Upload folder using huggingface_hub
#4
by
CreatorPhan
- opened
- README.md +1 -2
- adapter_model.bin +1 -1
- checkpoint-100/README.md +2 -4
- checkpoint-100/adapter_model.bin +1 -1
- checkpoint-100/optimizer.pt +2 -2
- checkpoint-100/rng_state.pth +1 -1
- checkpoint-100/trainer_state.json +203 -206
- checkpoint-100/training_args.bin +2 -2
- training_args.bin +2 -2
README.md
CHANGED
|
@@ -5,7 +5,6 @@ library_name: peft
|
|
| 5 |
|
| 6 |
|
| 7 |
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
-
- quant_method: bitsandbytes
|
| 9 |
- load_in_8bit: True
|
| 10 |
- load_in_4bit: False
|
| 11 |
- llm_int8_threshold: 6.0
|
|
@@ -18,4 +17,4 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 18 |
### Framework versions
|
| 19 |
|
| 20 |
|
| 21 |
-
- PEFT 0.
|
|
|
|
| 5 |
|
| 6 |
|
| 7 |
The following `bitsandbytes` quantization config was used during training:
|
|
|
|
| 8 |
- load_in_8bit: True
|
| 9 |
- load_in_4bit: False
|
| 10 |
- llm_int8_threshold: 6.0
|
|
|
|
| 17 |
### Framework versions
|
| 18 |
|
| 19 |
|
| 20 |
+
- PEFT 0.4.0
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 39409357
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c57a457d1ce624c042cf5f076804ef0ce9ab9b3e07fef0a29dce7b58311b6c6
|
| 3 |
size 39409357
|
checkpoint-100/README.md
CHANGED
|
@@ -5,7 +5,6 @@ library_name: peft
|
|
| 5 |
|
| 6 |
|
| 7 |
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
-
- quant_method: bitsandbytes
|
| 9 |
- load_in_8bit: True
|
| 10 |
- load_in_4bit: False
|
| 11 |
- llm_int8_threshold: 6.0
|
|
@@ -17,7 +16,6 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 17 |
- bnb_4bit_compute_dtype: float32
|
| 18 |
|
| 19 |
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
-
- quant_method: bitsandbytes
|
| 21 |
- load_in_8bit: True
|
| 22 |
- load_in_4bit: False
|
| 23 |
- llm_int8_threshold: 6.0
|
|
@@ -29,6 +27,6 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 29 |
- bnb_4bit_compute_dtype: float32
|
| 30 |
### Framework versions
|
| 31 |
|
| 32 |
-
- PEFT 0.
|
| 33 |
|
| 34 |
-
- PEFT 0.
|
|
|
|
| 5 |
|
| 6 |
|
| 7 |
The following `bitsandbytes` quantization config was used during training:
|
|
|
|
| 8 |
- load_in_8bit: True
|
| 9 |
- load_in_4bit: False
|
| 10 |
- llm_int8_threshold: 6.0
|
|
|
|
| 16 |
- bnb_4bit_compute_dtype: float32
|
| 17 |
|
| 18 |
The following `bitsandbytes` quantization config was used during training:
|
|
|
|
| 19 |
- load_in_8bit: True
|
| 20 |
- load_in_4bit: False
|
| 21 |
- llm_int8_threshold: 6.0
|
|
|
|
| 27 |
- bnb_4bit_compute_dtype: float32
|
| 28 |
### Framework versions
|
| 29 |
|
| 30 |
+
- PEFT 0.4.0
|
| 31 |
|
| 32 |
+
- PEFT 0.4.0
|
checkpoint-100/adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 39409357
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:044e3971abe97a7f3b70569e01834ebc3e8f0ca21682fb1b3a18c6f125bb4e8c
|
| 3 |
size 39409357
|
checkpoint-100/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 sha256:1ea58d9361a5148e15d11e309e14ba6ee6dba57f3cc0ab6da8fd515b70f6ce64
|
| 3 |
+
size 78784709
|
checkpoint-100/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7508d4b8dd267de5cc58e972da25236687927651336a28f292c92f7f23951475
|
| 3 |
size 14575
|
checkpoint-100/trainer_state.json
CHANGED
|
@@ -1,619 +1,616 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps": 500,
|
| 6 |
"global_step": 100,
|
| 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": 0.00099375,
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
"learning_rate": 0.0009875,
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 2
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
"learning_rate": 0.00098125,
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 3
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
"learning_rate": 0.000975,
|
| 32 |
-
"loss": 2.
