Upload folder using huggingface_hub
Browse files- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- checkpoint-150/adapter_config.json +3 -3
- checkpoint-150/adapter_model.safetensors +1 -1
- checkpoint-150/optimizer.pt +1 -1
- checkpoint-150/trainer_state.json +772 -772
- checkpoint-164/adapter_config.json +3 -3
- checkpoint-164/adapter_model.safetensors +1 -1
- checkpoint-164/optimizer.pt +1 -1
- checkpoint-164/trainer_state.json +842 -842
adapter_config.json
CHANGED
|
@@ -30,11 +30,11 @@
|
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
"o_proj",
|
| 33 |
-
"q_proj",
|
| 34 |
-
"gate_proj",
|
| 35 |
-
"down_proj",
|
| 36 |
"k_proj",
|
|
|
|
|
|
|
| 37 |
"v_proj",
|
|
|
|
| 38 |
"up_proj"
|
| 39 |
],
|
| 40 |
"target_parameters": null,
|
|
|
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
"o_proj",
|
|
|
|
|
|
|
|
|
|
| 33 |
"k_proj",
|
| 34 |
+
"down_proj",
|
| 35 |
+
"q_proj",
|
| 36 |
"v_proj",
|
| 37 |
+
"gate_proj",
|
| 38 |
"up_proj"
|
| 39 |
],
|
| 40 |
"target_parameters": null,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83946192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d7799ff7ae7f290e67eada9d323c6418a3a9db26bdb2158f039838076d95d1f
|
| 3 |
size 83946192
|
checkpoint-150/adapter_config.json
CHANGED
|
@@ -30,11 +30,11 @@
|
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
"o_proj",
|
| 33 |
-
"q_proj",
|
| 34 |
-
"gate_proj",
|
| 35 |
-
"down_proj",
|
| 36 |
"k_proj",
|
|
|
|
|
|
|
| 37 |
"v_proj",
|
|
|
|
| 38 |
"up_proj"
|
| 39 |
],
|
| 40 |
"target_parameters": null,
|
|
|
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
"o_proj",
|
|
|
|
|
|
|
|
|
|
| 33 |
"k_proj",
|
| 34 |
+
"down_proj",
|
| 35 |
+
"q_proj",
|
| 36 |
"v_proj",
|
| 37 |
+
"gate_proj",
|
| 38 |
"up_proj"
|
| 39 |
],
|
| 40 |
"target_parameters": null,
|
checkpoint-150/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83946192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:762e73134d8cb3ce48b4da442522cb72cca6085e8d761c4956332ea9159b92c5
|
| 3 |
size 83946192
|
checkpoint-150/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 85728997
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ecadd10ced5ddfdc8d4f885fbd10e4abbd52e6bb17ae093c71dec91b5ed759e5
|
| 3 |
size 85728997
|
checkpoint-150/trainer_state.json
CHANGED
|
@@ -10,1536 +10,1536 @@
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"entropy": 0.
|
| 14 |
"epoch": 0.01225114854517611,
|
| 15 |
-
"grad_norm":
|
| 16 |
"learning_rate": 0.0002,
|
| 17 |
-
"loss": 0.
|
| 18 |
-
"mean_token_accuracy": 0.
|
| 19 |
-
"num_tokens":
|
| 20 |
"step": 1
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"entropy": 0.
|
| 24 |
"epoch": 0.02450229709035222,
|
| 25 |
-
"grad_norm":
|
| 26 |
"learning_rate": 0.00019878048780487805,
|
| 27 |
-
"loss":
|
| 28 |
-
"mean_token_accuracy":
|
| 29 |
-
"num_tokens":
|
| 30 |
"step": 2
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"entropy": 0.
|
| 34 |
"epoch": 0.036753445635528334,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 0.0001975609756097561,
|
| 37 |
-
"loss":
|
| 38 |
-
"mean_token_accuracy":
|
| 39 |
-
"num_tokens":
|
| 40 |
"step": 3
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"entropy": 0.
|
| 44 |
"epoch": 0.04900459418070444,
|
| 45 |
-
"grad_norm":
|
| 46 |
"learning_rate": 0.00019634146341463416,
|
| 47 |
-
"loss": 0.
|
| 48 |
-
"mean_token_accuracy": 0.
|
| 49 |
-
"num_tokens":
|
| 50 |
"step": 4
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"entropy": 0.
|
| 54 |
"epoch": 0.06125574272588055,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 0.0001951219512195122,
|
| 57 |
-
"loss":
|
| 58 |
-
"mean_token_accuracy":
|
| 59 |
-
"num_tokens":
|
| 60 |
"step": 5
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"entropy": 0.
|
| 64 |
"epoch": 0.07350689127105667,
|
| 65 |
-
"grad_norm":
|
| 66 |
"learning_rate": 0.00019390243902439025,
|
| 67 |
-
"loss":
|
| 68 |
-
"mean_token_accuracy":
|
| 69 |
-
"num_tokens":
|
| 70 |
"step": 6
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"entropy": 0.
|
| 74 |
"epoch": 0.08575803981623277,
|
| 75 |
-
"grad_norm":
|
| 76 |
"learning_rate": 0.0001926829268292683,
|
| 77 |
-
"loss":
|
| 78 |
-
"mean_token_accuracy":
|
| 79 |
-
"num_tokens":
|
| 80 |
"step": 7
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"entropy": 0.
|
| 84 |
"epoch": 0.09800918836140889,
|
| 85 |
-
"grad_norm":
|
| 86 |
"learning_rate": 0.00019146341463414633,
|
| 87 |
-
"loss":
|
| 88 |
-
"mean_token_accuracy":
|
| 89 |
-
"num_tokens":
|
| 90 |
"step": 8
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"entropy": 0.
|
| 94 |
"epoch": 0.11026033690658499,
|
| 95 |
-
"grad_norm": 0.
|
| 96 |
"learning_rate": 0.0001902439024390244,
|
| 97 |
-
"loss": 0.
|
| 98 |
-
"mean_token_accuracy": 0.
|
| 99 |
-
"num_tokens":
|
| 100 |
"step": 9
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"entropy": 0.
|
| 104 |
"epoch": 0.1225114854517611,
|
| 105 |
-
"grad_norm":
|
| 106 |
"learning_rate": 0.00018902439024390244,
|
| 107 |
-
"loss": 0.
|
| 108 |
-
"mean_token_accuracy": 0.
|
| 109 |
-
"num_tokens":
|
| 110 |
"step": 10
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"entropy": 0.
|
| 114 |
"epoch": 0.13476263399693722,
|
| 115 |
-
"grad_norm": 0.
|
| 116 |
"learning_rate": 0.0001878048780487805,
|
| 117 |
-
"loss":
|
| 118 |
-
"mean_token_accuracy":
|
| 119 |
-
"num_tokens":
|
| 120 |
"step": 11
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"entropy": 0.
|
| 124 |
"epoch": 0.14701378254211334,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
"learning_rate": 0.00018658536585365856,
|
| 127 |
-
"loss":
|
| 128 |
-
"mean_token_accuracy":
|
| 129 |
-
"num_tokens":
|
| 130 |
"step": 12
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"entropy": 0.
|
| 134 |
"epoch": 0.15926493108728942,
|
| 135 |
-
"grad_norm": 0.
|
| 136 |
"learning_rate": 0.0001853658536585366,
|
| 137 |
-
"loss":
|
| 138 |
-
"mean_token_accuracy":
|
| 139 |
-
"num_tokens":
|
| 140 |
"step": 13
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"entropy": 0.
|
| 144 |
"epoch": 0.17151607963246554,
|
| 145 |
-
"grad_norm":
|
| 146 |
"learning_rate": 0.00018414634146341464,
|
| 147 |
-
"loss":
|
| 148 |
-
"mean_token_accuracy":
|
| 149 |
-
"num_tokens":
|
| 150 |
"step": 14
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"entropy": 0.
|
| 154 |
"epoch": 0.18376722817764166,
|
| 155 |
-
"grad_norm":
|
| 156 |
"learning_rate": 0.0001829268292682927,
|
| 157 |
-
"loss":
|
| 158 |
-
"mean_token_accuracy":
|
| 159 |
-
"num_tokens":
|
| 160 |
"step": 15
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"entropy": 0.
|
| 164 |
"epoch": 0.19601837672281777,
|
| 165 |
-
"grad_norm":
|
| 166 |
"learning_rate": 0.00018170731707317075,
|
| 167 |
-
"loss":
|
| 168 |
-
"mean_token_accuracy":
|
| 169 |
-
"num_tokens":
|
| 170 |
"step": 16
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"entropy": 0.
|
| 174 |
"epoch": 0.2082695252679939,
|
| 175 |
-
"grad_norm":
|
| 176 |
"learning_rate": 0.0001804878048780488,
|
| 177 |
-
"loss":
|
| 178 |
-
"mean_token_accuracy":
|
| 179 |
-
"num_tokens":
|
| 180 |
"step": 17
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"entropy": 0.
|
| 184 |
"epoch": 0.22052067381316998,
|
| 185 |
-
"grad_norm":
|
| 186 |
"learning_rate": 0.00017926829268292684,
|
| 187 |
-
"loss":
|
| 188 |
-
"mean_token_accuracy":
|
| 189 |
-
"num_tokens":
|
| 190 |
"step": 18
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"entropy": 0.
|
| 194 |
"epoch": 0.2327718223583461,
|
| 195 |
-
"grad_norm":
|
| 196 |
"learning_rate": 0.00017804878048780488,
|
| 197 |
-
"loss":
|
| 198 |
-
"mean_token_accuracy":
|
| 199 |
-
"num_tokens":
|
| 200 |
"step": 19
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"entropy": 0.
|
| 204 |
"epoch": 0.2450229709035222,
|
| 205 |
-
"grad_norm": 0.
|
| 206 |
"learning_rate": 0.00017682926829268295,
|
| 207 |
-
"loss":
|
| 208 |
-
"mean_token_accuracy":
|
| 209 |
-
"num_tokens":
|
| 210 |
"step": 20
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"entropy": 0.
|
| 214 |
"epoch": 0.2572741194486983,
|
| 215 |
-
"grad_norm":
|
| 216 |
"learning_rate": 0.000175609756097561,
|
| 217 |
-
"loss": 0.
|
| 218 |
-
"mean_token_accuracy": 0.
|
| 219 |
-
"num_tokens":
|
| 220 |
"step": 21
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"entropy": 0.
|
| 224 |
"epoch": 0.26952526799387444,
|
| 225 |
-
"grad_norm":
|
| 226 |
"learning_rate": 0.00017439024390243903,
|
| 227 |
-
"loss":
|
| 228 |
-
"mean_token_accuracy":
|
| 229 |
-
"num_tokens":
|
| 230 |
"step": 22
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"entropy": 0.
|
| 234 |
"epoch": 0.28177641653905056,
|
| 235 |
-
"grad_norm":
|
| 236 |
"learning_rate": 0.00017317073170731708,
|
| 237 |
-
"loss":
|
| 238 |
-
"mean_token_accuracy":
|
| 239 |
-
"num_tokens":
|
| 240 |
"step": 23
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"entropy": 0.
|
| 244 |
"epoch": 0.29402756508422667,
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
"learning_rate": 0.00017195121951219512,
|
| 247 |
-
"loss":
|
| 248 |
-
"mean_token_accuracy":
|
| 249 |
-
"num_tokens":
|
| 250 |
"step": 24
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"entropy": 0.
|
| 254 |
"epoch": 0.30627871362940273,
|
| 255 |
-
"grad_norm": 0.
|
| 256 |
"learning_rate": 0.0001707317073170732,
|
| 257 |
-
"loss":
|
| 258 |
-
"mean_token_accuracy":
|
| 259 |
-
"num_tokens":
|
| 260 |
"step": 25
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"entropy": 0.
|
| 264 |
"epoch": 0.31852986217457885,
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
"learning_rate": 0.00016951219512195123,
|
| 267 |
-
"loss":
|
| 268 |
-
"mean_token_accuracy":
|
| 269 |
-
"num_tokens":
|
| 270 |
"step": 26
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"entropy": 0.
|
| 274 |
"epoch": 0.33078101071975496,
|
| 275 |
-
"grad_norm":
|
| 276 |
"learning_rate": 0.00016829268292682927,
|
| 277 |
-
"loss": 0.
|
| 278 |
-
"mean_token_accuracy":
|
| 279 |
-
"num_tokens":
|
| 280 |
"step": 27
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"entropy": 0.
|
| 284 |
"epoch": 0.3430321592649311,
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
"learning_rate": 0.00016707317073170731,
|
| 287 |
-
"loss": 0.
|
| 288 |
-
"mean_token_accuracy": 0.
|
| 289 |
-
"num_tokens":
|
| 290 |
"step": 28
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"entropy": 0.
|
| 294 |
"epoch": 0.3552833078101072,
|
| 295 |
-
"grad_norm": 0.
|
| 296 |
"learning_rate": 0.00016585365853658536,
|
| 297 |
-
"loss": 0.
|
| 298 |
-
"mean_token_accuracy":
|
| 299 |
-
"num_tokens":
|
| 300 |
"step": 29
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"entropy": 0.
|
| 304 |
"epoch": 0.3675344563552833,
|
| 305 |
-
"grad_norm": 0.
|
| 306 |
"learning_rate": 0.00016463414634146343,
|
| 307 |
-
"loss":
|
| 308 |
-
"mean_token_accuracy":
|
| 309 |
-
"num_tokens":
|
| 310 |
"step": 30
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"entropy": 0.
|
| 314 |
"epoch": 0.37978560490045943,
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
"learning_rate": 0.00016341463414634147,
|
| 317 |
-
"loss":
|
| 318 |
-
"mean_token_accuracy":
|
| 319 |
-
"num_tokens":
|
| 320 |
"step": 31
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"entropy": 0.
|
| 324 |
"epoch": 0.39203675344563554,
|
| 325 |
-
"grad_norm":
|
| 326 |
"learning_rate": 0.00016219512195121954,
|
| 327 |
-
"loss":
|
| 328 |
-
"mean_token_accuracy":
|
| 329 |
-
"num_tokens":
|
| 330 |
"step": 32
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"entropy": 0.
|
| 334 |
"epoch": 0.40428790199081166,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 0.00016097560975609758,
|
| 337 |
-
"loss":
|
| 338 |
-
"mean_token_accuracy":
|
| 339 |
-
"num_tokens":
|
| 340 |
"step": 33
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"entropy": 0.
|
| 344 |
"epoch": 0.4165390505359878,
|
| 345 |
-
"grad_norm":
|
| 346 |
"learning_rate": 0.00015975609756097562,
|
| 347 |
-
"loss":
|
| 348 |
-
"mean_token_accuracy":
|
| 349 |
-
"num_tokens":
|
| 350 |
"step": 34
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"entropy": 0.
|
| 354 |
"epoch": 0.42879019908116384,
|
| 355 |
-
"grad_norm":
|
| 356 |
"learning_rate": 0.00015853658536585366,
|
| 357 |
-
"loss": 0.
|
| 358 |
-
"mean_token_accuracy":
|
| 359 |
-
"num_tokens":
|
| 360 |
"step": 35
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"entropy": 0.
|
| 364 |
"epoch": 0.44104134762633995,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
"learning_rate": 0.00015731707317073173,
|
| 367 |
-
"loss": 0.
|
| 368 |
-
"mean_token_accuracy":
|
| 369 |
-
"num_tokens":
|
| 370 |
"step": 36
|
| 371 |
},
|
| 372 |
{
|
| 373 |
-
"entropy": 0.
|
| 374 |
"epoch": 0.45329249617151607,
|
| 375 |
-
"grad_norm": 0.
|
| 376 |
"learning_rate": 0.00015609756097560978,
|
| 377 |
-
"loss": 0.
|
| 378 |
-
"mean_token_accuracy": 0.
|
| 379 |
-
"num_tokens":
|
| 380 |
"step": 37
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"entropy": 0.
|
| 384 |
"epoch": 0.4655436447166922,
|
| 385 |
-
"grad_norm":
|
| 386 |
"learning_rate": 0.00015487804878048782,
|
| 387 |
-
"loss":
|
| 388 |
-
"mean_token_accuracy":
|
| 389 |
-
"num_tokens":
|
| 390 |
"step": 38
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"entropy": 0.
|
| 394 |
"epoch": 0.4777947932618683,
|
| 395 |
-
"grad_norm":
|
| 396 |
"learning_rate": 0.00015365853658536586,
|
| 397 |
-
"loss":
|
| 398 |
-
"mean_token_accuracy":
|
| 399 |
-
"num_tokens":
|
| 400 |
"step": 39
|
| 401 |
},
|
| 402 |
{
|
| 403 |
-
"entropy": 0.
|
| 404 |
"epoch": 0.4900459418070444,
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
"learning_rate": 0.0001524390243902439,
|
| 407 |
-
"loss": 0.
|
| 408 |
-
"mean_token_accuracy": 0.
|
| 409 |
-
"num_tokens":
|
| 410 |
"step": 40
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"entropy": 0.
|
| 414 |
"epoch": 0.5022970903522205,
|
| 415 |
-
"grad_norm": 0.
|
| 416 |
"learning_rate": 0.00015121951219512197,
|
| 417 |
-
"loss":
|
| 418 |
-
"mean_token_accuracy":
|
| 419 |
-
"num_tokens":
|
| 420 |
"step": 41
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"entropy": 0.
|
| 424 |
"epoch": 0.5145482388973966,
|
| 425 |
-
"grad_norm":
|
| 426 |
"learning_rate": 0.00015000000000000001,
|
| 427 |
-
"loss":
|
| 428 |
-
"mean_token_accuracy":
|
| 429 |
-
"num_tokens":
|
| 430 |
"step": 42
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"entropy": 0.
|
| 434 |
"epoch": 0.5267993874425727,
|
| 435 |
-
"grad_norm":
|
| 436 |
"learning_rate": 0.00014878048780487806,
|
| 437 |
-
"loss":
|
| 438 |
-
"mean_token_accuracy":
|
| 439 |
-
"num_tokens":
|
| 440 |
"step": 43
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"entropy": 0.
|
| 444 |
"epoch": 0.5390505359877489,
|
| 445 |
-
"grad_norm":
|
| 446 |
"learning_rate": 0.0001475609756097561,
|
| 447 |
-
"loss":
|
| 448 |
-
"mean_token_accuracy":
|
| 449 |
-
"num_tokens":
|
| 450 |
"step": 44
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"entropy": 0.
|
| 454 |
"epoch": 0.5513016845329249,
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
"learning_rate": 0.00014634146341463414,
|
| 457 |
-
"loss": 0.
|
| 458 |
-
"mean_token_accuracy": 0.
|
| 459 |
-
"num_tokens":
|
| 460 |
"step": 45
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"entropy": 0.
|
| 464 |
"epoch": 0.5635528330781011,
|
| 465 |
-
"grad_norm":
|
| 466 |
"learning_rate": 0.0001451219512195122,
|
| 467 |
-
"loss":
|
| 468 |
-
"mean_token_accuracy":
|
| 469 |
-
"num_tokens":
|
| 470 |
"step": 46
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"entropy": 0.
|
| 474 |
"epoch": 0.5758039816232772,
|
| 475 |
-
"grad_norm":
|
| 476 |
"learning_rate": 0.00014390243902439025,
|
| 477 |
-
"loss":
|
| 478 |
-
"mean_token_accuracy":
|
| 479 |
-
"num_tokens":
|
| 480 |
"step": 47
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"entropy": 0.
|
| 484 |
"epoch": 0.5880551301684533,
|
| 485 |
-
"grad_norm":
|
| 486 |
"learning_rate": 0.0001426829268292683,
|
| 487 |
-
"loss":
|
| 488 |
-
"mean_token_accuracy":
|
| 489 |
-
"num_tokens":
|
| 490 |
"step": 48
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"entropy": 0.
|
| 494 |
"epoch": 0.6003062787136294,
|
| 495 |
-
"grad_norm":
|
| 496 |
"learning_rate": 0.00014146341463414634,
|
| 497 |
-
"loss":
|
| 498 |
-
"mean_token_accuracy":
|
| 499 |
-
"num_tokens":
|
| 500 |
"step": 49
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"entropy": 0.
|
| 504 |
"epoch": 0.6125574272588055,
|
| 505 |
-
"grad_norm":
|
| 506 |
"learning_rate": 0.00014024390243902438,
|
| 507 |
-
"loss":
|
| 508 |
-
"mean_token_accuracy":
|
| 509 |
-
"num_tokens":
|
| 510 |
"step": 50
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 0.6125574272588055,
|
| 514 |
-
"eval_entropy": 0.
|
| 515 |
-
"eval_loss": 0.
|
| 516 |
-
"eval_mean_token_accuracy": 0.
|
| 517 |
-
"eval_num_tokens":
|
| 518 |
-
"eval_runtime":
|
| 519 |
-
"eval_samples_per_second": 1.
|
| 520 |
-
"eval_steps_per_second": 1.
|
| 521 |
"step": 50
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"entropy": 0.
|
| 525 |
"epoch": 0.6248085758039816,
|
| 526 |
-
"grad_norm":
|
| 527 |
"learning_rate": 0.00013902439024390245,
|
| 528 |
-
"loss":
|
| 529 |
-
"mean_token_accuracy":
|
| 530 |
-
"num_tokens":
|
| 531 |
"step": 51
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"entropy": 0.
|
| 535 |
"epoch": 0.6370597243491577,
|
| 536 |
-
"grad_norm":
|
| 537 |
"learning_rate": 0.0001378048780487805,
|
| 538 |
-
"loss":
|
| 539 |
-
"mean_token_accuracy":
|
| 540 |
-
"num_tokens":
|
| 541 |
"step": 52
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"entropy": 0.
|
| 545 |
"epoch": 0.6493108728943339,
|
| 546 |
-
"grad_norm":
|
| 547 |
"learning_rate": 0.00013658536585365856,
|
| 548 |
-
"loss": 0.
|
| 549 |
-
"mean_token_accuracy": 0.
|
| 550 |
-
"num_tokens":
|
| 551 |
"step": 53
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"entropy": 0.
|
| 555 |
"epoch": 0.6615620214395099,
|
| 556 |
-
"grad_norm": 0.
|
| 557 |
"learning_rate": 0.0001353658536585366,
|
| 558 |
-
"loss":
|
| 559 |
-
"mean_token_accuracy":
|
| 560 |
-
"num_tokens":
|
| 561 |
"step": 54
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"entropy": 0.
|
| 565 |
"epoch": 0.6738131699846861,
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
"learning_rate": 0.00013414634146341464,
|
| 568 |
-
"loss": 0.
|
| 569 |
-
"mean_token_accuracy": 0.
|
| 570 |
-
"num_tokens":
|
| 571 |
"step": 55
|
| 572 |
},
|
| 573 |
{
|
| 574 |
-
"entropy": 0.
|
| 575 |
"epoch": 0.6860643185298622,
|
| 576 |
-
"grad_norm":
|
| 577 |
"learning_rate": 0.0001329268292682927,
|
| 578 |
-
"loss":
|
| 579 |
-
"mean_token_accuracy":
|
| 580 |
-
"num_tokens":
|
| 581 |
"step": 56
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"entropy": 0.
|
| 585 |
"epoch": 0.6983154670750383,
|
| 586 |
-
"grad_norm":
|
| 587 |
"learning_rate": 0.00013170731707317076,
|
| 588 |
-
"loss":
|
| 589 |
-
"mean_token_accuracy":
|
| 590 |
-
"num_tokens":
|
| 591 |
"step": 57
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"entropy": 0.
|
| 595 |
"epoch": 0.7105666156202144,
|
| 596 |
-
"grad_norm":
|
| 597 |
"learning_rate": 0.0001304878048780488,
|
| 598 |
-
"loss": 0.
|
| 599 |
-
"mean_token_accuracy": 0.
|
| 600 |
-
"num_tokens":
|
| 601 |
"step": 58
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"entropy": 0.
|
| 605 |
"epoch": 0.7228177641653905,
|
| 606 |
-
"grad_norm":
|
| 607 |
"learning_rate": 0.00012926829268292684,
|
| 608 |
-
"loss": 0.
|
| 609 |
-
"mean_token_accuracy": 0.
|
| 610 |
-
"num_tokens":
|
| 611 |
"step": 59
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"entropy": 0.
|
| 615 |
"epoch": 0.7350689127105666,
|
| 616 |
-
"grad_norm":
|
| 617 |
"learning_rate": 0.00012804878048780488,
|
| 618 |
-
"loss":
|
| 619 |
-
"mean_token_accuracy":
|
| 620 |
-
"num_tokens":
|
| 621 |
"step": 60
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"entropy": 0.
|
| 625 |
"epoch": 0.7473200612557427,
|
| 626 |
-
"grad_norm":
|
| 627 |
"learning_rate": 0.00012682926829268293,
|
| 628 |
-
"loss":
|
| 629 |
-
"mean_token_accuracy":
|
| 630 |
-
"num_tokens":
|
| 631 |
"step": 61
|
| 632 |
},
|
| 633 |
{
|
| 634 |
-
"entropy": 0.
|
| 635 |
"epoch": 0.7595712098009189,
|
| 636 |
-
"grad_norm":
|
| 637 |
"learning_rate": 0.000125609756097561,
|
| 638 |
-
"loss":
|
| 639 |
-
"mean_token_accuracy":
|
| 640 |
-
"num_tokens":
|
| 641 |
"step": 62
|
| 642 |
},
|
| 643 |
{
|
| 644 |
-
"entropy": 0.
|
| 645 |
"epoch": 0.7718223583460949,
|
| 646 |
-
"grad_norm":
|
| 647 |
"learning_rate": 0.00012439024390243904,
|
| 648 |
-
"loss":
|
| 649 |
-
"mean_token_accuracy":
|
| 650 |
-
"num_tokens":
|
| 651 |
"step": 63
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"entropy": 0.
|
| 655 |
"epoch": 0.7840735068912711,
|
| 656 |
-
"grad_norm":
|
| 657 |
"learning_rate": 0.00012317073170731708,
|
| 658 |
-
"loss":
|
| 659 |
-
"mean_token_accuracy":
|
| 660 |
-
"num_tokens":
|
| 661 |
"step": 64
|
| 662 |
},
|
| 663 |
{
|
| 664 |
-
"entropy": 0.
|
| 665 |
"epoch": 0.7963246554364471,
|
| 666 |
-
"grad_norm":
|
| 667 |
"learning_rate": 0.00012195121951219512,
|
| 668 |
-
"loss":
|
| 669 |
-
"mean_token_accuracy":
|
| 670 |
-
"num_tokens":
|
| 671 |
"step": 65
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"entropy": 0.
|
| 675 |
"epoch": 0.8085758039816233,
|
| 676 |
-
"grad_norm": 0.
|
| 677 |
"learning_rate": 0.00012073170731707318,
|
| 678 |
-
"loss":
|
| 679 |
-
"mean_token_accuracy":
|
| 680 |
-
"num_tokens":
|
| 681 |
"step": 66
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"entropy": 0.
|
| 685 |
"epoch": 0.8208269525267994,
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
"learning_rate": 0.00011951219512195122,
|
| 688 |
-
"loss": 0.
|
| 689 |
-
"mean_token_accuracy": 0.
|
| 690 |
-
"num_tokens":
|
| 691 |
"step": 67
|
| 692 |
},
|
| 693 |
{
|
| 694 |
-
"entropy": 0.
|
| 695 |
"epoch": 0.8330781010719756,
|
| 696 |
-
"grad_norm": 0.
|
| 697 |
"learning_rate": 0.00011829268292682926,
|
| 698 |
-
"loss":
|
| 699 |
-
"mean_token_accuracy":
|
| 700 |
-
"num_tokens":
|
| 701 |
"step": 68
|
| 702 |
},
|
| 703 |
{
|
| 704 |
-
"entropy": 0.
|
| 705 |
"epoch": 0.8453292496171516,
|
| 706 |
-
"grad_norm": 0.
|
| 707 |
"learning_rate": 0.00011707317073170732,
|
| 708 |
-
"loss":
|
| 709 |
-
"mean_token_accuracy":
|
| 710 |
-
"num_tokens":
|
| 711 |
"step": 69
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"entropy": 0.
|
| 715 |
"epoch": 0.8575803981623277,
|
| 716 |
-
"grad_norm": 0.
|
| 717 |
"learning_rate": 0.00011585365853658536,
|
| 718 |
-
"loss": 0.
|
| 719 |
-
"mean_token_accuracy": 0.
|
| 720 |
-
"num_tokens":
|
| 721 |
"step": 70
|
| 722 |
},
|
| 723 |
{
|
| 724 |
-
"entropy": 0.
|
| 725 |
"epoch": 0.8698315467075038,
|
| 726 |
-
"grad_norm": 0.
|
| 727 |
"learning_rate": 0.00011463414634146342,
|
| 728 |
-
"loss": 0.
|
| 729 |
-
"mean_token_accuracy": 0.
|
| 730 |
-
"num_tokens":
|
| 731 |
"step": 71
|
| 732 |
},
|
| 733 |
{
|
| 734 |
-
"entropy": 0.
|
| 735 |
"epoch": 0.8820826952526799,
|
| 736 |
-
"grad_norm": 0.
|
| 737 |
"learning_rate": 0.00011341463414634146,
|
| 738 |
-
"loss":
|
| 739 |
-
"mean_token_accuracy":
|
| 740 |
-
"num_tokens":
|
| 741 |
"step": 72
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"entropy": 0.
|
| 745 |
"epoch": 0.8943338437978561,
|
| 746 |
-
"grad_norm": 0.
|
| 747 |
"learning_rate": 0.00011219512195121953,
|
| 748 |
-
"loss":
|
| 749 |
-
"mean_token_accuracy":
|
| 750 |
-
"num_tokens":
|
| 751 |
"step": 73
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"entropy": 0.
|
| 755 |
"epoch": 0.9065849923430321,
|
| 756 |
-
"grad_norm":
|
| 757 |
"learning_rate": 0.00011097560975609757,
|
| 758 |
-
"loss":
|
| 759 |
-
"mean_token_accuracy":
|
| 760 |
-
"num_tokens":
|
| 761 |
"step": 74
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"entropy": 0.
|
| 765 |
"epoch": 0.9188361408882083,
|
| 766 |
-
"grad_norm":
|
| 767 |
"learning_rate": 0.00010975609756097563,
|
| 768 |
-
"loss":
|
| 769 |
-
"mean_token_accuracy":
|
| 770 |
-
"num_tokens":
|
| 771 |
"step": 75
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"entropy": 0.
|
| 775 |
"epoch": 0.9310872894333844,
|
| 776 |
-
"grad_norm":
|
| 777 |
"learning_rate": 0.00010853658536585367,
|
| 778 |
-
"loss":
|
| 779 |
-
"mean_token_accuracy":
|
| 780 |
-
"num_tokens":
|
| 781 |
"step": 76
|
| 782 |
},
|
| 783 |
{
|
| 784 |
-
"entropy": 0.
|
| 785 |
"epoch": 0.9433384379785605,
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
"learning_rate": 0.00010731707317073172,
|
| 788 |
-
"loss":
|
| 789 |
-
"mean_token_accuracy":
|
| 790 |
-
"num_tokens":
|
| 791 |
"step": 77
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"entropy": 0.
|
| 795 |
"epoch": 0.9555895865237366,
|
| 796 |
-
"grad_norm":
|
| 797 |
"learning_rate": 0.00010609756097560977,
|
| 798 |
-
"loss":
|
| 799 |
-
"mean_token_accuracy":
|
| 800 |
-
"num_tokens":
|
| 801 |
"step": 78
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"entropy": 0.
