Upload folder using huggingface_hub
Browse files- README.md +1 -36
- adapter_config.json +6 -3
- adapter_model.safetensors +1 -1
- optimizer.pt +2 -2
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +220 -688
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -199,41 +199,6 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
|
|
| 199 |
[More Information Needed]
|
| 200 |
|
| 201 |
|
| 202 |
-
## Training procedure
|
| 203 |
-
|
| 204 |
-
|
| 205 |
-
The following `bitsandbytes` quantization config was used during training:
|
| 206 |
-
- quant_method: bitsandbytes
|
| 207 |
-
- load_in_8bit: False
|
| 208 |
-
- load_in_4bit: True
|
| 209 |
-
- llm_int8_threshold: 6.0
|
| 210 |
-
- llm_int8_skip_modules: None
|
| 211 |
-
- llm_int8_enable_fp32_cpu_offload: False
|
| 212 |
-
- llm_int8_has_fp16_weight: False
|
| 213 |
-
- bnb_4bit_quant_type: nf4
|
| 214 |
-
- bnb_4bit_use_double_quant: False
|
| 215 |
-
- bnb_4bit_compute_dtype: float16
|
| 216 |
-
|
| 217 |
-
### Framework versions
|
| 218 |
-
|
| 219 |
-
|
| 220 |
-
- PEFT 0.6.3.dev0
|
| 221 |
-
## Training procedure
|
| 222 |
-
|
| 223 |
-
|
| 224 |
-
The following `bitsandbytes` quantization config was used during training:
|
| 225 |
-
- quant_method: bitsandbytes
|
| 226 |
-
- load_in_8bit: False
|
| 227 |
-
- load_in_4bit: True
|
| 228 |
-
- llm_int8_threshold: 6.0
|
| 229 |
-
- llm_int8_skip_modules: None
|
| 230 |
-
- llm_int8_enable_fp32_cpu_offload: False
|
| 231 |
-
- llm_int8_has_fp16_weight: False
|
| 232 |
-
- bnb_4bit_quant_type: nf4
|
| 233 |
-
- bnb_4bit_use_double_quant: False
|
| 234 |
-
- bnb_4bit_compute_dtype: float16
|
| 235 |
-
|
| 236 |
### Framework versions
|
| 237 |
|
| 238 |
-
|
| 239 |
-
- PEFT 0.6.3.dev0
|
|
|
|
| 199 |
[More Information Needed]
|
| 200 |
|
| 201 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
### Framework versions
|
| 203 |
|
| 204 |
+
- PEFT 0.7.1.dev0
|
|
|
adapter_config.json
CHANGED
|
@@ -8,18 +8,21 @@
|
|
| 8 |
"init_lora_weights": true,
|
| 9 |
"layers_pattern": null,
|
| 10 |
"layers_to_transform": null,
|
|
|
|
| 11 |
"lora_alpha": 16,
|
| 12 |
"lora_dropout": 0.1,
|
|
|
|
|
|
|
| 13 |
"modules_to_save": null,
|
| 14 |
"peft_type": "LORA",
|
| 15 |
"r": 64,
|
| 16 |
"rank_pattern": {},
|
| 17 |
"revision": null,
|
| 18 |
"target_modules": [
|
| 19 |
-
"dense_4h_to_h",
|
| 20 |
-
"query_key_value",
|
| 21 |
"dense_h_to_4h",
|
| 22 |
-
"
|
|
|
|
|
|
|
| 23 |
],
|
| 24 |
"task_type": "CAUSAL_LM"
|
| 25 |
}
|
|
|
|
| 8 |
"init_lora_weights": true,
|
| 9 |
"layers_pattern": null,
|
| 10 |
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
"lora_alpha": 16,
|
| 13 |
"lora_dropout": 0.1,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
"modules_to_save": null,
|
| 17 |
"peft_type": "LORA",
|
| 18 |
"r": 64,
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
|
|
|
|
|
|
| 22 |
"dense_h_to_4h",
|
| 23 |
+
"query_key_value",
|
| 24 |
+
"dense",
|
| 25 |
+
"dense_4h_to_h"
|
| 26 |
],
|
| 27 |
"task_type": "CAUSAL_LM"
|
| 28 |
}
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 134235712
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64f47eb614eeeac02198516c8e09fcfbc5db7f3c337305b2b9461ad77a8adf1f
|
| 3 |
size 134235712
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58c667c86b48949dd4dfb32952fa167e0583f8e5686af20a6e7232dd094f3d48
|
| 3 |
+
size 268514874
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:337e893fff51aad090563eae7b251d31e2705a96eb109dfc6fdadfb33b6a6240
|
| 3 |
size 14244
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68ca608482c17f9314e1c94cd309a18be088851d4c0591a9306e6a01c952c9f3
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,919 +1,451 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 2.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 2
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 4
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 0.
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 6
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 0.
|
| 32 |
-
"loss": 2.
|
| 33 |
"step": 8
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 0.
|
| 38 |
-
"loss": 2.
|
| 39 |
"step": 10
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate": 0.
