Commit
•
a78c29a
1
Parent(s):
d4f5af0
Upload folder using huggingface_hub
Browse files- added_tokens.json +5 -0
- config.json +305 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +24 -0
- tokenizer.model +3 -0
- tokenizer_config.json +33 -0
- trainer_state.json +1416 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</img>": 32001,
|
| 3 |
+
"<ImageContent>": 32002,
|
| 4 |
+
"<img>": 32000
|
| 5 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,305 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_commit_hash": null,
|
| 3 |
+
"_name_or_path": "/mnt/petrelfs/share_data/wangweiyun/recognize_anything_annotation/asm_ckpt/asm_v2",
|
| 4 |
+
"aligned": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"UnifiedHuskyFlattenCatForCaption"
|
| 7 |
+
],
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_size": 4096,
|
| 11 |
+
"initializer_factor": 1.0,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"model_type": "blip-2",
|
| 14 |
+
"num_query_tokens": 32,
|
| 15 |
+
"output_size": [
|
| 16 |
+
7,
|
| 17 |
+
7
|
| 18 |
+
],
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"prompt_length": 5,
|
| 21 |
+
"qformer_config": {
|
| 22 |
+
"_name_or_path": "",
|
| 23 |
+
"add_cross_attention": false,
|
| 24 |
+
"architectures": null,
|
| 25 |
+
"attention_probs_dropout_prob": 0.1,
|
| 26 |
+
"bad_words_ids": null,
|
| 27 |
+
"begin_suppress_tokens": null,
|
| 28 |
+
"bos_token_id": null,
|
| 29 |
+
"chunk_size_feed_forward": 0,
|
| 30 |
+
"classifier_dropout": null,
|
| 31 |
+
"cross_attention_frequency": 2,
|
| 32 |
+
"cross_attention_hidden_size": null,
|
| 33 |
+
"decoder_start_token_id": null,
|
| 34 |
+
"diversity_penalty": 0.0,
|
| 35 |
+
"do_sample": false,
|
| 36 |
+
"early_stopping": false,
|
| 37 |
+
"encoder_hidden_size": 1408,
|
| 38 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 39 |
+
"eos_token_id": null,
|
| 40 |
+
"exponential_decay_length_penalty": null,
|
| 41 |
+
"finetuning_task": null,
|
| 42 |
+
"forced_bos_token_id": null,
|
| 43 |
+
"forced_eos_token_id": null,
|
| 44 |
+
"hidden_act": "gelu",
|
| 45 |
+
"hidden_dropout_prob": 0.1,
|
| 46 |
+
"hidden_size": 768,
|
| 47 |
+
"id2label": {
|
| 48 |
+
"0": "LABEL_0",
|
| 49 |
+
"1": "LABEL_1"
|
| 50 |
+
},
|
| 51 |
+
"initializer_range": 0.02,
|
| 52 |
+
"intermediate_size": 3072,
|
| 53 |
+
"is_decoder": false,
|
| 54 |
+
"is_encoder_decoder": false,
|
| 55 |
+
"label2id": {
|
| 56 |
+
"LABEL_0": 0,
|
| 57 |
+
"LABEL_1": 1
|
| 58 |
+
},
|
| 59 |
+
"layer_norm_eps": 1e-12,
|
| 60 |
+
"length_penalty": 1.0,
|
| 61 |
+
"max_length": 20,
|
| 62 |
+
"max_position_embeddings": 512,
|
| 63 |
+
"min_length": 0,
|
| 64 |
+
"model_type": "blip_2_qformer",
|
| 65 |
+
"no_repeat_ngram_size": 0,
|
| 66 |
+
"num_attention_heads": 12,
|
| 67 |
+
"num_beam_groups": 1,
|
| 68 |
+
"num_beams": 1,
|
| 69 |
+
"num_hidden_layers": 12,
|
| 70 |
+
"num_return_sequences": 1,
|
| 71 |
+
"output_attentions": false,
|
| 72 |
+
"output_hidden_states": false,
|
| 73 |
+
"output_scores": false,
|
| 74 |
+
"pad_token_id": 0,
|
| 75 |
+
"position_embedding_type": "absolute",
|
| 76 |
+
"prefix": null,
|
| 77 |
+
"problem_type": null,
|
| 78 |
+
"pruned_heads": {},
|
| 79 |
+
"remove_invalid_values": false,
|
| 80 |
+
"repetition_penalty": 1.0,
|
| 81 |
+
"return_dict": true,
|
| 82 |
+
"return_dict_in_generate": false,
|
| 83 |
+
"sep_token_id": null,
|
| 84 |
+
"suppress_tokens": null,
|
| 85 |
+
"task_specific_params": null,
|
| 86 |
+
"temperature": 1.0,
|
| 87 |
+
"tf_legacy_loss": false,
|
| 88 |
+
"tie_encoder_decoder": false,
|
| 89 |
+
"tie_word_embeddings": true,
|
| 90 |
+
"tokenizer_class": null,
|
| 91 |
+
"top_k": 50,
|
| 92 |
+
"top_p": 1.0,
|
| 93 |
+
"torch_dtype": null,
|
| 94 |
+
"torchscript": false,
|
| 95 |
+
"transformers_version": "4.28.0",
|
| 96 |
+
"typical_p": 1.0,
|
| 97 |
+
"use_bfloat16": false,
|
| 98 |
+
"vocab_size": 30522
|
| 99 |
+
},
|
| 100 |
+
"sampling_ratio": -1,
|
| 101 |
+
"text_config": {
|
| 102 |
+
"_name_or_path": "",
|
| 103 |
+
"add_cross_attention": false,
|
| 104 |
+
"architectures": [
|
| 105 |
+
"LlamaForCausalLM"
|
| 106 |
+
],
|
| 107 |
+
"bad_words_ids": null,
|
| 108 |
+
"begin_suppress_tokens": null,
|
| 109 |
+
"bos_token_id": 1,
|
| 110 |
+
"chunk_size_feed_forward": 0,
|
| 111 |
+
"cross_attention_hidden_size": null,
|
| 112 |
+
"decoder_start_token_id": null,
|
| 113 |
+
"diversity_penalty": 0.0,
|
| 114 |
+
"do_sample": false,
|
| 115 |
+
"early_stopping": false,
|
| 116 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 117 |
+
"eos_token_id": 2,
|
| 118 |
+
"exponential_decay_length_penalty": null,
|
| 119 |
+
"finetuning_task": null,
|
| 120 |
+
"forced_bos_token_id": null,
|
| 121 |
+
"forced_eos_token_id": null,
|
| 122 |
+
"hidden_act": "silu",
|
| 123 |
+
"hidden_size": 4096,
|
| 124 |
+
"id2label": {
|
| 125 |
+
"0": "LABEL_0",
|
| 126 |
+
"1": "LABEL_1"
|
| 127 |
+
},
|
| 128 |
+
"initializer_range": 0.02,
|
| 129 |
+
"intermediate_size": 11008,
|
| 130 |
+
"is_decoder": false,
|
| 131 |
+
"is_encoder_decoder": false,
|
| 132 |
+
"label2id": {
|
| 133 |
+
"LABEL_0": 0,
|
| 134 |
+
"LABEL_1": 1
|
| 135 |
+
},
|
| 136 |
+
"length_penalty": 1.0,
|
| 137 |
+
"max_length": 20,
|
| 138 |
+
"max_position_embeddings": 2048,
|
| 139 |
+
"max_sequence_length": 2048,
|
| 140 |
+
"min_length": 0,
|
| 141 |
+
"model_type": "llama",
|
| 142 |
+
"no_repeat_ngram_size": 0,
|
| 143 |
+
"num_attention_heads": 32,
|
| 144 |
+
"num_beam_groups": 1,
|
| 145 |
+
"num_beams": 1,
|
| 146 |
+
"num_hidden_layers": 32,
|
| 147 |
+
"num_return_sequences": 1,
|
| 148 |
+
"output_attentions": false,
|
| 149 |
+
"output_hidden_states": false,
|
| 150 |
+
"output_scores": false,
|
| 151 |
+
"pad_token_id": 0,
|
| 152 |
+
"prefix": null,
|
| 153 |
+
"problem_type": null,
|
| 154 |
+
"pruned_heads": {},
|
| 155 |
+
"remove_invalid_values": false,
|
| 156 |
+
"repetition_penalty": 1.0,
|
| 157 |
+
"return_dict": true,
|
| 158 |
+
"return_dict_in_generate": false,
|
| 159 |
+
"rms_norm_eps": 1e-06,
|
| 160 |
+
"sep_token_id": null,
|
| 161 |
+
"suppress_tokens": null,
|
| 162 |
+
"task_specific_params": null,
|
| 163 |
+
"temperature": 1.0,
|
| 164 |
+
"tf_legacy_loss": false,
|
| 165 |
+
"tie_encoder_decoder": false,
|
| 166 |
+
"tie_word_embeddings": false,
|
| 167 |
+
"tokenizer_class": null,
|
| 168 |
+
"top_k": 50,
|
| 169 |
+
"top_p": 1.0,
|
| 170 |
+
"torch_dtype": "float16",
|
| 171 |
+
"torchscript": false,
|
| 172 |
+
"transformers_version": "4.28.0",
|
| 173 |
+
"typical_p": 1.0,
|
| 174 |
+
"use_bfloat16": false,
