Sara Price
commited on
Training in progress, step 400, checkpoint
Browse files- last-checkpoint/model-00001-of-00006.safetensors +1 -1
- last-checkpoint/model-00002-of-00006.safetensors +1 -1
- last-checkpoint/model-00003-of-00006.safetensors +1 -1
- last-checkpoint/model-00004-of-00006.safetensors +1 -1
- last-checkpoint/model-00005-of-00006.safetensors +1 -1
- last-checkpoint/model-00006-of-00006.safetensors +1 -1
- last-checkpoint/trainer_state.json +61 -901
- last-checkpoint/training_args.bin +1 -1
last-checkpoint/model-00001-of-00006.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4840658560
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef7402095373828868c6a8c82feed83c288050530180d57b6352e806cf58b1f6
|
| 3 |
size 4840658560
|
last-checkpoint/model-00002-of-00006.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4857206856
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f62da095bfe2f7bab3bc886fc3bc2b6de500f06f57f897eafb8dc0efb326fd8
|
| 3 |
size 4857206856
|
last-checkpoint/model-00003-of-00006.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4857206904
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6b096a5676e65903cff59acf69d696927f6cc30a379206e9cf4eeece59d6d7a
|
| 3 |
size 4857206904
|
last-checkpoint/model-00004-of-00006.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4857206904
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de37c4d3e07cda4c12efbb271b44e386d980542786ad5bd58846677dc3229213
|
| 3 |
size 4857206904
|
last-checkpoint/model-00005-of-00006.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4857206904
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64218f8c0e4a08ca5ebf894c332331d6e36968327167e22813cd4ea6e69a0e27
|
| 3 |
size 4857206904
|
last-checkpoint/model-00006-of-00006.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2684734256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e551d45159d773a5d9a73c93ba8e731328c39b98762114a539357d30098dd964
|
| 3 |
size 2684734256
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,980 +1,140 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.1,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 2.
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 50
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1,
|
| 20 |
-
"eval_loss": 0.
|
| 21 |
-
"eval_runtime": 2.
|
| 22 |
-
"eval_samples_per_second": 69.
|
| 23 |
-
"eval_steps_per_second": 3.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.2,
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"loss": 0.
|
| 31 |
"step": 100
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.2,
|
| 35 |
-
"eval_loss": 0.
|
| 36 |
-
"eval_runtime": 2.
|
| 37 |
-
"eval_samples_per_second": 69.
|
| 38 |
-
"eval_steps_per_second": 3.
|
| 39 |
"step": 100
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.3,
|
| 43 |
-
"grad_norm":
|
| 44 |
-
"learning_rate":
|
| 45 |
-
"loss": 0.
|
| 46 |
"step": 150
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 0.3,
|
| 50 |
-
"eval_loss": 0.
|
| 51 |
-
"eval_runtime": 2.
|
| 52 |
-
"eval_samples_per_second": 69.
|
| 53 |
-
"eval_steps_per_second": 3.
|
| 54 |
"step": 150
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 0.4,
|
| 58 |
-
"grad_norm":
|
| 59 |
-
"learning_rate":
|
| 60 |
-
"loss": 0.
|
| 61 |
"step": 200
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 0.4,
|
| 65 |
-
"eval_loss": 0.
|
| 66 |
-
"eval_runtime": 2.
|
| 67 |
-
"eval_samples_per_second": 69.
|
| 68 |
-
"eval_steps_per_second": 3.
|
| 69 |
"step": 200
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.5,
|
| 73 |
-
"grad_norm":
|
| 74 |
-
"learning_rate":
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 250
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 0.5,
|
| 80 |
-
"eval_loss": 0.
|
| 81 |
-
"eval_runtime": 2.
|
| 82 |
-
"eval_samples_per_second": 69.
|
| 83 |
-
"eval_steps_per_second": 3.
|
| 84 |
"step": 250
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 0.6,
|
| 88 |
-
"grad_norm":
|
| 89 |
-
"learning_rate": 1.
|
| 90 |
-
"loss": 0.
|
| 91 |
"step": 300
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 0.6,
|
| 95 |
-
"eval_loss": 0.
|
| 96 |
-
"eval_runtime": 2.
|
| 97 |
-
"eval_samples_per_second": 69.
|
| 98 |
-
"eval_steps_per_second": 3.
|
| 99 |
"step": 300
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.7,
|
| 103 |
-
"grad_norm":
|
| 104 |
-
"learning_rate":
|
| 105 |
-
"loss": 0.
|
| 106 |
"step": 350
|
| 107 |
},
|
| 108 |
{
|
| 109 |
"epoch": 0.7,
|
| 110 |
-
"eval_loss": 0.
|
| 111 |
-
"eval_runtime": 2.
|
| 112 |
-
"eval_samples_per_second": 69.
|
| 113 |
-
"eval_steps_per_second": 3.
|
| 114 |
"step": 350
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.8,
|
| 118 |
-
"grad_norm": 2.
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.8,
|
| 125 |
-
"eval_loss": 0.
|
| 126 |
-
"eval_runtime": 2.
