Upload 4 files
Browse files- tokenizer_config.json +198 -0
- trainer_state.json +1499 -0
- training_args.bin +3 -0
- vocab.json +0 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,198 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
}
|
| 181 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_end|>",
|
| 184 |
+
"<|im_start|>"
|
| 185 |
+
],
|
| 186 |
+
"bos_token": null,
|
| 187 |
+
"clean_up_tokenization_spaces": false,
|
| 188 |
+
"eos_token": "<|im_end|>",
|
| 189 |
+
"errors": "replace",
|
| 190 |
+
"extra_special_tokens": {},
|
| 191 |
+
"model_max_length": 1024,
|
| 192 |
+
"pad_token": "<|endoftext|>",
|
| 193 |
+
"padding_side": "right",
|
| 194 |
+
"split_special_tokens": false,
|
| 195 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 196 |
+
"truncation": true,
|
| 197 |
+
"unk_token": null
|
| 198 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1499 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 6666,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0144014401440144,
|
| 14 |
+
"grad_norm": 10.98869514465332,
|
| 15 |
+
"learning_rate": 4.347826086956522e-07,
|
| 16 |
+
"loss": 0.6974,
|
| 17 |
+
"step": 32
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0288028802880288,
|
| 21 |
+
"grad_norm": 10.853715896606445,
|
| 22 |
+
"learning_rate": 9.145427286356823e-07,
|
| 23 |
+
"loss": 0.6333,
|
| 24 |
+
"step": 64
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.043204320432043204,
|
| 28 |
+
"grad_norm": 13.330265045166016,
|
| 29 |
+
"learning_rate": 1.3943028485757123e-06,
|
| 30 |
+
"loss": 0.5058,
|
| 31 |
+
"step": 96
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0576057605760576,
|
| 35 |
+
"grad_norm": 11.676138877868652,
|
| 36 |
+
"learning_rate": 1.8740629685157422e-06,
|
| 37 |
+
"loss": 0.5039,
|
| 38 |
+
"step": 128
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.07200720072007201,
|
| 42 |
+
"grad_norm": 11.024947166442871,
|
| 43 |
+
"learning_rate": 2.3388305847076464e-06,
|
| 44 |
+
"loss": 0.5272,
|
| 45 |
+
"step": 160
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.08640864086408641,
|
| 49 |
+
"grad_norm": 11.500349998474121,
|
| 50 |
+
"learning_rate": 2.8185907046476763e-06,
|
| 51 |
+
"loss": 0.5414,
|
| 52 |
+
"step": 192
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.10081008100810081,
|
| 56 |
+
"grad_norm": 14.144675254821777,
|
| 57 |
+
"learning_rate": 3.2983508245877066e-06,
|
| 58 |
+
"loss": 0.5167,
|
| 59 |
+
"step": 224
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.1152115211521152,
|
| 63 |
+
"grad_norm": 15.339383125305176,
|
| 64 |
+
"learning_rate": 3.763118440779611e-06,
|
| 65 |
+
"loss": 0.4782,
|
| 66 |
+
"step": 256
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.12961296129612962,
|
| 70 |
+
"grad_norm": 14.547229766845703,
|
| 71 |
+
"learning_rate": 4.242878560719641e-06,
|
| 72 |
+
"loss": 0.5071,
|
| 73 |
+
"step": 288
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.14401440144014402,
|
| 77 |
+
"grad_norm": 16.514537811279297,
|
| 78 |
+
"learning_rate": 4.722638680659671e-06,
|
| 79 |
+
"loss": 0.5238,
|
| 80 |
+
"step": 320
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.15841584158415842,
|
| 84 |
+
"grad_norm": 13.728464126586914,
|
| 85 |
+
"learning_rate": 5.2023988005997004e-06,
|
| 86 |
+
"loss": 0.6176,
|
| 87 |
+
"step": 352
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.17281728172817282,
|
| 91 |
+
"grad_norm": 9.859703063964844,
|
| 92 |
+
"learning_rate": 5.682158920539731e-06,
|
| 93 |
+
"loss": 0.5022,
|
| 94 |
+
"step": 384
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.18721872187218722,
|
| 98 |
+
"grad_norm": 16.529651641845703,
|
| 99 |
+
"learning_rate": 6.16191904047976e-06,
|
| 100 |
+
"loss": 0.5606,
|
| 101 |
+
"step": 416
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.20162016201620162,
|
| 105 |
+
"grad_norm": 12.64522647857666,
|
| 106 |
+
"learning_rate": 6.6416791604197905e-06,
|
| 107 |
+
"loss": 0.5115,
|
| 108 |
+
"step": 448
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.21602160216021601,
|
| 112 |
+
"grad_norm": 9.362441062927246,
|
| 113 |
+
"learning_rate": 7.121439280359821e-06,
|
| 114 |
+
"loss": 0.5371,
|
| 115 |
+
"step": 480
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.2304230423042304,
|
| 119 |
+
"grad_norm": 12.86221694946289,
|
| 120 |
+
"learning_rate": 7.60119940029985e-06,
|
| 121 |
+
"loss": 0.5343,
|
| 122 |
+
"step": 512
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.2448244824482448,
|
| 126 |
+
"grad_norm": 12.512419700622559,
|
| 127 |
+
"learning_rate": 8.065967016491755e-06,
|
| 128 |
+
"loss": 0.6191,
|
| 129 |
+
"step": 544
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.25922592259225924,
|
| 133 |
+
"grad_norm": 14.35505199432373,
|
| 134 |
+
"learning_rate": 8.53073463268366e-06,
|
| 135 |
+
"loss": 0.5668,
|
| 136 |
+
"step": 576
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.27362736273627364,
|
| 140 |
+
"grad_norm": 14.951635360717773,
|
| 141 |
+
"learning_rate": 9.010494752623688e-06,
|
| 142 |
+
"loss": 0.6024,
|
| 143 |
+
"step": 608
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.28802880288028804,
|
| 147 |
+
"grad_norm": 11.853578567504883,
|
| 148 |
+
"learning_rate": 9.490254872563718e-06,
|
| 149 |
+
"loss": 0.6092,
|
| 150 |
+
"step": 640
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.30243024302430244,
|
| 154 |
+
"grad_norm": 9.165077209472656,
|
| 155 |
+
"learning_rate": 9.970014992503749e-06,
|
| 156 |
+
"loss": 0.5587,
|
| 157 |
+
"step": 672
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.31683168316831684,
|
| 161 |
+
"grad_norm": 12.137311935424805,
|
| 162 |
+
"learning_rate": 9.999382956748588e-06,
|
| 163 |
+
"loss": 0.5444,
|
| 164 |
+
"step": 704
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.33123312331233123,
|
| 168 |
+
"grad_norm": 15.218328475952148,
|
| 169 |
+
"learning_rate": 9.997364717027728e-06,
|
| 170 |
+
"loss": 0.5909,
|
| 171 |
+
"step": 736
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.34563456345634563,
