End of training
Browse files- README.md +11 -11
- adapter_1/adapter_config.json +2 -2
- adapter_1/adapter_model.safetensors +1 -1
- adapter_config.json +2 -2
- adapter_model.safetensors +1 -1
- trainer_state.json +493 -493
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -49,16 +49,16 @@ The following hyperparameters were used during training:
|
|
| 49 |
|
| 50 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
-
| 0.
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.2212
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 49 |
|
| 50 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 0.0726 | 1.0 | 546 | 0.2210 |
|
| 53 |
+
| 0.0419 | 2.0 | 1092 | 0.2139 |
|
| 54 |
+
| 0.0322 | 3.0 | 1638 | 0.1935 |
|
| 55 |
+
| 0.0175 | 4.0 | 2184 | 0.1896 |
|
| 56 |
+
| 0.0266 | 5.0 | 2730 | 0.1927 |
|
| 57 |
+
| 0.0178 | 6.0 | 3276 | 0.2013 |
|
| 58 |
+
| 0.0081 | 7.0 | 3822 | 0.1979 |
|
| 59 |
+
| 0.0081 | 8.0 | 4368 | 0.2113 |
|
| 60 |
+
| 0.0018 | 9.0 | 4914 | 0.2146 |
|
| 61 |
+
| 0.0015 | 10.0 | 5460 | 0.2212 |
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
adapter_1/adapter_config.json
CHANGED
|
@@ -23,8 +23,8 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"
|
| 27 |
-
"
|
| 28 |
],
|
| 29 |
"task_type": null,
|
| 30 |
"use_dora": false,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
+
"v_proj",
|
| 27 |
+
"q_proj"
|
| 28 |
],
|
| 29 |
"task_type": null,
|
| 30 |
"use_dora": false,
|
adapter_1/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14176064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b28887bcb66a49c5a999c6eb3d3de767b7cce7fddc9b4c82e33787479d628aa8
|
| 3 |
size 14176064
|
adapter_config.json
CHANGED
|
@@ -23,8 +23,8 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"
|
| 27 |
-
"
|
| 28 |
],
|
| 29 |
"task_type": null,
|
| 30 |
"use_dora": false,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
+
"v_proj",
|
| 27 |
+
"q_proj"
|
| 28 |
],
|
| 29 |
"task_type": null,
|
| 30 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14176064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c76afb1d27a22a98d2438cd165915fee99c1936be55a353c091f790c722bdd2
|
| 3 |
size 14176064
|
trainer_state.json
CHANGED
|
@@ -10,1617 +10,1617 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss":
|
| 16 |
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09,
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss":
|
| 23 |
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.14,
|
| 27 |
-
"grad_norm": 1.
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.18,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.23,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 125
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.27,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 150
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 175
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.37,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.41,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 225
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.46,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 250
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5,
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 0.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.55,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 300
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.6,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 325
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.64,
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 350
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.69,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 375
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.73,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.78,
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.82,
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.87,
|
| 139 |
-
"grad_norm":
|
| 140 |
"learning_rate": 0.0009219963031423291,
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 475
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.92,
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
"learning_rate": 0.0009173752310536044,
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 500
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.96,
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
"learning_rate": 0.0009127541589648799,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 525
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 1.0,
|
| 160 |
-
"eval_loss": 0.
|
| 161 |
-
"eval_runtime":
|
| 162 |
-
"eval_samples_per_second": 4.
|
| 163 |
-
"eval_steps_per_second": 0.
|
| 164 |
"step": 546
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 1.01,
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
"learning_rate": 0.0009081330868761552,
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 550
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.05,
|
| 175 |
-
"grad_norm": 0.
|
| 176 |
"learning_rate": 0.0009035120147874307,
|
| 177 |
-
"loss": 0.
|
| 178 |
"step": 575
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 1.1,
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
"learning_rate": 0.000898890942698706,
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 600
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.14,
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
"learning_rate": 0.0008942698706099815,
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 625
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.19,
|
| 196 |
-
"grad_norm": 0.
|
| 197 |
"learning_rate": 0.0008896487985212569,
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 650
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 1.24,
|
| 203 |
-
"grad_norm": 0.
|
| 204 |
"learning_rate": 0.0008850277264325323,
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 675
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.28,
|
| 210 |
-
"grad_norm": 0.
|
| 211 |
"learning_rate": 0.0008804066543438077,
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 700
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 1.33,
|
| 217 |
-
"grad_norm":
|
| 218 |
"learning_rate": 0.0008757855822550833,
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 725
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 1.37,
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
"learning_rate": 0.0008711645101663586,
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 750
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 1.42,
|
| 231 |
-
"grad_norm": 0.
|
| 232 |
"learning_rate": 0.0008665434380776341,
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 775
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 1.47,
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
"learning_rate": 0.0008619223659889095,
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 800
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 1.51,
|
| 245 |
-
"grad_norm":
|
| 246 |
"learning_rate": 0.0008573012939001849,
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 825
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 1.56,
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
"learning_rate": 0.0008526802218114603,
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 850
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 1.6,
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
"learning_rate": 0.0008480591497227357,
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 875
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 1.65,
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
"learning_rate": 0.0008434380776340112,
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 900
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.69,
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
"learning_rate": 0.0008388170055452865,
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 925
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.74,
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
"learning_rate": 0.000834195933456562,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 1.79,
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
"learning_rate": 0.0008295748613678373,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 975
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.83,
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
"learning_rate": 0.0008249537892791128,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 1000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 1.88,
|
| 301 |
-
"grad_norm":
|
| 302 |
"learning_rate": 0.0008203327171903881,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 1025
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 1.92,
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
"learning_rate": 0.0008157116451016636,
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 1050
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 1.97,
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
"learning_rate": 0.000811090573012939,
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 1075
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 2.0,
|
| 322 |
-
"eval_loss": 0.
|
| 323 |
-
"eval_runtime": 174.
|
| 324 |
-
"eval_samples_per_second": 4.
|
| 325 |
"eval_steps_per_second": 0.781,
|
| 326 |
"step": 1092
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 2.01,
|
| 330 |
-
"grad_norm": 0.
|
| 331 |
"learning_rate": 0.0008064695009242144,
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 1100
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 2.06,
|
| 337 |
-
"grad_norm": 0.
|
| 338 |
"learning_rate": 0.0008018484288354898,
|
| 339 |
-
"loss": 0.
|
| 340 |
"step": 1125
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 2.11,
|
| 344 |
-
"grad_norm": 0.
|
| 345 |
"learning_rate": 0.0007972273567467652,
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1150
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 2.15,
|
| 351 |
-
"grad_norm":
|
| 352 |
"learning_rate": 0.0007926062846580406,
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 1175
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 2.2,
|
| 358 |
-
"grad_norm": 0.
|
| 359 |
"learning_rate": 0.0007879852125693162,
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 1200
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 2.24,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
"learning_rate": 0.0007833641404805915,
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 1225
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 2.29,
|
| 372 |
-
"grad_norm": 0.
|
| 373 |
"learning_rate": 0.000778743068391867,
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 1250
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.34,
|
| 379 |
-
"grad_norm": 0.
|
| 380 |
"learning_rate": 0.0007741219963031424,
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 1275
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 2.38,
|
| 386 |
-
"grad_norm":
|
| 387 |
"learning_rate": 0.0007695009242144178,
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1300
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 2.43,
|
| 393 |
-
"grad_norm": 0.
|
| 394 |
"learning_rate": 0.0007648798521256932,
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 1325
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 2.47,
|
| 400 |
-
"grad_norm": 0.
|
| 401 |
"learning_rate": 0.0007602587800369686,
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 1350
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 2.52,
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
"learning_rate": 0.0007556377079482441,
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 1375
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 2.56,
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
"learning_rate": 0.0007510166358595194,
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 1400
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 2.61,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
"learning_rate": 0.0007463955637707949,
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 1425
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 2.66,
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
"learning_rate": 0.0007417744916820702,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1450
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 2.7,
|
| 435 |
-
"grad_norm":
|
| 436 |
"learning_rate": 0.0007371534195933457,
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 1475
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 2.75,
|
| 442 |
-
"grad_norm": 0.
|
| 443 |
"learning_rate": 0.000732532347504621,
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 1500
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 2.79,
|
| 449 |
-
"grad_norm":
|
| 450 |
"learning_rate": 0.0007279112754158965,
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 1525
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 2.84,
|
| 456 |
-
"grad_norm": 0.
|
| 457 |
"learning_rate": 0.0007232902033271719,
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 1550
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 2.88,
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
"learning_rate": 0.0007186691312384473,
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 1575
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 2.93,
|
| 470 |
-
"grad_norm": 0.
|
| 471 |
"learning_rate": 0.0007140480591497227,
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1600
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 2.98,
|
| 477 |
-
"grad_norm": 0.
|
| 478 |
"learning_rate": 0.0007094269870609981,
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 1625
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 3.0,
|
| 484 |
-
"eval_loss": 0.
|
| 485 |
-
"eval_runtime": 175.
|
| 486 |
-
"eval_samples_per_second": 4.
|
| 487 |
-
"eval_steps_per_second": 0.
|
| 488 |
"step": 1638
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 3.02,
|
| 492 |
-
"grad_norm": 0.
|
| 493 |
"learning_rate": 0.0007048059149722735,
|
| 494 |
-
"loss": 0.
|
| 495 |
"step": 1650
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 3.07,
|
| 499 |
-
"grad_norm": 0.
|
| 500 |
"learning_rate": 0.000700184842883549,
|
| 501 |
-
"loss": 0.
|
| 502 |
"step": 1675
|
| 503 |
},
|
| 504 |
{
|
| 505 |
"epoch": 3.11,
|
| 506 |
-
"grad_norm":
|
| 507 |
"learning_rate": 0.0006955637707948245,
|
| 508 |
-
"loss": 0.
|
| 509 |
"step": 1700
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 3.16,
|
| 513 |
-
"grad_norm":
|
| 514 |
"learning_rate": 0.0006909426987060999,
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 1725
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 3.21,
|
| 520 |
-
"grad_norm": 0.
|
| 521 |
"learning_rate": 0.0006863216266173753,
|
| 522 |
-
"loss": 0.
|
| 523 |
"step": 1750
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 3.25,
|
| 527 |
-
"grad_norm": 0.
|
| 528 |
"learning_rate": 0.0006817005545286507,
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 1775
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 3.3,
|
| 534 |
-
"grad_norm": 0.
|
| 535 |
"learning_rate": 0.0006770794824399261,
|
| 536 |
-
"loss": 0.
|
| 537 |
"step": 1800
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 3.34,
|
| 541 |
-
"grad_norm": 0.
|
| 542 |
"learning_rate": 0.0006724584103512015,
|
| 543 |
-
"loss": 0.
|
| 544 |
"step": 1825
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 3.39,
|
| 548 |
-
"grad_norm": 0.
|
| 549 |
"learning_rate": 0.000667837338262477,
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 1850
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 3.43,
|
| 555 |
-
"grad_norm": 0.
|
| 556 |
"learning_rate": 0.0006632162661737523,
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 1875
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.48,
|
| 562 |
-
"grad_norm": 0.
|
| 563 |
"learning_rate": 0.0006585951940850278,
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 1900
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 3.53,
|
| 569 |
-
"grad_norm": 0.
|
| 570 |
"learning_rate": 0.0006539741219963031,
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 1925
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 3.57,
|
| 576 |
-
"grad_norm": 0.
|
| 577 |
"learning_rate": 0.0006493530499075786,
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 1950
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 3.62,
|
| 583 |
-
"grad_norm": 0.
|
| 584 |
"learning_rate": 0.0006447319778188539,
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 1975
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 3.66,
|
| 590 |
-
"grad_norm": 0.
|
| 591 |
"learning_rate": 0.0006401109057301294,
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 2000
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 3.71,
|
| 597 |
-
"grad_norm": 0.
|
| 598 |
"learning_rate": 0.0006354898336414048,
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 2025
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 3.75,
|
| 604 |
-
"grad_norm": 0.
|
| 605 |
"learning_rate": 0.0006308687615526802,
|
| 606 |
-
"loss": 0.
|
| 607 |
"step": 2050
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 3.8,
|
| 611 |
-
"grad_norm": 0.
|
| 612 |
"learning_rate": 0.0006262476894639556,
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 2075
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 3.85,
|
| 618 |
-
"grad_norm": 0.
|
| 619 |
"learning_rate": 0.000621626617375231,
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 2100
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 3.89,
|
| 625 |
-
"grad_norm": 0.
|
| 626 |
"learning_rate": 0.0006170055452865064,
|
| 627 |
-
"loss": 0.
|
| 628 |
"step": 2125
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 3.94,
|
| 632 |
-
"grad_norm": 0.
|
| 633 |
"learning_rate": 0.000612384473197782,
|
| 634 |
-
"loss": 0.
|
| 635 |
"step": 2150
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 3.98,
|
| 639 |
-
"grad_norm": 0.
|
| 640 |
"learning_rate": 0.0006077634011090574,
|
| 641 |
-
"loss": 0.
|
| 642 |
"step": 2175
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 4.0,
|
| 646 |
-
"eval_loss": 0.
|
| 647 |
-
"eval_runtime": 176.
