Commit
•
4a73888
1
Parent(s):
416fb1b
First Commit For XLM-Language Detection Model
Browse files- .gitattributes +1 -0
- config.json +201 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +3 -0
- tokenizer_config.json +19 -0
- trainer_state.json +1536 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 32 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 33 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 32 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 33 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
|
@@ -0,0 +1,201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "xlm-roberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "nl",
|
| 15 |
+
"1": "fr",
|
| 16 |
+
"2": "fi",
|
| 17 |
+
"3": "de",
|
| 18 |
+
"4": "en",
|
| 19 |
+
"5": "el",
|
| 20 |
+
"6": "hr",
|
| 21 |
+
"7": "ro",
|
| 22 |
+
"8": "pt",
|
| 23 |
+
"9": "es",
|
| 24 |
+
"10": "cs",
|
| 25 |
+
"11": "da",
|
| 26 |
+
"12": "it",
|
| 27 |
+
"13": "hu",
|
| 28 |
+
"14": "ar",
|
| 29 |
+
"15": "sv",
|
| 30 |
+
"16": "pl",
|
| 31 |
+
"17": "id",
|
| 32 |
+
"18": "ms",
|
| 33 |
+
"19": "bg",
|
| 34 |
+
"20": "pam",
|
| 35 |
+
"21": "th",
|
| 36 |
+
"22": "ko",
|
| 37 |
+
"23": "zh",
|
| 38 |
+
"24": "zh-Hant",
|
| 39 |
+
"25": "vi",
|
| 40 |
+
"26": "bs",
|
| 41 |
+
"27": "ur",
|
| 42 |
+
"28": "und",
|
| 43 |
+
"29": "lb",
|
| 44 |
+
"30": "mr",
|
| 45 |
+
"31": "haw",
|
| 46 |
+
"32": "te",
|
| 47 |
+
"33": "la",
|
| 48 |
+
"34": "ug",
|
| 49 |
+
"35": "az",
|
| 50 |
+
"36": "sk",
|
| 51 |
+
"37": "tr",
|
| 52 |
+
"38": "gn",
|
| 53 |
+
"39": "ja",
|
| 54 |
+
"40": "is",
|
| 55 |
+
"41": "af",
|
| 56 |
+
"42": "gu",
|
| 57 |
+
"43": "uz",
|
| 58 |
+
"44": "hi",
|
| 59 |
+
"45": "rw",
|
| 60 |
+
"46": "sq",
|
| 61 |
+
"47": "co",
|
| 62 |
+
"48": "sl",
|
| 63 |
+
"49": "kk",
|
| 64 |
+
"50": "ru",
|
| 65 |
+
"51": "uk",
|
| 66 |
+
"52": "tt",
|
| 67 |
+
"53": "ky",
|
| 68 |
+
"54": "mk",
|
| 69 |
+
"55": "sr",
|
| 70 |
+
"56": "mi",
|
| 71 |
+
"57": "ga",
|
| 72 |
+
"58": "ml",
|
| 73 |
+
"59": "st",
|
| 74 |
+
"60": "no",
|
| 75 |
+
"61": "ku",
|
| 76 |
+
"62": "fil",
|
| 77 |
+
"63": "sd",
|
| 78 |
+
"64": "yo",
|
| 79 |
+
"65": "fy",
|
| 80 |
+
"66": "tk",
|
| 81 |
+
"67": "ln",
|
| 82 |
+
"68": "et",
|
| 83 |
+
"69": "lv",
|
| 84 |
+
"70": "hy",
|
| 85 |
+
"71": "ka",
|
| 86 |
+
"72": "mn",
|
| 87 |
+
"73": "tg",
|
| 88 |
+
"74": "ca",
|
| 89 |
+
"75": "lt",
|
| 90 |
+
"76": "kn",
|
| 91 |
+
"77": "ps",
|
| 92 |
+
"78": "mg",
|
| 93 |
+
"79": "hmn",
|
| 94 |
+
"80": "ceb",
|
| 95 |
+
"81": "bn",
|
| 96 |
+
"82": "my",
|
| 97 |
+
"83": "gl"
|
| 98 |
+
},
|
| 99 |
+
"initializer_range": 0.02,
|
| 100 |
+
"intermediate_size": 3072,
|
| 101 |
+
"label2id": {
|
| 102 |
+
"af": 41,
|
| 103 |
+
"ar": 14,
|
| 104 |
+
"az": 35,
|
| 105 |
+
"bg": 19,
|
| 106 |
+
"bn": 81,
|
| 107 |
+
"bs": 26,
|
| 108 |
+
"ca": 74,
|
| 109 |
+
"ceb": 80,
|
| 110 |
+
"co": 47,
|
| 111 |
+
"cs": 10,
|
| 112 |
+
"da": 11,
|
| 113 |
+
"de": 3,
|
| 114 |
+
"el": 5,
|
| 115 |
+
"en": 4,
|
| 116 |
+
"es": 9,
|
| 117 |
+
"et": 68,
|
| 118 |
+
"fi": 2,
|
| 119 |
+
"fil": 62,
|
| 120 |
+
"fr": 1,
|
| 121 |
+
"fy": 65,
|
| 122 |
+
"ga": 57,
|
| 123 |
+
"gl": 83,
|
| 124 |
+
"gn": 38,
|
| 125 |
+
"gu": 42,
|
| 126 |
+
"haw": 31,
|
| 127 |
+
"hi": 44,
|
| 128 |
+
"hmn": 79,
|
| 129 |
+
"hr": 6,
|
| 130 |
+
"hu": 13,
|
| 131 |
+
"hy": 70,
|
| 132 |
+
"id": 17,
|
| 133 |
+
"is": 40,
|
| 134 |
+
"it": 12,
|
| 135 |
+
"ja": 39,
|
| 136 |
+
"ka": 71,
|
| 137 |
+
"kk": 49,
|
| 138 |
+
"kn": 76,
|
| 139 |
+
"ko": 22,
|
| 140 |
+
"ku": 61,
|
| 141 |
+
"ky": 53,
|
| 142 |
+
"la": 33,
|
| 143 |
+
"lb": 29,
|
| 144 |
+
"ln": 67,
|
| 145 |
+
"lt": 75,
|
| 146 |
+
"lv": 69,
|
| 147 |
+
"mg": 78,
|
| 148 |
+
"mi": 56,
|
| 149 |
+
"mk": 54,
|
| 150 |
+
"ml": 58,
|
| 151 |
+
"mn": 72,
|
| 152 |
+
"mr": 30,
|
| 153 |
+
"ms": 18,
|
| 154 |
+
"my": 82,
|
| 155 |
+
"nl": 0,
|
| 156 |
+
"no": 60,
|
| 157 |
+
"pam": 20,
|
| 158 |
+
"pl": 16,
|
| 159 |
+
"ps": 77,
|
| 160 |
+
"pt": 8,
|
| 161 |
+
"ro": 7,
|
| 162 |
+
"ru": 50,
|
| 163 |
+
"rw": 45,
|
| 164 |
+
"sd": 63,
|
| 165 |
+
"sk": 36,
|
| 166 |
+
"sl": 48,
|
| 167 |
+
"sq": 46,
|
| 168 |
+
"sr": 55,
|
| 169 |
+
"st": 59,
|
| 170 |
+
"sv": 15,
|
| 171 |
+
"te": 32,
|
| 172 |
+
"tg": 73,
|
| 173 |
+
"th": 21,
|
| 174 |
+
"tk": 66,
|
| 175 |
+
"tr": 37,
|
| 176 |
+
"tt": 52,
|
| 177 |
+
"ug": 34,
|
| 178 |
+
"uk": 51,
|
| 179 |
+
"und": 28,
|
| 180 |
+
"ur": 27,
|
| 181 |
+
"uz": 43,
|
| 182 |
+
"vi": 25,
|
| 183 |
+
"yo": 64,
|
| 184 |
+
"zh": 23,
|
| 185 |
+
"zh-Hant": 24
|
| 186 |
+
},
|
| 187 |
+
"layer_norm_eps": 1e-05,
|
| 188 |
+
"max_position_embeddings": 514,
|
| 189 |
+
"model_type": "xlm-roberta",
|
| 190 |
+
"num_attention_heads": 12,
|
| 191 |
+
"num_hidden_layers": 12,
|
| 192 |
+
"output_past": true,
|
| 193 |
+
"pad_token_id": 1,
|
| 194 |
+
"position_embedding_type": "absolute",
|
| 195 |
+
"problem_type": "single_label_classification",
|
| 196 |
+
"torch_dtype": "float32",
|
| 197 |
+
"transformers_version": "4.28.0",
|
| 198 |
+
"type_vocab_size": 1,
|
| 199 |
+
"use_cache": true,
|
| 200 |
+
"vocab_size": 250002
|
| 201 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:65e0b66b1664d4b9d4bd7279d4319e9209620a1bbd22c4eddd551dfd0624aca6
|
| 3 |
+
size 1112506357
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d3aa98a3b3f5e56f2d8d704ca75f50ff43b963b6c72afce6913b9ca8d157f00
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:abd10b24342b4b92b8a20d32efae471f6f6f22de4dac48c80ff55b1c2ea32c0d
|
| 3 |
+
size 627
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46afe88da5fd71bdbab5cfab5e84c1adce59c246ea5f9341bbecef061891d0a7
|
| 3 |
+
size 17082913
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"mask_token": {
|
| 7 |
+
"__type": "AddedToken",
|
| 8 |
+
"content": "<mask>",
|
| 9 |
+
"lstrip": true,
|
| 10 |
+
"normalized": true,
|
| 11 |
+
"rstrip": false,
|