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 12
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 14
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate": 0.
|
| 56 |
-
"loss": 2.
|
| 57 |
"step": 16
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"learning_rate": 0.
|
| 62 |
-
"loss": 2.
|
| 63 |
"step": 18
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
-
"learning_rate": 0.
|
| 68 |
-
"loss":
|
| 69 |
"step": 20
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"learning_rate": 0.
|
| 74 |
-
"loss": 2.
|
| 75 |
"step": 22
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 0.
|
| 79 |
-
"learning_rate": 0.
|
| 80 |
-
"loss": 2.
|
| 81 |
"step": 24
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
-
"learning_rate": 0.
|
| 86 |
-
"loss": 2.
|
| 87 |
"step": 26
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss": 2.
|
| 93 |
"step": 28
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"learning_rate": 0.
|
| 98 |
-
"loss": 2.
|
| 99 |
"step": 30
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
-
"learning_rate": 0.
|
| 104 |
-
"loss": 2.
|
| 105 |
"step": 32
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
-
"learning_rate": 0.
|
| 110 |
-
"loss": 2.
|
| 111 |
"step": 34
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch": 0.
|
| 115 |
-
"learning_rate": 0.
|
| 116 |
-
"loss": 2.
|
| 117 |
"step": 36
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 0.
|
| 121 |
-
"learning_rate": 0.
|
| 122 |
-
"loss": 2.
|
| 123 |
"step": 38
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
-
"learning_rate": 0.
|
| 128 |
-
"loss": 2.
|
| 129 |
"step": 40
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 2.
|
| 135 |
"step": 42
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"learning_rate": 0.
|
| 140 |
-
"loss": 2.
|
| 141 |
"step": 44
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch": 0.
|
| 145 |
-
"learning_rate": 0.
|
| 146 |
-
"loss": 2.
|
| 147 |
"step": 46
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
-
"learning_rate": 0.
|
| 152 |
-
"loss": 2.
|
| 153 |
"step": 48
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch":
|
| 157 |
-
"learning_rate": 0.
|
| 158 |
-
"loss": 2.
|
| 159 |
"step": 50
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch":
|
| 163 |
-
"learning_rate": 0.
|
| 164 |
-
"loss": 2.
|
| 165 |
"step": 52
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch":
|
| 169 |
-
"learning_rate": 0.
|
| 170 |
-
"loss": 2.
|
| 171 |
"step": 54
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch":
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss": 2.
|
| 177 |
"step": 56
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"learning_rate": 0.
|
| 182 |
-
"loss": 2.
|
| 183 |
"step": 58
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate": 0.
|
| 188 |
-
"loss": 2.
|
| 189 |
"step": 60
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 2.
|
| 195 |
"step": 62
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"learning_rate": 0.
|
| 200 |
-
"loss": 2.
|
| 201 |
"step": 64
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"learning_rate": 0.
|
| 206 |
-
"loss": 2.
|
| 207 |
"step": 66
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch":
|
| 211 |
-
"learning_rate": 0.
|
| 212 |
-
"loss": 2.
|
| 213 |
"step": 68
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch":
|
| 217 |
-
"learning_rate": 0.
|
| 218 |
-
"loss": 2.
|
| 219 |
"step": 70
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"learning_rate": 0.
|
| 224 |
-
"loss": 2.
|
| 225 |
"step": 72
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch":
|
| 229 |
-
"learning_rate": 0.
|
| 230 |
-
"loss": 2.
|
| 231 |
"step": 74
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch":
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss": 2.
|
| 237 |
"step": 76
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch":
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss": 2.
|
| 243 |
"step": 78
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch":
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss": 2.
|
| 249 |
"step": 80
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"learning_rate":
|
| 254 |
-
"loss": 2.
|
| 255 |
"step": 82
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss": 2.
|
| 261 |
"step": 84
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch":
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 2.
|
| 267 |
"step": 86
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss": 2.
|
| 273 |
"step": 88
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"learning_rate":
|
| 278 |
-
"loss": 2.
|
| 279 |
"step": 90
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch":
|
| 283 |
-
"learning_rate":
|
| 284 |
-
"loss": 2.
|
| 285 |
"step": 92
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss": 2.
|
| 291 |
"step": 94
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch":
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss": 2.
|
| 297 |
"step": 96
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss": 2.
|
| 303 |
"step": 98
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch":
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss":
|
| 309 |
"step": 100
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch":
|
| 313 |
-
"learning_rate":
|
| 314 |
-
"loss": 2.
|
| 315 |
"step": 102
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch":
|
| 319 |
-
"learning_rate":
|
| 320 |
-
"loss": 2.
|
| 321 |
"step": 104
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
-
"learning_rate":
|
| 326 |
-
"loss": 2.
|
| 327 |
"step": 106
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss": 2.
|
| 333 |
"step": 108
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss":
|
| 339 |
"step": 110
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 2.
|
| 345 |
"step": 112
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"learning_rate":
|
| 350 |
-
"loss": 2.
|
| 351 |
"step": 114
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch":
|
| 355 |
-
"learning_rate":
|
| 356 |
-
"loss":
|
| 357 |
"step": 116
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
-
"learning_rate":
|
| 362 |
-
"loss": 2.
|
| 363 |
"step": 118
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch":
|
| 367 |
-
"learning_rate":
|
| 368 |
-
"loss":
|
| 369 |
"step": 120
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
-
"learning_rate":
|
| 374 |
-
"loss":
|
| 375 |
"step": 122
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch":
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 2.
|
| 381 |
"step": 124
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss": 2.
|
| 387 |
"step": 126
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch":
|
| 391 |
-
"learning_rate":
|
| 392 |
-
"loss": 2.
|
| 393 |
"step": 128
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch":
|
| 397 |
-
"learning_rate":
|
| 398 |
-
"loss": 2.
|
| 399 |
"step": 130
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch":
|
| 403 |
-
"learning_rate":
|
| 404 |
-
"loss": 2.
|
| 405 |
"step": 132
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch":
|
| 409 |
-
"learning_rate":
|
| 410 |
-
"loss": 2.
|
| 411 |
"step": 134
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch":
|
| 415 |
-
"learning_rate":
|
| 416 |
-
"loss": 2.
|
| 417 |
"step": 136
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"learning_rate":
|
| 422 |
-
"loss": 2.