|
| 175 |
+
"use_cache": true,
|
| 176 |
+
"vocab_size": 32003
|
| 177 |
+
},
|
| 178 |
+
"tie_word_embeddings": false,
|
| 179 |
+
"torch_dtype": "bfloat16",
|
| 180 |
+
"transformers_version": null,
|
| 181 |
+
"use_cache": false,
|
| 182 |
+
"use_decoder_only_language_model": true,
|
| 183 |
+
"use_window_attn": false,
|
| 184 |
+
"vision_config": {
|
| 185 |
+
"_name_or_path": "",
|
| 186 |
+
"add_cross_attention": false,
|
| 187 |
+
"architectures": null,
|
| 188 |
+
"attention_dropout": 0.0,
|
| 189 |
+
"bad_words_ids": null,
|
| 190 |
+
"begin_suppress_tokens": null,
|
| 191 |
+
"bos_token_id": null,
|
| 192 |
+
"chunk_size_feed_forward": 0,
|
| 193 |
+
"cross_attention_hidden_size": null,
|
| 194 |
+
"decoder_start_token_id": null,
|
| 195 |
+
"diversity_penalty": 0.0,
|
| 196 |
+
"do_sample": false,
|
| 197 |
+
"dropout": 0.0,
|
| 198 |
+
"early_stopping": false,
|
| 199 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 200 |
+
"eos_token_id": null,
|
| 201 |
+
"exponential_decay_length_penalty": null,
|
| 202 |
+
"finetuning_task": null,
|
| 203 |
+
"forced_bos_token_id": null,
|
| 204 |
+
"forced_eos_token_id": null,
|
| 205 |
+
"hidden_act": "gelu",
|
| 206 |
+
"hidden_size": 1408,
|
| 207 |
+
"id2label": {
|
| 208 |
+
"0": "LABEL_0",
|
| 209 |
+
"1": "LABEL_1"
|
| 210 |
+
},
|
| 211 |
+
"image_size": 364,
|
| 212 |
+
"initializer_factor": 1.0,
|
| 213 |
+
"initializer_range": 1e-10,
|
| 214 |
+
"intermediate_size": 6144,
|
| 215 |
+
"is_decoder": false,
|
| 216 |
+
"is_encoder_decoder": false,
|
| 217 |
+
"label2id": {
|
| 218 |
+
"LABEL_0": 0,
|
| 219 |
+
"LABEL_1": 1
|
| 220 |
+
},
|
| 221 |
+
"layer_norm_eps": 1e-05,
|
| 222 |
+
"length_penalty": 1.0,
|
| 223 |
+
"max_length": 20,
|
| 224 |
+
"min_length": 0,
|
| 225 |
+
"model_type": "blip_2_vision_model",
|
| 226 |
+
"no_repeat_ngram_size": 0,
|
| 227 |
+
"num_attention_heads": 16,
|
| 228 |
+
"num_beam_groups": 1,
|
| 229 |
+
"num_beams": 1,
|
| 230 |
+
"num_channels": 3,
|
| 231 |
+
"num_hidden_layers": 39,
|
| 232 |
+
"num_return_sequences": 1,
|
| 233 |
+
"output_attentions": false,
|
| 234 |
+
"output_hidden_states": false,
|
| 235 |
+
"output_scores": false,
|
| 236 |
+
"pad_token_id": null,
|
| 237 |
+
"patch_size": 14,
|
| 238 |
+
"prefix": null,
|
| 239 |
+
"problem_type": null,
|
| 240 |
+
"projection_dim": 512,
|
| 241 |
+
"pruned_heads": {},
|
| 242 |
+
"qkv_bias": true,
|
| 243 |
+
"remove_invalid_values": false,
|
| 244 |
+
"repetition_penalty": 1.0,
|
| 245 |
+
"return_dict": true,
|
| 246 |
+
"return_dict_in_generate": false,
|
| 247 |
+
"sep_token_id": null,
|
| 248 |
+
"suppress_tokens": null,
|
| 249 |
+
"task_specific_params": null,
|
| 250 |
+
"temperature": 1.0,
|
| 251 |
+
"tf_legacy_loss": false,
|
| 252 |
+
"tie_encoder_decoder": false,
|
| 253 |
+
"tie_word_embeddings": true,
|
| 254 |
+
"tokenizer_class": null,
|
| 255 |
+
"top_k": 50,
|
| 256 |
+
"top_p": 1.0,
|
| 257 |
+
"torch_dtype": null,
|
| 258 |
+
"torchscript": false,
|
| 259 |
+
"transformers_version": "4.28.0",
|
| 260 |
+
"typical_p": 1.0,
|
| 261 |
+
"use_bfloat16": false
|
| 262 |
+
},
|
| 263 |
+
"window_attn": [
|
| 264 |
+
false,
|
| 265 |
+
false,
|
| 266 |
+
false,
|
| 267 |
+
false,
|
| 268 |
+
false,
|
| 269 |
+
false,
|
| 270 |
+
false,
|
| 271 |
+
false,
|
| 272 |
+
false,
|
| 273 |
+
false,
|
| 274 |
+
false,
|
| 275 |
+
false,
|
| 276 |
+
false,
|
| 277 |
+
false,
|
| 278 |
+
false,
|
| 279 |
+
false,
|
| 280 |
+
false,
|
| 281 |
+
false,
|
| 282 |
+
false,
|
| 283 |
+
false,
|
| 284 |
+
false,
|
| 285 |
+
false,
|
| 286 |
+
false,
|
| 287 |
+
false,
|
| 288 |
+
false,
|
| 289 |
+
false,
|
| 290 |
+
false,
|
| 291 |
+
false,
|
| 292 |
+
false,
|
| 293 |
+
false,
|
| 294 |
+
false,
|
| 295 |
+
false,
|
| 296 |
+
false,
|
| 297 |
+
false,
|
| 298 |
+
false,
|
| 299 |
+
false,
|
| 300 |
+
false,
|
| 301 |
+
false,
|
| 302 |
+
false
|
| 303 |
+
],
|
| 304 |
+
"window_size": 16
|
| 305 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebee5957ad90d1b2f18fd40cd2999defaf772de629b8161e97a0305bf76bfee2
|
| 3 |
+
size 16527237973
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "<s>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": false,
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "</s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 22 |
+
"pad_token": null,
|
| 23 |
+
"sp_model_kwargs": {},
|
| 24 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 25 |
+
"unk_token": {
|
| 26 |
+
"__type": "AddedToken",
|
| 27 |
+
"content": "<unk>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": true,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1416 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.8600301010535368,
|
| 5 |
+
"global_step": 2000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 7.163323782234957e-07,
|
| 13 |
+
"loss": 2.0323,
|
| 14 |
+
"step": 10,
|
| 15 |
+
"train_LM": 0.9931640625
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.01,
|
| 19 |
+
"learning_rate": 1.4326647564469915e-06,
|
| 20 |
+
"loss": 2.0619,
|
| 21 |
+
"step": 20,
|
| 22 |
+
"train_LM": 1.0081787109375
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.01,
|
| 26 |
+
"learning_rate": 2.1489971346704872e-06,
|
| 27 |
+
"loss": 2.0231,
|
| 28 |
+
"step": 30,
|
| 29 |
+
"train_LM": 0.9273681640625
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.02,
|
| 33 |
+
"learning_rate": 2.865329512893983e-06,
|
| 34 |
+
"loss": 1.9714,
|
| 35 |
+
"step": 40,
|
| 36 |
+
"train_LM": 1.0142822265625
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.02,
|
| 40 |
+
"learning_rate": 3.5816618911174783e-06,
|
| 41 |
+
"loss": 1.8703,
|
| 42 |
+
"step": 50,
|
| 43 |
+
"train_LM": 0.8978271484375
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.03,
|
| 47 |
+
"learning_rate": 4.2979942693409744e-06,
|
| 48 |
+
"loss": 1.7548,
|
| 49 |
+
"step": 60,
|
| 50 |
+
"train_LM": 0.8167724609375
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.03,
|
| 54 |
+
"learning_rate": 5.01432664756447e-06,
|
| 55 |
+
"loss": 1.6451,
|
| 56 |
+
"step": 70,
|
| 57 |
+
"train_LM": 0.801025390625
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.03,
|
| 61 |
+
"learning_rate": 5.730659025787966e-06,
|
| 62 |
+
"loss": 1.5975,
|
| 63 |
+
"step": 80,
|
| 64 |
+
"train_LM": 0.773193359375
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.04,
|
| 68 |
+
"learning_rate": 6.446991404011461e-06,
|
| 69 |
+
"loss": 1.5418,
|
| 70 |
+
"step": 90,
|
| 71 |
+
"train_LM": 0.7550048828125
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.04,
|