|
| 175 |
+
"grad_norm": 13.624066352844238,
|
| 176 |
+
"learning_rate": 9.993943105133823e-06,
|
| 177 |
+
"loss": 0.6135,
|
| 178 |
+
"step": 768
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.36003600360036003,
|
| 182 |
+
"grad_norm": 12.48714542388916,
|
| 183 |
+
"learning_rate": 9.989119081932283e-06,
|
| 184 |
+
"loss": 0.5868,
|
| 185 |
+
"step": 800
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.37443744374437443,
|
| 189 |
+
"grad_norm": 20.65450668334961,
|
| 190 |
+
"learning_rate": 9.9828940021171e-06,
|
| 191 |
+
"loss": 0.6172,
|
| 192 |
+
"step": 832
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.38883888388838883,
|
| 196 |
+
"grad_norm": 15.066054344177246,
|
| 197 |
+
"learning_rate": 9.975269613830395e-06,
|
| 198 |
+
"loss": 0.5682,
|
| 199 |
+
"step": 864
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.40324032403240323,
|
| 203 |
+
"grad_norm": 10.448845863342285,
|
| 204 |
+
"learning_rate": 9.966248058171527e-06,
|
| 205 |
+
"loss": 0.6659,
|
| 206 |
+
"step": 896
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.41764176417641763,
|
| 210 |
+
"grad_norm": 6.550926208496094,
|
| 211 |
+
"learning_rate": 9.955831868595796e-06,
|
| 212 |
+
"loss": 0.6053,
|
| 213 |
+
"step": 928
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.43204320432043203,
|
| 217 |
+
"grad_norm": 18.164133071899414,
|
| 218 |
+
"learning_rate": 9.94402397020302e-06,
|
| 219 |
+
"loss": 0.5109,
|
| 220 |
+
"step": 960
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.4464446444644464,
|
| 224 |
+
"grad_norm": 26.886669158935547,
|
| 225 |
+
"learning_rate": 9.930827678916084e-06,
|
| 226 |
+
"loss": 0.6006,
|
| 227 |
+
"step": 992
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.4608460846084608,
|
| 231 |
+
"grad_norm": 9.457767486572266,
|
| 232 |
+
"learning_rate": 9.916246700549754e-06,
|
| 233 |
+
"loss": 0.5794,
|
| 234 |
+
"step": 1024
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.4752475247524752,
|
| 238 |
+
"grad_norm": 15.5580472946167,
|
| 239 |
+
"learning_rate": 9.900285129770016e-06,
|
| 240 |
+
"loss": 0.5606,
|
| 241 |
+
"step": 1056
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.4896489648964896,
|
| 245 |
+
"grad_norm": 9.84897232055664,
|
| 246 |
+
"learning_rate": 9.882947448944177e-06,
|
| 247 |
+
"loss": 0.6579,
|
| 248 |
+
"step": 1088
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.504050405040504,
|
| 252 |
+
"grad_norm": 10.959239959716797,
|
| 253 |
+
"learning_rate": 9.864238526882147e-06,
|
| 254 |
+
"loss": 0.5455,
|
| 255 |
+
"step": 1120
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.5184518451845185,
|
| 259 |
+
"grad_norm": 15.946135520935059,
|
| 260 |
+
"learning_rate": 9.844163617469138e-06,
|
| 261 |
+
"loss": 0.6002,
|
| 262 |
+
"step": 1152
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.5328532853285328,
|
| 266 |
+
"grad_norm": 12.9840726852417,
|
| 267 |
+
"learning_rate": 9.822728358190274e-06,
|
| 268 |
+
"loss": 0.6541,
|
| 269 |
+
"step": 1184
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.5472547254725473,
|
| 273 |
+
"grad_norm": 11.688780784606934,
|
| 274 |
+
"learning_rate": 9.799938768547452e-06,
|
| 275 |
+
"loss": 0.6294,
|
| 276 |
+
"step": 1216
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.5616561656165616,
|
| 280 |
+
"grad_norm": 13.91913890838623,
|
| 281 |
+
"learning_rate": 9.77580124836893e-06,
|
| 282 |
+
"loss": 0.5856,
|
| 283 |
+
"step": 1248
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.5760576057605761,
|
| 287 |
+
"grad_norm": 13.728696823120117,
|
| 288 |
+
"learning_rate": 9.750322576012119e-06,
|
| 289 |
+
"loss": 0.622,
|
| 290 |
+
"step": 1280
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5904590459045904,
|
| 294 |
+
"grad_norm": 8.887660026550293,
|
| 295 |
+
"learning_rate": 9.723509906460054e-06,
|
| 296 |
+
"loss": 0.6663,
|
| 297 |
+
"step": 1312
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.6048604860486049,
|
| 301 |
+
"grad_norm": 12.27409839630127,
|
| 302 |
+
"learning_rate": 9.69537076931213e-06,
|
| 303 |
+
"loss": 0.5707,
|
| 304 |
+
"step": 1344
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.6192619261926192,
|
| 308 |
+
"grad_norm": 23.663471221923828,
|
| 309 |
+
"learning_rate": 9.665913066669608e-06,
|
| 310 |
+
"loss": 0.6018,
|
| 311 |
+
"step": 1376
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.6336633663366337,
|
| 315 |
+
"grad_norm": 14.472834587097168,
|
| 316 |
+
"learning_rate": 9.635145070916541e-06,
|
| 317 |
+
"loss": 0.5736,
|
| 318 |
+
"step": 1408
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.648064806480648,
|
| 322 |
+
"grad_norm": 13.856266975402832,
|
| 323 |
+
"learning_rate": 9.603075422396685e-06,
|
| 324 |
+
"loss": 0.6117,
|
| 325 |
+
"step": 1440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.6624662466246625,
|
| 329 |
+
"grad_norm": 11.307412147521973,
|
| 330 |
+
"learning_rate": 9.569713126987122e-06,
|
| 331 |
+
"loss": 0.6123,
|
| 332 |
+
"step": 1472
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.6768676867686768,
|
| 336 |
+
"grad_norm": 12.995174407958984,
|
| 337 |
+
"learning_rate": 9.535067553569175e-06,
|
| 338 |
+
"loss": 0.5784,
|
| 339 |
+
"step": 1504
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.6912691269126913,
|
| 343 |
+
"grad_norm": 21.58072853088379,
|
| 344 |
+
"learning_rate": 9.499148431397448e-06,
|
| 345 |
+
"loss": 0.6081,
|
| 346 |
+
"step": 1536
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.7056705670567057,
|
| 350 |
+
"grad_norm": 16.922595977783203,
|
| 351 |
+
"learning_rate": 9.461965847367611e-06,
|
| 352 |
+
"loss": 0.6303,
|
| 353 |
+
"step": 1568
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.7200720072007201,
|
| 357 |
+
"grad_norm": 17.760387420654297,
|
| 358 |
+
"learning_rate": 9.423530243183783e-06,
|
| 359 |
+
"loss": 0.5761,
|
| 360 |
+
"step": 1600
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.7344734473447345,
|
| 364 |
+
"grad_norm": 11.714215278625488,
|
| 365 |
+
"learning_rate": 9.385111038077417e-06,