|
| 648 |
-
"eval_samples_per_second": 4.
|
| 649 |
-
"eval_steps_per_second": 0.
|
| 650 |
"step": 2184
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 4.03,
|
| 654 |
-
"grad_norm": 0.
|
| 655 |
"learning_rate": 0.0006031423290203328,
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 2200
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 4.08,
|
| 661 |
-
"grad_norm": 0.
|
| 662 |
"learning_rate": 0.0005985212569316082,
|
| 663 |
-
"loss": 0.
|
| 664 |
"step": 2225
|
| 665 |
},
|
| 666 |
{
|
| 667 |
"epoch": 4.12,
|
| 668 |
-
"grad_norm": 0.
|
| 669 |
"learning_rate": 0.0005939001848428836,
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 2250
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 4.17,
|
| 675 |
-
"grad_norm": 0.
|
| 676 |
"learning_rate": 0.000589279112754159,
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 2275
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 4.21,
|
| 682 |
-
"grad_norm": 0.
|
| 683 |
"learning_rate": 0.0005846580406654344,
|
| 684 |
-
"loss": 0.
|
| 685 |
"step": 2300
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 4.26,
|
| 689 |
-
"grad_norm":
|
| 690 |
"learning_rate": 0.0005800369685767099,
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2325
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 4.3,
|
| 696 |
-
"grad_norm": 0.
|
| 697 |
"learning_rate": 0.0005754158964879852,
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 2350
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 4.35,
|
| 703 |
-
"grad_norm": 0.
|
| 704 |
"learning_rate": 0.0005707948243992607,
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 2375
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 4.4,
|
| 710 |
-
"grad_norm":
|
| 711 |
"learning_rate": 0.000566173752310536,
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 2400
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 4.44,
|
| 717 |
-
"grad_norm": 0.
|
| 718 |
"learning_rate": 0.0005615526802218115,
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 2425
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 4.49,
|
| 724 |
-
"grad_norm": 0.
|
| 725 |
"learning_rate": 0.0005569316081330868,
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 2450
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 4.53,
|
| 731 |
-
"grad_norm": 0.
|
| 732 |
"learning_rate": 0.0005523105360443623,
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2475
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 4.58,
|
| 738 |
-
"grad_norm":
|
| 739 |
"learning_rate": 0.0005476894639556377,
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 2500
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 4.62,
|
| 745 |
-
"grad_norm": 0.
|
| 746 |
"learning_rate": 0.0005430683918669131,
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 2525
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 4.67,
|
| 752 |
-
"grad_norm": 0.
|
| 753 |
"learning_rate": 0.0005384473197781885,
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 2550
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 4.72,
|
| 759 |
-
"grad_norm": 0.
|
| 760 |
"learning_rate": 0.0005338262476894639,
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 2575
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 4.76,
|
| 766 |
-
"grad_norm": 0.
|
| 767 |
"learning_rate": 0.0005292051756007393,
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 2600
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 4.81,
|
| 773 |
-
"grad_norm": 0.
|
| 774 |
"learning_rate": 0.0005245841035120147,
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 2625
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 4.85,
|
| 780 |
-
"grad_norm": 0.
|
| 781 |
"learning_rate": 0.0005199630314232903,
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 2650
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 4.9,
|
| 787 |
-
"grad_norm": 0.
|
| 788 |
"learning_rate": 0.0005153419593345657,
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 2675
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 4.95,
|
| 794 |
-
"grad_norm": 0.
|
| 795 |
"learning_rate": 0.0005107208872458411,
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 2700
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 4.99,
|
| 801 |
-
"grad_norm": 0.
|
| 802 |
"learning_rate": 0.0005060998151571165,
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 2725
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 5.0,
|
| 808 |
-
"eval_loss": 0.
|
| 809 |
-
"eval_runtime":
|
| 810 |
-
"eval_samples_per_second": 4.
|
| 811 |
-
"eval_steps_per_second": 0.
|
| 812 |
"step": 2730
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 5.04,
|
| 816 |
-
"grad_norm": 0.
|
| 817 |
"learning_rate": 0.0005014787430683919,
|
| 818 |
-
"loss": 0.
|
| 819 |
"step": 2750
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 5.08,
|
| 823 |
-
"grad_norm": 0.
|
| 824 |
"learning_rate": 0.0004968576709796673,
|
| 825 |
-
"loss": 0.
|
| 826 |
"step": 2775
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 5.13,
|
| 830 |
-
"grad_norm": 0.
|
| 831 |
"learning_rate": 0.0004922365988909427,
|
| 832 |
-
"loss": 0.
|
| 833 |
"step": 2800
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 5.17,
|
| 837 |
-
"grad_norm": 0.
|
| 838 |
"learning_rate": 0.0004876155268022181,
|
| 839 |
-
"loss": 0.
|
| 840 |
"step": 2825
|
| 841 |
},
|
| 842 |
{
|
| 843 |
"epoch": 5.22,
|
| 844 |
-
"grad_norm": 0.
|
| 845 |
"learning_rate": 0.0004829944547134935,
|
| 846 |
-
"loss": 0.
|
| 847 |
"step": 2850
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 5.27,
|
| 851 |
-
"grad_norm": 0.
|
| 852 |
"learning_rate": 0.000478373382624769,
|
| 853 |
-
"loss": 0.
|
| 854 |
"step": 2875
|
| 855 |
},
|
| 856 |
{
|
| 857 |
"epoch": 5.31,
|
| 858 |
-
"grad_norm": 0.
|
| 859 |
"learning_rate": 0.0004737523105360444,
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 2900
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 5.36,
|
| 865 |
-
"grad_norm": 0.
|
| 866 |
"learning_rate": 0.0004691312384473198,
|
| 867 |
-
"loss": 0.
|
| 868 |
"step": 2925
|
| 869 |
},
|
| 870 |
{
|
| 871 |
"epoch": 5.4,
|
| 872 |
-
"grad_norm":
|
| 873 |
"learning_rate": 0.0004645101663585952,
|
| 874 |
-
"loss": 0.
|
| 875 |
"step": 2950
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 5.45,
|
| 879 |
-
"grad_norm": 0.
|
| 880 |
"learning_rate": 0.0004598890942698706,
|
| 881 |
-
"loss": 0.
|
| 882 |
"step": 2975
|
| 883 |
},
|
| 884 |
{
|
| 885 |
"epoch": 5.49,
|
| 886 |
-
"grad_norm": 0.
|
| 887 |
"learning_rate": 0.00045526802218114607,
|
| 888 |
-
"loss": 0.
|
| 889 |
"step": 3000
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 5.54,
|
| 893 |
-
"grad_norm": 0.
|
| 894 |
"learning_rate": 0.0004506469500924215,
|
| 895 |
-
"loss": 0.
|
| 896 |
"step": 3025
|
| 897 |
},
|
| 898 |
{
|
| 899 |
"epoch": 5.59,
|
| 900 |
-
"grad_norm": 0.
|
| 901 |
"learning_rate": 0.0004460258780036969,
|
| 902 |
-
"loss": 0.
|
| 903 |
"step": 3050
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 5.63,
|
| 907 |
-
"grad_norm": 0.
|
| 908 |
"learning_rate": 0.0004414048059149723,
|
| 909 |
-
"loss": 0.
|
| 910 |
"step": 3075
|
| 911 |
},
|
| 912 |
{
|
| 913 |
"epoch": 5.68,
|
| 914 |
-
"grad_norm": 0.
|
| 915 |
"learning_rate": 0.0004367837338262477,
|
| 916 |
-
"loss": 0.
|
| 917 |
"step": 3100
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 5.72,
|
| 921 |
-
"grad_norm": 0.
|
| 922 |
"learning_rate": 0.0004321626617375231,
|
| 923 |
-
"loss": 0.
|
| 924 |
"step": 3125
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 5.77,
|
| 928 |
-
"grad_norm": 0.
|
| 929 |
"learning_rate": 0.0004275415896487985,
|
| 930 |
-
"loss": 0.
|
| 931 |
"step": 3150
|
| 932 |
},
|
| 933 |
{
|
| 934 |
"epoch": 5.82,
|
| 935 |
-
"grad_norm": 0.
|
| 936 |
"learning_rate": 0.0004229205175600739,
|
| 937 |
"loss": 0.0127,
|
| 938 |
"step": 3175
|
| 939 |
},
|
| 940 |
{
|
| 941 |
"epoch": 5.86,
|
| 942 |
-
"grad_norm": 0.
|
| 943 |
"learning_rate": 0.00041829944547134933,
|
| 944 |
-
"loss": 0.
|
| 945 |
"step": 3200
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 5.91,
|
| 949 |
-
"grad_norm": 0.
|
| 950 |
"learning_rate": 0.00041367837338262474,
|
| 951 |
-
"loss": 0.
|
| 952 |
"step": 3225
|
| 953 |
},
|
| 954 |
{
|
| 955 |
"epoch": 5.95,
|
| 956 |
-
"grad_norm": 0.
|
| 957 |
"learning_rate": 0.0004090573012939002,
|
| 958 |
-
"loss": 0.
|
| 959 |
"step": 3250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
"epoch": 6.0,
|
| 963 |
-
"grad_norm": 0.
|
| 964 |
"learning_rate": 0.0004044362292051756,
|
| 965 |
-
"loss": 0.
|
| 966 |
"step": 3275
|
| 967 |
},
|
| 968 |
{
|
| 969 |
"epoch": 6.0,
|
| 970 |
-
"eval_loss": 0.
|
| 971 |
-
"eval_runtime":
|
| 972 |
-
"eval_samples_per_second": 4.
|
| 973 |
-
"eval_steps_per_second": 0.
|
| 974 |
"step": 3276
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 6.04,
|
| 978 |
-
"grad_norm": 0.
|
| 979 |
"learning_rate": 0.000399815157116451,
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 3300
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 6.09,
|
| 985 |
-
"grad_norm": 0.
|
| 986 |
"learning_rate": 0.0003951940850277264,
|
| 987 |
-
"loss": 0.
|
| 988 |
"step": 3325
|
| 989 |
},
|
| 990 |
{
|
| 991 |
"epoch": 6.14,
|
| 992 |
-
"grad_norm": 0.
|
| 993 |
"learning_rate": 0.0003905730129390019,
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 3350
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 6.18,
|
| 999 |
-
"grad_norm": 0.
|
| 1000 |
"learning_rate": 0.0003859519408502773,
|
| 1001 |
-
"loss": 0.
|
| 1002 |
"step": 3375
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
"epoch": 6.23,
|
| 1006 |
-
"grad_norm": 0.
|
| 1007 |
"learning_rate": 0.0003813308687615527,
|
| 1008 |
-
"loss": 0.
|
| 1009 |
"step": 3400
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
"epoch": 6.27,
|
| 1013 |
-
"grad_norm": 0.
|
| 1014 |
"learning_rate": 0.0003767097966728281,
|
| 1015 |
-
"loss": 0.
|
| 1016 |
"step": 3425
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
"epoch": 6.32,
|
| 1020 |
-
"grad_norm": 0.
|
| 1021 |
"learning_rate": 0.0003720887245841035,
|
| 1022 |
-
"loss": 0.
|
| 1023 |
"step": 3450
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
"epoch": 6.36,
|
| 1027 |
-
"grad_norm": 0.
|
| 1028 |
"learning_rate": 0.0003674676524953789,
|
| 1029 |
-
"loss": 0.
|
| 1030 |
"step": 3475
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
"epoch": 6.41,
|
| 1034 |
-
"grad_norm": 0.
|
| 1035 |
"learning_rate": 0.0003628465804066544,
|
| 1036 |
-
"loss": 0.
|
| 1037 |
"step": 3500
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
"epoch": 6.46,
|
| 1041 |
-
"grad_norm": 0.
|
| 1042 |
"learning_rate": 0.0003582255083179298,
|
| 1043 |
-
"loss": 0.
|
| 1044 |
"step": 3525
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
"epoch": 6.5,
|
| 1048 |
-
"grad_norm": 0.
|
| 1049 |
"learning_rate": 0.0003536044362292052,
|
| 1050 |
-
"loss": 0.
|
| 1051 |
"step": 3550
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 6.55,
|
| 1055 |
-
"grad_norm": 0.
|
| 1056 |
"learning_rate": 0.0003489833641404806,
|
| 1057 |
-
"loss": 0.
|
| 1058 |
"step": 3575
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
"epoch": 6.59,
|
| 1062 |
-
"grad_norm": 0.
|
| 1063 |
"learning_rate": 0.000344362292051756,
|
| 1064 |
-
"loss": 0.
|
| 1065 |
"step": 3600
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
"epoch": 6.64,
|
| 1069 |
-
"grad_norm": 0.
|
| 1070 |
"learning_rate": 0.0003397412199630314,
|
| 1071 |
-
"loss": 0.
|
| 1072 |
"step": 3625
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
"epoch": 6.68,
|
| 1076 |
-
"grad_norm": 0.
|
| 1077 |
"learning_rate": 0.0003351201478743068,
|
| 1078 |
-
"loss": 0.
|
| 1079 |
"step": 3650
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 6.73,
|
| 1083 |
-
"grad_norm": 0.
|
| 1084 |
"learning_rate": 0.00033049907578558223,
|
| 1085 |
-
"loss": 0.