| 75 |
+
"learning_rate": 7.1633237822349565e-06,
|
| 76 |
+
"loss": 1.519,
|
| 77 |
+
"step": 100,
|
| 78 |
+
"train_LM": 0.7457275390625
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.05,
|
| 82 |
+
"learning_rate": 7.879656160458452e-06,
|
| 83 |
+
"loss": 1.4836,
|
| 84 |
+
"step": 110,
|
| 85 |
+
"train_LM": 0.7720947265625
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.05,
|
| 89 |
+
"learning_rate": 8.595988538681949e-06,
|
| 90 |
+
"loss": 1.4816,
|
| 91 |
+
"step": 120,
|
| 92 |
+
"train_LM": 0.7650146484375
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.06,
|
| 96 |
+
"learning_rate": 9.312320916905444e-06,
|
| 97 |
+
"loss": 1.4378,
|
| 98 |
+
"step": 130,
|
| 99 |
+
"train_LM": 0.6661376953125
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.06,
|
| 103 |
+
"learning_rate": 1.002865329512894e-05,
|
| 104 |
+
"loss": 1.4387,
|
| 105 |
+
"step": 140,
|
| 106 |
+
"train_LM": 0.7376708984375
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.06,
|
| 110 |
+
"learning_rate": 1.0744985673352435e-05,
|
| 111 |
+
"loss": 1.4177,
|
| 112 |
+
"step": 150,
|
| 113 |
+
"train_LM": 0.724609375
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.07,
|
| 117 |
+
"learning_rate": 1.1461318051575932e-05,
|
| 118 |
+
"loss": 1.3805,
|
| 119 |
+
"step": 160,
|
| 120 |
+
"train_LM": 0.70050048828125
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.07,
|
| 124 |
+
"learning_rate": 1.2177650429799429e-05,
|
| 125 |
+
"loss": 1.3812,
|
| 126 |
+
"step": 170,
|
| 127 |
+
"train_LM": 0.7005615234375
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.08,
|
| 131 |
+
"learning_rate": 1.2893982808022922e-05,
|
| 132 |
+
"loss": 1.3465,
|
| 133 |
+
"step": 180,
|
| 134 |
+
"train_LM": 0.6685791015625
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.08,
|
| 138 |
+
"learning_rate": 1.361031518624642e-05,
|
| 139 |
+
"loss": 1.3491,
|
| 140 |
+
"step": 190,
|
| 141 |
+
"train_LM": 0.677490234375
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.09,
|
| 145 |
+
"learning_rate": 1.4326647564469913e-05,
|
| 146 |
+
"loss": 1.3211,
|
| 147 |
+
"step": 200,
|
| 148 |
+
"train_LM": 0.664306640625
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.09,
|
| 152 |
+
"learning_rate": 1.5042979942693412e-05,
|
| 153 |
+
"loss": 1.2896,
|
| 154 |
+
"step": 210,
|
| 155 |
+
"train_LM": 0.65985107421875
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.09,
|
| 159 |
+
"learning_rate": 1.5759312320916904e-05,
|
| 160 |
+
"loss": 1.2866,
|
| 161 |
+
"step": 220,
|
| 162 |
+
"train_LM": 0.65362548828125
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.1,
|
| 166 |
+
"learning_rate": 1.64756446991404e-05,
|
| 167 |
+
"loss": 1.2578,
|
| 168 |
+
"step": 230,
|
| 169 |
+
"train_LM": 0.5946044921875
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.1,
|
| 173 |
+
"learning_rate": 1.7191977077363898e-05,
|
| 174 |
+
"loss": 1.2582,
|
| 175 |
+
"step": 240,
|
| 176 |
+
"train_LM": 0.61859130859375
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.11,
|
| 180 |
+
"learning_rate": 1.7908309455587395e-05,
|
| 181 |
+
"loss": 1.2398,
|
| 182 |
+
"step": 250,
|
| 183 |
+
"train_LM": 0.613525390625
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.11,
|
| 187 |
+
"learning_rate": 1.862464183381089e-05,
|
| 188 |
+
"loss": 1.2373,
|
| 189 |
+
"step": 260,
|
| 190 |
+
"train_LM": 0.6136474609375
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.12,
|
| 194 |
+
"learning_rate": 1.9340974212034385e-05,
|
| 195 |
+
"loss": 1.2273,
|
| 196 |
+
"step": 270,
|
| 197 |
+
"train_LM": 0.61920166015625
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.12,
|
| 201 |
+
"learning_rate": 2.005730659025788e-05,
|
| 202 |
+
"loss": 1.226,
|
| 203 |
+
"step": 280,
|
| 204 |
+
"train_LM": 0.62518310546875
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.12,
|
| 208 |
+
"learning_rate": 2.0773638968481376e-05,
|
| 209 |
+
"loss": 1.2234,
|
| 210 |
+
"step": 290,
|
| 211 |
+
"train_LM": 0.6243896484375
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.13,
|
| 215 |
+
"learning_rate": 2.148997134670487e-05,
|
| 216 |
+
"loss": 1.2205,
|
| 217 |
+
"step": 300,
|
| 218 |
+
"train_LM": 0.5906982421875
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.13,
|
| 222 |
+
"learning_rate": 2.2206303724928367e-05,
|
| 223 |
+
"loss": 1.2122,
|
| 224 |
+
"step": 310,
|
| 225 |
+
"train_LM": 0.61639404296875
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.14,
|
| 229 |
+
"learning_rate": 2.2922636103151864e-05,
|
| 230 |
+
"loss": 1.203,
|
| 231 |
+
"step": 320,
|
| 232 |
+
"train_LM": 0.5921630859375
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.14,
|
| 236 |
+
"learning_rate": 2.363896848137536e-05,
|
| 237 |
+
"loss": 1.2039,
|
| 238 |
+
"step": 330,
|
| 239 |
+
"train_LM": 0.60321044921875
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.15,
|
| 243 |
+
"learning_rate": 2.4355300859598858e-05,
|
| 244 |
+
"loss": 1.1911,
|
| 245 |
+
"step": 340,
|
| 246 |
+
"train_LM": 0.60760498046875
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.15,
|
| 250 |
+
"learning_rate": 2.507163323782235e-05,
|
| 251 |
+
"loss": 1.1849,
|
| 252 |
+
"step": 350,
|
| 253 |
+
"train_LM": 0.6004638671875
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.15,
|
| 257 |
+
"learning_rate": 2.5787965616045845e-05,
|
| 258 |
+
"loss": 1.1938,
|
| 259 |
+
"step": 360,
|
| 260 |
+
"train_LM": 0.6033935546875
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.16,
|
| 264 |
+
"learning_rate": 2.6504297994269345e-05,
|
| 265 |
+
"loss": 1.1869,
|
| 266 |
+
"step": 370,
|
| 267 |
+
"train_LM": 0.5823974609375
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.16,
|
| 271 |
+
"learning_rate": 2.722063037249284e-05,
|
| 272 |
+
"loss": 1.1832,
|
| 273 |
+
"step": 380,
|
| 274 |
+
"train_LM": 0.58441162109375
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.17,
|
| 278 |
+
"learning_rate": 2.7936962750716332e-05,
|
| 279 |
+
"loss": 1.1731,
|
| 280 |
+
"step": 390,
|
| 281 |
+
"train_LM": 0.5772705078125
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.17,
|
| 285 |
+
"learning_rate": 2.8653295128939826e-05,
|
| 286 |
+
"loss": 1.1723,
|
| 287 |
+
"step": 400,
|
| 288 |
+
"train_LM": 0.58673095703125
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.18,
|
| 292 |
+
"learning_rate": 2.9369627507163327e-05,
|
| 293 |
+
"loss": 1.1723,
|
| 294 |
+
"step": 410,
|
| 295 |
+
"train_LM": 0.5994873046875
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.18,
|
| 299 |
+
"learning_rate": 3.0085959885386824e-05,
