 |
+
1152
|
| 97 |
+
],
|
| 98 |
+
[
|
| 99 |
+
1536,
|
| 100 |
+
1536
|
| 101 |
+
],
|
| 102 |
+
[
|
| 103 |
+
1536,
|
| 104 |
+
1920
|
| 105 |
+
],
|
| 106 |
+
[
|
| 107 |
+
1536,
|
| 108 |
+
2304
|
| 109 |
+
],
|
| 110 |
+
[
|
| 111 |
+
1920,
|
| 112 |
+
384
|
| 113 |
+
],
|
| 114 |
+
[
|
| 115 |
+
1920,
|
| 116 |
+
768
|
| 117 |
+
],
|
| 118 |
+
[
|
| 119 |
+
1920,
|
| 120 |
+
1152
|
| 121 |
+
],
|
| 122 |
+
[
|
| 123 |
+
1920,
|
| 124 |
+
1536
|
| 125 |
+
],
|
| 126 |
+
[
|
| 127 |
+
1920,
|
| 128 |
+
1920
|
| 129 |
+
],
|
| 130 |
+
[
|
| 131 |
+
1920,
|
| 132 |
+
2304
|
| 133 |
+
],
|
| 134 |
+
[
|
| 135 |
+
2304,
|
| 136 |
+
384
|
| 137 |
+
],
|
| 138 |
+
[
|
| 139 |
+
2304,
|
| 140 |
+
768
|
| 141 |
+
],
|
| 142 |
+
[
|
| 143 |
+
2304,
|
| 144 |
+
1152
|
| 145 |
+
],
|
| 146 |
+
[
|
| 147 |
+
2304,
|
| 148 |
+
1536
|
| 149 |
+
],
|
| 150 |
+
[
|
| 151 |
+
2304,
|
| 152 |
+
1920
|
| 153 |
+
],
|
| 154 |
+
[
|
| 155 |
+
2304,
|
| 156 |
+
2304
|
| 157 |
+
]
|
| 158 |
+
],
|
| 159 |
+
"image_split_resolution": null,
|
| 160 |
+
"image_token_index": 151646,
|
| 161 |
+
"initializer_range": 0.02,
|
| 162 |
+
"intermediate_size": 18944,
|
| 163 |
+
"max_position_embeddings": 32768,
|
| 164 |
+
"max_window_layers": 28,
|
| 165 |
+
"mm_hidden_size": 1152,
|
| 166 |
+
"mm_newline_position": "one_token",
|
| 167 |
+
"mm_patch_merge_type": "spatial_unpad",
|
| 168 |
+
"mm_projector_lr": null,
|
| 169 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 170 |
+
"mm_resampler_type": null,
|
| 171 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 172 |
+
"mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_slowfast,mm_slowfast_projector,mm_language_model",
|
| 173 |
+
"mm_use_im_patch_token": false,
|
| 174 |
+
"mm_use_im_start_end": false,
|
| 175 |
+
"mm_vision_select_feature": "patch",
|
| 176 |
+
"mm_vision_select_layer": -2,
|
| 177 |
+
"mm_vision_tower": "google/siglip-so400m-patch14-384",
|
| 178 |
+
"mm_vision_tower_lr": 2e-06,
|
| 179 |
+
"model_type": "llava_qwen",
|
| 180 |
+
"num_attention_heads": 28,
|
| 181 |
+
"num_hidden_layers": 28,
|
| 182 |
+
"num_key_value_heads": 4,
|
| 183 |
+
"pos_skipping_range": 4096,
|
| 184 |
+
"rms_norm_eps": 1e-06,
|
| 185 |
+
"rope_scaling": null,
|
| 186 |
+
"rope_theta": 1000000.0,
|
| 187 |
+
"sliding_window": null,
|
| 188 |
+
"tie_word_embeddings": false,
|
| 189 |
+
"tokenizer_model_max_length": 32768,
|
| 190 |
+
"tokenizer_padding_side": "right",
|
| 191 |
+
"torch_dtype": "bfloat16",
|
| 192 |
+
"transformers_version": "4.44.0",
|
| 193 |
+
"use_cache": true,
|
| 194 |
+
"use_mm_proj": true,
|
| 195 |
+
"use_pos_skipping": false,
|
| 196 |
+
"use_sliding_window": false,
|
| 197 |
+
"vision_tower_pretrained": null,
|
| 198 |
+
"vocab_size": 151647
|
| 199 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"attn_implementation": null,
|
| 3 |
+
"bos_token_id": 151643,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
151645,
|
| 7 |
+
151643
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 151643,
|
| 10 |
+
"repetition_penalty": 1.05,
|
| 11 |
+
"temperature": 0.7,
|
| 12 |
+
"top_k": 20,
|
| 13 |
+
"top_p": 0.8,
|
| 14 |
+
"transformers_version": "4.44.0"
|
| 15 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4a93ab3164589039dbf997f792a63e9479c21d9ee5114f015ed08394bd89dff
|
| 3 |
+
size 4874678976
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46242bb38d2d5fea01af36b2eca977172508ec7ebf0890f9dba21496f8a2efb1
|
| 3 |
+
size 4932751008
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:526f6eb71e500fb64043ca9687bcc23a18c4f1192c59d75198eaa324cfdd98b2
|
| 3 |
+
size 4994571904
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:711a5863b9db01e510c4bf998f4ffac1ed0aae4dee7228f16d0771af84ef4cf3
|
| 3 |
+
size 1348907608
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|im_end|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"pad_token": {
|
| 14 |
+
"content": "<|endoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
}
|
| 20 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "<image>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"additional_special_tokens": [
|
| 38 |
+
"<|im_start|>",
|
| 39 |
+
"<|im_end|>"
|
| 40 |
+
],
|
| 41 |
+
"bos_token": null,
|
| 42 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 43 |
+
"clean_up_tokenization_spaces": false,
|
| 44 |
+
"eos_token": "<|im_end|>",
|
| 45 |
+
"errors": "replace",
|
| 46 |
+
"model_max_length": 32768,
|
| 47 |
+
"pad_token": "<|endoftext|>",
|
| 48 |
+
"padding_side": "right",
|
| 49 |
+
"processor_class": "LlavaProcessor",
|
| 50 |
+
"split_special_tokens": false,
|
| 51 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 52 |
+
"unk_token": null
|
| 53 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1883 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 263,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0038022813688212928,
|
| 13 |
+
"grad_norm": 5.838922807698357,
|
| 14 |
+
"learning_rate": 1.25e-06,
|
| 15 |
+
"loss": 0.4415,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0076045627376425855,
|
| 20 |
+
"grad_norm": 6.450188921175257,
|
| 21 |
+
"learning_rate": 2.5e-06,
|
| 22 |
+
"loss": 0.2714,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.011406844106463879,
|
| 27 |
+
"grad_norm": 6.328807457221606,
|
| 28 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 29 |
+
"loss": 0.2392,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.015209125475285171,
|
| 34 |
+
"grad_norm": 3.1900170578862403,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.1902,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.019011406844106463,
|
| 41 |
+
"grad_norm": 4.396787628400217,
|
| 42 |
+
"learning_rate": 6.25e-06,
|
| 43 |
+
"loss": 0.2287,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.022813688212927757,
|
| 48 |
+
"grad_norm": 2.328876214447226,
|
| 49 |
+
"learning_rate": 7.500000000000001e-06,
|
| 50 |
+
"loss": 0.1073,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.026615969581749048,
|
| 55 |
+
"grad_norm": 1.9893066003363478,
|
| 56 |
+
"learning_rate": 8.750000000000001e-06,
|
| 57 |
+
"loss": 0.0892,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.030418250950570342,
|
| 62 |
+
"grad_norm": 3.4491162158646227,
|
| 63 |
+
"learning_rate": 1e-05,
|
| 64 |
+
"loss": 0.1859,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.034220532319391636,
|
| 69 |
+
"grad_norm": 3.391007553648945,
|
| 70 |
+
"learning_rate": 9.999620550574155e-06,
|
| 71 |
+
"loss": 0.1162,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03802281368821293,
|
| 76 |
+
"grad_norm": 5.243929471727456,
|
| 77 |
+
"learning_rate": 9.99848225988936e-06,
|
| 78 |
+
"loss": 0.2052,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04182509505703422,
|
| 83 |
+
"grad_norm": 2.556605157640159,
|
| 84 |
+
"learning_rate": 9.996585300715117e-06,
|
| 85 |
+
"loss": 0.1279,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.045627376425855515,
|
| 90 |
+
"grad_norm": 2.3229950077008232,
|
| 91 |
+
"learning_rate": 9.99392996097145e-06,
|
| 92 |
+
"loss": 0.2329,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.049429657794676805,
|
| 97 |
+
"grad_norm": 3.4538715709146754,
|
| 98 |
+