|
| 33 |
"step": 4
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
"learning_rate": 0.00096875,
|
| 38 |
-
"loss": 2.
|
| 39 |
"step": 5
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
"learning_rate": 0.0009625,
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 6
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
"learning_rate": 0.0009562500000000001,
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 7
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
"learning_rate": 0.00095,
|
| 56 |
-
"loss": 2.
|
| 57 |
"step": 8
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
"learning_rate": 0.00094375,
|
| 62 |
-
"loss": 2.
|
| 63 |
"step": 9
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
"learning_rate": 0.0009375,
|
| 68 |
-
"loss": 2.
|
| 69 |
"step": 10
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
"learning_rate": 0.00093125,
|
| 74 |
-
"loss": 2.
|
| 75 |
"step": 11
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 0.
|
| 79 |
"learning_rate": 0.000925,
|
| 80 |
-
"loss": 2.
|
| 81 |
"step": 12
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
"learning_rate": 0.00091875,
|
| 86 |
-
"loss": 2.
|
| 87 |
"step": 13
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
"learning_rate": 0.0009125,
|
| 92 |
-
"loss": 2.
|
| 93 |
"step": 14
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
"learning_rate": 0.00090625,
|
| 98 |
-
"loss": 2.
|
| 99 |
"step": 15
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
"learning_rate": 0.0009000000000000001,
|
| 104 |
-
"loss": 2.
|
| 105 |
"step": 16
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
"learning_rate": 0.00089375,
|
| 110 |
-
"loss": 2.
|
| 111 |
"step": 17
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch": 0.
|
| 115 |
"learning_rate": 0.0008874999999999999,
|
| 116 |
-
"loss": 2.
|
| 117 |
"step": 18
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 0.
|
| 121 |
"learning_rate": 0.00088125,
|
| 122 |
-
"loss": 2.
|
| 123 |
"step": 19
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
"learning_rate": 0.000875,
|
| 128 |
-
"loss": 2.
|
| 129 |
"step": 20
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
"learning_rate": 0.0008687500000000001,
|
| 134 |
-
"loss": 2.
|
| 135 |
"step": 21
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
"learning_rate": 0.0008625000000000001,
|
| 140 |
-
"loss": 2.
|
| 141 |
"step": 22
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch": 0.
|
| 145 |
"learning_rate": 0.00085625,
|
| 146 |
-
"loss": 2.
|
| 147 |
"step": 23
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
"learning_rate": 0.00085,
|
| 152 |
-
"loss": 2.
|
| 153 |
"step": 24
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 0.
|
| 157 |
"learning_rate": 0.00084375,
|
| 158 |
-
"loss": 2.
|
| 159 |
"step": 25
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 0.
|
| 163 |
"learning_rate": 0.0008375,
|
| 164 |
-
"loss": 2.
|
| 165 |
"step": 26
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch": 0.
|
| 169 |
"learning_rate": 0.0008312500000000001,
|
| 170 |
-
"loss": 2.
|
| 171 |
"step": 27
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
"learning_rate": 0.000825,
|
| 176 |
-
"loss": 2.
|
| 177 |
"step": 28
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
"learning_rate": 0.00081875,
|
| 182 |
-
"loss": 2.
|
| 183 |
"step": 29
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch": 0.
|
| 187 |
"learning_rate": 0.0008125000000000001,
|
| 188 |
-
"loss": 2.
|
| 189 |
"step": 30
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 0.
|
| 193 |
"learning_rate": 0.00080625,
|
| 194 |
-
"loss": 2.
|
| 195 |
"step": 31
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
"learning_rate": 0.0008,
|
| 200 |
-
"loss": 2.
|
| 201 |
"step": 32
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 0.
|
| 205 |
"learning_rate": 0.00079375,
|
| 206 |
-
"loss": 2.
|
| 207 |
"step": 33
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 0.
|
| 211 |
"learning_rate": 0.0007875,
|
| 212 |
-
"loss": 2.
|
| 213 |
"step": 34
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
"learning_rate": 0.00078125,
|
| 218 |
-
"loss": 2.
|
| 219 |
"step": 35
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
"learning_rate": 0.0007750000000000001,
|
| 224 |
-
"loss": 2.
|
| 225 |
"step": 36
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch": 0.
|
| 229 |
"learning_rate": 0.00076875,
|
| 230 |
-
"loss": 2.
|
| 231 |
"step": 37
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch": 0.