|
| 1086 |
"step": 3675
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
"epoch": 6.78,
|
| 1090 |
-
"grad_norm": 0.
|
| 1091 |
"learning_rate": 0.00032587800369685764,
|
| 1092 |
-
"loss": 0.
|
| 1093 |
"step": 3700
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
"epoch": 6.82,
|
| 1097 |
-
"grad_norm": 0.
|
| 1098 |
"learning_rate": 0.0003212569316081331,
|
| 1099 |
-
"loss": 0.
|
| 1100 |
"step": 3725
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 6.87,
|
| 1104 |
-
"grad_norm": 0.
|
| 1105 |
"learning_rate": 0.0003166358595194085,
|
| 1106 |
-
"loss": 0.
|
| 1107 |
"step": 3750
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
"epoch": 6.91,
|
| 1111 |
-
"grad_norm": 0.
|
| 1112 |
"learning_rate": 0.0003120147874306839,
|
| 1113 |
-
"loss": 0.
|
| 1114 |
"step": 3775
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
"epoch": 6.96,
|
| 1118 |
-
"grad_norm": 0.
|
| 1119 |
"learning_rate": 0.0003073937153419594,
|
| 1120 |
-
"loss": 0.
|
| 1121 |
"step": 3800
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
"epoch": 7.0,
|
| 1125 |
-
"eval_loss": 0.
|
| 1126 |
-
"eval_runtime":
|
| 1127 |
-
"eval_samples_per_second": 4.
|
| 1128 |
-
"eval_steps_per_second": 0.
|
| 1129 |
"step": 3822
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 7.01,
|
| 1133 |
-
"grad_norm": 0.
|
| 1134 |
"learning_rate": 0.0003027726432532348,
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 3825
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 7.05,
|
| 1140 |
-
"grad_norm": 0.
|
| 1141 |
"learning_rate": 0.0002981515711645102,
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 3850
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 7.1,
|
| 1147 |
-
"grad_norm": 0.
|
| 1148 |
"learning_rate": 0.0002935304990757856,
|
| 1149 |
-
"loss": 0.
|
| 1150 |
"step": 3875
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
"epoch": 7.14,
|
| 1154 |
-
"grad_norm": 0.
|
| 1155 |
"learning_rate": 0.000288909426987061,
|
| 1156 |
-
"loss": 0.
|
| 1157 |
"step": 3900
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 7.19,
|
| 1161 |
-
"grad_norm": 0.
|
| 1162 |
"learning_rate": 0.0002842883548983364,
|
| 1163 |
-
"loss": 0.
|
| 1164 |
"step": 3925
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 7.23,
|
| 1168 |
-
"grad_norm": 0.
|
| 1169 |
"learning_rate": 0.0002796672828096118,
|
| 1170 |
-
"loss": 0.
|
| 1171 |
"step": 3950
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 7.28,
|
| 1175 |
-
"grad_norm": 0.
|
| 1176 |
"learning_rate": 0.0002750462107208873,
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 3975
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 7.33,
|
| 1182 |
-
"grad_norm": 0.
|
| 1183 |
"learning_rate": 0.0002704251386321627,
|
| 1184 |
-
"loss": 0.
|
| 1185 |
"step": 4000
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 7.37,
|
| 1189 |
-
"grad_norm": 0.
|
| 1190 |
"learning_rate": 0.0002658040665434381,
|
| 1191 |
-
"loss": 0.
|
| 1192 |
"step": 4025
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 7.42,
|
| 1196 |
-
"grad_norm": 0.
|
| 1197 |
"learning_rate": 0.0002611829944547135,
|
| 1198 |
-
"loss": 0.
|
| 1199 |
"step": 4050
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 7.46,
|
| 1203 |
-
"grad_norm": 0.
|
| 1204 |
"learning_rate": 0.0002565619223659889,
|
| 1205 |
-
"loss": 0.
|
| 1206 |
"step": 4075
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 7.51,
|
| 1210 |
-
"grad_norm":
|
| 1211 |
"learning_rate": 0.0002519408502772643,
|
| 1212 |
-
"loss": 0.
|
| 1213 |
"step": 4100
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 7.55,
|
| 1217 |
-
"grad_norm": 0.
|
| 1218 |
"learning_rate": 0.0002473197781885397,
|
| 1219 |
-
"loss": 0.
|
| 1220 |
"step": 4125
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 7.6,
|
| 1224 |
-
"grad_norm": 0.
|
| 1225 |
"learning_rate": 0.0002426987060998152,
|
| 1226 |
-
"loss": 0.
|
| 1227 |
"step": 4150
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 7.65,
|
| 1231 |
-
"grad_norm": 0.
|
| 1232 |
"learning_rate": 0.0002380776340110906,
|
| 1233 |
-
"loss": 0.
|
| 1234 |
"step": 4175
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
"epoch": 7.69,
|
| 1238 |
-
"grad_norm": 0.
|
| 1239 |
"learning_rate": 0.000233456561922366,
|
| 1240 |
-
"loss": 0.
|
| 1241 |
"step": 4200
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 7.74,
|
| 1245 |
-
"grad_norm": 0.
|
| 1246 |
"learning_rate": 0.0002288354898336414,
|
| 1247 |
-
"loss": 0.
|
| 1248 |
"step": 4225
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
"epoch": 7.78,
|
| 1252 |
-
"grad_norm": 0.
|
| 1253 |
"learning_rate": 0.00022421441774491682,
|
| 1254 |
-
"loss": 0.
|
| 1255 |
"step": 4250
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
"epoch": 7.83,
|
| 1259 |
-
"grad_norm": 0.
|
| 1260 |
"learning_rate": 0.00021959334565619225,
|
| 1261 |
-
"loss": 0.
|
| 1262 |
"step": 4275
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
"epoch": 7.88,
|
| 1266 |
-
"grad_norm": 0.
|
| 1267 |
"learning_rate": 0.00021497227356746766,
|
| 1268 |
-
"loss": 0.
|
| 1269 |
"step": 4300
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 7.92,
|
| 1273 |
-
"grad_norm": 0.
|
| 1274 |
"learning_rate": 0.00021035120147874306,
|
| 1275 |
-
"loss": 0.
|
| 1276 |
"step": 4325
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
"epoch": 7.97,
|
| 1280 |
-
"grad_norm": 0.
|
| 1281 |
"learning_rate": 0.00020573012939001847,
|
| 1282 |
-
"loss": 0.
|
| 1283 |
"step": 4350
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
"epoch": 8.0,
|
| 1287 |
-
"eval_loss": 0.
|
| 1288 |
-
"eval_runtime":
|
| 1289 |
-
"eval_samples_per_second": 4.
|
| 1290 |
-
"eval_steps_per_second": 0.
|
| 1291 |
"step": 4368
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
"epoch": 8.01,
|
| 1295 |
-
"grad_norm": 0.
|
| 1296 |
"learning_rate": 0.00020110905730129388,
|
| 1297 |
-
"loss": 0.
|
| 1298 |
"step": 4375
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
"epoch": 8.06,
|
| 1302 |
-
"grad_norm": 0.
|
| 1303 |
"learning_rate": 0.00019648798521256934,
|
| 1304 |
-
"loss": 0.
|
| 1305 |
"step": 4400
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
"epoch": 8.1,
|
| 1309 |
-
"grad_norm": 0.
|
| 1310 |
"learning_rate": 0.00019186691312384475,
|
| 1311 |
-
"loss": 0.
|
| 1312 |
"step": 4425
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
"epoch": 8.15,
|
| 1316 |
-
"grad_norm": 0.
|
| 1317 |
"learning_rate": 0.00018724584103512016,
|
| 1318 |
-
"loss": 0.
|
| 1319 |
"step": 4450
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
"epoch": 8.2,
|
| 1323 |
-
"grad_norm": 0.
|
| 1324 |
"learning_rate": 0.00018262476894639556,
|
| 1325 |
-
"loss": 0.
|
| 1326 |
"step": 4475
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
"epoch": 8.24,
|
| 1330 |
-
"grad_norm": 0.
|
| 1331 |
"learning_rate": 0.00017800369685767097,
|
| 1332 |
-
"loss": 0.
|
| 1333 |
"step": 4500
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
"epoch": 8.29,
|
| 1337 |
-
"grad_norm": 0.
|
| 1338 |
"learning_rate": 0.0001733826247689464,
|
| 1339 |
-
"loss": 0.
|
| 1340 |
"step": 4525
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
"epoch": 8.33,
|
| 1344 |
-
"grad_norm": 0.
|
| 1345 |
"learning_rate": 0.0001687615526802218,
|
| 1346 |
-
"loss": 0.
|
| 1347 |
"step": 4550
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
"epoch": 8.38,
|
| 1351 |
-
"grad_norm": 0.
|
| 1352 |
"learning_rate": 0.00016414048059149722,
|
| 1353 |
-
"loss": 0.
|
| 1354 |
"step": 4575
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
"epoch": 8.42,
|
| 1358 |
-
"grad_norm": 0.
|
| 1359 |
"learning_rate": 0.00015951940850277263,
|
| 1360 |
-
"loss": 0.
|
| 1361 |
"step": 4600
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
"epoch": 8.47,
|
| 1365 |
-
"grad_norm": 0.
|
| 1366 |
"learning_rate": 0.0001548983364140481,
|
| 1367 |
-
"loss": 0.
|
| 1368 |
"step": 4625
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
"epoch": 8.52,
|
| 1372 |
-
"grad_norm": 0.
|
| 1373 |
"learning_rate": 0.0001502772643253235,
|
| 1374 |
-
"loss": 0.
|
| 1375 |
"step": 4650
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
"epoch": 8.56,
|
| 1379 |
-
"grad_norm": 0.
|
| 1380 |
"learning_rate": 0.0001456561922365989,
|
| 1381 |
-
"loss": 0.
|
| 1382 |
"step": 4675
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
"epoch": 8.61,
|
| 1386 |
-
"grad_norm": 0.
|
| 1387 |
"learning_rate": 0.0001410351201478743,
|
| 1388 |
-
"loss": 0.
|
| 1389 |
"step": 4700
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 8.65,
|
| 1393 |
-
"grad_norm": 0.
|
| 1394 |
"learning_rate": 0.00013641404805914972,
|
| 1395 |
-
"loss": 0.
|
| 1396 |
"step": 4725
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
"epoch": 8.7,
|
| 1400 |
-
"grad_norm": 0.
|
| 1401 |
"learning_rate": 0.00013179297597042515,
|
| 1402 |
-
"loss": 0.
|
| 1403 |
"step": 4750
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
"epoch": 8.75,
|
| 1407 |
-
"grad_norm": 0.
|
| 1408 |
"learning_rate": 0.00012717190388170056,
|
| 1409 |
-
"loss": 0.
|
| 1410 |
"step": 4775
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
"epoch": 8.79,
|
| 1414 |
-
"grad_norm": 0.
|
| 1415 |
"learning_rate": 0.00012255083179297597,
|
| 1416 |
-
"loss": 0.
|
| 1417 |
"step": 4800
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
"epoch": 8.84,
|
| 1421 |
-
"grad_norm": 0.
|
| 1422 |
"learning_rate": 0.00011792975970425139,
|
| 1423 |
-
"loss": 0.
|
| 1424 |
"step": 4825
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
"epoch": 8.88,
|
| 1428 |
-
"grad_norm": 0.
|
| 1429 |
"learning_rate": 0.00011330868761552681,
|
| 1430 |
-
"loss": 0.
|
| 1431 |
"step": 4850
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
"epoch": 8.93,
|
| 1435 |
-
"grad_norm": 0.
|
| 1436 |
"learning_rate": 0.00010868761552680221,
|
| 1437 |
-
"loss": 0.
|
| 1438 |
"step": 4875
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
"epoch": 8.97,
|
| 1442 |
-
"grad_norm": 0.
|
| 1443 |
"learning_rate": 0.00010406654343807764,
|
| 1444 |
-
"loss": 0.
|
| 1445 |
"step": 4900
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
"epoch": 9.0,
|
| 1449 |
-
"eval_loss": 0.
|
| 1450 |
-
"eval_runtime":
|
| 1451 |
-
"eval_samples_per_second": 4.
|
| 1452 |
-
"eval_steps_per_second": 0.
|
| 1453 |
"step": 4914
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
"epoch": 9.02,
|
| 1457 |
-
"grad_norm": 0.
|
| 1458 |
"learning_rate": 9.944547134935306e-05,
|
| 1459 |
-
"loss": 0.
|
| 1460 |
"step": 4925
|
| 1461 |
},
|
| 1462 |
{
|
| 1463 |
"epoch": 9.07,
|
| 1464 |
-
"grad_norm": 0.
|
| 1465 |
"learning_rate": 9.482439926062846e-05,
|
| 1466 |
"loss": 0.0013,
|
| 1467 |
"step": 4950
|
| 1468 |
},
|
| 1469 |
{
|
| 1470 |
"epoch": 9.11,
|
| 1471 |
-
"grad_norm": 0.
|
| 1472 |
"learning_rate": 9.020332717190388e-05,
|
| 1473 |
-
"loss": 0.
|
| 1474 |
"step": 4975
|
| 1475 |
},
|
| 1476 |
{
|
| 1477 |
"epoch": 9.16,
|
| 1478 |
-
"grad_norm": 0.