| 12 |
+
"single_word": false
|
| 13 |
+
},
|
| 14 |
+
"model_max_length": 512,
|
| 15 |
+
"pad_token": "<pad>",
|
| 16 |
+
"sep_token": "</s>",
|
| 17 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 18 |
+
"unk_token": "<unk>"
|
| 19 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1536 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.6260504201680672,
|
| 5 |
+
"global_step": 2500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 13 |
+
"loss": 4.6151,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 4.000000000000001e-06,
|
| 19 |
+
"loss": 4.6009,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 6e-06,
|
| 25 |
+
"loss": 4.5892,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 8.000000000000001e-06,
|
| 31 |
+
"loss": 4.4933,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"learning_rate": 1e-05,
|
| 37 |
+
"loss": 4.4838,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"learning_rate": 1.2e-05,
|
| 43 |
+
"loss": 4.0367,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07,
|
| 48 |
+
"learning_rate": 1.4e-05,
|
| 49 |
+
"loss": 3.6247,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.08,
|
| 54 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 55 |
+
"loss": 3.329,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.09,
|
| 60 |
+
"learning_rate": 1.8e-05,
|
| 61 |
+
"loss": 3.1007,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.11,
|
| 66 |
+
"learning_rate": 2e-05,
|
| 67 |
+
"loss": 2.8089,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.12,
|
| 72 |
+
"learning_rate": 1.992743105950653e-05,
|
| 73 |
+
"loss": 2.5354,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.13,
|
| 78 |
+
"learning_rate": 1.9854862119013064e-05,
|
| 79 |
+
"loss": 2.2403,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.14,
|
| 84 |
+
"learning_rate": 1.9782293178519596e-05,
|
| 85 |
+
"loss": 2.7885,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15,
|
| 90 |
+
"learning_rate": 1.9709724238026125e-05,
|
| 91 |
+
"loss": 2.4836,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.16,
|
| 96 |
+
"learning_rate": 1.9637155297532658e-05,
|
| 97 |
+
"loss": 2.3569,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.17,
|
| 102 |
+
"learning_rate": 1.9564586357039187e-05,
|
| 103 |
+
"loss": 2.4299,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.18,
|
| 108 |
+
"learning_rate": 1.949201741654572e-05,
|
| 109 |
+
"loss": 2.0131,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.19,
|
| 114 |
+
"learning_rate": 1.9419448476052253e-05,
|
| 115 |
+
"loss": 1.8533,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.2,
|
| 120 |
+
"learning_rate": 1.9346879535558782e-05,
|
| 121 |
+
"loss": 2.0193,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.21,
|
| 126 |
+
"learning_rate": 1.9274310595065315e-05,
|
| 127 |
+
"loss": 2.0059,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.22,
|
| 132 |
+
"learning_rate": 1.9201741654571844e-05,
|
| 133 |
+
"loss": 1.6821,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.23,
|
| 138 |
+
"learning_rate": 1.9129172714078376e-05,
|
| 139 |
+
"loss": 1.8958,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.24,
|
| 144 |
+
"learning_rate": 1.905660377358491e-05,
|
| 145 |
+
"loss": 1.9811,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.25,
|
| 150 |
+
"learning_rate": 1.898403483309144e-05,
|
| 151 |
+
"loss": 2.0177,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.26,
|
| 156 |
+
"learning_rate": 1.8911465892597968e-05,
|
| 157 |
+
"loss": 1.9885,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.27,
|
| 162 |
+
"learning_rate": 1.88388969521045e-05,
|
| 163 |
+
"loss": 1.8754,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.28,
|
| 168 |
+
"learning_rate": 1.8766328011611033e-05,
|
| 169 |
+
"loss": 1.5953,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.29,
|
| 174 |
+
"learning_rate": 1.8693759071117562e-05,
|
| 175 |
+
"loss": 1.9287,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.3,
|
| 180 |
+
"learning_rate": 1.8621190130624095e-05,
|
| 181 |
+
"loss": 1.7173,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.32,
|
| 186 |
+
"learning_rate": 1.8548621190130624e-05,
|
| 187 |
+
"loss": 1.8167,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.33,
|
| 192 |
+
"learning_rate": 1.8476052249637157e-05,
|
| 193 |
+
"loss": 1.3269,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.34,
|
| 198 |
+
"learning_rate": 1.840348330914369e-05,
|
| 199 |
+
"loss": 1.5738,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.35,
|
| 204 |
+
"learning_rate": 1.833091436865022e-05,
|
| 205 |
+
"loss": 1.5026,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.36,
|
| 210 |
+
"learning_rate": 1.825834542815675e-05,
|
| 211 |
+
"loss": 1.6312,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.37,
|
| 216 |
+
"learning_rate": 1.8185776487663284e-05,
|
| 217 |
+
"loss": 1.8379,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.38,
|
| 222 |
+
"learning_rate": 1.8113207547169813e-05,
|
| 223 |
+
"loss": 1.242,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.39,
|
| 228 |
+
"learning_rate": 1.8040638606676342e-05,
|
| 229 |
+
"loss": 1.5145,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.4,
|
| 234 |
+
"learning_rate": 1.7968069666182875e-05,
|
| 235 |
+
"loss": 1.326,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.41,
|
| 240 |
+
"learning_rate": 1.7895500725689404e-05,
|
| 241 |
+
"loss": 1.2579,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.42,
|
| 246 |
+
"learning_rate": 1.7822931785195937e-05,
|
| 247 |
+
"loss": 1.4427,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.43,
|
| 252 |
+
"learning_rate": 1.775036284470247e-05,
|
| 253 |
+
"loss": 1.2246,
|
| 254 |
+
"step": 410
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.44,
|
| 258 |
+
"learning_rate": 1.7677793904209e-05,
|
| 259 |
+
"loss": 1.3649,
|