|
| 127 |
-
"eval_samples_per_second":
|
| 128 |
-
"eval_steps_per_second": 3.548,
|
| 129 |
-
"step": 400
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.9,
|
| 133 |
-
"grad_norm": 4.2005743980407715,
|
| 134 |
-
"learning_rate": 1.9950307753654016e-05,
|
| 135 |
-
"loss": 0.7883,
|
| 136 |
-
"step": 450
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.9,
|
| 140 |
-
"eval_loss": 0.7130094766616821,
|
| 141 |
-
"eval_runtime": 2.3009,
|
| 142 |
-
"eval_samples_per_second": 67.798,
|
| 143 |
-
"eval_steps_per_second": 3.477,
|
| 144 |
-
"step": 450
|
| 145 |
-
},
|
| 146 |
-
{
|
| 147 |
-
"epoch": 1.0,
|
| 148 |
-
"grad_norm": 2.7577402591705322,
|
| 149 |
-
"learning_rate": 1.9888308262251286e-05,
|
| 150 |
-
"loss": 0.7819,
|
| 151 |
-
"step": 500
|
| 152 |
-
},
|
| 153 |
-
{
|
| 154 |
-
"epoch": 1.0,
|
| 155 |
-
"eval_loss": 0.7020623683929443,
|
| 156 |
-
"eval_runtime": 2.9849,
|
| 157 |
-
"eval_samples_per_second": 52.263,
|
| 158 |
-
"eval_steps_per_second": 2.68,
|
| 159 |
-
"step": 500
|
| 160 |
-
},
|
| 161 |
-
{
|
| 162 |
-
"epoch": 1.1,
|
| 163 |
-
"grad_norm": 2.6879160404205322,
|
| 164 |
-
"learning_rate": 1.9801724878485438e-05,
|
| 165 |
-
"loss": 0.4665,
|
| 166 |
-
"step": 550
|
| 167 |
-
},
|
| 168 |
-
{
|
| 169 |
-
"epoch": 1.1,
|
| 170 |
-
"eval_loss": 0.7426314353942871,
|
| 171 |
-
"eval_runtime": 3.4217,
|
| 172 |
-
"eval_samples_per_second": 45.592,
|
| 173 |
-
"eval_steps_per_second": 2.338,
|
| 174 |
-
"step": 550
|
| 175 |
-
},
|
| 176 |
-
{
|
| 177 |
-
"epoch": 1.2,
|
| 178 |
-
"grad_norm": 2.7690176963806152,
|
| 179 |
-
"learning_rate": 1.969077286229078e-05,
|
| 180 |
-
"loss": 0.4805,
|
| 181 |
-
"step": 600
|
| 182 |
-
},
|
| 183 |
-
{
|
| 184 |
-
"epoch": 1.2,
|
| 185 |
-
"eval_loss": 0.7398442029953003,
|
| 186 |
-
"eval_runtime": 2.6618,
|
| 187 |
-
"eval_samples_per_second": 58.606,
|
| 188 |
-
"eval_steps_per_second": 3.005,
|
| 189 |
-
"step": 600
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"epoch": 1.3,
|
| 193 |
-
"grad_norm": 2.0741443634033203,
|
| 194 |
-
"learning_rate": 1.955572805786141e-05,
|
| 195 |
-
"loss": 0.4674,
|
| 196 |
-
"step": 650
|
| 197 |
-
},
|
| 198 |
-
{
|
| 199 |
-
"epoch": 1.3,
|
| 200 |
-
"eval_loss": 0.7346900701522827,
|
| 201 |
-
"eval_runtime": 2.2666,
|
| 202 |
-
"eval_samples_per_second": 68.825,
|
| 203 |
-
"eval_steps_per_second": 3.529,
|
| 204 |
-
"step": 650
|
| 205 |
-
},
|
| 206 |
-
{
|
| 207 |
-
"epoch": 1.4,
|
| 208 |
-
"grad_norm": 1.9551373720169067,
|
| 209 |
-
"learning_rate": 1.9396926207859085e-05,
|
| 210 |
-
"loss": 0.4792,
|
| 211 |
-
"step": 700
|
| 212 |
-
},
|
| 213 |
-
{
|
| 214 |
-
"epoch": 1.4,
|
| 215 |
-
"eval_loss": 0.7401903867721558,
|
| 216 |
-
"eval_runtime": 2.2602,
|
| 217 |
-
"eval_samples_per_second": 69.019,
|
| 218 |
-
"eval_steps_per_second": 3.539,
|
| 219 |
-
"step": 700
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 1.5,
|
| 223 |
-
"grad_norm": 1.8908940553665161,
|
| 224 |
-
"learning_rate": 1.921476211870408e-05,
|
| 225 |
-
"loss": 0.4717,
|
| 226 |
-
"step": 750
|
| 227 |
-
},
|
| 228 |
-
{
|
| 229 |
-
"epoch": 1.5,
|
| 230 |
-
"eval_loss": 0.7299663424491882,
|
| 231 |
-
"eval_runtime": 2.2563,
|
| 232 |
-
"eval_samples_per_second": 69.14,
|
| 233 |
-
"eval_steps_per_second": 3.546,
|
| 234 |
-
"step": 750
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 1.6,
|
| 238 |
-
"grad_norm": 2.781102418899536,
|
| 239 |
-
"learning_rate": 1.900968867902419e-05,
|
| 240 |
-
"loss": 0.4765,
|
| 241 |
-
"step": 800
|
| 242 |
-
},
|
| 243 |
-
{
|
| 244 |
-
"epoch": 1.6,
|
| 245 |
-
"eval_loss": 0.7422571182250977,
|
| 246 |
-
"eval_runtime": 2.2558,
|
| 247 |
-
"eval_samples_per_second": 69.156,
|
| 248 |
-
"eval_steps_per_second": 3.546,
|
| 249 |
-
"step": 800
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 1.7,
|
| 253 |
-
"grad_norm": 2.0381083488464355,
|
| 254 |
-
"learning_rate": 1.8782215733702286e-05,
|
| 255 |
-
"loss": 0.4982,
|
| 256 |
-
"step": 850
|
| 257 |
-
},
|
| 258 |
-
{
|
| 259 |
-
"epoch": 1.7,
|
| 260 |
-
"eval_loss": 0.7189474701881409,
|
| 261 |
-
"eval_runtime": 2.8175,
|
| 262 |
-
"eval_samples_per_second": 55.369,
|
| 263 |
-
"eval_steps_per_second": 2.839,
|
| 264 |
-
"step": 850
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 1.8,
|
| 268 |
-
"grad_norm": 2.6000654697418213,