|
| 235 |
"learning_rate": 0.0007624999999999999,
|
| 236 |
-
"loss": 2.
|
| 237 |
"step": 38
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 0.
|
| 241 |
"learning_rate": 0.00075625,
|
| 242 |
-
"loss": 2.
|
| 243 |
"step": 39
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch": 0
|
| 247 |
"learning_rate": 0.00075,
|
| 248 |
-
"loss": 2.
|
| 249 |
"step": 40
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
"learning_rate": 0.00074375,
|
| 254 |
-
"loss": 2.
|
| 255 |
"step": 41
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
"learning_rate": 0.0007375000000000001,
|
| 260 |
-
"loss": 2.
|
| 261 |
"step": 42
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch":
|
| 265 |
"learning_rate": 0.00073125,
|
| 266 |
-
"loss": 2.
|
| 267 |
"step": 43
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
"learning_rate": 0.000725,
|
| 272 |
-
"loss": 2.
|
| 273 |
"step": 44
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
"learning_rate": 0.00071875,
|
| 278 |
-
"loss": 2.
|
| 279 |
"step": 45
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch":
|
| 283 |
"learning_rate": 0.0007125,
|
| 284 |
-
"loss": 2.
|
| 285 |
"step": 46
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
"learning_rate": 0.0007062500000000001,
|
| 290 |
-
"loss": 2.
|
| 291 |
"step": 47
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch":
|
| 295 |
"learning_rate": 0.0007,
|
| 296 |
-
"loss": 2.
|
| 297 |
"step": 48
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
"learning_rate": 0.00069375,
|
| 302 |
-
"loss": 2.
|
| 303 |
"step": 49
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch":
|
| 307 |
"learning_rate": 0.0006875,
|
| 308 |
-
"loss": 2.
|
| 309 |
"step": 50
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch":
|
| 313 |
"learning_rate": 0.00068125,
|
| 314 |
-
"loss": 2.
|
| 315 |
"step": 51
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch":
|
| 319 |
"learning_rate": 0.000675,
|
| 320 |
-
"loss": 2.
|
| 321 |
"step": 52
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
"learning_rate": 0.00066875,
|
| 326 |
-
"loss": 2.
|
| 327 |
"step": 53
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
"learning_rate": 0.0006625,
|
| 332 |
-
"loss": 2.
|
| 333 |
"step": 54
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch":
|
| 337 |
"learning_rate": 0.00065625,
|
| 338 |
-
"loss": 2.
|
| 339 |
"step": 55
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
"learning_rate": 0.0006500000000000001,
|
| 344 |
-
"loss": 2.
|
| 345 |
"step": 56
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
"learning_rate": 0.00064375,
|
| 350 |
-
"loss": 2.
|
| 351 |
"step": 57
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch":
|
| 355 |
"learning_rate": 0.0006374999999999999,
|
| 356 |
-
"loss": 2.
|
| 357 |
"step": 58
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
"learning_rate": 0.00063125,
|
| 362 |
-
"loss": 2.
|
| 363 |
"step": 59
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch":
|
| 367 |
"learning_rate": 0.000625,
|
| 368 |
-
"loss": 2.
|
| 369 |
"step": 60
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
"learning_rate": 0.00061875,
|
| 374 |
-
"loss": 2.
|
| 375 |
"step": 61
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch":
|
| 379 |
"learning_rate": 0.0006125000000000001,
|
| 380 |
-
"loss": 2.
|
| 381 |
"step": 62
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
"learning_rate": 0.00060625,
|
| 386 |
-
"loss": 2.
|
| 387 |
"step": 63
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch":
|
| 391 |
"learning_rate": 0.0006,
|
| 392 |
-
"loss": 2.
|
| 393 |
"step": 64
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch":
|
| 397 |
"learning_rate": 0.00059375,
|
| 398 |
-
"loss": 2.
|
| 399 |
"step": 65
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch":
|
| 403 |
"learning_rate": 0.0005875,
|
| 404 |
-
"loss": 2.
|
| 405 |
"step": 66
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch":
|
| 409 |
"learning_rate": 0.0005812500000000001,
|
| 410 |
-
"loss": 2.
|
| 411 |
"step": 67
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch":
|
| 415 |
"learning_rate": 0.000575,
|
| 416 |
-
"loss": 2.
|
| 417 |
"step": 68
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
"learning_rate": 0.00056875,
|
| 422 |
-
"loss": 2.