|
| 366 |
+
"loss": 0.5852,
|
| 367 |
+
"step": 1632
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.7488748874887489,
|
| 371 |
+
"grad_norm": 12.36257266998291,
|
| 372 |
+
"learning_rate": 9.344240422449167e-06,
|
| 373 |
+
"loss": 0.5965,
|
| 374 |
+
"step": 1664
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.7632763276327633,
|
| 378 |
+
"grad_norm": 14.300825119018555,
|
| 379 |
+
"learning_rate": 9.302149846608464e-06,
|
| 380 |
+
"loss": 0.5316,
|
| 381 |
+
"step": 1696
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.7776777677767777,
|
| 385 |
+
"grad_norm": 16.075668334960938,
|
| 386 |
+
"learning_rate": 9.25885113053368e-06,
|
| 387 |
+
"loss": 0.6228,
|
| 388 |
+
"step": 1728
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.7920792079207921,
|
| 392 |
+
"grad_norm": 11.793745040893555,
|
| 393 |
+
"learning_rate": 9.214356433476091e-06,
|
| 394 |
+
"loss": 0.5644,
|
| 395 |
+
"step": 1760
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.8064806480648065,
|
| 399 |
+
"grad_norm": 12.965546607971191,
|
| 400 |
+
"learning_rate": 9.168678250545255e-06,
|
| 401 |
+
"loss": 0.5646,
|
| 402 |
+
"step": 1792
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.8208820882088209,
|
| 406 |
+
"grad_norm": 9.327117919921875,
|
| 407 |
+
"learning_rate": 9.121829409200145e-06,
|
| 408 |
+
"loss": 0.5961,
|
| 409 |
+
"step": 1824
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.8352835283528353,
|
| 413 |
+
"grad_norm": 9.750507354736328,
|
| 414 |
+
"learning_rate": 9.073823065646882e-06,
|
| 415 |
+
"loss": 0.6158,
|
| 416 |
+
"step": 1856
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.8496849684968497,
|
| 420 |
+
"grad_norm": 17.131214141845703,
|
| 421 |
+
"learning_rate": 9.024672701144184e-06,
|
| 422 |
+
"loss": 0.5422,
|
| 423 |
+
"step": 1888
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.8640864086408641,
|
| 427 |
+
"grad_norm": 9.129823684692383,
|
| 428 |
+
"learning_rate": 8.97439211821753e-06,
|
| 429 |
+
"loss": 0.5846,
|
| 430 |
+
"step": 1920
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.8784878487848785,
|
| 434 |
+
"grad_norm": 9.996623039245605,
|
| 435 |
+
"learning_rate": 8.922995436783104e-06,
|
| 436 |
+
"loss": 0.5824,
|
| 437 |
+
"step": 1952
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8928892889288929,
|
| 441 |
+
"grad_norm": 12.214370727539062,
|
| 442 |
+
"learning_rate": 8.870497090182593e-06,
|
| 443 |
+
"loss": 0.5921,
|
| 444 |
+
"step": 1984
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.9072907290729073,
|
| 448 |
+
"grad_norm": 12.267783164978027,
|
| 449 |
+
"learning_rate": 8.816911821129992e-06,
|
| 450 |
+
"loss": 0.5496,
|
| 451 |
+
"step": 2016
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.9216921692169217,
|
| 455 |
+
"grad_norm": 11.49367618560791,
|
| 456 |
+
"learning_rate": 8.762254677571517e-06,
|
| 457 |
+
"loss": 0.6221,
|
| 458 |
+
"step": 2048
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.9360936093609361,
|
| 462 |
+
"grad_norm": 11.387689590454102,
|
| 463 |
+
"learning_rate": 8.706541008459798e-06,
|
| 464 |
+
"loss": 0.5685,
|
| 465 |
+
"step": 2080
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.9504950495049505,
|
| 469 |
+
"grad_norm": 12.342247009277344,
|
| 470 |
+
"learning_rate": 8.649786459443555e-06,
|
| 471 |
+
"loss": 0.5922,
|
| 472 |
+
"step": 2112
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.9648964896489649,
|
| 476 |
+
"grad_norm": 17.48115348815918,
|
| 477 |
+
"learning_rate": 8.59200696847395e-06,
|
| 478 |
+
"loss": 0.6309,
|
| 479 |
+
"step": 2144
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.9792979297929792,
|
| 483 |
+
"grad_norm": 13.321036338806152,
|
| 484 |
+
"learning_rate": 8.533218761328843e-06,
|
| 485 |
+
"loss": 0.5513,
|
| 486 |
+
"step": 2176
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.9936993699369937,
|
| 490 |
+
"grad_norm": 13.628349304199219,
|
| 491 |
+
"learning_rate": 8.473438347056239e-06,
|
| 492 |
+
"loss": 0.5934,
|
| 493 |
+
"step": 2208
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.008100810081008,
|
| 497 |
+
"grad_norm": 6.996334075927734,
|
| 498 |
+
"learning_rate": 8.412682513338176e-06,
|
| 499 |
+
"loss": 0.4909,
|
| 500 |
+
"step": 2240
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.0225022502250225,
|
| 504 |
+
"grad_norm": 10.135323524475098,
|
| 505 |
+
"learning_rate": 8.35096832177636e-06,
|
| 506 |
+
"loss": 0.3238,
|
| 507 |
+
"step": 2272
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.036903690369037,
|
| 511 |
+
"grad_norm": 7.344214916229248,
|
| 512 |
+
"learning_rate": 8.28831310310089e-06,
|
| 513 |
+
"loss": 0.297,
|
| 514 |
+
"step": 2304
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.0513051305130514,
|
| 518 |
+
"grad_norm": 8.131421089172363,
|
| 519 |
+
"learning_rate": 8.224734452303397e-06,
|
| 520 |
+
"loss": 0.3335,
|
| 521 |
+
"step": 2336
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.0657065706570656,
|
| 525 |
+
"grad_norm": 8.626580238342285,
|
| 526 |
+
"learning_rate": 8.160250223695987e-06,
|
| 527 |
+
"loss": 0.3357,
|
| 528 |
+
"step": 2368
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.08010801080108,
|
| 532 |
+
"grad_norm": 20.845579147338867,
|
| 533 |
+
"learning_rate": 8.094878525897325e-06,
|
| 534 |
+
"loss": 0.3235,
|
| 535 |
+
"step": 2400
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.0945094509450946,
|
| 539 |
+
"grad_norm": 13.4508695602417,
|
| 540 |
+
"learning_rate": 8.028637716747355e-06,
|
| 541 |
+
"loss": 0.3331,
|
| 542 |
+
"step": 2432
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.108910891089109,
|
| 546 |
+
"grad_norm": 11.288851737976074,
|
| 547 |
+
"learning_rate": 7.961546398151988e-06,
|
| 548 |
+
"loss": 0.3425,
|
| 549 |
+
"step": 2464
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1233123312331232,
|
| 553 |
+
"grad_norm": 12.70616626739502,
|
| 554 |
+
"learning_rate": 7.893623410859282e-06,
|
| 555 |
+
"loss": 0.343,
|
| 556 |
+
"step": 2496