|
| 1479 |
"learning_rate": 8.558225508317929e-05,
|
| 1480 |
-
"loss": 0.
|
| 1481 |
"step": 5000
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
"epoch": 9.2,
|
| 1485 |
-
"grad_norm": 0.
|
| 1486 |
"learning_rate": 8.096118299445473e-05,
|
| 1487 |
-
"loss": 0.
|
| 1488 |
"step": 5025
|
| 1489 |
},
|
| 1490 |
{
|
| 1491 |
"epoch": 9.25,
|
| 1492 |
-
"grad_norm": 0.
|
| 1493 |
"learning_rate": 7.634011090573013e-05,
|
| 1494 |
-
"loss": 0.
|
| 1495 |
"step": 5050
|
| 1496 |
},
|
| 1497 |
{
|
| 1498 |
"epoch": 9.29,
|
| 1499 |
-
"grad_norm": 0.
|
| 1500 |
"learning_rate": 7.171903881700554e-05,
|
| 1501 |
-
"loss": 0.
|
| 1502 |
"step": 5075
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
"epoch": 9.34,
|
| 1506 |
-
"grad_norm": 0.
|
| 1507 |
"learning_rate": 6.709796672828096e-05,
|
| 1508 |
-
"loss": 0.
|
| 1509 |
"step": 5100
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
"epoch": 9.39,
|
| 1513 |
-
"grad_norm": 0.
|
| 1514 |
"learning_rate": 6.247689463955638e-05,
|
| 1515 |
-
"loss": 0.
|
| 1516 |
"step": 5125
|
| 1517 |
},
|
| 1518 |
{
|
| 1519 |
"epoch": 9.43,
|
| 1520 |
-
"grad_norm": 0.
|
| 1521 |
"learning_rate": 5.785582255083179e-05,
|
| 1522 |
-
"loss": 0.
|
| 1523 |
"step": 5150
|
| 1524 |
},
|
| 1525 |
{
|
| 1526 |
"epoch": 9.48,
|
| 1527 |
-
"grad_norm": 0.
|
| 1528 |
"learning_rate": 5.323475046210721e-05,
|
| 1529 |
-
"loss": 0.
|
| 1530 |
"step": 5175
|
| 1531 |
},
|
| 1532 |
{
|
| 1533 |
"epoch": 9.52,
|
| 1534 |
-
"grad_norm": 0.
|
| 1535 |
"learning_rate": 4.8613678373382625e-05,
|
| 1536 |
-
"loss": 0.
|
| 1537 |
"step": 5200
|
| 1538 |
},
|
| 1539 |
{
|
| 1540 |
"epoch": 9.57,
|
| 1541 |
-
"grad_norm": 0.
|
| 1542 |
"learning_rate": 4.3992606284658045e-05,
|
| 1543 |
-
"loss": 0.
|
| 1544 |
"step": 5225
|
| 1545 |
},
|
| 1546 |
{
|
| 1547 |
"epoch": 9.62,
|
| 1548 |
-
"grad_norm": 0.
|
| 1549 |
"learning_rate": 3.937153419593346e-05,
|
| 1550 |
"loss": 0.001,
|
| 1551 |
"step": 5250
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 9.66,
|
| 1555 |
-
"grad_norm": 0.
|
| 1556 |
"learning_rate": 3.4750462107208874e-05,
|
| 1557 |
-
"loss": 0.
|
| 1558 |
"step": 5275
|
| 1559 |
},
|
| 1560 |
{
|
| 1561 |
"epoch": 9.71,
|
| 1562 |
-
"grad_norm": 0.
|
| 1563 |
"learning_rate": 3.012939001848429e-05,
|
| 1564 |
-
"loss": 0.
|
| 1565 |
"step": 5300
|
| 1566 |
},
|
| 1567 |
{
|
| 1568 |
"epoch": 9.75,
|
| 1569 |
-
"grad_norm": 0.
|
| 1570 |
"learning_rate": 2.5508317929759705e-05,
|
| 1571 |
-
"loss": 0.
|
| 1572 |
"step": 5325
|
| 1573 |
},
|
| 1574 |
{
|
| 1575 |
"epoch": 9.8,
|
| 1576 |
-
"grad_norm": 0.
|
| 1577 |
"learning_rate": 2.088724584103512e-05,
|
| 1578 |
-
"loss": 0.
|
| 1579 |
"step": 5350
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
"epoch": 9.84,
|
| 1583 |
-
"grad_norm": 0.
|
| 1584 |
"learning_rate": 1.6266173752310537e-05,
|
| 1585 |
-
"loss": 0.
|
| 1586 |
"step": 5375
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
"epoch": 9.89,
|
| 1590 |
-
"grad_norm": 0.
|
| 1591 |
"learning_rate": 1.1645101663585952e-05,
|
| 1592 |
-
"loss": 0.
|
| 1593 |
"step": 5400
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
"epoch": 9.94,
|
| 1597 |
-
"grad_norm": 0.
|
| 1598 |
"learning_rate": 7.024029574861368e-06,
|
| 1599 |
-
"loss": 0.
|
| 1600 |
"step": 5425
|
| 1601 |
},
|
| 1602 |
{
|
| 1603 |
"epoch": 9.98,
|
| 1604 |
-
"grad_norm": 0.
|
| 1605 |
"learning_rate": 2.402957486136784e-06,
|
| 1606 |
-
"loss": 0.
|
| 1607 |
"step": 5450
|
| 1608 |
},
|
| 1609 |
{
|
| 1610 |
"epoch": 10.0,
|
| 1611 |
-
"eval_loss": 0.
|
| 1612 |
-
"eval_runtime":
|
| 1613 |
-
"eval_samples_per_second": 4.
|
| 1614 |
-
"eval_steps_per_second": 0.
|
| 1615 |
"step": 5460
|
| 1616 |
},
|
| 1617 |
{
|
| 1618 |
"epoch": 10.0,
|
| 1619 |
"step": 5460,
|
| 1620 |
"total_flos": 9.7789895073792e+18,
|
| 1621 |
-
"train_loss": 0.
|
| 1622 |
-
"train_runtime":
|
| 1623 |
-
"train_samples_per_second": 3.
|
| 1624 |
"train_steps_per_second": 0.509
|
| 1625 |
}
|
| 1626 |
],
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 0.4075450599193573,
|
| 14 |
+
"learning_rate": 0.0005,
|
| 15 |
+
"loss": 0.0571,
|
| 16 |
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09,
|
| 20 |
+
"grad_norm": 0.5959680676460266,
|
| 21 |
+
"learning_rate": 0.001,
|
| 22 |
+
"loss": 0.0853,
|
| 23 |
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.14,
|
| 27 |
+
"grad_norm": 1.1371592283248901,
|
| 28 |
+
"learning_rate": 0.0009955637707948243,
|
| 29 |
+
"loss": 0.0983,
|
| 30 |
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.18,
|
| 34 |
+
"grad_norm": 1.7857468128204346,
|
| 35 |
+
"learning_rate": 0.0009911275415896488,
|
| 36 |
+
"loss": 0.0689,
|
| 37 |
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.23,
|
| 41 |
+
"grad_norm": 1.947630763053894,
|
| 42 |
+
"learning_rate": 0.0009865064695009243,
|
| 43 |
+
"loss": 0.0834,
|
| 44 |
"step": 125
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.27,
|
| 48 |
+
"grad_norm": 0.750166654586792,
|
| 49 |
+
"learning_rate": 0.0009818853974121996,
|
| 50 |
+
"loss": 0.0655,
|
| 51 |
"step": 150
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32,
|
| 55 |
+
"grad_norm": 3.860727548599243,
|
| 56 |
+
"learning_rate": 0.000977264325323475,
|
| 57 |
+
"loss": 0.0554,
|
| 58 |
"step": 175
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.37,
|
| 62 |
+
"grad_norm": 0.3753944933414459,
|
| 63 |
+
"learning_rate": 0.0009726432532347505,
|
| 64 |
+
"loss": 0.0785,
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.41,
|
| 69 |
+
"grad_norm": 0.4372863471508026,
|
| 70 |
+
"learning_rate": 0.0009680221811460259,
|
| 71 |
+
"loss": 0.063,
|
| 72 |
"step": 225
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.46,
|
| 76 |
+
"grad_norm": 0.31646546721458435,
|
| 77 |
+
"learning_rate": 0.0009634011090573013,
|
| 78 |
+
"loss": 0.0487,
|
| 79 |
"step": 250
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5,
|
| 83 |
+
"grad_norm": 0.8565055131912231,
|
| 84 |
+
"learning_rate": 0.0009587800369685768,
|
| 85 |
+
"loss": 0.0586,
|
| 86 |
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.55,
|
| 90 |
+
"grad_norm": 0.5980587601661682,
|
| 91 |
+
"learning_rate": 0.0009541589648798521,
|
| 92 |
+
"loss": 0.0683,
|
| 93 |
"step": 300
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.6,
|
| 97 |
+
"grad_norm": 0.2764056324958801,
|
| 98 |
+
"learning_rate": 0.0009495378927911276,
|
| 99 |
+
"loss": 0.0697,
|
| 100 |
"step": 325
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.64,
|
| 104 |
+
"grad_norm": 1.5360766649246216,
|
| 105 |
+
"learning_rate": 0.0009449168207024029,
|
| 106 |
+
"loss": 0.0639,
|
| 107 |
"step": 350
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.69,
|
| 111 |
+
"grad_norm": 0.38272273540496826,
|
| 112 |
+
"learning_rate": 0.0009402957486136784,
|
| 113 |
+
"loss": 0.0816,
|
| 114 |
"step": 375
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.73,
|
| 118 |
+
"grad_norm": 0.362632155418396,
|
| 119 |
+
"learning_rate": 0.0009356746765249538,
|
| 120 |
+
"loss": 0.0734,
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.78,
|
| 125 |
+
"grad_norm": 1.5462536811828613,
|
| 126 |
+
"learning_rate": 0.0009310536044362292,
|
| 127 |
+
"loss": 0.2255,
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.82,
|
| 132 |
+
"grad_norm": 0.6713312268257141,
|
| 133 |
+
"learning_rate": 0.0009264325323475047,
|
| 134 |
+
"loss": 0.1296,
|
| 135 |
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.87,
|
| 139 |
+
"grad_norm": 0.8870647549629211,
|
| 140 |
"learning_rate": 0.0009219963031423291,
|
| 141 |
+
"loss": 0.0904,
|
| 142 |
"step": 475
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.92,
|
| 146 |
+
"grad_norm": 0.4918694496154785,
|
| 147 |
"learning_rate": 0.0009173752310536044,
|
| 148 |
+
"loss": 0.0689,
|
| 149 |
"step": 500
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.96,
|
| 153 |
+
"grad_norm": 0.3674885630607605,
|
| 154 |
"learning_rate": 0.0009127541589648799,
|
| 155 |
+
"loss": 0.0726,
|
| 156 |
"step": 525
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 1.0,
|
| 160 |
+
"eval_loss": 0.2210056632757187,
|
| 161 |
+
"eval_runtime": 173.4286,
|
| 162 |
+
"eval_samples_per_second": 4.688,
|
| 163 |
+
"eval_steps_per_second": 0.784,
|
| 164 |
"step": 546
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 1.01,
|
| 168 |
+
"grad_norm": 0.8691617250442505,
|
| 169 |
"learning_rate": 0.0009081330868761552,
|
| 170 |
+
"loss": 0.0421,
|
| 171 |
"step": 550
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.05,
|
| 175 |
+
"grad_norm": 0.8861550688743591,
|
| 176 |
"learning_rate": 0.0009035120147874307,
|
| 177 |
+
"loss": 0.058,
|
| 178 |
"step": 575
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 1.1,
|
| 182 |
+
"grad_norm": 0.10453300923109055,
|
| 183 |
"learning_rate": 0.000898890942698706,
|
| 184 |
+
"loss": 0.0391,
|
| 185 |
"step": 600
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.14,
|
| 189 |
+
"grad_norm": 0.11498710513114929,
|
| 190 |
"learning_rate": 0.0008942698706099815,
|
| 191 |
+
"loss": 0.0413,
|
| 192 |
"step": 625
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.19,
|
| 196 |
+
"grad_norm": 0.3542003333568573,
|
| 197 |
"learning_rate": 0.0008896487985212569,
|
| 198 |
+
"loss": 0.0764,
|
| 199 |
"step": 650
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 1.24,
|
| 203 |
+
"grad_norm": 0.5665566921234131,
|
| 204 |
"learning_rate": 0.0008850277264325323,
|
| 205 |
+
"loss": 0.0464,
|
| 206 |
"step": 675
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.28,
|
| 210 |
+
"grad_norm": 0.37183037400245667,
|
| 211 |
"learning_rate": 0.0008804066543438077,
|
| 212 |
+
"loss": 0.0557,
|
| 213 |
"step": 700
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 1.33,
|
| 217 |
+
"grad_norm": 2.33689546585083,
|
| 218 |
"learning_rate": 0.0008757855822550833,
|
| 219 |
+
"loss": 0.0723,
|
| 220 |
"step": 725
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 1.37,
|
| 224 |
+
"grad_norm": 0.47746214270591736,
|
| 225 |
"learning_rate": 0.0008711645101663586,
|
| 226 |
+
"loss": 0.0613,
|
| 227 |
"step": 750
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 1.42,
|
| 231 |
+
"grad_norm": 0.5573539137840271,
|
| 232 |
"learning_rate": 0.0008665434380776341,
|
| 233 |
+
"loss": 0.0581,
|
| 234 |
"step": 775
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 1.47,
|
| 238 |
+
"grad_norm": 0.5228638648986816,
|
| 239 |
"learning_rate": 0.0008619223659889095,