"learning_rate": 9.990516643685222e-06,
|
| 99 |
+
"loss": 0.0859,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.053231939163498096,
|
| 104 |
+
"grad_norm": 2.144930306306445,
|
| 105 |
+
"learning_rate": 9.98634586692894e-06,
|
| 106 |
+
"loss": 0.1541,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.057034220532319393,
|
| 111 |
+
"grad_norm": 1.6344989568819797,
|
| 112 |
+
"learning_rate": 9.981418263742148e-06,
|
| 113 |
+
"loss": 0.1433,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.060836501901140684,
|
| 118 |
+
"grad_norm": 1.2978801538360458,
|
| 119 |
+
"learning_rate": 9.975734582035323e-06,
|
| 120 |
+
"loss": 0.0885,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06463878326996197,
|
| 125 |
+
"grad_norm": 2.0950541376792366,
|
| 126 |
+
"learning_rate": 9.96929568447637e-06,
|
| 127 |
+
"loss": 0.1319,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06844106463878327,
|
| 132 |
+
"grad_norm": 1.0425316398679343,
|
| 133 |
+
"learning_rate": 9.96210254835968e-06,
|
| 134 |
+
"loss": 0.1279,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07224334600760456,
|
| 139 |
+
"grad_norm": 1.7860823823073846,
|
| 140 |
+
"learning_rate": 9.954156265457801e-06,
|
| 141 |
+
"loss": 0.1857,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07604562737642585,
|
| 146 |
+
"grad_norm": 1.7415782346194582,
|
| 147 |
+
"learning_rate": 9.945458041855732e-06,
|
| 148 |
+
"loss": 0.0717,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07984790874524715,
|
| 153 |
+
"grad_norm": 1.0879866640913596,
|
| 154 |
+
"learning_rate": 9.936009197767847e-06,
|
| 155 |
+
"loss": 0.1016,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08365019011406843,
|
| 160 |
+
"grad_norm": 2.2439360478661645,
|
| 161 |
+
"learning_rate": 9.925811167337533e-06,
|
| 162 |
+
"loss": 0.0935,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08745247148288973,
|
| 167 |
+
"grad_norm": 1.8496647469351228,
|
| 168 |
+
"learning_rate": 9.91486549841951e-06,
|
| 169 |
+
"loss": 0.1117,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09125475285171103,
|
| 174 |
+
"grad_norm": 1.1499446513004061,
|
| 175 |
+
"learning_rate": 9.903173852344889e-06,
|
| 176 |
+
"loss": 0.0786,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09505703422053231,
|
| 181 |
+
"grad_norm": 1.0813157058702787,
|
| 182 |
+
"learning_rate": 9.890738003669029e-06,
|
| 183 |
+
"loss": 0.0466,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09885931558935361,
|
| 188 |
+
"grad_norm": 1.1206071268561641,
|
| 189 |
+
"learning_rate": 9.877559839902185e-06,
|
| 190 |
+
"loss": 0.0473,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10266159695817491,
|
| 195 |
+
"grad_norm": 1.2183071418887779,
|
| 196 |
+
"learning_rate": 9.863641361223025e-06,
|
| 197 |
+
"loss": 0.0625,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10646387832699619,
|
| 202 |
+
"grad_norm": 1.7316764015595658,
|
| 203 |
+
"learning_rate": 9.848984680175049e-06,
|
| 204 |
+
"loss": 0.0485,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11026615969581749,
|
| 209 |
+
"grad_norm": 0.6990239085195359,
|
| 210 |
+
"learning_rate": 9.833592021345938e-06,
|
| 211 |
+
"loss": 0.0573,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11406844106463879,
|
| 216 |
+
"grad_norm": 2.2785738442682413,
|
| 217 |
+
"learning_rate": 9.817465721029916e-06,
|
| 218 |
+
"loss": 0.0796,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11787072243346007,
|
| 223 |
+
"grad_norm": 1.0584228821124018,
|
| 224 |
+
"learning_rate": 9.800608226873143e-06,
|
| 225 |
+
"loss": 0.0403,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12167300380228137,
|
| 230 |
+
"grad_norm": 2.5879877206601636,
|
| 231 |
+
"learning_rate": 9.783022097502204e-06,
|
| 232 |
+
"loss": 0.1062,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12547528517110265,
|
| 237 |
+
"grad_norm": 2.5381277344322433,
|
| 238 |
+
"learning_rate": 9.764710002135784e-06,
|
| 239 |
+
"loss": 0.0769,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12927756653992395,
|
| 244 |
+
"grad_norm": 1.0747071356178863,
|
| 245 |
+
"learning_rate": 9.745674720179507e-06,
|
| 246 |
+
"loss": 0.0544,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.13307984790874525,
|
| 251 |
+
"grad_norm": 3.102732036616969,
|
| 252 |
+
"learning_rate": 9.7259191408041e-06,
|
| 253 |
+
"loss": 0.0926,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13688212927756654,
|
| 258 |
+
"grad_norm": 1.0789665373993582,
|
| 259 |
+
"learning_rate": 9.705446262506858e-06,
|
| 260 |
+
"loss": 0.041,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14068441064638784,
|
| 265 |
+
"grad_norm": 1.872380809657841,
|
| 266 |
+
"learning_rate": 9.684259192656554e-06,
|
| 267 |
+
"loss": 0.0987,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1444866920152091,
|
| 272 |
+
"grad_norm": 1.6725384234341496,
|
| 273 |
+
"learning_rate": 9.66236114702178e-06,
|
| 274 |
+
"loss": 0.0664,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1482889733840304,
|
| 279 |
+
"grad_norm": 1.3155529867243292,
|
| 280 |
+
"learning_rate": 9.639755449282874e-06,
|
| 281 |
+
"loss": 0.0471,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1520912547528517,
|
| 286 |
+
"grad_norm": 0.7375858425847438,
|
| 287 |
+
"learning_rate": 9.616445530527448e-06,
|
| 288 |
+
"loss": 0.0542,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.155893536121673,
|
| 293 |
+
"grad_norm": 2.5798050001278865,
|
| 294 |
+
"learning_rate": 9.592434928729617e-06,
|
| 295 |
+
"loss": 0.0851,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1596958174904943,
|
| 300 |
+
"grad_norm": 0.9563326747806046,
|
| 301 |
+
"learning_rate": 9.567727288213005e-06,
|
| 302 |
+
"loss": 0.0701,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1634980988593156,
|
| 307 |
+
"grad_norm": 1.2201713738820383,
|
| 308 |
+
"learning_rate": 9.542326359097619e-06,
|
| 309 |
+
"loss": 0.0483,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16730038022813687,
|
| 314 |
+
"grad_norm": 1.5349602348665532,
|
| 315 |
+
"learning_rate": 9.516235996730645e-06,
|
| 316 |
+
"loss": 0.0493,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.17110266159695817,
|
| 321 |
+
"grad_norm": 1.2129366778348925,
|
| 322 |
+
"learning_rate": 9.489460161101291e-06,
|
| 323 |
+
"loss": 0.0423,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.17490494296577946,
|
| 328 |
+
"grad_norm": 1.5701039195877315,
|
| 329 |
+
"learning_rate": 9.46200291623974e-06,
|
| 330 |
+
"loss": 0.0594,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.17870722433460076,
|
| 335 |
+
"grad_norm": 1.7001361570347544,
|
| 336 |
+
"learning_rate": 9.43386842960031e-06,
|
| 337 |
+
"loss": 0.056,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.18250950570342206,
|
| 342 |
+
"grad_norm": 0.8564787838977387,
|
| 343 |
+
"learning_rate": 9.405060971428924e-06,
|
| 344 |
+
"loss": 0.0498,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.18631178707224336,
|
| 349 |
+
"grad_norm": 2.4376315388128056,
|
| 350 |
+
"learning_rate": 9.375584914114963e-06,
|
| 351 |
+
"loss": 0.0717,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.19011406844106463,
|
| 356 |
+
"grad_norm": 3.0710870034344704,