|
| 805 |
"epoch": 0.9678407350689127,
|
| 806 |
-
"grad_norm":
|
| 807 |
"learning_rate": 0.00010487804878048781,
|
| 808 |
-
"loss":
|
| 809 |
-
"mean_token_accuracy":
|
| 810 |
-
"num_tokens":
|
| 811 |
"step": 79
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"entropy": 0.
|
| 815 |
"epoch": 0.9800918836140888,
|
| 816 |
-
"grad_norm":
|
| 817 |
"learning_rate": 0.00010365853658536586,
|
| 818 |
-
"loss": 0.
|
| 819 |
-
"mean_token_accuracy": 0.
|
| 820 |
-
"num_tokens":
|
| 821 |
"step": 80
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"entropy": 0.
|
| 825 |
"epoch": 0.9923430321592649,
|
| 826 |
-
"grad_norm":
|
| 827 |
"learning_rate": 0.0001024390243902439,
|
| 828 |
-
"loss":
|
| 829 |
-
"mean_token_accuracy":
|
| 830 |
-
"num_tokens":
|
| 831 |
"step": 81
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"entropy": 0.
|
| 835 |
"epoch": 1.0,
|
| 836 |
-
"grad_norm": 0.
|
| 837 |
"learning_rate": 0.00010121951219512196,
|
| 838 |
-
"loss":
|
| 839 |
-
"mean_token_accuracy":
|
| 840 |
-
"num_tokens":
|
| 841 |
"step": 82
|
| 842 |
},
|
| 843 |
{
|
| 844 |
-
"entropy": 0.
|
| 845 |
"epoch": 1.0122511485451762,
|
| 846 |
-
"grad_norm": 0.
|
| 847 |
"learning_rate": 0.0001,
|
| 848 |
-
"loss": 0.
|
| 849 |
-
"mean_token_accuracy": 0.
|
| 850 |
-
"num_tokens":
|
| 851 |
"step": 83
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"entropy": 0.
|
| 855 |
"epoch": 1.0245022970903521,
|
| 856 |
-
"grad_norm": 0.
|
| 857 |
"learning_rate": 9.878048780487805e-05,
|
| 858 |
-
"loss":
|
| 859 |
-
"mean_token_accuracy":
|
| 860 |
-
"num_tokens":
|
| 861 |
"step": 84
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"entropy": 0.
|
| 865 |
"epoch": 1.0367534456355283,
|
| 866 |
-
"grad_norm": 0.
|
| 867 |
"learning_rate": 9.75609756097561e-05,
|
| 868 |
-
"loss":
|
| 869 |
-
"mean_token_accuracy":
|
| 870 |
-
"num_tokens":
|
| 871 |
"step": 85
|
| 872 |
},
|
| 873 |
{
|
| 874 |
-
"entropy": 0.
|
| 875 |
"epoch": 1.0490045941807045,
|
| 876 |
-
"grad_norm": 0.
|
| 877 |
"learning_rate": 9.634146341463415e-05,
|
| 878 |
-
"loss":
|
| 879 |
-
"mean_token_accuracy":
|
| 880 |
-
"num_tokens":
|
| 881 |
"step": 86
|
| 882 |
},
|
| 883 |
{
|
| 884 |
-
"entropy": 0.
|
| 885 |
"epoch": 1.0612557427258806,
|
| 886 |
-
"grad_norm": 0.
|
| 887 |
"learning_rate": 9.51219512195122e-05,
|
| 888 |
-
"loss":
|
| 889 |
-
"mean_token_accuracy":
|
| 890 |
-
"num_tokens":
|
| 891 |
"step": 87
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"entropy": 0.
|
| 895 |
"epoch": 1.0735068912710566,
|
| 896 |
-
"grad_norm": 0.
|
| 897 |
"learning_rate": 9.390243902439024e-05,
|
| 898 |
-
"loss":
|
| 899 |
-
"mean_token_accuracy":
|
| 900 |
-
"num_tokens":
|
| 901 |
"step": 88
|
| 902 |
},
|
| 903 |
{
|
| 904 |
-
"entropy": 0.
|
| 905 |
"epoch": 1.0857580398162328,
|
| 906 |
-
"grad_norm": 0.
|
| 907 |
"learning_rate": 9.26829268292683e-05,
|
| 908 |
-
"loss": 0.
|
| 909 |
-
"mean_token_accuracy": 0.
|
| 910 |
-
"num_tokens":
|
| 911 |
"step": 89
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"entropy": 0.
|
| 915 |
"epoch": 1.098009188361409,
|
| 916 |
-
"grad_norm": 0.
|
| 917 |
"learning_rate": 9.146341463414635e-05,
|
| 918 |
-
"loss":
|
| 919 |
-
"mean_token_accuracy":
|
| 920 |
-
"num_tokens":
|
| 921 |
"step": 90
|
| 922 |
},
|
| 923 |
{
|
| 924 |
-
"entropy": 0.
|
| 925 |
"epoch": 1.110260336906585,
|
| 926 |
-
"grad_norm": 0.
|
| 927 |
"learning_rate": 9.02439024390244e-05,
|
| 928 |
-
"loss": 0.
|
| 929 |
-
"mean_token_accuracy": 0.
|
| 930 |
-
"num_tokens":
|
| 931 |
"step": 91
|
| 932 |
},
|
| 933 |
{
|
| 934 |
-
"entropy": 0.
|
| 935 |
"epoch": 1.122511485451761,
|
| 936 |
-
"grad_norm": 0.
|
| 937 |
"learning_rate": 8.902439024390244e-05,
|
| 938 |
-
"loss": 0.
|
| 939 |
-
"mean_token_accuracy": 0.
|
| 940 |
-
"num_tokens":
|
| 941 |
"step": 92
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"entropy": 0.
|
| 945 |
"epoch": 1.1347626339969372,
|
| 946 |
-
"grad_norm": 0.
|
| 947 |
"learning_rate": 8.78048780487805e-05,
|
| 948 |
-
"loss": 0.
|
| 949 |
-
"mean_token_accuracy":
|
| 950 |
-
"num_tokens":
|
| 951 |
"step": 93
|
| 952 |
},
|
| 953 |
{
|
| 954 |
-
"entropy": 0.
|
| 955 |
"epoch": 1.1470137825421134,
|
| 956 |
-
"grad_norm": 0.
|
| 957 |
"learning_rate": 8.658536585365854e-05,
|
| 958 |
-
"loss": 0.
|
| 959 |
-
"mean_token_accuracy":
|
| 960 |
-
"num_tokens":
|
| 961 |
"step": 94
|
| 962 |
},
|
| 963 |
{
|
| 964 |
-
"entropy": 0.
|
| 965 |
"epoch": 1.1592649310872893,
|
| 966 |
-
"grad_norm":
|
| 967 |
"learning_rate": 8.53658536585366e-05,
|
| 968 |
-
"loss": 0.
|
| 969 |
-
"mean_token_accuracy":
|
| 970 |
-
"num_tokens":
|
| 971 |
"step": 95
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"entropy": 0.
|
| 975 |
"epoch": 1.1715160796324655,
|
| 976 |
-
"grad_norm": 0.
|
| 977 |
"learning_rate": 8.414634146341464e-05,
|
| 978 |
-
"loss": 0.
|
| 979 |
-
"mean_token_accuracy": 0.
|
| 980 |
-
"num_tokens":
|
| 981 |
"step": 96
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"entropy": 0.
|
| 985 |
"epoch": 1.1837672281776417,
|
| 986 |
-
"grad_norm": 0.
|
| 987 |
"learning_rate": 8.292682926829268e-05,
|
| 988 |
-
"loss":
|
| 989 |
-
"mean_token_accuracy":
|
| 990 |
-
"num_tokens":
|
| 991 |
"step": 97
|
| 992 |
},
|
| 993 |
{
|
| 994 |
-
"entropy": 0.
|
| 995 |
"epoch": 1.1960183767228179,
|
| 996 |
-
"grad_norm": 0.
|
| 997 |
"learning_rate": 8.170731707317073e-05,
|
| 998 |
-
"loss": 0.
|
| 999 |
-
"mean_token_accuracy": 0.
|
| 1000 |
-
"num_tokens":
|
| 1001 |
"step": 98
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
-
"entropy": 0.
|
| 1005 |
"epoch": 1.2082695252679938,
|
| 1006 |
-
"grad_norm": 0.
|
| 1007 |
"learning_rate": 8.048780487804879e-05,
|
| 1008 |
-
"loss":
|
| 1009 |
-
"mean_token_accuracy":
|
| 1010 |
-
"num_tokens":
|
| 1011 |
"step": 99
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"entropy": 0.
|
| 1015 |
"epoch": 1.22052067381317,
|
| 1016 |
-
"grad_norm":
|
| 1017 |
"learning_rate": 7.926829268292683e-05,
|
| 1018 |
-
"loss": 0.
|
| 1019 |
-
"mean_token_accuracy":
|
| 1020 |
-
"num_tokens":
|
| 1021 |
"step": 100
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 1.22052067381317,
|
| 1025 |
-
"eval_entropy": 0.
|
| 1026 |
-
"eval_loss": 0.
|
| 1027 |
-
"eval_mean_token_accuracy": 0.
|
| 1028 |
-
"eval_num_tokens":
|
| 1029 |
-
"eval_runtime":
|
| 1030 |
-
"eval_samples_per_second": 1.
|
| 1031 |
-
"eval_steps_per_second": 1.
|
| 1032 |
"step": 100
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
-
"entropy": 0.
|
| 1036 |
"epoch": 1.2327718223583461,
|
| 1037 |
-
"grad_norm": 0.
|
| 1038 |
"learning_rate": 7.804878048780489e-05,
|
| 1039 |
-
"loss":
|
| 1040 |
-
"mean_token_accuracy":
|
| 1041 |
-
"num_tokens":
|
| 1042 |
"step": 101
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
-
"entropy": 0.
|
| 1046 |
"epoch": 1.245022970903522,
|
| 1047 |
-
"grad_norm": 0.
|
| 1048 |
"learning_rate": 7.682926829268293e-05,
|
| 1049 |
-
"loss":
|
| 1050 |
-
"mean_token_accuracy":
|
| 1051 |
-
"num_tokens":
|
| 1052 |
"step": 102
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
-
"entropy": 0.
|
| 1056 |
"epoch": 1.2572741194486983,
|
| 1057 |
-
"grad_norm": 0.
|
| 1058 |
"learning_rate": 7.560975609756099e-05,
|
| 1059 |
-
"loss":
|
| 1060 |
-
"mean_token_accuracy":
|
| 1061 |
-
"num_tokens":
|
| 1062 |
"step": 103
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
-
"entropy": 0.
|
| 1066 |
"epoch": 1.2695252679938744,
|
| 1067 |
-
"grad_norm":
|
| 1068 |
"learning_rate": 7.439024390243903e-05,
|
| 1069 |
-
"loss":
|
| 1070 |
-
"mean_token_accuracy":
|
| 1071 |
-
"num_tokens":
|
| 1072 |
"step": 104
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
-
"entropy": 0.
|
| 1076 |
"epoch": 1.2817764165390506,
|
| 1077 |
-
"grad_norm": 0.
|
| 1078 |
"learning_rate": 7.317073170731707e-05,
|
| 1079 |
-
"loss": 0.
|
| 1080 |
-
"mean_token_accuracy": 0.
|
| 1081 |
-
"num_tokens":
|
| 1082 |
"step": 105
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
-
"entropy": 0.
|
| 1086 |
"epoch": 1.2940275650842268,
|
| 1087 |
-
"grad_norm": 0.
|
| 1088 |
"learning_rate": 7.195121951219513e-05,
|
| 1089 |
-
"loss":
|
| 1090 |
-
"mean_token_accuracy":
|
| 1091 |
-
"num_tokens":
|
| 1092 |
"step": 106
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
-
"entropy": 0.
|
| 1096 |
"epoch": 1.3062787136294027,
|
| 1097 |
-
"grad_norm":
|
| 1098 |
"learning_rate": 7.073170731707317e-05,
|
| 1099 |
-
"loss":
|
| 1100 |
-
"mean_token_accuracy":
|
| 1101 |
-
"num_tokens":
|
| 1102 |
"step": 107
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
-
"entropy": 0.
|
| 1106 |
"epoch": 1.318529862174579,
|
| 1107 |
-
"grad_norm": 0.
|
| 1108 |
"learning_rate": 6.951219512195122e-05,
|
| 1109 |
-
"loss":
|
| 1110 |
-
"mean_token_accuracy":
|
| 1111 |
-
"num_tokens":
|
| 1112 |
"step": 108
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"entropy": 0.
|
| 1116 |
"epoch": 1.3307810107197549,
|
| 1117 |
-
"grad_norm": 0.
|
| 1118 |
"learning_rate": 6.829268292682928e-05,
|
| 1119 |
-
"loss":
|
| 1120 |
-
"mean_token_accuracy":
|
| 1121 |
-
"num_tokens":
|
| 1122 |
"step": 109
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"entropy": 0.
|
| 1126 |
"epoch": 1.343032159264931,
|
| 1127 |
-
"grad_norm": 0.
|
| 1128 |
"learning_rate": 6.707317073170732e-05,
|
| 1129 |
-
"loss":
|
| 1130 |
-
"mean_token_accuracy":
|
| 1131 |
-
"num_tokens":
|
| 1132 |
"step": 110
|
| 1133 |
},
|
| 1134 |
{
|
| 1135 |
-
"entropy": 0.
|
| 1136 |
"epoch": 1.3552833078101072,
|
| 1137 |
-
"grad_norm": 0.
|
| 1138 |
"learning_rate": 6.585365853658538e-05,
|
| 1139 |
-
"loss":
|
| 1140 |
-
"mean_token_accuracy":
|
| 1141 |
-
"num_tokens":
|
| 1142 |
"step": 111
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
-
"entropy": 0.
|
| 1146 |
"epoch": 1.3675344563552834,
|
| 1147 |
-
"grad_norm": 0.
|
| 1148 |
"learning_rate": 6.463414634146342e-05,
|
| 1149 |
-
"loss": 0.
|
| 1150 |
-
"mean_token_accuracy": 0.
|
| 1151 |
-
"num_tokens":
|
| 1152 |
"step": 112
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
-
"entropy": 0.
|
| 1156 |
"epoch": 1.3797856049004595,
|
| 1157 |
-
"grad_norm": 0.
|
| 1158 |
"learning_rate": 6.341463414634146e-05,
|
| 1159 |
-
"loss": 0.
|
| 1160 |
-
"mean_token_accuracy": 0.
|
| 1161 |
-
"num_tokens":
|
| 1162 |
"step": 113
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
-
"entropy": 0.
|
| 1166 |
"epoch": 1.3920367534456355,
|
| 1167 |
-
"grad_norm":
|
| 1168 |
"learning_rate": 6.219512195121952e-05,
|
| 1169 |
-
"loss":
|
| 1170 |
-
"mean_token_accuracy":
|
| 1171 |
-
"num_tokens":
|
| 1172 |
"step": 114
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"entropy": 0.
|
| 1176 |
"epoch": 1.4042879019908117,
|
| 1177 |
-
"grad_norm": 0.
|
| 1178 |
"learning_rate": 6.097560975609756e-05,
|
| 1179 |
-
"loss":
|
| 1180 |
-
"mean_token_accuracy":
|
| 1181 |
-
"num_tokens":
|
| 1182 |
"step": 115
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
-
"entropy": 0.
|
| 1186 |
"epoch": 1.4165390505359878,
|
| 1187 |
-
"grad_norm": 0.
|
| 1188 |
"learning_rate": 5.975609756097561e-05,
|
| 1189 |
-
"loss":
|
| 1190 |
-
"mean_token_accuracy":
|
| 1191 |
-
"num_tokens":
|
| 1192 |
"step": 116
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
-
"entropy": 0.
|
| 1196 |
"epoch": 1.4287901990811638,
|
| 1197 |
-
"grad_norm":
|
| 1198 |
"learning_rate": 5.853658536585366e-05,
|
| 1199 |
-
"loss":
|
| 1200 |
-
"mean_token_accuracy":
|
| 1201 |
-
"num_tokens":
|
| 1202 |
"step": 117
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
-
"entropy": 0.
|
| 1206 |
"epoch": 1.44104134762634,
|
| 1207 |
-
"grad_norm":
|
| 1208 |
"learning_rate": 5.731707317073171e-05,
|
| 1209 |
-
"loss":
|
| 1210 |
-
"mean_token_accuracy":
|
| 1211 |
-
"num_tokens":
|
| 1212 |
"step": 118
|
| 1213 |
},
|
| 1214 |
{
|
| 1215 |
-
"entropy": 0.
|
| 1216 |
"epoch": 1.4532924961715161,
|
| 1217 |
-
"grad_norm":
|
| 1218 |
"learning_rate": 5.6097560975609764e-05,
|
| 1219 |
-
"loss":
|
| 1220 |
-
"mean_token_accuracy":
|
| 1221 |
-
"num_tokens":
|
| 1222 |
"step": 119
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
-
"entropy": 0.
|
| 1226 |
"epoch": 1.4655436447166923,
|
| 1227 |
-
"grad_norm": 0.
|
| 1228 |
"learning_rate": 5.487804878048781e-05,
|
| 1229 |
-
"loss":
|
| 1230 |
-
"mean_token_accuracy":
|
| 1231 |
-
"num_tokens":
|
| 1232 |
"step": 120
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
-
"entropy": 0.
|
| 1236 |
"epoch": 1.4777947932618682,
|
| 1237 |
-
"grad_norm": 0.
|
| 1238 |
"learning_rate": 5.365853658536586e-05,
|
| 1239 |
-
"loss":
|
| 1240 |
-
"mean_token_accuracy":
|
| 1241 |
-
"num_tokens":
|
| 1242 |
"step": 121
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"entropy": 0.
|
| 1246 |
"epoch": 1.4900459418070444,
|
| 1247 |
-
"grad_norm":
|
| 1248 |
"learning_rate": 5.2439024390243904e-05,
|
| 1249 |
-
"loss":
|
| 1250 |
-
"mean_token_accuracy":
|
| 1251 |
-
"num_tokens":
|
| 1252 |
"step": 122
|
| 1253 |
},
|
| 1254 |
{
|
| 1255 |
-
"entropy": 0.
|
| 1256 |
"epoch": 1.5022970903522204,
|
| 1257 |
-
"grad_norm": 0.
|
| 1258 |
"learning_rate": 5.121951219512195e-05,
|
| 1259 |
-
"loss":
|
| 1260 |
-
"mean_token_accuracy":
|
| 1261 |
-
"num_tokens":
|
| 1262 |
"step": 123
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
-
"entropy": 0.
|
| 1266 |
"epoch": 1.5145482388973965,
|
| 1267 |
-
"grad_norm": 0.
|
| 1268 |
"learning_rate": 5e-05,
|
| 1269 |
-
"loss":
|
| 1270 |
-
"mean_token_accuracy":
|
| 1271 |
-
"num_tokens":
|
| 1272 |
"step": 124
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
-
"entropy": 0.
|
| 1276 |
"epoch": 1.5267993874425727,
|
| 1277 |
-
"grad_norm":
|
| 1278 |
"learning_rate": 4.878048780487805e-05,
|
| 1279 |
-
"loss":
|
| 1280 |
-
"mean_token_accuracy":
|
| 1281 |
-
"num_tokens":
|
| 1282 |
"step": 125
|
| 1283 |
},
|
| 1284 |
{
|
| 1285 |
-
"entropy": 0.
|
| 1286 |
"epoch": 1.5390505359877489,
|
| 1287 |
-
"grad_norm": 0.
|
| 1288 |
"learning_rate": 4.75609756097561e-05,
|
| 1289 |
-
"loss":
|
| 1290 |
-
"mean_token_accuracy":
|
| 1291 |
-
"num_tokens":
|
| 1292 |
"step": 126
|
| 1293 |
},
|
| 1294 |
{
|
| 1295 |
-
"entropy": 0.
|
| 1296 |
"epoch": 1.551301684532925,
|
| 1297 |
-
"grad_norm": 0.
|
| 1298 |
"learning_rate": 4.634146341463415e-05,
|
| 1299 |
-
"loss": 0.
|
| 1300 |
-
"mean_token_accuracy":
|
| 1301 |
-
"num_tokens":
|
| 1302 |
"step": 127
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
-
"entropy": 0.
|
| 1306 |
"epoch": 1.5635528330781012,
|
| 1307 |
-
"grad_norm": 0.
|
| 1308 |
"learning_rate": 4.51219512195122e-05,
|
| 1309 |
-
"loss":
|
| 1310 |
-
"mean_token_accuracy":
|
| 1311 |
-
"num_tokens":
|
| 1312 |
"step": 128
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
-
"entropy": 0.
|
| 1316 |
"epoch": 1.5758039816232772,
|
| 1317 |
-
"grad_norm": 0.
|
| 1318 |
"learning_rate": 4.390243902439025e-05,
|
| 1319 |
-
"loss":
|
| 1320 |
-
"mean_token_accuracy":
|
| 1321 |
-
"num_tokens":
|
| 1322 |
"step": 129
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
-
"entropy": 0.
|
| 1326 |
"epoch": 1.5880551301684533,
|
| 1327 |
-
"grad_norm": 0.
|
| 1328 |
"learning_rate": 4.26829268292683e-05,
|
| 1329 |
-
"loss":
|
| 1330 |
-
"mean_token_accuracy":
|
| 1331 |
-
"num_tokens":
|
| 1332 |
"step": 130
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
-
"entropy": 0.
|
| 1336 |
"epoch": 1.6003062787136293,
|
| 1337 |
-
"grad_norm": 0.
|
| 1338 |
"learning_rate": 4.146341463414634e-05,
|
| 1339 |
-
"loss": 0.
|
| 1340 |
-
"mean_token_accuracy": 0.
|
| 1341 |
-
"num_tokens":
|
| 1342 |
"step": 131
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
-
"entropy": 0.
|
| 1346 |
"epoch": 1.6125574272588055,
|
| 1347 |
-
"grad_norm": 0.
|
| 1348 |
"learning_rate": 4.0243902439024395e-05,
|
| 1349 |
-
"loss": 0.
|
| 1350 |
-
"mean_token_accuracy": 0.
|
| 1351 |
-
"num_tokens":
|
| 1352 |
"step": 132
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
-
"entropy": 0.
|
| 1356 |
"epoch": 1.6248085758039816,
|
| 1357 |
-
"grad_norm": 0.
|
| 1358 |
"learning_rate": 3.9024390243902444e-05,
|
| 1359 |
-
"loss":
|
| 1360 |
-
"mean_token_accuracy":
|
| 1361 |
-
"num_tokens":
|
| 1362 |
"step": 133
|
| 1363 |
},
|
| 1364 |
{
|
| 1365 |
-
"entropy": 0.
|
| 1366 |
"epoch": 1.6370597243491578,
|
| 1367 |
-
"grad_norm": 0.
|
| 1368 |
"learning_rate": 3.780487804878049e-05,
|
| 1369 |
-
"loss":
|
| 1370 |
-
"mean_token_accuracy":
|
| 1371 |
-
"num_tokens":
|
| 1372 |
"step": 134
|
| 1373 |
},
|
| 1374 |
{
|
| 1375 |
-
"entropy": 0.
|
| 1376 |
"epoch": 1.649310872894334,
|
| 1377 |
-
"grad_norm": 0.
|
| 1378 |
"learning_rate": 3.6585365853658535e-05,
|
| 1379 |
-
"loss":
|
| 1380 |
-
"mean_token_accuracy":
|
| 1381 |
-
"num_tokens":
|
| 1382 |
"step": 135
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
-
"entropy": 0.
|
| 1386 |
"epoch": 1.66156202143951,
|
| 1387 |
-
"grad_norm": 0.
|
| 1388 |
"learning_rate": 3.5365853658536584e-05,
|
| 1389 |
-
"loss":
|
| 1390 |
-
"mean_token_accuracy":
|
| 1391 |
-
"num_tokens":
|
| 1392 |
"step": 136
|
| 1393 |
},
|
| 1394 |
{
|
| 1395 |
-
"entropy": 0.
|
| 1396 |
"epoch": 1.673813169984686,
|
| 1397 |
-
"grad_norm": 0.
|
| 1398 |
"learning_rate": 3.414634146341464e-05,
|
| 1399 |
-
"loss":
|
| 1400 |
-
"mean_token_accuracy":
|
| 1401 |
-
"num_tokens":
|
| 1402 |
"step": 137
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
-
"entropy": 0.
|
| 1406 |
"epoch": 1.686064318529862,
|
| 1407 |
-
"grad_norm": 0.
|
| 1408 |
"learning_rate": 3.292682926829269e-05,
|
| 1409 |
-
"loss": 0.
|
| 1410 |
-
"mean_token_accuracy": 0.
|
| 1411 |
-
"num_tokens":
|
| 1412 |
"step": 138
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
-
"entropy": 0.
|
| 1416 |
"epoch": 1.6983154670750382,
|
| 1417 |
-
"grad_norm": 0.
|
| 1418 |
"learning_rate": 3.170731707317073e-05,
|
| 1419 |
-
"loss":
|
| 1420 |
-
"mean_token_accuracy":
|
| 1421 |
-
"num_tokens":
|
| 1422 |
"step": 139
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
-
"entropy": 0.
|
| 1426 |
"epoch": 1.7105666156202144,
|
| 1427 |
-
"grad_norm": 0.
|
| 1428 |
"learning_rate": 3.048780487804878e-05,
|
| 1429 |
-
"loss":
|
| 1430 |
-
"mean_token_accuracy":
|
| 1431 |
-
"num_tokens":
|
| 1432 |
"step": 140
|
| 1433 |
},
|
| 1434 |
{
|
| 1435 |
-
"entropy": 0.
|
| 1436 |
"epoch": 1.7228177641653906,
|
| 1437 |
-
"grad_norm": 0.
|
| 1438 |
"learning_rate": 2.926829268292683e-05,
|
| 1439 |
-
"loss":
|
| 1440 |
-
"mean_token_accuracy":
|
| 1441 |
-
"num_tokens":
|
| 1442 |
"step": 141
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
-
"entropy": 0.
|
| 1446 |
"epoch": 1.7350689127105667,
|
| 1447 |
-
"grad_norm": 0.
|
| 1448 |
"learning_rate": 2.8048780487804882e-05,
|
| 1449 |
-
"loss":
|
| 1450 |
-
"mean_token_accuracy":
|
| 1451 |
-
"num_tokens":
|
| 1452 |
"step": 142
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
-
"entropy": 0.
|
| 1456 |
"epoch": 1.7473200612557427,
|
| 1457 |
-
"grad_norm": 0.
|
| 1458 |
"learning_rate": 2.682926829268293e-05,
|
| 1459 |
-
"loss":
|
| 1460 |
-
"mean_token_accuracy":
|
| 1461 |
-
"num_tokens":
|
| 1462 |
"step": 143
|
| 1463 |
},
|
| 1464 |
{
|
| 1465 |
-
"entropy": 0.
|
| 1466 |
"epoch": 1.7595712098009189,
|
| 1467 |
-
"grad_norm": 0.
|
| 1468 |
"learning_rate": 2.5609756097560977e-05,
|
| 1469 |
-
"loss":
|
| 1470 |
-
"mean_token_accuracy":
|
| 1471 |
-
"num_tokens":
|
| 1472 |
"step": 144
|
| 1473 |
},
|
| 1474 |
{
|
| 1475 |
-
"entropy": 0.
|
| 1476 |
"epoch": 1.7718223583460948,
|
| 1477 |
-
"grad_norm": 0.
|
| 1478 |
"learning_rate": 2.4390243902439026e-05,
|
| 1479 |
-
"loss": 0.
|
| 1480 |
-
"mean_token_accuracy": 0.
|
| 1481 |
-
"num_tokens":
|
| 1482 |
"step": 145
|
| 1483 |
},
|
| 1484 |
{
|
| 1485 |
-
"entropy": 0.
|
| 1486 |
"epoch": 1.784073506891271,
|
| 1487 |
-
"grad_norm": 0.
|
| 1488 |
"learning_rate": 2.3170731707317075e-05,
|
| 1489 |
-
"loss":
|
| 1490 |
-
"mean_token_accuracy":
|
| 1491 |
-
"num_tokens":
|
| 1492 |
"step": 146
|
| 1493 |
},
|
| 1494 |
{
|
| 1495 |
-
"entropy": 0.
|
| 1496 |
"epoch": 1.7963246554364471,
|
| 1497 |
-
"grad_norm": 0.
|
| 1498 |
"learning_rate": 2.1951219512195124e-05,
|
| 1499 |
-
"loss":
|
| 1500 |
-
"mean_token_accuracy":
|
| 1501 |
-
"num_tokens":
|
| 1502 |
"step": 147
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
-
"entropy": 0.
|
| 1506 |
"epoch": 1.8085758039816233,
|
| 1507 |
-
"grad_norm": 0.
|
| 1508 |
"learning_rate": 2.073170731707317e-05,
|
| 1509 |
-
"loss":
|
| 1510 |
-
"mean_token_accuracy":
|
| 1511 |
-
"num_tokens":
|
| 1512 |
"step": 148
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
-
"entropy": 0.
|
| 1516 |
"epoch": 1.8208269525267995,
|
| 1517 |
-
"grad_norm":
|
| 1518 |
"learning_rate": 1.9512195121951222e-05,
|
| 1519 |
-
"loss":
|
| 1520 |
-
"mean_token_accuracy":
|
| 1521 |
-
"num_tokens":
|
| 1522 |
"step": 149
|
| 1523 |
},
|
| 1524 |
{
|
| 1525 |
-
"entropy": 0.
|
| 1526 |
"epoch": 1.8330781010719757,
|
| 1527 |
-
"grad_norm": 0.
|
| 1528 |
"learning_rate": 1.8292682926829268e-05,
|
| 1529 |
-
"loss": 0.
|
| 1530 |
-
"mean_token_accuracy":
|
| 1531 |
-
"num_tokens":
|
| 1532 |
"step": 150
|
| 1533 |
},
|
| 1534 |
{
|
| 1535 |
"epoch": 1.8330781010719757,
|
| 1536 |
-
"eval_entropy": 0.
|
| 1537 |
-
"eval_loss": 0.
|
| 1538 |
-
"eval_mean_token_accuracy": 0.
|
| 1539 |
-
"eval_num_tokens":
|
| 1540 |
-
"eval_runtime":
|
| 1541 |
-
"eval_samples_per_second": 1.
|
| 1542 |
-
"eval_steps_per_second": 1.