|
| 269 |
-
"learning_rate": 1.8532908816321557e-05,
|
| 270 |
-
"loss": 0.4798,
|
| 271 |
-
"step": 900
|
| 272 |
-
},
|
| 273 |
-
{
|
| 274 |
-
"epoch": 1.8,
|
| 275 |
-
"eval_loss": 0.7210726141929626,
|
| 276 |
-
"eval_runtime": 4.1738,
|
| 277 |
-
"eval_samples_per_second": 37.376,
|
| 278 |
-
"eval_steps_per_second": 1.917,
|
| 279 |
-
"step": 900
|
| 280 |
-
},
|
| 281 |
-
{
|
| 282 |
-
"epoch": 1.9,
|
| 283 |
-
"grad_norm": 2.4208157062530518,
|
| 284 |
-
"learning_rate": 1.826238774315995e-05,
|
| 285 |
-
"loss": 0.4977,
|
| 286 |
-
"step": 950
|
| 287 |
-
},
|
| 288 |
-
{
|
| 289 |
-
"epoch": 1.9,
|
| 290 |
-
"eval_loss": 0.7182486057281494,
|
| 291 |
-
"eval_runtime": 3.4468,
|
| 292 |
-
"eval_samples_per_second": 45.259,
|
| 293 |
-
"eval_steps_per_second": 2.321,
|
| 294 |
-
"step": 950
|
| 295 |
-
},
|
| 296 |
-
{
|
| 297 |
-
"epoch": 2.0,
|
| 298 |
-
"grad_norm": 2.4245738983154297,
|
| 299 |
-
"learning_rate": 1.7971325072229227e-05,
|
| 300 |
-
"loss": 0.4679,
|
| 301 |
-
"step": 1000
|
| 302 |
-
},
|
| 303 |
-
{
|
| 304 |
-
"epoch": 2.0,
|
| 305 |
-
"eval_loss": 0.7213618755340576,
|
| 306 |
-
"eval_runtime": 2.7343,
|
| 307 |
-
"eval_samples_per_second": 57.053,
|
| 308 |
-
"eval_steps_per_second": 2.926,
|
| 309 |
-
"step": 1000
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 2.1,
|
| 313 |
-
"grad_norm": 2.163320302963257,
|
| 314 |
-
"learning_rate": 1.766044443118978e-05,
|
| 315 |
-
"loss": 0.2305,
|
| 316 |
-
"step": 1050
|
| 317 |
-
},
|
| 318 |
-
{
|
| 319 |
-
"epoch": 2.1,
|
| 320 |
-
"eval_loss": 0.8159535527229309,
|
| 321 |
-
"eval_runtime": 2.2585,
|
| 322 |
-
"eval_samples_per_second": 69.074,
|
| 323 |
-
"eval_steps_per_second": 3.542,
|
| 324 |
-
"step": 1050
|
| 325 |
-
},
|
| 326 |
-
{
|
| 327 |
-
"epoch": 2.2,
|
| 328 |
-
"grad_norm": 1.630889654159546,
|
| 329 |
-
"learning_rate": 1.7330518718298263e-05,
|
| 330 |
-
"loss": 0.2266,
|
| 331 |
-
"step": 1100
|
| 332 |
-
},
|
| 333 |
-
{
|
| 334 |
-
"epoch": 2.2,
|
| 335 |
-
"eval_loss": 0.8228777647018433,
|
| 336 |
-
"eval_runtime": 2.295,
|
| 337 |
-
"eval_samples_per_second": 67.975,
|
| 338 |
-
"eval_steps_per_second": 3.486,
|
| 339 |
-
"step": 1100
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 2.3,
|
| 343 |
-
"grad_norm": 2.260448455810547,
|
| 344 |
-
"learning_rate": 1.698236818086073e-05,
|
| 345 |
-
"loss": 0.2317,
|
| 346 |
-
"step": 1150
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 2.3,
|
| 350 |
-
"eval_loss": 0.8190523982048035,
|
| 351 |
-
"eval_runtime": 2.2571,
|
| 352 |
-
"eval_samples_per_second": 69.114,
|
| 353 |
-
"eval_steps_per_second": 3.544,
|
| 354 |
-
"step": 1150
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 2.4,
|
| 358 |
-
"grad_norm": 1.3863478899002075,
|
| 359 |
-
"learning_rate": 1.6616858375968596e-05,
|
| 360 |
-
"loss": 0.2342,
|
| 361 |
-
"step": 1200
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"epoch": 2.4,
|
| 365 |
-
"eval_loss": 0.8243977427482605,
|
| 366 |
-
"eval_runtime": 2.2587,
|
| 367 |
-
"eval_samples_per_second": 69.067,
|
| 368 |
-
"eval_steps_per_second": 3.542,
|
| 369 |
-
"step": 1200
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 2.5,
|
| 373 |
-
"grad_norm": 2.200590133666992,
|
| 374 |
-
"learning_rate": 1.6234898018587336e-05,
|
| 375 |
-
"loss": 0.239,
|
| 376 |
-
"step": 1250
|
| 377 |
-
},
|
| 378 |
-
{
|
| 379 |
-
"epoch": 2.5,
|
| 380 |
-
"eval_loss": 0.8106646537780762,
|
| 381 |
-
"eval_runtime": 2.4225,
|
| 382 |
-
"eval_samples_per_second": 64.395,
|
| 383 |
-
"eval_steps_per_second": 3.302,
|
| 384 |
-
"step": 1250
|
| 385 |
-
},
|
| 386 |
-
{
|
| 387 |
-
"epoch": 2.6,
|
| 388 |
-
"grad_norm": 1.6970518827438354,
|
| 389 |
-
"learning_rate": 1.5837436722347902e-05,
|
| 390 |
-
"loss": 0.2391,
|
| 391 |
-
"step": 1300
|
| 392 |
-
},
|
| 393 |
-
{
|
| 394 |
-
"epoch": 2.6,
|
| 395 |
-
"eval_loss": 0.830450713634491,
|
| 396 |
-
"eval_runtime": 3.5213,
|
| 397 |
-
"eval_samples_per_second": 44.302,
|
| 398 |
-
"eval_steps_per_second": 2.272,
|
| 399 |
-
"step": 1300
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 2.7,
|
| 403 |
-
"grad_norm": 1.6799397468566895,
|
| 404 |
-
"learning_rate": 1.5425462638657597e-05,
|
| 405 |
-
"loss": 0.2348,
|
| 406 |
-
"step": 1350
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 2.7,
|
| 410 |
-