|
| 240 |
+
"loss": 0.0438,
|
| 241 |
"step": 800
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 1.51,
|
| 245 |
+
"grad_norm": 1.0585103034973145,
|
| 246 |
"learning_rate": 0.0008573012939001849,
|
| 247 |
+
"loss": 0.0357,
|
| 248 |
"step": 825
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 1.56,
|
| 252 |
+
"grad_norm": 0.13868175446987152,
|
| 253 |
"learning_rate": 0.0008526802218114603,
|
| 254 |
+
"loss": 0.0374,
|
| 255 |
"step": 850
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 1.6,
|
| 259 |
+
"grad_norm": 0.43853959441185,
|
| 260 |
"learning_rate": 0.0008480591497227357,
|
| 261 |
+
"loss": 0.0482,
|
| 262 |
"step": 875
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 1.65,
|
| 266 |
+
"grad_norm": 0.47208574414253235,
|
| 267 |
"learning_rate": 0.0008434380776340112,
|
| 268 |
+
"loss": 0.0551,
|
| 269 |
"step": 900
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.69,
|
| 273 |
+
"grad_norm": 0.2631681561470032,
|
| 274 |
"learning_rate": 0.0008388170055452865,
|
| 275 |
+
"loss": 0.0717,
|
| 276 |
"step": 925
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.74,
|
| 280 |
+
"grad_norm": 0.23163950443267822,
|
| 281 |
"learning_rate": 0.000834195933456562,
|
| 282 |
+
"loss": 0.0415,
|
| 283 |
"step": 950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 1.79,
|
| 287 |
+
"grad_norm": 0.45487725734710693,
|
| 288 |
"learning_rate": 0.0008295748613678373,
|
| 289 |
+
"loss": 0.0392,
|
| 290 |
"step": 975
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.83,
|
| 294 |
+
"grad_norm": 0.40454888343811035,
|
| 295 |
"learning_rate": 0.0008249537892791128,
|
| 296 |
+
"loss": 0.0342,
|
| 297 |
"step": 1000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 1.88,
|
| 301 |
+
"grad_norm": 0.10719649493694305,
|
| 302 |
"learning_rate": 0.0008203327171903881,
|
| 303 |
+
"loss": 0.0499,
|
| 304 |
"step": 1025
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 1.92,
|
| 308 |
+
"grad_norm": 0.5795795917510986,
|
| 309 |
"learning_rate": 0.0008157116451016636,
|
| 310 |
+
"loss": 0.0553,
|
| 311 |
"step": 1050
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 1.97,
|
| 315 |
+
"grad_norm": 0.2069532871246338,
|
| 316 |
"learning_rate": 0.000811090573012939,
|
| 317 |
+
"loss": 0.0419,
|
| 318 |
"step": 1075
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 2.0,
|
| 322 |
+
"eval_loss": 0.21386997401714325,
|
| 323 |
+
"eval_runtime": 174.0952,
|
| 324 |
+
"eval_samples_per_second": 4.67,
|
| 325 |
"eval_steps_per_second": 0.781,
|
| 326 |
"step": 1092
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 2.01,
|
| 330 |
+
"grad_norm": 0.170976459980011,
|
| 331 |
"learning_rate": 0.0008064695009242144,
|
| 332 |
+
"loss": 0.0373,
|
| 333 |
"step": 1100
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 2.06,
|
| 337 |
+
"grad_norm": 0.10965342819690704,
|
| 338 |
"learning_rate": 0.0008018484288354898,
|
| 339 |
+
"loss": 0.0286,
|
| 340 |
"step": 1125
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 2.11,
|
| 344 |
+
"grad_norm": 0.02158469147980213,
|
| 345 |
"learning_rate": 0.0007972273567467652,
|
| 346 |
+
"loss": 0.05,
|
| 347 |
"step": 1150
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 2.15,
|
| 351 |
+
"grad_norm": 1.0225136280059814,
|
| 352 |
"learning_rate": 0.0007926062846580406,
|
| 353 |
+
"loss": 0.0423,
|
| 354 |
"step": 1175
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 2.2,
|
| 358 |
+
"grad_norm": 0.09866318106651306,
|
| 359 |
"learning_rate": 0.0007879852125693162,
|
| 360 |
+
"loss": 0.0376,
|
| 361 |
"step": 1200
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 2.24,
|
| 365 |
+
"grad_norm": 0.23199380934238434,
|
| 366 |
"learning_rate": 0.0007833641404805915,
|
| 367 |
+
"loss": 0.0293,
|
| 368 |
"step": 1225
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 2.29,
|
| 372 |
+
"grad_norm": 0.05752483755350113,
|
| 373 |
"learning_rate": 0.000778743068391867,
|
| 374 |
+
"loss": 0.0381,
|
| 375 |
"step": 1250
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.34,
|
| 379 |
+
"grad_norm": 0.13506996631622314,
|
| 380 |
"learning_rate": 0.0007741219963031424,
|
| 381 |
+
"loss": 0.0394,
|
| 382 |
"step": 1275
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 2.38,
|
| 386 |
+
"grad_norm": 1.1013309955596924,
|
| 387 |
"learning_rate": 0.0007695009242144178,
|
| 388 |
+
"loss": 0.0394,
|
| 389 |
"step": 1300
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 2.43,
|
| 393 |
+
"grad_norm": 0.43956679105758667,
|
| 394 |
"learning_rate": 0.0007648798521256932,
|
| 395 |
+
"loss": 0.0459,
|
| 396 |
"step": 1325
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 2.47,
|
| 400 |
+
"grad_norm": 0.39061295986175537,
|
| 401 |
"learning_rate": 0.0007602587800369686,
|
| 402 |
+
"loss": 0.037,
|
| 403 |
"step": 1350
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 2.52,
|
| 407 |
+
"grad_norm": 0.2657981216907501,
|
| 408 |
"learning_rate": 0.0007556377079482441,
|
| 409 |
+
"loss": 0.0327,
|
| 410 |
"step": 1375
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 2.56,
|
| 414 |
+
"grad_norm": 0.4138255715370178,
|
| 415 |
"learning_rate": 0.0007510166358595194,
|
| 416 |
+
"loss": 0.0307,
|
| 417 |
"step": 1400
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 2.61,
|
| 421 |
+
"grad_norm": 0.32367995381355286,
|
| 422 |
"learning_rate": 0.0007463955637707949,
|
| 423 |
+
"loss": 0.0335,
|
| 424 |
"step": 1425
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 2.66,
|
| 428 |
+
"grad_norm": 0.5355994701385498,
|
| 429 |
"learning_rate": 0.0007417744916820702,
|
| 430 |
+
"loss": 0.0262,
|
| 431 |
"step": 1450
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 2.7,
|
| 435 |
+
"grad_norm": 3.182929039001465,
|
| 436 |
"learning_rate": 0.0007371534195933457,
|
| 437 |
+
"loss": 0.0302,
|
| 438 |
"step": 1475
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 2.75,
|
| 442 |
+
"grad_norm": 0.9068237543106079,
|
| 443 |
"learning_rate": 0.000732532347504621,
|
| 444 |
+
"loss": 0.0318,
|
| 445 |
"step": 1500
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 2.79,
|
| 449 |
+
"grad_norm": 0.804796576499939,
|
| 450 |
"learning_rate": 0.0007279112754158965,
|
| 451 |
+
"loss": 0.0462,
|
| 452 |
"step": 1525
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 2.84,
|
| 456 |
+
"grad_norm": 0.40627536177635193,
|
| 457 |
"learning_rate": 0.0007232902033271719,
|
| 458 |
+
"loss": 0.0226,
|
| 459 |
"step": 1550
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 2.88,
|
| 463 |
+
"grad_norm": 0.2852160632610321,
|
| 464 |
"learning_rate": 0.0007186691312384473,
|
| 465 |
+
"loss": 0.0327,
|
| 466 |
"step": 1575
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 2.93,
|
| 470 |
+
"grad_norm": 0.5738157629966736,
|
| 471 |
"learning_rate": 0.0007140480591497227,
|
| 472 |
+
"loss": 0.0317,
|
| 473 |
"step": 1600
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 2.98,
|
| 477 |
+
"grad_norm": 0.2782443165779114,
|
| 478 |
"learning_rate": 0.0007094269870609981,
|
| 479 |
+
"loss": 0.0322,
|
| 480 |
"step": 1625
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 3.0,
|
| 484 |
+
"eval_loss": 0.1934811770915985,
|
| 485 |
+
"eval_runtime": 175.4238,
|
| 486 |
+
"eval_samples_per_second": 4.634,
|
| 487 |
+
"eval_steps_per_second": 0.775,
|
| 488 |
"step": 1638
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 3.02,
|
| 492 |
+
"grad_norm": 0.027267010882496834,
|
| 493 |
"learning_rate": 0.0007048059149722735,
|
| 494 |
+
"loss": 0.0248,
|
| 495 |
"step": 1650
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 3.07,
|
| 499 |
+
"grad_norm": 0.23983055353164673,
|
| 500 |
"learning_rate": 0.000700184842883549,
|
| 501 |
+
"loss": 0.0252,
|
| 502 |
"step": 1675
|
| 503 |
},
|
| 504 |
{
|
| 505 |
"epoch": 3.11,
|
| 506 |
+
"grad_norm": 0.03389419987797737,
|
| 507 |
"learning_rate": 0.0006955637707948245,
|
| 508 |
+
"loss": 0.0216,
|
| 509 |
"step": 1700
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 3.16,
|
| 513 |
+
"grad_norm": 2.448323965072632,
|
| 514 |
"learning_rate": 0.0006909426987060999,
|
| 515 |
+
"loss": 0.0402,
|
| 516 |
"step": 1725
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 3.21,
|
| 520 |
+
"grad_norm": 0.5986452102661133,
|
| 521 |
"learning_rate": 0.0006863216266173753,
|
| 522 |
+
"loss": 0.0349,
|
| 523 |
"step": 1750
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 3.25,
|
| 527 |
+
"grad_norm": 0.046656377613544464,
|
| 528 |
"learning_rate": 0.0006817005545286507,
|
| 529 |
+
"loss": 0.0179,
|
| 530 |
"step": 1775
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 3.3,
|
| 534 |
+
"grad_norm": 0.2432301789522171,
|
| 535 |
"learning_rate": 0.0006770794824399261,
|
| 536 |
+
"loss": 0.0261,
|
| 537 |
"step": 1800
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 3.34,
|
| 541 |
+
"grad_norm": 0.4144662022590637,
|
| 542 |
"learning_rate": 0.0006724584103512015,
|
| 543 |
+
"loss": 0.0256,
|
| 544 |
"step": 1825
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 3.39,
|
| 548 |
+
"grad_norm": 0.27171510457992554,
|
| 549 |
"learning_rate": 0.000667837338262477,
|
| 550 |
+
"loss": 0.0322,
|
| 551 |
"step": 1850
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 3.43,
|
| 555 |
+
"grad_norm": 0.1022319346666336,
|
| 556 |
"learning_rate": 0.0006632162661737523,
|
| 557 |
+
"loss": 0.0293,
|
| 558 |
"step": 1875
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.48,
|
| 562 |
+
"grad_norm": 0.16478094458580017,
|
| 563 |
"learning_rate": 0.0006585951940850278,
|
| 564 |
+
"loss": 0.0178,
|
| 565 |
"step": 1900
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 3.53,
|
| 569 |
+
"grad_norm": 0.1675555408000946,
|
| 570 |
"learning_rate": 0.0006539741219963031,
|
| 571 |
+
"loss": 0.0174,
|
| 572 |
"step": 1925
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 3.57,
|
| 576 |
+
"grad_norm": 0.39023590087890625,
|
| 577 |
"learning_rate": 0.0006493530499075786,
|
| 578 |
+
"loss": 0.0149,
|
| 579 |
"step": 1950
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 3.62,
|
| 583 |
+
"grad_norm": 0.025721503421664238,
|
| 584 |
"learning_rate": 0.0006447319778188539,
|
| 585 |
+
"loss": 0.0231,
|
| 586 |
"step": 1975
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 3.66,
|
| 590 |
+
"grad_norm": 0.3088337182998657,
|
| 591 |
"learning_rate": 0.0006401109057301294,
|
| 592 |
+
"loss": 0.0283,
|
| 593 |
"step": 2000
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 3.71,