|
| 1543 |
"step": 150
|
| 1544 |
}
|
| 1545 |
],
|
|
@@ -1560,7 +1560,7 @@
|
|
| 1560 |
"attributes": {}
|
| 1561 |
}
|
| 1562 |
},
|
| 1563 |
-
"total_flos":
|
| 1564 |
"train_batch_size": 1,
|
| 1565 |
"trial_name": null,
|
| 1566 |
"trial_params": null
|
|
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"entropy": 0.22490596678107977,
|
| 14 |
"epoch": 0.01225114854517611,
|
| 15 |
+
"grad_norm": 8.875,
|
| 16 |
"learning_rate": 0.0002,
|
| 17 |
+
"loss": 0.1875426322221756,
|
| 18 |
+
"mean_token_accuracy": 0.9461580626666546,
|
| 19 |
+
"num_tokens": 6770.0,
|
| 20 |
"step": 1
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"entropy": 0.24707041680812836,
|
| 24 |
"epoch": 0.02450229709035222,
|
| 25 |
+
"grad_norm": 8.4375,
|
| 26 |
"learning_rate": 0.00019878048780487805,
|
| 27 |
+
"loss": 0.16050274670124054,
|
| 28 |
+
"mean_token_accuracy": 0.9445944800972939,
|
| 29 |
+
"num_tokens": 14234.0,
|
| 30 |
"step": 2
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"entropy": 0.32129648607224226,
|
| 34 |
"epoch": 0.036753445635528334,
|
| 35 |
+
"grad_norm": 2.75,
|
| 36 |
"learning_rate": 0.0001975609756097561,
|
| 37 |
+
"loss": 0.09863867610692978,
|
| 38 |
+
"mean_token_accuracy": 0.9659304060041904,
|
| 39 |
+
"num_tokens": 20673.0,
|
| 40 |
"step": 3
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"entropy": 0.32960685156285763,
|
| 44 |
"epoch": 0.04900459418070444,
|
| 45 |
+
"grad_norm": 1.671875,
|
| 46 |
"learning_rate": 0.00019634146341463416,
|
| 47 |
+
"loss": 0.08542143553495407,
|
| 48 |
+
"mean_token_accuracy": 0.9690693095326424,
|
| 49 |
+
"num_tokens": 26890.0,
|
| 50 |
"step": 4
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"entropy": 0.2677983660250902,
|
| 54 |
"epoch": 0.06125574272588055,
|
| 55 |
+
"grad_norm": 1.359375,
|
| 56 |
"learning_rate": 0.0001951219512195122,
|
| 57 |
+
"loss": 0.08666501939296722,
|
| 58 |
+
"mean_token_accuracy": 0.968298003077507,
|
| 59 |
+
"num_tokens": 35017.0,
|
| 60 |
"step": 5
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"entropy": 0.3096502358093858,
|
| 64 |
"epoch": 0.07350689127105667,
|
| 65 |
+
"grad_norm": 0.66015625,
|
| 66 |
"learning_rate": 0.00019390243902439025,
|
| 67 |
+
"loss": 0.07875043898820877,
|
| 68 |
+
"mean_token_accuracy": 0.969221331179142,
|
| 69 |
+
"num_tokens": 41478.0,
|
| 70 |
"step": 6
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"entropy": 0.3156957607716322,
|
| 74 |
"epoch": 0.08575803981623277,
|
| 75 |
+
"grad_norm": 2.0,
|
| 76 |
"learning_rate": 0.0001926829268292683,
|
| 77 |
+
"loss": 0.07807251811027527,
|
| 78 |
+
"mean_token_accuracy": 0.9681689888238907,
|
| 79 |
+
"num_tokens": 48204.0,
|
| 80 |
"step": 7
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"entropy": 0.2759731076657772,
|
| 84 |
"epoch": 0.09800918836140889,
|
| 85 |
+
"grad_norm": 1.1328125,
|
| 86 |
"learning_rate": 0.00019146341463414633,
|
| 87 |
+
"loss": 0.07681904733181,
|
| 88 |
+
"mean_token_accuracy": 0.9719767943024635,
|
| 89 |
+
"num_tokens": 54668.0,
|
| 90 |
"step": 8
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"entropy": 0.24453612882643938,
|
| 94 |
"epoch": 0.11026033690658499,
|
| 95 |
+
"grad_norm": 0.875,
|
| 96 |
"learning_rate": 0.0001902439024390244,
|
| 97 |
+
"loss": 0.07310224324464798,
|
| 98 |
+
"mean_token_accuracy": 0.96934475004673,
|
| 99 |
+
"num_tokens": 61929.0,
|
| 100 |
"step": 9
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"entropy": 0.25852775294333696,
|
| 104 |
"epoch": 0.1225114854517611,
|
| 105 |
+
"grad_norm": 1.4921875,
|
| 106 |
"learning_rate": 0.00018902439024390244,
|
| 107 |
+
"loss": 0.07384984195232391,
|
| 108 |
+
"mean_token_accuracy": 0.9701811708509922,
|
| 109 |
+
"num_tokens": 69036.0,
|
| 110 |
"step": 10
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"entropy": 0.27396084927022457,
|
| 114 |
"epoch": 0.13476263399693722,
|
| 115 |
+
"grad_norm": 0.94140625,
|
| 116 |
"learning_rate": 0.0001878048780487805,
|
| 117 |
+
"loss": 0.10277765244245529,
|
| 118 |
+
"mean_token_accuracy": 0.9634475558996201,
|
| 119 |
+
"num_tokens": 76394.0,
|
| 120 |
"step": 11
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"entropy": 0.3001147015020251,
|
| 124 |
"epoch": 0.14701378254211334,
|
| 125 |
+
"grad_norm": 0.84765625,
|
| 126 |
"learning_rate": 0.00018658536585365856,
|
| 127 |
+
"loss": 0.08927591890096664,
|
| 128 |
+
"mean_token_accuracy": 0.9625685028731823,
|
| 129 |
+
"num_tokens": 84073.0,
|
| 130 |
"step": 12
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"entropy": 0.29679975286126137,
|
| 134 |
"epoch": 0.15926493108728942,
|
| 135 |
+
"grad_norm": 0.8359375,
|
| 136 |
"learning_rate": 0.0001853658536585366,
|
| 137 |
+
"loss": 0.10607243329286575,
|
| 138 |
+
"mean_token_accuracy": 0.9608454070985317,
|
| 139 |
+
"num_tokens": 91135.0,
|
| 140 |
"step": 13
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"entropy": 0.28288435423746705,
|
| 144 |
"epoch": 0.17151607963246554,
|
| 145 |
+
"grad_norm": 0.69921875,
|
| 146 |
"learning_rate": 0.00018414634146341464,
|
| 147 |
+
"loss": 0.07875586301088333,
|
| 148 |
+
"mean_token_accuracy": 0.9699672348797321,
|
| 149 |
+
"num_tokens": 97740.0,
|
| 150 |
"step": 14
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"entropy": 0.2927901232615113,
|
| 154 |
"epoch": 0.18376722817764166,
|
| 155 |
+
"grad_norm": 0.8515625,
|
| 156 |
"learning_rate": 0.0001829268292682927,
|
| 157 |
+
"loss": 0.08531365543603897,
|
| 158 |
+
"mean_token_accuracy": 0.9704407565295696,
|
| 159 |
+
"num_tokens": 104242.0,
|
| 160 |
"step": 15
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"entropy": 0.2802786426618695,
|
| 164 |
"epoch": 0.19601837672281777,
|
| 165 |
+
"grad_norm": 2.609375,
|
| 166 |
"learning_rate": 0.00018170731707317075,
|
| 167 |
+
"loss": 0.10073477029800415,
|
| 168 |
+
"mean_token_accuracy": 0.9676352478563786,
|
| 169 |
+
"num_tokens": 112245.0,
|
| 170 |
"step": 16
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"entropy": 0.30663597770035267,
|
| 174 |
"epoch": 0.2082695252679939,
|
| 175 |
+
"grad_norm": 2.3125,
|
| 176 |
"learning_rate": 0.0001804878048780488,
|
| 177 |
+
"loss": 0.11375448107719421,
|
| 178 |
+
"mean_token_accuracy": 0.9604234844446182,
|
| 179 |
+
"num_tokens": 119086.0,
|
| 180 |
"step": 17
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"entropy": 0.31370354909449816,
|
| 184 |
"epoch": 0.22052067381316998,
|
| 185 |
+
"grad_norm": 2.3125,
|
| 186 |
"learning_rate": 0.00017926829268292684,
|
| 187 |
+
"loss": 0.10826913267374039,
|
| 188 |
+
"mean_token_accuracy": 0.9599097929894924,
|
| 189 |
+
"num_tokens": 126539.0,
|
| 190 |
"step": 18
|
| 191 |
},
|
| 192 |
{
|
| 193 |
+
"entropy": 0.3203959669917822,
|
| 194 |
"epoch": 0.2327718223583461,
|
| 195 |
+
"grad_norm": 1.3203125,
|
| 196 |
"learning_rate": 0.00017804878048780488,
|
| 197 |
+
"loss": 0.075275719165802,
|
| 198 |
+
"mean_token_accuracy": 0.9775180667638779,
|
| 199 |
+
"num_tokens": 133104.0,
|
| 200 |
"step": 19
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"entropy": 0.32591533567756414,
|
| 204 |
"epoch": 0.2450229709035222,
|
| 205 |
+
"grad_norm": 0.921875,
|
| 206 |
"learning_rate": 0.00017682926829268295,
|
| 207 |
+
"loss": 0.09778374433517456,
|
| 208 |
+
"mean_token_accuracy": 0.9647064991295338,
|
| 209 |
+
"num_tokens": 139853.0,
|
| 210 |
"step": 20
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"entropy": 0.3228916050866246,
|
| 214 |
"epoch": 0.2572741194486983,
|
| 215 |
+
"grad_norm": 1.9453125,
|
| 216 |
"learning_rate": 0.000175609756097561,
|
| 217 |
+
"loss": 0.11495943367481232,
|
| 218 |
+
"mean_token_accuracy": 0.9573761746287346,
|
| 219 |
+
"num_tokens": 146482.0,
|
| 220 |
"step": 21
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"entropy": 0.3363859634846449,
|
| 224 |
"epoch": 0.26952526799387444,
|
| 225 |
+
"grad_norm": 1.5,
|
| 226 |
"learning_rate": 0.00017439024390243903,
|
| 227 |
+
"loss": 0.10473912209272385,
|
| 228 |
+
"mean_token_accuracy": 0.9618786759674549,
|
| 229 |
+
"num_tokens": 153819.0,
|
| 230 |
"step": 22
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"entropy": 0.3069695383310318,
|
| 234 |
"epoch": 0.28177641653905056,
|
| 235 |
+
"grad_norm": 1.3203125,
|
| 236 |
"learning_rate": 0.00017317073170731708,
|
| 237 |
+
"loss": 0.09256276488304138,
|
| 238 |
+
"mean_token_accuracy": 0.9625396281480789,
|
| 239 |
+
"num_tokens": 160972.0,
|
| 240 |
"step": 23
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"entropy": 0.3574997428804636,
|
| 244 |
"epoch": 0.29402756508422667,
|
| 245 |
+
"grad_norm": 0.62890625,
|
| 246 |
"learning_rate": 0.00017195121951219512,
|
| 247 |
+
"loss": 0.09489140659570694,
|
| 248 |
+
"mean_token_accuracy": 0.9578843042254448,
|
| 249 |
+
"num_tokens": 167730.0,
|
| 250 |
"step": 24
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"entropy": 0.3444826593622565,
|
| 254 |
"epoch": 0.30627871362940273,
|
| 255 |
+
"grad_norm": 0.56640625,
|
| 256 |
"learning_rate": 0.0001707317073170732,
|
| 257 |
+
"loss": 0.09492132067680359,
|
| 258 |
+
"mean_token_accuracy": 0.9603794105350971,
|
| 259 |
+
"num_tokens": 174078.0,
|
| 260 |
"step": 25
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"entropy": 0.328093777410686,
|
| 264 |
"epoch": 0.31852986217457885,
|
| 265 |
+
"grad_norm": 0.97265625,
|
| 266 |
"learning_rate": 0.00016951219512195123,
|
| 267 |
+
"loss": 0.08727280050516129,
|
| 268 |
+
"mean_token_accuracy": 0.9661480598151684,
|
| 269 |
+
"num_tokens": 180867.0,
|
| 270 |
"step": 26
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"entropy": 0.3213672311976552,
|
| 274 |
"epoch": 0.33078101071975496,
|
| 275 |
+
"grad_norm": 1.203125,
|
| 276 |
"learning_rate": 0.00016829268292682927,
|
| 277 |
+
"loss": 0.07705243676900864,
|
| 278 |
+
"mean_token_accuracy": 0.9675347730517387,
|
| 279 |
+
"num_tokens": 187459.0,
|
| 280 |
"step": 27
|
| 281 |
},
|
| 282 |
{
|
| 283 |
+
"entropy": 0.3209801884368062,
|
| 284 |
"epoch": 0.3430321592649311,
|
| 285 |
+
"grad_norm": 0.49609375,
|
| 286 |
"learning_rate": 0.00016707317073170731,
|
| 287 |
+
"loss": 0.08744930475950241,
|
| 288 |
+
"mean_token_accuracy": 0.9658873043954372,
|
| 289 |
+
"num_tokens": 194265.0,
|
| 290 |
"step": 28
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"entropy": 0.2975130006670952,
|
| 294 |
"epoch": 0.3552833078101072,
|
| 295 |
+
"grad_norm": 0.578125,
|
| 296 |
"learning_rate": 0.00016585365853658536,
|
| 297 |
+
"loss": 0.08422811329364777,
|
| 298 |
+
"mean_token_accuracy": 0.9715595282614231,
|
| 299 |
+
"num_tokens": 201332.0,
|
| 300 |
"step": 29
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"entropy": 0.29833013843744993,
|
| 304 |
"epoch": 0.3675344563552833,
|
| 305 |
+
"grad_norm": 0.82421875,
|
| 306 |
"learning_rate": 0.00016463414634146343,
|
| 307 |
+
"loss": 0.08079958707094193,
|
| 308 |
+
"mean_token_accuracy": 0.9676232784986496,
|
| 309 |
+
"num_tokens": 208902.0,
|
| 310 |
"step": 30
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"entropy": 0.31810148898512125,
|
| 314 |
"epoch": 0.37978560490045943,
|
| 315 |
+
"grad_norm": 0.6796875,
|
| 316 |
"learning_rate": 0.00016341463414634147,
|
| 317 |
+
"loss": 0.09296617656946182,
|
| 318 |
+
"mean_token_accuracy": 0.9628731682896614,
|
| 319 |
+
"num_tokens": 214635.0,
|
| 320 |
"step": 31
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"entropy": 0.2774961022660136,
|
| 324 |
"epoch": 0.39203675344563554,
|
| 325 |
+
"grad_norm": 1.2109375,
|
| 326 |
"learning_rate": 0.00016219512195121954,
|
| 327 |
+
"loss": 0.08057809621095657,
|
| 328 |
+
"mean_token_accuracy": 0.9683544635772705,
|
| 329 |
+
"num_tokens": 222703.0,
|
| 330 |
"step": 32
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"entropy": 0.2500351797789335,
|
| 334 |
"epoch": 0.40428790199081166,
|
| 335 |
+
"grad_norm": 0.6953125,
|
| 336 |
"learning_rate": 0.00016097560975609758,
|
| 337 |
+
"loss": 0.07790188491344452,
|
| 338 |
+
"mean_token_accuracy": 0.9730625562369823,
|
| 339 |
+
"num_tokens": 230136.0,
|
| 340 |
"step": 33
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"entropy": 0.27261121198534966,
|
| 344 |
"epoch": 0.4165390505359878,
|
| 345 |
+
"grad_norm": 1.21875,
|
| 346 |
"learning_rate": 0.00015975609756097562,
|
| 347 |
+
"loss": 0.08459997177124023,
|
| 348 |
+
"mean_token_accuracy": 0.9683701656758785,
|
| 349 |
+
"num_tokens": 236711.0,
|
| 350 |
"step": 34
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"entropy": 0.25461648125201464,
|
| 354 |
"epoch": 0.42879019908116384,
|
| 355 |
+
"grad_norm": 1.5078125,
|
| 356 |
"learning_rate": 0.00015853658536585366,
|
| 357 |
+
"loss": 0.09788602590560913,
|
| 358 |
+
"mean_token_accuracy": 0.9601947516202927,
|
| 359 |
+
"num_tokens": 243492.0,
|
| 360 |
"step": 35
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"entropy": 0.250462488271296,
|
| 364 |
"epoch": 0.44104134762633995,
|
| 365 |
+
"grad_norm": 0.62109375,
|
| 366 |
"learning_rate": 0.00015731707317073173,
|
| 367 |
+
"loss": 0.09664106369018555,
|
| 368 |
+
"mean_token_accuracy": 0.9635641165077686,
|
| 369 |
+
"num_tokens": 250330.0,
|
| 370 |
"step": 36
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"entropy": 0.26719998102635145,
|
| 374 |
"epoch": 0.45329249617151607,
|
| 375 |
+
"grad_norm": 0.609375,
|
| 376 |
"learning_rate": 0.00015609756097560978,
|
| 377 |
+
"loss": 0.08978135138750076,
|
| 378 |
+
"mean_token_accuracy": 0.9730992764234543,
|
| 379 |
+
"num_tokens": 257503.0,
|
| 380 |
"step": 37
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"entropy": 0.25437645614147186,
|
| 384 |
"epoch": 0.4655436447166922,
|
| 385 |
+
"grad_norm": 1.0859375,
|
| 386 |
"learning_rate": 0.00015487804878048782,
|
| 387 |
+
"loss": 0.08938639611005783,
|
| 388 |
+
"mean_token_accuracy": 0.9675878100097179,
|
| 389 |
+
"num_tokens": 264436.0,
|
| 390 |
"step": 38
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"entropy": 0.2722023595124483,
|
| 394 |
"epoch": 0.4777947932618683,
|
| 395 |
+
"grad_norm": 1.375,
|
| 396 |
"learning_rate": 0.00015365853658536586,
|
| 397 |
+
"loss": 0.07785381376743317,
|
| 398 |
+
"mean_token_accuracy": 0.9736072942614555,
|
| 399 |
+
"num_tokens": 270483.0,
|
| 400 |
"step": 39
|
| 401 |
},
|
| 402 |
{
|
| 403 |
+
"entropy": 0.3116175327450037,
|
| 404 |
"epoch": 0.4900459418070444,
|
| 405 |
+
"grad_norm": 0.65625,
|
| 406 |
"learning_rate": 0.0001524390243902439,
|
| 407 |
+
"loss": 0.09019558876752853,
|
| 408 |
+
"mean_token_accuracy": 0.9605641178786755,
|
| 409 |
+
"num_tokens": 276329.0,
|
| 410 |
"step": 40
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"entropy": 0.28687036503106356,
|
| 414 |
"epoch": 0.5022970903522205,
|
| 415 |
+
"grad_norm": 0.62890625,
|
| 416 |
"learning_rate": 0.00015121951219512197,
|
| 417 |
+
"loss": 0.0810370221734047,
|
| 418 |
+
"mean_token_accuracy": 0.9663555175065994,
|
| 419 |
+
"num_tokens": 281636.0,
|
| 420 |
"step": 41
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"entropy": 0.2999298516660929,
|
| 424 |
"epoch": 0.5145482388973966,
|
| 425 |
+
"grad_norm": 1.1484375,
|
| 426 |
"learning_rate": 0.00015000000000000001,
|
| 427 |
+
"loss": 0.06981078535318375,
|
| 428 |
+
"mean_token_accuracy": 0.9718391671776772,
|
| 429 |
+
"num_tokens": 287849.0,
|
| 430 |
"step": 42
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"entropy": 0.3097079414874315,
|
| 434 |
"epoch": 0.5267993874425727,
|
| 435 |
+
"grad_norm": 1.5546875,
|
| 436 |
"learning_rate": 0.00014878048780487806,
|
| 437 |
+
"loss": 0.09350281953811646,
|
| 438 |
+
"mean_token_accuracy": 0.9683773033320904,
|
| 439 |
+
"num_tokens": 294425.0,
|
| 440 |
"step": 43
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"entropy": 0.2796417009085417,
|
| 444 |
"epoch": 0.5390505359877489,
|
| 445 |
+
"grad_norm": 1.25,
|
| 446 |
"learning_rate": 0.0001475609756097561,
|
| 447 |
+
"loss": 0.09558023512363434,
|
| 448 |
+
"mean_token_accuracy": 0.9602576456964016,
|
| 449 |
+
"num_tokens": 301451.0,
|
| 450 |
"step": 44
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"entropy": 0.265599487349391,
|
| 454 |
"epoch": 0.5513016845329249,
|
| 455 |
+
"grad_norm": 0.59375,
|
| 456 |
"learning_rate": 0.00014634146341463414,
|
| 457 |
+
"loss": 0.07772304862737656,
|
| 458 |
+
"mean_token_accuracy": 0.9693298228085041,
|
| 459 |
+
"num_tokens": 308208.0,
|
| 460 |
"step": 45
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"entropy": 0.29693579114973545,
|
| 464 |
"epoch": 0.5635528330781011,
|
| 465 |
+
"grad_norm": 1.2109375,
|
| 466 |
"learning_rate": 0.0001451219512195122,
|
| 467 |
+
"loss": 0.09863201528787613,
|
| 468 |
+
"mean_token_accuracy": 0.963471919298172,
|
| 469 |
+
"num_tokens": 314427.0,
|
| 470 |
"step": 46
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"entropy": 0.2665130514651537,
|
| 474 |
"epoch": 0.5758039816232772,
|
| 475 |
+
"grad_norm": 0.80078125,
|
| 476 |
"learning_rate": 0.00014390243902439025,
|
| 477 |
+
"loss": 0.08794506639242172,
|
| 478 |
+
"mean_token_accuracy": 0.9714972339570522,
|
| 479 |
+
"num_tokens": 321146.0,
|
| 480 |
"step": 47
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"entropy": 0.2665897011756897,
|
| 484 |
"epoch": 0.5880551301684533,
|
| 485 |
+
"grad_norm": 0.765625,
|
| 486 |
"learning_rate": 0.0001426829268292683,
|
| 487 |
+
"loss": 0.07602453231811523,
|
| 488 |
+
"mean_token_accuracy": 0.9719848223030567,
|
| 489 |
+
"num_tokens": 326952.0,
|
| 490 |
"step": 48
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"entropy": 0.2814077762886882,
|
| 494 |
"epoch": 0.6003062787136294,
|
| 495 |
+
"grad_norm": 0.74609375,
|
| 496 |
"learning_rate": 0.00014146341463414634,
|
| 497 |
+
"loss": 0.08512163907289505,
|
| 498 |
+
"mean_token_accuracy": 0.9680779539048672,
|
| 499 |
+
"num_tokens": 333716.0,
|
| 500 |
"step": 49
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"entropy": 0.311913987621665,
|
| 504 |
"epoch": 0.6125574272588055,
|
| 505 |
+
"grad_norm": 0.52734375,
|
| 506 |
"learning_rate": 0.00014024390243902438,
|
| 507 |
+
"loss": 0.0735088661313057,
|
| 508 |
+
"mean_token_accuracy": 0.9693484716117382,
|
| 509 |
+
"num_tokens": 339075.0,
|
| 510 |
"step": 50
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 0.6125574272588055,
|
| 514 |
+
"eval_entropy": 0.2772115924652072,
|
| 515 |
+
"eval_loss": 0.08680303394794464,
|
| 516 |
+
"eval_mean_token_accuracy": 0.9665399781171826,
|
| 517 |
+
"eval_num_tokens": 339075.0,
|
| 518 |
+
"eval_runtime": 64.122,
|
| 519 |
+
"eval_samples_per_second": 1.076,
|
| 520 |
+
"eval_steps_per_second": 1.076,
|
| 521 |
"step": 50
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"entropy": 0.2713254941627383,
|
| 525 |
"epoch": 0.6248085758039816,
|
| 526 |
+
"grad_norm": 0.76953125,
|
| 527 |
"learning_rate": 0.00013902439024390245,
|
| 528 |
+
"loss": 0.07364857941865921,
|
| 529 |
+
"mean_token_accuracy": 0.9685694649815559,
|
| 530 |
+
"num_tokens": 346211.0,
|
| 531 |
"step": 51
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"entropy": 0.27622572146356106,
|
| 535 |
"epoch": 0.6370597243491577,
|
| 536 |
+
"grad_norm": 1.953125,
|
| 537 |
"learning_rate": 0.0001378048780487805,
|
| 538 |
+
"loss": 0.08796638250350952,
|
| 539 |
+
"mean_token_accuracy": 0.9678671807050705,
|
| 540 |
+
"num_tokens": 353743.0,
|
| 541 |
"step": 52
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"entropy": 0.3153565675020218,
|
| 545 |
"epoch": 0.6493108728943339,
|
| 546 |
+
"grad_norm": 1.8125,
|
| 547 |
"learning_rate": 0.00013658536585365856,
|
| 548 |
+
"loss": 0.09189874678850174,
|
| 549 |
+
"mean_token_accuracy": 0.9696005284786224,
|
| 550 |
+
"num_tokens": 360765.0,
|
| 551 |
"step": 53
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"entropy": 0.2793878586962819,
|
| 555 |
"epoch": 0.6615620214395099,
|
| 556 |
+
"grad_norm": 0.94140625,
|
| 557 |
"learning_rate": 0.0001353658536585366,
|
| 558 |
+
"loss": 0.0844489261507988,
|
| 559 |
+
"mean_token_accuracy": 0.9678577370941639,
|
| 560 |
+
"num_tokens": 366800.0,
|
| 561 |
"step": 54
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"entropy": 0.31044898089021444,
|
| 565 |
"epoch": 0.6738131699846861,
|
| 566 |
+
"grad_norm": 0.9375,
|
| 567 |
"learning_rate": 0.00013414634146341464,
|
| 568 |
+
"loss": 0.07886157184839249,
|
| 569 |
+
"mean_token_accuracy": 0.9673981033265591,
|
| 570 |
+
"num_tokens": 373439.0,
|
| 571 |
"step": 55
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"entropy": 0.27184910606592894,
|
| 575 |
"epoch": 0.6860643185298622,
|
| 576 |
+
"grad_norm": 1.3359375,
|
| 577 |
"learning_rate": 0.0001329268292682927,
|
| 578 |
+
"loss": 0.0787871852517128,
|
| 579 |
+
"mean_token_accuracy": 0.9677317887544632,
|
| 580 |
+
"num_tokens": 380492.0,
|
| 581 |
"step": 56
|
| 582 |
},
|
| 583 |
{
|
| 584 |
+
"entropy": 0.31349051371216774,
|
| 585 |
"epoch": 0.6983154670750383,
|
| 586 |
+
"grad_norm": 1.59375,
|
| 587 |
"learning_rate": 0.00013170731707317076,
|
| 588 |
+
"loss": 0.08862332254648209,
|
| 589 |
+
"mean_token_accuracy": 0.9652546346187592,
|
| 590 |
+
"num_tokens": 386711.0,
|
| 591 |
"step": 57
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"entropy": 0.2799685625359416,
|
| 595 |
"epoch": 0.7105666156202144,
|
| 596 |
+
"grad_norm": 1.5078125,
|
| 597 |
"learning_rate": 0.0001304878048780488,
|
| 598 |
+
"loss": 0.10028493404388428,
|
| 599 |
+
"mean_token_accuracy": 0.9606899172067642,
|
| 600 |
+
"num_tokens": 394124.0,
|
| 601 |
"step": 58
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"entropy": 0.2792940763756633,
|
| 605 |
"epoch": 0.7228177641653905,
|
| 606 |
+
"grad_norm": 1.5859375,
|
| 607 |
"learning_rate": 0.00012926829268292684,
|
| 608 |
+
"loss": 0.07462260127067566,
|
| 609 |
+
"mean_token_accuracy": 0.9740471467375755,
|
| 610 |
+
"num_tokens": 401499.0,
|
| 611 |
"step": 59
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"entropy": 0.29724057391285896,
|
| 615 |
"epoch": 0.7350689127105666,
|
| 616 |
+
"grad_norm": 0.95703125,
|
| 617 |
"learning_rate": 0.00012804878048780488,
|
| 618 |
+
"loss": 0.06339482963085175,
|
| 619 |
+
"mean_token_accuracy": 0.9754546955227852,
|
| 620 |
+
"num_tokens": 407443.0,
|
| 621 |
"step": 60
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"entropy": 0.2698040744289756,
|
| 625 |
"epoch": 0.7473200612557427,
|
| 626 |
+
"grad_norm": 0.60546875,
|
| 627 |
"learning_rate": 0.00012682926829268293,
|
| 628 |
+
"loss": 0.10221480578184128,
|
| 629 |
+
"mean_token_accuracy": 0.9670109152793884,
|
| 630 |
+
"num_tokens": 415471.0,
|
| 631 |
"step": 61
|
| 632 |
},
|
| 633 |
{
|
| 634 |
+
"entropy": 0.2995635373517871,
|
| 635 |
"epoch": 0.7595712098009189,
|
| 636 |
+
"grad_norm": 1.75,
|
| 637 |
"learning_rate": 0.000125609756097561,
|
| 638 |
+
"loss": 0.08588436245918274,
|
| 639 |
+
"mean_token_accuracy": 0.9686382673680782,
|
| 640 |
+
"num_tokens": 422504.0,
|
| 641 |
"step": 62
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"entropy": 0.2458120621740818,
|
| 645 |
"epoch": 0.7718223583460949,
|
| 646 |
+
"grad_norm": 2.1875,
|
| 647 |
"learning_rate": 0.00012439024390243904,
|
| 648 |
+
"loss": 0.08629653602838516,
|
| 649 |
+
"mean_token_accuracy": 0.966422975063324,
|
| 650 |
+
"num_tokens": 430143.0,
|
| 651 |
"step": 63
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"entropy": 0.2900782600045204,
|
| 655 |
"epoch": 0.7840735068912711,
|
| 656 |
+
"grad_norm": 0.9296875,
|
| 657 |
"learning_rate": 0.00012317073170731708,
|
| 658 |
+
"loss": 0.08716308325529099,
|
| 659 |
+
"mean_token_accuracy": 0.965714868158102,
|
| 660 |
+
"num_tokens": 435664.0,
|
| 661 |
"step": 64
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"entropy": 0.29250922333449125,
|
| 665 |
"epoch": 0.7963246554364471,
|
| 666 |
+
"grad_norm": 0.59375,
|
| 667 |
"learning_rate": 0.00012195121951219512,
|
| 668 |
+
"loss": 0.08158774673938751,
|
| 669 |
+
"mean_token_accuracy": 0.9694335348904133,
|
| 670 |
+
"num_tokens": 442457.0,
|
| 671 |
"step": 65
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"entropy": 0.3083174014464021,
|
| 675 |
"epoch": 0.8085758039816233,
|
| 676 |
+
"grad_norm": 0.82421875,
|
| 677 |
"learning_rate": 0.00012073170731707318,
|
| 678 |
+
"loss": 0.0988016203045845,
|
| 679 |
+
"mean_token_accuracy": 0.9648039489984512,
|
| 680 |
+
"num_tokens": 449983.0,
|
| 681 |
"step": 66