|
| 357 |
+
"learning_rate": 9.345444731527642e-06,
|
| 358 |
+
"loss": 0.0597,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.19391634980988592,
|
| 363 |
+
"grad_norm": 1.1595171258313646,
|
| 364 |
+
"learning_rate": 9.31464499833695e-06,
|
| 365 |
+
"loss": 0.0484,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.19771863117870722,
|
| 370 |
+
"grad_norm": 2.036685842781906,
|
| 371 |
+
"learning_rate": 9.283190389319315e-06,
|
| 372 |
+
"loss": 0.0533,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.20152091254752852,
|
| 377 |
+
"grad_norm": 1.6358538836833496,
|
| 378 |
+
"learning_rate": 9.251085678648072e-06,
|
| 379 |
+
"loss": 0.063,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.20532319391634982,
|
| 384 |
+
"grad_norm": 2.341827886552907,
|
| 385 |
+
"learning_rate": 9.218335739168833e-06,
|
| 386 |
+
"loss": 0.0614,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.20912547528517111,
|
| 391 |
+
"grad_norm": 1.6561642430586991,
|
| 392 |
+
"learning_rate": 9.18494554165989e-06,
|
| 393 |
+
"loss": 0.0335,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.21292775665399238,
|
| 398 |
+
"grad_norm": 1.395413976072295,
|
| 399 |
+
"learning_rate": 9.150920154077753e-06,
|
| 400 |
+
"loss": 0.0536,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.21673003802281368,
|
| 405 |
+
"grad_norm": 2.657218435113838,
|
| 406 |
+
"learning_rate": 9.116264740787937e-06,
|
| 407 |
+
"loss": 0.0697,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.22053231939163498,
|
| 412 |
+
"grad_norm": 3.3032959131412634,
|
| 413 |
+
"learning_rate": 9.08098456178111e-06,
|
| 414 |
+
"loss": 0.0833,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.22433460076045628,
|
| 419 |
+
"grad_norm": 1.8964099123844205,
|
| 420 |
+
"learning_rate": 9.045084971874738e-06,
|
| 421 |
+
"loss": 0.0568,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.22813688212927757,
|
| 426 |
+
"grad_norm": 1.5300857159970949,
|
| 427 |
+
"learning_rate": 9.008571419900334e-06,
|
| 428 |
+
"loss": 0.0599,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.23193916349809887,
|
| 433 |
+
"grad_norm": 1.5891594516335554,
|
| 434 |
+
"learning_rate": 8.97144944787643e-06,
|
| 435 |
+
"loss": 0.0454,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.23574144486692014,
|
| 440 |
+
"grad_norm": 1.3527045916224263,
|
| 441 |
+
"learning_rate": 8.933724690167417e-06,
|
| 442 |
+
"loss": 0.0429,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.23954372623574144,
|
| 447 |
+
"grad_norm": 3.4176684375694446,
|
| 448 |
+
"learning_rate": 8.895402872628352e-06,
|
| 449 |
+
"loss": 0.0535,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.24334600760456274,
|
| 454 |
+
"grad_norm": 0.9708842431312176,
|
| 455 |
+
"learning_rate": 8.856489811735904e-06,
|
| 456 |
+
"loss": 0.05,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.24714828897338403,
|
| 461 |
+
"grad_norm": 1.4542670605196113,
|
| 462 |
+
"learning_rate": 8.816991413705515e-06,
|
| 463 |
+
"loss": 0.0553,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2509505703422053,
|
| 468 |
+
"grad_norm": 1.3407992355297549,
|
| 469 |
+
"learning_rate": 8.776913673594968e-06,
|
| 470 |
+
"loss": 0.0337,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.25475285171102663,
|
| 475 |
+
"grad_norm": 2.282768165416241,
|
| 476 |
+
"learning_rate": 8.736262674394455e-06,
|
| 477 |
+
"loss": 0.054,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2585551330798479,
|
| 482 |
+
"grad_norm": 2.120541919990877,
|
| 483 |
+
"learning_rate": 8.695044586103297e-06,
|
| 484 |
+
"loss": 0.0528,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2623574144486692,
|
| 489 |
+
"grad_norm": 2.340243548218491,
|
| 490 |
+
"learning_rate": 8.653265664793466e-06,
|
| 491 |
+
"loss": 0.0559,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2661596958174905,
|
| 496 |
+
"grad_norm": 2.6043724991846915,
|
| 497 |
+
"learning_rate": 8.610932251660046e-06,
|
| 498 |
+
"loss": 0.0788,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.26996197718631176,
|
| 503 |
+
"grad_norm": 1.981130327340551,
|
| 504 |
+
"learning_rate": 8.568050772058763e-06,
|
| 505 |
+
"loss": 0.0631,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2737642585551331,
|
| 510 |
+
"grad_norm": 3.0833275950321535,
|
| 511 |
+
"learning_rate": 8.524627734530738e-06,
|
| 512 |
+
"loss": 0.066,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.27756653992395436,
|
| 517 |
+
"grad_norm": 2.2997642624470354,
|
| 518 |
+
"learning_rate": 8.480669729814635e-06,
|
| 519 |
+
"loss": 0.056,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2813688212927757,
|
| 524 |
+
"grad_norm": 1.4420430072448926,
|
| 525 |
+
"learning_rate": 8.436183429846314e-06,
|
| 526 |
+
"loss": 0.0425,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.28517110266159695,
|
| 531 |
+
"grad_norm": 1.6013511726718204,
|
| 532 |
+
"learning_rate": 8.39117558674617e-06,
|
| 533 |
+
"loss": 0.0501,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2889733840304182,
|
| 538 |
+
"grad_norm": 1.4338725585293788,
|
| 539 |
+
"learning_rate": 8.345653031794292e-06,
|
| 540 |
+
"loss": 0.0422,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.29277566539923955,
|
| 545 |
+
"grad_norm": 3.890683595146634,
|
| 546 |
+
"learning_rate": 8.299622674393615e-06,
|
| 547 |
+
"loss": 0.1042,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2965779467680608,
|
| 552 |
+
"grad_norm": 2.8479412113795406,
|
| 553 |
+
"learning_rate": 8.25309150102121e-06,
|
| 554 |
+
"loss": 0.065,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.30038022813688214,
|
| 559 |
+
"grad_norm": 1.431054324105267,
|
| 560 |
+
"learning_rate": 8.206066574167893e-06,
|
| 561 |
+
"loss": 0.0458,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3041825095057034,
|
| 566 |
+
"grad_norm": 0.8730610840802391,
|
| 567 |
+
"learning_rate": 8.158555031266255e-06,
|
| 568 |
+
"loss": 0.04,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.30798479087452474,
|
| 573 |
+
"grad_norm": 2.448239293063525,
|
| 574 |
+
"learning_rate": 8.110564083607371e-06,
|
| 575 |
+
"loss": 0.0759,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.311787072243346,
|
| 580 |
+
"grad_norm": 1.8456070372447848,
|
| 581 |
+
"learning_rate": 8.06210101524625e-06,
|
| 582 |
+
"loss": 0.0576,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3155893536121673,
|
| 587 |
+
"grad_norm": 1.834734351428776,
|
| 588 |
+
"learning_rate": 8.013173181896283e-06,
|
| 589 |
+
"loss": 0.06,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3193916349809886,
|
| 594 |
+
"grad_norm": 1.563283754929812,
|
| 595 |
+
"learning_rate": 7.963788009812775e-06,
|
| 596 |
+
"loss": 0.0513,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3231939163498099,
|
| 601 |
+
"grad_norm": 1.1475605231367754,
|
| 602 |
+
"learning_rate": 7.913952994665805e-06,
|
| 603 |
+
"loss": 0.0543,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3269961977186312,
|
| 608 |
+
"grad_norm": 1.4206944686043848,
|
| 609 |
+
"learning_rate": 7.863675700402527e-06,
|
| 610 |
+
"loss": 0.0374,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.33079847908745247,
|
| 615 |