|
| 597 |
+
"grad_norm": 0.06729228049516678,
|
| 598 |
"learning_rate": 0.0006354898336414048,
|
| 599 |
+
"loss": 0.0204,
|
| 600 |
"step": 2025
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 3.75,
|
| 604 |
+
"grad_norm": 0.18552298843860626,
|
| 605 |
"learning_rate": 0.0006308687615526802,
|
| 606 |
+
"loss": 0.0274,
|
| 607 |
"step": 2050
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 3.8,
|
| 611 |
+
"grad_norm": 0.08045148104429245,
|
| 612 |
"learning_rate": 0.0006262476894639556,
|
| 613 |
+
"loss": 0.0218,
|
| 614 |
"step": 2075
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 3.85,
|
| 618 |
+
"grad_norm": 0.6443850994110107,
|
| 619 |
"learning_rate": 0.000621626617375231,
|
| 620 |
+
"loss": 0.0207,
|
| 621 |
"step": 2100
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 3.89,
|
| 625 |
+
"grad_norm": 0.6463542580604553,
|
| 626 |
"learning_rate": 0.0006170055452865064,
|
| 627 |
+
"loss": 0.0322,
|
| 628 |
"step": 2125
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 3.94,
|
| 632 |
+
"grad_norm": 0.2903934419155121,
|
| 633 |
"learning_rate": 0.000612384473197782,
|
| 634 |
+
"loss": 0.031,
|
| 635 |
"step": 2150
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 3.98,
|
| 639 |
+
"grad_norm": 0.1343035101890564,
|
| 640 |
"learning_rate": 0.0006077634011090574,
|
| 641 |
+
"loss": 0.0175,
|
| 642 |
"step": 2175
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 4.0,
|
| 646 |
+
"eval_loss": 0.1896440088748932,
|
| 647 |
+
"eval_runtime": 176.3159,
|
| 648 |
+
"eval_samples_per_second": 4.611,
|
| 649 |
+
"eval_steps_per_second": 0.771,
|
| 650 |
"step": 2184
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 4.03,
|
| 654 |
+
"grad_norm": 0.10466930270195007,
|
| 655 |
"learning_rate": 0.0006031423290203328,
|
| 656 |
+
"loss": 0.0215,
|
| 657 |
"step": 2200
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 4.08,
|
| 661 |
+
"grad_norm": 0.35988566279411316,
|
| 662 |
"learning_rate": 0.0005985212569316082,
|
| 663 |
+
"loss": 0.0193,
|
| 664 |
"step": 2225
|
| 665 |
},
|
| 666 |
{
|
| 667 |
"epoch": 4.12,
|
| 668 |
+
"grad_norm": 0.16410423815250397,
|
| 669 |
"learning_rate": 0.0005939001848428836,
|
| 670 |
+
"loss": 0.0143,
|
| 671 |
"step": 2250
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 4.17,
|
| 675 |
+
"grad_norm": 0.2650511562824249,
|
| 676 |
"learning_rate": 0.000589279112754159,
|
| 677 |
+
"loss": 0.0268,
|
| 678 |
"step": 2275
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 4.21,
|
| 682 |
+
"grad_norm": 0.2793768048286438,
|
| 683 |
"learning_rate": 0.0005846580406654344,
|
| 684 |
+
"loss": 0.0159,
|
| 685 |
"step": 2300
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 4.26,
|
| 689 |
+
"grad_norm": 2.7625114917755127,
|
| 690 |
"learning_rate": 0.0005800369685767099,
|
| 691 |
+
"loss": 0.0226,
|
| 692 |
"step": 2325
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 4.3,
|
| 696 |
+
"grad_norm": 0.45461520552635193,
|
| 697 |
"learning_rate": 0.0005754158964879852,
|
| 698 |
+
"loss": 0.0137,
|
| 699 |
"step": 2350
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 4.35,
|
| 703 |
+
"grad_norm": 0.28511613607406616,
|
| 704 |
"learning_rate": 0.0005707948243992607,
|
| 705 |
+
"loss": 0.0184,
|
| 706 |
"step": 2375
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 4.4,
|
| 710 |
+
"grad_norm": 0.5333670377731323,
|
| 711 |
"learning_rate": 0.000566173752310536,
|
| 712 |
+
"loss": 0.0186,
|
| 713 |
"step": 2400
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 4.44,
|
| 717 |
+
"grad_norm": 0.41222718358039856,
|
| 718 |
"learning_rate": 0.0005615526802218115,
|
| 719 |
+
"loss": 0.011,
|
| 720 |
"step": 2425
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 4.49,
|
| 724 |
+
"grad_norm": 0.27146583795547485,
|
| 725 |
"learning_rate": 0.0005569316081330868,
|
| 726 |
+
"loss": 0.0165,
|
| 727 |
"step": 2450
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 4.53,
|
| 731 |
+
"grad_norm": 0.29553595185279846,
|
| 732 |
"learning_rate": 0.0005523105360443623,
|
| 733 |
+
"loss": 0.0138,
|
| 734 |
"step": 2475
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 4.58,
|
| 738 |
+
"grad_norm": 0.13532432913780212,
|
| 739 |
"learning_rate": 0.0005476894639556377,
|
| 740 |
+
"loss": 0.0167,
|
| 741 |
"step": 2500
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 4.62,
|
| 745 |
+
"grad_norm": 0.10051342844963074,
|
| 746 |
"learning_rate": 0.0005430683918669131,
|
| 747 |
+
"loss": 0.0152,
|
| 748 |
"step": 2525
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 4.67,
|
| 752 |
+
"grad_norm": 0.023720353841781616,
|
| 753 |
"learning_rate": 0.0005384473197781885,
|
| 754 |
+
"loss": 0.0155,
|
| 755 |
"step": 2550
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 4.72,
|
| 759 |
+
"grad_norm": 0.2686695456504822,
|
| 760 |
"learning_rate": 0.0005338262476894639,
|
| 761 |
+
"loss": 0.0125,
|
| 762 |
"step": 2575
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 4.76,
|
| 766 |
+
"grad_norm": 0.33857473731040955,
|
| 767 |
"learning_rate": 0.0005292051756007393,
|
| 768 |
+
"loss": 0.0332,
|
| 769 |
"step": 2600
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 4.81,
|
| 773 |
+
"grad_norm": 0.0131806880235672,
|
| 774 |
"learning_rate": 0.0005245841035120147,
|
| 775 |
+
"loss": 0.014,
|
| 776 |
"step": 2625
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 4.85,
|
| 780 |
+
"grad_norm": 0.4342842698097229,
|
| 781 |
"learning_rate": 0.0005199630314232903,
|
| 782 |
+
"loss": 0.016,
|
| 783 |
"step": 2650
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 4.9,
|
| 787 |
+
"grad_norm": 0.005540889222174883,
|
| 788 |
"learning_rate": 0.0005153419593345657,
|
| 789 |
+
"loss": 0.0134,
|
| 790 |
"step": 2675
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 4.95,
|
| 794 |
+
"grad_norm": 0.004122666083276272,
|
| 795 |
"learning_rate": 0.0005107208872458411,
|
| 796 |
+
"loss": 0.0223,
|
| 797 |
"step": 2700
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 4.99,
|
| 801 |
+
"grad_norm": 0.14384405314922333,
|
| 802 |
"learning_rate": 0.0005060998151571165,
|
| 803 |
+
"loss": 0.0266,
|
| 804 |
"step": 2725
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 5.0,
|
| 808 |
+
"eval_loss": 0.19267761707305908,
|
| 809 |
+
"eval_runtime": 179.8301,
|
| 810 |
+
"eval_samples_per_second": 4.521,
|
| 811 |
+
"eval_steps_per_second": 0.756,
|
| 812 |
"step": 2730
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 5.04,
|
| 816 |
+
"grad_norm": 0.3819844126701355,
|
| 817 |
"learning_rate": 0.0005014787430683919,
|
| 818 |
+
"loss": 0.0166,
|
| 819 |
"step": 2750
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 5.08,
|
| 823 |
+
"grad_norm": 0.27138832211494446,
|
| 824 |
"learning_rate": 0.0004968576709796673,
|
| 825 |
+
"loss": 0.0056,
|
| 826 |
"step": 2775
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 5.13,
|
| 830 |
+
"grad_norm": 0.36033156514167786,
|
| 831 |
"learning_rate": 0.0004922365988909427,
|
| 832 |
+
"loss": 0.0084,
|
| 833 |
"step": 2800
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 5.17,
|
| 837 |
+
"grad_norm": 0.3422500789165497,
|
| 838 |
"learning_rate": 0.0004876155268022181,
|
| 839 |
+
"loss": 0.0089,
|
| 840 |
"step": 2825
|
| 841 |
},
|
| 842 |
{
|
| 843 |
"epoch": 5.22,
|
| 844 |
+
"grad_norm": 0.12272176891565323,
|
| 845 |
"learning_rate": 0.0004829944547134935,
|
| 846 |
+
"loss": 0.0079,
|
| 847 |
"step": 2850
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 5.27,
|
| 851 |
+
"grad_norm": 0.03446231782436371,
|
| 852 |
"learning_rate": 0.000478373382624769,
|
| 853 |
+
"loss": 0.011,
|
| 854 |
"step": 2875
|
| 855 |
},
|
| 856 |
{
|
| 857 |
"epoch": 5.31,
|
| 858 |
+
"grad_norm": 0.2042599320411682,
|
| 859 |
"learning_rate": 0.0004737523105360444,
|
| 860 |
+
"loss": 0.0091,
|
| 861 |
"step": 2900
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 5.36,
|
| 865 |
+
"grad_norm": 0.18888217210769653,
|
| 866 |
"learning_rate": 0.0004691312384473198,
|
| 867 |
+
"loss": 0.0146,
|
| 868 |
"step": 2925
|
| 869 |
},
|
| 870 |
{
|
| 871 |
"epoch": 5.4,
|
| 872 |
+
"grad_norm": 4.216693878173828,
|
| 873 |
"learning_rate": 0.0004645101663585952,
|
| 874 |
+
"loss": 0.0162,
|
| 875 |
"step": 2950
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 5.45,
|
| 879 |
+
"grad_norm": 0.20249082148075104,
|
| 880 |
"learning_rate": 0.0004598890942698706,
|
| 881 |
+
"loss": 0.0193,
|
| 882 |
"step": 2975
|
| 883 |
},
|
| 884 |
{
|
| 885 |
"epoch": 5.49,
|
| 886 |
+
"grad_norm": 0.37886273860931396,
|
| 887 |
"learning_rate": 0.00045526802218114607,
|
| 888 |
+
"loss": 0.0163,
|
| 889 |
"step": 3000
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 5.54,
|
| 893 |
+
"grad_norm": 0.24141408503055573,
|
| 894 |
"learning_rate": 0.0004506469500924215,
|
| 895 |
+
"loss": 0.0147,
|
| 896 |
"step": 3025
|
| 897 |
},
|
| 898 |
{
|
| 899 |
"epoch": 5.59,
|
| 900 |
+
"grad_norm": 0.23406554758548737,
|
| 901 |
"learning_rate": 0.0004460258780036969,
|
| 902 |
+
"loss": 0.0145,
|
| 903 |
"step": 3050
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 5.63,
|
| 907 |
+
"grad_norm": 0.355023056268692,
|
| 908 |
"learning_rate": 0.0004414048059149723,
|
| 909 |
+
"loss": 0.0144,
|
| 910 |
"step": 3075
|
| 911 |
},
|
| 912 |
{
|
| 913 |
"epoch": 5.68,
|
| 914 |
+
"grad_norm": 0.18628603219985962,
|
| 915 |
"learning_rate": 0.0004367837338262477,
|
| 916 |
+
"loss": 0.0105,
|
| 917 |
"step": 3100
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 5.72,
|
| 921 |
+
"grad_norm": 0.328931987285614,
|
| 922 |
"learning_rate": 0.0004321626617375231,
|
| 923 |
+
"loss": 0.0107,
|
| 924 |
"step": 3125
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 5.77,
|
| 928 |
+
"grad_norm": 0.004133810754865408,
|
| 929 |
"learning_rate": 0.0004275415896487985,
|
| 930 |
+
"loss": 0.01,
|
| 931 |
"step": 3150
|
| 932 |
},
|
| 933 |
{
|
| 934 |
"epoch": 5.82,
|
| 935 |
+
"grad_norm": 0.036314379423856735,
|
| 936 |
"learning_rate": 0.0004229205175600739,
|
| 937 |
"loss": 0.0127,
|
| 938 |
"step": 3175
|
| 939 |
},
|
| 940 |
{
|
| 941 |
"epoch": 5.86,
|
| 942 |
+
"grad_norm": 0.27704620361328125,
|
| 943 |
"learning_rate": 0.00041829944547134933,
|
| 944 |
+
"loss": 0.0111,
|
| 945 |
"step": 3200
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 5.91,
|
| 949 |
+
"grad_norm": 0.5109962821006775,
|
| 950 |
"learning_rate": 0.00041367837338262474,
|
| 951 |
+
"loss": 0.0157,
|
| 952 |