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"entropy": 0.25693165976554155,
|
| 685 |
"epoch": 0.8208269525267994,
|
| 686 |
+
"grad_norm": 0.74609375,
|
| 687 |
"learning_rate": 0.00011951219512195122,
|
| 688 |
+
"loss": 0.07928164303302765,
|
| 689 |
+
"mean_token_accuracy": 0.9698546566069126,
|
| 690 |
+
"num_tokens": 457640.0,
|
| 691 |
"step": 67
|
| 692 |
},
|
| 693 |
{
|
| 694 |
+
"entropy": 0.2752681290730834,
|
| 695 |
"epoch": 0.8330781010719756,
|
| 696 |
+
"grad_norm": 0.97265625,
|
| 697 |
"learning_rate": 0.00011829268292682926,
|
| 698 |
+
"loss": 0.07464170455932617,
|
| 699 |
+
"mean_token_accuracy": 0.9697864800691605,
|
| 700 |
+
"num_tokens": 464050.0,
|
| 701 |
"step": 68
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"entropy": 0.27110164798796177,
|
| 705 |
"epoch": 0.8453292496171516,
|
| 706 |
+
"grad_norm": 0.71875,
|
| 707 |
"learning_rate": 0.00011707317073170732,
|
| 708 |
+
"loss": 0.0718315988779068,
|
| 709 |
+
"mean_token_accuracy": 0.9709942191839218,
|
| 710 |
+
"num_tokens": 469546.0,
|
| 711 |
"step": 69
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"entropy": 0.3264527218416333,
|
| 715 |
"epoch": 0.8575803981623277,
|
| 716 |
+
"grad_norm": 0.62109375,
|
| 717 |
"learning_rate": 0.00011585365853658536,
|
| 718 |
+
"loss": 0.0866687223315239,
|
| 719 |
+
"mean_token_accuracy": 0.9700192771852016,
|
| 720 |
+
"num_tokens": 475365.0,
|
| 721 |
"step": 70
|
| 722 |
},
|
| 723 |
{
|
| 724 |
+
"entropy": 0.3122966531664133,
|
| 725 |
"epoch": 0.8698315467075038,
|
| 726 |
+
"grad_norm": 0.67578125,
|
| 727 |
"learning_rate": 0.00011463414634146342,
|
| 728 |
+
"loss": 0.06088244915008545,
|
| 729 |
+
"mean_token_accuracy": 0.9754119366407394,
|
| 730 |
+
"num_tokens": 481830.0,
|
| 731 |
"step": 71
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"entropy": 0.3018254106864333,
|
| 735 |
"epoch": 0.8820826952526799,
|
| 736 |
+
"grad_norm": 0.56640625,
|
| 737 |
"learning_rate": 0.00011341463414634146,
|
| 738 |
+
"loss": 0.08657931536436081,
|
| 739 |
+
"mean_token_accuracy": 0.9676030017435551,
|
| 740 |
+
"num_tokens": 487767.0,
|
| 741 |
"step": 72
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"entropy": 0.3276115320622921,
|
| 745 |
"epoch": 0.8943338437978561,
|
| 746 |
+
"grad_norm": 0.5078125,
|
| 747 |
"learning_rate": 0.00011219512195121953,
|
| 748 |
+
"loss": 0.08024603128433228,
|
| 749 |
+
"mean_token_accuracy": 0.9690204374492168,
|
| 750 |
+
"num_tokens": 494428.0,
|
| 751 |
"step": 73
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"entropy": 0.32397411670535803,
|
| 755 |
"epoch": 0.9065849923430321,
|
| 756 |
+
"grad_norm": 1.1015625,
|
| 757 |
"learning_rate": 0.00011097560975609757,
|
| 758 |
+
"loss": 0.07867392897605896,
|
| 759 |
+
"mean_token_accuracy": 0.9685576297342777,
|
| 760 |
+
"num_tokens": 500828.0,
|
| 761 |
"step": 74
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"entropy": 0.319146528840065,
|
| 765 |
"epoch": 0.9188361408882083,
|
| 766 |
+
"grad_norm": 0.97265625,
|
| 767 |
"learning_rate": 0.00010975609756097563,
|
| 768 |
+
"loss": 0.08432602882385254,
|
| 769 |
+
"mean_token_accuracy": 0.9689616709947586,
|
| 770 |
+
"num_tokens": 507523.0,
|
| 771 |
"step": 75
|
| 772 |
},
|
| 773 |
{
|
| 774 |
+
"entropy": 0.3080446803942323,
|
| 775 |
"epoch": 0.9310872894333844,
|
| 776 |
+
"grad_norm": 1.265625,
|
| 777 |
"learning_rate": 0.00010853658536585367,
|
| 778 |
+
"loss": 0.0796058252453804,
|
| 779 |
+
"mean_token_accuracy": 0.9683922417461872,
|
| 780 |
+
"num_tokens": 513607.0,
|
| 781 |
"step": 76
|
| 782 |
},
|
| 783 |
{
|
| 784 |
+
"entropy": 0.2667541950941086,
|
| 785 |
"epoch": 0.9433384379785605,
|
| 786 |
+
"grad_norm": 0.59375,
|
| 787 |
"learning_rate": 0.00010731707317073172,
|
| 788 |
+
"loss": 0.06495777517557144,
|
| 789 |
+
"mean_token_accuracy": 0.977863471955061,
|
| 790 |
+
"num_tokens": 521376.0,
|
| 791 |
"step": 77
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"entropy": 0.27901614736765623,
|
| 795 |
"epoch": 0.9555895865237366,
|
| 796 |
+
"grad_norm": 1.0859375,
|
| 797 |
"learning_rate": 0.00010609756097560977,
|
| 798 |
+
"loss": 0.08389777690172195,
|
| 799 |
+
"mean_token_accuracy": 0.967527512460947,
|
| 800 |
+
"num_tokens": 528624.0,
|
| 801 |
"step": 78
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"entropy": 0.2754220822826028,
|
| 805 |
"epoch": 0.9678407350689127,
|
| 806 |
+
"grad_norm": 1.3515625,
|
| 807 |
"learning_rate": 0.00010487804878048781,
|
| 808 |
+
"loss": 0.0762331560254097,
|
| 809 |
+
"mean_token_accuracy": 0.9713698588311672,
|
| 810 |
+
"num_tokens": 534817.0,
|
| 811 |
"step": 79
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"entropy": 0.2981132147833705,
|
| 815 |
"epoch": 0.9800918836140888,
|
| 816 |
+
"grad_norm": 1.4375,
|
| 817 |
"learning_rate": 0.00010365853658536586,
|
| 818 |
+
"loss": 0.07953717559576035,
|
| 819 |
+
"mean_token_accuracy": 0.967929158359766,
|
| 820 |
+
"num_tokens": 541716.0,
|
| 821 |
"step": 80
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"entropy": 0.30576920323073864,
|
| 825 |
"epoch": 0.9923430321592649,
|
| 826 |
+
"grad_norm": 1.0234375,
|
| 827 |
"learning_rate": 0.0001024390243902439,
|
| 828 |
+
"loss": 0.07800528407096863,
|
| 829 |
+
"mean_token_accuracy": 0.971219640225172,
|
| 830 |
+
"num_tokens": 548000.0,
|
| 831 |
"step": 81
|
| 832 |
},
|
| 833 |
{
|
| 834 |
+
"entropy": 0.24986045509576799,
|
| 835 |
"epoch": 1.0,
|
| 836 |
+
"grad_norm": 0.58203125,
|
| 837 |
"learning_rate": 0.00010121951219512196,
|
| 838 |
+
"loss": 0.05879032611846924,
|
| 839 |
+
"mean_token_accuracy": 0.9748349964618683,
|
| 840 |
+
"num_tokens": 552608.0,
|
| 841 |
"step": 82
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"entropy": 0.2518839007243514,
|
| 845 |
"epoch": 1.0122511485451762,
|
| 846 |
+
"grad_norm": 0.49609375,
|
| 847 |
"learning_rate": 0.0001,
|
| 848 |
+
"loss": 0.047237373888492584,
|
| 849 |
+
"mean_token_accuracy": 0.9874232485890388,
|
| 850 |
+
"num_tokens": 559523.0,
|
| 851 |
"step": 83
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"entropy": 0.2561075631529093,
|
| 855 |
"epoch": 1.0245022970903521,
|
| 856 |
+
"grad_norm": 0.65234375,
|
| 857 |
"learning_rate": 9.878048780487805e-05,
|
| 858 |
+
"loss": 0.04376941919326782,
|
| 859 |
+
"mean_token_accuracy": 0.9896520264446735,
|
| 860 |
+
"num_tokens": 566232.0,
|
| 861 |
"step": 84
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"entropy": 0.2935391655191779,
|
| 865 |
"epoch": 1.0367534456355283,
|
| 866 |
+
"grad_norm": 0.486328125,
|
| 867 |
"learning_rate": 9.75609756097561e-05,
|
| 868 |
+
"loss": 0.052017997950315475,
|
| 869 |
+
"mean_token_accuracy": 0.9823879115283489,
|
| 870 |
+
"num_tokens": 573965.0,
|
| 871 |
"step": 85
|
| 872 |
},
|
| 873 |
{
|
| 874 |
+
"entropy": 0.21971730748191476,
|
| 875 |
"epoch": 1.0490045941807045,
|
| 876 |
+
"grad_norm": 0.330078125,
|
| 877 |
"learning_rate": 9.634146341463415e-05,
|
| 878 |
+
"loss": 0.04022914543747902,
|
| 879 |
+
"mean_token_accuracy": 0.9874378368258476,
|
| 880 |
+
"num_tokens": 580768.0,
|
| 881 |
"step": 86
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"entropy": 0.23719595093280077,
|
| 885 |
"epoch": 1.0612557427258806,
|
| 886 |
+
"grad_norm": 0.68359375,
|
| 887 |
"learning_rate": 9.51219512195122e-05,
|
| 888 |
+
"loss": 0.04782414808869362,
|
| 889 |
+
"mean_token_accuracy": 0.9846052750945091,
|
| 890 |
+
"num_tokens": 588097.0,
|
| 891 |
"step": 87
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"entropy": 0.25634779036045074,
|
| 895 |
"epoch": 1.0735068912710566,
|
| 896 |
+
"grad_norm": 0.291015625,
|
| 897 |
"learning_rate": 9.390243902439024e-05,
|
| 898 |
+
"loss": 0.03357430174946785,
|
| 899 |
+
"mean_token_accuracy": 0.9895204566419125,
|
| 900 |
+
"num_tokens": 594215.0,
|
| 901 |
"step": 88
|
| 902 |
},
|
| 903 |
{
|
| 904 |
+
"entropy": 0.26507470663636923,
|
| 905 |
"epoch": 1.0857580398162328,
|
| 906 |
+
"grad_norm": 0.89453125,
|
| 907 |
"learning_rate": 9.26829268292683e-05,
|
| 908 |
+
"loss": 0.0427095852792263,
|
| 909 |
+
"mean_token_accuracy": 0.984734483063221,
|
| 910 |
+
"num_tokens": 600018.0,
|
| 911 |
"step": 89
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"entropy": 0.25531507655978203,
|
| 915 |
"epoch": 1.098009188361409,
|
| 916 |
+
"grad_norm": 0.357421875,
|
| 917 |
"learning_rate": 9.146341463414635e-05,
|
| 918 |
+
"loss": 0.04051242396235466,
|
| 919 |
+
"mean_token_accuracy": 0.9878104776144028,
|
| 920 |
+
"num_tokens": 606254.0,
|
| 921 |
"step": 90
|
| 922 |
},
|
| 923 |
{
|
| 924 |
+
"entropy": 0.26176126673817635,
|
| 925 |
"epoch": 1.110260336906585,
|
| 926 |
+
"grad_norm": 0.55078125,
|
| 927 |
"learning_rate": 9.02439024390244e-05,
|
| 928 |
+
"loss": 0.03882109373807907,
|
| 929 |
+
"mean_token_accuracy": 0.9838540144264698,
|
| 930 |
+
"num_tokens": 612316.0,
|
| 931 |
"step": 91
|
| 932 |
},
|
| 933 |
{
|
| 934 |
+
"entropy": 0.2165100760757923,
|
| 935 |
"epoch": 1.122511485451761,
|
| 936 |
+
"grad_norm": 0.3671875,
|
| 937 |
"learning_rate": 8.902439024390244e-05,
|
| 938 |
+
"loss": 0.03010629303753376,
|
| 939 |
+
"mean_token_accuracy": 0.9918084405362606,
|
| 940 |
+
"num_tokens": 619629.0,
|
| 941 |
"step": 92
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"entropy": 0.24866555724292994,
|
| 945 |
"epoch": 1.1347626339969372,
|
| 946 |
+
"grad_norm": 0.7578125,
|
| 947 |
"learning_rate": 8.78048780487805e-05,
|
| 948 |
+
"loss": 0.03892926499247551,
|
| 949 |
+
"mean_token_accuracy": 0.984953761100769,
|
| 950 |
+
"num_tokens": 625947.0,
|
| 951 |
"step": 93
|
| 952 |
},
|
| 953 |
{
|
| 954 |
+
"entropy": 0.21699398616328835,
|
| 955 |
"epoch": 1.1470137825421134,
|
| 956 |
+
"grad_norm": 0.53125,
|
| 957 |
"learning_rate": 8.658536585365854e-05,
|
| 958 |
+
"loss": 0.040178049355745316,
|
| 959 |
+
"mean_token_accuracy": 0.986099898815155,
|
| 960 |
+
"num_tokens": 632906.0,
|
| 961 |
"step": 94
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"entropy": 0.2104594809934497,
|
| 965 |
"epoch": 1.1592649310872893,
|
| 966 |
+
"grad_norm": 1.4375,
|
| 967 |
"learning_rate": 8.53658536585366e-05,
|
| 968 |
+
"loss": 0.05103502795100212,
|
| 969 |
+
"mean_token_accuracy": 0.9873828142881393,
|
| 970 |
+
"num_tokens": 639769.0,
|
| 971 |
"step": 95
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"entropy": 0.21941981185227633,
|
| 975 |
"epoch": 1.1715160796324655,
|
| 976 |
+
"grad_norm": 0.984375,
|
| 977 |
"learning_rate": 8.414634146341464e-05,
|
| 978 |
+
"loss": 0.03593335300683975,
|
| 979 |
+
"mean_token_accuracy": 0.9901031330227852,
|
| 980 |
+
"num_tokens": 646347.0,
|
| 981 |
"step": 96
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"entropy": 0.23086606059223413,
|
| 985 |
"epoch": 1.1837672281776417,
|
| 986 |
+
"grad_norm": 0.65625,
|
| 987 |
"learning_rate": 8.292682926829268e-05,
|
| 988 |
+
"loss": 0.034123439341783524,
|
| 989 |
+
"mean_token_accuracy": 0.9874096475541592,
|
| 990 |
+
"num_tokens": 652247.0,
|
| 991 |
"step": 97
|
| 992 |
},
|
| 993 |
{
|
| 994 |
+
"entropy": 0.21858725044876337,
|
| 995 |
"epoch": 1.1960183767228179,
|
| 996 |
+
"grad_norm": 0.3515625,
|
| 997 |
"learning_rate": 8.170731707317073e-05,
|
| 998 |
+
"loss": 0.03983831778168678,
|
| 999 |
+
"mean_token_accuracy": 0.9883633032441139,
|
| 1000 |
+
"num_tokens": 659620.0,
|
| 1001 |
"step": 98
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
+
"entropy": 0.2186456574127078,
|
| 1005 |
"epoch": 1.2082695252679938,
|
| 1006 |
+
"grad_norm": 0.50390625,
|
| 1007 |
"learning_rate": 8.048780487804879e-05,
|
| 1008 |
+
"loss": 0.03659169375896454,
|
| 1009 |
+
"mean_token_accuracy": 0.9874354675412178,
|
| 1010 |
+
"num_tokens": 667017.0,
|
| 1011 |
"step": 99
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
+
"entropy": 0.21289387485012412,
|
| 1015 |
"epoch": 1.22052067381317,
|
| 1016 |
+
"grad_norm": 1.2890625,
|
| 1017 |
"learning_rate": 7.926829268292683e-05,
|
| 1018 |
+
"loss": 0.09039004892110825,
|
| 1019 |
+
"mean_token_accuracy": 0.9841732494533062,
|
| 1020 |
+
"num_tokens": 673866.0,
|
| 1021 |
"step": 100
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 1.22052067381317,
|
| 1025 |
+
"eval_entropy": 0.22615607968275098,
|
| 1026 |
+
"eval_loss": 0.0748714804649353,
|
| 1027 |
+
"eval_mean_token_accuracy": 0.9701917439267256,
|
| 1028 |
+
"eval_num_tokens": 673866.0,
|
| 1029 |
+
"eval_runtime": 64.1728,
|
| 1030 |
+
"eval_samples_per_second": 1.075,
|
| 1031 |
+
"eval_steps_per_second": 1.075,
|
| 1032 |
"step": 100
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
+
"entropy": 0.20847708079963923,
|
| 1036 |
"epoch": 1.2327718223583461,
|
| 1037 |
+
"grad_norm": 0.9453125,
|
| 1038 |
"learning_rate": 7.804878048780489e-05,
|
| 1039 |
+
"loss": 0.032662514597177505,
|
| 1040 |
+
"mean_token_accuracy": 0.9919092357158661,
|
| 1041 |
+
"num_tokens": 681308.0,
|
| 1042 |
"step": 101
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
+
"entropy": 0.23787071648985147,
|
| 1046 |
"epoch": 1.245022970903522,
|
| 1047 |
+
"grad_norm": 0.859375,
|
| 1048 |
"learning_rate": 7.682926829268293e-05,
|
| 1049 |
+
"loss": 0.044949762523174286,
|
| 1050 |
+
"mean_token_accuracy": 0.987742405384779,
|
| 1051 |
+
"num_tokens": 687496.0,
|
| 1052 |
"step": 102
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
+
"entropy": 0.21969830617308617,
|
| 1056 |
"epoch": 1.2572741194486983,
|
| 1057 |
+
"grad_norm": 0.8671875,
|
| 1058 |
"learning_rate": 7.560975609756099e-05,
|
| 1059 |
+
"loss": 0.036048222333192825,
|
| 1060 |
+
"mean_token_accuracy": 0.98578891903162,
|
| 1061 |
+
"num_tokens": 694818.0,
|
| 1062 |
"step": 103
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
+
"entropy": 0.228535583242774,
|
| 1066 |
"epoch": 1.2695252679938744,
|
| 1067 |
+
"grad_norm": 1.7109375,
|
| 1068 |
"learning_rate": 7.439024390243903e-05,
|
| 1069 |
+
"loss": 0.050321951508522034,
|
| 1070 |
+
"mean_token_accuracy": 0.9846261814236641,
|
| 1071 |
+
"num_tokens": 701351.0,
|
| 1072 |
"step": 104
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
+
"entropy": 0.21918219700455666,
|
| 1076 |
"epoch": 1.2817764165390506,
|
| 1077 |
+
"grad_norm": 0.57421875,
|
| 1078 |
"learning_rate": 7.317073170731707e-05,
|
| 1079 |
+
"loss": 0.03220512717962265,
|
| 1080 |
+
"mean_token_accuracy": 0.9897662363946438,
|
| 1081 |
+
"num_tokens": 707212.0,
|
| 1082 |
"step": 105
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
+
"entropy": 0.21648676693439484,
|
| 1086 |
"epoch": 1.2940275650842268,
|
| 1087 |
+
"grad_norm": 0.921875,
|
| 1088 |
"learning_rate": 7.195121951219513e-05,
|
| 1089 |
+
"loss": 0.031827542930841446,
|
| 1090 |
+
"mean_token_accuracy": 0.9904872179031372,
|
| 1091 |
+
"num_tokens": 714524.0,
|
| 1092 |
"step": 106
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
+
"entropy": 0.20004846714437008,
|
| 1096 |
"epoch": 1.3062787136294027,
|
| 1097 |
+
"grad_norm": 1.0234375,
|
| 1098 |
"learning_rate": 7.073170731707317e-05,
|
| 1099 |
+
"loss": 0.03981270268559456,
|
| 1100 |
+
"mean_token_accuracy": 0.9861926138401031,
|
| 1101 |
+
"num_tokens": 722033.0,
|
| 1102 |
"step": 107
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
+
"entropy": 0.21497153211385012,
|
| 1106 |
"epoch": 1.318529862174579,
|
| 1107 |
+
"grad_norm": 0.53515625,
|
| 1108 |
"learning_rate": 6.951219512195122e-05,
|
| 1109 |
+
"loss": 0.03612194582819939,
|
| 1110 |
+
"mean_token_accuracy": 0.9883794784545898,
|
| 1111 |
+
"num_tokens": 728835.0,
|
| 1112 |
"step": 108
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
+
"entropy": 0.22441515233367682,
|
| 1116 |
"epoch": 1.3307810107197549,
|
| 1117 |
+
"grad_norm": 0.66796875,
|
| 1118 |
"learning_rate": 6.829268292682928e-05,
|
| 1119 |
+
"loss": 0.037204962223768234,
|
| 1120 |
+
"mean_token_accuracy": 0.9865190424025059,
|
| 1121 |
+
"num_tokens": 735463.0,
|
| 1122 |
"step": 109
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
+
"entropy": 0.21172351390123367,
|
| 1126 |
"epoch": 1.343032159264931,
|
| 1127 |
+
"grad_norm": 0.314453125,
|
| 1128 |
"learning_rate": 6.707317073170732e-05,
|
| 1129 |
+
"loss": 0.03260833024978638,
|
| 1130 |
+
"mean_token_accuracy": 0.9877017810940742,
|
| 1131 |
+
"num_tokens": 742536.0,
|
| 1132 |
"step": 110
|
| 1133 |
},
|
| 1134 |
{
|
| 1135 |
+
"entropy": 0.19597876677289605,
|
| 1136 |
"epoch": 1.3552833078101072,
|
| 1137 |
+
"grad_norm": 0.419921875,
|
| 1138 |
"learning_rate": 6.585365853658538e-05,
|
| 1139 |
+
"loss": 0.0339697040617466,
|
| 1140 |
+
"mean_token_accuracy": 0.990579642355442,
|
| 1141 |
+
"num_tokens": 749606.0,
|
| 1142 |
"step": 111
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
+
"entropy": 0.21933963894844055,
|
| 1146 |
"epoch": 1.3675344563552834,
|
| 1147 |
+
"grad_norm": 0.53515625,
|
| 1148 |
"learning_rate": 6.463414634146342e-05,
|
| 1149 |
+
"loss": 0.028515402227640152,
|
| 1150 |
+
"mean_token_accuracy": 0.9883383698761463,
|
| 1151 |
+
"num_tokens": 755287.0,
|
| 1152 |
"step": 112
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
+
"entropy": 0.21494697034358978,
|
| 1156 |
"epoch": 1.3797856049004595,
|
| 1157 |
+
"grad_norm": 0.37890625,
|
| 1158 |
"learning_rate": 6.341463414634146e-05,
|
| 1159 |
+
"loss": 0.03924579173326492,
|
| 1160 |
+
"mean_token_accuracy": 0.9876385144889355,
|
| 1161 |
+
"num_tokens": 763515.0,
|
| 1162 |
"step": 113
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
+
"entropy": 0.22842750838026404,
|
| 1166 |
"epoch": 1.3920367534456355,
|
| 1167 |
+
"grad_norm": 1.1484375,
|
| 1168 |
"learning_rate": 6.219512195121952e-05,
|
| 1169 |
+
"loss": 0.0367334708571434,
|
| 1170 |
+
"mean_token_accuracy": 0.9872251562774181,
|
| 1171 |
+
"num_tokens": 769660.0,
|
| 1172 |
"step": 114
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"entropy": 0.2147415135987103,
|
| 1176 |
"epoch": 1.4042879019908117,
|
| 1177 |
+
"grad_norm": 0.921875,
|
| 1178 |
"learning_rate": 6.097560975609756e-05,
|
| 1179 |
+
"loss": 0.030023006722331047,
|
| 1180 |
+
"mean_token_accuracy": 0.9890519753098488,
|
| 1181 |
+
"num_tokens": 777068.0,
|
| 1182 |
"step": 115
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
+
"entropy": 0.2247378919273615,
|
| 1186 |
"epoch": 1.4165390505359878,
|
| 1187 |
+
"grad_norm": 0.9375,
|
| 1188 |
"learning_rate": 5.975609756097561e-05,
|
| 1189 |
+
"loss": 0.03915408253669739,
|
| 1190 |
+
"mean_token_accuracy": 0.9883266240358353,
|
| 1191 |
+
"num_tokens": 783422.0,
|
| 1192 |
"step": 116
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"entropy": 0.19090860895812511,
|
| 1196 |
"epoch": 1.4287901990811638,
|
| 1197 |
+
"grad_norm": 0.765625,
|
| 1198 |
"learning_rate": 5.853658536585366e-05,
|
| 1199 |
+
"loss": 0.037202730774879456,
|
| 1200 |
+
"mean_token_accuracy": 0.9874398410320282,
|
| 1201 |
+
"num_tokens": 790851.0,
|
| 1202 |
"step": 117
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
+
"entropy": 0.2285028137266636,
|
| 1206 |
"epoch": 1.44104134762634,
|
| 1207 |
+
"grad_norm": 1.9140625,
|
| 1208 |
"learning_rate": 5.731707317073171e-05,
|
| 1209 |
+
"loss": 0.043229859322309494,
|
| 1210 |
+
"mean_token_accuracy": 0.9905107729136944,
|
| 1211 |
+
"num_tokens": 797801.0,
|
| 1212 |
"step": 118
|
| 1213 |
},
|
| 1214 |
{
|
| 1215 |
+
"entropy": 0.2443255502730608,
|
| 1216 |
"epoch": 1.4532924961715161,
|
| 1217 |
+
"grad_norm": 0.365234375,
|
| 1218 |
"learning_rate": 5.6097560975609764e-05,
|
| 1219 |
+
"loss": 0.04100143164396286,
|
| 1220 |
+
"mean_token_accuracy": 0.9880562499165535,
|
| 1221 |
+
"num_tokens": 804371.0,
|
| 1222 |
"step": 119
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
+
"entropy": 0.19626039918512106,
|
| 1226 |
"epoch": 1.4655436447166923,
|
| 1227 |
+
"grad_norm": 0.83984375,
|
| 1228 |
"learning_rate": 5.487804878048781e-05,
|
| 1229 |
+
"loss": 0.038516998291015625,
|
| 1230 |
+
"mean_token_accuracy": 0.988171175122261,
|
| 1231 |
+
"num_tokens": 812335.0,
|
| 1232 |
"step": 120
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
+
"entropy": 0.2181866616010666,
|
| 1236 |
"epoch": 1.4777947932618682,
|
| 1237 |
+
"grad_norm": 0.53515625,
|
| 1238 |
"learning_rate": 5.365853658536586e-05,
|
| 1239 |
+
"loss": 0.02816646918654442,
|
| 1240 |
+
"mean_token_accuracy": 0.9916124008595943,
|
| 1241 |
+
"num_tokens": 818577.0,
|
| 1242 |
"step": 121
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"entropy": 0.20635052677243948,
|
| 1246 |
"epoch": 1.4900459418070444,
|
| 1247 |
+
"grad_norm": 0.74609375,
|
| 1248 |
"learning_rate": 5.2439024390243904e-05,
|
| 1249 |
+
"loss": 0.04106622561812401,
|
| 1250 |
+
"mean_token_accuracy": 0.9839451834559441,
|
| 1251 |
+
"num_tokens": 825535.0,
|
| 1252 |
"step": 122
|
| 1253 |
},
|
| 1254 |
{
|
| 1255 |
+
"entropy": 0.21835408825427294,
|
| 1256 |
"epoch": 1.5022970903522204,
|
| 1257 |
+
"grad_norm": 0.427734375,
|
| 1258 |
"learning_rate": 5.121951219512195e-05,
|
| 1259 |
+
"loss": 0.026341412216424942,
|
| 1260 |
+
"mean_token_accuracy": 0.9940293915569782,
|
| 1261 |
+
"num_tokens": 831505.0,
|
| 1262 |
"step": 123
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
+
"entropy": 0.21729151718318462,
|
| 1266 |
"epoch": 1.5145482388973965,
|
| 1267 |
+
"grad_norm": 0.455078125,
|
| 1268 |
"learning_rate": 5e-05,
|
| 1269 |
+
"loss": 0.028432821854948997,
|
| 1270 |
+
"mean_token_accuracy": 0.9925089627504349,
|
| 1271 |
+
"num_tokens": 838385.0,
|
| 1272 |
"step": 124
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
+
"entropy": 0.23625962156802416,
|
| 1276 |
"epoch": 1.5267993874425727,
|
| 1277 |
+
"grad_norm": 0.72265625,
|
| 1278 |
"learning_rate": 4.878048780487805e-05,
|
| 1279 |
+
"loss": 0.03885198384523392,
|
| 1280 |
+
"mean_token_accuracy": 0.9883155077695847,
|
| 1281 |
+
"num_tokens": 845433.0,
|
| 1282 |
"step": 125
|
| 1283 |
},
|
| 1284 |
{
|
| 1285 |
+
"entropy": 0.21153692342340946,
|
| 1286 |
"epoch": 1.5390505359877489,
|
| 1287 |
+
"grad_norm": 0.66796875,
|
| 1288 |
"learning_rate": 4.75609756097561e-05,
|
| 1289 |
+
"loss": 0.03570759296417236,
|
| 1290 |
+
"mean_token_accuracy": 0.9910184219479561,
|
| 1291 |
+
"num_tokens": 852471.0,
|
| 1292 |
"step": 126
|
| 1293 |
},
|
| 1294 |
{
|
| 1295 |
+
"entropy": 0.23752436228096485,
|
| 1296 |
"epoch": 1.551301684532925,
|
| 1297 |
+
"grad_norm": 0.640625,
|
| 1298 |
"learning_rate": 4.634146341463415e-05,
|
| 1299 |
+
"loss": 0.028638798743486404,
|
| 1300 |
+
"mean_token_accuracy": 0.9928638078272343,
|
| 1301 |
+
"num_tokens": 858702.0,
|
| 1302 |
"step": 127
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
+
"entropy": 0.2128417994827032,
|
| 1306 |
"epoch": 1.5635528330781012,
|
| 1307 |
+
"grad_norm": 0.8828125,
|
| 1308 |
"learning_rate": 4.51219512195122e-05,
|
| 1309 |
+
"loss": 0.029636576771736145,
|
| 1310 |
+
"mean_token_accuracy": 0.9910452663898468,
|
| 1311 |
+
"num_tokens": 865325.0,
|
| 1312 |
"step": 128
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
+
"entropy": 0.216589767485857,
|
| 1316 |
"epoch": 1.5758039816232772,
|
| 1317 |
+
"grad_norm": 0.453125,
|
| 1318 |
"learning_rate": 4.390243902439025e-05,
|
| 1319 |
+
"loss": 0.03238631784915924,
|
| 1320 |
+
"mean_token_accuracy": 0.9904623441398144,
|
| 1321 |
+
"num_tokens": 871341.0,
|
| 1322 |
"step": 129
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
+
"entropy": 0.19242106284946203,
|
| 1326 |
"epoch": 1.5880551301684533,
|
| 1327 |
+
"grad_norm": 0.392578125,
|
| 1328 |
"learning_rate": 4.26829268292683e-05,
|
| 1329 |
+
"loss": 0.0261989776045084,
|
| 1330 |
+
"mean_token_accuracy": 0.9925210140645504,
|
| 1331 |
+
"num_tokens": 878973.0,
|
| 1332 |
"step": 130
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
+
"entropy": 0.22208478767424822,
|
| 1336 |
"epoch": 1.6003062787136293,
|
| 1337 |
+
"grad_norm": 0.328125,
|
| 1338 |
"learning_rate": 4.146341463414634e-05,
|
| 1339 |
+
"loss": 0.029643766582012177,
|
| 1340 |
+
"mean_token_accuracy": 0.9926025420427322,
|
| 1341 |
+
"num_tokens": 885517.0,
|
| 1342 |
"step": 131
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
+
"entropy": 0.19283092580735683,
|
| 1346 |
"epoch": 1.6125574272588055,
|
| 1347 |
+
"grad_norm": 0.423828125,
|
| 1348 |
"learning_rate": 4.0243902439024395e-05,
|
| 1349 |
+
"loss": 0.03948017954826355,
|
| 1350 |
+
"mean_token_accuracy": 0.9875317811965942,
|
| 1351 |
+
"num_tokens": 893273.0,
|
| 1352 |
"step": 132