"eval_loss": 0.8164975047111511,
|
| 411 |
-
"eval_runtime": 4.3755,
|
| 412 |
-
"eval_samples_per_second": 35.653,
|
| 413 |
-
"eval_steps_per_second": 1.828,
|
| 414 |
-
"step": 1350
|
| 415 |
-
},
|
| 416 |
-
{
|
| 417 |
-
"epoch": 2.8,
|
| 418 |
-
"grad_norm": 1.6129848957061768,
|
| 419 |
-
"learning_rate": 1.5000000000000002e-05,
|
| 420 |
-
"loss": 0.2383,
|
| 421 |
-
"step": 1400
|
| 422 |
-
},
|
| 423 |
-
{
|
| 424 |
-
"epoch": 2.8,
|
| 425 |
-
"eval_loss": 0.8099900484085083,
|
| 426 |
-
"eval_runtime": 4.1168,
|
| 427 |
-
"eval_samples_per_second": 37.894,
|
| 428 |
-
"eval_steps_per_second": 1.943,
|
| 429 |
-
"step": 1400
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 2.9,
|
| 433 |
-
"grad_norm": 1.9729114770889282,
|
| 434 |
-
"learning_rate": 1.4562106573531632e-05,
|
| 435 |
-
"loss": 0.2357,
|
| 436 |
-
"step": 1450
|
| 437 |
-
},
|
| 438 |
-
{
|
| 439 |
-
"epoch": 2.9,
|
| 440 |
-
"eval_loss": 0.8040891289710999,
|
| 441 |
-
"eval_runtime": 2.2676,
|
| 442 |
-
"eval_samples_per_second": 68.796,
|
| 443 |
-
"eval_steps_per_second": 3.528,
|
| 444 |
-
"step": 1450
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 3.0,
|
| 448 |
-
"grad_norm": 1.3552831411361694,
|
| 449 |
-
"learning_rate": 1.4112871031306118e-05,
|
| 450 |
-
"loss": 0.2418,
|
| 451 |
-
"step": 1500
|
| 452 |
-
},
|
| 453 |
-
{
|
| 454 |
-
"epoch": 3.0,
|
| 455 |
-
"eval_loss": 0.7948845028877258,
|
| 456 |
-
"eval_runtime": 2.2537,
|
| 457 |
-
"eval_samples_per_second": 69.22,
|
| 458 |
-
"eval_steps_per_second": 3.55,
|
| 459 |
-
"step": 1500
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 3.1,
|
| 463 |
-
"grad_norm": 1.5692390203475952,
|
| 464 |
-
"learning_rate": 1.3653410243663953e-05,
|
| 465 |
-
"loss": 0.1423,
|
| 466 |
-
"step": 1550
|
| 467 |
-
},
|
| 468 |
-
{
|
| 469 |
-
"epoch": 3.1,
|
| 470 |
-
"eval_loss": 0.8620080947875977,
|
| 471 |
-
"eval_runtime": 2.2569,
|
| 472 |
-
"eval_samples_per_second": 69.12,
|
| 473 |
-
"eval_steps_per_second": 3.545,
|
| 474 |
-
"step": 1550
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 3.2,
|
| 478 |
-
"grad_norm": 1.1803256273269653,
|
| 479 |
-
"learning_rate": 1.3184866502516846e-05,
|
| 480 |
-
"loss": 0.1409,
|
| 481 |
-
"step": 1600
|
| 482 |
-
},
|
| 483 |
-
{
|
| 484 |
-
"epoch": 3.2,
|
| 485 |
-
"eval_loss": 0.8848384022712708,
|
| 486 |
-
"eval_runtime": 2.2588,
|
| 487 |
-
"eval_samples_per_second": 69.062,
|
| 488 |
-
"eval_steps_per_second": 3.542,
|
| 489 |
-
"step": 1600
|
| 490 |
-
},
|
| 491 |
-
{
|
| 492 |
-
"epoch": 3.3,
|
| 493 |
-
"grad_norm": 1.1109460592269897,
|
| 494 |
-
"learning_rate": 1.2708404681430054e-05,
|
| 495 |
-
"loss": 0.1496,
|
| 496 |
-
"step": 1650
|
| 497 |
-
},
|
| 498 |
-
{
|
| 499 |
-
"epoch": 3.3,
|
| 500 |
-
"eval_loss": 0.8655369281768799,
|
| 501 |
-
"eval_runtime": 4.4935,
|
| 502 |
-
"eval_samples_per_second": 34.717,
|
| 503 |
-
"eval_steps_per_second": 1.78,
|
| 504 |
-
"step": 1650
|
| 505 |
-
},
|
| 506 |
-
{
|
| 507 |
-
"epoch": 3.4,
|
| 508 |
-
"grad_norm": 1.0697747468948364,
|
| 509 |
-
"learning_rate": 1.2225209339563144e-05,
|
| 510 |
-
"loss": 0.1491,
|
| 511 |
-
"step": 1700
|
| 512 |
-
},
|
| 513 |
-
{
|
| 514 |
-
"epoch": 3.4,
|
| 515 |
-
"eval_loss": 0.8790720701217651,
|
| 516 |
-
"eval_runtime": 2.9231,
|
| 517 |
-
"eval_samples_per_second": 53.369,
|
| 518 |
-
"eval_steps_per_second": 2.737,
|
| 519 |
-
"step": 1700
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 3.5,
|
| 523 |
-
"grad_norm": 1.5176475048065186,
|
| 524 |
-
"learning_rate": 1.1736481776669307e-05,
|
| 525 |
-
"loss": 0.1475,
|
| 526 |
-
"step": 1750
|
| 527 |
-
},
|
| 528 |
-
{
|
| 529 |
-
"epoch": 3.5,
|
| 530 |
-
"eval_loss": 0.8695181608200073,
|
| 531 |
-
"eval_runtime": 2.8158,
|
| 532 |
-
"eval_samples_per_second": 55.401,
|
| 533 |
-
"eval_steps_per_second": 2.841,
|
| 534 |
-
"step": 1750
|
| 535 |
-
},
|
| 536 |
-
{
|
| 537 |
-
"epoch": 3.6,
|
| 538 |
-
"grad_norm": 2.0232934951782227,
|
| 539 |
-
"learning_rate": 1.1243437046474854e-05,
|
| 540 |
-
"loss": 0.1481,
|
| 541 |
-
"step": 1800
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 3.6,
|
| 545 |
-
"eval_loss": 0.8817379474639893,
|
| 546 |
-
"eval_runtime": 3.4544,
|
| 547 |
-
"eval_samples_per_second": 45.159,
|
| 548 |
-
"eval_steps_per_second": 2.316,
|
| 549 |
-
"step": 1800
|
| 550 |
-
},
|