"step": 3225
|
| 953 |
},
|
| 954 |
{
|
| 955 |
"epoch": 5.95,
|
| 956 |
+
"grad_norm": 0.09048620611429214,
|
| 957 |
"learning_rate": 0.0004090573012939002,
|
| 958 |
+
"loss": 0.0184,
|
| 959 |
"step": 3250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
"epoch": 6.0,
|
| 963 |
+
"grad_norm": 0.010707640089094639,
|
| 964 |
"learning_rate": 0.0004044362292051756,
|
| 965 |
+
"loss": 0.0178,
|
| 966 |
"step": 3275
|
| 967 |
},
|
| 968 |
{
|
| 969 |
"epoch": 6.0,
|
| 970 |
+
"eval_loss": 0.20126062631607056,
|
| 971 |
+
"eval_runtime": 179.6758,
|
| 972 |
+
"eval_samples_per_second": 4.525,
|
| 973 |
+
"eval_steps_per_second": 0.757,
|
| 974 |
"step": 3276
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 6.04,
|
| 978 |
+
"grad_norm": 0.013095181435346603,
|
| 979 |
"learning_rate": 0.000399815157116451,
|
| 980 |
+
"loss": 0.0106,
|
| 981 |
"step": 3300
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 6.09,
|
| 985 |
+
"grad_norm": 0.15969859063625336,
|
| 986 |
"learning_rate": 0.0003951940850277264,
|
| 987 |
+
"loss": 0.0098,
|
| 988 |
"step": 3325
|
| 989 |
},
|
| 990 |
{
|
| 991 |
"epoch": 6.14,
|
| 992 |
+
"grad_norm": 0.09395785629749298,
|
| 993 |
"learning_rate": 0.0003905730129390019,
|
| 994 |
+
"loss": 0.012,
|
| 995 |
"step": 3350
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 6.18,
|
| 999 |
+
"grad_norm": 0.010071701370179653,
|
| 1000 |
"learning_rate": 0.0003859519408502773,
|
| 1001 |
+
"loss": 0.0071,
|
| 1002 |
"step": 3375
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
"epoch": 6.23,
|
| 1006 |
+
"grad_norm": 0.005003762431442738,
|
| 1007 |
"learning_rate": 0.0003813308687615527,
|
| 1008 |
+
"loss": 0.0093,
|
| 1009 |
"step": 3400
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
"epoch": 6.27,
|
| 1013 |
+
"grad_norm": 0.01751079224050045,
|
| 1014 |
"learning_rate": 0.0003767097966728281,
|
| 1015 |
+
"loss": 0.0093,
|
| 1016 |
"step": 3425
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
"epoch": 6.32,
|
| 1020 |
+
"grad_norm": 0.048858534544706345,
|
| 1021 |
"learning_rate": 0.0003720887245841035,
|
| 1022 |
+
"loss": 0.0092,
|
| 1023 |
"step": 3450
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
"epoch": 6.36,
|
| 1027 |
+
"grad_norm": 0.05492233484983444,
|
| 1028 |
"learning_rate": 0.0003674676524953789,
|
| 1029 |
+
"loss": 0.0137,
|
| 1030 |
"step": 3475
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
"epoch": 6.41,
|
| 1034 |
+
"grad_norm": 0.011647823266685009,
|
| 1035 |
"learning_rate": 0.0003628465804066544,
|
| 1036 |
+
"loss": 0.008,
|
| 1037 |
"step": 3500
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
"epoch": 6.46,
|
| 1041 |
+
"grad_norm": 0.02023889683187008,
|
| 1042 |
"learning_rate": 0.0003582255083179298,
|
| 1043 |
+
"loss": 0.0064,
|
| 1044 |
"step": 3525
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
"epoch": 6.5,
|
| 1048 |
+
"grad_norm": 0.1093795895576477,
|
| 1049 |
"learning_rate": 0.0003536044362292052,
|
| 1050 |
+
"loss": 0.0087,
|
| 1051 |
"step": 3550
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 6.55,
|
| 1055 |
+
"grad_norm": 0.33639025688171387,
|
| 1056 |
"learning_rate": 0.0003489833641404806,
|
| 1057 |
+
"loss": 0.0127,
|
| 1058 |
"step": 3575
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
"epoch": 6.59,
|
| 1062 |
+
"grad_norm": 0.08823797106742859,
|
| 1063 |
"learning_rate": 0.000344362292051756,
|
| 1064 |
+
"loss": 0.0112,
|
| 1065 |
"step": 3600
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
"epoch": 6.64,
|
| 1069 |
+
"grad_norm": 0.052434779703617096,
|
| 1070 |
"learning_rate": 0.0003397412199630314,
|
| 1071 |
+
"loss": 0.0116,
|
| 1072 |
"step": 3625
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
"epoch": 6.68,
|
| 1076 |
+
"grad_norm": 0.1535090208053589,
|
| 1077 |
"learning_rate": 0.0003351201478743068,
|
| 1078 |
+
"loss": 0.011,
|
| 1079 |
"step": 3650
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 6.73,
|
| 1083 |
+
"grad_norm": 0.2711283564567566,
|
| 1084 |
"learning_rate": 0.00033049907578558223,
|
| 1085 |
+
"loss": 0.007,
|
| 1086 |
"step": 3675
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
"epoch": 6.78,
|
| 1090 |
+
"grad_norm": 0.006919647566974163,
|
| 1091 |
"learning_rate": 0.00032587800369685764,
|
| 1092 |
+
"loss": 0.0098,
|
| 1093 |
"step": 3700
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
"epoch": 6.82,
|
| 1097 |
+
"grad_norm": 0.03872460126876831,
|
| 1098 |
"learning_rate": 0.0003212569316081331,
|
| 1099 |
+
"loss": 0.0092,
|
| 1100 |
"step": 3725
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 6.87,
|
| 1104 |
+
"grad_norm": 0.0396348237991333,
|
| 1105 |
"learning_rate": 0.0003166358595194085,
|
| 1106 |
+
"loss": 0.0126,
|
| 1107 |
"step": 3750
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
"epoch": 6.91,
|
| 1111 |
+
"grad_norm": 0.008865280076861382,
|
| 1112 |
"learning_rate": 0.0003120147874306839,
|
| 1113 |
+
"loss": 0.0097,
|
| 1114 |
"step": 3775
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
"epoch": 6.96,
|
| 1118 |
+
"grad_norm": 0.2857593894004822,
|
| 1119 |
"learning_rate": 0.0003073937153419594,
|
| 1120 |
+
"loss": 0.0081,
|
| 1121 |
"step": 3800
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
"epoch": 7.0,
|
| 1125 |
+
"eval_loss": 0.19787272810935974,
|
| 1126 |
+
"eval_runtime": 178.7624,
|
| 1127 |
+
"eval_samples_per_second": 4.548,
|
| 1128 |
+
"eval_steps_per_second": 0.761,
|
| 1129 |
"step": 3822
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 7.01,
|
| 1133 |
+
"grad_norm": 0.162245973944664,
|
| 1134 |
"learning_rate": 0.0003027726432532348,
|
| 1135 |
+
"loss": 0.0091,
|
| 1136 |
"step": 3825
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 7.05,
|
| 1140 |
+
"grad_norm": 0.08275479078292847,
|
| 1141 |
"learning_rate": 0.0002981515711645102,
|
| 1142 |
+
"loss": 0.0064,
|
| 1143 |
"step": 3850
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 7.1,
|
| 1147 |
+
"grad_norm": 0.04956310614943504,
|
| 1148 |
"learning_rate": 0.0002935304990757856,
|
| 1149 |
+
"loss": 0.0033,
|
| 1150 |
"step": 3875
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
"epoch": 7.14,
|
| 1154 |
+
"grad_norm": 0.2950696647167206,
|
| 1155 |
"learning_rate": 0.000288909426987061,
|
| 1156 |
+
"loss": 0.0059,
|
| 1157 |
"step": 3900
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 7.19,
|
| 1161 |
+
"grad_norm": 0.16667646169662476,
|
| 1162 |
"learning_rate": 0.0002842883548983364,
|
| 1163 |
+
"loss": 0.0065,
|
| 1164 |
"step": 3925
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 7.23,
|
| 1168 |
+
"grad_norm": 0.018928788602352142,
|
| 1169 |
"learning_rate": 0.0002796672828096118,
|
| 1170 |
+
"loss": 0.0053,
|
| 1171 |
"step": 3950
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 7.28,
|
| 1175 |
+
"grad_norm": 0.01914687640964985,
|
| 1176 |
"learning_rate": 0.0002750462107208873,
|
| 1177 |
+
"loss": 0.0058,
|
| 1178 |
"step": 3975
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 7.33,
|
| 1182 |
+
"grad_norm": 0.009565665386617184,
|
| 1183 |
"learning_rate": 0.0002704251386321627,
|
| 1184 |
+
"loss": 0.0042,
|
| 1185 |
"step": 4000
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 7.37,
|
| 1189 |
+
"grad_norm": 0.10117679834365845,
|
| 1190 |
"learning_rate": 0.0002658040665434381,
|
| 1191 |
+
"loss": 0.0081,
|
| 1192 |
"step": 4025
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 7.42,
|
| 1196 |
+
"grad_norm": 0.10825569927692413,
|
| 1197 |
"learning_rate": 0.0002611829944547135,
|
| 1198 |
+
"loss": 0.0088,
|
| 1199 |
"step": 4050
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 7.46,
|
| 1203 |
+
"grad_norm": 0.008808852173388004,
|
| 1204 |
"learning_rate": 0.0002565619223659889,
|
| 1205 |
+
"loss": 0.0052,
|
| 1206 |
"step": 4075
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 7.51,
|
| 1210 |
+
"grad_norm": 0.0186983160674572,
|
| 1211 |
"learning_rate": 0.0002519408502772643,
|
| 1212 |
+
"loss": 0.0051,
|
| 1213 |
"step": 4100
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 7.55,
|
| 1217 |
+
"grad_norm": 0.07354945689439774,
|
| 1218 |
"learning_rate": 0.0002473197781885397,
|
| 1219 |
+
"loss": 0.0055,
|
| 1220 |
"step": 4125
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 7.6,
|
| 1224 |
+
"grad_norm": 0.0021155644208192825,
|
| 1225 |
"learning_rate": 0.0002426987060998152,
|
| 1226 |
+
"loss": 0.0044,
|
| 1227 |
"step": 4150
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 7.65,
|
| 1231 |
+
"grad_norm": 0.08616074174642563,
|
| 1232 |
"learning_rate": 0.0002380776340110906,
|
| 1233 |
+
"loss": 0.0037,
|
| 1234 |
"step": 4175
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
"epoch": 7.69,
|
| 1238 |
+
"grad_norm": 0.009911403059959412,
|
| 1239 |
"learning_rate": 0.000233456561922366,
|
| 1240 |
+
"loss": 0.0073,
|
| 1241 |
"step": 4200
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 7.74,
|
| 1245 |
+
"grad_norm": 0.36762863397598267,
|
| 1246 |
"learning_rate": 0.0002288354898336414,
|
| 1247 |
+
"loss": 0.004,
|
| 1248 |
"step": 4225
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
"epoch": 7.78,
|
| 1252 |
+
"grad_norm": 0.0590713806450367,
|
| 1253 |
"learning_rate": 0.00022421441774491682,
|
| 1254 |
+
"loss": 0.0034,
|
| 1255 |
"step": 4250
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
"epoch": 7.83,
|
| 1259 |
+
"grad_norm": 0.0876949205994606,
|
| 1260 |
"learning_rate": 0.00021959334565619225,
|
| 1261 |
+
"loss": 0.0061,
|
| 1262 |
"step": 4275
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
"epoch": 7.88,
|
| 1266 |
+
"grad_norm": 0.2488565295934677,
|
| 1267 |
"learning_rate": 0.00021497227356746766,
|
| 1268 |
+
"loss": 0.0047,
|
| 1269 |
"step": 4300
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 7.92,
|
| 1273 |
+
"grad_norm": 0.16184526681900024,
|
| 1274 |
"learning_rate": 0.00021035120147874306,
|
| 1275 |
+
"loss": 0.0064,
|
| 1276 |
"step": 4325
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
"epoch": 7.97,
|
| 1280 |
+
"grad_norm": 0.025223182514309883,
|
| 1281 |
"learning_rate": 0.00020573012939001847,
|
| 1282 |
+
"loss": 0.0081,
|
| 1283 |
"step": 4350
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
"epoch": 8.0,
|
| 1287 |
+
"eval_loss": 0.21132159233093262,
|
| 1288 |
+
"eval_runtime": 178.6799,
|
| 1289 |
+
"eval_samples_per_second": 4.55,
|
| 1290 |
+
"eval_steps_per_second": 0.761,
|
| 1291 |
"step": 4368
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