|
| 423 |
"step": 138
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
-
"learning_rate":
|
| 428 |
-
"loss": 2.
|
| 429 |
"step": 140
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch":
|
| 433 |
-
"learning_rate":
|
| 434 |
-
"loss": 2.
|
| 435 |
"step": 142
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
-
"learning_rate": 0.00011131138261952845,
|
| 440 |
-
"loss": 1.9611,
|
| 441 |
-
"step": 144
|
| 442 |
-
},
|
| 443 |
-
{
|
| 444 |
-
"epoch": 1.45,
|
| 445 |
-
"learning_rate": 0.0001091635999850655,
|
| 446 |
-
"loss": 1.9342,
|
| 447 |
-
"step": 146
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 1.47,
|
| 451 |
-
"learning_rate": 0.00010701154543160541,
|
| 452 |
-
"loss": 1.9842,
|
| 453 |
-
"step": 148
|
| 454 |
-
},
|
| 455 |
-
{
|
| 456 |
-
"epoch": 1.49,
|
| 457 |
-
"learning_rate": 0.00010485622221144484,
|
| 458 |
-
"loss": 2.0796,
|
| 459 |
-
"step": 150
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 1.51,
|
| 463 |
-
"learning_rate": 0.00010269863510067872,
|
| 464 |
-
"loss": 1.9456,
|
| 465 |
-
"step": 152
|
| 466 |
-
},
|
| 467 |
-
{
|
| 468 |
-
"epoch": 1.53,
|
| 469 |
-
"learning_rate": 0.00010053978993079045,
|
| 470 |
-
"loss": 2.2969,
|
| 471 |
-
"step": 154
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 1.55,
|
| 475 |
-
"learning_rate": 9.838069311974986e-05,
|
| 476 |
-
"loss": 2.1745,
|
| 477 |
-
"step": 156
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 1.57,
|
| 481 |
-
"learning_rate": 9.622235120283769e-05,
|
| 482 |
-
"loss": 2.0905,
|
| 483 |
-
"step": 158
|
| 484 |
-
},
|
| 485 |
-
{
|
| 486 |
-
"epoch": 1.59,
|
| 487 |
-
"learning_rate": 9.406577036341548e-05,
|
| 488 |
-
"loss": 2.1926,
|
| 489 |
-
"step": 160
|
| 490 |
-
},
|
| 491 |
-
{
|
| 492 |
-
"epoch": 1.61,
|
| 493 |
-
"learning_rate": 9.19119559638596e-05,
|
| 494 |
-
"loss": 2.1519,
|
| 495 |
-
"step": 162
|
| 496 |
-
},
|
| 497 |
-
{
|
| 498 |
-
"epoch": 1.63,
|
| 499 |
-
"learning_rate": 8.976191207687775e-05,
|
| 500 |
-
"loss": 2.0953,
|
| 501 |
-
"step": 164
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 1.65,
|
| 505 |
-
"learning_rate": 8.7616641017427e-05,
|
| 506 |
-
"loss": 2.0602,
|
| 507 |
-
"step": 166
|
| 508 |
-
},
|
| 509 |
-
{
|
| 510 |
-
"epoch": 1.67,
|
| 511 |
-
"learning_rate": 8.5477142875451e-05,
|
| 512 |
-
"loss": 2.1189,
|
| 513 |
-
"step": 168
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 1.69,
|
| 517 |
-
"learning_rate": 8.334441504965455e-05,
|
| 518 |
-
"loss": 1.8349,
|
| 519 |
-
"step": 170
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 1.71,
|
| 523 |
-
"learning_rate": 8.1219451782533e-05,
|
| 524 |
-
"loss": 2.0207,
|
| 525 |
-
"step": 172
|
| 526 |
-
},
|
| 527 |
-
{
|
| 528 |
-
"epoch": 1.73,
|
| 529 |
-
"learning_rate": 7.91032436968725e-05,
|
| 530 |
-
"loss": 2.0835,
|
| 531 |
-
"step": 174
|
| 532 |
-
},
|
| 533 |
-
{
|
| 534 |
-
"epoch": 1.75,
|
| 535 |
-
"learning_rate": 7.699677733393826e-05,
|
| 536 |
-
"loss": 2.1297,
|
| 537 |
-
"step": 176
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"epoch": 1.77,
|
| 541 |
-
"learning_rate": 7.490103469356513e-05,
|
| 542 |
-
"loss": 2.2253,
|
| 543 |
-
"step": 178
|
| 544 |
-
},
|
| 545 |
-
{
|
| 546 |
-
"epoch": 1.79,
|
| 547 |
-
"learning_rate": 7.281699277636572e-05,
|
| 548 |
-
"loss": 2.1392,
|
| 549 |
-
"step": 180
|
| 550 |
-
},
|
| 551 |
-
{
|
| 552 |
-
"epoch": 1.81,
|
| 553 |
-
"learning_rate": 7.07456231282686e-05,
|
| 554 |
-
"loss": 2.2366,
|
| 555 |
-
"step": 182
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 1.83,
|
| 559 |
-
"learning_rate": 6.868789138759976e-05,
|
| 560 |
-
"loss": 2.2417,
|
| 561 |
-
"step": 184
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 1.85,
|
| 565 |
-
"learning_rate": 6.664475683491796e-05,
|
| 566 |
-
"loss": 2.1114,
|
| 567 |
-
"step": 186
|
| 568 |
-
},
|
| 569 |
-
{
|
| 570 |
-
"epoch": 1.87,
|
| 571 |
-
"learning_rate": 6.461717194581393e-05,
|
| 572 |
-
"loss": 2.1858,
|
| 573 |
-
"step": 188
|
| 574 |
-
},
|
| 575 |
-
{
|
| 576 |
-
"epoch": 1.89,
|
| 577 |
-
"learning_rate": 6.260608194688206e-05,
|
| 578 |
-
"loss": 2.2032,
|
| 579 |
-
"step": 190
|
| 580 |
-
},
|
| 581 |
-
{
|
| 582 |
-
"epoch": 1.91,
|
| 583 |