| 260 |
+
"step": 420
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.45,
|
| 264 |
+
"learning_rate": 1.760522496371553e-05,
|
| 265 |
+
"loss": 1.5865,
|
| 266 |
+
"step": 430
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.46,
|
| 270 |
+
"learning_rate": 1.7532656023222064e-05,
|
| 271 |
+
"loss": 1.0269,
|
| 272 |
+
"step": 440
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.47,
|
| 276 |
+
"learning_rate": 1.7460087082728593e-05,
|
| 277 |
+
"loss": 1.2287,
|
| 278 |
+
"step": 450
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.48,
|
| 282 |
+
"learning_rate": 1.7387518142235126e-05,
|
| 283 |
+
"loss": 1.231,
|
| 284 |
+
"step": 460
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.49,
|
| 288 |
+
"learning_rate": 1.7314949201741655e-05,
|
| 289 |
+
"loss": 1.4882,
|
| 290 |
+
"step": 470
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5,
|
| 294 |
+
"learning_rate": 1.7242380261248185e-05,
|
| 295 |
+
"loss": 1.3311,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.51,
|
| 300 |
+
"learning_rate": 1.716981132075472e-05,
|
| 301 |
+
"loss": 0.8074,
|
| 302 |
+
"step": 490
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.53,
|
| 306 |
+
"learning_rate": 1.709724238026125e-05,
|
| 307 |
+
"loss": 1.3885,
|
| 308 |
+
"step": 500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.54,
|
| 312 |
+
"learning_rate": 1.702467343976778e-05,
|
| 313 |
+
"loss": 0.7269,
|
| 314 |
+
"step": 510
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.55,
|
| 318 |
+
"learning_rate": 1.6952104499274312e-05,
|
| 319 |
+
"loss": 1.0517,
|
| 320 |
+
"step": 520
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.56,
|
| 324 |
+
"learning_rate": 1.6879535558780844e-05,
|
| 325 |
+
"loss": 1.0979,
|
| 326 |
+
"step": 530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.57,
|
| 330 |
+
"learning_rate": 1.6806966618287374e-05,
|
| 331 |
+
"loss": 1.2165,
|
| 332 |
+
"step": 540
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.58,
|
| 336 |
+
"learning_rate": 1.6734397677793906e-05,
|
| 337 |
+
"loss": 0.9271,
|
| 338 |
+
"step": 550
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.59,
|
| 342 |
+
"learning_rate": 1.6661828737300436e-05,
|
| 343 |
+
"loss": 1.1317,
|
| 344 |
+
"step": 560
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.6,
|
| 348 |
+
"learning_rate": 1.6589259796806968e-05,
|
| 349 |
+
"loss": 1.1018,
|
| 350 |
+
"step": 570
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.61,
|
| 354 |
+
"learning_rate": 1.65166908563135e-05,
|
| 355 |
+
"loss": 1.054,
|
| 356 |
+
"step": 580
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.62,
|
| 360 |
+
"learning_rate": 1.644412191582003e-05,
|
| 361 |
+
"loss": 1.0754,
|
| 362 |
+
"step": 590
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.63,
|
| 366 |
+
"learning_rate": 1.6371552975326563e-05,
|
| 367 |
+
"loss": 0.8159,
|
| 368 |
+
"step": 600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.64,
|
| 372 |
+
"learning_rate": 1.6298984034833092e-05,
|
| 373 |
+
"loss": 0.929,
|
| 374 |
+
"step": 610
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.65,
|
| 378 |
+
"learning_rate": 1.6226415094339625e-05,
|
| 379 |
+
"loss": 1.1673,
|
| 380 |
+
"step": 620
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.66,
|
| 384 |
+
"learning_rate": 1.6153846153846154e-05,
|
| 385 |
+
"loss": 0.8993,
|
| 386 |
+
"step": 630
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.67,
|
| 390 |
+
"learning_rate": 1.6081277213352687e-05,
|
| 391 |
+
"loss": 0.8426,
|
| 392 |
+
"step": 640
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.68,
|
| 396 |
+
"learning_rate": 1.6008708272859216e-05,
|
| 397 |
+
"loss": 0.9685,
|
| 398 |
+
"step": 650
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.69,
|
| 402 |
+
"learning_rate": 1.593613933236575e-05,
|
| 403 |
+
"loss": 0.7475,
|
| 404 |
+
"step": 660
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.7,
|
| 408 |
+
"learning_rate": 1.586357039187228e-05,
|
| 409 |
+
"loss": 0.4813,
|
| 410 |
+
"step": 670
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.71,
|
| 414 |
+
"learning_rate": 1.579100145137881e-05,
|
| 415 |
+
"loss": 0.8771,
|
| 416 |
+
"step": 680
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.72,
|
| 420 |
+
"learning_rate": 1.5718432510885343e-05,
|
| 421 |
+
"loss": 0.8842,
|
| 422 |
+
"step": 690
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.74,
|
| 426 |
+
"learning_rate": 1.5645863570391872e-05,
|
| 427 |
+
"loss": 1.0122,
|
| 428 |
+
"step": 700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.75,
|
| 432 |
+
"learning_rate": 1.5573294629898405e-05,
|
| 433 |
+
"loss": 0.6247,
|
| 434 |
+
"step": 710
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.76,
|
| 438 |
+
"learning_rate": 1.5500725689404938e-05,
|
| 439 |
+
"loss": 0.9389,
|
| 440 |
+
"step": 720
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.77,
|
| 444 |
+
"learning_rate": 1.5428156748911467e-05,
|
| 445 |
+
"loss": 0.3921,
|
| 446 |
+
"step": 730
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.78,
|
| 450 |
+
"learning_rate": 1.5355587808417996e-05,
|
| 451 |
+
"loss": 0.695,
|
| 452 |
+
"step": 740
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.79,
|
| 456 |
+
"learning_rate": 1.5283018867924532e-05,
|
| 457 |
+
"loss": 0.4541,
|
| 458 |
+
"step": 750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.8,
|
| 462 |
+
"learning_rate": 1.5210449927431061e-05,
|
| 463 |
+
"loss": 0.6469,
|
| 464 |
+
"step": 760
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.81,
|
| 468 |
+
"learning_rate": 1.5137880986937592e-05,
|
| 469 |
+
"loss": 0.6391,
|
| 470 |
+
"step": 770
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.82,
|
| 474 |
+
"learning_rate": 1.5065312046444123e-05,
|
| 475 |
+
"loss": 0.6501,
|
| 476 |
+
"step": 780
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.83,