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.1377137713771377,
|
| 560 |
+
"grad_norm": 14.446159362792969,
|
| 561 |
+
"learning_rate": 7.824887829168522e-06,
|
| 562 |
+
"loss": 0.3426,
|
| 563 |
+
"step": 2528
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.1521152115211521,
|
| 567 |
+
"grad_norm": 9.85571575164795,
|
| 568 |
+
"learning_rate": 7.755358955573747e-06,
|
| 569 |
+
"loss": 0.3229,
|
| 570 |
+
"step": 2560
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.1665166516651666,
|
| 574 |
+
"grad_norm": 16.067195892333984,
|
| 575 |
+
"learning_rate": 7.685056315343165e-06,
|
| 576 |
+
"loss": 0.3246,
|
| 577 |
+
"step": 2592
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.1809180918091808,
|
| 581 |
+
"grad_norm": 20.33487319946289,
|
| 582 |
+
"learning_rate": 7.613999651036016e-06,
|
| 583 |
+
"loss": 0.374,
|
| 584 |
+
"step": 2624
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.1953195319531953,
|
| 588 |
+
"grad_norm": 12.800278663635254,
|
| 589 |
+
"learning_rate": 7.542208916958433e-06,
|
| 590 |
+
"loss": 0.3008,
|
| 591 |
+
"step": 2656
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.2097209720972097,
|
| 595 |
+
"grad_norm": 10.589587211608887,
|
| 596 |
+
"learning_rate": 7.469704273559807e-06,
|
| 597 |
+
"loss": 0.3738,
|
| 598 |
+
"step": 2688
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.2241224122412242,
|
| 602 |
+
"grad_norm": 9.341863632202148,
|
| 603 |
+
"learning_rate": 7.396506081771295e-06,
|
| 604 |
+
"loss": 0.3035,
|
| 605 |
+
"step": 2720
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.2385238523852384,
|
| 609 |
+
"grad_norm": 8.244071960449219,
|
| 610 |
+
"learning_rate": 7.322634897288008e-06,
|
| 611 |
+
"loss": 0.3748,
|
| 612 |
+
"step": 2752
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.2529252925292529,
|
| 616 |
+
"grad_norm": 6.00961971282959,
|
| 617 |
+
"learning_rate": 7.248111464796508e-06,
|
| 618 |
+
"loss": 0.2727,
|
| 619 |
+
"step": 2784
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.2673267326732673,
|
| 623 |
+
"grad_norm": 9.179043769836426,
|
| 624 |
+
"learning_rate": 7.172956712149234e-06,
|
| 625 |
+
"loss": 0.3004,
|
| 626 |
+
"step": 2816
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.2817281728172818,
|
| 630 |
+
"grad_norm": 9.150934219360352,
|
| 631 |
+
"learning_rate": 7.0971917444875015e-06,
|
| 632 |
+
"loss": 0.3635,
|
| 633 |
+
"step": 2848
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.296129612961296,
|
| 637 |
+
"grad_norm": 9.036107063293457,
|
| 638 |
+
"learning_rate": 7.020837838314691e-06,
|
| 639 |
+
"loss": 0.2943,
|
| 640 |
+
"step": 2880
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.3105310531053105,
|
| 644 |
+
"grad_norm": 9.950891494750977,
|
| 645 |
+
"learning_rate": 6.94391643552134e-06,
|
| 646 |
+
"loss": 0.3727,
|
| 647 |
+
"step": 2912
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.324932493249325,
|
| 651 |
+
"grad_norm": 15.246482849121094,
|
| 652 |
+
"learning_rate": 6.866449137363768e-06,
|
| 653 |
+
"loss": 0.2992,
|
| 654 |
+
"step": 2944
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.3393339333933394,
|
| 658 |
+
"grad_norm": 10.054420471191406,
|
| 659 |
+
"learning_rate": 6.788457698397973e-06,
|
| 660 |
+
"loss": 0.3322,
|
| 661 |
+
"step": 2976
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.3537353735373538,
|
| 665 |
+
"grad_norm": 10.79068374633789,
|
| 666 |
+
"learning_rate": 6.709964020370445e-06,
|
| 667 |
+
"loss": 0.3245,
|
| 668 |
+
"step": 3008
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.368136813681368,
|
| 672 |
+
"grad_norm": 10.584324836730957,
|
| 673 |
+
"learning_rate": 6.630990146067687e-06,
|
| 674 |
+
"loss": 0.3418,
|
| 675 |
+
"step": 3040
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.3825382538253825,
|
| 679 |
+
"grad_norm": 16.110841751098633,
|
| 680 |
+
"learning_rate": 6.554047211421132e-06,
|
| 681 |
+
"loss": 0.3437,
|
| 682 |
+
"step": 3072
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.396939693969397,
|
| 686 |
+
"grad_norm": 11.55324649810791,
|
| 687 |
+
"learning_rate": 6.4741928832357855e-06,
|
| 688 |
+
"loss": 0.3011,
|
| 689 |
+
"step": 3104
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.4113411341134112,
|
| 693 |
+
"grad_norm": 15.144320487976074,
|
| 694 |
+
"learning_rate": 6.393924568602145e-06,
|
| 695 |
+
"loss": 0.337,
|
| 696 |
+
"step": 3136
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.4257425742574257,
|
| 700 |
+
"grad_norm": 10.759925842285156,
|
| 701 |
+
"learning_rate": 6.313264808664494e-06,
|
| 702 |
+
"loss": 0.3149,
|
| 703 |
+
"step": 3168
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.4401440144014401,
|
| 707 |
+
"grad_norm": 14.752134323120117,
|
| 708 |
+
"learning_rate": 6.232236254493746e-06,
|
| 709 |
+
"loss": 0.3622,
|
| 710 |
+
"step": 3200
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.4545454545454546,
|
| 714 |
+
"grad_norm": 11.484366416931152,
|
| 715 |
+
"learning_rate": 6.150861660726515e-06,
|
| 716 |
+
"loss": 0.3112,
|
| 717 |
+
"step": 3232
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.468946894689469,
|
| 721 |
+
"grad_norm": 13.652670860290527,
|
| 722 |
+
"learning_rate": 6.069163879175092e-06,
|
| 723 |
+
"loss": 0.3403,
|
| 724 |
+
"step": 3264
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.4833483348334833,
|
| 728 |
+
"grad_norm": 9.8715181350708,
|
| 729 |
+
"learning_rate": 5.9871658524101565e-06,
|
| 730 |
+
"loss": 0.3205,
|
| 731 |
+
"step": 3296
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.4977497749774977,
|
| 735 |
+
"grad_norm": 11.28494644165039,
|
| 736 |
+
"learning_rate": 5.9048906073179824e-06,
|
| 737 |
+
"loss": 0.304,
|
| 738 |
+
"step": 3328
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.5121512151215122,
|
| 742 |
+
"grad_norm": 13.887341499328613,
|
| 743 |
+
"learning_rate": 5.822361248633973e-06,
|
| 744 |
+
"loss": 0.3027,
|
| 745 |
+
"step": 3360
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.5265526552655264,
|
| 749 |
+
"grad_norm": 7.218605041503906,
|
| 750 |