|
| 423 |
"step": 69
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
"learning_rate": 0.0005625000000000001,
|
| 428 |
-
"loss": 2.
|
| 429 |
"step": 70
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch":
|
| 433 |
"learning_rate": 0.00055625,
|
| 434 |
-
"loss": 2.
|
| 435 |
"step": 71
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
"learning_rate": 0.00055,
|
| 440 |
-
"loss": 2.
|
| 441 |
"step": 72
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch":
|
| 445 |
"learning_rate": 0.00054375,
|
| 446 |
-
"loss": 2.
|
| 447 |
"step": 73
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch":
|
| 451 |
"learning_rate": 0.0005375,
|
| 452 |
-
"loss": 2.
|
| 453 |
"step": 74
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
"learning_rate": 0.00053125,
|
| 458 |
-
"loss": 2.
|
| 459 |
"step": 75
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
"learning_rate": 0.0005250000000000001,
|
| 464 |
-
"loss": 2.
|
| 465 |
"step": 76
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
"learning_rate": 0.00051875,
|
| 470 |
-
"loss": 2.
|
| 471 |
"step": 77
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch":
|
| 475 |
"learning_rate": 0.0005124999999999999,
|
| 476 |
-
"loss": 2.
|
| 477 |
"step": 78
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch":
|
| 481 |
"learning_rate": 0.00050625,
|
| 482 |
-
"loss": 2.
|
| 483 |
"step": 79
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch": 1.
|
| 487 |
"learning_rate": 0.0005,
|
| 488 |
-
"loss": 2.
|
| 489 |
"step": 80
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch":
|
| 493 |
"learning_rate": 0.00049375,
|
| 494 |
-
"loss": 2.
|
| 495 |
"step": 81
|
| 496 |
},
|
| 497 |
{
|
| 498 |
-
"epoch":
|
| 499 |
"learning_rate": 0.0004875,
|
| 500 |
-
"loss": 2.
|
| 501 |
"step": 82
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch":
|
| 505 |
"learning_rate": 0.00048125,
|
| 506 |
-
"loss": 2.
|
| 507 |
"step": 83
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch":
|
| 511 |
"learning_rate": 0.000475,
|
| 512 |
-
"loss": 2.
|
| 513 |
"step": 84
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch":
|
| 517 |
"learning_rate": 0.00046875,
|
| 518 |
-
"loss": 2.
|
| 519 |
"step": 85
|
| 520 |
},
|
| 521 |
{
|
| 522 |
-
"epoch":
|
| 523 |
"learning_rate": 0.0004625,
|
| 524 |
-
"loss": 2.
|
| 525 |
"step": 86
|
| 526 |
},
|
| 527 |
{
|
| 528 |
-
"epoch":
|
| 529 |
"learning_rate": 0.00045625,
|
| 530 |
-
"loss": 2.
|
| 531 |
"step": 87
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
"learning_rate": 0.00045000000000000004,
|
| 536 |
-
"loss": 2.
|
| 537 |
"step": 88
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch":
|
| 541 |
"learning_rate": 0.00044374999999999997,
|
| 542 |
-
"loss": 2.
|
| 543 |
"step": 89
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"epoch":
|
| 547 |
"learning_rate": 0.0004375,
|
| 548 |
-
"loss": 2.
|
| 549 |
"step": 90
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch":
|
| 553 |
"learning_rate": 0.00043125000000000005,
|
| 554 |
-
"loss": 2.
|
| 555 |
"step": 91
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch":
|
| 559 |
"learning_rate": 0.000425,
|
| 560 |
-
"loss": 2.
|
| 561 |
"step": 92
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
"learning_rate": 0.00041875,
|
| 566 |
-
"loss": 2.
|
| 567 |
"step": 93
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch":
|
| 571 |
"learning_rate": 0.0004125,
|
| 572 |
-
"loss": 2.
|
| 573 |
"step": 94
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch":
|
| 577 |
"learning_rate": 0.00040625000000000004,
|
| 578 |
-
"loss": 2.
|
| 579 |
"step": 95
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch":
|
| 583 |
"learning_rate": 0.0004,
|
| 584 |
-
"loss": 2.
|
| 585 |
"step": 96
|
| 586 |
},
|
| 587 |
{
|
| 588 |
-
"epoch":
|
| 589 |
"learning_rate": 0.00039375,
|
| 590 |
-
"loss": 2.
|
| 591 |
"step": 97
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
"learning_rate": 0.00038750000000000004,
|
| 596 |
-
"loss": 2.