|
| 480 |
+
"learning_rate": 1.4992743105950653e-05,
|
| 481 |
+
"loss": 0.9943,
|
| 482 |
+
"step": 790
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.84,
|
| 486 |
+
"learning_rate": 1.4920174165457187e-05,
|
| 487 |
+
"loss": 0.5541,
|
| 488 |
+
"step": 800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.85,
|
| 492 |
+
"learning_rate": 1.4847605224963716e-05,
|
| 493 |
+
"loss": 0.7708,
|
| 494 |
+
"step": 810
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.86,
|
| 498 |
+
"learning_rate": 1.4775036284470247e-05,
|
| 499 |
+
"loss": 0.9068,
|
| 500 |
+
"step": 820
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.87,
|
| 504 |
+
"learning_rate": 1.4702467343976778e-05,
|
| 505 |
+
"loss": 0.8202,
|
| 506 |
+
"step": 830
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.88,
|
| 510 |
+
"learning_rate": 1.462989840348331e-05,
|
| 511 |
+
"loss": 0.6628,
|
| 512 |
+
"step": 840
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.89,
|
| 516 |
+
"learning_rate": 1.4557329462989842e-05,
|
| 517 |
+
"loss": 0.768,
|
| 518 |
+
"step": 850
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.9,
|
| 522 |
+
"learning_rate": 1.4484760522496373e-05,
|
| 523 |
+
"loss": 0.829,
|
| 524 |
+
"step": 860
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.91,
|
| 528 |
+
"learning_rate": 1.4412191582002904e-05,
|
| 529 |
+
"loss": 0.9561,
|
| 530 |
+
"step": 870
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.92,
|
| 534 |
+
"learning_rate": 1.4339622641509435e-05,
|
| 535 |
+
"loss": 0.6783,
|
| 536 |
+
"step": 880
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.93,
|
| 540 |
+
"learning_rate": 1.4267053701015967e-05,
|
| 541 |
+
"loss": 0.7017,
|
| 542 |
+
"step": 890
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.95,
|
| 546 |
+
"learning_rate": 1.4194484760522498e-05,
|
| 547 |
+
"loss": 0.6462,
|
| 548 |
+
"step": 900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.96,
|
| 552 |
+
"learning_rate": 1.4121915820029029e-05,
|
| 553 |
+
"loss": 0.648,
|
| 554 |
+
"step": 910
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.97,
|
| 558 |
+
"learning_rate": 1.4049346879535558e-05,
|
| 559 |
+
"loss": 0.4731,
|
| 560 |
+
"step": 920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.98,
|
| 564 |
+
"learning_rate": 1.3976777939042093e-05,
|
| 565 |
+
"loss": 0.6257,
|
| 566 |
+
"step": 930
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.99,
|
| 570 |
+
"learning_rate": 1.3904208998548622e-05,
|
| 571 |
+
"loss": 0.4245,
|
| 572 |
+
"step": 940
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.0,
|
| 576 |
+
"learning_rate": 1.3831640058055153e-05,
|
| 577 |
+
"loss": 0.687,
|
| 578 |
+
"step": 950
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.0,
|
| 582 |
+
"eval_Accuracy": 0.9114854517611026,
|
| 583 |
+
"eval_F1": 0.8950475194731157,
|
| 584 |
+
"eval_loss": 0.5604398846626282,
|
| 585 |
+
"eval_runtime": 98.5377,
|
| 586 |
+
"eval_samples_per_second": 33.135,
|
| 587 |
+
"eval_steps_per_second": 2.08,
|
| 588 |
+
"step": 952
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.01,
|
| 592 |
+
"learning_rate": 1.3759071117561684e-05,
|
| 593 |
+
"loss": 0.3493,
|
| 594 |
+
"step": 960
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.02,
|
| 598 |
+
"learning_rate": 1.3686502177068215e-05,
|
| 599 |
+
"loss": 0.9263,
|
| 600 |
+
"step": 970
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 1.03,
|
| 604 |
+
"learning_rate": 1.3613933236574748e-05,
|
| 605 |
+
"loss": 0.4981,
|
| 606 |
+
"step": 980
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 1.04,
|
| 610 |
+
"learning_rate": 1.3541364296081278e-05,
|
| 611 |
+
"loss": 0.4574,
|
| 612 |
+
"step": 990
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.05,
|
| 616 |
+
"learning_rate": 1.346879535558781e-05,
|
| 617 |
+
"loss": 0.6115,
|
| 618 |
+
"step": 1000
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.06,
|
| 622 |
+
"learning_rate": 1.339622641509434e-05,
|
| 623 |
+
"loss": 0.5973,
|
| 624 |
+
"step": 1010
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.07,
|
| 628 |
+
"learning_rate": 1.3323657474600873e-05,
|
| 629 |
+
"loss": 0.4881,
|
| 630 |
+
"step": 1020
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 1.08,
|
| 634 |
+
"learning_rate": 1.3251088534107404e-05,
|
| 635 |
+
"loss": 0.4824,
|
| 636 |
+
"step": 1030
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.09,
|
| 640 |
+
"learning_rate": 1.3178519593613935e-05,
|
| 641 |
+
"loss": 0.7372,
|
| 642 |
+
"step": 1040
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.1,
|
| 646 |
+
"learning_rate": 1.3105950653120464e-05,
|
| 647 |
+
"loss": 0.7515,
|
| 648 |
+
"step": 1050
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.11,
|
| 652 |
+
"learning_rate": 1.3033381712626995e-05,
|
| 653 |
+
"loss": 0.446,
|
| 654 |
+
"step": 1060
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.12,
|
| 658 |
+
"learning_rate": 1.2960812772133528e-05,
|
| 659 |
+
"loss": 0.6326,
|
| 660 |
+
"step": 1070
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.13,
|
| 664 |
+
"learning_rate": 1.2888243831640059e-05,
|
| 665 |
+
"loss": 0.7661,
|
| 666 |
+
"step": 1080
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.14,
|
| 670 |
+
"learning_rate": 1.281567489114659e-05,
|
| 671 |
+
"loss": 0.4974,
|
| 672 |
+
"step": 1090
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.16,
|
| 676 |
+
"learning_rate": 1.274310595065312e-05,
|
| 677 |
+
"loss": 0.7276,
|
| 678 |
+
"step": 1100
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.17,
|
| 682 |
+
"learning_rate": 1.2670537010159653e-05,
|
| 683 |
+
"loss": 0.3897,
|
| 684 |
+
"step": 1110
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.18,
|
| 688 |
+
"learning_rate": 1.2597968069666184e-05,
|
| 689 |
+
"loss": 0.7384,
|
| 690 |
+
"step": 1120
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.19,