-
"learning_rate": 6.061242437507131e-05,
|
| 584 |
-
"loss": 2.1147,
|
| 585 |
-
"step": 192
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"epoch": 1.93,
|
| 589 |
-
"learning_rate": 5.863712864062089e-05,
|
| 590 |
-
"loss": 1.9731,
|
| 591 |
-
"step": 194
|
| 592 |
-
},
|
| 593 |
-
{
|
| 594 |
-
"epoch": 1.95,
|
| 595 |
-
"learning_rate": 5.668111559378471e-05,
|
| 596 |
-
"loss": 2.0307,
|
| 597 |
-
"step": 196
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 1.97,
|
| 601 |
-
"learning_rate": 5.474529709554612e-05,
|
| 602 |
-
"loss": 2.0143,
|
| 603 |
-
"step": 198
|
| 604 |
-
},
|
| 605 |
-
{
|
| 606 |
-
"epoch": 1.99,
|
| 607 |
-
"learning_rate": 5.283057559252341e-05,
|
| 608 |
-
"loss": 1.8633,
|
| 609 |
-
"step": 200
|
| 610 |
-
},
|
| 611 |
-
{
|
| 612 |
-
"epoch": 2.01,
|
| 613 |
-
"learning_rate": 5.0937843696263966e-05,
|
| 614 |
-
"loss": 1.8288,
|
| 615 |
-
"step": 202
|
| 616 |
-
},
|
| 617 |
-
{
|
| 618 |
-
"epoch": 2.03,
|
| 619 |
-
"learning_rate": 4.9067983767123736e-05,
|
| 620 |
-
"loss": 2.2213,
|
| 621 |
-
"step": 204
|
| 622 |
-
},
|
| 623 |
-
{
|
| 624 |
-
"epoch": 2.05,
|
| 625 |
-
"learning_rate": 4.722186750292511e-05,
|
| 626 |
-
"loss": 2.1991,
|
| 627 |
-
"step": 206
|
| 628 |
-
},
|
| 629 |
-
{
|
| 630 |
-
"epoch": 2.07,
|
| 631 |
-
"learning_rate": 4.540035553258619e-05,
|
| 632 |
-
"loss": 1.986,
|
| 633 |
-
"step": 208
|
| 634 |
-
},
|
| 635 |
-
{
|
| 636 |
-
"epoch": 2.09,
|
| 637 |
-
"learning_rate": 4.360429701490934e-05,
|
| 638 |
-
"loss": 2.1149,
|
| 639 |
-
"step": 210
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 2.11,
|
| 643 |
-
"learning_rate": 4.183452924271776e-05,
|
| 644 |
-
"loss": 2.0299,
|
| 645 |
-
"step": 212
|
| 646 |
-
},
|
| 647 |
-
{
|
| 648 |
-
"epoch": 2.13,
|
| 649 |
-
"learning_rate": 4.009187725252309e-05,
|
| 650 |
-
"loss": 2.182,
|
| 651 |
-
"step": 214
|
| 652 |
-
},
|
| 653 |
-
{
|
| 654 |
-
"epoch": 2.15,
|
| 655 |
-
"learning_rate": 3.8377153439907266e-05,
|
| 656 |
-
"loss": 2.125,
|
| 657 |
-
"step": 216
|
| 658 |
-
},
|
| 659 |
-
{
|
| 660 |
-
"epoch": 2.17,
|
| 661 |
-
"learning_rate": 3.669115718079702e-05,
|
| 662 |
-
"loss": 2.0164,
|
| 663 |
-
"step": 218
|
| 664 |
-
},
|
| 665 |
-
{
|
| 666 |
-
"epoch": 2.19,
|
| 667 |
-
"learning_rate": 3.503467445880789e-05,
|
| 668 |
-
"loss": 1.9398,
|
| 669 |
-
"step": 220
|
| 670 |
-
},
|
| 671 |
-
{
|
| 672 |
-
"epoch": 2.21,
|
| 673 |
-
"learning_rate": 3.340847749883191e-05,
|
| 674 |
-
"loss": 1.9637,
|
| 675 |
-
"step": 222
|
| 676 |
-
},
|
| 677 |
-
{
|
| 678 |
-
"epoch": 2.23,
|
| 679 |
-
"learning_rate": 3.1813324407038825e-05,
|
| 680 |
-
"loss": 1.99,
|
| 681 |
-
"step": 224
|
| 682 |
-
},
|
| 683 |
-
{
|
| 684 |
-
"epoch": 2.25,
|
| 685 |
-
"learning_rate": 3.0249958817459722e-05,
|
| 686 |
-
"loss": 1.9515,
|
| 687 |
-
"step": 226
|
| 688 |
-
},
|
| 689 |
-
{
|
| 690 |
-
"epoch": 2.27,
|
| 691 |
-
"learning_rate": 2.8719109545317103e-05,
|
| 692 |
-
"loss": 2.0532,
|
| 693 |
-
"step": 228
|
| 694 |
-
},
|
| 695 |
-
{
|
| 696 |
-
"epoch": 2.29,
|
| 697 |
-
"learning_rate": 2.722149024726307e-05,
|
| 698 |
-
"loss": 2.1069,
|
| 699 |
-
"step": 230
|
| 700 |
-
},
|
| 701 |
-
{
|
| 702 |
-
"epoch": 2.31,
|
| 703 |
-
"learning_rate": 2.5757799088684654e-05,
|
| 704 |
-
"loss": 2.0917,
|
| 705 |
-
"step": 232
|
| 706 |
-
},
|
| 707 |
-
{
|
| 708 |
-
"epoch": 2.33,
|
| 709 |
-
"learning_rate": 2.432871841823047e-05,
|
| 710 |
-
"loss": 2.2022,
|
| 711 |
-
"step": 234
|
| 712 |
-
},
|
| 713 |
-
{
|
| 714 |
-
"epoch": 2.35,
|
| 715 |
-
"learning_rate": 2.2934914449711087e-05,
|
| 716 |
-
"loss": 2.1084,
|
| 717 |
-
"step": 236
|
| 718 |
-
},
|
| 719 |
-
{
|
| 720 |
-
"epoch": 2.37,
|
| 721 |
-
"learning_rate": 2.157703695152109e-05,
|
| 722 |
-
"loss": 2.1929,
|
| 723 |
-
"step": 238
|
| 724 |
-
},
|
| 725 |
-
{
|
| 726 |
-
"epoch": 2.39,
|
| 727 |
-
"learning_rate": 2.025571894372794e-05,
|
| 728 |
-
"loss": 2.0758,
|
| 729 |
-
"step": 240
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"epoch": 2.41,
|
| 733 |
-
"learning_rate": 1.897157640296825e-05,
|
| 734 |
-
"loss": 2.0788,
|
| 735 |
-
"step": 242
|
| 736 |
-
},
|
| 737 |
-
{
|
| 738 |
-
"epoch": 2.43,
|
| 739 |
-