+
"grad_norm": 40.37566470088141,
|
| 616 |
+
"learning_rate": 7.812963758099118e-06,
|
| 617 |
+
"loss": 0.2393,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.33460076045627374,
|
| 622 |
+
"grad_norm": 0.781171019162742,
|
| 623 |
+
"learning_rate": 7.76182486480253e-06,
|
| 624 |
+
"loss": 0.0439,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.33840304182509506,
|
| 629 |
+
"grad_norm": 1.2238520077458774,
|
| 630 |
+
"learning_rate": 7.710266782362248e-06,
|
| 631 |
+
"loss": 0.0363,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.34220532319391633,
|
| 636 |
+
"grad_norm": 3.0641582178280307,
|
| 637 |
+
"learning_rate": 7.658297336252181e-06,
|
| 638 |
+
"loss": 0.0662,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.34600760456273766,
|
| 643 |
+
"grad_norm": 1.965932071811284,
|
| 644 |
+
"learning_rate": 7.605924414382926e-06,
|
| 645 |
+
"loss": 0.0472,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.34980988593155893,
|
| 650 |
+
"grad_norm": 1.755513579015423,
|
| 651 |
+
"learning_rate": 7.553155965904535e-06,
|
| 652 |
+
"loss": 0.0355,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.35361216730038025,
|
| 657 |
+
"grad_norm": 1.7758852391959217,
|
| 658 |
+
"learning_rate": 7.500000000000001e-06,
|
| 659 |
+
"loss": 0.0439,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3574144486692015,
|
| 664 |
+
"grad_norm": 1.3555739799872628,
|
| 665 |
+
"learning_rate": 7.4464645846696186e-06,
|
| 666 |
+
"loss": 0.0503,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3612167300380228,
|
| 671 |
+
"grad_norm": 1.4562802703339102,
|
| 672 |
+
"learning_rate": 7.392557845506433e-06,
|
| 673 |
+
"loss": 0.0451,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3650190114068441,
|
| 678 |
+
"grad_norm": 1.242339996611598,
|
| 679 |
+
"learning_rate": 7.3382879644629345e-06,
|
| 680 |
+
"loss": 0.0372,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3688212927756654,
|
| 685 |
+
"grad_norm": 1.7658082693457402,
|
| 686 |
+
"learning_rate": 7.283663178609204e-06,
|
| 687 |
+
"loss": 0.0516,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3726235741444867,
|
| 692 |
+
"grad_norm": 2.308009029051408,
|
| 693 |
+
"learning_rate": 7.2286917788826926e-06,
|
| 694 |
+
"loss": 0.049,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.376425855513308,
|
| 699 |
+
"grad_norm": 1.5974409986244866,
|
| 700 |
+
"learning_rate": 7.173382108829826e-06,
|
| 701 |
+
"loss": 0.052,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.38022813688212925,
|
| 706 |
+
"grad_norm": 0.8994670069496103,
|
| 707 |
+
"learning_rate": 7.117742563339622e-06,
|
| 708 |
+
"loss": 0.0256,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3840304182509506,
|
| 713 |
+
"grad_norm": 2.4678061096370465,
|
| 714 |
+
"learning_rate": 7.061781587369518e-06,
|
| 715 |
+
"loss": 0.0666,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.38783269961977185,
|
| 720 |
+
"grad_norm": 0.866063867737264,
|
| 721 |
+
"learning_rate": 7.005507674663594e-06,
|
| 722 |
+
"loss": 0.0404,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3916349809885932,
|
| 727 |
+
"grad_norm": 1.2929123997330905,
|
| 728 |
+
"learning_rate": 6.948929366463397e-06,
|
| 729 |
+
"loss": 0.0416,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.39543726235741444,
|
| 734 |
+
"grad_norm": 1.6302913701612327,
|
| 735 |
+
"learning_rate": 6.892055250211552e-06,
|
| 736 |
+
"loss": 0.0444,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.39923954372623577,
|
| 741 |
+
"grad_norm": 2.728101647559586,
|
| 742 |
+
"learning_rate": 6.834893958248361e-06,
|
| 743 |
+
"loss": 0.0475,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.40304182509505704,
|
| 748 |
+
"grad_norm": 2.2488039758111915,
|
| 749 |
+
"learning_rate": 6.77745416650159e-06,
|
| 750 |
+
"loss": 0.0659,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4068441064638783,
|
| 755 |
+
"grad_norm": 1.977313537391127,
|
| 756 |
+
"learning_rate": 6.719744593169642e-06,
|
| 757 |
+
"loss": 0.052,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.41064638783269963,
|
| 762 |
+
"grad_norm": 5.706465703956973,
|
| 763 |
+
"learning_rate": 6.6617739973982985e-06,
|
| 764 |
+
"loss": 0.2998,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4144486692015209,
|
| 769 |
+
"grad_norm": 3.8000626899961825,
|
| 770 |
+
"learning_rate": 6.6035511779512764e-06,
|
| 771 |
+
"loss": 0.1164,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.41825095057034223,
|
| 776 |
+
"grad_norm": 10.502568472372902,
|
| 777 |
+
"learning_rate": 6.545084971874738e-06,
|
| 778 |
+
"loss": 0.123,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4220532319391635,
|
| 783 |
+
"grad_norm": 1.8313837197728655,
|
| 784 |
+
"learning_rate": 6.486384253156014e-06,
|
| 785 |
+
"loss": 0.1134,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.42585551330798477,
|
| 790 |
+
"grad_norm": 2.0327190996550177,
|
| 791 |
+
"learning_rate": 6.427457931376712e-06,
|
| 792 |
+
"loss": 0.0888,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4296577946768061,
|
| 797 |
+
"grad_norm": 1.7161432735601356,
|
| 798 |
+
"learning_rate": 6.368314950360416e-06,
|
| 799 |
+
"loss": 0.0902,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.43346007604562736,
|
| 804 |
+
"grad_norm": 1.074310172184068,
|
| 805 |
+
"learning_rate": 6.308964286815203e-06,
|
| 806 |
+
"loss": 0.0581,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4372623574144487,
|
| 811 |
+
"grad_norm": 1.6451467195844793,
|
| 812 |
+
"learning_rate": 6.249414948971154e-06,
|
| 813 |
+
"loss": 0.0503,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.44106463878326996,
|
| 818 |
+
"grad_norm": 4.814077108982163,
|
| 819 |
+
"learning_rate": 6.189675975213094e-06,
|
| 820 |
+
"loss": 0.1629,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4448669201520912,
|
| 825 |
+
"grad_norm": 1.4504882892841855,
|
| 826 |
+
"learning_rate": 6.129756432708739e-06,
|
| 827 |
+
"loss": 0.077,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.44866920152091255,
|
| 832 |
+
"grad_norm": 2.5745840097932517,
|
| 833 |
+
"learning_rate": 6.0696654160324875e-06,
|
| 834 |
+
"loss": 0.0785,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4524714828897338,
|
| 839 |
+
"grad_norm": 1.2717081467380555,
|
| 840 |
+
"learning_rate": 6.009412045785051e-06,
|
| 841 |
+
"loss": 0.0721,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.45627376425855515,
|
| 846 |
+
"grad_norm": 1.3500755816952386,
|
| 847 |
+
"learning_rate": 5.9490054672091305e-06,
|
| 848 |
+
"loss": 0.0585,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4600760456273764,
|
| 853 |
+
"grad_norm": 0.6582503184547626,
|
| 854 |
+
"learning_rate": 5.888454848801345e-06,
|
| 855 |
+
"loss": 0.0647,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.46387832699619774,
|
| 860 |
+
"grad_norm": 2.057266551706436,
|
| 861 |
+
"learning_rate": 5.82776938092065e-06,
|
| 862 |
+
"loss": 0.1195,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.467680608365019,
|
| 867 |
+
"grad_norm": 1.717649125812939,
|
| 868 |
+
"learning_rate": 5.766958274393428e-06,
|
| 869 |
+
"loss": 0.0762,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4714828897338403,