|
| 694 |
+
"learning_rate": 1.2525399129172715e-05,
|
| 695 |
+
"loss": 0.7007,
|
| 696 |
+
"step": 1130
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.2,
|
| 700 |
+
"learning_rate": 1.2452830188679246e-05,
|
| 701 |
+
"loss": 0.6739,
|
| 702 |
+
"step": 1140
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.21,
|
| 706 |
+
"learning_rate": 1.2380261248185777e-05,
|
| 707 |
+
"loss": 0.5027,
|
| 708 |
+
"step": 1150
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.22,
|
| 712 |
+
"learning_rate": 1.230769230769231e-05,
|
| 713 |
+
"loss": 0.5931,
|
| 714 |
+
"step": 1160
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.23,
|
| 718 |
+
"learning_rate": 1.223512336719884e-05,
|
| 719 |
+
"loss": 0.3967,
|
| 720 |
+
"step": 1170
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.24,
|
| 724 |
+
"learning_rate": 1.216255442670537e-05,
|
| 725 |
+
"loss": 0.5215,
|
| 726 |
+
"step": 1180
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 1.25,
|
| 730 |
+
"learning_rate": 1.2089985486211901e-05,
|
| 731 |
+
"loss": 0.5034,
|
| 732 |
+
"step": 1190
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.26,
|
| 736 |
+
"learning_rate": 1.2017416545718435e-05,
|
| 737 |
+
"loss": 0.4521,
|
| 738 |
+
"step": 1200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.27,
|
| 742 |
+
"learning_rate": 1.1944847605224965e-05,
|
| 743 |
+
"loss": 0.255,
|
| 744 |
+
"step": 1210
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.28,
|
| 748 |
+
"learning_rate": 1.1872278664731495e-05,
|
| 749 |
+
"loss": 0.2349,
|
| 750 |
+
"step": 1220
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 1.29,
|
| 754 |
+
"learning_rate": 1.1799709724238026e-05,
|
| 755 |
+
"loss": 0.6059,
|
| 756 |
+
"step": 1230
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 1.3,
|
| 760 |
+
"learning_rate": 1.1727140783744557e-05,
|
| 761 |
+
"loss": 0.3443,
|
| 762 |
+
"step": 1240
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.31,
|
| 766 |
+
"learning_rate": 1.165457184325109e-05,
|
| 767 |
+
"loss": 0.3367,
|
| 768 |
+
"step": 1250
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.32,
|
| 772 |
+
"learning_rate": 1.1582002902757621e-05,
|
| 773 |
+
"loss": 0.4613,
|
| 774 |
+
"step": 1260
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.33,
|
| 778 |
+
"learning_rate": 1.1509433962264152e-05,
|
| 779 |
+
"loss": 0.3399,
|
| 780 |
+
"step": 1270
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.34,
|
| 784 |
+
"learning_rate": 1.1436865021770683e-05,
|
| 785 |
+
"loss": 0.5677,
|
| 786 |
+
"step": 1280
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.36,
|
| 790 |
+
"learning_rate": 1.1364296081277216e-05,
|
| 791 |
+
"loss": 0.8075,
|
| 792 |
+
"step": 1290
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 1.37,
|
| 796 |
+
"learning_rate": 1.1291727140783746e-05,
|
| 797 |
+
"loss": 0.4384,
|
| 798 |
+
"step": 1300
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 1.38,
|
| 802 |
+
"learning_rate": 1.1219158200290277e-05,
|
| 803 |
+
"loss": 0.3861,
|
| 804 |
+
"step": 1310
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.39,
|
| 808 |
+
"learning_rate": 1.1146589259796807e-05,
|
| 809 |
+
"loss": 0.1245,
|
| 810 |
+
"step": 1320
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 1.4,
|
| 814 |
+
"learning_rate": 1.1074020319303338e-05,
|
| 815 |
+
"loss": 0.6668,
|
| 816 |
+
"step": 1330
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.41,
|
| 820 |
+
"learning_rate": 1.100145137880987e-05,
|
| 821 |
+
"loss": 0.7036,
|
| 822 |
+
"step": 1340
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.42,
|
| 826 |
+
"learning_rate": 1.0928882438316401e-05,
|
| 827 |
+
"loss": 0.205,
|
| 828 |
+
"step": 1350
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.43,
|
| 832 |
+
"learning_rate": 1.0856313497822932e-05,
|
| 833 |
+
"loss": 0.588,
|
| 834 |
+
"step": 1360
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.44,
|
| 838 |
+
"learning_rate": 1.0783744557329463e-05,
|
| 839 |
+
"loss": 0.7293,
|
| 840 |
+
"step": 1370
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.45,
|
| 844 |
+
"learning_rate": 1.0711175616835996e-05,
|
| 845 |
+
"loss": 0.602,
|
| 846 |
+
"step": 1380
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 1.46,
|
| 850 |
+
"learning_rate": 1.0638606676342527e-05,
|
| 851 |
+
"loss": 0.3027,
|
| 852 |
+
"step": 1390
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.47,
|
| 856 |
+
"learning_rate": 1.0566037735849058e-05,
|
| 857 |
+
"loss": 0.4086,
|
| 858 |
+
"step": 1400
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.48,
|
| 862 |
+
"learning_rate": 1.0493468795355589e-05,
|
| 863 |
+
"loss": 0.4706,
|
| 864 |
+
"step": 1410
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.49,
|
| 868 |
+
"learning_rate": 1.0420899854862121e-05,
|
| 869 |
+
"loss": 0.2685,
|
| 870 |
+
"step": 1420
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.5,
|
| 874 |
+
"learning_rate": 1.0348330914368652e-05,
|
| 875 |
+
"loss": 0.4058,
|
| 876 |
+
"step": 1430
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.51,
|
| 880 |
+
"learning_rate": 1.0275761973875183e-05,
|
| 881 |
+
"loss": 0.4807,
|
| 882 |
+
"step": 1440
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 1.52,
|
| 886 |
+
"learning_rate": 1.0203193033381712e-05,
|
| 887 |
+
"loss": 0.3557,
|
| 888 |
+
"step": 1450
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 1.53,
|
| 892 |
+
"learning_rate": 1.0130624092888243e-05,
|
| 893 |
+
"loss": 0.3329,
|
| 894 |
+
"step": 1460
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.54,
|
| 898 |
+
"learning_rate": 1.0058055152394776e-05,
|
| 899 |
+
"loss": 0.2153,
|
| 900 |
+
"step": 1470
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.55,
|
| 904 |
+
"learning_rate": 9.985486211901307e-06,
|
| 905 |
+
"loss": 0.7515,
|
| 906 |
+
"step": 1480
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.57,