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
+
"entropy": 0.18790056556463242,
|
| 1356 |
"epoch": 1.6248085758039816,
|
| 1357 |
+
"grad_norm": 0.625,
|
| 1358 |
"learning_rate": 3.9024390243902444e-05,
|
| 1359 |
+
"loss": 0.025747017934918404,
|
| 1360 |
+
"mean_token_accuracy": 0.9934940375387669,
|
| 1361 |
+
"num_tokens": 900019.0,
|
| 1362 |
"step": 133
|
| 1363 |
},
|
| 1364 |
{
|
| 1365 |
+
"entropy": 0.20814241049811244,
|
| 1366 |
"epoch": 1.6370597243491578,
|
| 1367 |
+
"grad_norm": 0.376953125,
|
| 1368 |
"learning_rate": 3.780487804878049e-05,
|
| 1369 |
+
"loss": 0.03998865559697151,
|
| 1370 |
+
"mean_token_accuracy": 0.9876968078315258,
|
| 1371 |
+
"num_tokens": 906633.0,
|
| 1372 |
"step": 134
|
| 1373 |
},
|
| 1374 |
{
|
| 1375 |
+
"entropy": 0.1975369704887271,
|
| 1376 |
"epoch": 1.649310872894334,
|
| 1377 |
+
"grad_norm": 0.3203125,
|
| 1378 |
"learning_rate": 3.6585365853658535e-05,
|
| 1379 |
+
"loss": 0.031131668016314507,
|
| 1380 |
+
"mean_token_accuracy": 0.9915927015244961,
|
| 1381 |
+
"num_tokens": 913990.0,
|
| 1382 |
"step": 135
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
+
"entropy": 0.23459685500711203,
|
| 1386 |
"epoch": 1.66156202143951,
|
| 1387 |
+
"grad_norm": 0.76171875,
|
| 1388 |
"learning_rate": 3.5365853658536584e-05,
|
| 1389 |
+
"loss": 0.03373259678483009,
|
| 1390 |
+
"mean_token_accuracy": 0.9898596629500389,
|
| 1391 |
+
"num_tokens": 919248.0,
|
| 1392 |
"step": 136
|
| 1393 |
},
|
| 1394 |
{
|
| 1395 |
+
"entropy": 0.1909911371767521,
|
| 1396 |
"epoch": 1.673813169984686,
|
| 1397 |
+
"grad_norm": 0.60546875,
|
| 1398 |
"learning_rate": 3.414634146341464e-05,
|
| 1399 |
+
"loss": 0.037791188806295395,
|
| 1400 |
+
"mean_token_accuracy": 0.9897548258304596,
|
| 1401 |
+
"num_tokens": 926248.0,
|
| 1402 |
"step": 137
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
+
"entropy": 0.2332595670595765,
|
| 1406 |
"epoch": 1.686064318529862,
|
| 1407 |
+
"grad_norm": 0.89453125,
|
| 1408 |
"learning_rate": 3.292682926829269e-05,
|
| 1409 |
+
"loss": 0.03799242898821831,
|
| 1410 |
+
"mean_token_accuracy": 0.9867184162139893,
|
| 1411 |
+
"num_tokens": 932490.0,
|
| 1412 |
"step": 138
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
+
"entropy": 0.22243124432861805,
|
| 1416 |
"epoch": 1.6983154670750382,
|
| 1417 |
+
"grad_norm": 0.61328125,
|
| 1418 |
"learning_rate": 3.170731707317073e-05,
|
| 1419 |
+
"loss": 0.04291514679789543,
|
| 1420 |
+
"mean_token_accuracy": 0.9877815246582031,
|
| 1421 |
+
"num_tokens": 938756.0,
|
| 1422 |
"step": 139
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
+
"entropy": 0.20778016652911901,
|
| 1426 |
"epoch": 1.7105666156202144,
|
| 1427 |
+
"grad_norm": 0.41796875,
|
| 1428 |
"learning_rate": 3.048780487804878e-05,
|
| 1429 |
+
"loss": 0.023588458076119423,
|
| 1430 |
+
"mean_token_accuracy": 0.9942950084805489,
|
| 1431 |
+
"num_tokens": 945866.0,
|
| 1432 |
"step": 140
|
| 1433 |
},
|
| 1434 |
{
|
| 1435 |
+
"entropy": 0.18776059616357088,
|
| 1436 |
"epoch": 1.7228177641653906,
|
| 1437 |
+
"grad_norm": 0.41796875,
|
| 1438 |
"learning_rate": 2.926829268292683e-05,
|
| 1439 |
+
"loss": 0.03229852020740509,
|
| 1440 |
+
"mean_token_accuracy": 0.9909596405923367,
|
| 1441 |
+
"num_tokens": 952865.0,
|
| 1442 |
"step": 141
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
+
"entropy": 0.18707702960819006,
|
| 1446 |
"epoch": 1.7350689127105667,
|
| 1447 |
+
"grad_norm": 0.609375,
|
| 1448 |
"learning_rate": 2.8048780487804882e-05,
|
| 1449 |
+
"loss": 0.03691868111491203,
|
| 1450 |
+
"mean_token_accuracy": 0.9900590926408768,
|
| 1451 |
+
"num_tokens": 959190.0,
|
| 1452 |
"step": 142
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
+
"entropy": 0.1914756903424859,
|
| 1456 |
"epoch": 1.7473200612557427,
|
| 1457 |
+
"grad_norm": 0.408203125,
|
| 1458 |
"learning_rate": 2.682926829268293e-05,
|
| 1459 |
+
"loss": 0.03487441688776016,
|
| 1460 |
+
"mean_token_accuracy": 0.9909356310963631,
|
| 1461 |
+
"num_tokens": 966059.0,
|
| 1462 |
"step": 143
|
| 1463 |
},
|
| 1464 |
{
|
| 1465 |
+
"entropy": 0.20852853963151574,
|
| 1466 |
"epoch": 1.7595712098009189,
|
| 1467 |
+
"grad_norm": 0.380859375,
|
| 1468 |
"learning_rate": 2.5609756097560977e-05,
|
| 1469 |
+
"loss": 0.03023841790854931,
|
| 1470 |
+
"mean_token_accuracy": 0.9922478385269642,
|
| 1471 |
+
"num_tokens": 973553.0,
|
| 1472 |
"step": 144
|
| 1473 |
},
|
| 1474 |
{
|
| 1475 |
+
"entropy": 0.18278094567358494,
|
| 1476 |
"epoch": 1.7718223583460948,
|
| 1477 |
+
"grad_norm": 0.65625,
|
| 1478 |
"learning_rate": 2.4390243902439026e-05,
|
| 1479 |
+
"loss": 0.03335383161902428,
|
| 1480 |
+
"mean_token_accuracy": 0.9902437664568424,
|
| 1481 |
+
"num_tokens": 980748.0,
|
| 1482 |
"step": 145
|
| 1483 |
},
|
| 1484 |
{
|
| 1485 |
+
"entropy": 0.2156418706290424,
|
| 1486 |
"epoch": 1.784073506891271,
|
| 1487 |
+
"grad_norm": 0.58984375,
|
| 1488 |
"learning_rate": 2.3170731707317075e-05,
|
| 1489 |
+
"loss": 0.026211977005004883,
|
| 1490 |
+
"mean_token_accuracy": 0.9913386814296246,
|
| 1491 |
+
"num_tokens": 987018.0,
|
| 1492 |
"step": 146
|
| 1493 |
},
|
| 1494 |
{
|
| 1495 |
+
"entropy": 0.2084086169488728,
|
| 1496 |
"epoch": 1.7963246554364471,
|
| 1497 |
+
"grad_norm": 0.37890625,
|
| 1498 |
"learning_rate": 2.1951219512195124e-05,
|
| 1499 |
+
"loss": 0.029074503108859062,
|
| 1500 |
+
"mean_token_accuracy": 0.9920879267156124,
|
| 1501 |
+
"num_tokens": 993841.0,
|
| 1502 |
"step": 147
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
+
"entropy": 0.2162067350000143,
|
| 1506 |
"epoch": 1.8085758039816233,
|
| 1507 |
+
"grad_norm": 0.38671875,
|
| 1508 |
"learning_rate": 2.073170731707317e-05,
|
| 1509 |
+
"loss": 0.027591165155172348,
|
| 1510 |
+
"mean_token_accuracy": 0.9916894063353539,
|
| 1511 |
+
"num_tokens": 1000318.0,
|
| 1512 |
"step": 148
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
+
"entropy": 0.22895692195743322,
|
| 1516 |
"epoch": 1.8208269525267995,
|
| 1517 |
+
"grad_norm": 1.421875,
|
| 1518 |
"learning_rate": 1.9512195121951222e-05,
|
| 1519 |
+
"loss": 0.034101299941539764,
|
| 1520 |
+
"mean_token_accuracy": 0.9889856353402138,
|
| 1521 |
+
"num_tokens": 1005747.0,
|
| 1522 |
"step": 149
|
| 1523 |
},
|
| 1524 |
{
|
| 1525 |
+
"entropy": 0.21029841899871826,
|
| 1526 |
"epoch": 1.8330781010719757,
|
| 1527 |
+
"grad_norm": 0.59375,
|
| 1528 |
"learning_rate": 1.8292682926829268e-05,
|
| 1529 |
+
"loss": 0.04408642649650574,
|
| 1530 |
+
"mean_token_accuracy": 0.988445583730936,
|
| 1531 |
+
"num_tokens": 1013365.0,
|
| 1532 |
"step": 150
|
| 1533 |
},
|
| 1534 |
{
|
| 1535 |
"epoch": 1.8330781010719757,
|
| 1536 |
+
"eval_entropy": 0.21028992522885834,
|
| 1537 |
+
"eval_loss": 0.06481878459453583,
|
| 1538 |
+
"eval_mean_token_accuracy": 0.9753203677094501,
|
| 1539 |
+
"eval_num_tokens": 1013365.0,
|
| 1540 |
+
"eval_runtime": 64.1224,
|
| 1541 |
+
"eval_samples_per_second": 1.076,
|
| 1542 |
+
"eval_steps_per_second": 1.076,
|
| 1543 |
"step": 150
|
| 1544 |
}
|
| 1545 |
],
|
|
|
|
| 1560 |
"attributes": {}
|
| 1561 |
}
|
| 1562 |
},
|
| 1563 |
+
"total_flos": 4.588638959382528e+16,
|
| 1564 |
"train_batch_size": 1,
|
| 1565 |
"trial_name": null,
|
| 1566 |
"trial_params": null
|
checkpoint-164/adapter_config.json
CHANGED
|
@@ -30,11 +30,11 @@
|
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
"o_proj",
|
| 33 |
-
"q_proj",
|
| 34 |
-
"gate_proj",
|
| 35 |
-
"down_proj",
|
| 36 |
"k_proj",
|
|
|
|
|
|
|
| 37 |
"v_proj",
|
|
|
|
| 38 |
"up_proj"
|
| 39 |
],
|
| 40 |
"target_parameters": null,
|
|
|
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
"o_proj",
|
|
|
|
|
|
|
|
|
|
| 33 |
"k_proj",
|
| 34 |
+
"down_proj",
|
| 35 |
+
"q_proj",
|
| 36 |
"v_proj",
|
| 37 |
+
"gate_proj",
|
| 38 |
"up_proj"
|
| 39 |
],
|
| 40 |
"target_parameters": null,
|
checkpoint-164/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83946192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d7799ff7ae7f290e67eada9d323c6418a3a9db26bdb2158f039838076d95d1f
|
| 3 |
size 83946192
|
checkpoint-164/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 85728997
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d6acb58aa21bd3fa4437c7d5eee173fd6389daa33a68a747aec69733fea274a
|
| 3 |
size 85728997
|
checkpoint-164/trainer_state.json
CHANGED
|
@@ -10,1676 +10,1676 @@
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"entropy": 0.
|
| 14 |
"epoch": 0.01225114854517611,
|
| 15 |
-
"grad_norm":
|
| 16 |
"learning_rate": 0.0002,
|
| 17 |
-
"loss": 0.
|
| 18 |
-
"mean_token_accuracy": 0.
|
| 19 |
-
"num_tokens":
|
| 20 |
"step": 1
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"entropy": 0.
|
| 24 |
"epoch": 0.02450229709035222,
|
| 25 |
-
"grad_norm":
|
| 26 |
"learning_rate": 0.00019878048780487805,
|
| 27 |
-
"loss":
|
| 28 |
-
"mean_token_accuracy":
|
| 29 |
-
"num_tokens":
|
| 30 |
"step": 2
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"entropy": 0.
|
| 34 |
"epoch": 0.036753445635528334,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 0.0001975609756097561,
|
| 37 |
-
"loss":
|
| 38 |
-
"mean_token_accuracy":
|
| 39 |
-
"num_tokens":
|
| 40 |
"step": 3
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"entropy": 0.
|
| 44 |
"epoch": 0.04900459418070444,
|
| 45 |
-
"grad_norm":
|
| 46 |
"learning_rate": 0.00019634146341463416,
|
| 47 |
-
"loss": 0.
|
| 48 |
-
"mean_token_accuracy": 0.
|
| 49 |
-
"num_tokens":
|
| 50 |
"step": 4
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"entropy": 0.
|
| 54 |
"epoch": 0.06125574272588055,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 0.0001951219512195122,
|
| 57 |
-
"loss":
|
| 58 |
-
"mean_token_accuracy":
|
| 59 |
-
"num_tokens":
|
| 60 |
"step": 5
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"entropy": 0.
|
| 64 |
"epoch": 0.07350689127105667,
|
| 65 |
-
"grad_norm":
|
| 66 |
"learning_rate": 0.00019390243902439025,
|
| 67 |
-
"loss":
|
| 68 |
-
"mean_token_accuracy":
|
| 69 |
-
"num_tokens":
|
| 70 |
"step": 6
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"entropy": 0.
|
| 74 |
"epoch": 0.08575803981623277,
|
| 75 |
-
"grad_norm":
|
| 76 |
"learning_rate": 0.0001926829268292683,
|
| 77 |
-
"loss":
|
| 78 |
-
"mean_token_accuracy":
|
| 79 |
-
"num_tokens":
|
| 80 |
"step": 7
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"entropy": 0.
|
| 84 |
"epoch": 0.09800918836140889,
|
| 85 |
-
"grad_norm":
|
| 86 |
"learning_rate": 0.00019146341463414633,
|
| 87 |
-
"loss":
|
| 88 |
-
"mean_token_accuracy":
|
| 89 |
-
"num_tokens":
|
| 90 |
"step": 8
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"entropy": 0.
|
| 94 |
"epoch": 0.11026033690658499,
|
| 95 |
-
"grad_norm": 0.
|
| 96 |
"learning_rate": 0.0001902439024390244,
|
| 97 |
-
"loss": 0.
|
| 98 |
-
"mean_token_accuracy": 0.
|
| 99 |
-
"num_tokens":
|
| 100 |
"step": 9
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"entropy": 0.
|
| 104 |
"epoch": 0.1225114854517611,
|
| 105 |
-
"grad_norm":
|
| 106 |
"learning_rate": 0.00018902439024390244,
|
| 107 |
-
"loss": 0.
|
| 108 |
-
"mean_token_accuracy": 0.
|
| 109 |
-
"num_tokens":
|
| 110 |
"step": 10
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"entropy": 0.
|
| 114 |
"epoch": 0.13476263399693722,
|
| 115 |
-
"grad_norm": 0.
|
| 116 |
"learning_rate": 0.0001878048780487805,
|
| 117 |
-
"loss":
|
| 118 |
-
"mean_token_accuracy":
|
| 119 |
-
"num_tokens":
|
| 120 |
"step": 11
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"entropy": 0.
|
| 124 |
"epoch": 0.14701378254211334,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
"learning_rate": 0.00018658536585365856,
|
| 127 |
-
"loss":
|
| 128 |
-
"mean_token_accuracy":
|
| 129 |
-
"num_tokens":
|
| 130 |
"step": 12
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"entropy": 0.
|
| 134 |
"epoch": 0.15926493108728942,
|
| 135 |
-
"grad_norm": 0.
|
| 136 |
"learning_rate": 0.0001853658536585366,
|
| 137 |
-
"loss":
|
| 138 |
-
"mean_token_accuracy":
|
| 139 |
-
"num_tokens":
|
| 140 |
"step": 13
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"entropy": 0.
|
| 144 |
"epoch": 0.17151607963246554,
|
| 145 |
-
"grad_norm":
|
| 146 |
"learning_rate": 0.00018414634146341464,
|
| 147 |
-
"loss":
|
| 148 |
-
"mean_token_accuracy":
|
| 149 |
-
"num_tokens":
|
| 150 |
"step": 14
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"entropy": 0.
|
| 154 |
"epoch": 0.18376722817764166,
|
| 155 |
-
"grad_norm":
|
| 156 |
"learning_rate": 0.0001829268292682927,
|
| 157 |
-
"loss":
|
| 158 |
-
"mean_token_accuracy":
|
| 159 |
-
"num_tokens":
|
| 160 |
"step": 15
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"entropy": 0.
|
| 164 |
"epoch": 0.19601837672281777,
|
| 165 |
-
"grad_norm":
|
| 166 |
"learning_rate": 0.00018170731707317075,
|
| 167 |
-
"loss":
|
| 168 |
-
"mean_token_accuracy":
|
| 169 |
-
"num_tokens":
|
| 170 |
"step": 16
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"entropy": 0.
|
| 174 |
"epoch": 0.2082695252679939,
|
| 175 |
-
"grad_norm":
|
| 176 |
"learning_rate": 0.0001804878048780488,
|
| 177 |
-
"loss":
|
| 178 |
-
"mean_token_accuracy":
|
| 179 |
-
"num_tokens":
|
| 180 |
"step": 17
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"entropy": 0.
|
| 184 |
"epoch": 0.22052067381316998,
|
| 185 |
-
"grad_norm":
|
| 186 |
"learning_rate": 0.00017926829268292684,
|
| 187 |
-
"loss":
|
| 188 |
-
"mean_token_accuracy":
|
| 189 |
-
"num_tokens":
|
| 190 |
"step": 18
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"entropy": 0.
|
| 194 |
"epoch": 0.2327718223583461,
|
| 195 |
-
"grad_norm":
|
| 196 |
"learning_rate": 0.00017804878048780488,
|
| 197 |
-
"loss":
|
| 198 |
-
"mean_token_accuracy":
|
| 199 |
-
"num_tokens":
|
| 200 |
"step": 19
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"entropy": 0.
|
| 204 |
"epoch": 0.2450229709035222,
|
| 205 |
-
"grad_norm": 0.
|
| 206 |
"learning_rate": 0.00017682926829268295,
|
| 207 |
-
"loss":
|
| 208 |
-
"mean_token_accuracy":
|
| 209 |
-
"num_tokens":
|
| 210 |
"step": 20
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"entropy": 0.
|
| 214 |
"epoch": 0.2572741194486983,
|
| 215 |
-
"grad_norm":
|
| 216 |
"learning_rate": 0.000175609756097561,
|
| 217 |
-
"loss": 0.
|
| 218 |
-
"mean_token_accuracy": 0.
|
| 219 |
-
"num_tokens":
|
| 220 |
"step": 21
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"entropy": 0.
|
| 224 |
"epoch": 0.26952526799387444,
|
| 225 |
-
"grad_norm":
|
| 226 |
"learning_rate": 0.00017439024390243903,
|
| 227 |
-
"loss":
|
| 228 |
-
"mean_token_accuracy":
|
| 229 |
-
"num_tokens":
|
| 230 |
"step": 22
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"entropy": 0.
|
| 234 |
"epoch": 0.28177641653905056,
|
| 235 |
-
"grad_norm":
|
| 236 |
"learning_rate": 0.00017317073170731708,
|
| 237 |
-
"loss":
|
| 238 |
-
"mean_token_accuracy":
|
| 239 |
-
"num_tokens":
|
| 240 |
"step": 23
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"entropy": 0.
|
| 244 |
"epoch": 0.29402756508422667,
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
"learning_rate": 0.00017195121951219512,
|
| 247 |
-
"loss":
|
| 248 |
-
"mean_token_accuracy":
|
| 249 |
-
"num_tokens":
|
| 250 |
"step": 24
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"entropy": 0.
|
| 254 |
"epoch": 0.30627871362940273,
|
| 255 |
-
"grad_norm": 0.
|
| 256 |
"learning_rate": 0.0001707317073170732,
|
| 257 |
-
"loss":
|
| 258 |
-
"mean_token_accuracy":
|
| 259 |
-
"num_tokens":
|
| 260 |
"step": 25
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"entropy": 0.
|
| 264 |
"epoch": 0.31852986217457885,
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
"learning_rate": 0.00016951219512195123,
|
| 267 |
-
"loss":
|
| 268 |
-
"mean_token_accuracy":
|
| 269 |
-
"num_tokens":
|
| 270 |
"step": 26
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"entropy": 0.
|
| 274 |
"epoch": 0.33078101071975496,
|
| 275 |
-
"grad_norm":
|
| 276 |
"learning_rate": 0.00016829268292682927,
|
| 277 |
-
"loss": 0.
|
| 278 |
-
"mean_token_accuracy":
|
| 279 |
-
"num_tokens":
|
| 280 |
"step": 27
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"entropy": 0.
|
| 284 |
"epoch": 0.3430321592649311,
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
"learning_rate": 0.00016707317073170731,
|
| 287 |
-
"loss": 0.
|
| 288 |
-
"mean_token_accuracy": 0.
|
| 289 |
-
"num_tokens":
|
| 290 |
"step": 28
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"entropy": 0.
|
| 294 |
"epoch": 0.3552833078101072,
|
| 295 |
-
"grad_norm": 0.
|
| 296 |
"learning_rate": 0.00016585365853658536,
|
| 297 |
-
"loss": 0.
|
| 298 |
-
"mean_token_accuracy":
|
| 299 |
-
"num_tokens":
|
| 300 |
"step": 29
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"entropy": 0.
|
| 304 |
"epoch": 0.3675344563552833,
|
| 305 |
-
"grad_norm": 0.
|
| 306 |
"learning_rate": 0.00016463414634146343,
|
| 307 |
-
"loss":
|
| 308 |
-
"mean_token_accuracy":
|
| 309 |
-
"num_tokens":
|
| 310 |
"step": 30
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"entropy": 0.
|
| 314 |
"epoch": 0.37978560490045943,
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
"learning_rate": 0.00016341463414634147,
|
| 317 |
-
"loss":
|
| 318 |
-
"mean_token_accuracy":
|
| 319 |
-
"num_tokens":
|
| 320 |
"step": 31
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"entropy": 0.
|
| 324 |
"epoch": 0.39203675344563554,
|
| 325 |
-
"grad_norm":
|
| 326 |
"learning_rate": 0.00016219512195121954,
|
| 327 |
-
"loss":
|
| 328 |
-
"mean_token_accuracy":
|
| 329 |
-
"num_tokens":
|
| 330 |
"step": 32
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"entropy": 0.
|
| 334 |
"epoch": 0.40428790199081166,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 0.00016097560975609758,
|
| 337 |
-
"loss":
|
| 338 |
-
"mean_token_accuracy":
|
| 339 |
-
"num_tokens":
|
| 340 |
"step": 33
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"entropy": 0.
|
| 344 |
"epoch": 0.4165390505359878,
|
| 345 |
-
"grad_norm":
|
| 346 |
"learning_rate": 0.00015975609756097562,
|
| 347 |
-
"loss":
|
| 348 |
-
"mean_token_accuracy":
|
| 349 |
-
"num_tokens":
|
| 350 |
"step": 34
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"entropy": 0.
|
| 354 |
"epoch": 0.42879019908116384,
|
| 355 |
-
"grad_norm":
|
| 356 |
"learning_rate": 0.00015853658536585366,
|
| 357 |
-
"loss": 0.
|
| 358 |
-
"mean_token_accuracy":
|
| 359 |
-
"num_tokens":
|
| 360 |
"step": 35
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"entropy": 0.
|
| 364 |
"epoch": 0.44104134762633995,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
"learning_rate": 0.00015731707317073173,
|
| 367 |
-
"loss": 0.
|
| 368 |
-
"mean_token_accuracy":
|
| 369 |
-
"num_tokens":
|
| 370 |
"step": 36
|
| 371 |
},
|
| 372 |
{
|
| 373 |
-
"entropy": 0.
|
| 374 |
"epoch": 0.45329249617151607,
|
| 375 |
-
"grad_norm": 0.
|
| 376 |
"learning_rate": 0.00015609756097560978,
|
| 377 |
-
"loss": 0.
|
| 378 |
-
"mean_token_accuracy": 0.
|
| 379 |
-
"num_tokens":
|
| 380 |
"step": 37
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"entropy": 0.
|
| 384 |
"epoch": 0.4655436447166922,
|
| 385 |
-
"grad_norm":
|
| 386 |
"learning_rate": 0.00015487804878048782,
|
| 387 |
-
"loss":
|
| 388 |
-
"mean_token_accuracy":
|
| 389 |
-
"num_tokens":
|
| 390 |
"step": 38
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"entropy": 0.
|
| 394 |
"epoch": 0.4777947932618683,
|
| 395 |
-
"grad_norm":
|
| 396 |
"learning_rate": 0.00015365853658536586,
|
| 397 |
-
"loss":
|
| 398 |
-
"mean_token_accuracy":
|
| 399 |
-
"num_tokens":
|
| 400 |
"step": 39
|
| 401 |
},
|
| 402 |
{
|
| 403 |
-
"entropy": 0.
|
| 404 |
"epoch": 0.4900459418070444,
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
"learning_rate": 0.0001524390243902439,
|
| 407 |
-
"loss": 0.
|
| 408 |
-
"mean_token_accuracy": 0.
|
| 409 |
-
"num_tokens":
|
| 410 |
"step": 40
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"entropy": 0.
|
| 414 |
"epoch": 0.5022970903522205,
|
| 415 |
-
"grad_norm": 0.
|
| 416 |
"learning_rate": 0.00015121951219512197,
|
| 417 |
-
"loss":
|
| 418 |
-
"mean_token_accuracy":
|
| 419 |
-
"num_tokens":
|
| 420 |
"step": 41
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"entropy": 0.
|
| 424 |
"epoch": 0.5145482388973966,
|
| 425 |
-
"grad_norm":
|
| 426 |
"learning_rate": 0.00015000000000000001,
|
| 427 |
-
"loss":
|
| 428 |
-
"mean_token_accuracy":
|
| 429 |
-
"num_tokens":
|
| 430 |
"step": 42
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"entropy": 0.
|
| 434 |
"epoch": 0.5267993874425727,
|
| 435 |
-
"grad_norm":
|
| 436 |
"learning_rate": 0.00014878048780487806,
|
| 437 |
-
"loss":
|
| 438 |
-
"mean_token_accuracy":
|
| 439 |
-
"num_tokens":
|
| 440 |
"step": 43
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"entropy": 0.
|
| 444 |
"epoch": 0.5390505359877489,
|
| 445 |
-
"grad_norm":
|
| 446 |
"learning_rate": 0.0001475609756097561,
|
| 447 |
-
"loss":
|
| 448 |
-
"mean_token_accuracy":
|
| 449 |
-
"num_tokens":
|
| 450 |
"step": 44
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"entropy": 0.
|
| 454 |
"epoch": 0.5513016845329249,
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
"learning_rate": 0.00014634146341463414,
|
| 457 |
-
"loss": 0.
|
| 458 |
-
"mean_token_accuracy": 0.
|
| 459 |
-
"num_tokens":
|
| 460 |
"step": 45
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"entropy": 0.
|
| 464 |
"epoch": 0.5635528330781011,
|
| 465 |
-
"grad_norm":
|
| 466 |
"learning_rate": 0.0001451219512195122,
|
| 467 |
-
"loss":
|
| 468 |
-
"mean_token_accuracy":
|
| 469 |
-
"num_tokens":
|
| 470 |
"step": 46
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"entropy": 0.
|
| 474 |
"epoch": 0.5758039816232772,
|
| 475 |
-
"grad_norm":
|
| 476 |
"learning_rate": 0.00014390243902439025,
|
| 477 |
-
"loss":
|
| 478 |
-
"mean_token_accuracy":
|
| 479 |
-
"num_tokens":
|
| 480 |
"step": 47
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"entropy": 0.
|
| 484 |
"epoch": 0.5880551301684533,
|
| 485 |
-
"grad_norm":
|
| 486 |
"learning_rate": 0.0001426829268292683,
|
| 487 |
-
"loss":
|
| 488 |
-
"mean_token_accuracy":
|
| 489 |
-
"num_tokens":
|
| 490 |
"step": 48
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"entropy": 0.
|
| 494 |
"epoch": 0.6003062787136294,
|
| 495 |
-
"grad_norm":
|
| 496 |
"learning_rate": 0.00014146341463414634,
|
| 497 |
-
"loss":
|
| 498 |
-
"mean_token_accuracy":
|
| 499 |
-
"num_tokens":
|
| 500 |
"step": 49
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"entropy": 0.
|
| 504 |
"epoch": 0.6125574272588055,
|
| 505 |
-
"grad_norm":
|
| 506 |
"learning_rate": 0.00014024390243902438,
|
| 507 |
-
"loss":
|
| 508 |
-
"mean_token_accuracy":
|
| 509 |
-
"num_tokens":
|
| 510 |
"step": 50
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 0.6125574272588055,
|
| 514 |
-
"eval_entropy": 0.
|
| 515 |
-
"eval_loss": 0.
|
| 516 |
-
"eval_mean_token_accuracy": 0.
|
| 517 |
-
"eval_num_tokens":
|
| 518 |
-
"eval_runtime":
|
| 519 |
-
"eval_samples_per_second": 1.
|
| 520 |
-
"eval_steps_per_second": 1.
|
| 521 |
"step": 50
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"entropy": 0.
|
| 525 |
"epoch": 0.6248085758039816,
|
| 526 |
-
"grad_norm":
|
| 527 |
"learning_rate": 0.00013902439024390245,
|
| 528 |
-
"loss":
|
| 529 |
-
"mean_token_accuracy":
|
| 530 |
-
"num_tokens":
|
| 531 |
"step": 51
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"entropy": 0.
|
| 535 |
"epoch": 0.6370597243491577,
|
| 536 |
-
"grad_norm":
|
| 537 |
"learning_rate": 0.0001378048780487805,
|
| 538 |
-
"loss":
|
| 539 |
-
"mean_token_accuracy":
|
| 540 |
-
"num_tokens":
|
| 541 |
"step": 52
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"entropy": 0.
|
| 545 |
"epoch": 0.6493108728943339,
|
| 546 |
-
"grad_norm":
|
| 547 |
"learning_rate": 0.00013658536585365856,
|
| 548 |
-
"loss": 0.
|
| 549 |
-
"mean_token_accuracy": 0.
|
| 550 |
-
"num_tokens":
|
| 551 |
"step": 53
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"entropy": 0.
|
| 555 |
"epoch": 0.6615620214395099,
|
| 556 |
-
"grad_norm": 0.
|
| 557 |
"learning_rate": 0.0001353658536585366,
|
| 558 |
-
"loss":
|
| 559 |
-
"mean_token_accuracy":
|
| 560 |
-
"num_tokens":
|
| 561 |
"step": 54
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"entropy": 0.
|
| 565 |
"epoch": 0.6738131699846861,
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
"learning_rate": 0.00013414634146341464,
|
| 568 |
-
"loss": 0.
|
| 569 |
-
"mean_token_accuracy": 0.