"epoch": 8.01,
|
| 1295 |
+
"grad_norm": 0.04916756972670555,
|
| 1296 |
"learning_rate": 0.00020110905730129388,
|
| 1297 |
+
"loss": 0.0049,
|
| 1298 |
"step": 4375
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
"epoch": 8.06,
|
| 1302 |
+
"grad_norm": 0.010703769512474537,
|
| 1303 |
"learning_rate": 0.00019648798521256934,
|
| 1304 |
+
"loss": 0.0034,
|
| 1305 |
"step": 4400
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
"epoch": 8.1,
|
| 1309 |
+
"grad_norm": 0.004313566256314516,
|
| 1310 |
"learning_rate": 0.00019186691312384475,
|
| 1311 |
+
"loss": 0.003,
|
| 1312 |
"step": 4425
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
"epoch": 8.15,
|
| 1316 |
+
"grad_norm": 0.18936963379383087,
|
| 1317 |
"learning_rate": 0.00018724584103512016,
|
| 1318 |
+
"loss": 0.004,
|
| 1319 |
"step": 4450
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
"epoch": 8.2,
|
| 1323 |
+
"grad_norm": 0.0596047043800354,
|
| 1324 |
"learning_rate": 0.00018262476894639556,
|
| 1325 |
+
"loss": 0.0027,
|
| 1326 |
"step": 4475
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
"epoch": 8.24,
|
| 1330 |
+
"grad_norm": 0.0016723590670153499,
|
| 1331 |
"learning_rate": 0.00017800369685767097,
|
| 1332 |
+
"loss": 0.0037,
|
| 1333 |
"step": 4500
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
"epoch": 8.29,
|
| 1337 |
+
"grad_norm": 0.026407798752188683,
|
| 1338 |
"learning_rate": 0.0001733826247689464,
|
| 1339 |
+
"loss": 0.0026,
|
| 1340 |
"step": 4525
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
"epoch": 8.33,
|
| 1344 |
+
"grad_norm": 0.004466090817004442,
|
| 1345 |
"learning_rate": 0.0001687615526802218,
|
| 1346 |
+
"loss": 0.0044,
|
| 1347 |
"step": 4550
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
"epoch": 8.38,
|
| 1351 |
+
"grad_norm": 0.013297215104103088,
|
| 1352 |
"learning_rate": 0.00016414048059149722,
|
| 1353 |
+
"loss": 0.0044,
|
| 1354 |
"step": 4575
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
"epoch": 8.42,
|
| 1358 |
+
"grad_norm": 0.013365192338824272,
|
| 1359 |
"learning_rate": 0.00015951940850277263,
|
| 1360 |
+
"loss": 0.0033,
|
| 1361 |
"step": 4600
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
"epoch": 8.47,
|
| 1365 |
+
"grad_norm": 0.32592836022377014,
|
| 1366 |
"learning_rate": 0.0001548983364140481,
|
| 1367 |
+
"loss": 0.0056,
|
| 1368 |
"step": 4625
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
"epoch": 8.52,
|
| 1372 |
+
"grad_norm": 0.023310931399464607,
|
| 1373 |
"learning_rate": 0.0001502772643253235,
|
| 1374 |
+
"loss": 0.0017,
|
| 1375 |
"step": 4650
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
"epoch": 8.56,
|
| 1379 |
+
"grad_norm": 0.0938984677195549,
|
| 1380 |
"learning_rate": 0.0001456561922365989,
|
| 1381 |
+
"loss": 0.0028,
|
| 1382 |
"step": 4675
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
"epoch": 8.61,
|
| 1386 |
+
"grad_norm": 0.006782053969800472,
|
| 1387 |
"learning_rate": 0.0001410351201478743,
|
| 1388 |
+
"loss": 0.0019,
|
| 1389 |
"step": 4700
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 8.65,
|
| 1393 |
+
"grad_norm": 0.08395280689001083,
|
| 1394 |
"learning_rate": 0.00013641404805914972,
|
| 1395 |
+
"loss": 0.0024,
|
| 1396 |
"step": 4725
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
"epoch": 8.7,
|
| 1400 |
+
"grad_norm": 0.04261644929647446,
|
| 1401 |
"learning_rate": 0.00013179297597042515,
|
| 1402 |
+
"loss": 0.0029,
|
| 1403 |
"step": 4750
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
"epoch": 8.75,
|
| 1407 |
+
"grad_norm": 0.020602483302354813,
|
| 1408 |
"learning_rate": 0.00012717190388170056,
|
| 1409 |
+
"loss": 0.0025,
|
| 1410 |
"step": 4775
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
"epoch": 8.79,
|
| 1414 |
+
"grad_norm": 0.0013005019864067435,
|
| 1415 |
"learning_rate": 0.00012255083179297597,
|
| 1416 |
+
"loss": 0.0024,
|
| 1417 |
"step": 4800
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
"epoch": 8.84,
|
| 1421 |
+
"grad_norm": 0.0019000261090695858,
|
| 1422 |
"learning_rate": 0.00011792975970425139,
|
| 1423 |
+
"loss": 0.004,
|
| 1424 |
"step": 4825
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
"epoch": 8.88,
|
| 1428 |
+
"grad_norm": 0.02021609991788864,
|
| 1429 |
"learning_rate": 0.00011330868761552681,
|
| 1430 |
+
"loss": 0.0023,
|
| 1431 |
"step": 4850
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
"epoch": 8.93,
|
| 1435 |
+
"grad_norm": 0.012654704973101616,
|
| 1436 |
"learning_rate": 0.00010868761552680221,
|
| 1437 |
+
"loss": 0.0036,
|
| 1438 |
"step": 4875
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
"epoch": 8.97,
|
| 1442 |
+
"grad_norm": 0.009410886093974113,
|
| 1443 |
"learning_rate": 0.00010406654343807764,
|
| 1444 |
+
"loss": 0.0018,
|
| 1445 |
"step": 4900
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
"epoch": 9.0,
|
| 1449 |
+
"eval_loss": 0.2146490514278412,
|
| 1450 |
+
"eval_runtime": 176.7137,
|
| 1451 |
+
"eval_samples_per_second": 4.601,
|
| 1452 |
+
"eval_steps_per_second": 0.77,
|
| 1453 |
"step": 4914
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
"epoch": 9.02,
|
| 1457 |
+
"grad_norm": 0.00876565556973219,
|
| 1458 |
"learning_rate": 9.944547134935306e-05,
|
| 1459 |
+
"loss": 0.0024,
|
| 1460 |
"step": 4925
|
| 1461 |
},
|
| 1462 |
{
|
| 1463 |
"epoch": 9.07,
|
| 1464 |
+
"grad_norm": 0.001896819332614541,
|
| 1465 |
"learning_rate": 9.482439926062846e-05,
|
| 1466 |
"loss": 0.0013,
|
| 1467 |
"step": 4950
|
| 1468 |
},
|
| 1469 |
{
|
| 1470 |
"epoch": 9.11,
|
| 1471 |
+
"grad_norm": 0.007586441468447447,
|
| 1472 |
"learning_rate": 9.020332717190388e-05,
|
| 1473 |
+
"loss": 0.0017,
|
| 1474 |
"step": 4975
|
| 1475 |
},
|
| 1476 |
{
|
| 1477 |
"epoch": 9.16,
|
| 1478 |
+
"grad_norm": 0.006564935203641653,
|
| 1479 |
"learning_rate": 8.558225508317929e-05,
|
| 1480 |
+
"loss": 0.003,
|
| 1481 |
"step": 5000
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
"epoch": 9.2,
|
| 1485 |
+
"grad_norm": 0.005424303933978081,
|
| 1486 |
"learning_rate": 8.096118299445473e-05,
|
| 1487 |
+
"loss": 0.0014,
|
| 1488 |
"step": 5025
|
| 1489 |
},
|
| 1490 |
{
|
| 1491 |
"epoch": 9.25,
|
| 1492 |
+
"grad_norm": 0.0165091622620821,
|
| 1493 |
"learning_rate": 7.634011090573013e-05,
|
| 1494 |
+
"loss": 0.0027,
|
| 1495 |
"step": 5050
|
| 1496 |
},
|
| 1497 |
{
|
| 1498 |
"epoch": 9.29,
|
| 1499 |
+
"grad_norm": 0.09231999516487122,
|
| 1500 |
"learning_rate": 7.171903881700554e-05,
|
| 1501 |
+
"loss": 0.0018,
|
| 1502 |
"step": 5075
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
"epoch": 9.34,
|
| 1506 |
+
"grad_norm": 0.16238878667354584,
|
| 1507 |
"learning_rate": 6.709796672828096e-05,
|
| 1508 |
+
"loss": 0.0015,
|
| 1509 |
"step": 5100
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
"epoch": 9.39,
|
| 1513 |
+
"grad_norm": 0.04476441815495491,
|
| 1514 |
"learning_rate": 6.247689463955638e-05,
|
| 1515 |
+
"loss": 0.0011,
|
| 1516 |
"step": 5125
|
| 1517 |
},
|
| 1518 |
{
|
| 1519 |
"epoch": 9.43,
|
| 1520 |
+
"grad_norm": 0.00874653086066246,
|
| 1521 |
"learning_rate": 5.785582255083179e-05,
|
| 1522 |
+
"loss": 0.0008,
|
| 1523 |
"step": 5150
|
| 1524 |
},
|
| 1525 |
{
|
| 1526 |
"epoch": 9.48,
|
| 1527 |
+
"grad_norm": 0.010477906093001366,
|
| 1528 |
"learning_rate": 5.323475046210721e-05,
|
| 1529 |
+
"loss": 0.0021,
|
| 1530 |
"step": 5175
|
| 1531 |
},
|
| 1532 |
{
|
| 1533 |
"epoch": 9.52,
|
| 1534 |
+
"grad_norm": 0.00953985471278429,
|
| 1535 |
"learning_rate": 4.8613678373382625e-05,
|
| 1536 |
+
"loss": 0.0017,
|
| 1537 |
"step": 5200
|
| 1538 |
},
|
| 1539 |
{
|
| 1540 |
"epoch": 9.57,
|
| 1541 |
+
"grad_norm": 0.0022518846672028303,
|
| 1542 |
"learning_rate": 4.3992606284658045e-05,
|
| 1543 |
+
"loss": 0.0019,
|
| 1544 |
"step": 5225
|
| 1545 |
},
|
| 1546 |
{
|
| 1547 |
"epoch": 9.62,
|
| 1548 |
+
"grad_norm": 0.037685129791498184,
|
| 1549 |
"learning_rate": 3.937153419593346e-05,
|
| 1550 |
"loss": 0.001,
|
| 1551 |
"step": 5250
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 9.66,
|
| 1555 |
+
"grad_norm": 0.08190955966711044,
|
| 1556 |
"learning_rate": 3.4750462107208874e-05,
|
| 1557 |
+
"loss": 0.0017,
|
| 1558 |
"step": 5275
|
| 1559 |
},
|
| 1560 |
{
|
| 1561 |
"epoch": 9.71,
|
| 1562 |
+
"grad_norm": 0.017375241965055466,
|
| 1563 |
"learning_rate": 3.012939001848429e-05,
|
| 1564 |
+
"loss": 0.0016,
|
| 1565 |
"step": 5300
|
| 1566 |
},
|
| 1567 |
{
|
| 1568 |
"epoch": 9.75,
|
| 1569 |
+
"grad_norm": 0.03486447408795357,
|
| 1570 |
"learning_rate": 2.5508317929759705e-05,
|
| 1571 |
+
"loss": 0.0012,
|
| 1572 |
"step": 5325
|
| 1573 |
},
|
| 1574 |
{
|
| 1575 |
"epoch": 9.8,
|
| 1576 |
+
"grad_norm": 0.0786125510931015,
|
| 1577 |
"learning_rate": 2.088724584103512e-05,
|
| 1578 |
+
"loss": 0.0012,
|
| 1579 |
"step": 5350
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
"epoch": 9.84,
|
| 1583 |
+
"grad_norm": 0.09049534052610397,
|
| 1584 |
"learning_rate": 1.6266173752310537e-05,
|
| 1585 |
+
"loss": 0.0012,
|
| 1586 |
"step": 5375
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
"epoch": 9.89,
|
| 1590 |
+
"grad_norm": 0.012832165695726871,
|
| 1591 |
"learning_rate": 1.1645101663585952e-05,
|
| 1592 |
+
"loss": 0.0014,
|
| 1593 |
"step": 5400
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
"epoch": 9.94,
|
| 1597 |
+
"grad_norm": 0.006516186986118555,
|
| 1598 |
"learning_rate": 7.024029574861368e-06,
|
| 1599 |
+
"loss": 0.0015,
|
| 1600 |
"step": 5425
|
| 1601 |
},
|
| 1602 |
{
|
| 1603 |
"epoch": 9.98,
|
| 1604 |
+
"grad_norm": 0.02494051493704319,
|
| 1605 |
"learning_rate": 2.402957486136784e-06,
|
| 1606 |
+
"loss": 0.0015,
|
| 1607 |
"step": 5450
|
| 1608 |
},
|
| 1609 |
{
|
| 1610 |
"epoch": 10.0,
|
| 1611 |
+
"eval_loss": 0.221242755651474,
|
| 1612 |
+
"eval_runtime": 176.7742,
|
| 1613 |
+
"eval_samples_per_second": 4.599,
|
| 1614 |
+
"eval_steps_per_second": 0.769,
|
| 1615 |
"step": 5460
|
| 1616 |
},
|
| 1617 |
{
|
| 1618 |
"epoch": 10.0,
|
| 1619 |
"step": 5460,
|
| 1620 |
"total_flos": 9.7789895073792e+18,
|
| 1621 |
+
"train_loss": 0.024085146698745945,
|
| 1622 |
+
"train_runtime": 10729.5864,
|
| 1623 |
+
"train_samples_per_second": 3.05,
|
| 1624 |
"train_steps_per_second": 0.509
|
| 1625 |
}
|
| 1626 |
],
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5112
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0bd466f2e7cc2734735d5e5c8279377c63a93a1f9a6913853ad505a1a2013446
|
| 3 |
size 5112
|