|
| 910 |
+
"learning_rate": 9.912917271407838e-06,
|
| 911 |
+
"loss": 0.3258,
|
| 912 |
+
"step": 1490
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.58,
|
| 916 |
+
"learning_rate": 9.84034833091437e-06,
|
| 917 |
+
"loss": 0.1792,
|
| 918 |
+
"step": 1500
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 1.59,
|
| 922 |
+
"learning_rate": 9.7677793904209e-06,
|
| 923 |
+
"loss": 0.5118,
|
| 924 |
+
"step": 1510
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.6,
|
| 928 |
+
"learning_rate": 9.69521044992743e-06,
|
| 929 |
+
"loss": 0.3737,
|
| 930 |
+
"step": 1520
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 1.61,
|
| 934 |
+
"learning_rate": 9.622641509433963e-06,
|
| 935 |
+
"loss": 0.3137,
|
| 936 |
+
"step": 1530
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.62,
|
| 940 |
+
"learning_rate": 9.550072568940494e-06,
|
| 941 |
+
"loss": 0.473,
|
| 942 |
+
"step": 1540
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 1.63,
|
| 946 |
+
"learning_rate": 9.477503628447025e-06,
|
| 947 |
+
"loss": 0.5268,
|
| 948 |
+
"step": 1550
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.64,
|
| 952 |
+
"learning_rate": 9.404934687953556e-06,
|
| 953 |
+
"loss": 0.4378,
|
| 954 |
+
"step": 1560
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.65,
|
| 958 |
+
"learning_rate": 9.332365747460089e-06,
|
| 959 |
+
"loss": 0.4031,
|
| 960 |
+
"step": 1570
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.66,
|
| 964 |
+
"learning_rate": 9.259796806966618e-06,
|
| 965 |
+
"loss": 0.7125,
|
| 966 |
+
"step": 1580
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 1.67,
|
| 970 |
+
"learning_rate": 9.187227866473151e-06,
|
| 971 |
+
"loss": 0.3404,
|
| 972 |
+
"step": 1590
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 1.68,
|
| 976 |
+
"learning_rate": 9.114658925979682e-06,
|
| 977 |
+
"loss": 0.4025,
|
| 978 |
+
"step": 1600
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 1.69,
|
| 982 |
+
"learning_rate": 9.042089985486213e-06,
|
| 983 |
+
"loss": 0.6217,
|
| 984 |
+
"step": 1610
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.7,
|
| 988 |
+
"learning_rate": 8.969521044992744e-06,
|
| 989 |
+
"loss": 0.4097,
|
| 990 |
+
"step": 1620
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.71,
|
| 994 |
+
"learning_rate": 8.896952104499275e-06,
|
| 995 |
+
"loss": 0.4337,
|
| 996 |
+
"step": 1630
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.72,
|
| 1000 |
+
"learning_rate": 8.824383164005806e-06,
|
| 1001 |
+
"loss": 0.2964,
|
| 1002 |
+
"step": 1640
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 1.73,
|
| 1006 |
+
"learning_rate": 8.751814223512337e-06,
|
| 1007 |
+
"loss": 0.5013,
|
| 1008 |
+
"step": 1650
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 1.74,
|
| 1012 |
+
"learning_rate": 8.67924528301887e-06,
|
| 1013 |
+
"loss": 0.6303,
|
| 1014 |
+
"step": 1660
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 1.75,
|
| 1018 |
+
"learning_rate": 8.6066763425254e-06,
|
| 1019 |
+
"loss": 0.3938,
|
| 1020 |
+
"step": 1670
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 1.76,
|
| 1024 |
+
"learning_rate": 8.534107402031931e-06,
|
| 1025 |
+
"loss": 0.1885,
|
| 1026 |
+
"step": 1680
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.78,
|
| 1030 |
+
"learning_rate": 8.461538461538462e-06,
|
| 1031 |
+
"loss": 0.3159,
|
| 1032 |
+
"step": 1690
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.79,
|
| 1036 |
+
"learning_rate": 8.388969521044995e-06,
|
| 1037 |
+
"loss": 0.5674,
|
| 1038 |
+
"step": 1700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.8,
|
| 1042 |
+
"learning_rate": 8.316400580551524e-06,
|
| 1043 |
+
"loss": 0.1875,
|
| 1044 |
+
"step": 1710
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.81,
|
| 1048 |
+
"learning_rate": 8.243831640058055e-06,
|
| 1049 |
+
"loss": 0.4107,
|
| 1050 |
+
"step": 1720
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 1.82,
|
| 1054 |
+
"learning_rate": 8.171262699564588e-06,
|
| 1055 |
+
"loss": 0.693,
|
| 1056 |
+
"step": 1730
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 1.83,
|
| 1060 |
+
"learning_rate": 8.098693759071119e-06,
|
| 1061 |
+
"loss": 0.5109,
|
| 1062 |
+
"step": 1740
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.84,
|
| 1066 |
+
"learning_rate": 8.02612481857765e-06,
|
| 1067 |
+
"loss": 0.3853,
|
| 1068 |
+
"step": 1750
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.85,
|
| 1072 |
+
"learning_rate": 7.95355587808418e-06,
|
| 1073 |
+
"loss": 0.4673,
|
| 1074 |
+
"step": 1760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.86,
|
| 1078 |
+
"learning_rate": 7.880986937590711e-06,
|
| 1079 |
+
"loss": 0.2653,
|
| 1080 |
+
"step": 1770
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.87,
|
| 1084 |
+
"learning_rate": 7.808417997097242e-06,
|
| 1085 |
+
"loss": 0.3816,
|
| 1086 |
+
"step": 1780
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 1.88,
|
| 1090 |
+
"learning_rate": 7.735849056603775e-06,
|
| 1091 |
+
"loss": 0.2759,
|
| 1092 |
+
"step": 1790
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 1.89,
|
| 1096 |
+
"learning_rate": 7.663280116110306e-06,
|
| 1097 |
+
"loss": 0.2267,
|
| 1098 |
+
"step": 1800
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.9,
|
| 1102 |
+
"learning_rate": 7.590711175616836e-06,
|
| 1103 |
+
"loss": 0.3894,
|
| 1104 |
+
"step": 1810
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.91,
|
| 1108 |
+
"learning_rate": 7.518142235123368e-06,
|
| 1109 |
+
"loss": 0.1902,
|
| 1110 |
+
"step": 1820
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 1.92,
|
| 1114 |
+
"learning_rate": 7.445573294629899e-06,
|
| 1115 |
+
"loss": 0.3042,
|
| 1116 |
+
"step": 1830
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.93,
|
| 1120 |
+
"learning_rate": 7.373004354136431e-06,
|
| 1121 |
+
"loss": 0.2981,
|
| 1122 |
+