|
| 570 |
-
"num_tokens":
|
| 571 |
"step": 55
|
| 572 |
},
|
| 573 |
{
|
| 574 |
-
"entropy": 0.
|
| 575 |
"epoch": 0.6860643185298622,
|
| 576 |
-
"grad_norm":
|
| 577 |
"learning_rate": 0.0001329268292682927,
|
| 578 |
-
"loss":
|
| 579 |
-
"mean_token_accuracy":
|
| 580 |
-
"num_tokens":
|
| 581 |
"step": 56
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"entropy": 0.
|
| 585 |
"epoch": 0.6983154670750383,
|
| 586 |
-
"grad_norm":
|
| 587 |
"learning_rate": 0.00013170731707317076,
|
| 588 |
-
"loss":
|
| 589 |
-
"mean_token_accuracy":
|
| 590 |
-
"num_tokens":
|
| 591 |
"step": 57
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"entropy": 0.
|
| 595 |
"epoch": 0.7105666156202144,
|
| 596 |
-
"grad_norm":
|
| 597 |
"learning_rate": 0.0001304878048780488,
|
| 598 |
-
"loss": 0.
|
| 599 |
-
"mean_token_accuracy": 0.
|
| 600 |
-
"num_tokens":
|
| 601 |
"step": 58
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"entropy": 0.
|
| 605 |
"epoch": 0.7228177641653905,
|
| 606 |
-
"grad_norm":
|
| 607 |
"learning_rate": 0.00012926829268292684,
|
| 608 |
-
"loss": 0.
|
| 609 |
-
"mean_token_accuracy": 0.
|
| 610 |
-
"num_tokens":
|
| 611 |
"step": 59
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"entropy": 0.
|
| 615 |
"epoch": 0.7350689127105666,
|
| 616 |
-
"grad_norm":
|
| 617 |
"learning_rate": 0.00012804878048780488,
|
| 618 |
-
"loss":
|
| 619 |
-
"mean_token_accuracy":
|
| 620 |
-
"num_tokens":
|
| 621 |
"step": 60
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"entropy": 0.
|
| 625 |
"epoch": 0.7473200612557427,
|
| 626 |
-
"grad_norm":
|
| 627 |
"learning_rate": 0.00012682926829268293,
|
| 628 |
-
"loss":
|
| 629 |
-
"mean_token_accuracy":
|
| 630 |
-
"num_tokens":
|
| 631 |
"step": 61
|
| 632 |
},
|
| 633 |
{
|
| 634 |
-
"entropy": 0.
|
| 635 |
"epoch": 0.7595712098009189,
|
| 636 |
-
"grad_norm":
|
| 637 |
"learning_rate": 0.000125609756097561,
|
| 638 |
-
"loss":
|
| 639 |
-
"mean_token_accuracy":
|
| 640 |
-
"num_tokens":
|
| 641 |
"step": 62
|
| 642 |
},
|
| 643 |
{
|
| 644 |
-
"entropy": 0.
|
| 645 |
"epoch": 0.7718223583460949,
|
| 646 |
-
"grad_norm":
|
| 647 |
"learning_rate": 0.00012439024390243904,
|
| 648 |
-
"loss":
|
| 649 |
-
"mean_token_accuracy":
|
| 650 |
-
"num_tokens":
|
| 651 |
"step": 63
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"entropy": 0.
|
| 655 |
"epoch": 0.7840735068912711,
|
| 656 |
-
"grad_norm":
|
| 657 |
"learning_rate": 0.00012317073170731708,
|
| 658 |
-
"loss":
|
| 659 |
-
"mean_token_accuracy":
|
| 660 |
-
"num_tokens":
|
| 661 |
"step": 64
|
| 662 |
},
|
| 663 |
{
|
| 664 |
-
"entropy": 0.
|
| 665 |
"epoch": 0.7963246554364471,
|
| 666 |
-
"grad_norm":
|
| 667 |
"learning_rate": 0.00012195121951219512,
|
| 668 |
-
"loss":
|
| 669 |
-
"mean_token_accuracy":
|
| 670 |
-
"num_tokens":
|
| 671 |
"step": 65
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"entropy": 0.
|
| 675 |
"epoch": 0.8085758039816233,
|
| 676 |
-
"grad_norm": 0.
|
| 677 |
"learning_rate": 0.00012073170731707318,
|
| 678 |
-
"loss":
|
| 679 |
-
"mean_token_accuracy":
|
| 680 |
-
"num_tokens":
|
| 681 |
"step": 66
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"entropy": 0.
|
| 685 |
"epoch": 0.8208269525267994,
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
"learning_rate": 0.00011951219512195122,
|
| 688 |
-
"loss": 0.
|
| 689 |
-
"mean_token_accuracy": 0.
|
| 690 |
-
"num_tokens":
|
| 691 |
"step": 67
|
| 692 |
},
|
| 693 |
{
|
| 694 |
-
"entropy": 0.
|
| 695 |
"epoch": 0.8330781010719756,
|
| 696 |
-
"grad_norm": 0.
|
| 697 |
"learning_rate": 0.00011829268292682926,
|
| 698 |
-
"loss":
|
| 699 |
-
"mean_token_accuracy":
|
| 700 |
-
"num_tokens":
|
| 701 |
"step": 68
|
| 702 |
},
|
| 703 |
{
|
| 704 |
-
"entropy": 0.
|
| 705 |
"epoch": 0.8453292496171516,
|
| 706 |
-
"grad_norm": 0.
|
| 707 |
"learning_rate": 0.00011707317073170732,
|
| 708 |
-
"loss":
|
| 709 |
-
"mean_token_accuracy":
|
| 710 |
-
"num_tokens":
|
| 711 |
"step": 69
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"entropy": 0.
|
| 715 |
"epoch": 0.8575803981623277,
|
| 716 |
-
"grad_norm": 0.
|
| 717 |
"learning_rate": 0.00011585365853658536,
|
| 718 |
-
"loss": 0.
|
| 719 |
-
"mean_token_accuracy": 0.
|
| 720 |
-
"num_tokens":
|
| 721 |
"step": 70
|
| 722 |
},
|
| 723 |
{
|
| 724 |
-
"entropy": 0.
|
| 725 |
"epoch": 0.8698315467075038,
|
| 726 |
-
"grad_norm": 0.
|
| 727 |
"learning_rate": 0.00011463414634146342,
|
| 728 |
-
"loss": 0.
|
| 729 |
-
"mean_token_accuracy": 0.
|
| 730 |
-
"num_tokens":
|
| 731 |
"step": 71
|
| 732 |
},
|
| 733 |
{
|
| 734 |
-
"entropy": 0.
|
| 735 |
"epoch": 0.8820826952526799,
|
| 736 |
-
"grad_norm": 0.
|
| 737 |
"learning_rate": 0.00011341463414634146,
|
| 738 |
-
"loss":
|
| 739 |
-
"mean_token_accuracy":
|
| 740 |
-
"num_tokens":
|
| 741 |
"step": 72
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"entropy": 0.
|
| 745 |
"epoch": 0.8943338437978561,
|
| 746 |
-
"grad_norm": 0.
|
| 747 |
"learning_rate": 0.00011219512195121953,
|
| 748 |
-
"loss":
|
| 749 |
-
"mean_token_accuracy":
|
| 750 |
-
"num_tokens":
|
| 751 |
"step": 73
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"entropy": 0.
|
| 755 |
"epoch": 0.9065849923430321,
|
| 756 |
-
"grad_norm":
|
| 757 |
"learning_rate": 0.00011097560975609757,
|
| 758 |
-
"loss":
|
| 759 |
-
"mean_token_accuracy":
|
| 760 |
-
"num_tokens":
|
| 761 |
"step": 74
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"entropy": 0.
|
| 765 |
"epoch": 0.9188361408882083,
|
| 766 |
-
"grad_norm":
|
| 767 |
"learning_rate": 0.00010975609756097563,
|
| 768 |
-
"loss":
|
| 769 |
-
"mean_token_accuracy":
|
| 770 |
-
"num_tokens":
|
| 771 |
"step": 75
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"entropy": 0.
|
| 775 |
"epoch": 0.9310872894333844,
|
| 776 |
-
"grad_norm":
|
| 777 |
"learning_rate": 0.00010853658536585367,
|
| 778 |
-
"loss":
|
| 779 |
-
"mean_token_accuracy":
|
| 780 |
-
"num_tokens":
|
| 781 |
"step": 76
|
| 782 |
},
|
| 783 |
{
|
| 784 |
-
"entropy": 0.
|
| 785 |
"epoch": 0.9433384379785605,
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
"learning_rate": 0.00010731707317073172,
|
| 788 |
-
"loss":
|
| 789 |
-
"mean_token_accuracy":
|
| 790 |
-
"num_tokens":
|
| 791 |
"step": 77
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"entropy": 0.
|
| 795 |
"epoch": 0.9555895865237366,
|
| 796 |
-
"grad_norm":
|
| 797 |
"learning_rate": 0.00010609756097560977,
|
| 798 |
-
"loss":
|
| 799 |
-
"mean_token_accuracy":
|
| 800 |
-
"num_tokens":
|
| 801 |
"step": 78
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"entropy": 0.
|
| 805 |
"epoch": 0.9678407350689127,
|
| 806 |
-
"grad_norm":
|
| 807 |
"learning_rate": 0.00010487804878048781,
|
| 808 |
-
"loss":
|
| 809 |
-
"mean_token_accuracy":
|
| 810 |
-
"num_tokens":
|
| 811 |
"step": 79
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"entropy": 0.
|
| 815 |
"epoch": 0.9800918836140888,
|
| 816 |
-
"grad_norm":
|
| 817 |
"learning_rate": 0.00010365853658536586,
|
| 818 |
-
"loss": 0.
|
| 819 |
-
"mean_token_accuracy": 0.
|
| 820 |
-
"num_tokens":
|
| 821 |
"step": 80
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"entropy": 0.
|
| 825 |
"epoch": 0.9923430321592649,
|
| 826 |
-
"grad_norm":
|
| 827 |
"learning_rate": 0.0001024390243902439,
|
| 828 |
-
"loss":
|
| 829 |
-
"mean_token_accuracy":
|
| 830 |
-
"num_tokens":
|
| 831 |
"step": 81
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"entropy": 0.
|
| 835 |
"epoch": 1.0,
|
| 836 |
-
"grad_norm": 0.
|
| 837 |
"learning_rate": 0.00010121951219512196,
|
| 838 |
-
"loss":
|
| 839 |
-
"mean_token_accuracy":
|
| 840 |
-
"num_tokens":
|
| 841 |
"step": 82
|
| 842 |
},
|
| 843 |
{
|
| 844 |
-
"entropy": 0.
|
| 845 |
"epoch": 1.0122511485451762,
|
| 846 |
-
"grad_norm": 0.
|
| 847 |
"learning_rate": 0.0001,
|
| 848 |
-
"loss": 0.
|
| 849 |
-
"mean_token_accuracy": 0.
|
| 850 |
-
"num_tokens":
|
| 851 |
"step": 83
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"entropy": 0.
|
| 855 |
"epoch": 1.0245022970903521,
|
| 856 |
-
"grad_norm": 0.
|
| 857 |
"learning_rate": 9.878048780487805e-05,
|
| 858 |
-
"loss":
|
| 859 |
-
"mean_token_accuracy":
|
| 860 |
-
"num_tokens":
|
| 861 |
"step": 84
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"entropy": 0.
|
| 865 |
"epoch": 1.0367534456355283,
|
| 866 |
-
"grad_norm": 0.
|
| 867 |
"learning_rate": 9.75609756097561e-05,
|
| 868 |
-
"loss":
|
| 869 |
-
"mean_token_accuracy":
|
| 870 |
-
"num_tokens":
|
| 871 |
"step": 85
|
| 872 |
},
|
| 873 |
{
|
| 874 |
-
"entropy": 0.
|
| 875 |
"epoch": 1.0490045941807045,
|
| 876 |
-
"grad_norm": 0.
|
| 877 |
"learning_rate": 9.634146341463415e-05,
|
| 878 |
-
"loss":
|
| 879 |
-
"mean_token_accuracy":
|
| 880 |
-
"num_tokens":
|
| 881 |
"step": 86
|
| 882 |
},
|
| 883 |
{
|
| 884 |
-
"entropy": 0.
|
| 885 |
"epoch": 1.0612557427258806,
|
| 886 |
-
"grad_norm": 0.
|
| 887 |
"learning_rate": 9.51219512195122e-05,
|
| 888 |
-
"loss":
|
| 889 |
-
"mean_token_accuracy":
|
| 890 |
-
"num_tokens":
|
| 891 |
"step": 87
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"entropy": 0.
|
| 895 |
"epoch": 1.0735068912710566,
|
| 896 |
-
"grad_norm": 0.
|
| 897 |
"learning_rate": 9.390243902439024e-05,
|
| 898 |
-
"loss":
|
| 899 |
-
"mean_token_accuracy":
|
| 900 |
-
"num_tokens":
|
| 901 |
"step": 88
|
| 902 |
},
|
| 903 |
{
|
| 904 |
-
"entropy": 0.
|
| 905 |
"epoch": 1.0857580398162328,
|
| 906 |
-
"grad_norm": 0.
|
| 907 |
"learning_rate": 9.26829268292683e-05,
|
| 908 |
-
"loss": 0.
|
| 909 |
-
"mean_token_accuracy": 0.
|
| 910 |
-
"num_tokens":
|
| 911 |
"step": 89
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"entropy": 0.
|
| 915 |
"epoch": 1.098009188361409,
|
| 916 |
-
"grad_norm": 0.
|
| 917 |
"learning_rate": 9.146341463414635e-05,
|
| 918 |
-
"loss":
|
| 919 |
-
"mean_token_accuracy":
|
| 920 |
-
"num_tokens":
|
| 921 |
"step": 90
|
| 922 |
},
|
| 923 |
{
|
| 924 |
-
"entropy": 0.
|
| 925 |
"epoch": 1.110260336906585,
|
| 926 |
-
"grad_norm": 0.
|
| 927 |
"learning_rate": 9.02439024390244e-05,
|
| 928 |
-
"loss": 0.
|
| 929 |
-
"mean_token_accuracy": 0.
|
| 930 |
-
"num_tokens":
|
| 931 |
"step": 91
|
| 932 |
},
|
| 933 |
{
|
| 934 |
-
"entropy": 0.
|
| 935 |
"epoch": 1.122511485451761,
|
| 936 |
-
"grad_norm": 0.
|
| 937 |
"learning_rate": 8.902439024390244e-05,
|
| 938 |
-
"loss": 0.
|
| 939 |
-
"mean_token_accuracy": 0.
|
| 940 |
-
"num_tokens":
|
| 941 |
"step": 92
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"entropy": 0.
|
| 945 |
"epoch": 1.1347626339969372,
|
| 946 |
-
"grad_norm": 0.
|
| 947 |
"learning_rate": 8.78048780487805e-05,
|
| 948 |
-
"loss": 0.
|
| 949 |
-
"mean_token_accuracy":
|
| 950 |
-
"num_tokens":
|
| 951 |
"step": 93
|
| 952 |
},
|
| 953 |
{
|
| 954 |
-
"entropy": 0.
|
| 955 |
"epoch": 1.1470137825421134,
|
| 956 |
-
"grad_norm": 0.
|
| 957 |
"learning_rate": 8.658536585365854e-05,
|
| 958 |
-
"loss": 0.
|
| 959 |
-
"mean_token_accuracy":
|
| 960 |
-
"num_tokens":
|
| 961 |
"step": 94
|
| 962 |
},
|
| 963 |
{
|
| 964 |
-
"entropy": 0.
|
| 965 |
"epoch": 1.1592649310872893,
|
| 966 |
-
"grad_norm":
|
| 967 |
"learning_rate": 8.53658536585366e-05,
|
| 968 |
-
"loss": 0.
|
| 969 |
-
"mean_token_accuracy":
|
| 970 |
-
"num_tokens":
|
| 971 |
"step": 95
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"entropy": 0.
|
| 975 |
"epoch": 1.1715160796324655,
|
| 976 |
-
"grad_norm": 0.
|
| 977 |
"learning_rate": 8.414634146341464e-05,
|
| 978 |
-
"loss": 0.
|
| 979 |
-
"mean_token_accuracy": 0.
|
| 980 |
-
"num_tokens":
|
| 981 |
"step": 96
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"entropy": 0.
|
| 985 |
"epoch": 1.1837672281776417,
|
| 986 |
-
"grad_norm": 0.
|
| 987 |
"learning_rate": 8.292682926829268e-05,
|
| 988 |
-
"loss":
|
| 989 |
-
"mean_token_accuracy":
|
| 990 |
-
"num_tokens":
|
| 991 |
"step": 97
|
| 992 |
},
|
| 993 |
{
|
| 994 |
-
"entropy": 0.
|
| 995 |
"epoch": 1.1960183767228179,
|
| 996 |
-
"grad_norm": 0.
|
| 997 |
"learning_rate": 8.170731707317073e-05,
|
| 998 |
-
"loss": 0.
|
| 999 |
-
"mean_token_accuracy": 0.
|
| 1000 |
-
"num_tokens":
|
| 1001 |
"step": 98
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
-
"entropy": 0.
|
| 1005 |
"epoch": 1.2082695252679938,
|
| 1006 |
-
"grad_norm": 0.
|
| 1007 |
"learning_rate": 8.048780487804879e-05,
|
| 1008 |
-
"loss":
|
| 1009 |
-
"mean_token_accuracy":
|
| 1010 |
-
"num_tokens":
|
| 1011 |
"step": 99
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"entropy": 0.
|
| 1015 |
"epoch": 1.22052067381317,
|
| 1016 |
-
"grad_norm":
|
| 1017 |
"learning_rate": 7.926829268292683e-05,
|
| 1018 |
-
"loss": 0.
|
| 1019 |
-
"mean_token_accuracy":
|
| 1020 |
-
"num_tokens":
|
| 1021 |
"step": 100
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 1.22052067381317,
|
| 1025 |
-
"eval_entropy": 0.
|
| 1026 |
-
"eval_loss": 0.
|
| 1027 |
-
"eval_mean_token_accuracy": 0.
|
| 1028 |
-
"eval_num_tokens":
|
| 1029 |
-
"eval_runtime":
|
| 1030 |
-
"eval_samples_per_second": 1.
|
| 1031 |
-
"eval_steps_per_second": 1.
|
| 1032 |
"step": 100
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
-
"entropy": 0.
|
| 1036 |
"epoch": 1.2327718223583461,
|
| 1037 |
-
"grad_norm": 0.
|
| 1038 |
"learning_rate": 7.804878048780489e-05,
|
| 1039 |
-
"loss":
|
| 1040 |
-
"mean_token_accuracy":
|
| 1041 |
-
"num_tokens":
|
| 1042 |
"step": 101
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
-
"entropy": 0.
|
| 1046 |
"epoch": 1.245022970903522,
|
| 1047 |
-
"grad_norm": 0.
|
| 1048 |
"learning_rate": 7.682926829268293e-05,
|
| 1049 |
-
"loss":
|
| 1050 |
-
"mean_token_accuracy":
|
| 1051 |
-
"num_tokens":
|
| 1052 |
"step": 102
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
-
"entropy": 0.
|
| 1056 |
"epoch": 1.2572741194486983,
|
| 1057 |
-
"grad_norm": 0.
|
| 1058 |
"learning_rate": 7.560975609756099e-05,
|
| 1059 |
-
"loss":
|
| 1060 |
-
"mean_token_accuracy":
|
| 1061 |
-
"num_tokens":
|
| 1062 |
"step": 103
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
-
"entropy": 0.
|
| 1066 |
"epoch": 1.2695252679938744,
|
| 1067 |
-
"grad_norm":
|
| 1068 |
"learning_rate": 7.439024390243903e-05,
|
| 1069 |
-
"loss":
|
| 1070 |
-
"mean_token_accuracy":
|
| 1071 |
-
"num_tokens":
|
| 1072 |
"step": 104
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
-
"entropy": 0.
|
| 1076 |
"epoch": 1.2817764165390506,
|
| 1077 |
-
"grad_norm": 0.
|
| 1078 |
"learning_rate": 7.317073170731707e-05,
|
| 1079 |
-
"loss": 0.
|
| 1080 |
-
"mean_token_accuracy": 0.
|
| 1081 |
-
"num_tokens":
|
| 1082 |
"step": 105
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
-
"entropy": 0.
|
| 1086 |
"epoch": 1.2940275650842268,
|
| 1087 |
-
"grad_norm": 0.
|
| 1088 |
"learning_rate": 7.195121951219513e-05,
|
| 1089 |
-
"loss":
|
| 1090 |
-
"mean_token_accuracy":
|
| 1091 |
-
"num_tokens":
|
| 1092 |
"step": 106
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
-
"entropy": 0.
|
| 1096 |
"epoch": 1.3062787136294027,
|
| 1097 |
-
"grad_norm":
|
| 1098 |
"learning_rate": 7.073170731707317e-05,
|
| 1099 |
-
"loss":
|
| 1100 |
-
"mean_token_accuracy":
|
| 1101 |
-
"num_tokens":
|
| 1102 |
"step": 107
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
-
"entropy": 0.
|
| 1106 |
"epoch": 1.318529862174579,
|
| 1107 |
-
"grad_norm": 0.
|
| 1108 |
"learning_rate": 6.951219512195122e-05,
|
| 1109 |
-
"loss":
|
| 1110 |
-
"mean_token_accuracy":
|
| 1111 |
-
"num_tokens":
|
| 1112 |
"step": 108
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"entropy": 0.
|
| 1116 |
"epoch": 1.3307810107197549,
|
| 1117 |
-
"grad_norm": 0.
|
| 1118 |
"learning_rate": 6.829268292682928e-05,
|
| 1119 |
-
"loss":
|
| 1120 |
-
"mean_token_accuracy":
|
| 1121 |
-
"num_tokens":
|
| 1122 |
"step": 109
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"entropy": 0.
|
| 1126 |
"epoch": 1.343032159264931,
|
| 1127 |
-
"grad_norm": 0.
|
| 1128 |
"learning_rate": 6.707317073170732e-05,
|
| 1129 |
-
"loss":
|
| 1130 |
-
"mean_token_accuracy":
|
| 1131 |
-
"num_tokens":
|
| 1132 |
"step": 110
|
| 1133 |
},
|
| 1134 |
{
|
| 1135 |
-
"entropy": 0.
|
| 1136 |
"epoch": 1.3552833078101072,
|
| 1137 |
-
"grad_norm": 0.
|
| 1138 |
"learning_rate": 6.585365853658538e-05,
|
| 1139 |
-
"loss":
|
| 1140 |
-
"mean_token_accuracy":
|
| 1141 |
-
"num_tokens":
|
| 1142 |
"step": 111
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
-
"entropy": 0.
|
| 1146 |
"epoch": 1.3675344563552834,
|
| 1147 |
-
"grad_norm": 0.
|
| 1148 |
"learning_rate": 6.463414634146342e-05,
|
| 1149 |
-
"loss": 0.
|
| 1150 |
-
"mean_token_accuracy": 0.
|
| 1151 |
-
"num_tokens":
|
| 1152 |
"step": 112
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
-
"entropy": 0.
|
| 1156 |
"epoch": 1.3797856049004595,
|
| 1157 |
-
"grad_norm": 0.
|
| 1158 |
"learning_rate": 6.341463414634146e-05,
|
| 1159 |
-
"loss": 0.
|
| 1160 |
-
"mean_token_accuracy": 0.
|
| 1161 |
-
"num_tokens":
|
| 1162 |
"step": 113
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
-
"entropy": 0.
|
| 1166 |
"epoch": 1.3920367534456355,
|
| 1167 |
-
"grad_norm":
|
| 1168 |
"learning_rate": 6.219512195121952e-05,
|
| 1169 |
-
"loss":
|
| 1170 |
-
"mean_token_accuracy":
|
| 1171 |
-
"num_tokens":
|
| 1172 |
"step": 114
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"entropy": 0.
|
| 1176 |
"epoch": 1.4042879019908117,
|
| 1177 |
-
"grad_norm": 0.
|
| 1178 |
"learning_rate": 6.097560975609756e-05,
|
| 1179 |
-
"loss":
|
| 1180 |
-
"mean_token_accuracy":
|
| 1181 |
-
"num_tokens":
|
| 1182 |
"step": 115
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
-
"entropy": 0.
|
| 1186 |
"epoch": 1.4165390505359878,
|
| 1187 |
-
"grad_norm": 0.
|
| 1188 |
"learning_rate": 5.975609756097561e-05,
|
| 1189 |
-
"loss":
|
| 1190 |
-
"mean_token_accuracy":
|
| 1191 |
-
"num_tokens":
|
| 1192 |
"step": 116
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
-
"entropy": 0.
|
| 1196 |
"epoch": 1.4287901990811638,
|
| 1197 |
-
"grad_norm":
|
| 1198 |
"learning_rate": 5.853658536585366e-05,
|
| 1199 |
-
"loss":
|
| 1200 |
-
"mean_token_accuracy":
|
| 1201 |
-
"num_tokens":
|
| 1202 |
"step": 117
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
-
"entropy": 0.
|
| 1206 |
"epoch": 1.44104134762634,
|
| 1207 |
-
"grad_norm":
|
| 1208 |
"learning_rate": 5.731707317073171e-05,
|
| 1209 |
-
"loss":
|
| 1210 |
-
"mean_token_accuracy":
|
| 1211 |
-
"num_tokens":
|
| 1212 |
"step": 118
|
| 1213 |
},
|
| 1214 |
{
|
| 1215 |
-
"entropy": 0.
|
| 1216 |
"epoch": 1.4532924961715161,
|
| 1217 |
-
"grad_norm":
|
| 1218 |
"learning_rate": 5.6097560975609764e-05,
|
| 1219 |
-
"loss":
|
| 1220 |
-
"mean_token_accuracy":
|
| 1221 |
-
"num_tokens":
|
| 1222 |
"step": 119
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
-
"entropy": 0.
|
| 1226 |
"epoch": 1.4655436447166923,
|
| 1227 |
-
"grad_norm": 0.
|
| 1228 |
"learning_rate": 5.487804878048781e-05,
|
| 1229 |
-
"loss":
|
| 1230 |
-
"mean_token_accuracy":
|
| 1231 |
-
"num_tokens":
|
| 1232 |
"step": 120
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
-
"entropy": 0.
|
| 1236 |
"epoch": 1.4777947932618682,
|
| 1237 |
-
"grad_norm": 0.
|
| 1238 |
"learning_rate": 5.365853658536586e-05,
|
| 1239 |
-
"loss":
|
| 1240 |
-
"mean_token_accuracy":
|
| 1241 |
-
"num_tokens":
|
| 1242 |
"step": 121
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"entropy": 0.
|
| 1246 |
"epoch": 1.4900459418070444,
|
| 1247 |
-
"grad_norm":
|
| 1248 |
"learning_rate": 5.2439024390243904e-05,
|
| 1249 |
-
"loss":
|
| 1250 |
-
"mean_token_accuracy":
|
| 1251 |
-
"num_tokens":
|
| 1252 |
"step": 122
|
| 1253 |
},
|
| 1254 |
{
|
| 1255 |
-
"entropy": 0.
|
| 1256 |
"epoch": 1.5022970903522204,
|
| 1257 |
-
"grad_norm": 0.
|
| 1258 |
"learning_rate": 5.121951219512195e-05,
|
| 1259 |
-
"loss":
|
| 1260 |
-
"mean_token_accuracy":
|
| 1261 |
-
"num_tokens":
|
| 1262 |
"step": 123
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
-
"entropy": 0.
|
| 1266 |
"epoch": 1.5145482388973965,
|
| 1267 |
-
"grad_norm": 0.
|
| 1268 |
"learning_rate": 5e-05,
|
| 1269 |
-
"loss":
|
| 1270 |
-
"mean_token_accuracy":
|
| 1271 |
-
"num_tokens":
|
| 1272 |
"step": 124
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
-
"entropy": 0.
|
| 1276 |
"epoch": 1.5267993874425727,
|
| 1277 |
-
"grad_norm":
|
| 1278 |
"learning_rate": 4.878048780487805e-05,
|
| 1279 |
-
"loss":
|
| 1280 |
-
"mean_token_accuracy":
|
| 1281 |
-
"num_tokens":
|
| 1282 |
"step": 125
|
| 1283 |
},
|
| 1284 |
{
|
| 1285 |
-
"entropy": 0.
|
| 1286 |
"epoch": 1.5390505359877489,
|
| 1287 |
-
"grad_norm": 0.
|
| 1288 |
"learning_rate": 4.75609756097561e-05,
|
| 1289 |
-
"loss":
|
| 1290 |
-
"mean_token_accuracy":
|
| 1291 |
-
"num_tokens":
|
| 1292 |
"step": 126
|
| 1293 |
},
|
| 1294 |
{
|
| 1295 |
-
"entropy": 0.
|
| 1296 |
"epoch": 1.551301684532925,
|
| 1297 |
-
"grad_norm": 0.
|
| 1298 |
"learning_rate": 4.634146341463415e-05,
|
| 1299 |
-
"loss": 0.
|
| 1300 |
-
"mean_token_accuracy":
|
| 1301 |
-
"num_tokens":
|
| 1302 |
"step": 127
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
-
"entropy": 0.
|
| 1306 |
"epoch": 1.5635528330781012,
|
| 1307 |
-
"grad_norm": 0.
|
| 1308 |
"learning_rate": 4.51219512195122e-05,
|
| 1309 |
-
"loss":
|
| 1310 |
-
"mean_token_accuracy":
|
| 1311 |
-
"num_tokens":
|
| 1312 |
"step": 128
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
-
"entropy": 0.
|
| 1316 |
"epoch": 1.5758039816232772,
|
| 1317 |
-
"grad_norm": 0.
|
| 1318 |
"learning_rate": 4.390243902439025e-05,
|
| 1319 |
-
"loss":
|
| 1320 |
-
"mean_token_accuracy":
|
| 1321 |
-
"num_tokens":
|
| 1322 |
"step": 129
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
-
"entropy": 0.
|
| 1326 |
"epoch": 1.5880551301684533,
|
| 1327 |
-
"grad_norm": 0.
|
| 1328 |
"learning_rate": 4.26829268292683e-05,
|
| 1329 |
-
"loss":
|
| 1330 |
-
"mean_token_accuracy":
|
| 1331 |
-
"num_tokens":
|
| 1332 |
"step": 130
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
-
"entropy": 0.
|
| 1336 |
"epoch": 1.6003062787136293,
|
| 1337 |
-
"grad_norm": 0.
|
| 1338 |
"learning_rate": 4.146341463414634e-05,
|
| 1339 |
-
"loss": 0.
|
| 1340 |
-
"mean_token_accuracy": 0.
|
| 1341 |
-
"num_tokens":
|
| 1342 |
"step": 131
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
-
"entropy": 0.
|
| 1346 |
"epoch": 1.6125574272588055,
|
| 1347 |
-
"grad_norm": 0.
|
| 1348 |
"learning_rate": 4.0243902439024395e-05,
|
| 1349 |
-
"loss": 0.
|
| 1350 |
-
"mean_token_accuracy": 0.
|
| 1351 |
-
"num_tokens":
|
| 1352 |
"step": 132
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
-
"entropy": 0.
|
| 1356 |
"epoch": 1.6248085758039816,
|
| 1357 |
-
"grad_norm": 0.