| 551 |
-
{
|
| 552 |
-
"epoch": 3.7,
|
| 553 |
-
"grad_norm": 2.0155882835388184,
|
| 554 |
-
"learning_rate": 1.0747300935864245e-05,
|
| 555 |
-
"loss": 0.1514,
|
| 556 |
-
"step": 1850
|
| 557 |
-
},
|
| 558 |
-
{
|
| 559 |
-
"epoch": 3.7,
|
| 560 |
-
"eval_loss": 0.8860240578651428,
|
| 561 |
-
"eval_runtime": 2.2629,
|
| 562 |
-
"eval_samples_per_second": 68.939,
|
| 563 |
"eval_steps_per_second": 3.535,
|
| 564 |
-
"step":
|
| 565 |
-
},
|
| 566 |
-
{
|
| 567 |
-
"epoch": 3.8,
|
| 568 |
-
"grad_norm": 1.2712018489837646,
|
| 569 |
-
"learning_rate": 1.0249306917380731e-05,
|
| 570 |
-
"loss": 0.1421,
|
| 571 |
-
"step": 1900
|
| 572 |
-
},
|
| 573 |
-
{
|
| 574 |
-
"epoch": 3.8,
|
| 575 |
-
"eval_loss": 0.8792973756790161,
|
| 576 |
-
"eval_runtime": 2.2679,
|
| 577 |
-
"eval_samples_per_second": 68.786,
|
| 578 |
-
"eval_steps_per_second": 3.528,
|
| 579 |
-
"step": 1900
|
| 580 |
-
},
|
| 581 |
-
{
|
| 582 |
-
"epoch": 3.9,
|
| 583 |
-
"grad_norm": 1.3542224168777466,
|
| 584 |
-
"learning_rate": 9.750693082619274e-06,
|
| 585 |
-
"loss": 0.1446,
|
| 586 |
-
"step": 1950
|
| 587 |
-
},
|
| 588 |
-
{
|
| 589 |
-
"epoch": 3.9,
|
| 590 |
-
"eval_loss": 0.8789901733398438,
|
| 591 |
-
"eval_runtime": 2.2568,
|
| 592 |
-
"eval_samples_per_second": 69.126,
|
| 593 |
-
"eval_steps_per_second": 3.545,
|
| 594 |
-
"step": 1950
|
| 595 |
-
},
|
| 596 |
-
{
|
| 597 |
-
"epoch": 4.0,
|
| 598 |
-
"grad_norm": 1.1753352880477905,
|
| 599 |
-
"learning_rate": 9.252699064135759e-06,
|
| 600 |
-
"loss": 0.1458,
|
| 601 |
-
"step": 2000
|
| 602 |
-
},
|
| 603 |
-
{
|
| 604 |
-
"epoch": 4.0,
|
| 605 |
-
"eval_loss": 0.8896375894546509,
|
| 606 |
-
"eval_runtime": 2.2586,
|
| 607 |
-
"eval_samples_per_second": 69.069,
|
| 608 |
-
"eval_steps_per_second": 3.542,
|
| 609 |
-
"step": 2000
|
| 610 |
-
},
|
| 611 |
-
{
|
| 612 |
-
"epoch": 4.1,
|
| 613 |
-
"grad_norm": 1.1664059162139893,
|
| 614 |
-
"learning_rate": 8.756562953525151e-06,
|
| 615 |
-
"loss": 0.0943,
|
| 616 |
-
"step": 2050
|
| 617 |
-
},
|
| 618 |
-
{
|
| 619 |
-
"epoch": 4.1,
|
| 620 |
-
"eval_loss": 0.9453464150428772,
|
| 621 |
-
"eval_runtime": 2.7602,
|
| 622 |
-
"eval_samples_per_second": 56.517,
|
| 623 |
-
"eval_steps_per_second": 2.898,
|
| 624 |
-
"step": 2050
|
| 625 |
-
},
|
| 626 |
-
{
|
| 627 |
-
"epoch": 4.2,
|
| 628 |
-
"grad_norm": 0.8652101755142212,
|
| 629 |
-
"learning_rate": 8.263518223330698e-06,
|
| 630 |
-
"loss": 0.0936,
|
| 631 |
-
"step": 2100
|
| 632 |
-
},
|
| 633 |
-
{
|
| 634 |
-
"epoch": 4.2,
|
| 635 |
-
"eval_loss": 0.9263865947723389,
|
| 636 |
-
"eval_runtime": 2.7019,
|
| 637 |
-
"eval_samples_per_second": 57.738,
|
| 638 |
-
"eval_steps_per_second": 2.961,
|
| 639 |
-
"step": 2100
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 4.3,
|
| 643 |
-
"grad_norm": 1.1194976568222046,
|
| 644 |
-
"learning_rate": 7.774790660436857e-06,
|
| 645 |
-
"loss": 0.0944,
|
| 646 |
-
"step": 2150
|
| 647 |
-
},
|
| 648 |
-
{
|
| 649 |
-
"epoch": 4.3,
|
| 650 |
-
"eval_loss": 0.9188054203987122,
|
| 651 |
-
"eval_runtime": 2.4797,
|
| 652 |
-
"eval_samples_per_second": 62.91,
|
| 653 |
-
"eval_steps_per_second": 3.226,
|
| 654 |
-
"step": 2150
|
| 655 |
-
},
|
| 656 |
-
{
|
| 657 |
-
"epoch": 4.4,
|
| 658 |
-
"grad_norm": 0.8499676585197449,
|
| 659 |
-
"learning_rate": 7.291595318569951e-06,
|
| 660 |
-
"loss": 0.0946,
|
| 661 |
-
"step": 2200
|
| 662 |
-
},
|
| 663 |
-
{
|
| 664 |
-
"epoch": 4.4,
|
| 665 |
-
"eval_loss": 0.9282132387161255,
|
| 666 |
-
"eval_runtime": 4.1021,
|
| 667 |
-
"eval_samples_per_second": 38.03,
|
| 668 |
-
"eval_steps_per_second": 1.95,
|
| 669 |
-
"step": 2200
|
| 670 |
-
},
|
| 671 |
-
{
|
| 672 |
-
"epoch": 4.5,
|
| 673 |
-
"grad_norm": 0.9869304299354553,
|
| 674 |
-
"learning_rate": 6.815133497483157e-06,
|
| 675 |
-
"loss": 0.0914,
|
| 676 |
-
"step": 2250
|
| 677 |
-
},
|
| 678 |
-
{
|
| 679 |
-
"epoch": 4.5,
|
| 680 |
-
"eval_loss": 0.9421446323394775,
|
| 681 |
-
"eval_runtime": 2.2639,
|
| 682 |
-
"eval_samples_per_second": 68.908,
|
| 683 |
-
"eval_steps_per_second": 3.534,
|
| 684 |
-
"step": 2250
|
| 685 |
-
},
|
| 686 |
-
{
|
| 687 |
-
"epoch": 4.6,
|
| 688 |
-
"grad_norm": 0.7122153043746948,
|
| 689 |
-
"learning_rate": 6.34658975633605e-06,
|
| 690 |
-
"loss": 0.0927,
|
| 691 |
-
"step": 2300
|
| 692 |
-
},
|
| 693 |
-
{
|
| 694 |
-