+
"learning_rate": 5.7396009524543274e-06,
|
| 751 |
+
"loss": 0.3081,
|
| 752 |
+
"step": 3392
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5409540954095409,
|
| 756 |
+
"grad_norm": 7.5199503898620605,
|
| 757 |
+
"learning_rate": 5.656632959727683e-06,
|
| 758 |
+
"loss": 0.2977,
|
| 759 |
+
"step": 3424
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5553555355535553,
|
| 763 |
+
"grad_norm": 9.384359359741211,
|
| 764 |
+
"learning_rate": 5.57348056972852e-06,
|
| 765 |
+
"loss": 0.3032,
|
| 766 |
+
"step": 3456
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.5697569756975698,
|
| 770 |
+
"grad_norm": 14.385334014892578,
|
| 771 |
+
"learning_rate": 5.492772883806706e-06,
|
| 772 |
+
"loss": 0.3244,
|
| 773 |
+
"step": 3488
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.5841584158415842,
|
| 777 |
+
"grad_norm": 3.620002269744873,
|
| 778 |
+
"learning_rate": 5.40932574467665e-06,
|
| 779 |
+
"loss": 0.2992,
|
| 780 |
+
"step": 3520
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.5985598559855987,
|
| 784 |
+
"grad_norm": 9.094705581665039,
|
| 785 |
+
"learning_rate": 5.325763657690609e-06,
|
| 786 |
+
"loss": 0.2942,
|
| 787 |
+
"step": 3552
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.612961296129613,
|
| 791 |
+
"grad_norm": 10.146318435668945,
|
| 792 |
+
"learning_rate": 5.242110088958073e-06,
|
| 793 |
+
"loss": 0.3395,
|
| 794 |
+
"step": 3584
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.6273627362736274,
|
| 798 |
+
"grad_norm": 13.460920333862305,
|
| 799 |
+
"learning_rate": 5.158388530278656e-06,
|
| 800 |
+
"loss": 0.3097,
|
| 801 |
+
"step": 3616
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.6417641764176416,
|
| 805 |
+
"grad_norm": 11.407035827636719,
|
| 806 |
+
"learning_rate": 5.074622492545074e-06,
|
| 807 |
+
"loss": 0.3159,
|
| 808 |
+
"step": 3648
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.656165616561656,
|
| 812 |
+
"grad_norm": 10.230562210083008,
|
| 813 |
+
"learning_rate": 4.9908354991407666e-06,
|
| 814 |
+
"loss": 0.327,
|
| 815 |
+
"step": 3680
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.6705670567056705,
|
| 819 |
+
"grad_norm": 11.073262214660645,
|
| 820 |
+
"learning_rate": 4.9070510793339835e-06,
|
| 821 |
+
"loss": 0.2981,
|
| 822 |
+
"step": 3712
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.684968496849685,
|
| 826 |
+
"grad_norm": 12.47003173828125,
|
| 827 |
+
"learning_rate": 4.823292761670264e-06,
|
| 828 |
+
"loss": 0.3065,
|
| 829 |
+
"step": 3744
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.6993699369936994,
|
| 833 |
+
"grad_norm": 25.917009353637695,
|
| 834 |
+
"learning_rate": 4.74219897937266e-06,
|
| 835 |
+
"loss": 0.2992,
|
| 836 |
+
"step": 3776
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.7137713771377139,
|
| 840 |
+
"grad_norm": 11.52535629272461,
|
| 841 |
+
"learning_rate": 4.658560774737667e-06,
|
| 842 |
+
"loss": 0.3474,
|
| 843 |
+
"step": 3808
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.7281728172817283,
|
| 847 |
+
"grad_norm": 9.606013298034668,
|
| 848 |
+
"learning_rate": 4.5750184539003665e-06,
|
| 849 |
+
"loss": 0.3436,
|
| 850 |
+
"step": 3840
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.7425742574257426,
|
| 854 |
+
"grad_norm": 15.367290496826172,
|
| 855 |
+
"learning_rate": 4.4915954774194676e-06,
|
| 856 |
+
"loss": 0.2932,
|
| 857 |
+
"step": 3872
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.756975697569757,
|
| 861 |
+
"grad_norm": 7.993281364440918,
|
| 862 |
+
"learning_rate": 4.408315272339104e-06,
|
| 863 |
+
"loss": 0.3203,
|
| 864 |
+
"step": 3904
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.7713771377137713,
|
| 868 |
+
"grad_norm": 8.027710914611816,
|
| 869 |
+
"learning_rate": 4.325201225609999e-06,
|
| 870 |
+
"loss": 0.3139,
|
| 871 |
+
"step": 3936
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.7857785778577857,
|
| 875 |
+
"grad_norm": 10.957657814025879,
|
| 876 |
+
"learning_rate": 4.242276677521877e-06,
|
| 877 |
+
"loss": 0.3453,
|
| 878 |
+
"step": 3968
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.8001800180018002,
|
| 882 |
+
"grad_norm": 10.544370651245117,
|
| 883 |
+
"learning_rate": 4.159564915148997e-06,
|
| 884 |
+
"loss": 0.2853,
|
| 885 |
+
"step": 4000
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.8145814581458146,
|
| 889 |
+
"grad_norm": 12.427223205566406,
|
| 890 |
+
"learning_rate": 4.077089165810611e-06,
|
| 891 |
+
"loss": 0.3355,
|
| 892 |
+
"step": 4032
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.828982898289829,
|
| 896 |
+
"grad_norm": 18.81423568725586,
|
| 897 |
+
"learning_rate": 3.994872590548211e-06,
|
| 898 |
+
"loss": 0.328,
|
| 899 |
+
"step": 4064
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.8433843384338435,
|
| 903 |
+
"grad_norm": 9.081976890563965,
|
| 904 |
+
"learning_rate": 3.9129382776213945e-06,
|
| 905 |
+
"loss": 0.293,
|
| 906 |
+
"step": 4096
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.8577857785778578,
|
| 910 |
+
"grad_norm": 8.164251327514648,
|
| 911 |
+
"learning_rate": 3.831309236024159e-06,
|
| 912 |
+
"loss": 0.2782,
|
| 913 |
+
"step": 4128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.8721872187218722,
|
| 917 |
+
"grad_norm": 9.661165237426758,
|
| 918 |
+
"learning_rate": 3.7500083890234606e-06,
|
| 919 |
+
"loss": 0.3296,
|
| 920 |
+
"step": 4160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.8865886588658864,
|
| 924 |
+
"grad_norm": 14.657001495361328,
|
| 925 |
+
"learning_rate": 3.66905856772185e-06,
|
| 926 |
+
"loss": 0.2631,
|
| 927 |
+
"step": 4192
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.900990099009901,
|
| 931 |
+
"grad_norm": 16.178007125854492,
|
| 932 |
+
"learning_rate": 3.5884825046459805e-06,
|
| 933 |
+
"loss": 0.2765,
|
| 934 |
+
"step": 4224
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.9153915391539154,
|
| 938 |
+
"grad_norm": 7.737805366516113,
|
| 939 |
+
"learning_rate": 3.508302827362805e-06,
|
| 940 |
+
"loss": 0.317,
|
| 941 |
+
"step": 4256