"learning_rate": 1.772520797528988e-05,
|
| 740 |
-
"loss": 1.966,
|
| 741 |
-
"step": 244
|
| 742 |
-
},
|
| 743 |
-
{
|
| 744 |
-
"epoch": 2.45,
|
| 745 |
-
"learning_rate": 1.65171946970729e-05,
|
| 746 |
-
"loss": 2.0296,
|
| 747 |
-
"step": 246
|
| 748 |
-
},
|
| 749 |
-
{
|
| 750 |
-
"epoch": 2.47,
|
| 751 |
-
"learning_rate": 1.534809972415998e-05,
|
| 752 |
-
"loss": 1.967,
|
| 753 |
-
"step": 248
|
| 754 |
-
},
|
| 755 |
-
{
|
| 756 |
-
"epoch": 2.49,
|
| 757 |
-
"learning_rate": 1.4218468069322578e-05,
|
| 758 |
-
"loss": 2.0128,
|
| 759 |
-
"step": 250
|
| 760 |
-
},
|
| 761 |
-
{
|
| 762 |
-
"epoch": 2.51,
|
| 763 |
-
"learning_rate": 1.3128826348184887e-05,
|
| 764 |
-
"loss": 1.8105,
|
| 765 |
-
"step": 252
|
| 766 |
-
},
|
| 767 |
-
{
|
| 768 |
-
"epoch": 2.53,
|
| 769 |
-
"learning_rate": 1.2079682533724379e-05,
|
| 770 |
-
"loss": 2.1251,
|
| 771 |
-
"step": 254
|
| 772 |
-
},
|
| 773 |
-
{
|
| 774 |
-
"epoch": 2.55,
|
| 775 |
-
"learning_rate": 1.1071525719463095e-05,
|
| 776 |
-
"loss": 2.2662,
|
| 777 |
-
"step": 256
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"epoch": 2.57,
|
| 781 |
-
"learning_rate": 1.010482589146048e-05,
|
| 782 |
-
"loss": 2.172,
|
| 783 |
-
"step": 258
|
| 784 |
-
},
|
| 785 |
-
{
|
| 786 |
-
"epoch": 2.59,
|
| 787 |
-
"learning_rate": 9.180033709213454e-06,
|
| 788 |
-
"loss": 2.1301,
|
| 789 |
-
"step": 260
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 2.61,
|
| 793 |
-
"learning_rate": 8.297580295566575e-06,
|
| 794 |
-
"loss": 2.0283,
|
| 795 |
-
"step": 262
|
| 796 |
-
},
|
| 797 |
-
{
|
| 798 |
-
"epoch": 2.63,
|
| 799 |
-
"learning_rate": 7.457877035729588e-06,
|
| 800 |
-
"loss": 2.092,
|
| 801 |
-
"step": 264
|
| 802 |
-
},
|
| 803 |
-
{
|
| 804 |
-
"epoch": 2.65,
|
| 805 |
-
"learning_rate": 6.661315385496425e-06,
|
| 806 |
-
"loss": 2.1478,
|
| 807 |
-
"step": 266
|
| 808 |
-
},
|
| 809 |
-
{
|
| 810 |
-
"epoch": 2.67,
|
| 811 |
-
"learning_rate": 5.908266688755049e-06,
|
| 812 |
-
"loss": 2.0034,
|
| 813 |
-
"step": 268
|
| 814 |
-
},
|
| 815 |
-
{
|
| 816 |
-
"epoch": 2.68,
|
| 817 |
-
"learning_rate": 5.199082004372957e-06,
|
| 818 |
-
"loss": 2.0571,
|
| 819 |
-
"step": 270
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 2.7,
|
| 823 |
-
"learning_rate": 4.534091942539475e-06,
|
| 824 |
-
"loss": 1.8802,
|
| 825 |
-
"step": 272
|
| 826 |
-
},
|
| 827 |
-
{
|
| 828 |
-
"epoch": 2.72,
|
| 829 |
-
"learning_rate": 3.913606510640644e-06,
|
| 830 |
-
"loss": 2.1067,
|
| 831 |
-
"step": 274
|
| 832 |
-
},
|
| 833 |
-
{
|
| 834 |
-
"epoch": 2.74,
|
| 835 |
-
"learning_rate": 3.3379149687388867e-06,
|
| 836 |
-
"loss": 1.9134,
|
| 837 |
-
"step": 276
|
| 838 |
-
},
|
| 839 |
-
{
|
| 840 |
-
"epoch": 2.76,
|
| 841 |
-
"learning_rate": 2.8072856947248037e-06,
|
| 842 |
-
"loss": 2.1052,
|
| 843 |
-
"step": 278
|
| 844 |
-
},
|
| 845 |
-
{
|
| 846 |
-
"epoch": 2.78,
|
| 847 |
-
"learning_rate": 2.3219660592038285e-06,
|
| 848 |
-
"loss": 1.9601,
|
| 849 |
-
"step": 280
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 2.8,
|
| 853 |
-
"learning_rate": 1.882182310176095e-06,
|
| 854 |
-
"loss": 2.0907,
|
| 855 |
-
"step": 282
|
| 856 |
-
},
|
| 857 |
-
{
|
| 858 |
-
"epoch": 2.82,
|
| 859 |
-
"learning_rate": 1.488139467563354e-06,
|
| 860 |
-
"loss": 2.0213,
|
| 861 |
-
"step": 284
|
| 862 |
-
},
|
| 863 |
-
{
|
| 864 |
-
"epoch": 2.84,
|
| 865 |
-
"learning_rate": 1.1400212276321376e-06,
|
| 866 |
-
"loss": 2.2126,
|
| 867 |
-
"step": 286
|
| 868 |
-
},
|
| 869 |
-
{
|
| 870 |
-
"epoch": 2.86,
|
| 871 |
-
"learning_rate": 8.379898773574924e-07,
|
| 872 |
-
"loss": 2.2776,
|
| 873 |
-
"step": 288
|
| 874 |
-
},
|
| 875 |
-
{
|
| 876 |
-
"epoch": 2.88,
|
| 877 |
-
"learning_rate": 5.821862187675775e-07,
|
| 878 |
-
"loss": 2.0965,
|
| 879 |
-
"step": 290
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 2.9,
|
| 883 |
-
"learning_rate": 3.727295033040035e-07,
|
| 884 |
-
"loss": 2.0892,
|
| 885 |
-
"step": 292
|
| 886 |
-
},
|
| 887 |
-
{
|
| 888 |
-
"epoch": 2.92,
|
| 889 |
-
"learning_rate": 2.0971737622883515e-07,
|
| 890 |
-
"loss": 1.9065,
|
| 891 |
-
"step": 294
|
| 892 |
-
},
|
| 893 |
-
{
|
| 894 |
-
"epoch": 2.94,
|
| 895 |
-
"learning_rate": 9.32258311039269e-08,