|
| 1358 |
"learning_rate": 3.9024390243902444e-05,
|
| 1359 |
-
"loss":
|
| 1360 |
-
"mean_token_accuracy":
|
| 1361 |
-
"num_tokens":
|
| 1362 |
"step": 133
|
| 1363 |
},
|
| 1364 |
{
|
| 1365 |
-
"entropy": 0.
|
| 1366 |
"epoch": 1.6370597243491578,
|
| 1367 |
-
"grad_norm": 0.
|
| 1368 |
"learning_rate": 3.780487804878049e-05,
|
| 1369 |
-
"loss":
|
| 1370 |
-
"mean_token_accuracy":
|
| 1371 |
-
"num_tokens":
|
| 1372 |
"step": 134
|
| 1373 |
},
|
| 1374 |
{
|
| 1375 |
-
"entropy": 0.
|
| 1376 |
"epoch": 1.649310872894334,
|
| 1377 |
-
"grad_norm": 0.
|
| 1378 |
"learning_rate": 3.6585365853658535e-05,
|
| 1379 |
-
"loss":
|
| 1380 |
-
"mean_token_accuracy":
|
| 1381 |
-
"num_tokens":
|
| 1382 |
"step": 135
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
-
"entropy": 0.
|
| 1386 |
"epoch": 1.66156202143951,
|
| 1387 |
-
"grad_norm": 0.
|
| 1388 |
"learning_rate": 3.5365853658536584e-05,
|
| 1389 |
-
"loss":
|
| 1390 |
-
"mean_token_accuracy":
|
| 1391 |
-
"num_tokens":
|
| 1392 |
"step": 136
|
| 1393 |
},
|
| 1394 |
{
|
| 1395 |
-
"entropy": 0.
|
| 1396 |
"epoch": 1.673813169984686,
|
| 1397 |
-
"grad_norm": 0.
|
| 1398 |
"learning_rate": 3.414634146341464e-05,
|
| 1399 |
-
"loss":
|
| 1400 |
-
"mean_token_accuracy":
|
| 1401 |
-
"num_tokens":
|
| 1402 |
"step": 137
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
-
"entropy": 0.
|
| 1406 |
"epoch": 1.686064318529862,
|
| 1407 |
-
"grad_norm": 0.
|
| 1408 |
"learning_rate": 3.292682926829269e-05,
|
| 1409 |
-
"loss": 0.
|
| 1410 |
-
"mean_token_accuracy": 0.
|
| 1411 |
-
"num_tokens":
|
| 1412 |
"step": 138
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
-
"entropy": 0.
|
| 1416 |
"epoch": 1.6983154670750382,
|
| 1417 |
-
"grad_norm": 0.
|
| 1418 |
"learning_rate": 3.170731707317073e-05,
|
| 1419 |
-
"loss":
|
| 1420 |
-
"mean_token_accuracy":
|
| 1421 |
-
"num_tokens":
|
| 1422 |
"step": 139
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
-
"entropy": 0.
|
| 1426 |
"epoch": 1.7105666156202144,
|
| 1427 |
-
"grad_norm": 0.
|
| 1428 |
"learning_rate": 3.048780487804878e-05,
|
| 1429 |
-
"loss":
|
| 1430 |
-
"mean_token_accuracy":
|
| 1431 |
-
"num_tokens":
|
| 1432 |
"step": 140
|
| 1433 |
},
|
| 1434 |
{
|
| 1435 |
-
"entropy": 0.
|
| 1436 |
"epoch": 1.7228177641653906,
|
| 1437 |
-
"grad_norm": 0.
|
| 1438 |
"learning_rate": 2.926829268292683e-05,
|
| 1439 |
-
"loss":
|
| 1440 |
-
"mean_token_accuracy":
|
| 1441 |
-
"num_tokens":
|
| 1442 |
"step": 141
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
-
"entropy": 0.
|
| 1446 |
"epoch": 1.7350689127105667,
|
| 1447 |
-
"grad_norm": 0.
|
| 1448 |
"learning_rate": 2.8048780487804882e-05,
|
| 1449 |
-
"loss":
|
| 1450 |
-
"mean_token_accuracy":
|
| 1451 |
-
"num_tokens":
|
| 1452 |
"step": 142
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
-
"entropy": 0.
|
| 1456 |
"epoch": 1.7473200612557427,
|
| 1457 |
-
"grad_norm": 0.
|
| 1458 |
"learning_rate": 2.682926829268293e-05,
|
| 1459 |
-
"loss":
|
| 1460 |
-
"mean_token_accuracy":
|
| 1461 |
-
"num_tokens":
|
| 1462 |
"step": 143
|
| 1463 |
},
|
| 1464 |
{
|
| 1465 |
-
"entropy": 0.
|
| 1466 |
"epoch": 1.7595712098009189,
|
| 1467 |
-
"grad_norm": 0.
|
| 1468 |
"learning_rate": 2.5609756097560977e-05,
|
| 1469 |
-
"loss":
|
| 1470 |
-
"mean_token_accuracy":
|
| 1471 |
-
"num_tokens":
|
| 1472 |
"step": 144
|
| 1473 |
},
|
| 1474 |
{
|
| 1475 |
-
"entropy": 0.
|
| 1476 |
"epoch": 1.7718223583460948,
|
| 1477 |
-
"grad_norm": 0.
|
| 1478 |
"learning_rate": 2.4390243902439026e-05,
|
| 1479 |
-
"loss": 0.
|
| 1480 |
-
"mean_token_accuracy": 0.
|
| 1481 |
-
"num_tokens":
|
| 1482 |
"step": 145
|
| 1483 |
},
|
| 1484 |
{
|
| 1485 |
-
"entropy": 0.
|
| 1486 |
"epoch": 1.784073506891271,
|
| 1487 |
-
"grad_norm": 0.
|
| 1488 |
"learning_rate": 2.3170731707317075e-05,
|
| 1489 |
-
"loss":
|
| 1490 |
-
"mean_token_accuracy":
|
| 1491 |
-
"num_tokens":
|
| 1492 |
"step": 146
|
| 1493 |
},
|
| 1494 |
{
|
| 1495 |
-
"entropy": 0.
|
| 1496 |
"epoch": 1.7963246554364471,
|
| 1497 |
-
"grad_norm": 0.
|
| 1498 |
"learning_rate": 2.1951219512195124e-05,
|
| 1499 |
-
"loss":
|
| 1500 |
-
"mean_token_accuracy":
|
| 1501 |
-
"num_tokens":
|
| 1502 |
"step": 147
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
-
"entropy": 0.
|
| 1506 |
"epoch": 1.8085758039816233,
|
| 1507 |
-
"grad_norm": 0.
|
| 1508 |
"learning_rate": 2.073170731707317e-05,
|
| 1509 |
-
"loss":
|
| 1510 |
-
"mean_token_accuracy":
|
| 1511 |
-
"num_tokens":
|
| 1512 |
"step": 148
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
-
"entropy": 0.
|
| 1516 |
"epoch": 1.8208269525267995,
|
| 1517 |
-
"grad_norm":
|
| 1518 |
"learning_rate": 1.9512195121951222e-05,
|
| 1519 |
-
"loss":
|
| 1520 |
-
"mean_token_accuracy":
|
| 1521 |
-
"num_tokens":
|
| 1522 |
"step": 149
|
| 1523 |
},
|
| 1524 |
{
|
| 1525 |
-
"entropy": 0.
|
| 1526 |
"epoch": 1.8330781010719757,
|
| 1527 |
-
"grad_norm": 0.
|
| 1528 |
"learning_rate": 1.8292682926829268e-05,
|
| 1529 |
-
"loss": 0.
|
| 1530 |
-
"mean_token_accuracy":
|
| 1531 |
-
"num_tokens":
|
| 1532 |
"step": 150
|
| 1533 |
},
|
| 1534 |
{
|
| 1535 |
"epoch": 1.8330781010719757,
|
| 1536 |
-
"eval_entropy": 0.
|
| 1537 |
-
"eval_loss": 0.
|
| 1538 |
-
"eval_mean_token_accuracy": 0.
|
| 1539 |
-
"eval_num_tokens":
|
| 1540 |
-
"eval_runtime":
|
| 1541 |
-
"eval_samples_per_second": 1.
|
| 1542 |
-
"eval_steps_per_second": 1.
|
| 1543 |
"step": 150
|
| 1544 |
},
|
| 1545 |
{
|
| 1546 |
-
"entropy": 0.
|
| 1547 |
"epoch": 1.8453292496171516,
|
| 1548 |
-
"grad_norm": 0.
|
| 1549 |
"learning_rate": 1.707317073170732e-05,
|
| 1550 |
-
"loss":
|
| 1551 |
-
"mean_token_accuracy":
|
| 1552 |
-
"num_tokens":
|
| 1553 |
"step": 151
|
| 1554 |
},
|
| 1555 |
{
|
| 1556 |
-
"entropy": 0.
|
| 1557 |
"epoch": 1.8575803981623276,
|
| 1558 |
-
"grad_norm": 0.
|
| 1559 |
"learning_rate": 1.5853658536585366e-05,
|
| 1560 |
-
"loss":
|
| 1561 |
-
"mean_token_accuracy":
|
| 1562 |
-
"num_tokens":
|
| 1563 |
"step": 152
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
-
"entropy": 0.
|
| 1567 |
"epoch": 1.8698315467075037,
|
| 1568 |
-
"grad_norm": 0.
|
| 1569 |
"learning_rate": 1.4634146341463415e-05,
|
| 1570 |
-
"loss":
|
| 1571 |
-
"mean_token_accuracy":
|
| 1572 |
-
"num_tokens":
|
| 1573 |
"step": 153
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
-
"entropy": 0.
|
| 1577 |
"epoch": 1.88208269525268,
|
| 1578 |
-
"grad_norm": 0.
|
| 1579 |
"learning_rate": 1.3414634146341466e-05,
|
| 1580 |
-
"loss":
|
| 1581 |
-
"mean_token_accuracy":
|
| 1582 |
-
"num_tokens":
|
| 1583 |
"step": 154
|
| 1584 |
},
|
| 1585 |
{
|
| 1586 |
-
"entropy": 0.
|
| 1587 |
"epoch": 1.894333843797856,
|
| 1588 |
-
"grad_norm": 0.
|
| 1589 |
"learning_rate": 1.2195121951219513e-05,
|
| 1590 |
-
"loss": 0.
|
| 1591 |
-
"mean_token_accuracy": 0.
|
| 1592 |
-
"num_tokens":
|
| 1593 |
"step": 155
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
-
"entropy": 0.
|
| 1597 |
"epoch": 1.9065849923430322,
|
| 1598 |
-
"grad_norm": 0.
|
| 1599 |
"learning_rate": 1.0975609756097562e-05,
|
| 1600 |
-
"loss": 0.
|
| 1601 |
-
"mean_token_accuracy": 0.
|
| 1602 |
-
"num_tokens":
|
| 1603 |
"step": 156
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
-
"entropy": 0.
|
| 1607 |
"epoch": 1.9188361408882084,
|
| 1608 |
-
"grad_norm": 0.
|
| 1609 |
"learning_rate": 9.756097560975611e-06,
|
| 1610 |
-
"loss":
|
| 1611 |
-
"mean_token_accuracy":
|
| 1612 |
-
"num_tokens":
|
| 1613 |
"step": 157
|
| 1614 |
},
|
| 1615 |
{
|
| 1616 |
-
"entropy": 0.
|
| 1617 |
"epoch": 1.9310872894333844,
|
| 1618 |
-
"grad_norm": 0.
|
| 1619 |
"learning_rate": 8.53658536585366e-06,
|
| 1620 |
-
"loss":
|
| 1621 |
-
"mean_token_accuracy":
|
| 1622 |
-
"num_tokens":
|
| 1623 |
"step": 158
|
| 1624 |
},
|
| 1625 |
{
|
| 1626 |
-
"entropy": 0.
|
| 1627 |
"epoch": 1.9433384379785605,
|
| 1628 |
-
"grad_norm": 0.
|
| 1629 |
"learning_rate": 7.317073170731707e-06,
|
| 1630 |
-
"loss":
|
| 1631 |
-
"mean_token_accuracy":
|
| 1632 |
-
"num_tokens":
|
| 1633 |
"step": 159
|
| 1634 |
},
|
| 1635 |
{
|
| 1636 |
-
"entropy": 0.
|
| 1637 |
"epoch": 1.9555895865237365,
|
| 1638 |
-
"grad_norm": 0.
|
| 1639 |
"learning_rate": 6.0975609756097564e-06,
|
| 1640 |
-
"loss":
|
| 1641 |
-
"mean_token_accuracy":
|
| 1642 |
-
"num_tokens":
|
| 1643 |
"step": 160
|
| 1644 |
},
|
| 1645 |
{
|
| 1646 |
-
"entropy": 0.
|
| 1647 |
"epoch": 1.9678407350689127,
|
| 1648 |
-
"grad_norm": 0.
|
| 1649 |
"learning_rate": 4.8780487804878055e-06,
|
| 1650 |
-
"loss":
|
| 1651 |
-
"mean_token_accuracy":
|
| 1652 |
-
"num_tokens":
|
| 1653 |
"step": 161
|
| 1654 |
},
|
| 1655 |
{
|
| 1656 |
-
"entropy": 0.
|
| 1657 |
"epoch": 1.9800918836140888,
|
| 1658 |
-
"grad_norm": 0.
|
| 1659 |
"learning_rate": 3.6585365853658537e-06,
|
| 1660 |
-
"loss":
|
| 1661 |
-
"mean_token_accuracy":
|
| 1662 |
-
"num_tokens":
|
| 1663 |
"step": 162
|
| 1664 |
},
|
| 1665 |
{
|
| 1666 |
-
"entropy": 0.
|
| 1667 |
"epoch": 1.992343032159265,
|
| 1668 |
-
"grad_norm": 0.
|
| 1669 |
"learning_rate": 2.4390243902439027e-06,
|
| 1670 |
-
"loss":
|
| 1671 |
-
"mean_token_accuracy":
|
| 1672 |
-
"num_tokens":
|
| 1673 |
"step": 163
|
| 1674 |
},
|
| 1675 |
{
|
| 1676 |
-
"entropy": 0.
|
| 1677 |
"epoch": 2.0,
|
| 1678 |
-
"grad_norm": 0.
|
| 1679 |
"learning_rate": 1.2195121951219514e-06,
|
| 1680 |
-
"loss": 0.
|
| 1681 |
-
"mean_token_accuracy":
|
| 1682 |
-
"num_tokens":
|
| 1683 |
"step": 164
|
| 1684 |
}
|
| 1685 |
],
|
|
@@ -1700,7 +1700,7 @@
|
|
| 1700 |
"attributes": {}
|
| 1701 |
}
|
| 1702 |
},
|
| 1703 |
-
"total_flos":
|
| 1704 |
"train_batch_size": 1,
|
| 1705 |
"trial_name": null,
|
| 1706 |
"trial_params": null
|
|
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"entropy": 0.22490596678107977,
|
| 14 |
"epoch": 0.01225114854517611,
|
| 15 |
+
"grad_norm": 8.875,
|
| 16 |
"learning_rate": 0.0002,
|
| 17 |
+
"loss": 0.1875426322221756,
|
| 18 |
+
"mean_token_accuracy": 0.9461580626666546,
|
| 19 |
+
"num_tokens": 6770.0,
|
| 20 |
"step": 1
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"entropy": 0.24707041680812836,
|
| 24 |
"epoch": 0.02450229709035222,
|
| 25 |
+
"grad_norm": 8.4375,
|
| 26 |
"learning_rate": 0.00019878048780487805,
|
| 27 |
+
"loss": 0.16050274670124054,
|
| 28 |
+
"mean_token_accuracy": 0.9445944800972939,
|
| 29 |
+
"num_tokens": 14234.0,
|
| 30 |
"step": 2
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"entropy": 0.32129648607224226,
|
| 34 |
"epoch": 0.036753445635528334,
|
| 35 |
+
"grad_norm": 2.75,
|
| 36 |
"learning_rate": 0.0001975609756097561,
|
| 37 |
+
"loss": 0.09863867610692978,
|
| 38 |
+
"mean_token_accuracy": 0.9659304060041904,
|
| 39 |
+
"num_tokens": 20673.0,
|
| 40 |
"step": 3
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"entropy": 0.32960685156285763,
|
| 44 |
"epoch": 0.04900459418070444,
|
| 45 |
+
"grad_norm": 1.671875,
|
| 46 |
"learning_rate": 0.00019634146341463416,
|
| 47 |
+
"loss": 0.08542143553495407,
|
| 48 |
+
"mean_token_accuracy": 0.9690693095326424,
|
| 49 |
+
"num_tokens": 26890.0,
|
| 50 |
"step": 4
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"entropy": 0.2677983660250902,
|
| 54 |
"epoch": 0.06125574272588055,
|
| 55 |
+
"grad_norm": 1.359375,
|
| 56 |
"learning_rate": 0.0001951219512195122,
|
| 57 |
+
"loss": 0.08666501939296722,
|
| 58 |
+
"mean_token_accuracy": 0.968298003077507,
|
| 59 |
+
"num_tokens": 35017.0,
|
| 60 |
"step": 5
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"entropy": 0.3096502358093858,
|
| 64 |
"epoch": 0.07350689127105667,
|
| 65 |
+
"grad_norm": 0.66015625,
|
| 66 |
"learning_rate": 0.00019390243902439025,
|
| 67 |
+
"loss": 0.07875043898820877,
|
| 68 |
+
"mean_token_accuracy": 0.969221331179142,
|
| 69 |
+
"num_tokens": 41478.0,
|
| 70 |
"step": 6
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"entropy": 0.3156957607716322,
|
| 74 |
"epoch": 0.08575803981623277,
|
| 75 |
+
"grad_norm": 2.0,
|
| 76 |
"learning_rate": 0.0001926829268292683,
|
| 77 |
+
"loss": 0.07807251811027527,
|
| 78 |
+
"mean_token_accuracy": 0.9681689888238907,
|
| 79 |
+
"num_tokens": 48204.0,
|
| 80 |
"step": 7
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"entropy": 0.2759731076657772,
|
| 84 |
"epoch": 0.09800918836140889,
|
| 85 |
+
"grad_norm": 1.1328125,
|
| 86 |
"learning_rate": 0.00019146341463414633,
|
| 87 |
+
"loss": 0.07681904733181,
|
| 88 |
+
"mean_token_accuracy": 0.9719767943024635,
|
| 89 |
+
"num_tokens": 54668.0,
|
| 90 |
"step": 8
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"entropy": 0.24453612882643938,
|
| 94 |
"epoch": 0.11026033690658499,
|
| 95 |
+
"grad_norm": 0.875,
|
| 96 |
"learning_rate": 0.0001902439024390244,
|
| 97 |
+
"loss": 0.07310224324464798,
|
| 98 |
+
"mean_token_accuracy": 0.96934475004673,
|
| 99 |
+
"num_tokens": 61929.0,
|
| 100 |
"step": 9
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"entropy": 0.25852775294333696,
|
| 104 |
"epoch": 0.1225114854517611,
|
| 105 |
+
"grad_norm": 1.4921875,
|
| 106 |
"learning_rate": 0.00018902439024390244,
|
| 107 |
+
"loss": 0.07384984195232391,
|
| 108 |
+
"mean_token_accuracy": 0.9701811708509922,
|
| 109 |
+
"num_tokens": 69036.0,
|
| 110 |
"step": 10
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"entropy": 0.27396084927022457,
|
| 114 |
"epoch": 0.13476263399693722,
|
| 115 |
+
"grad_norm": 0.94140625,
|
| 116 |
"learning_rate": 0.0001878048780487805,
|
| 117 |
+
"loss": 0.10277765244245529,
|
| 118 |
+
"mean_token_accuracy": 0.9634475558996201,
|
| 119 |
+
"num_tokens": 76394.0,
|
| 120 |
"step": 11
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"entropy": 0.3001147015020251,
|
| 124 |
"epoch": 0.14701378254211334,
|
| 125 |
+
"grad_norm": 0.84765625,
|
| 126 |
"learning_rate": 0.00018658536585365856,
|
| 127 |
+
"loss": 0.08927591890096664,
|
| 128 |
+
"mean_token_accuracy": 0.9625685028731823,
|
| 129 |
+
"num_tokens": 84073.0,
|
| 130 |
"step": 12
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"entropy": 0.29679975286126137,
|
| 134 |
"epoch": 0.15926493108728942,
|
| 135 |
+
"grad_norm": 0.8359375,
|
| 136 |
"learning_rate": 0.0001853658536585366,
|
| 137 |
+
"loss": 0.10607243329286575,
|
| 138 |
+
"mean_token_accuracy": 0.9608454070985317,
|
| 139 |
+
"num_tokens": 91135.0,
|
| 140 |
"step": 13
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"entropy": 0.28288435423746705,
|
| 144 |
"epoch": 0.17151607963246554,
|
| 145 |
+
"grad_norm": 0.69921875,
|
| 146 |
"learning_rate": 0.00018414634146341464,
|
| 147 |
+
"loss": 0.07875586301088333,
|
| 148 |
+
"mean_token_accuracy": 0.9699672348797321,
|
| 149 |
+
"num_tokens": 97740.0,
|
| 150 |
"step": 14
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"entropy": 0.2927901232615113,
|
| 154 |
"epoch": 0.18376722817764166,
|
| 155 |
+
"grad_norm": 0.8515625,
|
| 156 |
"learning_rate": 0.0001829268292682927,
|
| 157 |
+
"loss": 0.08531365543603897,
|
| 158 |
+
"mean_token_accuracy": 0.9704407565295696,
|
| 159 |
+
"num_tokens": 104242.0,
|
| 160 |
"step": 15
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"entropy": 0.2802786426618695,
|
| 164 |
"epoch": 0.19601837672281777,
|
| 165 |
+
"grad_norm": 2.609375,
|
| 166 |
"learning_rate": 0.00018170731707317075,
|
| 167 |
+
"loss": 0.10073477029800415,
|
| 168 |
+
"mean_token_accuracy": 0.9676352478563786,
|
| 169 |
+
"num_tokens": 112245.0,
|
| 170 |
"step": 16
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"entropy": 0.30663597770035267,
|
| 174 |
"epoch": 0.2082695252679939,
|
| 175 |
+
"grad_norm": 2.3125,
|
| 176 |
"learning_rate": 0.0001804878048780488,
|
| 177 |
+
"loss": 0.11375448107719421,
|
| 178 |
+
"mean_token_accuracy": 0.9604234844446182,
|
| 179 |
+
"num_tokens": 119086.0,
|
| 180 |
"step": 17
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"entropy": 0.31370354909449816,
|
| 184 |
"epoch": 0.22052067381316998,
|
| 185 |
+
"grad_norm": 2.3125,
|
| 186 |
"learning_rate": 0.00017926829268292684,
|
| 187 |
+
"loss": 0.10826913267374039,
|
| 188 |
+
"mean_token_accuracy": 0.9599097929894924,
|
| 189 |
+
"num_tokens": 126539.0,
|
| 190 |
"step": 18
|
| 191 |
},
|
| 192 |
{
|
| 193 |
+
"entropy": 0.3203959669917822,
|
| 194 |
"epoch": 0.2327718223583461,
|
| 195 |
+
"grad_norm": 1.3203125,
|
| 196 |
"learning_rate": 0.00017804878048780488,
|
| 197 |
+
"loss": 0.075275719165802,
|
| 198 |
+
"mean_token_accuracy": 0.9775180667638779,
|
| 199 |
+
"num_tokens": 133104.0,
|
| 200 |
"step": 19
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"entropy": 0.32591533567756414,
|
| 204 |
"epoch": 0.2450229709035222,
|
| 205 |
+
"grad_norm": 0.921875,
|
| 206 |
"learning_rate": 0.00017682926829268295,
|
| 207 |
+
"loss": 0.09778374433517456,
|
| 208 |
+
"mean_token_accuracy": 0.9647064991295338,
|
| 209 |
+
"num_tokens": 139853.0,
|
| 210 |
"step": 20
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"entropy": 0.3228916050866246,
|
| 214 |
"epoch": 0.2572741194486983,
|
| 215 |
+
"grad_norm": 1.9453125,
|
| 216 |
"learning_rate": 0.000175609756097561,
|
| 217 |
+
"loss": 0.11495943367481232,
|
| 218 |
+
"mean_token_accuracy": 0.9573761746287346,
|
| 219 |
+
"num_tokens": 146482.0,
|
| 220 |
"step": 21
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"entropy": 0.3363859634846449,
|
| 224 |
"epoch": 0.26952526799387444,
|
| 225 |
+
"grad_norm": 1.5,
|
| 226 |
"learning_rate": 0.00017439024390243903,
|
| 227 |
+
"loss": 0.10473912209272385,
|
| 228 |
+
"mean_token_accuracy": 0.9618786759674549,
|
| 229 |
+
"num_tokens": 153819.0,
|
| 230 |
"step": 22
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"entropy": 0.3069695383310318,
|
| 234 |
"epoch": 0.28177641653905056,
|
| 235 |
+
"grad_norm": 1.3203125,
|
| 236 |
"learning_rate": 0.00017317073170731708,
|
| 237 |
+
"loss": 0.09256276488304138,
|
| 238 |
+
"mean_token_accuracy": 0.9625396281480789,
|
| 239 |
+
"num_tokens": 160972.0,
|
| 240 |
"step": 23
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"entropy": 0.3574997428804636,
|
| 244 |
"epoch": 0.29402756508422667,
|
| 245 |
+
"grad_norm": 0.62890625,
|
| 246 |
"learning_rate": 0.00017195121951219512,
|
| 247 |
+
"loss": 0.09489140659570694,
|
| 248 |
+
"mean_token_accuracy": 0.9578843042254448,
|
| 249 |
+
"num_tokens": 167730.0,
|
| 250 |
"step": 24
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"entropy": 0.3444826593622565,
|
| 254 |
"epoch": 0.30627871362940273,
|
| 255 |
+
"grad_norm": 0.56640625,
|
| 256 |
"learning_rate": 0.0001707317073170732,
|
| 257 |
+
"loss": 0.09492132067680359,
|
| 258 |
+
"mean_token_accuracy": 0.9603794105350971,
|
| 259 |
+
"num_tokens": 174078.0,
|
| 260 |
"step": 25
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"entropy": 0.328093777410686,
|
| 264 |
"epoch": 0.31852986217457885,
|
| 265 |
+
"grad_norm": 0.97265625,
|
| 266 |
"learning_rate": 0.00016951219512195123,
|
| 267 |
+
"loss": 0.08727280050516129,
|
| 268 |
+
"mean_token_accuracy": 0.9661480598151684,
|
| 269 |
+
"num_tokens": 180867.0,
|
| 270 |
"step": 26
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"entropy": 0.3213672311976552,
|
| 274 |
"epoch": 0.33078101071975496,
|
| 275 |
+
"grad_norm": 1.203125,
|
| 276 |
"learning_rate": 0.00016829268292682927,
|
| 277 |
+
"loss": 0.07705243676900864,
|
| 278 |
+
"mean_token_accuracy": 0.9675347730517387,
|
| 279 |
+
"num_tokens": 187459.0,
|
| 280 |
"step": 27
|
| 281 |
},
|
| 282 |
{
|
| 283 |
+
"entropy": 0.3209801884368062,
|
| 284 |
"epoch": 0.3430321592649311,
|
| 285 |
+
"grad_norm": 0.49609375,
|
| 286 |
"learning_rate": 0.00016707317073170731,
|
| 287 |
+
"loss": 0.08744930475950241,
|
| 288 |
+
"mean_token_accuracy": 0.9658873043954372,
|
| 289 |
+
"num_tokens": 194265.0,
|
| 290 |
"step": 28
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"entropy": 0.2975130006670952,
|
| 294 |
"epoch": 0.3552833078101072,
|
| 295 |
+
"grad_norm": 0.578125,
|
| 296 |
"learning_rate": 0.00016585365853658536,
|
| 297 |
+
"loss": 0.08422811329364777,
|
| 298 |
+
"mean_token_accuracy": 0.9715595282614231,
|
| 299 |
+
"num_tokens": 201332.0,
|
| 300 |
"step": 29
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"entropy": 0.29833013843744993,
|
| 304 |
"epoch": 0.3675344563552833,
|
| 305 |
+
"grad_norm": 0.82421875,
|
| 306 |
"learning_rate": 0.00016463414634146343,
|
| 307 |
+
"loss": 0.08079958707094193,
|
| 308 |
+
"mean_token_accuracy": 0.9676232784986496,
|
| 309 |
+
"num_tokens": 208902.0,
|
| 310 |
"step": 30
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"entropy": 0.31810148898512125,
|
| 314 |
"epoch": 0.37978560490045943,
|
| 315 |
+
"grad_norm": 0.6796875,
|
| 316 |
"learning_rate": 0.00016341463414634147,
|
| 317 |
+
"loss": 0.09296617656946182,
|
| 318 |
+
"mean_token_accuracy": 0.9628731682896614,
|
| 319 |
+
"num_tokens": 214635.0,
|
| 320 |
"step": 31
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"entropy": 0.2774961022660136,
|
| 324 |
"epoch": 0.39203675344563554,
|
| 325 |
+
"grad_norm": 1.2109375,
|
| 326 |
"learning_rate": 0.00016219512195121954,
|
| 327 |
+
"loss": 0.08057809621095657,
|
| 328 |
+
"mean_token_accuracy": 0.9683544635772705,
|
| 329 |
+
"num_tokens": 222703.0,
|
| 330 |
"step": 32
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"entropy": 0.2500351797789335,
|
| 334 |
"epoch": 0.40428790199081166,
|
| 335 |
+
"grad_norm": 0.6953125,
|
| 336 |
"learning_rate": 0.00016097560975609758,
|
| 337 |
+
"loss": 0.07790188491344452,
|
| 338 |
+
"mean_token_accuracy": 0.9730625562369823,
|
| 339 |
+
"num_tokens": 230136.0,
|
| 340 |
"step": 33
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"entropy": 0.27261121198534966,
|
| 344 |
"epoch": 0.4165390505359878,
|
| 345 |
+
"grad_norm": 1.21875,
|
| 346 |
"learning_rate": 0.00015975609756097562,
|
| 347 |
+
"loss": 0.08459997177124023,
|
| 348 |
+
"mean_token_accuracy": 0.9683701656758785,
|
| 349 |
+
"num_tokens": 236711.0,
|
| 350 |
"step": 34
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"entropy": 0.25461648125201464,
|
| 354 |
"epoch": 0.42879019908116384,
|
| 355 |
+
"grad_norm": 1.5078125,
|
| 356 |
"learning_rate": 0.00015853658536585366,
|
| 357 |
+
"loss": 0.09788602590560913,
|
| 358 |
+
"mean_token_accuracy": 0.9601947516202927,
|
| 359 |
+
"num_tokens": 243492.0,
|
| 360 |
"step": 35
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"entropy": 0.250462488271296,
|
| 364 |
"epoch": 0.44104134762633995,
|
| 365 |
+
"grad_norm": 0.62109375,
|
| 366 |
"learning_rate": 0.00015731707317073173,
|
| 367 |
+
"loss": 0.09664106369018555,
|
| 368 |
+
"mean_token_accuracy": 0.9635641165077686,
|
| 369 |
+
"num_tokens": 250330.0,
|
| 370 |
"step": 36
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"entropy": 0.26719998102635145,
|
| 374 |
"epoch": 0.45329249617151607,
|
| 375 |
+
"grad_norm": 0.609375,
|
| 376 |
"learning_rate": 0.00015609756097560978,
|
| 377 |
+
"loss": 0.08978135138750076,
|
| 378 |
+
"mean_token_accuracy": 0.9730992764234543,
|
| 379 |
+
"num_tokens": 257503.0,
|
| 380 |
"step": 37
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"entropy": 0.25437645614147186,
|
| 384 |
"epoch": 0.4655436447166922,
|
| 385 |
+
"grad_norm": 1.0859375,
|
| 386 |
"learning_rate": 0.00015487804878048782,
|
| 387 |
+
"loss": 0.08938639611005783,
|
| 388 |
+
"mean_token_accuracy": 0.9675878100097179,
|
| 389 |
+
"num_tokens": 264436.0,
|
| 390 |
"step": 38
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"entropy": 0.2722023595124483,
|
| 394 |