"epoch": 4.6,
|
| 695 |
-
"eval_loss": 0.9317699074745178,
|
| 696 |
-
"eval_runtime": 2.2762,
|
| 697 |
-
"eval_samples_per_second": 68.535,
|
| 698 |
-
"eval_steps_per_second": 3.515,
|
| 699 |
-
"step": 2300
|
| 700 |
-
},
|
| 701 |
-
{
|
| 702 |
-
"epoch": 4.7,
|
| 703 |
-
"grad_norm": 0.7879806756973267,
|
| 704 |
-
"learning_rate": 5.887128968693887e-06,
|
| 705 |
-
"loss": 0.0926,
|
| 706 |
-
"step": 2350
|
| 707 |
-
},
|
| 708 |
-
{
|
| 709 |
-
"epoch": 4.7,
|
| 710 |
-
"eval_loss": 0.9379280209541321,
|
| 711 |
-
"eval_runtime": 2.2587,
|
| 712 |
-
"eval_samples_per_second": 69.065,
|
| 713 |
-
"eval_steps_per_second": 3.542,
|
| 714 |
-
"step": 2350
|
| 715 |
-
},
|
| 716 |
-
{
|
| 717 |
-
"epoch": 4.8,
|
| 718 |
-
"grad_norm": 0.7589277625083923,
|
| 719 |
-
"learning_rate": 5.43789342646837e-06,
|
| 720 |
-
"loss": 0.096,
|
| 721 |
-
"step": 2400
|
| 722 |
-
},
|
| 723 |
-
{
|
| 724 |
-
"epoch": 4.8,
|
| 725 |
-
"eval_loss": 0.9322577118873596,
|
| 726 |
-
"eval_runtime": 2.2566,
|
| 727 |
-
"eval_samples_per_second": 69.13,
|
| 728 |
-
"eval_steps_per_second": 3.545,
|
| 729 |
-
"step": 2400
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"epoch": 4.9,
|
| 733 |
-
"grad_norm": 1.010057806968689,
|
| 734 |
-
"learning_rate": 5.000000000000003e-06,
|
| 735 |
-
"loss": 0.093,
|
| 736 |
-
"step": 2450
|
| 737 |
-
},
|
| 738 |
-
{
|
| 739 |
-
"epoch": 4.9,
|
| 740 |
-
"eval_loss": 0.9507681131362915,
|
| 741 |
-
"eval_runtime": 2.3695,
|
| 742 |
-
"eval_samples_per_second": 65.835,
|
| 743 |
-
"eval_steps_per_second": 3.376,
|
| 744 |
-
"step": 2450
|
| 745 |
-
},
|
| 746 |
-
{
|
| 747 |
-
"epoch": 5.0,
|
| 748 |
-
"grad_norm": 0.8985171914100647,
|
| 749 |
-
"learning_rate": 4.5745373613424075e-06,
|
| 750 |
-
"loss": 0.0923,
|
| 751 |
-
"step": 2500
|
| 752 |
-
},
|
| 753 |
-
{
|
| 754 |
-
"epoch": 5.0,
|
| 755 |
-
"eval_loss": 0.9445481896400452,
|
| 756 |
-
"eval_runtime": 2.455,
|
| 757 |
-
"eval_samples_per_second": 63.545,
|
| 758 |
-
"eval_steps_per_second": 3.259,
|
| 759 |
-
"step": 2500
|
| 760 |
-
},
|
| 761 |
-
{
|
| 762 |
-
"epoch": 5.1,
|
| 763 |
-
"grad_norm": 0.7004356384277344,
|
| 764 |
-
"learning_rate": 4.162563277652104e-06,
|
| 765 |
-
"loss": 0.0534,
|
| 766 |
-
"step": 2550
|
| 767 |
-
},
|
| 768 |
-
{
|
| 769 |
-
"epoch": 5.1,
|
| 770 |
-
"eval_loss": 1.019740343093872,
|
| 771 |
-
"eval_runtime": 3.8097,
|
| 772 |
-
"eval_samples_per_second": 40.949,
|
| 773 |
-
"eval_steps_per_second": 2.1,
|
| 774 |
-
"step": 2550
|
| 775 |
-
},
|
| 776 |
-
{
|
| 777 |
-
"epoch": 5.2,
|
| 778 |
-
"grad_norm": 1.2868136167526245,
|
| 779 |
-
"learning_rate": 3.7651019814126656e-06,
|
| 780 |
-
"loss": 0.0552,
|
| 781 |
-
"step": 2600
|
| 782 |
-
},
|
| 783 |
-
{
|
| 784 |
-
"epoch": 5.2,
|
| 785 |
-
"eval_loss": 1.0226831436157227,
|
| 786 |
-
"eval_runtime": 3.5746,
|
| 787 |
-
"eval_samples_per_second": 43.641,
|
| 788 |
-
"eval_steps_per_second": 2.238,
|
| 789 |
-
"step": 2600
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 5.3,
|
| 793 |
-
"grad_norm": 1.0688791275024414,
|
| 794 |
-
"learning_rate": 3.3831416240314085e-06,
|
| 795 |
-
"loss": 0.0523,
|
| 796 |
-
"step": 2650
|
| 797 |
-
},
|
| 798 |
-
{
|
| 799 |
-
"epoch": 5.3,
|
| 800 |
-
"eval_loss": 1.0200960636138916,
|
| 801 |
-
"eval_runtime": 2.26,
|
| 802 |
-
"eval_samples_per_second": 69.026,
|
| 803 |
-
"eval_steps_per_second": 3.54,
|
| 804 |
-
"step": 2650
|
| 805 |
-
},
|
| 806 |
-
{
|
| 807 |
-
"epoch": 5.4,
|
| 808 |
-
"grad_norm": 0.46982139348983765,
|
| 809 |
-
"learning_rate": 3.017631819139273e-06,
|
| 810 |
-
"loss": 0.0534,
|
| 811 |
-
"step": 2700
|
| 812 |
-
},
|
| 813 |
-
{
|
| 814 |
-
"epoch": 5.4,
|
| 815 |
-
"eval_loss": 1.0137168169021606,
|
| 816 |
-
"eval_runtime": 2.2646,
|
| 817 |
-
"eval_samples_per_second": 68.885,
|
| 818 |
-
"eval_steps_per_second": 3.533,
|
| 819 |
-
"step": 2700
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 5.5,
|
| 823 |
-
"grad_norm": 0.4713106155395508,
|
| 824 |
-
"learning_rate": 2.669481281701739e-06,
|
| 825 |
-
"loss": 0.0541,
|
| 826 |
-
"step": 2750
|
| 827 |
-
},
|
| 828 |
-
{
|
| 829 |
-
"epoch": 5.5,
|
| 830 |
-
"eval_loss": 1.0238293409347534,
|
| 831 |
-
"eval_runtime": 2.2586,
|
| 832 |
-
"eval_samples_per_second": 69.068,
|
| 833 |
-
"eval_steps_per_second": 3.542,
|
| 834 |
-
"step": 2750