|
| 597 |
"step": 98
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch":
|
| 601 |
"learning_rate": 0.00038124999999999997,
|
| 602 |
-
"loss": 2.
|
| 603 |
"step": 99
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch":
|
| 607 |
"learning_rate": 0.000375,
|
| 608 |
-
"loss": 2.
|
| 609 |
"step": 100
|
| 610 |
}
|
| 611 |
],
|
| 612 |
-
"logging_steps": 1,
|
| 613 |
"max_steps": 160,
|
| 614 |
-
"num_train_epochs":
|
| 615 |
-
"
|
| 616 |
-
"total_flos": 2.588289853513728e+16,
|
| 617 |
"trial_name": null,
|
| 618 |
"trial_params": null
|
| 619 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.488335925349922,
|
|
|
|
| 5 |
"global_step": 100,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
"learning_rate": 0.00099375,
|
| 13 |
+
"loss": 2.9857,
|
| 14 |
"step": 1
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
"learning_rate": 0.0009875,
|
| 19 |
+
"loss": 2.8497,
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
"learning_rate": 0.00098125,
|
| 25 |
+
"loss": 2.8853,
|
| 26 |
"step": 3
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
"learning_rate": 0.000975,
|
| 31 |
+
"loss": 2.7689,
|
| 32 |
"step": 4
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.12,
|
| 36 |
"learning_rate": 0.00096875,
|
| 37 |
+
"loss": 2.7474,
|
| 38 |
"step": 5
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
"learning_rate": 0.0009625,
|
| 43 |
+
"loss": 2.7695,
|
| 44 |
"step": 6
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
"learning_rate": 0.0009562500000000001,
|
| 49 |
+
"loss": 2.6897,
|
| 50 |
"step": 7
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
"learning_rate": 0.00095,
|
| 55 |
+
"loss": 2.6285,
|
| 56 |
"step": 8
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.22,
|
| 60 |
"learning_rate": 0.00094375,
|
| 61 |
+
"loss": 2.7107,
|
| 62 |
"step": 9
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
"learning_rate": 0.0009375,
|
| 67 |
+
"loss": 2.6897,
|
| 68 |
"step": 10
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.27,
|
| 72 |
"learning_rate": 0.00093125,
|
| 73 |
+
"loss": 2.6634,
|
| 74 |
"step": 11
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.3,
|
| 78 |
"learning_rate": 0.000925,
|
| 79 |
+
"loss": 2.635,
|
| 80 |
"step": 12
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.32,
|
| 84 |
"learning_rate": 0.00091875,
|
| 85 |
+
"loss": 2.643,
|
| 86 |
"step": 13
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.35,
|
| 90 |
"learning_rate": 0.0009125,
|
| 91 |
+
"loss": 2.7166,
|
| 92 |
"step": 14
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.37,
|
| 96 |
"learning_rate": 0.00090625,
|
| 97 |
+
"loss": 2.7354,
|
| 98 |
"step": 15
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.4,
|
| 102 |
"learning_rate": 0.0009000000000000001,
|
| 103 |
+
"loss": 2.615,
|
| 104 |
"step": 16
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.42,
|
| 108 |
"learning_rate": 0.00089375,
|
| 109 |
+
"loss": 2.7225,
|
| 110 |
"step": 17
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.45,
|
| 114 |
"learning_rate": 0.0008874999999999999,
|
| 115 |
+
"loss": 2.6189,
|
| 116 |
"step": 18
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.47,
|
| 120 |
"learning_rate": 0.00088125,
|
| 121 |
+
"loss": 2.7269,
|
| 122 |
"step": 19
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.5,
|
| 126 |
"learning_rate": 0.000875,
|
| 127 |
+
"loss": 2.6312,
|
| 128 |
"step": 20
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.52,
|
| 132 |
"learning_rate": 0.0008687500000000001,
|
| 133 |
+
"loss": 2.6758,
|
| 134 |
"step": 21
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.55,
|
| 138 |
"learning_rate": 0.0008625000000000001,
|
| 139 |
+
"loss": 2.6997,
|
| 140 |
"step": 22
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.57,
|
| 144 |
"learning_rate": 0.00085625,
|
| 145 |
+
"loss": 2.6878,
|
| 146 |
"step": 23
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.6,
|
| 150 |
"learning_rate": 0.00085,
|
| 151 |
+