|
| 896 |
-
"loss": 1.9225,
|
| 897 |
-
"step": 296
|
| 898 |
-
},
|
| 899 |
-
{
|
| 900 |
-
"epoch": 2.96,
|
| 901 |
-
"learning_rate": 2.3309174364027907e-08,
|
| 902 |
-
"loss": 1.9352,
|
| 903 |
-
"step": 298
|
| 904 |
-
},
|
| 905 |
-
{
|
| 906 |
-
"epoch": 2.98,
|
| 907 |
"learning_rate": 0.0,
|
| 908 |
-
"loss":
|
| 909 |
-
"step":
|
| 910 |
}
|
| 911 |
],
|
| 912 |
"logging_steps": 2,
|
| 913 |
-
"max_steps":
|
| 914 |
"num_train_epochs": 3,
|
| 915 |
"save_steps": 500,
|
| 916 |
-
"total_flos":
|
| 917 |
"trial_name": null,
|
| 918 |
"trial_params": null
|
| 919 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.953846153846154,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 144,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 8e-05,
|
| 14 |
+
"loss": 2.331,
|
| 15 |
"step": 2
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 0.00016,
|
| 20 |
+
"loss": 2.4252,
|
| 21 |
"step": 4
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 0.0001999744599547812,
|
| 26 |
+
"loss": 2.5627,
|
| 27 |
"step": 6
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.16,
|
| 31 |
+
"learning_rate": 0.00019977021786163598,
|
| 32 |
+
"loss": 2.47,
|
| 33 |
"step": 8
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.21,
|
| 37 |
+
"learning_rate": 0.00019936215093023884,
|
| 38 |
+
"loss": 2.4888,
|
| 39 |
"step": 10
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.25,
|
| 43 |
+
"learning_rate": 0.00019875109281794825,
|
| 44 |
+
"loss": 2.6307,
|
| 45 |
"step": 12
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.29,
|
| 49 |
+
"learning_rate": 0.00019793829188147406,
|
| 50 |
+
"loss": 2.2032,
|
| 51 |
"step": 14
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.33,
|
| 55 |
+
"learning_rate": 0.00019692540862655585,
|
| 56 |
+
"loss": 2.2254,
|
| 57 |
"step": 16
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.37,
|
| 61 |
+
"learning_rate": 0.00019571451231564525,
|
| 62 |
+
"loss": 2.3134,
|
| 63 |
"step": 18
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.41,
|
| 67 |
+
"learning_rate": 0.00019430807674052092,
|
| 68 |
+
"loss": 2.4247,
|
| 69 |
"step": 20
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.45,
|
| 73 |
+
"learning_rate": 0.00019270897516847403,
|
| 74 |
+
"loss": 2.4047,
|
| 75 |
"step": 22
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 0.49,
|
| 79 |
+
"learning_rate": 0.00019092047447238773,
|
| 80 |
+
"loss": 2.1955,
|
| 81 |
"step": 24
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 0.53,
|
| 85 |
+
"learning_rate": 0.00018894622845670283,
|
| 86 |
+
"loss": 2.1532,
|
| 87 |
"step": 26
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.57,
|
| 91 |
+
"learning_rate": 0.00018679027039290497,
|
| 92 |
+
"loss": 2.22,
|
| 93 |
"step": 28
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.62,
|
| 97 |
+
"learning_rate": 0.00018445700477978205,
|
| 98 |
+
"loss": 2.3129,
|
| 99 |
"step": 30
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 0.66,
|
| 103 |
+
"learning_rate": 0.00018195119834528534,
|
| 104 |
+
"loss": 2.319,
|
| 105 |
"step": 32
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 0.7,
|
| 109 |
+
"learning_rate": 0.00017927797030837768,
|
| 110 |
+
"loss": 2.3718,
|
| 111 |
"step": 34
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 0.74,
|
| 115 |
+
"learning_rate": 0.0001764427819207624,
|
| 116 |
+
"loss": 2.2181,
|
| 117 |
"step": 36
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 0.78,
|
| 121 |
+
"learning_rate": 0.00017345142530985887,
|
| 122 |
+
"loss": 2.2146,
|
| 123 |
"step": 38
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.82,
|
| 127 |
+
"learning_rate": 0.00017031001164581828,
|
| 128 |
+
"loss": 2.2694,
|
| 129 |
"step": 40
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.86,
|
| 133 |
+
"learning_rate": 0.0001670249586567531,
|
| 134 |
+
"loss": 2.475,
|
| 135 |
"step": 42
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.9,
|
| 139 |
+
"learning_rate": 0.0001636029775176862,
|
| 140 |
+
"loss": 2.4209,
|
| 141 |
"step": 44
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 0.94,
|
| 145 |
+
"learning_rate": 0.00016005105914000507,
|
| 146 |
+
"loss": 2.227,
|
| 147 |