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.9297929792979298,
|
| 945 |
+
"grad_norm": 9.958755493164062,
|
| 946 |
+
"learning_rate": 3.4285420521252533e-06,
|
| 947 |
+
"loss": 0.2975,
|
| 948 |
+
"step": 4288
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.9441944194419443,
|
| 952 |
+
"grad_norm": 13.483292579650879,
|
| 953 |
+
"learning_rate": 3.3492225775491582e-06,
|
| 954 |
+
"loss": 0.2776,
|
| 955 |
+
"step": 4320
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.9585958595859587,
|
| 959 |
+
"grad_norm": 11.092999458312988,
|
| 960 |
+
"learning_rate": 3.270366678323219e-06,
|
| 961 |
+
"loss": 0.3453,
|
| 962 |
+
"step": 4352
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.972997299729973,
|
| 966 |
+
"grad_norm": 11.395092964172363,
|
| 967 |
+
"learning_rate": 3.1919964989537755e-06,
|
| 968 |
+
"loss": 0.2678,
|
| 969 |
+
"step": 4384
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.9873987398739874,
|
| 973 |
+
"grad_norm": 10.83203411102295,
|
| 974 |
+
"learning_rate": 3.1141340475461316e-06,
|
| 975 |
+
"loss": 0.3074,
|
| 976 |
+
"step": 4416
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 2.0018001800180016,
|
| 980 |
+
"grad_norm": 6.892433166503906,
|
| 981 |
+
"learning_rate": 3.03680118962418e-06,
|
| 982 |
+
"loss": 0.3064,
|
| 983 |
+
"step": 4448
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.016201620162016,
|
| 987 |
+
"grad_norm": 6.377696514129639,
|
| 988 |
+
"learning_rate": 2.9600196419900795e-06,
|
| 989 |
+
"loss": 0.1179,
|
| 990 |
+
"step": 4480
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 2.0306030603060305,
|
| 994 |
+
"grad_norm": 8.865036964416504,
|
| 995 |
+
"learning_rate": 2.883810966625684e-06,
|
| 996 |
+
"loss": 0.1444,
|
| 997 |
+
"step": 4512
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 2.045004500450045,
|
| 1001 |
+
"grad_norm": 7.50280237197876,
|
| 1002 |
+
"learning_rate": 2.8081965646374582e-06,
|
| 1003 |
+
"loss": 0.1138,
|
| 1004 |
+
"step": 4544
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.0594059405940595,
|
| 1008 |
+
"grad_norm": 9.762062072753906,
|
| 1009 |
+
"learning_rate": 2.7331976702465647e-06,
|
| 1010 |
+
"loss": 0.1251,
|
| 1011 |
+
"step": 4576
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 2.073807380738074,
|
| 1015 |
+
"grad_norm": 10.026853561401367,
|
| 1016 |
+
"learning_rate": 2.658835344825821e-06,
|
| 1017 |
+
"loss": 0.1239,
|
| 1018 |
+
"step": 4608
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 2.0882088208820884,
|
| 1022 |
+
"grad_norm": 10.256850242614746,
|
| 1023 |
+
"learning_rate": 2.5851304709851855e-06,
|
| 1024 |
+
"loss": 0.1131,
|
| 1025 |
+
"step": 4640
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 2.102610261026103,
|
| 1029 |
+
"grad_norm": 11.005268096923828,
|
| 1030 |
+
"learning_rate": 2.5121037467074596e-06,
|
| 1031 |
+
"loss": 0.1224,
|
| 1032 |
+
"step": 4672
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 2.117011701170117,
|
| 1036 |
+
"grad_norm": 11.805505752563477,
|
| 1037 |
+
"learning_rate": 2.4397756795358287e-06,
|
| 1038 |
+
"loss": 0.1101,
|
| 1039 |
+
"step": 4704
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 2.1314131413141313,
|
| 1043 |
+
"grad_norm": 7.897261142730713,
|
| 1044 |
+
"learning_rate": 2.3703932820444233e-06,
|
| 1045 |
+
"loss": 0.1238,
|
| 1046 |
+
"step": 4736
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 2.1458145814581457,
|
| 1050 |
+
"grad_norm": 10.117232322692871,
|
| 1051 |
+
"learning_rate": 2.2994998626884623e-06,
|
| 1052 |
+
"loss": 0.1072,
|
| 1053 |
+
"step": 4768
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 2.16021602160216,
|
| 1057 |
+
"grad_norm": 9.720394134521484,
|
| 1058 |
+
"learning_rate": 2.2293648043808946e-06,
|
| 1059 |
+
"loss": 0.1377,
|
| 1060 |
+
"step": 4800
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 2.1746174617461747,
|
| 1064 |
+
"grad_norm": 9.53496265411377,
|
| 1065 |
+
"learning_rate": 2.1600078026201977e-06,
|
| 1066 |
+
"loss": 0.114,
|
| 1067 |
+
"step": 4832
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.189018901890189,
|
| 1071 |
+
"grad_norm": 9.775762557983398,
|
| 1072 |
+
"learning_rate": 2.091448334409112e-06,
|
| 1073 |
+
"loss": 0.1053,
|
| 1074 |
+
"step": 4864
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.2034203420342036,
|
| 1078 |
+
"grad_norm": 11.364226341247559,
|
| 1079 |
+
"learning_rate": 2.0237056527850555e-06,
|
| 1080 |
+
"loss": 0.0966,
|
| 1081 |
+
"step": 4896
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.217821782178218,
|
| 1085 |
+
"grad_norm": 17.391111373901367,
|
| 1086 |
+
"learning_rate": 1.95679878141344e-06,
|
| 1087 |
+
"loss": 0.1016,
|
| 1088 |
+
"step": 4928
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.232223222322232,
|
| 1092 |
+
"grad_norm": 10.15132999420166,
|
| 1093 |
+
"learning_rate": 1.8907465092453986e-06,
|
| 1094 |
+
"loss": 0.1184,
|
| 1095 |
+
"step": 4960
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 2.2466246624662465,
|
| 1099 |
+
"grad_norm": 12.572965621948242,
|
| 1100 |
+
"learning_rate": 1.8255673852414274e-06,
|
| 1101 |
+
"loss": 0.1015,
|
| 1102 |
+
"step": 4992
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 2.261026102610261,
|
| 1106 |
+
"grad_norm": 5.2437968254089355,
|
| 1107 |
+
"learning_rate": 1.7612797131624243e-06,
|
| 1108 |
+
"loss": 0.0993,
|
| 1109 |
+
"step": 5024
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.2754275427542754,
|
| 1113 |
+
"grad_norm": 8.918633460998535,
|
| 1114 |
+
"learning_rate": 1.6979015464295785e-06,
|
| 1115 |
+
"loss": 0.1079,
|
| 1116 |
+
"step": 5056
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 2.28982898289829,
|
| 1120 |
+
"grad_norm": 9.719175338745117,
|
| 1121 |
+
"learning_rate": 1.6354506830545625e-06,
|
| 1122 |
+
"loss": 0.1016,
|
| 1123 |
+
"step": 5088
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 2.3042304230423043,
|
| 1127 |
+
"grad_norm": 24.36786460876465,
|
| 1128 |
+
"learning_rate": 1.5739446606414522e-06,
|
| 1129 |
+
"loss": 0.1342,