"step": 1840
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.94,
|
| 1126 |
+
"learning_rate": 7.300435413642961e-06,
|
| 1127 |
+
"loss": 0.2297,
|
| 1128 |
+
"step": 1850
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.95,
|
| 1132 |
+
"learning_rate": 7.227866473149493e-06,
|
| 1133 |
+
"loss": 0.2528,
|
| 1134 |
+
"step": 1860
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.96,
|
| 1138 |
+
"learning_rate": 7.1552975326560235e-06,
|
| 1139 |
+
"loss": 0.2361,
|
| 1140 |
+
"step": 1870
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.97,
|
| 1144 |
+
"learning_rate": 7.082728592162555e-06,
|
| 1145 |
+
"loss": 0.327,
|
| 1146 |
+
"step": 1880
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 1.99,
|
| 1150 |
+
"learning_rate": 7.010159651669086e-06,
|
| 1151 |
+
"loss": 0.2838,
|
| 1152 |
+
"step": 1890
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 2.0,
|
| 1156 |
+
"learning_rate": 6.937590711175617e-06,
|
| 1157 |
+
"loss": 0.3027,
|
| 1158 |
+
"step": 1900
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 2.0,
|
| 1162 |
+
"eval_Accuracy": 0.935375191424196,
|
| 1163 |
+
"eval_F1": 0.9231380335717158,
|
| 1164 |
+
"eval_loss": 0.42070162296295166,
|
| 1165 |
+
"eval_runtime": 98.699,
|
| 1166 |
+
"eval_samples_per_second": 33.08,
|
| 1167 |
+
"eval_steps_per_second": 2.077,
|
| 1168 |
+
"step": 1904
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 2.01,
|
| 1172 |
+
"learning_rate": 6.865021770682149e-06,
|
| 1173 |
+
"loss": 0.3852,
|
| 1174 |
+
"step": 1910
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 2.02,
|
| 1178 |
+
"learning_rate": 6.792452830188679e-06,
|
| 1179 |
+
"loss": 0.4178,
|
| 1180 |
+
"step": 1920
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 2.03,
|
| 1184 |
+
"learning_rate": 6.719883889695211e-06,
|
| 1185 |
+
"loss": 0.4089,
|
| 1186 |
+
"step": 1930
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 2.04,
|
| 1190 |
+
"learning_rate": 6.647314949201742e-06,
|
| 1191 |
+
"loss": 0.3711,
|
| 1192 |
+
"step": 1940
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.05,
|
| 1196 |
+
"learning_rate": 6.574746008708274e-06,
|
| 1197 |
+
"loss": 0.4505,
|
| 1198 |
+
"step": 1950
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 2.06,
|
| 1202 |
+
"learning_rate": 6.502177068214805e-06,
|
| 1203 |
+
"loss": 0.2162,
|
| 1204 |
+
"step": 1960
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 2.07,
|
| 1208 |
+
"learning_rate": 6.429608127721336e-06,
|
| 1209 |
+
"loss": 0.2023,
|
| 1210 |
+
"step": 1970
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 2.08,
|
| 1214 |
+
"learning_rate": 6.357039187227867e-06,
|
| 1215 |
+
"loss": 0.2912,
|
| 1216 |
+
"step": 1980
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 2.09,
|
| 1220 |
+
"learning_rate": 6.2844702467343975e-06,
|
| 1221 |
+
"loss": 0.4556,
|
| 1222 |
+
"step": 1990
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 2.1,
|
| 1226 |
+
"learning_rate": 6.211901306240929e-06,
|
| 1227 |
+
"loss": 0.3712,
|
| 1228 |
+
"step": 2000
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 2.11,
|
| 1232 |
+
"learning_rate": 6.13933236574746e-06,
|
| 1233 |
+
"loss": 0.407,
|
| 1234 |
+
"step": 2010
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.12,
|
| 1238 |
+
"learning_rate": 6.066763425253992e-06,
|
| 1239 |
+
"loss": 0.3238,
|
| 1240 |
+
"step": 2020
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 2.13,
|
| 1244 |
+
"learning_rate": 5.994194484760523e-06,
|
| 1245 |
+
"loss": 0.118,
|
| 1246 |
+
"step": 2030
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 2.14,
|
| 1250 |
+
"learning_rate": 5.921625544267055e-06,
|
| 1251 |
+
"loss": 0.1698,
|
| 1252 |
+
"step": 2040
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 2.15,
|
| 1256 |
+
"learning_rate": 5.849056603773585e-06,
|
| 1257 |
+
"loss": 0.3557,
|
| 1258 |
+
"step": 2050
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 2.16,
|
| 1262 |
+
"learning_rate": 5.776487663280117e-06,
|
| 1263 |
+
"loss": 0.2857,
|
| 1264 |
+
"step": 2060
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 2.17,
|
| 1268 |
+
"learning_rate": 5.703918722786648e-06,
|
| 1269 |
+
"loss": 0.6379,
|
| 1270 |
+
"step": 2070
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.18,
|
| 1274 |
+
"learning_rate": 5.6313497822931794e-06,
|
| 1275 |
+
"loss": 0.7017,
|
| 1276 |
+
"step": 2080
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.2,
|
| 1280 |
+
"learning_rate": 5.55878084179971e-06,
|
| 1281 |
+
"loss": 0.2319,
|
| 1282 |
+
"step": 2090
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 2.21,
|
| 1286 |
+
"learning_rate": 5.486211901306241e-06,
|
| 1287 |
+
"loss": 0.2963,
|
| 1288 |
+
"step": 2100
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 2.22,
|
| 1292 |
+
"learning_rate": 5.413642960812773e-06,
|
| 1293 |
+
"loss": 0.284,
|
| 1294 |
+
"step": 2110
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 2.23,
|
| 1298 |
+
"learning_rate": 5.341074020319303e-06,
|
| 1299 |
+
"loss": 0.2335,
|
| 1300 |
+
"step": 2120
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 2.24,
|
| 1304 |
+
"learning_rate": 5.268505079825835e-06,
|
| 1305 |
+
"loss": 0.3575,
|
| 1306 |
+
"step": 2130
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.25,
|
| 1310 |
+
"learning_rate": 5.195936139332366e-06,
|
| 1311 |
+
"loss": 0.3061,
|
| 1312 |
+
"step": 2140
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 2.26,
|
| 1316 |
+
"learning_rate": 5.123367198838898e-06,
|
| 1317 |
+
"loss": 0.2509,
|
| 1318 |
+
"step": 2150
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 2.27,
|
| 1322 |
+
"learning_rate": 5.050798258345429e-06,
|
| 1323 |
+
"loss": 0.669,
|
| 1324 |
+
"step": 2160
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 2.28,
|
| 1328 |
+
"learning_rate": 4.97822931785196e-06,
|
| 1329 |
+
"loss": 0.3276,
|
| 1330 |
+
"step": 2170