"step": 46
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.98,
|
| 151 |
+
"learning_rate": 0.0001563764598894301,
|
| 152 |
+
"loss": 2.3764,
|
| 153 |
"step": 48
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 1.03,
|
| 157 |
+
"learning_rate": 0.00015258668676167546,
|
| 158 |
+
"loss": 2.3207,
|
| 159 |
"step": 50
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 1.07,
|
| 163 |
+
"learning_rate": 0.000148689482046087,
|
| 164 |
+
"loss": 2.1737,
|
| 165 |
"step": 52
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 1.11,
|
| 169 |
+
"learning_rate": 0.00014469280750858854,
|
| 170 |
+
"loss": 2.3004,
|
| 171 |
"step": 54
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 1.15,
|
| 175 |
+
"learning_rate": 0.00014060482812625055,
|
| 176 |
+
"loss": 2.3923,
|
| 177 |
"step": 56
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 1.19,
|
| 181 |
+
"learning_rate": 0.00013643389540670962,
|
| 182 |
+
"loss": 2.3233,
|
| 183 |
"step": 58
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 1.23,
|
| 187 |
+
"learning_rate": 0.0001321885303265172,
|
| 188 |
+
"loss": 2.2213,
|
| 189 |
"step": 60
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 1.27,
|
| 193 |
+
"learning_rate": 0.0001278774059232723,
|
| 194 |
+
"loss": 2.1091,
|
| 195 |
"step": 62
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 1.31,
|
| 199 |
+
"learning_rate": 0.0001235093295771032,
|
| 200 |
+
"loss": 2.1015,
|
| 201 |
"step": 64
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 1.35,
|
| 205 |
+
"learning_rate": 0.00011909322501769406,
|
| 206 |
+
"loss": 2.2032,
|
| 207 |
"step": 66
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 1.39,
|
| 211 |
+
"learning_rate": 0.00011463811409361667,
|
| 212 |
+
"loss": 2.3454,
|
| 213 |
"step": 68
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 1.44,
|
| 217 |
+
"learning_rate": 0.00011015309834121081,
|
| 218 |
+
"loss": 2.2736,
|
| 219 |
"step": 70
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 1.48,
|
| 223 |
+
"learning_rate": 0.00010564734039066699,
|
| 224 |
+
"loss": 2.308,
|
| 225 |
"step": 72
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 1.52,
|
| 229 |
+
"learning_rate": 0.00010113004524729799,
|
| 230 |
+
"loss": 2.1137,
|
| 231 |
"step": 74
|
| 232 |
},
|
| 233 |
{
|
| 234 |
+
"epoch": 1.56,
|
| 235 |
+
"learning_rate": 9.661044148624037e-05,
|
| 236 |
+
"loss": 2.1818,
|
| 237 |
"step": 76
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 1.6,
|
| 241 |
+
"learning_rate": 9.209776239900453e-05,
|
| 242 |
+
"loss": 2.3509,
|
| 243 |
"step": 78
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 1.64,
|
| 247 |
+
"learning_rate": 8.760122713038881e-05,
|
| 248 |
+
"loss": 2.3443,
|
| 249 |
"step": 80
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 1.68,
|
| 253 |
+
"learning_rate": 8.313002184429529e-05,
|
| 254 |
+
"loss": 2.401,
|
| 255 |
"step": 82
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 1.72,
|
| 259 |
+
"learning_rate": 7.869328095692312e-05,
|
| 260 |
+
"loss": 2.3839,
|
| 261 |
"step": 84
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 1.76,
|
| 265 |
+
"learning_rate": 7.430006847567972e-05,
|
| 266 |
+
"loss": 2.0637,
|
| 267 |
"step": 86
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 1.81,
|
| 271 |
+
"learning_rate": 6.995935948193294e-05,
|
| 272 |
+
"loss": 2.2678,
|
| 273 |
"step": 88
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"epoch": 1.85,
|
| 277 |
+
"learning_rate": 6.568002179543409e-05,
|
| 278 |
+
"loss": 2.1899,
|
| 279 |
"step": 90
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 1.89,
|
| 283 |
+
"learning_rate": 6.147079785787038e-05,
|
| 284 |
+
"loss": 2.3489,
|
| 285 |
"step": 92
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 1.93,
|
| 289 |
+
"learning_rate": 5.734028687255751e-05,
|
| 290 |
+
"loss": 2.2695,
|
| 291 |
"step": 94
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 1.97,
|
| 295 |
+
"learning_rate": 5.329692723675994e-05,
|
| 296 |
+
"loss": 2.046,
|
| 297 |
"step": 96
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 2.01,
|
| 301 |
+
"learning_rate": 4.934897930252886e-05,
|
| 302 |
+
"loss": 2.0616,
|
| 303 |
"step": 98