|
| 835 |
-
},
|
| 836 |
-
{
|
| 837 |
-
"epoch": 5.6,
|
| 838 |
-
"grad_norm": 0.5023716688156128,
|
| 839 |
-
"learning_rate": 2.339555568810221e-06,
|
| 840 |
-
"loss": 0.0518,
|
| 841 |
-
"step": 2800
|
| 842 |
-
},
|
| 843 |
-
{
|
| 844 |
-
"epoch": 5.6,
|
| 845 |
-
"eval_loss": 1.0314223766326904,
|
| 846 |
-
"eval_runtime": 2.2586,
|
| 847 |
-
"eval_samples_per_second": 69.069,
|
| 848 |
-
"eval_steps_per_second": 3.542,
|
| 849 |
-
"step": 2800
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 5.7,
|
| 853 |
-
"grad_norm": 0.9530413150787354,
|
| 854 |
-
"learning_rate": 2.0286749277707783e-06,
|
| 855 |
-
"loss": 0.053,
|
| 856 |
-
"step": 2850
|
| 857 |
-
},
|
| 858 |
-
{
|
| 859 |
-
"epoch": 5.7,
|
| 860 |
-
"eval_loss": 1.0339645147323608,
|
| 861 |
-
"eval_runtime": 3.1056,
|
| 862 |
-
"eval_samples_per_second": 50.232,
|
| 863 |
-
"eval_steps_per_second": 2.576,
|
| 864 |
-
"step": 2850
|
| 865 |
-
},
|
| 866 |
-
{
|
| 867 |
-
"epoch": 5.8,
|
| 868 |
-
"grad_norm": 0.38149017095565796,
|
| 869 |
-
"learning_rate": 1.7376122568400533e-06,
|
| 870 |
-
"loss": 0.053,
|
| 871 |
-
"step": 2900
|
| 872 |
-
},
|
| 873 |
-
{
|
| 874 |
-
"epoch": 5.8,
|
| 875 |
-
"eval_loss": 1.0364776849746704,
|
| 876 |
-
"eval_runtime": 3.5002,
|
| 877 |
-
"eval_samples_per_second": 44.569,
|
| 878 |
-
"eval_steps_per_second": 2.286,
|
| 879 |
-
"step": 2900
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 5.9,
|
| 883 |
-
"grad_norm": 0.5894300937652588,
|
| 884 |
-
"learning_rate": 1.467091183678444e-06,
|
| 885 |
-
"loss": 0.054,
|
| 886 |
-
"step": 2950
|
| 887 |
-
},
|
| 888 |
-
{
|
| 889 |
-
"epoch": 5.9,
|
| 890 |
-
"eval_loss": 1.0346895456314087,
|
| 891 |
-
"eval_runtime": 2.446,
|
| 892 |
-
"eval_samples_per_second": 63.778,
|
| 893 |
-
"eval_steps_per_second": 3.271,
|
| 894 |
-
"step": 2950
|
| 895 |
-
},
|
| 896 |
-
{
|
| 897 |
-
"epoch": 6.0,
|
| 898 |
-
"grad_norm": 0.565856397151947,
|
| 899 |
-
"learning_rate": 1.2177842662977136e-06,
|
| 900 |
-
"loss": 0.0527,
|
| 901 |
-
"step": 3000
|
| 902 |
-
},
|
| 903 |
-
{
|
| 904 |
-
"epoch": 6.0,
|
| 905 |
-
"eval_loss": 1.0345444679260254,
|
| 906 |
-
"eval_runtime": 3.2958,
|
| 907 |
-
"eval_samples_per_second": 47.333,
|
| 908 |
-
"eval_steps_per_second": 2.427,
|
| 909 |
-
"step": 3000
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 6.1,
|
| 913 |
-
"grad_norm": 0.31733566522598267,
|
| 914 |
-
"learning_rate": 9.903113209758098e-07,
|
| 915 |
-
"loss": 0.0437,
|
| 916 |
-
"step": 3050
|
| 917 |
-
},
|
| 918 |
-
{
|
| 919 |
-
"epoch": 6.1,
|
| 920 |
-
"eval_loss": 1.053617238998413,
|
| 921 |
-
"eval_runtime": 2.2676,
|
| 922 |
-
"eval_samples_per_second": 68.795,
|
| 923 |
-
"eval_steps_per_second": 3.528,
|
| 924 |
-
"step": 3050
|
| 925 |
-
},
|
| 926 |
-
{
|
| 927 |
-
"epoch": 6.2,
|
| 928 |
-
"grad_norm": 0.4090058207511902,
|
| 929 |
-
"learning_rate": 7.852378812959227e-07,
|
| 930 |
-
"loss": 0.0435,
|
| 931 |
-
"step": 3100
|
| 932 |
-
},
|
| 933 |
-
{
|
| 934 |
-
"epoch": 6.2,
|
| 935 |
-
"eval_loss": 1.0663079023361206,
|
| 936 |
-
"eval_runtime": 2.2592,
|
| 937 |
-
"eval_samples_per_second": 69.051,
|
| 938 |
-
"eval_steps_per_second": 3.541,
|
| 939 |
-
"step": 3100
|
| 940 |
-
},
|
| 941 |
-
{
|
| 942 |
-
"epoch": 6.3,
|
| 943 |
-
"grad_norm": 0.33440783619880676,
|
| 944 |
-
"learning_rate": 6.030737921409169e-07,
|
| 945 |
-
"loss": 0.0436,
|
| 946 |
-
"step": 3150
|
| 947 |
-
},
|
| 948 |
-
{
|
| 949 |
-
"epoch": 6.3,
|
| 950 |
-
"eval_loss": 1.075048565864563,
|
| 951 |
-
"eval_runtime": 2.2569,
|
| 952 |
-
"eval_samples_per_second": 69.123,
|
| 953 |
-
"eval_steps_per_second": 3.545,
|
| 954 |
-
"step": 3150
|
| 955 |
-
},
|
| 956 |
-
{
|
| 957 |
-
"epoch": 6.4,
|
| 958 |
-
"grad_norm": 0.398806631565094,
|
| 959 |
-
"learning_rate": 4.4427194213859216e-07,
|
| 960 |
-
"loss": 0.0442,
|
| 961 |
-
"step": 3200
|
| 962 |
-
},
|
| 963 |
-
{
|
| 964 |
-
"epoch": 6.4,
|
| 965 |
-
"eval_loss": 1.0798892974853516,
|
| 966 |
-
"eval_runtime": 2.2538,
|
| 967 |
-
"eval_samples_per_second": 69.216,
|
| 968 |
-
"eval_steps_per_second": 3.55,
|
| 969 |
-
"step": 3200
|
| 970 |
}
|
| 971 |
],
|
| 972 |
"logging_steps": 50,
|
| 973 |
-
"max_steps":
|
| 974 |
"num_input_tokens_seen": 0,
|
| 975 |
-
"num_train_epochs":
|
| 976 |
"save_steps": 400,
|
| 977 |
-
"total_flos": 1.