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 2.29,
|
| 1334 |
+
"learning_rate": 4.905660377358491e-06,
|
| 1335 |
+
"loss": 0.2792,
|
| 1336 |
+
"step": 2180
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 2.3,
|
| 1340 |
+
"learning_rate": 4.8330914368650224e-06,
|
| 1341 |
+
"loss": 0.3574,
|
| 1342 |
+
"step": 2190
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.31,
|
| 1346 |
+
"learning_rate": 4.760522496371553e-06,
|
| 1347 |
+
"loss": 0.1847,
|
| 1348 |
+
"step": 2200
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 2.32,
|
| 1352 |
+
"learning_rate": 4.687953555878084e-06,
|
| 1353 |
+
"loss": 0.3664,
|
| 1354 |
+
"step": 2210
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 2.33,
|
| 1358 |
+
"learning_rate": 4.615384615384616e-06,
|
| 1359 |
+
"loss": 0.313,
|
| 1360 |
+
"step": 2220
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 2.34,
|
| 1364 |
+
"learning_rate": 4.542815674891147e-06,
|
| 1365 |
+
"loss": 0.4165,
|
| 1366 |
+
"step": 2230
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 2.35,
|
| 1370 |
+
"learning_rate": 4.470246734397678e-06,
|
| 1371 |
+
"loss": 0.2627,
|
| 1372 |
+
"step": 2240
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 2.36,
|
| 1376 |
+
"learning_rate": 4.397677793904209e-06,
|
| 1377 |
+
"loss": 0.4678,
|
| 1378 |
+
"step": 2250
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 2.37,
|
| 1382 |
+
"learning_rate": 4.325108853410741e-06,
|
| 1383 |
+
"loss": 0.4254,
|
| 1384 |
+
"step": 2260
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 2.38,
|
| 1388 |
+
"learning_rate": 4.252539912917272e-06,
|
| 1389 |
+
"loss": 0.2166,
|
| 1390 |
+
"step": 2270
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 2.39,
|
| 1394 |
+
"learning_rate": 4.179970972423803e-06,
|
| 1395 |
+
"loss": 0.2313,
|
| 1396 |
+
"step": 2280
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.41,
|
| 1400 |
+
"learning_rate": 4.1074020319303345e-06,
|
| 1401 |
+
"loss": 0.2503,
|
| 1402 |
+
"step": 2290
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 2.42,
|
| 1406 |
+
"learning_rate": 4.0348330914368655e-06,
|
| 1407 |
+
"loss": 0.2127,
|
| 1408 |
+
"step": 2300
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 2.43,
|
| 1412 |
+
"learning_rate": 3.962264150943396e-06,
|
| 1413 |
+
"loss": 0.1339,
|
| 1414 |
+
"step": 2310
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 2.44,
|
| 1418 |
+
"learning_rate": 3.889695210449927e-06,
|
| 1419 |
+
"loss": 0.1467,
|
| 1420 |
+
"step": 2320
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 2.45,
|
| 1424 |
+
"learning_rate": 3.817126269956458e-06,
|
| 1425 |
+
"loss": 0.3517,
|
| 1426 |
+
"step": 2330
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.46,
|
| 1430 |
+
"learning_rate": 3.74455732946299e-06,
|
| 1431 |
+
"loss": 0.1513,
|
| 1432 |
+
"step": 2340
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 2.47,
|
| 1436 |
+
"learning_rate": 3.671988388969521e-06,
|
| 1437 |
+
"loss": 0.1561,
|
| 1438 |
+
"step": 2350
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.48,
|
| 1442 |
+
"learning_rate": 3.5994194484760525e-06,
|
| 1443 |
+
"loss": 0.2388,
|
| 1444 |
+
"step": 2360
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 2.49,
|
| 1448 |
+
"learning_rate": 3.526850507982584e-06,
|
| 1449 |
+
"loss": 0.5018,
|
| 1450 |
+
"step": 2370
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 2.5,
|
| 1454 |
+
"learning_rate": 3.454281567489115e-06,
|
| 1455 |
+
"loss": 0.3328,
|
| 1456 |
+
"step": 2380
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 2.51,
|
| 1460 |
+
"learning_rate": 3.381712626995646e-06,
|
| 1461 |
+
"loss": 0.4347,
|
| 1462 |
+
"step": 2390
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 2.52,
|
| 1466 |
+
"learning_rate": 3.3091436865021775e-06,
|
| 1467 |
+
"loss": 0.3993,
|
| 1468 |
+
"step": 2400
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 2.53,
|
| 1472 |
+
"learning_rate": 3.236574746008709e-06,
|
| 1473 |
+
"loss": 0.4217,
|
| 1474 |
+
"step": 2410
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 2.54,
|
| 1478 |
+
"learning_rate": 3.1640058055152394e-06,
|
| 1479 |
+
"loss": 0.651,
|
| 1480 |
+
"step": 2420
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 2.55,
|
| 1484 |
+
"learning_rate": 3.091436865021771e-06,
|
| 1485 |
+
"loss": 0.3912,
|
| 1486 |
+
"step": 2430
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.56,
|
| 1490 |
+
"learning_rate": 3.018867924528302e-06,
|
| 1491 |
+
"loss": 0.2198,
|
| 1492 |
+
"step": 2440
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 2.57,
|
| 1496 |
+
"learning_rate": 2.946298984034833e-06,
|
| 1497 |
+
"loss": 0.1841,
|
| 1498 |
+
"step": 2450
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 2.58,
|
| 1502 |
+
"learning_rate": 2.8737300435413645e-06,
|
| 1503 |
+
"loss": 0.2908,
|
| 1504 |
+
"step": 2460
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 2.59,
|
| 1508 |
+
"learning_rate": 2.801161103047896e-06,
|
| 1509 |
+
"loss": 0.2179,
|
| 1510 |
+
"step": 2470
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 2.61,
|
| 1514 |
+
"learning_rate": 2.728592162554427e-06,
|
| 1515 |
+
"loss": 0.1909,
|
| 1516 |
+
"step": 2480
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.62,
|
| 1520 |
+
"learning_rate": 2.6560232220609582e-06,
|
| 1521 |
+
"loss": 0.4162,
|
| 1522 |
+
"step": 2490
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 2.63,
|
| 1526 |
+
"learning_rate": 2.5834542815674896e-06,
|
| 1527 |
+
"loss": 0.3347,
|
| 1528 |
+
"step": 2500
|
| 1529 |
+
}
|
| 1530 |
+
],
|
| 1531 |
+
"max_steps": 2856,
|
| 1532 |
+
"num_train_epochs": 3,
|
| 1533 |
+
"total_flos": 5266095390720000.0,
|
| 1534 |
+
"trial_name": null,
|
| 1535 |
+
"trial_params": null
|
| 1536 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:468fb6e7e73029b88c773c61fd6dd90ca727d23a24badd21d835aa82b572fe40
|
| 3 |
+
size 3515
|