|
| 874 |
+
"grad_norm": 1.0979465765609746,
|
| 875 |
+
"learning_rate": 5.706030759115458e-06,
|
| 876 |
+
"loss": 0.074,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4752851711026616,
|
| 881 |
+
"grad_norm": 2.4576710255834184,
|
| 882 |
+
"learning_rate": 5.644996082651018e-06,
|
| 883 |
+
"loss": 0.09,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4790874524714829,
|
| 888 |
+
"grad_norm": 1.4141031874389902,
|
| 889 |
+
"learning_rate": 5.583863508829281e-06,
|
| 890 |
+
"loss": 0.0712,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4828897338403042,
|
| 895 |
+
"grad_norm": 1.3722960078778967,
|
| 896 |
+
"learning_rate": 5.522642316338268e-06,
|
| 897 |
+
"loss": 0.0774,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.4866920152091255,
|
| 902 |
+
"grad_norm": 1.9899235244969848,
|
| 903 |
+
"learning_rate": 5.46134179731651e-06,
|
| 904 |
+
"loss": 0.0727,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.49049429657794674,
|
| 909 |
+
"grad_norm": 0.9857281384665181,
|
| 910 |
+
"learning_rate": 5.399971255942708e-06,
|
| 911 |
+
"loss": 0.0709,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.49429657794676807,
|
| 916 |
+
"grad_norm": 1.873178689272728,
|
| 917 |
+
"learning_rate": 5.338540007023538e-06,
|
| 918 |
+
"loss": 0.086,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.49809885931558934,
|
| 923 |
+
"grad_norm": 1.2985642993407172,
|
| 924 |
+
"learning_rate": 5.27705737457985e-06,
|
| 925 |
+
"loss": 0.0729,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5019011406844106,
|
| 930 |
+
"grad_norm": 2.00845296448955,
|
| 931 |
+
"learning_rate": 5.2155326904314795e-06,
|
| 932 |
+
"loss": 0.0796,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5057034220532319,
|
| 937 |
+
"grad_norm": 1.3854160082227878,
|
| 938 |
+
"learning_rate": 5.153975292780852e-06,
|
| 939 |
+
"loss": 0.0655,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5095057034220533,
|
| 944 |
+
"grad_norm": 0.8110311967262005,
|
| 945 |
+
"learning_rate": 5.09239452479565e-06,
|
| 946 |
+
"loss": 0.0832,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5133079847908745,
|
| 951 |
+
"grad_norm": 1.349215941618463,
|
| 952 |
+
"learning_rate": 5.030799733190694e-06,
|
| 953 |
+
"loss": 0.0734,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5171102661596958,
|
| 958 |
+
"grad_norm": 1.7298488968449564,
|
| 959 |
+
"learning_rate": 4.9692002668093075e-06,
|
| 960 |
+
"loss": 0.0854,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5209125475285171,
|
| 965 |
+
"grad_norm": 1.9286164063763356,
|
| 966 |
+
"learning_rate": 4.907605475204352e-06,
|
| 967 |
+
"loss": 0.0967,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5247148288973384,
|
| 972 |
+
"grad_norm": 1.5511875987133978,
|
| 973 |
+
"learning_rate": 4.846024707219149e-06,
|
| 974 |
+
"loss": 0.0895,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5285171102661597,
|
| 979 |
+
"grad_norm": 1.8297476277504392,
|
| 980 |
+
"learning_rate": 4.784467309568524e-06,
|
| 981 |
+
"loss": 0.0861,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.532319391634981,
|
| 986 |
+
"grad_norm": 1.4103839066534454,
|
| 987 |
+
"learning_rate": 4.7229426254201504e-06,
|
| 988 |
+
"loss": 0.0783,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5361216730038023,
|
| 993 |
+
"grad_norm": 2.1074351624344136,
|
| 994 |
+
"learning_rate": 4.661459992976463e-06,
|
| 995 |
+
"loss": 0.0822,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5399239543726235,
|
| 1000 |
+
"grad_norm": 2.0662466538476827,
|
| 1001 |
+
"learning_rate": 4.6000287440572925e-06,
|
| 1002 |
+
"loss": 0.0923,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5437262357414449,
|
| 1007 |
+
"grad_norm": 1.9436401310479368,
|
| 1008 |
+
"learning_rate": 4.53865820268349e-06,
|
| 1009 |
+
"loss": 0.0973,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5475285171102662,
|
| 1014 |
+
"grad_norm": 1.6678903520906458,
|
| 1015 |
+
"learning_rate": 4.477357683661734e-06,
|
| 1016 |
+
"loss": 0.0877,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5513307984790875,
|
| 1021 |
+
"grad_norm": 0.809599618210663,
|
| 1022 |
+
"learning_rate": 4.41613649117072e-06,
|
| 1023 |
+
"loss": 0.0786,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5551330798479087,
|
| 1028 |
+
"grad_norm": 1.0977795425796129,
|
| 1029 |
+
"learning_rate": 4.355003917348985e-06,
|
| 1030 |
+
"loss": 0.0831,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.55893536121673,
|
| 1035 |
+
"grad_norm": 1.8349191027751834,
|
| 1036 |
+
"learning_rate": 4.293969240884545e-06,
|
| 1037 |
+
"loss": 0.0546,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5627376425855514,
|
| 1042 |
+
"grad_norm": 1.2008364879121194,
|
| 1043 |
+
"learning_rate": 4.233041725606573e-06,
|
| 1044 |
+
"loss": 0.0614,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5665399239543726,
|
| 1049 |
+
"grad_norm": 0.732201424861496,
|
| 1050 |
+
"learning_rate": 4.17223061907935e-06,
|
| 1051 |
+
"loss": 0.0734,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5703422053231939,
|
| 1056 |
+
"grad_norm": 1.225050397564832,
|
| 1057 |
+
"learning_rate": 4.111545151198657e-06,
|
| 1058 |
+
"loss": 0.0651,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5741444866920152,
|
| 1063 |
+
"grad_norm": 1.2302003045619856,
|
| 1064 |
+
"learning_rate": 4.050994532790871e-06,
|
| 1065 |
+
"loss": 0.0837,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5779467680608364,
|
| 1070 |
+
"grad_norm": 1.8397382613360544,
|
| 1071 |
+
"learning_rate": 3.99058795421495e-06,
|
| 1072 |
+
"loss": 0.0607,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5817490494296578,
|
| 1077 |
+
"grad_norm": 1.4093018088954288,
|
| 1078 |
+
"learning_rate": 3.930334583967514e-06,
|
| 1079 |
+
"loss": 0.0575,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5855513307984791,
|
| 1084 |
+
"grad_norm": 1.4975102213218823,
|
| 1085 |
+
"learning_rate": 3.870243567291263e-06,
|
| 1086 |
+
"loss": 0.0634,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5893536121673004,
|
| 1091 |
+
"grad_norm": 1.8275993786721776,
|
| 1092 |
+
"learning_rate": 3.8103240247869077e-06,
|
| 1093 |
+
"loss": 0.0477,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5931558935361216,
|
| 1098 |
+
"grad_norm": 1.5246149628230938,
|
| 1099 |
+
"learning_rate": 3.7505850510288455e-06,
|
| 1100 |
+
"loss": 0.0927,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.596958174904943,
|
| 1105 |
+
"grad_norm": 1.07378251359328,
|
| 1106 |
+
"learning_rate": 3.6910357131847986e-06,
|
| 1107 |
+
"loss": 0.0798,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6007604562737643,
|
| 1112 |
+
"grad_norm": 1.2099308034216252,
|
| 1113 |
+
"learning_rate": 3.6316850496395863e-06,
|
| 1114 |
+
"loss": 0.0724,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6045627376425855,
|
| 1119 |
+
"grad_norm": 1.0732935946375615,
|
| 1120 |
+
"learning_rate": 3.5725420686232903e-06,
|
| 1121 |
+
"loss": 0.0773,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6083650190114068,
|
| 1126 |
+
"grad_norm": 0.9225394975341887,
|
| 1127 |
+
"learning_rate": 3.513615746843987e-06,
|
| 1128 |
+