|
| 300 |
+
"loss": 1.1728,
|
| 301 |
+
"step": 420,
|
| 302 |
+
"train_LM": 0.57855224609375
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.18,
|
| 306 |
+
"learning_rate": 3.0802292263610314e-05,
|
| 307 |
+
"loss": 1.165,
|
| 308 |
+
"step": 430,
|
| 309 |
+
"train_LM": 0.5885009765625
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.19,
|
| 313 |
+
"learning_rate": 3.151862464183381e-05,
|
| 314 |
+
"loss": 1.1695,
|
| 315 |
+
"step": 440,
|
| 316 |
+
"train_LM": 0.57763671875
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.19,
|
| 320 |
+
"learning_rate": 3.223495702005731e-05,
|
| 321 |
+
"loss": 1.1596,
|
| 322 |
+
"step": 450,
|
| 323 |
+
"train_LM": 0.56146240234375
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.2,
|
| 327 |
+
"learning_rate": 3.29512893982808e-05,
|
| 328 |
+
"loss": 1.1488,
|
| 329 |
+
"step": 460,
|
| 330 |
+
"train_LM": 0.57391357421875
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.2,
|
| 334 |
+
"learning_rate": 3.36676217765043e-05,
|
| 335 |
+
"loss": 1.1643,
|
| 336 |
+
"step": 470,
|
| 337 |
+
"train_LM": 0.56817626953125
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.21,
|
| 341 |
+
"learning_rate": 3.4383954154727795e-05,
|
| 342 |
+
"loss": 1.1438,
|
| 343 |
+
"step": 480,
|
| 344 |
+
"train_LM": 0.56756591796875
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.21,
|
| 348 |
+
"learning_rate": 3.5100286532951296e-05,
|
| 349 |
+
"loss": 1.1518,
|
| 350 |
+
"step": 490,
|
| 351 |
+
"train_LM": 0.57366943359375
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.22,
|
| 355 |
+
"learning_rate": 3.581661891117479e-05,
|
| 356 |
+
"loss": 1.1613,
|
| 357 |
+
"step": 500,
|
| 358 |
+
"train_LM": 0.59564208984375
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.22,
|
| 362 |
+
"learning_rate": 3.653295128939828e-05,
|
| 363 |
+
"loss": 1.1456,
|
| 364 |
+
"step": 510,
|
| 365 |
+
"train_LM": 0.5662841796875
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.22,
|
| 369 |
+
"learning_rate": 3.724928366762178e-05,
|
| 370 |
+
"loss": 1.1426,
|
| 371 |
+
"step": 520,
|
| 372 |
+
"train_LM": 0.5601806640625
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.23,
|
| 376 |
+
"learning_rate": 3.796561604584528e-05,
|
| 377 |
+
"loss": 1.1473,
|
| 378 |
+
"step": 530,
|
| 379 |
+
"train_LM": 0.5787353515625
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 0.23,
|
| 383 |
+
"learning_rate": 3.868194842406877e-05,
|
| 384 |
+
"loss": 1.1421,
|
| 385 |
+
"step": 540,
|
| 386 |
+
"train_LM": 0.574462890625
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.24,
|
| 390 |
+
"learning_rate": 3.9398280802292264e-05,
|
| 391 |
+
"loss": 1.1396,
|
| 392 |
+
"step": 550,
|
| 393 |
+
"train_LM": 0.57293701171875
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.24,
|
| 397 |
+
"learning_rate": 4.011461318051576e-05,
|
| 398 |
+
"loss": 1.1366,
|
| 399 |
+
"step": 560,
|
| 400 |
+
"train_LM": 0.576904296875
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.25,
|
| 404 |
+
"learning_rate": 4.083094555873926e-05,
|
| 405 |
+
"loss": 1.1338,
|
| 406 |
+
"step": 570,
|
| 407 |
+
"train_LM": 0.5560302734375
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.25,
|
| 411 |
+
"learning_rate": 4.154727793696275e-05,
|
| 412 |
+
"loss": 1.1359,
|
| 413 |
+
"step": 580,
|
| 414 |
+
"train_LM": 0.5733642578125
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.25,
|
| 418 |
+
"learning_rate": 4.2263610315186246e-05,
|
| 419 |
+
"loss": 1.1333,
|
| 420 |
+
"step": 590,
|
| 421 |
+
"train_LM": 0.57745361328125
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.26,
|
| 425 |
+
"learning_rate": 4.297994269340974e-05,
|
| 426 |
+
"loss": 1.1396,
|
| 427 |
+
"step": 600,
|
| 428 |
+
"train_LM": 0.5357666015625
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.26,
|
| 432 |
+
"learning_rate": 4.369627507163324e-05,
|
| 433 |
+
"loss": 1.1312,
|
| 434 |
+
"step": 610,
|
| 435 |
+
"train_LM": 0.58441162109375
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.27,
|
| 439 |
+
"learning_rate": 4.441260744985673e-05,
|
| 440 |
+
"loss": 1.137,
|
| 441 |
+
"step": 620,
|
| 442 |
+
"train_LM": 0.56988525390625
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.27,
|
| 446 |
+
"learning_rate": 4.512893982808023e-05,
|
| 447 |
+
"loss": 1.1236,
|
| 448 |
+
"step": 630,
|
| 449 |
+
"train_LM": 0.5614013671875
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.28,
|
| 453 |
+
"learning_rate": 4.584527220630373e-05,
|
| 454 |
+
"loss": 1.111,
|
| 455 |
+
"step": 640,
|
| 456 |
+
"train_LM": 0.5447998046875
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.28,
|
| 460 |
+
"learning_rate": 4.656160458452722e-05,
|
| 461 |
+
"loss": 1.1196,
|
| 462 |
+
"step": 650,
|
| 463 |
+
"train_LM": 0.5562744140625
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.28,
|
| 467 |
+
"learning_rate": 4.727793696275072e-05,
|
| 468 |
+
"loss": 1.124,
|
| 469 |
+
"step": 660,
|
| 470 |
+
"train_LM": 0.55029296875
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.29,
|
| 474 |
+
"learning_rate": 4.7994269340974215e-05,
|
| 475 |
+
"loss": 1.1239,
|
| 476 |
+
"step": 670,
|
| 477 |
+
"train_LM": 0.55340576171875
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.29,
|
| 481 |
+
"learning_rate": 4.8710601719197715e-05,
|
| 482 |
+
"loss": 1.1119,
|
| 483 |
+
"step": 680,
|
| 484 |
+
"train_LM": 0.54693603515625
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.3,
|
| 488 |
+
"learning_rate": 4.942693409742121e-05,
|
| 489 |
+
"loss": 1.119,
|
| 490 |
+
"step": 690,
|
| 491 |
+
"train_LM": 0.5472412109375
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.3,
|
| 495 |
+
"learning_rate": 4.9999987475354104e-05,
|
| 496 |
+
"loss": 1.11,
|
| 497 |
+
"step": 700,
|
| 498 |
+
"train_LM": 0.550048828125
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.31,
|
| 502 |
+
"learning_rate": 4.9999549114065355e-05,
|
| 503 |
+
"loss": 1.1024,
|
| 504 |
+
"step": 710,
|
| 505 |
+
"train_LM": 0.531005859375
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.31,
|
| 509 |
+
"learning_rate": 4.999848453303098e-05,
|
| 510 |
+
"loss": 1.1101,
|
| 511 |
+
"step": 720,
|
| 512 |
+
"train_LM": 0.5552978515625
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.31,
|
| 516 |
+
"learning_rate": 4.9996793758917936e-05,
|
| 517 |
+
"loss": 1.0939,
|
| 518 |
+
"step": 730,
|
| 519 |
+
"train_LM": 0.5531005859375
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.32,
|
| 523 |
+
"learning_rate": 4.999447683407884e-05,
|
| 524 |
+
"loss": 1.1049,
|
| 525 |
+