|
| 978 |
"train_batch_size": 4,
|
| 979 |
"trial_name": null,
|
| 980 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.8,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 400,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.1,
|
| 13 |
+
"grad_norm": 10.53576374053955,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 1.6618,
|
| 16 |
"step": 50
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1,
|
| 20 |
+
"eval_loss": 0.7731789350509644,
|
| 21 |
+
"eval_runtime": 2.2494,
|
| 22 |
+
"eval_samples_per_second": 69.353,
|
| 23 |
+
"eval_steps_per_second": 3.557,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.2,
|
| 28 |
+
"grad_norm": 5.800010681152344,
|
| 29 |
+
"learning_rate": 4.000000000000001e-06,
|
| 30 |
+
"loss": 0.7629,
|
| 31 |
"step": 100
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.2,
|
| 35 |
+
"eval_loss": 0.6901325583457947,
|
| 36 |
+
"eval_runtime": 2.2539,
|
| 37 |
+
"eval_samples_per_second": 69.213,
|
| 38 |
+
"eval_steps_per_second": 3.549,
|
| 39 |
"step": 100
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.3,
|
| 43 |
+
"grad_norm": 4.960265636444092,
|
| 44 |
+
"learning_rate": 6e-06,
|
| 45 |
+
"loss": 0.7256,
|
| 46 |
"step": 150
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 0.3,
|
| 50 |
+
"eval_loss": 0.6716309785842896,
|
| 51 |
+
"eval_runtime": 2.2526,
|
| 52 |
+
"eval_samples_per_second": 69.254,
|
| 53 |
+
"eval_steps_per_second": 3.551,
|
| 54 |
"step": 150
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 0.4,
|
| 58 |
+
"grad_norm": 5.574848651885986,
|
| 59 |
+
"learning_rate": 8.000000000000001e-06,
|
| 60 |
+
"loss": 0.7243,
|
| 61 |
"step": 200
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 0.4,
|
| 65 |
+
"eval_loss": 0.6644517779350281,
|
| 66 |
+
"eval_runtime": 2.2546,
|
| 67 |
+
"eval_samples_per_second": 69.193,
|
| 68 |
+
"eval_steps_per_second": 3.548,
|
| 69 |
"step": 200
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.5,
|
| 73 |
+
"grad_norm": 3.0581891536712646,
|
| 74 |
+
"learning_rate": 1e-05,
|
| 75 |
+
"loss": 0.6918,
|
| 76 |
"step": 250
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 0.5,
|
| 80 |
+
"eval_loss": 0.6718080043792725,
|
| 81 |
+
"eval_runtime": 2.255,
|
| 82 |
+
"eval_samples_per_second": 69.18,
|
| 83 |
+
"eval_steps_per_second": 3.548,
|
| 84 |
"step": 250
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 0.6,
|
| 88 |
+
"grad_norm": 3.797400712966919,
|
| 89 |
+
"learning_rate": 1.2e-05,
|
| 90 |
+
"loss": 0.7433,
|
| 91 |
"step": 300
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 0.6,
|
| 95 |
+
"eval_loss": 0.67710280418396,
|
| 96 |
+
"eval_runtime": 2.2558,
|
| 97 |
+
"eval_samples_per_second": 69.155,
|
| 98 |
+
"eval_steps_per_second": 3.546,
|
| 99 |
"step": 300
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.7,
|
| 103 |
+
"grad_norm": 8.121636390686035,
|
| 104 |
+
"learning_rate": 1.4e-05,
|
| 105 |
+
"loss": 0.7523,
|
| 106 |
"step": 350
|
| 107 |
},
|
| 108 |
{
|
| 109 |
"epoch": 0.7,
|
| 110 |
+
"eval_loss": 0.680716335773468,
|
| 111 |
+
"eval_runtime": 2.2562,
|
| 112 |
+
"eval_samples_per_second": 69.144,
|
| 113 |
+
"eval_steps_per_second": 3.546,
|
| 114 |
"step": 350
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.8,
|
| 118 |
+
"grad_norm": 2.615454912185669,
|
| 119 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 120 |
+
"loss": 0.7322,
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.8,
|
| 125 |
+
"eval_loss": 0.6906686425209045,
|
| 126 |
+
"eval_runtime": 2.2633,
|
| 127 |
+
"eval_samples_per_second": 68.926,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"eval_steps_per_second": 3.535,
|
| 129 |
+
"step": 400
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
}
|
| 131 |
],
|
| 132 |
"logging_steps": 50,
|
| 133 |
+
"max_steps": 5000,
|
| 134 |
"num_input_tokens_seen": 0,
|
| 135 |
+
"num_train_epochs": 10,
|
| 136 |
"save_steps": 400,
|
| 137 |
+
"total_flos": 1.9232917507014656e+16,
|
| 138 |
"train_batch_size": 4,
|
| 139 |
"trial_name": null,
|
| 140 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:676922c8c88a048a8c76fcb1420a476fbd64daf190286d2f67fca7597684e8c6
|
| 3 |
size 5176
|