"loss": 0.0635,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6121673003802282,
|
| 1133 |
+
"grad_norm": 2.6158603947076577,
|
| 1134 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1135 |
+
"loss": 0.1059,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6159695817490495,
|
| 1140 |
+
"grad_norm": 0.9294415514571752,
|
| 1141 |
+
"learning_rate": 3.3964488220487252e-06,
|
| 1142 |
+
"loss": 0.0709,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6197718631178707,
|
| 1147 |
+
"grad_norm": 1.9222737047405196,
|
| 1148 |
+
"learning_rate": 3.3382260026017027e-06,
|
| 1149 |
+
"loss": 0.062,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.623574144486692,
|
| 1154 |
+
"grad_norm": 1.906035535415949,
|
| 1155 |
+
"learning_rate": 3.2802554068303595e-06,
|
| 1156 |
+
"loss": 0.0967,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6273764258555133,
|
| 1161 |
+
"grad_norm": 1.7676981737731525,
|
| 1162 |
+
"learning_rate": 3.22254583349841e-06,
|
| 1163 |
+
"loss": 0.0961,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6311787072243346,
|
| 1168 |
+
"grad_norm": 1.605378292764449,
|
| 1169 |
+
"learning_rate": 3.16510604175164e-06,
|
| 1170 |
+
"loss": 0.0507,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6349809885931559,
|
| 1175 |
+
"grad_norm": 1.1486017436180254,
|
| 1176 |
+
"learning_rate": 3.107944749788449e-06,
|
| 1177 |
+
"loss": 0.0653,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6387832699619772,
|
| 1182 |
+
"grad_norm": 1.9815488934112822,
|
| 1183 |
+
"learning_rate": 3.0510706335366034e-06,
|
| 1184 |
+
"loss": 0.0687,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6425855513307985,
|
| 1189 |
+
"grad_norm": 1.747055284984615,
|
| 1190 |
+
"learning_rate": 2.9944923253364066e-06,
|
| 1191 |
+
"loss": 0.0753,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6463878326996197,
|
| 1196 |
+
"grad_norm": 1.5044113934612662,
|
| 1197 |
+
"learning_rate": 2.9382184126304834e-06,
|
| 1198 |
+
"loss": 0.0751,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6501901140684411,
|
| 1203 |
+
"grad_norm": 1.4456809835756477,
|
| 1204 |
+
"learning_rate": 2.8822574366603804e-06,
|
| 1205 |
+
"loss": 0.0857,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6539923954372624,
|
| 1210 |
+
"grad_norm": 0.8421771448658065,
|
| 1211 |
+
"learning_rate": 2.8266178911701757e-06,
|
| 1212 |
+
"loss": 0.0754,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6577946768060836,
|
| 1217 |
+
"grad_norm": 1.6606796114651394,
|
| 1218 |
+
"learning_rate": 2.771308221117309e-06,
|
| 1219 |
+
"loss": 0.0633,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6615969581749049,
|
| 1224 |
+
"grad_norm": 1.0518360600901147,
|
| 1225 |
+
"learning_rate": 2.7163368213907975e-06,
|
| 1226 |
+
"loss": 0.0595,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6653992395437263,
|
| 1231 |
+
"grad_norm": 0.8118249216729664,
|
| 1232 |
+
"learning_rate": 2.6617120355370667e-06,
|
| 1233 |
+
"loss": 0.0721,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6692015209125475,
|
| 1238 |
+
"grad_norm": 2.735849675959911,
|
| 1239 |
+
"learning_rate": 2.607442154493568e-06,
|
| 1240 |
+
"loss": 0.0981,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6730038022813688,
|
| 1245 |
+
"grad_norm": 1.5818010804365756,
|
| 1246 |
+
"learning_rate": 2.5535354153303827e-06,
|
| 1247 |
+
"loss": 0.0944,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6768060836501901,
|
| 1252 |
+
"grad_norm": 2.106035867055531,
|
| 1253 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1254 |
+
"loss": 0.0563,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6806083650190115,
|
| 1259 |
+
"grad_norm": 1.1144497943115816,
|
| 1260 |
+
"learning_rate": 2.4468440340954664e-06,
|
| 1261 |
+
"loss": 0.08,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6844106463878327,
|
| 1266 |
+
"grad_norm": 0.6703641595119413,
|
| 1267 |
+
"learning_rate": 2.3940755856170744e-06,
|
| 1268 |
+
"loss": 0.0681,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.688212927756654,
|
| 1273 |
+
"grad_norm": 1.5423718248019078,
|
| 1274 |
+
"learning_rate": 2.341702663747819e-06,
|
| 1275 |
+
"loss": 0.0934,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6920152091254753,
|
| 1280 |
+
"grad_norm": 2.5974586850504173,
|
| 1281 |
+
"learning_rate": 2.289733217637753e-06,
|
| 1282 |
+
"loss": 0.0727,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6958174904942965,
|
| 1287 |
+
"grad_norm": 1.6582686566551548,
|
| 1288 |
+
"learning_rate": 2.238175135197471e-06,
|
| 1289 |
+
"loss": 0.0708,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6996197718631179,
|
| 1294 |
+
"grad_norm": 1.5364491430874354,
|
| 1295 |
+
"learning_rate": 2.1870362419008844e-06,
|
| 1296 |
+
"loss": 0.0896,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7034220532319392,
|
| 1301 |
+
"grad_norm": 0.8883156220926425,
|
| 1302 |
+
"learning_rate": 2.136324299597474e-06,
|
| 1303 |
+
"loss": 0.0805,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7072243346007605,
|
| 1308 |
+
"grad_norm": 0.722419818120629,
|
| 1309 |
+
"learning_rate": 2.0860470053341957e-06,
|
| 1310 |
+
"loss": 0.0677,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7110266159695817,
|
| 1315 |
+
"grad_norm": 0.8687543599479524,
|
| 1316 |
+
"learning_rate": 2.0362119901872262e-06,
|
| 1317 |
+
"loss": 0.0743,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.714828897338403,
|
| 1322 |
+
"grad_norm": 1.1765271610095978,
|
| 1323 |
+
"learning_rate": 1.9868268181037186e-06,
|
| 1324 |
+
"loss": 0.073,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7186311787072244,
|
| 1329 |
+
"grad_norm": 1.57171034016372,
|
| 1330 |
+
"learning_rate": 1.937898984753751e-06,
|
| 1331 |
+
"loss": 0.0774,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7224334600760456,
|
| 1336 |
+
"grad_norm": 2.5653255183794843,
|
| 1337 |
+
"learning_rate": 1.8894359163926312e-06,
|
| 1338 |
+
"loss": 0.0845,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7262357414448669,
|
| 1343 |
+
"grad_norm": 1.8207532400192823,
|
| 1344 |
+
"learning_rate": 1.8414449687337467e-06,
|
| 1345 |
+
"loss": 0.0866,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7300380228136882,
|
| 1350 |
+
"grad_norm": 2.494001462793861,
|
| 1351 |
+
"learning_rate": 1.7939334258321094e-06,
|
| 1352 |
+
"loss": 0.0745,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7338403041825095,
|
| 1357 |
+
"grad_norm": 1.2902772325556104,
|
| 1358 |
+
"learning_rate": 1.746908498978791e-06,
|
| 1359 |
+
"loss": 0.0755,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7376425855513308,
|
| 1364 |
+
"grad_norm": 1.6350962042870758,
|
| 1365 |
+
"learning_rate": 1.7003773256063882e-06,
|
| 1366 |
+
"loss": 0.0734,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7414448669201521,
|
| 1371 |
+
"grad_norm": 0.9676259463093386,
|
| 1372 |
+
"learning_rate": 1.6543469682057105e-06,
|
| 1373 |
+
"loss": 0.0816,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7452471482889734,
|
| 1378 |
+
"grad_norm": 0.5610374914180468,
|
| 1379 |
+
"learning_rate": 1.60882441325383e-06,
|
| 1380 |
+
"loss": 0.0754,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7490494296577946,