"step": 740,
|
| 526 |
+
"train_LM": 0.5413818359375
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.32,
|
| 530 |
+
"learning_rate": 4.9991533816550875e-05,
|
| 531 |
+
"loss": 1.1092,
|
| 532 |
+
"step": 750,
|
| 533 |
+
"train_LM": 0.5496826171875
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.33,
|
| 537 |
+
"learning_rate": 4.998796478005443e-05,
|
| 538 |
+
"loss": 1.1103,
|
| 539 |
+
"step": 760,
|
| 540 |
+
"train_LM": 0.5521240234375
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.33,
|
| 544 |
+
"learning_rate": 4.998376981399114e-05,
|
| 545 |
+
"loss": 1.0911,
|
| 546 |
+
"step": 770,
|
| 547 |
+
"train_LM": 0.52960205078125
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.34,
|
| 551 |
+
"learning_rate": 4.9978949023441736e-05,
|
| 552 |
+
"loss": 1.1009,
|
| 553 |
+
"step": 780,
|
| 554 |
+
"train_LM": 0.54736328125
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.34,
|
| 558 |
+
"learning_rate": 4.997350252916335e-05,
|
| 559 |
+
"loss": 1.1047,
|
| 560 |
+
"step": 790,
|
| 561 |
+
"train_LM": 0.566162109375
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.34,
|
| 565 |
+
"learning_rate": 4.9967430467586555e-05,
|
| 566 |
+
"loss": 1.1017,
|
| 567 |
+
"step": 800,
|
| 568 |
+
"train_LM": 0.55645751953125
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.35,
|
| 572 |
+
"learning_rate": 4.9960732990811876e-05,
|
| 573 |
+
"loss": 1.0986,
|
| 574 |
+
"step": 810,
|
| 575 |
+
"train_LM": 0.54345703125
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.35,
|
| 579 |
+
"learning_rate": 4.995341026660603e-05,
|
| 580 |
+
"loss": 1.0976,
|
| 581 |
+
"step": 820,
|
| 582 |
+
"train_LM": 0.5579833984375
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 0.36,
|
| 586 |
+
"learning_rate": 4.994546247839769e-05,
|
| 587 |
+
"loss": 1.1027,
|
| 588 |
+
"step": 830,
|
| 589 |
+
"train_LM": 0.53839111328125
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.36,
|
| 593 |
+
"learning_rate": 4.9936889825272945e-05,
|
| 594 |
+
"loss": 1.0909,
|
| 595 |
+
"step": 840,
|
| 596 |
+
"train_LM": 0.53472900390625
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.37,
|
| 600 |
+
"learning_rate": 4.9927692521970235e-05,
|
| 601 |
+
"loss": 1.0878,
|
| 602 |
+
"step": 850,
|
| 603 |
+
"train_LM": 0.5250244140625
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.37,
|
| 607 |
+
"learning_rate": 4.9917870798875055e-05,
|
| 608 |
+
"loss": 1.0945,
|
| 609 |
+
"step": 860,
|
| 610 |
+
"train_LM": 0.54498291015625
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.37,
|
| 614 |
+
"learning_rate": 4.99074249020141e-05,
|
| 615 |
+
"loss": 1.0837,
|
| 616 |
+
"step": 870,
|
| 617 |
+
"train_LM": 0.54376220703125
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.38,
|
| 621 |
+
"learning_rate": 4.9896355093049174e-05,
|
| 622 |
+
"loss": 1.094,
|
| 623 |
+
"step": 880,
|
| 624 |
+
"train_LM": 0.558349609375
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.38,
|
| 628 |
+
"learning_rate": 4.9884661649270595e-05,
|
| 629 |
+
"loss": 1.0913,
|
| 630 |
+
"step": 890,
|
| 631 |
+
"train_LM": 0.5499267578125
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.39,
|
| 635 |
+
"learning_rate": 4.987234486359025e-05,
|
| 636 |
+
"loss": 1.0946,
|
| 637 |
+
"step": 900,
|
| 638 |
+
"train_LM": 0.5506591796875
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.39,
|
| 642 |
+
"learning_rate": 4.98594050445343e-05,
|
| 643 |
+
"loss": 1.0881,
|
| 644 |
+
"step": 910,
|
| 645 |
+
"train_LM": 0.531005859375
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.4,
|
| 649 |
+
"learning_rate": 4.984584251623539e-05,
|
| 650 |
+
"loss": 1.0968,
|
| 651 |
+
"step": 920,
|
| 652 |
+
"train_LM": 0.55133056640625
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 0.4,
|
| 656 |
+
"learning_rate": 4.983165761842456e-05,
|
| 657 |
+
"loss": 1.087,
|
| 658 |
+
"step": 930,
|
| 659 |
+
"train_LM": 0.5438232421875
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.4,
|
| 663 |
+
"learning_rate": 4.981685070642274e-05,
|
| 664 |
+
"loss": 1.0843,
|
| 665 |
+
"step": 940,
|
| 666 |
+
"train_LM": 0.55621337890625
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.41,
|
| 670 |
+
"learning_rate": 4.980142215113186e-05,
|
| 671 |
+
"loss": 1.0778,
|
| 672 |
+
"step": 950,
|
| 673 |
+
"train_LM": 0.54248046875
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.41,
|
| 677 |
+
"learning_rate": 4.978537233902553e-05,
|
| 678 |
+
"loss": 1.0744,
|
| 679 |
+
"step": 960,
|
| 680 |
+
"train_LM": 0.53277587890625
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.42,
|
| 684 |
+
"learning_rate": 4.976870167213935e-05,
|
| 685 |
+
"loss": 1.078,
|
| 686 |
+
"step": 970,
|
| 687 |
+
"train_LM": 0.54443359375
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.42,
|
| 691 |
+
"learning_rate": 4.9751410568060905e-05,
|
| 692 |
+
"loss": 1.0804,
|
| 693 |
+
"step": 980,
|
| 694 |
+
"train_LM": 0.54437255859375
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 0.43,
|
| 698 |
+
"learning_rate": 4.973349945991923e-05,
|
| 699 |
+
"loss": 1.089,
|
| 700 |
+
"step": 990,
|
| 701 |
+
"train_LM": 0.53656005859375
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 0.43,
|
| 705 |
+
"learning_rate": 4.971496879637401e-05,
|
| 706 |
+
"loss": 1.0913,
|
| 707 |
+
"step": 1000,
|
| 708 |
+
"train_LM": 0.54193115234375
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.43,
|
| 712 |
+
"learning_rate": 4.9695819041604285e-05,
|
| 713 |
+
"loss": 1.0801,
|
| 714 |
+
"step": 1010,
|
| 715 |
+
"train_LM": 0.52960205078125
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.44,
|
| 719 |
+
"learning_rate": 4.967605067529692e-05,
|
| 720 |
+
"loss": 1.0743,
|
| 721 |
+
"step": 1020,
|
| 722 |
+
"train_LM": 0.51727294921875
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.44,
|
| 726 |
+
"learning_rate": 4.9655664192634484e-05,
|
| 727 |
+
"loss": 1.079,
|
| 728 |
+
"step": 1030,
|
| 729 |
+
"train_LM": 0.5628662109375
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.45,
|
| 733 |
+
"learning_rate": 4.963466010428291e-05,
|
| 734 |
+
"loss": 1.08,
|
| 735 |
+
"step": 1040,
|
| 736 |
+
"train_LM": 0.53753662109375
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.45,
|
| 740 |
+
"learning_rate": 4.961303893637867e-05,
|
| 741 |
+
"loss": 1.069,
|
| 742 |
+
"step": 1050,
|
| 743 |
+
"train_LM": 0.56597900390625
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.46,
|
| 747 |
+
"learning_rate": 4.9590801230515635e-05,
|
| 748 |
+
"loss": 1.0711,
|
| 749 |
+
"step": 1060,
|
| 750 |
+