"loss": 2.6433,
|
| 152 |
"step": 24
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.62,
|
| 156 |
"learning_rate": 0.00084375,
|
| 157 |
+
"loss": 2.7053,
|
| 158 |
"step": 25
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.65,
|
| 162 |
"learning_rate": 0.0008375,
|
| 163 |
+
"loss": 2.6745,
|
| 164 |
"step": 26
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.67,
|
| 168 |
"learning_rate": 0.0008312500000000001,
|
| 169 |
+
"loss": 2.7376,
|
| 170 |
"step": 27
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.7,
|
| 174 |
"learning_rate": 0.000825,
|
| 175 |
+
"loss": 2.6577,
|
| 176 |
"step": 28
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.72,
|
| 180 |
"learning_rate": 0.00081875,
|
| 181 |
+
"loss": 2.7428,
|
| 182 |
"step": 29
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.75,
|
| 186 |
"learning_rate": 0.0008125000000000001,
|
| 187 |
+
"loss": 2.6231,
|
| 188 |
"step": 30
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.77,
|
| 192 |
"learning_rate": 0.00080625,
|
| 193 |
+
"loss": 2.7078,
|
| 194 |
"step": 31
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.8,
|
| 198 |
"learning_rate": 0.0008,
|
| 199 |
+
"loss": 2.7057,
|
| 200 |
"step": 32
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.82,
|
| 204 |
"learning_rate": 0.00079375,
|
| 205 |
+
"loss": 2.6594,
|
| 206 |
"step": 33
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.85,
|
| 210 |
"learning_rate": 0.0007875,
|
| 211 |
+
"loss": 2.6718,
|
| 212 |
"step": 34
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.87,
|
| 216 |
"learning_rate": 0.00078125,
|
| 217 |
+
"loss": 2.6459,
|
| 218 |
"step": 35
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.9,
|
| 222 |
"learning_rate": 0.0007750000000000001,
|
| 223 |
+
"loss": 2.6558,
|
| 224 |
"step": 36
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.92,
|
| 228 |
"learning_rate": 0.00076875,
|
| 229 |
+
"loss": 2.6559,
|
| 230 |
"step": 37
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.95,
|
| 234 |
"learning_rate": 0.0007624999999999999,
|
| 235 |
+
"loss": 2.6488,
|
| 236 |
"step": 38
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.97,
|
| 240 |
"learning_rate": 0.00075625,
|
| 241 |
+
"loss": 2.5902,
|
| 242 |
"step": 39
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 1.0,
|
| 246 |
"learning_rate": 0.00075,
|
| 247 |
+
"loss": 2.6244,
|
| 248 |
"step": 40
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 1.02,
|
| 252 |
"learning_rate": 0.00074375,
|
| 253 |
+
"loss": 2.5278,
|
| 254 |
"step": 41
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 1.05,
|
| 258 |
"learning_rate": 0.0007375000000000001,
|
| 259 |
+
"loss": 2.5213,
|
| 260 |
"step": 42
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 1.07,
|
| 264 |
"learning_rate": 0.00073125,
|
| 265 |
+
"loss": 2.5515,
|
| 266 |
"step": 43
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 1.09,
|
| 270 |
"learning_rate": 0.000725,
|
| 271 |
+
"loss": 2.5196,
|
| 272 |
"step": 44
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 1.12,
|
| 276 |
"learning_rate": 0.00071875,
|
| 277 |
+
"loss": 2.5725,
|
| 278 |
"step": 45
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 1.14,
|
| 282 |
"learning_rate": 0.0007125,
|
| 283 |
+
"loss": 2.4971,
|
| 284 |
"step": 46
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 1.17,
|
| 288 |
"learning_rate": 0.0007062500000000001,
|
| 289 |
+
"loss": 2.4976,
|
| 290 |
"step": 47
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 1.19,
|
| 294 |
"learning_rate": 0.0007,
|
| 295 |
+
"loss": 2.5467,
|
| 296 |
"step": 48
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 1.22,
|
| 300 |
"learning_rate": 0.00069375,
|
| 301 |
+
"loss": 2.5302,
|
| 302 |
"step": 49
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 1.24,
|
| 306 |
"learning_rate": 0.0006875,
|
| 307 |
+
"loss": 2.5224,
|
| 308 |
"step": 50
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"epoch": 1.27,
|
| 312 |
"learning_rate": 0.00068125,