|
| 1130 |
+
"step": 5120
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.3186318631863188,
|
| 1134 |
+
"grad_norm": 7.500924587249756,
|
| 1135 |
+
"learning_rate": 1.5134007514617827e-06,
|
| 1136 |
+
"loss": 0.0955,
|
| 1137 |
+
"step": 5152
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 2.333033303330333,
|
| 1141 |
+
"grad_norm": 10.143550872802734,
|
| 1142 |
+
"learning_rate": 1.4538359576040923e-06,
|
| 1143 |
+
"loss": 0.1451,
|
| 1144 |
+
"step": 5184
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 2.3474347434743477,
|
| 1148 |
+
"grad_norm": 15.87746524810791,
|
| 1149 |
+
"learning_rate": 1.395267006199363e-06,
|
| 1150 |
+
"loss": 0.1329,
|
| 1151 |
+
"step": 5216
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 2.3618361836183617,
|
| 1155 |
+
"grad_norm": 12.571290016174316,
|
| 1156 |
+
"learning_rate": 1.33771034472367e-06,
|
| 1157 |
+
"loss": 0.0978,
|
| 1158 |
+
"step": 5248
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 2.376237623762376,
|
| 1162 |
+
"grad_norm": 11.44887924194336,
|
| 1163 |
+
"learning_rate": 1.2811821363793497e-06,
|
| 1164 |
+
"loss": 0.1037,
|
| 1165 |
+
"step": 5280
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 2.3906390639063906,
|
| 1169 |
+
"grad_norm": 11.863057136535645,
|
| 1170 |
+
"learning_rate": 1.2256982555560243e-06,
|
| 1171 |
+
"loss": 0.116,
|
| 1172 |
+
"step": 5312
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.405040504050405,
|
| 1176 |
+
"grad_norm": 5.990699768066406,
|
| 1177 |
+
"learning_rate": 1.171274283372703e-06,
|
| 1178 |
+
"loss": 0.1037,
|
| 1179 |
+
"step": 5344
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.4194419441944195,
|
| 1183 |
+
"grad_norm": 10.163458824157715,
|
| 1184 |
+
"learning_rate": 1.1179255033022624e-06,
|
| 1185 |
+
"loss": 0.1387,
|
| 1186 |
+
"step": 5376
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 2.433843384338434,
|
| 1190 |
+
"grad_norm": 11.073272705078125,
|
| 1191 |
+
"learning_rate": 1.0672833301104142e-06,
|
| 1192 |
+
"loss": 0.12,
|
| 1193 |
+
"step": 5408
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 2.4482448244824484,
|
| 1197 |
+
"grad_norm": 10.601714134216309,
|
| 1198 |
+
"learning_rate": 1.016094827126849e-06,
|
| 1199 |
+
"loss": 0.0957,
|
| 1200 |
+
"step": 5440
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 2.4626462646264624,
|
| 1204 |
+
"grad_norm": 7.653828144073486,
|
| 1205 |
+
"learning_rate": 9.660250941303178e-07,
|
| 1206 |
+
"loss": 0.113,
|
| 1207 |
+
"step": 5472
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 2.477047704770477,
|
| 1211 |
+
"grad_norm": 10.332706451416016,
|
| 1212 |
+
"learning_rate": 9.170881918256042e-07,
|
| 1213 |
+
"loss": 0.1001,
|
| 1214 |
+
"step": 5504
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 2.4914491449144913,
|
| 1218 |
+
"grad_norm": 7.20733642578125,
|
| 1219 |
+
"learning_rate": 8.692978627932148e-07,
|
| 1220 |
+
"loss": 0.0998,
|
| 1221 |
+
"step": 5536
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 2.5058505850585058,
|
| 1225 |
+
"grad_norm": 4.12467622756958,
|
| 1226 |
+
"learning_rate": 8.226675276301416e-07,
|
| 1227 |
+
"loss": 0.0855,
|
| 1228 |
+
"step": 5568
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 2.5202520252025202,
|
| 1232 |
+
"grad_norm": 9.155281066894531,
|
| 1233 |
+
"learning_rate": 7.772102811810689e-07,
|
| 1234 |
+
"loss": 0.122,
|
| 1235 |
+
"step": 5600
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 2.5346534653465347,
|
| 1239 |
+
"grad_norm": 5.481142997741699,
|
| 1240 |
+
"learning_rate": 7.329388888610384e-07,
|
| 1241 |
+
"loss": 0.0978,
|
| 1242 |
+
"step": 5632
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 2.549054905490549,
|
| 1246 |
+
"grad_norm": 7.971097946166992,
|
| 1247 |
+
"learning_rate": 6.898657830706367e-07,
|
| 1248 |
+
"loss": 0.1157,
|
| 1249 |
+
"step": 5664
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 2.5634563456345636,
|
| 1253 |
+
"grad_norm": 11.405919075012207,
|
| 1254 |
+
"learning_rate": 6.492928309381779e-07,
|
| 1255 |
+
"loss": 0.0979,
|
| 1256 |
+
"step": 5696
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.577857785778578,
|
| 1260 |
+
"grad_norm": 13.61896800994873,
|
| 1261 |
+
"learning_rate": 6.08613879617217e-07,
|
| 1262 |
+
"loss": 0.1041,
|
| 1263 |
+
"step": 5728
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.592259225922592,
|
| 1267 |
+
"grad_norm": 29.495868682861328,
|
| 1268 |
+
"learning_rate": 5.691681280788214e-07,
|
| 1269 |
+
"loss": 0.0992,
|
| 1270 |
+
"step": 5760
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.6066606660666065,
|
| 1274 |
+
"grad_norm": 11.616131782531738,
|
| 1275 |
+
"learning_rate": 5.309666535753417e-07,
|
| 1276 |
+
"loss": 0.1051,
|
| 1277 |
+
"step": 5792
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.621062106210621,
|
| 1281 |
+
"grad_norm": 8.599799156188965,
|
| 1282 |
+
"learning_rate": 4.940201839382114e-07,
|
| 1283 |
+
"loss": 0.1091,
|
| 1284 |
+
"step": 5824
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6354635463546354,
|
| 1288 |
+
"grad_norm": 12.70768928527832,
|
| 1289 |
+
"learning_rate": 4.5833909456532764e-07,
|
| 1290 |
+
"loss": 0.1111,
|
| 1291 |
+
"step": 5856
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.64986498649865,
|
| 1295 |
+
"grad_norm": 11.202815055847168,
|
| 1296 |
+
"learning_rate": 4.2393340550740844e-07,
|
| 1297 |
+
"loss": 0.1046,
|
| 1298 |
+
"step": 5888
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 2.6642664266426643,
|
| 1302 |
+
"grad_norm": 11.233574867248535,
|
| 1303 |
+
"learning_rate": 3.908127786541427e-07,
|
| 1304 |
+
"loss": 0.1022,
|
| 1305 |
+
"step": 5920
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 2.678667866786679,
|
| 1309 |
+
"grad_norm": 9.141092300415039,
|
| 1310 |
+
"learning_rate": 3.589865150209071e-07,
|
| 1311 |
+
"loss": 0.0984,
|
| 1312 |
+
"step": 5952
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 2.693069306930693,
|
| 1316 |
+
"grad_norm": 15.323395729064941,
|
| 1317 |
+
"learning_rate": 3.2846355213683456e-07,