"train_LM": 0.53118896484375
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.46,
|
| 754 |
+
"learning_rate": 4.956794754373148e-05,
|
| 755 |
+
"loss": 1.0853,
|
| 756 |
+
"step": 1070,
|
| 757 |
+
"train_LM": 0.540771484375
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.46,
|
| 761 |
+
"learning_rate": 4.954447844849371e-05,
|
| 762 |
+
"loss": 1.0756,
|
| 763 |
+
"step": 1080,
|
| 764 |
+
"train_LM": 0.5364990234375
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.47,
|
| 768 |
+
"learning_rate": 4.9520394532685364e-05,
|
| 769 |
+
"loss": 1.0706,
|
| 770 |
+
"step": 1090,
|
| 771 |
+
"train_LM": 0.5113525390625
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.47,
|
| 775 |
+
"learning_rate": 4.949569639959028e-05,
|
| 776 |
+
"loss": 1.0856,
|
| 777 |
+
"step": 1100,
|
| 778 |
+
"train_LM": 0.54327392578125
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.48,
|
| 782 |
+
"learning_rate": 4.9470384667877947e-05,
|
| 783 |
+
"loss": 1.0617,
|
| 784 |
+
"step": 1110,
|
| 785 |
+
"train_LM": 0.5179443359375
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.48,
|
| 789 |
+
"learning_rate": 4.944445997158805e-05,
|
| 790 |
+
"loss": 1.0779,
|
| 791 |
+
"step": 1120,
|
| 792 |
+
"train_LM": 0.5509033203125
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.49,
|
| 796 |
+
"learning_rate": 4.9417922960114583e-05,
|
| 797 |
+
"loss": 1.0668,
|
| 798 |
+
"step": 1130,
|
| 799 |
+
"train_LM": 0.538330078125
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.49,
|
| 803 |
+
"learning_rate": 4.9390774298189544e-05,
|
| 804 |
+
"loss": 1.0653,
|
| 805 |
+
"step": 1140,
|
| 806 |
+
"train_LM": 0.52734375
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.49,
|
| 810 |
+
"learning_rate": 4.936301466586633e-05,
|
| 811 |
+
"loss": 1.0694,
|
| 812 |
+
"step": 1150,
|
| 813 |
+
"train_LM": 0.51678466796875
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.5,
|
| 817 |
+
"learning_rate": 4.933464475850267e-05,
|
| 818 |
+
"loss": 1.0747,
|
| 819 |
+
"step": 1160,
|
| 820 |
+
"train_LM": 0.5343017578125
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.5,
|
| 824 |
+
"learning_rate": 4.930566528674323e-05,
|
| 825 |
+
"loss": 1.0688,
|
| 826 |
+
"step": 1170,
|
| 827 |
+
"train_LM": 0.5302734375
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.51,
|
| 831 |
+
"learning_rate": 4.927607697650182e-05,
|
| 832 |
+
"loss": 1.0663,
|
| 833 |
+
"step": 1180,
|
| 834 |
+
"train_LM": 0.51226806640625
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.51,
|
| 838 |
+
"learning_rate": 4.9245880568943134e-05,
|
| 839 |
+
"loss": 1.0716,
|
| 840 |
+
"step": 1190,
|
| 841 |
+
"train_LM": 0.5389404296875
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.52,
|
| 845 |
+
"learning_rate": 4.921507682046432e-05,
|
| 846 |
+
"loss": 1.0654,
|
| 847 |
+
"step": 1200,
|
| 848 |
+
"train_LM": 0.53076171875
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.52,
|
| 852 |
+
"learning_rate": 4.9183666502675885e-05,
|
| 853 |
+
"loss": 1.0718,
|
| 854 |
+
"step": 1210,
|
| 855 |
+
"train_LM": 0.521728515625
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.52,
|
| 859 |
+
"learning_rate": 4.915165040238249e-05,
|
| 860 |
+
"loss": 1.0618,
|
| 861 |
+
"step": 1220,
|
| 862 |
+
"train_LM": 0.53070068359375
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.53,
|
| 866 |
+
"learning_rate": 4.911902932156315e-05,
|
| 867 |
+
"loss": 1.0549,
|
| 868 |
+
"step": 1230,
|
| 869 |
+
"train_LM": 0.5350341796875
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.53,
|
| 873 |
+
"learning_rate": 4.9085804077351206e-05,
|
| 874 |
+
"loss": 1.0609,
|
| 875 |
+
"step": 1240,
|
| 876 |
+
"train_LM": 0.4913330078125
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.54,
|
| 880 |
+
"learning_rate": 4.905197550201384e-05,
|
| 881 |
+
"loss": 1.0628,
|
| 882 |
+
"step": 1250,
|
| 883 |
+
"train_LM": 0.5445556640625
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.54,
|
| 887 |
+
"learning_rate": 4.901754444293118e-05,
|
| 888 |
+
"loss": 1.0682,
|
| 889 |
+
"step": 1260,
|
| 890 |
+
"train_LM": 0.521484375
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.55,
|
| 894 |
+
"learning_rate": 4.898251176257517e-05,
|
| 895 |
+
"loss": 1.0526,
|
| 896 |
+
"step": 1270,
|
| 897 |
+
"train_LM": 0.51763916015625
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.55,
|
| 901 |
+
"learning_rate": 4.894687833848787e-05,
|
| 902 |
+
"loss": 1.0571,
|
| 903 |
+
"step": 1280,
|
| 904 |
+
"train_LM": 0.52374267578125
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.55,
|
| 908 |
+
"learning_rate": 4.891064506325953e-05,
|
| 909 |
+
"loss": 1.0637,
|
| 910 |
+
"step": 1290,
|
| 911 |
+
"train_LM": 0.55963134765625
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.56,
|
| 915 |
+
"learning_rate": 4.887381284450622e-05,
|
| 916 |
+
"loss": 1.0653,
|
| 917 |
+
"step": 1300,
|
| 918 |
+
"train_LM": 0.522216796875
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.56,
|
| 922 |
+
"learning_rate": 4.883638260484709e-05,
|
| 923 |
+
"loss": 1.0611,
|
| 924 |
+
"step": 1310,
|
| 925 |
+
"train_LM": 0.5406494140625
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.57,
|
| 929 |
+
"learning_rate": 4.8798355281881235e-05,
|
| 930 |
+
"loss": 1.068,
|
| 931 |
+
"step": 1320,
|
| 932 |
+
"train_LM": 0.5301513671875
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.57,
|
| 936 |
+
"learning_rate": 4.8759731828164284e-05,
|
| 937 |
+
"loss": 1.0563,
|
| 938 |
+
"step": 1330,
|
| 939 |
+
"train_LM": 0.52178955078125
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.58,
|
| 943 |
+
"learning_rate": 4.872051321118444e-05,
|
| 944 |
+
"loss": 1.0508,
|
| 945 |
+
"step": 1340,
|
| 946 |
+
"train_LM": 0.5125732421875
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.58,
|
| 950 |
+
"learning_rate": 4.868070041333833e-05,
|
| 951 |
+
"loss": 1.0569,
|
| 952 |
+
"step": 1350,
|
| 953 |
+
"train_LM": 0.53240966796875
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.58,
|
| 957 |
+
"learning_rate": 4.864029443190633e-05,
|
| 958 |
+
"loss": 1.0549,
|
| 959 |
+
"step": 1360,
|
| 960 |
+
"train_LM": 0.5390625
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.59,
|
| 964 |
+
"learning_rate": 4.859929627902765e-05,
|
| 965 |
+
"loss": 1.0572,
|
| 966 |
+
"step": 1370,
|
| 967 |
+
"train_LM": 0.536376953125
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.59,
|
| 971 |
+
"learning_rate": 4.8557706981674906e-05,
|
| 972 |
+
"loss": 1.0526,
|
| 973 |
+
"step": 1380,
|
| 974 |
+
"train_LM": 0.52813720703125