|
| 313 |
+
"loss": 2.5451,
|
| 314 |
"step": 51
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 1.29,
|
| 318 |
"learning_rate": 0.000675,
|
| 319 |
+
"loss": 2.4609,
|
| 320 |
"step": 52
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 1.32,
|
| 324 |
"learning_rate": 0.00066875,
|
| 325 |
+
"loss": 2.4897,
|
| 326 |
"step": 53
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 1.34,
|
| 330 |
"learning_rate": 0.0006625,
|
| 331 |
+
"loss": 2.5583,
|
| 332 |
"step": 54
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 1.37,
|
| 336 |
"learning_rate": 0.00065625,
|
| 337 |
+
"loss": 2.5675,
|
| 338 |
"step": 55
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 1.39,
|
| 342 |
"learning_rate": 0.0006500000000000001,
|
| 343 |
+
"loss": 2.5598,
|
| 344 |
"step": 56
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 1.42,
|
| 348 |
"learning_rate": 0.00064375,
|
| 349 |
+
"loss": 2.48,
|
| 350 |
"step": 57
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 1.44,
|
| 354 |
"learning_rate": 0.0006374999999999999,
|
| 355 |
+
"loss": 2.4761,
|
| 356 |
"step": 58
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 1.47,
|
| 360 |
"learning_rate": 0.00063125,
|
| 361 |
+
"loss": 2.5599,
|
| 362 |
"step": 59
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 1.49,
|
| 366 |
"learning_rate": 0.000625,
|
| 367 |
+
"loss": 2.5199,
|
| 368 |
"step": 60
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 1.52,
|
| 372 |
"learning_rate": 0.00061875,
|
| 373 |
+
"loss": 2.5161,
|
| 374 |
"step": 61
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 1.54,
|
| 378 |
"learning_rate": 0.0006125000000000001,
|
| 379 |
+
"loss": 2.5797,
|
| 380 |
"step": 62
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 1.57,
|
| 384 |
"learning_rate": 0.00060625,
|
| 385 |
+
"loss": 2.5345,
|
| 386 |
"step": 63
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 1.59,
|
| 390 |
"learning_rate": 0.0006,
|
| 391 |
+
"loss": 2.4726,
|
| 392 |
"step": 64
|
| 393 |
},
|
| 394 |
{
|
| 395 |
+
"epoch": 1.62,
|
| 396 |
"learning_rate": 0.00059375,
|
| 397 |
+
"loss": 2.448,
|
| 398 |
"step": 65
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 1.64,
|
| 402 |
"learning_rate": 0.0005875,
|
| 403 |
+
"loss": 2.4939,
|
| 404 |
"step": 66
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.67,
|
| 408 |
"learning_rate": 0.0005812500000000001,
|
| 409 |
+
"loss": 2.4881,
|
| 410 |
"step": 67
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 1.69,
|
| 414 |
"learning_rate": 0.000575,
|
| 415 |
+
"loss": 2.5731,
|
| 416 |
"step": 68
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 1.72,
|
| 420 |
"learning_rate": 0.00056875,
|
| 421 |
+
"loss": 2.5114,
|
| 422 |
"step": 69
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 1.74,
|
| 426 |
"learning_rate": 0.0005625000000000001,
|
| 427 |
+
"loss": 2.5049,
|
| 428 |
"step": 70
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 1.77,
|
| 432 |
"learning_rate": 0.00055625,
|
| 433 |
+
"loss": 2.4856,
|
| 434 |
"step": 71
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 1.79,
|
| 438 |
"learning_rate": 0.00055,
|
| 439 |
+
"loss": 2.5077,
|
| 440 |
"step": 72
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 1.82,
|
| 444 |
"learning_rate": 0.00054375,
|
| 445 |
+
"loss": 2.4801,
|
| 446 |
"step": 73
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 1.84,
|
| 450 |
"learning_rate": 0.0005375,
|
| 451 |
+
"loss": 2.5599,
|
| 452 |
"step": 74
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 1.87,
|
| 456 |
"learning_rate": 0.00053125,
|
| 457 |
+
"loss": 2.5269,
|
| 458 |
"step": 75
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 1.89,
|
| 462 |
"learning_rate": 0.0005250000000000001,
|
| 463 |
+
"loss": 2.4963,
|
| 464 |
"step": 76
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 1.92,
|
| 468 |
"learning_rate": 0.00051875,
|
| 469 |
+
"loss": 2.4595,
|
| 470 |
"step": 77
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"epoch": 1.94,