|
| 1318 |
+
"loss": 0.1279,
|
| 1319 |
+
"step": 5984
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 2.7074707470747077,
|
| 1323 |
+
"grad_norm": 10.427877426147461,
|
| 1324 |
+
"learning_rate": 2.9925246153496067e-07,
|
| 1325 |
+
"loss": 0.1169,
|
| 1326 |
+
"step": 6016
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 2.7218721872187217,
|
| 1330 |
+
"grad_norm": 12.140647888183594,
|
| 1331 |
+
"learning_rate": 2.713614463451364e-07,
|
| 1332 |
+
"loss": 0.1267,
|
| 1333 |
+
"step": 6048
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 2.736273627362736,
|
| 1337 |
+
"grad_norm": 13.221187591552734,
|
| 1338 |
+
"learning_rate": 2.4479833899041183e-07,
|
| 1339 |
+
"loss": 0.1198,
|
| 1340 |
+
"step": 6080
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.7506750675067506,
|
| 1344 |
+
"grad_norm": 5.667657852172852,
|
| 1345 |
+
"learning_rate": 2.19570598987513e-07,
|
| 1346 |
+
"loss": 0.1312,
|
| 1347 |
+
"step": 6112
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 2.765076507650765,
|
| 1351 |
+
"grad_norm": 0.735275149345398,
|
| 1352 |
+
"learning_rate": 1.9568531085204067e-07,
|
| 1353 |
+
"loss": 0.0699,
|
| 1354 |
+
"step": 6144
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 2.7794779477947795,
|
| 1358 |
+
"grad_norm": 4.510382652282715,
|
| 1359 |
+
"learning_rate": 1.731491821089848e-07,
|
| 1360 |
+
"loss": 0.0965,
|
| 1361 |
+
"step": 6176
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 2.793879387938794,
|
| 1365 |
+
"grad_norm": 13.006752967834473,
|
| 1366 |
+
"learning_rate": 1.5196854140909545e-07,
|
| 1367 |
+
"loss": 0.1026,
|
| 1368 |
+
"step": 6208
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 2.8082808280828084,
|
| 1372 |
+
"grad_norm": 10.59677791595459,
|
| 1373 |
+
"learning_rate": 1.321493367516574e-07,
|
| 1374 |
+
"loss": 0.1093,
|
| 1375 |
+
"step": 6240
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 2.8226822682268224,
|
| 1379 |
+
"grad_norm": 8.108485221862793,
|
| 1380 |
+
"learning_rate": 1.136971338141596e-07,
|
| 1381 |
+
"loss": 0.1191,
|
| 1382 |
+
"step": 6272
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.8370837083708373,
|
| 1386 |
+
"grad_norm": 11.026812553405762,
|
| 1387 |
+
"learning_rate": 9.661711438932686e-08,
|
| 1388 |
+
"loss": 0.1008,
|
| 1389 |
+
"step": 6304
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.8514851485148514,
|
| 1393 |
+
"grad_norm": 7.0519890785217285,
|
| 1394 |
+
"learning_rate": 8.09140749299564e-08,
|
| 1395 |
+
"loss": 0.1386,
|
| 1396 |
+
"step": 6336
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.865886588658866,
|
| 1400 |
+
"grad_norm": 27.898834228515625,
|
| 1401 |
+
"learning_rate": 6.659242520196562e-08,
|
| 1402 |
+
"loss": 0.103,
|
| 1403 |
+
"step": 6368
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 2.8802880288028803,
|
| 1407 |
+
"grad_norm": 7.882510185241699,
|
| 1408 |
+
"learning_rate": 5.365618704603392e-08,
|
| 1409 |
+
"loss": 0.0812,
|
| 1410 |
+
"step": 6400
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 2.8946894689468947,
|
| 1414 |
+
"grad_norm": 15.030416488647461,
|
| 1415 |
+
"learning_rate": 4.2108993248173855e-08,
|
| 1416 |
+
"loss": 0.102,
|
| 1417 |
+
"step": 6432
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 2.909090909090909,
|
| 1421 |
+
"grad_norm": 11.23343276977539,
|
| 1422 |
+
"learning_rate": 3.195408651956944e-08,
|
| 1423 |
+
"loss": 0.0976,
|
| 1424 |
+
"step": 6464
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.9234923492349236,
|
| 1428 |
+
"grad_norm": 6.837776184082031,
|
| 1429 |
+
"learning_rate": 2.3194318585945673e-08,
|
| 1430 |
+
"loss": 0.0938,
|
| 1431 |
+
"step": 6496
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 2.937893789378938,
|
| 1435 |
+
"grad_norm": 9.184088706970215,
|
| 1436 |
+
"learning_rate": 1.583214938674138e-08,
|
| 1437 |
+
"loss": 0.1087,
|
| 1438 |
+
"step": 6528
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.952295229522952,
|
| 1442 |
+
"grad_norm": 7.366724967956543,
|
| 1443 |
+
"learning_rate": 1.0034770609533285e-08,
|
| 1444 |
+
"loss": 0.1139,
|
| 1445 |
+
"step": 6560
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.9666966696669665,
|
| 1449 |
+
"grad_norm": 10.501226425170898,
|
| 1450 |
+
"learning_rate": 5.429794877803151e-09,
|
| 1451 |
+
"loss": 0.0928,
|
| 1452 |
+
"step": 6592
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 2.981098109810981,
|
| 1456 |
+
"grad_norm": 7.6273908615112305,
|
| 1457 |
+
"learning_rate": 2.2274065574556804e-09,
|
| 1458 |
+
"loss": 0.1124,
|
| 1459 |
+
"step": 6624
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 2.9954995499549955,
|
| 1463 |
+
"grad_norm": 5.487658977508545,
|
| 1464 |
+
"learning_rate": 4.2850495100610344e-10,
|
| 1465 |
+
"loss": 0.1158,
|
| 1466 |
+
"step": 6656
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 3.0,
|
| 1470 |
+
"step": 6666,
|
| 1471 |
+
"total_flos": 9491636163499776.0,
|
| 1472 |
+
"train_loss": 0.3371730904088448,
|
| 1473 |
+
"train_runtime": 4274.2917,
|
| 1474 |
+
"train_samples_per_second": 1.56,
|
| 1475 |
+
"train_steps_per_second": 1.56
|
| 1476 |
+
}
|
| 1477 |
+
],
|
| 1478 |
+
"logging_steps": 32,
|
| 1479 |
+
"max_steps": 6666,
|
| 1480 |
+
"num_input_tokens_seen": 0,
|
| 1481 |
+
"num_train_epochs": 3,
|
| 1482 |
+
"save_steps": 500,
|
| 1483 |
+
"stateful_callbacks": {
|
| 1484 |
+
"TrainerControl": {
|
| 1485 |
+
"args": {
|
| 1486 |
+
"should_epoch_stop": false,
|
| 1487 |
+
"should_evaluate": false,
|
| 1488 |
+
"should_log": false,
|
| 1489 |
+
"should_save": true,
|
| 1490 |
+
"should_training_stop": true
|
| 1491 |
+
},
|
| 1492 |
+
"attributes": {}
|
| 1493 |
+
}
|
| 1494 |
+
},
|
| 1495 |
+
"total_flos": 9491636163499776.0,
|
| 1496 |
+
"train_batch_size": 1,
|
| 1497 |
+
"trial_name": null,
|
| 1498 |
+
"trial_params": null
|
| 1499 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f6929da1178759e614ea0f5e12db1173429dad8fd8e288e950560f7021d95ff
|
| 3 |
+
size 5368
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|