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.6,
|
| 978 |
+
"learning_rate": 4.851552758162847e-05,
|
| 979 |
+
"loss": 1.0554,
|
| 980 |
+
"step": 1390,
|
| 981 |
+
"train_LM": 0.5244140625
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.6,
|
| 985 |
+
"learning_rate": 4.847275913545032e-05,
|
| 986 |
+
"loss": 1.057,
|
| 987 |
+
"step": 1400,
|
| 988 |
+
"train_LM": 0.54168701171875
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.61,
|
| 992 |
+
"learning_rate": 4.8429402714457586e-05,
|
| 993 |
+
"loss": 1.0611,
|
| 994 |
+
"step": 1410,
|
| 995 |
+
"train_LM": 0.58660888671875
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.61,
|
| 999 |
+
"learning_rate": 4.8385459404695755e-05,
|
| 1000 |
+
"loss": 1.0476,
|
| 1001 |
+
"step": 1420,
|
| 1002 |
+
"train_LM": 0.52166748046875
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.61,
|
| 1006 |
+
"learning_rate": 4.834093030691139e-05,
|
| 1007 |
+
"loss": 1.0513,
|
| 1008 |
+
"step": 1430,
|
| 1009 |
+
"train_LM": 0.509033203125
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.62,
|
| 1013 |
+
"learning_rate": 4.829581653652463e-05,
|
| 1014 |
+
"loss": 1.0501,
|
| 1015 |
+
"step": 1440,
|
| 1016 |
+
"train_LM": 0.517578125
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.62,
|
| 1020 |
+
"learning_rate": 4.82501192236012e-05,
|
| 1021 |
+
"loss": 1.0573,
|
| 1022 |
+
"step": 1450,
|
| 1023 |
+
"train_LM": 0.55438232421875
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.63,
|
| 1027 |
+
"learning_rate": 4.8203839512824145e-05,
|
| 1028 |
+
"loss": 1.0544,
|
| 1029 |
+
"step": 1460,
|
| 1030 |
+
"train_LM": 0.5068359375
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.63,
|
| 1034 |
+
"learning_rate": 4.8156978563465114e-05,
|
| 1035 |
+
"loss": 1.0505,
|
| 1036 |
+
"step": 1470,
|
| 1037 |
+
"train_LM": 0.5374755859375
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.64,
|
| 1041 |
+
"learning_rate": 4.8109537549355346e-05,
|
| 1042 |
+
"loss": 1.0478,
|
| 1043 |
+
"step": 1480,
|
| 1044 |
+
"train_LM": 0.5125732421875
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.64,
|
| 1048 |
+
"learning_rate": 4.806151765885627e-05,
|
| 1049 |
+
"loss": 1.0509,
|
| 1050 |
+
"step": 1490,
|
| 1051 |
+
"train_LM": 0.51953125
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.65,
|
| 1055 |
+
"learning_rate": 4.801292009482972e-05,
|
| 1056 |
+
"loss": 1.0453,
|
| 1057 |
+
"step": 1500,
|
| 1058 |
+
"train_LM": 0.54364013671875
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.65,
|
| 1062 |
+
"learning_rate": 4.796374607460782e-05,
|
| 1063 |
+
"loss": 1.0535,
|
| 1064 |
+
"step": 1510,
|
| 1065 |
+
"train_LM": 0.5399169921875
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.65,
|
| 1069 |
+
"learning_rate": 4.7913996829962494e-05,
|
| 1070 |
+
"loss": 1.0352,
|
| 1071 |
+
"step": 1520,
|
| 1072 |
+
"train_LM": 0.52886962890625
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.66,
|
| 1076 |
+
"learning_rate": 4.786367360707458e-05,
|
| 1077 |
+
"loss": 1.0477,
|
| 1078 |
+
"step": 1530,
|
| 1079 |
+
"train_LM": 0.5167236328125
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.66,
|
| 1083 |
+
"learning_rate": 4.7812777666502634e-05,
|
| 1084 |
+
"loss": 1.0496,
|
| 1085 |
+
"step": 1540,
|
| 1086 |
+
"train_LM": 0.51416015625
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.67,
|
| 1090 |
+
"learning_rate": 4.776131028315139e-05,
|
| 1091 |
+
"loss": 1.036,
|
| 1092 |
+
"step": 1550,
|
| 1093 |
+
"train_LM": 0.50091552734375
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.67,
|
| 1097 |
+
"learning_rate": 4.770927274623975e-05,
|
| 1098 |
+
"loss": 1.0462,
|
| 1099 |
+
"step": 1560,
|
| 1100 |
+
"train_LM": 0.557861328125
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.68,
|
| 1104 |
+
"learning_rate": 4.765666635926857e-05,
|
| 1105 |
+
"loss": 1.04,
|
| 1106 |
+
"step": 1570,
|
| 1107 |
+
"train_LM": 0.51849365234375
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.68,
|
| 1111 |
+
"learning_rate": 4.760349243998793e-05,
|
| 1112 |
+
"loss": 1.0457,
|
| 1113 |
+
"step": 1580,
|
| 1114 |
+
"train_LM": 0.51068115234375
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.68,
|
| 1118 |
+
"learning_rate": 4.754975232036418e-05,
|
| 1119 |
+
"loss": 1.048,
|
| 1120 |
+
"step": 1590,
|
| 1121 |
+
"train_LM": 0.53448486328125
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.69,
|
| 1125 |
+
"learning_rate": 4.749544734654657e-05,
|
| 1126 |
+
"loss": 1.0346,
|
| 1127 |
+
"step": 1600,
|
| 1128 |
+
"train_LM": 0.52642822265625
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.69,
|
| 1132 |
+
"learning_rate": 4.74405788788335e-05,
|
| 1133 |
+
"loss": 1.0532,
|
| 1134 |
+
"step": 1610,
|
| 1135 |
+
"train_LM": 0.53570556640625
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.7,
|
| 1139 |
+
"learning_rate": 4.738514829163848e-05,
|
| 1140 |
+
"loss": 1.0328,
|
| 1141 |
+
"step": 1620,
|
| 1142 |
+
"train_LM": 0.514404296875
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.7,
|
| 1146 |
+
"learning_rate": 4.732915697345567e-05,
|
| 1147 |
+
"loss": 1.0386,
|
| 1148 |
+
"step": 1630,
|
| 1149 |
+
"train_LM": 0.5050048828125
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.71,
|
| 1153 |
+
"learning_rate": 4.7272606326825144e-05,
|
| 1154 |
+
"loss": 1.0348,
|
| 1155 |
+
"step": 1640,
|
| 1156 |
+
"train_LM": 0.51141357421875
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.71,
|
| 1160 |
+
"learning_rate": 4.72154977682977e-05,
|
| 1161 |
+
"loss": 1.0445,
|
| 1162 |
+
"step": 1650,
|
| 1163 |
+
"train_LM": 0.53424072265625
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.71,
|
| 1167 |
+
"learning_rate": 4.715783272839943e-05,
|
| 1168 |
+
"loss": 1.0307,
|
| 1169 |
+
"step": 1660,
|
| 1170 |
+
"train_LM": 0.5203857421875
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.72,
|
| 1174 |
+
"learning_rate": 4.709961265159583e-05,
|
| 1175 |
+
"loss": 1.0361,
|
| 1176 |
+
"step": 1670,
|
| 1177 |
+
"train_LM": 0.51849365234375
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.72,
|
| 1181 |
+
"learning_rate": 4.704083899625568e-05,
|
| 1182 |
+
"loss": 1.0421,
|
| 1183 |
+
"step": 1680,
|
| 1184 |
+
"train_LM": 0.52880859375
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.73,
|
| 1188 |
+
"learning_rate": 4.698151323461448e-05,
|
| 1189 |
+
"loss": 1.0436,
|
| 1190 |
+
"step": 1690,
|
| 1191 |
+
"train_LM": 0.50469970703125
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.73,
|
| 1195 |
+
"learning_rate": 4.692163685273756e-05,
|