|
| 474 |
"learning_rate": 0.0005124999999999999,
|
| 475 |
+
"loss": 2.4801,
|
| 476 |
"step": 78
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 1.97,
|
| 480 |
"learning_rate": 0.00050625,
|
| 481 |
+
"loss": 2.5154,
|
| 482 |
"step": 79
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 1.99,
|
| 486 |
"learning_rate": 0.0005,
|
| 487 |
+
"loss": 2.5179,
|
| 488 |
"step": 80
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 2.02,
|
| 492 |
"learning_rate": 0.00049375,
|
| 493 |
+
"loss": 2.4121,
|
| 494 |
"step": 81
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 2.04,
|
| 498 |
"learning_rate": 0.0004875,
|
| 499 |
+
"loss": 2.354,
|
| 500 |
"step": 82
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"epoch": 2.07,
|
| 504 |
"learning_rate": 0.00048125,
|
| 505 |
+
"loss": 2.3033,
|
| 506 |
"step": 83
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 2.09,
|
| 510 |
"learning_rate": 0.000475,
|
| 511 |
+
"loss": 2.4023,
|
| 512 |
"step": 84
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 2.12,
|
| 516 |
"learning_rate": 0.00046875,
|
| 517 |
+
"loss": 2.3755,
|
| 518 |
"step": 85
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 2.14,
|
| 522 |
"learning_rate": 0.0004625,
|
| 523 |
+
"loss": 2.3892,
|
| 524 |
"step": 86
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 2.16,
|
| 528 |
"learning_rate": 0.00045625,
|
| 529 |
+
"loss": 2.3534,
|
| 530 |
"step": 87
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 2.19,
|
| 534 |
"learning_rate": 0.00045000000000000004,
|
| 535 |
+
"loss": 2.3634,
|
| 536 |
"step": 88
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 2.21,
|
| 540 |
"learning_rate": 0.00044374999999999997,
|
| 541 |
+
"loss": 2.3336,
|
| 542 |
"step": 89
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 2.24,
|
| 546 |
"learning_rate": 0.0004375,
|
| 547 |
+
"loss": 2.3646,
|
| 548 |
"step": 90
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 2.26,
|
| 552 |
"learning_rate": 0.00043125000000000005,
|
| 553 |
+
"loss": 2.3333,
|
| 554 |
"step": 91
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 2.29,
|
| 558 |
"learning_rate": 0.000425,
|
| 559 |
+
"loss": 2.3344,
|
| 560 |
"step": 92
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 2.31,
|
| 564 |
"learning_rate": 0.00041875,
|
| 565 |
+
"loss": 2.3308,
|
| 566 |
"step": 93
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 2.34,
|
| 570 |
"learning_rate": 0.0004125,
|
| 571 |
+
"loss": 2.3677,
|
| 572 |
"step": 94
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 2.36,
|
| 576 |
"learning_rate": 0.00040625000000000004,
|
| 577 |
+
"loss": 2.3755,
|
| 578 |
"step": 95
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 2.39,
|
| 582 |
"learning_rate": 0.0004,
|
| 583 |
+
"loss": 2.3721,
|
| 584 |
"step": 96
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 2.41,
|
| 588 |
"learning_rate": 0.00039375,
|
| 589 |
+
"loss": 2.3463,
|
| 590 |
"step": 97
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.44,
|
| 594 |
"learning_rate": 0.00038750000000000004,
|
| 595 |
+
"loss": 2.3536,
|
| 596 |
"step": 98
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 2.46,
|
| 600 |
"learning_rate": 0.00038124999999999997,
|
| 601 |
+
"loss": 2.3895,
|
| 602 |
"step": 99
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 2.49,
|
| 606 |
"learning_rate": 0.000375,
|
| 607 |
+
"loss": 2.3767,
|
| 608 |
"step": 100
|
| 609 |
}
|
| 610 |
],
|
|
|
|
| 611 |
"max_steps": 160,
|
| 612 |
+
"num_train_epochs": 4,
|
| 613 |
+
"total_flos": 5.95613039443968e+16,
|
|
|
|
| 614 |
"trial_name": null,
|
| 615 |
"trial_params": null
|
| 616 |
}
|
checkpoint-100/training_args.bin
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 sha256:29aa9882c4c56a1507dbc7c22ccd2d199e9b5b94a8c3844ada75fa3f9e9d5e41
|
| 3 |
+
size 3963
|
training_args.bin
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 sha256:29aa9882c4c56a1507dbc7c22ccd2d199e9b5b94a8c3844ada75fa3f9e9d5e41
|
| 3 |
+
size 3963
|