|
| 1385 |
+
"grad_norm": 0.7804394462668183,
|
| 1386 |
+
"learning_rate": 1.5638165701536866e-06,
|
| 1387 |
+
"loss": 0.0747,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.752851711026616,
|
| 1392 |
+
"grad_norm": 1.550496983980984,
|
| 1393 |
+
"learning_rate": 1.5193302701853674e-06,
|
| 1394 |
+
"loss": 0.0791,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7566539923954373,
|
| 1399 |
+
"grad_norm": 1.9010955783137817,
|
| 1400 |
+
"learning_rate": 1.475372265469265e-06,
|
| 1401 |
+
"loss": 0.0732,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7604562737642585,
|
| 1406 |
+
"grad_norm": 1.6079174758382067,
|
| 1407 |
+
"learning_rate": 1.4319492279412388e-06,
|
| 1408 |
+
"loss": 0.0708,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7642585551330798,
|
| 1413 |
+
"grad_norm": 0.6607128276682519,
|
| 1414 |
+
"learning_rate": 1.389067748339954e-06,
|
| 1415 |
+
"loss": 0.0715,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7680608365019012,
|
| 1420 |
+
"grad_norm": 1.3595868708508696,
|
| 1421 |
+
"learning_rate": 1.3467343352065349e-06,
|
| 1422 |
+
"loss": 0.0781,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7718631178707225,
|
| 1427 |
+
"grad_norm": 2.419173076809107,
|
| 1428 |
+
"learning_rate": 1.3049554138967052e-06,
|
| 1429 |
+
"loss": 0.0739,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7756653992395437,
|
| 1434 |
+
"grad_norm": 1.8044702216287307,
|
| 1435 |
+
"learning_rate": 1.2637373256055445e-06,
|
| 1436 |
+
"loss": 0.0825,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.779467680608365,
|
| 1441 |
+
"grad_norm": 0.9717656143179705,
|
| 1442 |
+
"learning_rate": 1.2230863264050308e-06,
|
| 1443 |
+
"loss": 0.0777,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7832699619771863,
|
| 1448 |
+
"grad_norm": 1.181299656152716,
|
| 1449 |
+
"learning_rate": 1.1830085862944851e-06,
|
| 1450 |
+
"loss": 0.0726,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7870722433460076,
|
| 1455 |
+
"grad_norm": 0.9219123710812074,
|
| 1456 |
+
"learning_rate": 1.1435101882640964e-06,
|
| 1457 |
+
"loss": 0.0832,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7908745247148289,
|
| 1462 |
+
"grad_norm": 0.6914642768758692,
|
| 1463 |
+
"learning_rate": 1.1045971273716476e-06,
|
| 1464 |
+
"loss": 0.0793,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7946768060836502,
|
| 1469 |
+
"grad_norm": 2.273763477220486,
|
| 1470 |
+
"learning_rate": 1.066275309832584e-06,
|
| 1471 |
+
"loss": 0.0715,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7984790874524715,
|
| 1476 |
+
"grad_norm": 1.6498287914408991,
|
| 1477 |
+
"learning_rate": 1.02855055212357e-06,
|
| 1478 |
+
"loss": 0.0669,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8022813688212928,
|
| 1483 |
+
"grad_norm": 1.102115306685056,
|
| 1484 |
+
"learning_rate": 9.91428580099667e-07,
|
| 1485 |
+
"loss": 0.0775,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.8060836501901141,
|
| 1490 |
+
"grad_norm": 2.4914475016820967,
|
| 1491 |
+
"learning_rate": 9.549150281252633e-07,
|
| 1492 |
+
"loss": 0.0677,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.8098859315589354,
|
| 1497 |
+
"grad_norm": 0.6217527292902375,
|
| 1498 |
+
"learning_rate": 9.190154382188921e-07,
|
| 1499 |
+
"loss": 0.0704,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.8136882129277566,
|
| 1504 |
+
"grad_norm": 0.776872411148349,
|
| 1505 |
+
"learning_rate": 8.837352592120646e-07,
|
| 1506 |
+
"loss": 0.0683,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8174904942965779,
|
| 1511 |
+
"grad_norm": 1.3830390856962926,
|
| 1512 |
+
"learning_rate": 8.490798459222477e-07,
|
| 1513 |
+
"loss": 0.0719,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8212927756653993,
|
| 1518 |
+
"grad_norm": 0.9016535949900154,
|
| 1519 |
+
"learning_rate": 8.150544583401116e-07,
|
| 1520 |
+
"loss": 0.0706,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8250950570342205,
|
| 1525 |
+
"grad_norm": 1.2565895090887536,
|
| 1526 |
+
"learning_rate": 7.816642608311692e-07,
|
| 1527 |
+
"loss": 0.0691,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.8288973384030418,
|
| 1532 |
+
"grad_norm": 0.4685806197849472,
|
| 1533 |
+
"learning_rate": 7.489143213519301e-07,
|
| 1534 |
+
"loss": 0.0749,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8326996197718631,
|
| 1539 |
+
"grad_norm": 1.8324078968655844,
|
| 1540 |
+
"learning_rate": 7.168096106806871e-07,
|
| 1541 |
+
"loss": 0.0842,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8365019011406845,
|
| 1546 |
+
"grad_norm": 0.615073356028593,
|
| 1547 |
+
"learning_rate": 6.853550016630517e-07,
|
| 1548 |
+
"loss": 0.0697,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8403041825095057,
|
| 1553 |
+
"grad_norm": 2.4222883575922154,
|
| 1554 |
+
"learning_rate": 6.545552684723583e-07,
|
| 1555 |
+
"loss": 0.0803,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.844106463878327,
|
| 1560 |
+
"grad_norm": 2.325809289721842,
|
| 1561 |
+
"learning_rate": 6.244150858850368e-07,
|
| 1562 |
+
"loss": 0.0969,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.8479087452471483,
|
| 1567 |
+
"grad_norm": 1.176358839322873,
|
| 1568 |
+
"learning_rate": 5.949390285710777e-07,
|
| 1569 |
+
"loss": 0.0599,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8517110266159695,
|
| 1574 |
+
"grad_norm": 0.7047528759218481,
|
| 1575 |
+
"learning_rate": 5.661315703996905e-07,
|
| 1576 |
+
"loss": 0.0807,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8555133079847909,
|
| 1581 |
+
"grad_norm": 1.1938640871812451,
|
| 1582 |
+
"learning_rate": 5.379970837602611e-07,
|
| 1583 |
+
"loss": 0.0712,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8593155893536122,
|
| 1588 |
+
"grad_norm": 2.0073147773750635,
|
| 1589 |
+
"learning_rate": 5.105398388987098e-07,
|
| 1590 |
+
"loss": 0.0723,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8631178707224335,
|
| 1595 |
+
"grad_norm": 1.910913263898643,
|
| 1596 |
+
"learning_rate": 4.837640032693558e-07,
|
| 1597 |
+
"loss": 0.0795,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8669201520912547,
|
| 1602 |
+
"grad_norm": 1.8337713325079474,
|
| 1603 |
+
"learning_rate": 4.576736409023813e-07,
|
| 1604 |
+
"loss": 0.068,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.870722433460076,
|
| 1609 |
+
"grad_norm": 1.2290049663038816,
|
| 1610 |
+
"learning_rate": 4.322727117869951e-07,
|
| 1611 |
+
"loss": 0.0768,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8745247148288974,
|
| 1616 |
+
"grad_norm": 1.4789730966126406,
|
| 1617 |
+
"learning_rate": 4.0756507127038494e-07,
|
| 1618 |
+
"loss": 0.0767,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.8783269961977186,
|
| 1623 |
+
"grad_norm": 1.6910881638069963,
|
| 1624 |
+
"learning_rate": 3.8355446947255293e-07,
|
| 1625 |
+
"loss": 0.0665,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8821292775665399,
|
| 1630 |
+
"grad_norm": 1.033604422841365,
|
| 1631 |
+
"learning_rate": 3.602445507171276e-07,
|
| 1632 |
+
"loss": 0.0778,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8859315589353612,
|
| 1637 |
+
"grad_norm": 0.8013760374315584,
|
| 1638 |
+