| 1196 |
+
"loss": 1.0332,
|
| 1197 |
+
"step": 1700,
|
| 1198 |
+
"train_LM": 0.51904296875
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.74,
|
| 1202 |
+
"learning_rate": 4.686121135048287e-05,
|
| 1203 |
+
"loss": 1.0395,
|
| 1204 |
+
"step": 1710,
|
| 1205 |
+
"train_LM": 0.51953125
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.74,
|
| 1209 |
+
"learning_rate": 4.6800238241463415e-05,
|
| 1210 |
+
"loss": 1.0391,
|
| 1211 |
+
"step": 1720,
|
| 1212 |
+
"train_LM": 0.5281982421875
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.74,
|
| 1216 |
+
"learning_rate": 4.673871905300935e-05,
|
| 1217 |
+
"loss": 1.0299,
|
| 1218 |
+
"step": 1730,
|
| 1219 |
+
"train_LM": 0.5076904296875
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.75,
|
| 1223 |
+
"learning_rate": 4.667665532612967e-05,
|
| 1224 |
+
"loss": 1.0351,
|
| 1225 |
+
"step": 1740,
|
| 1226 |
+
"train_LM": 0.49090576171875
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.75,
|
| 1230 |
+
"learning_rate": 4.661404861547368e-05,
|
| 1231 |
+
"loss": 1.0322,
|
| 1232 |
+
"step": 1750,
|
| 1233 |
+
"train_LM": 0.5260009765625
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.76,
|
| 1237 |
+
"learning_rate": 4.6550900489291985e-05,
|
| 1238 |
+
"loss": 1.0387,
|
| 1239 |
+
"step": 1760,
|
| 1240 |
+
"train_LM": 0.50994873046875
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.76,
|
| 1244 |
+
"learning_rate": 4.648721252939727e-05,
|
| 1245 |
+
"loss": 1.0296,
|
| 1246 |
+
"step": 1770,
|
| 1247 |
+
"train_LM": 0.50262451171875
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.77,
|
| 1251 |
+
"learning_rate": 4.642298633112462e-05,
|
| 1252 |
+
"loss": 1.0317,
|
| 1253 |
+
"step": 1780,
|
| 1254 |
+
"train_LM": 0.50982666015625
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.77,
|
| 1258 |
+
"learning_rate": 4.635822350329159e-05,
|
| 1259 |
+
"loss": 1.0305,
|
| 1260 |
+
"step": 1790,
|
| 1261 |
+
"train_LM": 0.52764892578125
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.77,
|
| 1265 |
+
"learning_rate": 4.629292566815791e-05,
|
| 1266 |
+
"loss": 1.0336,
|
| 1267 |
+
"step": 1800,
|
| 1268 |
+
"train_LM": 0.51715087890625
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.78,
|
| 1272 |
+
"learning_rate": 4.622709446138482e-05,
|
| 1273 |
+
"loss": 1.0357,
|
| 1274 |
+
"step": 1810,
|
| 1275 |
+
"train_LM": 0.4981689453125
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 0.78,
|
| 1279 |
+
"learning_rate": 4.616073153199413e-05,
|
| 1280 |
+
"loss": 1.0312,
|
| 1281 |
+
"step": 1820,
|
| 1282 |
+
"train_LM": 0.5146484375
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.79,
|
| 1286 |
+
"learning_rate": 4.609383854232689e-05,
|
| 1287 |
+
"loss": 1.0282,
|
| 1288 |
+
"step": 1830,
|
| 1289 |
+
"train_LM": 0.4881591796875
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.79,
|
| 1293 |
+
"learning_rate": 4.602641716800176e-05,
|
| 1294 |
+
"loss": 1.0285,
|
| 1295 |
+
"step": 1840,
|
| 1296 |
+
"train_LM": 0.5303955078125
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.8,
|
| 1300 |
+
"learning_rate": 4.595846909787302e-05,
|
| 1301 |
+
"loss": 1.036,
|
| 1302 |
+
"step": 1850,
|
| 1303 |
+
"train_LM": 0.5048828125
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.8,
|
| 1307 |
+
"learning_rate": 4.5889996033988314e-05,
|
| 1308 |
+
"loss": 1.0336,
|
| 1309 |
+
"step": 1860,
|
| 1310 |
+
"train_LM": 0.5347900390625
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.8,
|
| 1314 |
+
"learning_rate": 4.582099969154596e-05,
|
| 1315 |
+
"loss": 1.0266,
|
| 1316 |
+
"step": 1870,
|
| 1317 |
+
"train_LM": 0.51702880859375
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 0.81,
|
| 1321 |
+
"learning_rate": 4.5751481798852e-05,
|
| 1322 |
+
"loss": 1.0417,
|
| 1323 |
+
"step": 1880,
|
| 1324 |
+
"train_LM": 0.53302001953125
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.81,
|
| 1328 |
+
"learning_rate": 4.568144409727693e-05,
|
| 1329 |
+
"loss": 1.0201,
|
| 1330 |
+
"step": 1890,
|
| 1331 |
+
"train_LM": 0.53863525390625
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.82,
|
| 1335 |
+
"learning_rate": 4.561088834121205e-05,
|
| 1336 |
+
"loss": 1.0373,
|
| 1337 |
+
"step": 1900,
|
| 1338 |
+
"train_LM": 0.517333984375
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.82,
|
| 1342 |
+
"learning_rate": 4.553981629802555e-05,
|
| 1343 |
+
"loss": 1.0279,
|
| 1344 |
+
"step": 1910,
|
| 1345 |
+
"train_LM": 0.4990234375
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.83,
|
| 1349 |
+
"learning_rate": 4.54682297480182e-05,
|
| 1350 |
+
"loss": 1.0269,
|
| 1351 |
+
"step": 1920,
|
| 1352 |
+
"train_LM": 0.53338623046875
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.83,
|
| 1356 |
+
"learning_rate": 4.53961304843788e-05,
|
| 1357 |
+
"loss": 1.0265,
|
| 1358 |
+
"step": 1930,
|
| 1359 |
+
"train_LM": 0.49761962890625
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.83,
|
| 1363 |
+
"learning_rate": 4.532352031313922e-05,
|
| 1364 |
+
"loss": 1.023,
|
| 1365 |
+
"step": 1940,
|
| 1366 |
+
"train_LM": 0.50335693359375
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.84,
|
| 1370 |
+
"learning_rate": 4.5250401053129165e-05,
|
| 1371 |
+
"loss": 1.0333,
|
| 1372 |
+
"step": 1950,
|
| 1373 |
+
"train_LM": 0.5223388671875
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.84,
|
| 1377 |
+
"learning_rate": 4.5176774535930675e-05,
|
| 1378 |
+
"loss": 1.0349,
|
| 1379 |
+
"step": 1960,
|
| 1380 |
+
"train_LM": 0.4984130859375
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.85,
|
| 1384 |
+
"learning_rate": 4.5102642605832136e-05,
|
| 1385 |
+
"loss": 1.0307,
|
| 1386 |
+
"step": 1970,
|
| 1387 |
+
"train_LM": 0.504638671875
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.85,
|
| 1391 |
+
"learning_rate": 4.50280071197822e-05,
|
| 1392 |
+
"loss": 1.0163,
|
| 1393 |
+
"step": 1980,
|
| 1394 |
+
"train_LM": 0.5086669921875
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.86,
|
| 1398 |
+
"learning_rate": 4.495286994734317e-05,
|
| 1399 |
+
"loss": 1.0167,
|
| 1400 |
+
"step": 1990,
|
| 1401 |
+
"train_LM": 0.5035400390625
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.86,
|
| 1405 |
+
"learning_rate": 4.4877232970644245e-05,
|
| 1406 |
+
"loss": 1.0235,
|
| 1407 |
+
"step": 2000,
|
| 1408 |
+
"train_LM": 0.5247802734375
|
| 1409 |
+
}
|
| 1410 |
+
],
|
| 1411 |
+
"max_steps": 6975,
|
| 1412 |
+
"num_train_epochs": 3,
|
| 1413 |
+
"total_flos": 0.0,
|
| 1414 |
+
"trial_name": null,
|
| 1415 |
+
"trial_params": null
|
| 1416 |
+
}
|