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 2.05,
|
| 307 |
+
"learning_rate": 4.550450850127625e-05,
|
| 308 |
+
"loss": 2.2179,
|
| 309 |
"step": 100
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 2.09,
|
| 313 |
+
"learning_rate": 4.1771368866560665e-05,
|
| 314 |
+
"loss": 2.2652,
|
| 315 |
"step": 102
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 2.13,
|
| 319 |
+
"learning_rate": 3.815718698874672e-05,
|
| 320 |
+
"loss": 2.444,
|
| 321 |
"step": 104
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 2.17,
|
| 325 |
+
"learning_rate": 3.466934643431795e-05,
|
| 326 |
+
"loss": 2.2312,
|
| 327 |
"step": 106
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 2.22,
|
| 331 |
+
"learning_rate": 3.131497266167357e-05,
|
| 332 |
+
"loss": 2.2545,
|
| 333 |
"step": 108
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 2.26,
|
| 337 |
+
"learning_rate": 2.81009184642253e-05,
|
| 338 |
+
"loss": 1.9974,
|
| 339 |
"step": 110
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 2.3,
|
| 343 |
+
"learning_rate": 2.5033749970533015e-05,
|
| 344 |
+
"loss": 2.1236,
|
| 345 |
"step": 112
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 2.34,
|
| 349 |
+
"learning_rate": 2.2119733230080408e-05,
|
| 350 |
+
"loss": 2.0792,
|
| 351 |
"step": 114
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 2.38,
|
| 355 |
+
"learning_rate": 1.9364821412094857e-05,
|
| 356 |
+
"loss": 2.4105,
|
| 357 |
"step": 116
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 2.42,
|
| 361 |
+
"learning_rate": 1.6774642643563953e-05,
|
| 362 |
+
"loss": 2.3441,
|
| 363 |
"step": 118
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 2.46,
|
| 367 |
+
"learning_rate": 1.4354488511294417e-05,
|
| 368 |
+
"loss": 1.9161,
|
| 369 |
"step": 120
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 2.5,
|
| 373 |
+
"learning_rate": 1.2109303251503434e-05,
|
| 374 |
+
"loss": 1.8637,
|
| 375 |
"step": 122
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 2.54,
|
| 379 |
+
"learning_rate": 1.0043673649027518e-05,
|
| 380 |
+
"loss": 2.1511,
|
| 381 |
"step": 124
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 2.58,
|
| 385 |
+
"learning_rate": 8.161819666783888e-06,
|
| 386 |
+
"loss": 2.1222,
|
| 387 |
"step": 126
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 2.63,
|
| 391 |
+
"learning_rate": 6.467585824627887e-06,
|
| 392 |
+
"loss": 2.3351,
|
| 393 |
"step": 128
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 2.67,
|
| 397 |
+
"learning_rate": 4.964433345219355e-06,
|
| 398 |
+
"loss": 2.217,
|
| 399 |
"step": 130
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 2.71,
|
| 403 |
+
"learning_rate": 3.655433082942972e-06,
|
| 404 |
+
"loss": 2.362,
|
| 405 |
"step": 132
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 2.75,
|
| 409 |
+
"learning_rate": 2.5432592503288e-06,
|
| 410 |
+
"loss": 2.1953,
|
| 411 |
"step": 134
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 2.79,
|
| 415 |
+
"learning_rate": 1.6301839547892328e-06,
|
| 416 |
+
"loss": 2.1478,
|
| 417 |
"step": 136
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 2.83,
|
| 421 |
+
"learning_rate": 9.180725568338044e-07,
|
| 422 |
+
"loss": 2.1976,
|
| 423 |
"step": 138
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 2.87,
|
| 427 |
+
"learning_rate": 4.0837985924448984e-07,
|
| 428 |
+
"loss": 2.1874,
|
| 429 |
"step": 140
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 2.91,
|
| 433 |
+
"learning_rate": 1.0214713499706597e-07,
|
| 434 |
+
"loss": 2.3805,
|
| 435 |
"step": 142
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 2.95,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 439 |
"learning_rate": 0.0,
|
| 440 |
+
"loss": 2.2218,
|
| 441 |
+
"step": 144
|
| 442 |
}
|
| 443 |
],
|
| 444 |
"logging_steps": 2,
|
| 445 |
+
"max_steps": 144,
|
| 446 |
"num_train_epochs": 3,
|
| 447 |
"save_steps": 500,
|
| 448 |
+
"total_flos": 6408232920367104.0,
|
| 449 |
"trial_name": null,
|
| 450 |
"trial_params": null
|
| 451 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4600
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd87892201410819e4011448b9fe7093db86d6a66f410aa57e3570916e8f2bf6
|
| 3 |
size 4600
|