"learning_rate": 3.3763885297822153e-07,
|
| 1639 |
+
"loss": 0.0657,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.8897338403041825,
|
| 1644 |
+
"grad_norm": 2.429431976291132,
|
| 1645 |
+
"learning_rate": 3.1574080734344757e-07,
|
| 1646 |
+
"loss": 0.0644,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8935361216730038,
|
| 1651 |
+
"grad_norm": 0.9711502835652512,
|
| 1652 |
+
"learning_rate": 2.9455373749314285e-07,
|
| 1653 |
+
"loss": 0.0706,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8973384030418251,
|
| 1658 |
+
"grad_norm": 1.275326544398461,
|
| 1659 |
+
"learning_rate": 2.7408085919590265e-07,
|
| 1660 |
+
"loss": 0.0679,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.9011406844106464,
|
| 1665 |
+
"grad_norm": 1.2439206544006953,
|
| 1666 |
+
"learning_rate": 2.5432527982049424e-07,
|
| 1667 |
+
"loss": 0.0686,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9049429657794676,
|
| 1672 |
+
"grad_norm": 1.1151619016488201,
|
| 1673 |
+
"learning_rate": 2.3528999786421758e-07,
|
| 1674 |
+
"loss": 0.067,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.908745247148289,
|
| 1679 |
+
"grad_norm": 1.5637158385664465,
|
| 1680 |
+
"learning_rate": 2.1697790249779638e-07,
|
| 1681 |
+
"loss": 0.0653,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.9125475285171103,
|
| 1686 |
+
"grad_norm": 0.4492873956448081,
|
| 1687 |
+
"learning_rate": 1.9939177312685963e-07,
|
| 1688 |
+
"loss": 0.0782,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.9163498098859315,
|
| 1693 |
+
"grad_norm": 1.1564997677392477,
|
| 1694 |
+
"learning_rate": 1.825342789700846e-07,
|
| 1695 |
+
"loss": 0.0608,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.9201520912547528,
|
| 1700 |
+
"grad_norm": 1.8601658350989132,
|
| 1701 |
+
"learning_rate": 1.664079786540629e-07,
|
| 1702 |
+
"loss": 0.0853,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.9239543726235742,
|
| 1707 |
+
"grad_norm": 1.227671794720867,
|
| 1708 |
+
"learning_rate": 1.510153198249531e-07,
|
| 1709 |
+
"loss": 0.0681,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.9277566539923955,
|
| 1714 |
+
"grad_norm": 1.318746672359122,
|
| 1715 |
+
"learning_rate": 1.363586387769761e-07,
|
| 1716 |
+
"loss": 0.0716,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.9315589353612167,
|
| 1721 |
+
"grad_norm": 1.6242713478117103,
|
| 1722 |
+
"learning_rate": 1.22440160097817e-07,
|
| 1723 |
+
"loss": 0.0585,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.935361216730038,
|
| 1728 |
+
"grad_norm": 2.4521843943377455,
|
| 1729 |
+
"learning_rate": 1.0926199633097156e-07,
|
| 1730 |
+
"loss": 0.0748,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.9391634980988594,
|
| 1735 |
+
"grad_norm": 1.624621392645855,
|
| 1736 |
+
"learning_rate": 9.682614765511134e-08,
|
| 1737 |
+
"loss": 0.0778,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9429657794676806,
|
| 1742 |
+
"grad_norm": 0.4946954920558412,
|
| 1743 |
+
"learning_rate": 8.513450158049109e-08,
|
| 1744 |
+
"loss": 0.0787,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.9467680608365019,
|
| 1749 |
+
"grad_norm": 0.5126439315912313,
|
| 1750 |
+
"learning_rate": 7.418883266246734e-08,
|
| 1751 |
+
"loss": 0.0779,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.9505703422053232,
|
| 1756 |
+
"grad_norm": 1.9296492595340622,
|
| 1757 |
+
"learning_rate": 6.399080223215503e-08,
|
| 1758 |
+
"loss": 0.071,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.9543726235741445,
|
| 1763 |
+
"grad_norm": 2.381450276806091,
|
| 1764 |
+
"learning_rate": 5.454195814427021e-08,
|
| 1765 |
+
"loss": 0.0571,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9581749049429658,
|
| 1770 |
+
"grad_norm": 1.6084565235987887,
|
| 1771 |
+
"learning_rate": 4.584373454219859e-08,
|
| 1772 |
+
"loss": 0.0895,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9619771863117871,
|
| 1777 |
+
"grad_norm": 0.8689683317083515,
|
| 1778 |
+
"learning_rate": 3.7897451640321326e-08,
|
| 1779 |
+
"loss": 0.0666,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9657794676806084,
|
| 1784 |
+
"grad_norm": 1.3522488546526146,
|
| 1785 |
+
"learning_rate": 3.0704315523631956e-08,
|
| 1786 |
+
"loss": 0.0611,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9695817490494296,
|
| 1791 |
+
"grad_norm": 2.2795049559749754,
|
| 1792 |
+
"learning_rate": 2.426541796467785e-08,
|
| 1793 |
+
"loss": 0.0536,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.973384030418251,
|
| 1798 |
+
"grad_norm": 1.087776806550556,
|
| 1799 |
+
"learning_rate": 1.8581736257852756e-08,
|
| 1800 |
+
"loss": 0.0602,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9771863117870723,
|
| 1805 |
+
"grad_norm": 1.1862726730320716,
|
| 1806 |
+
"learning_rate": 1.3654133071059894e-08,
|
| 1807 |
+
"loss": 0.0722,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9809885931558935,
|
| 1812 |
+
"grad_norm": 0.7704198670078842,
|
| 1813 |
+
"learning_rate": 9.48335631477948e-09,
|
| 1814 |
+
"loss": 0.0639,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9847908745247148,
|
| 1819 |
+
"grad_norm": 1.0439502260084565,
|
| 1820 |
+
"learning_rate": 6.070039028550634e-09,
|
| 1821 |
+
"loss": 0.0623,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9885931558935361,
|
| 1826 |
+
"grad_norm": 0.9401759231345554,
|
| 1827 |
+
"learning_rate": 3.41469928488547e-09,
|
| 1828 |
+
"loss": 0.0735,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9923954372623575,
|
| 1833 |
+
"grad_norm": 1.8274467631316753,
|
| 1834 |
+
"learning_rate": 1.5177401106419853e-09,
|
| 1835 |
+
"loss": 0.0714,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9961977186311787,
|
| 1840 |
+
"grad_norm": 1.347810152487735,
|
| 1841 |
+
"learning_rate": 3.7944942584688947e-10,
|
| 1842 |
+
"loss": 0.0698,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 1.0,
|
| 1847 |
+
"grad_norm": 1.4596446787851405,
|
| 1848 |
+
"learning_rate": 0.0,
|
| 1849 |
+
"loss": 0.0874,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 1.0,
|
| 1854 |
+
"step": 263,
|
| 1855 |
+
"total_flos": 8314083559424.0,
|
| 1856 |
+
"train_loss": 0.04440210763889121,
|
| 1857 |
+
"train_runtime": 1286.9653,
|
| 1858 |
+
"train_samples_per_second": 1.632,
|
| 1859 |
+
"train_steps_per_second": 0.204
|
| 1860 |
+
}
|
| 1861 |
+
],
|
| 1862 |
+
"logging_steps": 1.0,
|
| 1863 |
+
"max_steps": 263,
|
| 1864 |
+
"num_input_tokens_seen": 0,
|
| 1865 |
+
"num_train_epochs": 1,
|
| 1866 |
+
"save_steps": 100,
|
| 1867 |
+
"stateful_callbacks": {
|
| 1868 |
+
"TrainerControl": {
|
| 1869 |
+
"args": {
|
| 1870 |
+
"should_epoch_stop": false,
|
| 1871 |
+
"should_evaluate": false,
|
| 1872 |
+
"should_log": false,
|
| 1873 |
+
"should_save": true,
|
| 1874 |
+
"should_training_stop": true
|
| 1875 |
+
},
|
| 1876 |
+
"attributes": {}
|
| 1877 |
+
}
|
| 1878 |
+
},
|
| 1879 |
+
"total_flos": 8314083559424.0,
|
| 1880 |
+
"train_batch_size": 1,
|
| 1881 |
+
"trial_name": null,
|
| 1882 |
+
"trial_params": null
|
| 1883 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b393855f7ba2a54cf50fda6b915e0adfd315f20cd0ac64f2eb6d6d3955c138f0
|
| 3 |
+
size 7992
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|