"epoch": 0.4777947932618683,
|
| 395 |
+
"grad_norm": 1.375,
|
| 396 |
"learning_rate": 0.00015365853658536586,
|
| 397 |
+
"loss": 0.07785381376743317,
|
| 398 |
+
"mean_token_accuracy": 0.9736072942614555,
|
| 399 |
+
"num_tokens": 270483.0,
|
| 400 |
"step": 39
|
| 401 |
},
|
| 402 |
{
|
| 403 |
+
"entropy": 0.3116175327450037,
|
| 404 |
"epoch": 0.4900459418070444,
|
| 405 |
+
"grad_norm": 0.65625,
|
| 406 |
"learning_rate": 0.0001524390243902439,
|
| 407 |
+
"loss": 0.09019558876752853,
|
| 408 |
+
"mean_token_accuracy": 0.9605641178786755,
|
| 409 |
+
"num_tokens": 276329.0,
|
| 410 |
"step": 40
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"entropy": 0.28687036503106356,
|
| 414 |
"epoch": 0.5022970903522205,
|
| 415 |
+
"grad_norm": 0.62890625,
|
| 416 |
"learning_rate": 0.00015121951219512197,
|
| 417 |
+
"loss": 0.0810370221734047,
|
| 418 |
+
"mean_token_accuracy": 0.9663555175065994,
|
| 419 |
+
"num_tokens": 281636.0,
|
| 420 |
"step": 41
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"entropy": 0.2999298516660929,
|
| 424 |
"epoch": 0.5145482388973966,
|
| 425 |
+
"grad_norm": 1.1484375,
|
| 426 |
"learning_rate": 0.00015000000000000001,
|
| 427 |
+
"loss": 0.06981078535318375,
|
| 428 |
+
"mean_token_accuracy": 0.9718391671776772,
|
| 429 |
+
"num_tokens": 287849.0,
|
| 430 |
"step": 42
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"entropy": 0.3097079414874315,
|
| 434 |
"epoch": 0.5267993874425727,
|
| 435 |
+
"grad_norm": 1.5546875,
|
| 436 |
"learning_rate": 0.00014878048780487806,
|
| 437 |
+
"loss": 0.09350281953811646,
|
| 438 |
+
"mean_token_accuracy": 0.9683773033320904,
|
| 439 |
+
"num_tokens": 294425.0,
|
| 440 |
"step": 43
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"entropy": 0.2796417009085417,
|
| 444 |
"epoch": 0.5390505359877489,
|
| 445 |
+
"grad_norm": 1.25,
|
| 446 |
"learning_rate": 0.0001475609756097561,
|
| 447 |
+
"loss": 0.09558023512363434,
|
| 448 |
+
"mean_token_accuracy": 0.9602576456964016,
|
| 449 |
+
"num_tokens": 301451.0,
|
| 450 |
"step": 44
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"entropy": 0.265599487349391,
|
| 454 |
"epoch": 0.5513016845329249,
|
| 455 |
+
"grad_norm": 0.59375,
|
| 456 |
"learning_rate": 0.00014634146341463414,
|
| 457 |
+
"loss": 0.07772304862737656,
|
| 458 |
+
"mean_token_accuracy": 0.9693298228085041,
|
| 459 |
+
"num_tokens": 308208.0,
|
| 460 |
"step": 45
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"entropy": 0.29693579114973545,
|
| 464 |
"epoch": 0.5635528330781011,
|
| 465 |
+
"grad_norm": 1.2109375,
|
| 466 |
"learning_rate": 0.0001451219512195122,
|
| 467 |
+
"loss": 0.09863201528787613,
|
| 468 |
+
"mean_token_accuracy": 0.963471919298172,
|
| 469 |
+
"num_tokens": 314427.0,
|
| 470 |
"step": 46
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"entropy": 0.2665130514651537,
|
| 474 |
"epoch": 0.5758039816232772,
|
| 475 |
+
"grad_norm": 0.80078125,
|
| 476 |
"learning_rate": 0.00014390243902439025,
|
| 477 |
+
"loss": 0.08794506639242172,
|
| 478 |
+
"mean_token_accuracy": 0.9714972339570522,
|
| 479 |
+
"num_tokens": 321146.0,
|
| 480 |
"step": 47
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"entropy": 0.2665897011756897,
|
| 484 |
"epoch": 0.5880551301684533,
|
| 485 |
+
"grad_norm": 0.765625,
|
| 486 |
"learning_rate": 0.0001426829268292683,
|
| 487 |
+
"loss": 0.07602453231811523,
|
| 488 |
+
"mean_token_accuracy": 0.9719848223030567,
|
| 489 |
+
"num_tokens": 326952.0,
|
| 490 |
"step": 48
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"entropy": 0.2814077762886882,
|
| 494 |
"epoch": 0.6003062787136294,
|
| 495 |
+
"grad_norm": 0.74609375,
|
| 496 |
"learning_rate": 0.00014146341463414634,
|
| 497 |
+
"loss": 0.08512163907289505,
|
| 498 |
+
"mean_token_accuracy": 0.9680779539048672,
|
| 499 |
+
"num_tokens": 333716.0,
|
| 500 |
"step": 49
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"entropy": 0.311913987621665,
|
| 504 |
"epoch": 0.6125574272588055,
|
| 505 |
+
"grad_norm": 0.52734375,
|
| 506 |
"learning_rate": 0.00014024390243902438,
|
| 507 |
+
"loss": 0.0735088661313057,
|
| 508 |
+
"mean_token_accuracy": 0.9693484716117382,
|
| 509 |
+
"num_tokens": 339075.0,
|
| 510 |
"step": 50
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 0.6125574272588055,
|
| 514 |
+
"eval_entropy": 0.2772115924652072,
|
| 515 |
+
"eval_loss": 0.08680303394794464,
|
| 516 |
+
"eval_mean_token_accuracy": 0.9665399781171826,
|
| 517 |
+
"eval_num_tokens": 339075.0,
|
| 518 |
+
"eval_runtime": 64.122,
|
| 519 |
+
"eval_samples_per_second": 1.076,
|
| 520 |
+
"eval_steps_per_second": 1.076,
|
| 521 |
"step": 50
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"entropy": 0.2713254941627383,
|
| 525 |
"epoch": 0.6248085758039816,
|
| 526 |
+
"grad_norm": 0.76953125,
|
| 527 |
"learning_rate": 0.00013902439024390245,
|
| 528 |
+
"loss": 0.07364857941865921,
|
| 529 |
+
"mean_token_accuracy": 0.9685694649815559,
|
| 530 |
+
"num_tokens": 346211.0,
|
| 531 |
"step": 51
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"entropy": 0.27622572146356106,
|
| 535 |
"epoch": 0.6370597243491577,
|
| 536 |
+
"grad_norm": 1.953125,
|
| 537 |
"learning_rate": 0.0001378048780487805,
|
| 538 |
+
"loss": 0.08796638250350952,
|
| 539 |
+
"mean_token_accuracy": 0.9678671807050705,
|
| 540 |
+
"num_tokens": 353743.0,
|
| 541 |
"step": 52
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"entropy": 0.3153565675020218,
|
| 545 |
"epoch": 0.6493108728943339,
|
| 546 |
+
"grad_norm": 1.8125,
|
| 547 |
"learning_rate": 0.00013658536585365856,
|
| 548 |
+
"loss": 0.09189874678850174,
|
| 549 |
+
"mean_token_accuracy": 0.9696005284786224,
|
| 550 |
+
"num_tokens": 360765.0,
|
| 551 |
"step": 53
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"entropy": 0.2793878586962819,
|
| 555 |
"epoch": 0.6615620214395099,
|
| 556 |
+
"grad_norm": 0.94140625,
|
| 557 |
"learning_rate": 0.0001353658536585366,
|
| 558 |
+
"loss": 0.0844489261507988,
|
| 559 |
+
"mean_token_accuracy": 0.9678577370941639,
|
| 560 |
+
"num_tokens": 366800.0,
|
| 561 |
"step": 54
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"entropy": 0.31044898089021444,
|
| 565 |
"epoch": 0.6738131699846861,
|
| 566 |
+
"grad_norm": 0.9375,
|
| 567 |
"learning_rate": 0.00013414634146341464,
|
| 568 |
+
"loss": 0.07886157184839249,
|
| 569 |
+
"mean_token_accuracy": 0.9673981033265591,
|
| 570 |
+
"num_tokens": 373439.0,
|
| 571 |
"step": 55
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"entropy": 0.27184910606592894,
|
| 575 |
"epoch": 0.6860643185298622,
|
| 576 |
+
"grad_norm": 1.3359375,
|
| 577 |
"learning_rate": 0.0001329268292682927,
|
| 578 |
+
"loss": 0.0787871852517128,
|
| 579 |
+
"mean_token_accuracy": 0.9677317887544632,
|
| 580 |
+
"num_tokens": 380492.0,
|
| 581 |
"step": 56
|
| 582 |
},
|
| 583 |
{
|
| 584 |
+
"entropy": 0.31349051371216774,
|
| 585 |
"epoch": 0.6983154670750383,
|
| 586 |
+
"grad_norm": 1.59375,
|
| 587 |
"learning_rate": 0.00013170731707317076,
|
| 588 |
+
"loss": 0.08862332254648209,
|
| 589 |
+
"mean_token_accuracy": 0.9652546346187592,
|
| 590 |
+
"num_tokens": 386711.0,
|
| 591 |
"step": 57
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"entropy": 0.2799685625359416,
|
| 595 |
"epoch": 0.7105666156202144,
|
| 596 |
+
"grad_norm": 1.5078125,
|
| 597 |
"learning_rate": 0.0001304878048780488,
|
| 598 |
+
"loss": 0.10028493404388428,
|
| 599 |
+
"mean_token_accuracy": 0.9606899172067642,
|
| 600 |
+
"num_tokens": 394124.0,
|
| 601 |
"step": 58
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"entropy": 0.2792940763756633,
|
| 605 |
"epoch": 0.7228177641653905,
|
| 606 |
+
"grad_norm": 1.5859375,
|
| 607 |
"learning_rate": 0.00012926829268292684,
|
| 608 |
+
"loss": 0.07462260127067566,
|
| 609 |
+
"mean_token_accuracy": 0.9740471467375755,
|
| 610 |
+
"num_tokens": 401499.0,
|
| 611 |
"step": 59
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"entropy": 0.29724057391285896,
|
| 615 |
"epoch": 0.7350689127105666,
|
| 616 |
+
"grad_norm": 0.95703125,
|
| 617 |
"learning_rate": 0.00012804878048780488,
|
| 618 |
+
"loss": 0.06339482963085175,
|
| 619 |
+
"mean_token_accuracy": 0.9754546955227852,
|
| 620 |
+
"num_tokens": 407443.0,
|
| 621 |
"step": 60
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"entropy": 0.2698040744289756,
|
| 625 |
"epoch": 0.7473200612557427,
|
| 626 |
+
"grad_norm": 0.60546875,
|
| 627 |
"learning_rate": 0.00012682926829268293,
|
| 628 |
+
"loss": 0.10221480578184128,
|
| 629 |
+
"mean_token_accuracy": 0.9670109152793884,
|
| 630 |
+
"num_tokens": 415471.0,
|
| 631 |
"step": 61
|
| 632 |
},
|
| 633 |
{
|
| 634 |
+
"entropy": 0.2995635373517871,
|
| 635 |
"epoch": 0.7595712098009189,
|
| 636 |
+
"grad_norm": 1.75,
|
| 637 |
"learning_rate": 0.000125609756097561,
|
| 638 |
+
"loss": 0.08588436245918274,
|
| 639 |
+
"mean_token_accuracy": 0.9686382673680782,
|
| 640 |
+
"num_tokens": 422504.0,
|
| 641 |
"step": 62
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"entropy": 0.2458120621740818,
|
| 645 |
"epoch": 0.7718223583460949,
|
| 646 |
+
"grad_norm": 2.1875,
|
| 647 |
"learning_rate": 0.00012439024390243904,
|
| 648 |
+
"loss": 0.08629653602838516,
|
| 649 |
+
"mean_token_accuracy": 0.966422975063324,
|
| 650 |
+
"num_tokens": 430143.0,
|
| 651 |
"step": 63
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"entropy": 0.2900782600045204,
|
| 655 |
"epoch": 0.7840735068912711,
|
| 656 |
+
"grad_norm": 0.9296875,
|
| 657 |
"learning_rate": 0.00012317073170731708,
|
| 658 |
+
"loss": 0.08716308325529099,
|
| 659 |
+
"mean_token_accuracy": 0.965714868158102,
|
| 660 |
+
"num_tokens": 435664.0,
|
| 661 |
"step": 64
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"entropy": 0.29250922333449125,
|
| 665 |
"epoch": 0.7963246554364471,
|
| 666 |
+
"grad_norm": 0.59375,
|
| 667 |
"learning_rate": 0.00012195121951219512,
|
| 668 |
+
"loss": 0.08158774673938751,
|
| 669 |
+
"mean_token_accuracy": 0.9694335348904133,
|
| 670 |
+
"num_tokens": 442457.0,
|
| 671 |
"step": 65
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"entropy": 0.3083174014464021,
|
| 675 |
"epoch": 0.8085758039816233,
|
| 676 |
+
"grad_norm": 0.82421875,
|
| 677 |
"learning_rate": 0.00012073170731707318,
|
| 678 |
+
"loss": 0.0988016203045845,
|
| 679 |
+
"mean_token_accuracy": 0.9648039489984512,
|
| 680 |
+
"num_tokens": 449983.0,
|
| 681 |
"step": 66
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"entropy": 0.25693165976554155,
|
| 685 |
"epoch": 0.8208269525267994,
|
| 686 |
+
"grad_norm": 0.74609375,
|
| 687 |
"learning_rate": 0.00011951219512195122,
|
| 688 |
+
"loss": 0.07928164303302765,
|
| 689 |
+
"mean_token_accuracy": 0.9698546566069126,
|
| 690 |
+
"num_tokens": 457640.0,
|
| 691 |
"step": 67
|
| 692 |
},
|
| 693 |
{
|
| 694 |
+
"entropy": 0.2752681290730834,
|
| 695 |
"epoch": 0.8330781010719756,
|
| 696 |
+
"grad_norm": 0.97265625,
|
| 697 |
"learning_rate": 0.00011829268292682926,
|
| 698 |
+
"loss": 0.07464170455932617,
|
| 699 |
+
"mean_token_accuracy": 0.9697864800691605,
|
| 700 |
+
"num_tokens": 464050.0,
|
| 701 |
"step": 68
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"entropy": 0.27110164798796177,
|
| 705 |
"epoch": 0.8453292496171516,
|
| 706 |
+
"grad_norm": 0.71875,
|
| 707 |
"learning_rate": 0.00011707317073170732,
|
| 708 |
+
"loss": 0.0718315988779068,
|
| 709 |
+
"mean_token_accuracy": 0.9709942191839218,
|
| 710 |
+
"num_tokens": 469546.0,
|
| 711 |
"step": 69
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"entropy": 0.3264527218416333,
|
| 715 |
"epoch": 0.8575803981623277,
|
| 716 |
+
"grad_norm": 0.62109375,
|
| 717 |
"learning_rate": 0.00011585365853658536,
|
| 718 |
+
"loss": 0.0866687223315239,
|
| 719 |
+
"mean_token_accuracy": 0.9700192771852016,
|
| 720 |
+
"num_tokens": 475365.0,
|
| 721 |
"step": 70
|
| 722 |
},
|
| 723 |
{
|
| 724 |
+
"entropy": 0.3122966531664133,
|
| 725 |
"epoch": 0.8698315467075038,
|
| 726 |
+
"grad_norm": 0.67578125,
|
| 727 |
"learning_rate": 0.00011463414634146342,
|
| 728 |
+
"loss": 0.06088244915008545,
|
| 729 |
+
"mean_token_accuracy": 0.9754119366407394,
|
| 730 |
+
"num_tokens": 481830.0,
|
| 731 |
"step": 71
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"entropy": 0.3018254106864333,
|
| 735 |
"epoch": 0.8820826952526799,
|
| 736 |
+
"grad_norm": 0.56640625,
|
| 737 |
"learning_rate": 0.00011341463414634146,
|
| 738 |
+
"loss": 0.08657931536436081,
|
| 739 |
+
"mean_token_accuracy": 0.9676030017435551,
|
| 740 |
+
"num_tokens": 487767.0,
|
| 741 |
"step": 72
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"entropy": 0.3276115320622921,
|
| 745 |
"epoch": 0.8943338437978561,
|
| 746 |
+
"grad_norm": 0.5078125,
|
| 747 |
"learning_rate": 0.00011219512195121953,
|
| 748 |
+
"loss": 0.08024603128433228,
|
| 749 |
+
"mean_token_accuracy": 0.9690204374492168,
|
| 750 |
+
"num_tokens": 494428.0,
|
| 751 |
"step": 73
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"entropy": 0.32397411670535803,
|
| 755 |
"epoch": 0.9065849923430321,
|
| 756 |
+
"grad_norm": 1.1015625,
|
| 757 |
"learning_rate": 0.00011097560975609757,
|
| 758 |
+
"loss": 0.07867392897605896,
|
| 759 |
+
"mean_token_accuracy": 0.9685576297342777,
|
| 760 |
+
"num_tokens": 500828.0,
|
| 761 |
"step": 74
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"entropy": 0.319146528840065,
|
| 765 |
"epoch": 0.9188361408882083,
|
| 766 |
+
"grad_norm": 0.97265625,
|
| 767 |
"learning_rate": 0.00010975609756097563,
|
| 768 |
+
"loss": 0.08432602882385254,
|
| 769 |
+
"mean_token_accuracy": 0.9689616709947586,
|
| 770 |
+
"num_tokens": 507523.0,
|
| 771 |
"step": 75
|
| 772 |
},
|
| 773 |
{
|
| 774 |
+
"entropy": 0.3080446803942323,
|
| 775 |
"epoch": 0.9310872894333844,
|
| 776 |
+
"grad_norm": 1.265625,
|
| 777 |
"learning_rate": 0.00010853658536585367,
|
| 778 |
+
"loss": 0.0796058252453804,
|
| 779 |
+
"mean_token_accuracy": 0.9683922417461872,
|
| 780 |
+
"num_tokens": 513607.0,
|
| 781 |
"step": 76
|
| 782 |
},
|
| 783 |
{
|
| 784 |
+
"entropy": 0.2667541950941086,
|
| 785 |
"epoch": 0.9433384379785605,
|
| 786 |
+
"grad_norm": 0.59375,
|
| 787 |
"learning_rate": 0.00010731707317073172,
|
| 788 |
+
"loss": 0.06495777517557144,
|
| 789 |
+
"mean_token_accuracy": 0.977863471955061,
|
| 790 |
+
"num_tokens": 521376.0,
|
| 791 |
"step": 77
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"entropy": 0.27901614736765623,
|
| 795 |
"epoch": 0.9555895865237366,
|
| 796 |
+
"grad_norm": 1.0859375,
|
| 797 |
"learning_rate": 0.00010609756097560977,
|
| 798 |
+
"loss": 0.08389777690172195,
|
| 799 |
+
"mean_token_accuracy": 0.967527512460947,
|
| 800 |
+
"num_tokens": 528624.0,
|
| 801 |
"step": 78
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"entropy": 0.2754220822826028,
|
| 805 |
"epoch": 0.9678407350689127,
|
| 806 |
+
"grad_norm": 1.3515625,
|
| 807 |
"learning_rate": 0.00010487804878048781,
|
| 808 |
+
"loss": 0.0762331560254097,
|
| 809 |
+
"mean_token_accuracy": 0.9713698588311672,
|
| 810 |
+
"num_tokens": 534817.0,
|
| 811 |
"step": 79
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"entropy": 0.2981132147833705,
|
| 815 |
"epoch": 0.9800918836140888,
|
| 816 |
+
"grad_norm": 1.4375,
|
| 817 |
"learning_rate": 0.00010365853658536586,
|
| 818 |
+
"loss": 0.07953717559576035,
|
| 819 |
+
"mean_token_accuracy": 0.967929158359766,
|
| 820 |
+
"num_tokens": 541716.0,
|
| 821 |
"step": 80
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"entropy": 0.30576920323073864,
|
| 825 |
"epoch": 0.9923430321592649,
|
| 826 |
+
"grad_norm": 1.0234375,
|
| 827 |
"learning_rate": 0.0001024390243902439,
|
| 828 |
+
"loss": 0.07800528407096863,
|
| 829 |
+
"mean_token_accuracy": 0.971219640225172,
|
| 830 |
+
"num_tokens": 548000.0,
|
| 831 |
"step": 81
|
| 832 |
},
|
| 833 |
{
|
| 834 |
+
"entropy": 0.24986045509576799,
|
| 835 |
"epoch": 1.0,
|
| 836 |
+
"grad_norm": 0.58203125,
|
| 837 |
"learning_rate": 0.00010121951219512196,
|
| 838 |
+
"loss": 0.05879032611846924,
|
| 839 |
+
"mean_token_accuracy": 0.9748349964618683,
|
| 840 |
+
"num_tokens": 552608.0,
|
| 841 |
"step": 82
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"entropy": 0.2518839007243514,
|
| 845 |
"epoch": 1.0122511485451762,
|
| 846 |
+
"grad_norm": 0.49609375,
|
| 847 |
"learning_rate": 0.0001,
|
| 848 |
+
"loss": 0.047237373888492584,
|
| 849 |
+
"mean_token_accuracy": 0.9874232485890388,
|
| 850 |
+
"num_tokens": 559523.0,
|
| 851 |
"step": 83
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"entropy": 0.2561075631529093,
|
| 855 |
"epoch": 1.0245022970903521,
|
| 856 |
+
"grad_norm": 0.65234375,
|
| 857 |
"learning_rate": 9.878048780487805e-05,
|
| 858 |
+
"loss": 0.04376941919326782,
|
| 859 |
+
"mean_token_accuracy": 0.9896520264446735,
|
| 860 |
+
"num_tokens": 566232.0,
|
| 861 |
"step": 84
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"entropy": 0.2935391655191779,
|
| 865 |
"epoch": 1.0367534456355283,
|
| 866 |
+
"grad_norm": 0.486328125,
|
| 867 |
"learning_rate": 9.75609756097561e-05,
|
| 868 |
+
"loss": 0.052017997950315475,
|
| 869 |
+
"mean_token_accuracy": 0.9823879115283489,
|
| 870 |
+
"num_tokens": 573965.0,
|
| 871 |
"step": 85
|
| 872 |
},
|
| 873 |
{
|
| 874 |
+
"entropy": 0.21971730748191476,
|
| 875 |
"epoch": 1.0490045941807045,
|
| 876 |
+
"grad_norm": 0.330078125,
|
| 877 |
"learning_rate": 9.634146341463415e-05,
|
| 878 |
+
"loss": 0.04022914543747902,
|
| 879 |
+
"mean_token_accuracy": 0.9874378368258476,
|
| 880 |
+
"num_tokens": 580768.0,
|
| 881 |
"step": 86
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"entropy": 0.23719595093280077,
|
| 885 |
"epoch": 1.0612557427258806,
|
| 886 |
+
"grad_norm": 0.68359375,
|
| 887 |
"learning_rate": 9.51219512195122e-05,
|
| 888 |
+
"loss": 0.04782414808869362,
|
| 889 |
+
"mean_token_accuracy": 0.9846052750945091,
|
| 890 |
+
"num_tokens": 588097.0,
|
| 891 |
"step": 87
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"entropy": 0.25634779036045074,
|
| 895 |
"epoch": 1.0735068912710566,
|
| 896 |
+
"grad_norm": 0.291015625,
|
| 897 |
"learning_rate": 9.390243902439024e-05,
|
| 898 |
+
"loss": 0.03357430174946785,
|
| 899 |
+
"mean_token_accuracy": 0.9895204566419125,
|
| 900 |
+
"num_tokens": 594215.0,
|
| 901 |
"step": 88
|
| 902 |
},
|
| 903 |
{
|
| 904 |
+
"entropy": 0.26507470663636923,
|
| 905 |
"epoch": 1.0857580398162328,
|
| 906 |
+
"grad_norm": 0.89453125,
|
| 907 |
"learning_rate": 9.26829268292683e-05,
|
| 908 |
+
"loss": 0.0427095852792263,
|
| 909 |
+
"mean_token_accuracy": 0.984734483063221,
|
| 910 |
+
"num_tokens": 600018.0,
|
| 911 |
"step": 89
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"entropy": 0.25531507655978203,
|
| 915 |
"epoch": 1.098009188361409,
|
| 916 |
+
"grad_norm": 0.357421875,
|
| 917 |
"learning_rate": 9.146341463414635e-05,
|
| 918 |
+
"loss": 0.04051242396235466,
|
| 919 |
+
"mean_token_accuracy": 0.9878104776144028,
|
| 920 |
+
"num_tokens": 606254.0,
|
| 921 |
"step": 90
|
| 922 |
},
|
| 923 |
{
|
| 924 |
+
"entropy": 0.26176126673817635,
|
| 925 |
"epoch": 1.110260336906585,
|
| 926 |
+
"grad_norm": 0.55078125,
|
| 927 |
"learning_rate": 9.02439024390244e-05,
|
| 928 |
+
"loss": 0.03882109373807907,
|
| 929 |
+
"mean_token_accuracy": 0.9838540144264698,
|
| 930 |
+
"num_tokens": 612316.0,
|
| 931 |
"step": 91
|
| 932 |
},
|
| 933 |
{
|
| 934 |
+
"entropy": 0.2165100760757923,
|
| 935 |
"epoch": 1.122511485451761,
|
| 936 |
+
"grad_norm": 0.3671875,
|
| 937 |
"learning_rate": 8.902439024390244e-05,
|
| 938 |
+
"loss": 0.03010629303753376,
|
| 939 |
+
"mean_token_accuracy": 0.9918084405362606,
|
| 940 |
+
"num_tokens": 619629.0,
|
| 941 |
"step": 92
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"entropy": 0.24866555724292994,
|
| 945 |
"epoch": 1.1347626339969372,
|
| 946 |
+
"grad_norm": 0.7578125,
|
| 947 |
"learning_rate": 8.78048780487805e-05,
|
| 948 |
+
"loss": 0.03892926499247551,
|
| 949 |
+
"mean_token_accuracy": 0.984953761100769,
|
| 950 |
+
"num_tokens": 625947.0,
|
| 951 |
"step": 93
|
| 952 |
},
|
| 953 |
{
|
| 954 |
+
"entropy": 0.21699398616328835,
|
| 955 |
"epoch": 1.1470137825421134,
|
| 956 |
+
"grad_norm": 0.53125,
|
| 957 |
"learning_rate": 8.658536585365854e-05,
|
| 958 |
+
"loss": 0.040178049355745316,
|
| 959 |
+
"mean_token_accuracy": 0.986099898815155,
|
| 960 |
+
"num_tokens": 632906.0,
|
| 961 |
"step": 94
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"entropy": 0.2104594809934497,
|
| 965 |
"epoch": 1.1592649310872893,
|
| 966 |
+
"grad_norm": 1.4375,
|
| 967 |
"learning_rate": 8.53658536585366e-05,
|
| 968 |
+
"loss": 0.05103502795100212,
|
| 969 |
+
"mean_token_accuracy": 0.9873828142881393,
|
| 970 |
+
"num_tokens": 639769.0,
|
| 971 |
"step": 95
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"entropy": 0.21941981185227633,
|
| 975 |
"epoch": 1.1715160796324655,
|
| 976 |
+
"grad_norm": 0.984375,
|
| 977 |
"learning_rate": 8.414634146341464e-05,
|
| 978 |
+
"loss": 0.03593335300683975,
|
| 979 |
+
"mean_token_accuracy": 0.9901031330227852,
|
| 980 |
+
"num_tokens": 646347.0,
|
| 981 |
"step": 96
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"entropy": 0.23086606059223413,
|
| 985 |
"epoch": 1.1837672281776417,
|
| 986 |
+
"grad_norm": 0.65625,
|
| 987 |
"learning_rate": 8.292682926829268e-05,
|
| 988 |
+
"loss": 0.034123439341783524,
|
| 989 |
+
"mean_token_accuracy": 0.9874096475541592,
|
| 990 |
+
"num_tokens": 652247.0,
|
| 991 |
"step": 97
|
| 992 |
},
|
| 993 |
{
|
| 994 |
+
"entropy": 0.21858725044876337,
|
| 995 |
"epoch": 1.1960183767228179,
|
| 996 |
+
"grad_norm": 0.3515625,
|
| 997 |
"learning_rate": 8.170731707317073e-05,
|
| 998 |
+
"loss": 0.03983831778168678,
|
| 999 |
+
"mean_token_accuracy": 0.9883633032441139,
|
| 1000 |
+
"num_tokens": 659620.0,
|
| 1001 |
"step": 98
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
+
"entropy": 0.2186456574127078,
|
| 1005 |
"epoch": 1.2082695252679938,
|
| 1006 |
+
"grad_norm": 0.50390625,
|
| 1007 |
"learning_rate": 8.048780487804879e-05,
|
| 1008 |
+
"loss": 0.03659169375896454,
|
| 1009 |
+
"mean_token_accuracy": 0.9874354675412178,
|
| 1010 |
+
"num_tokens": 667017.0,
|
| 1011 |
"step": 99
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
+
"entropy": 0.21289387485012412,
|
| 1015 |
"epoch": 1.22052067381317,
|
| 1016 |
+
"grad_norm": 1.2890625,
|
| 1017 |
"learning_rate": 7.926829268292683e-05,
|
| 1018 |
+
"loss": 0.09039004892110825,
|
| 1019 |
+
"mean_token_accuracy": 0.9841732494533062,
|
| 1020 |
+
"num_tokens": 673866.0,
|
| 1021 |
"step": 100
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 1.22052067381317,
|
| 1025 |
+
"eval_entropy": 0.22615607968275098,
|
| 1026 |
+
"eval_loss": 0.0748714804649353,
|
| 1027 |
+
"eval_mean_token_accuracy": 0.9701917439267256,
|
| 1028 |
+
"eval_num_tokens": 673866.0,
|
| 1029 |
+
"eval_runtime": 64.1728,
|
| 1030 |
+
"eval_samples_per_second": 1.075,
|
| 1031 |
+
"eval_steps_per_second": 1.075,
|
| 1032 |
"step": 100
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
+
"entropy": 0.20847708079963923,
|
| 1036 |
"epoch": 1.2327718223583461,
|
| 1037 |
+
"grad_norm": 0.9453125,
|
| 1038 |
"learning_rate": 7.804878048780489e-05,
|
| 1039 |
+
"loss": 0.032662514597177505,
|
| 1040 |
+
"mean_token_accuracy": 0.9919092357158661,
|
| 1041 |
+
"num_tokens": 681308.0,
|
| 1042 |
"step": 101
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
+
"entropy": 0.23787071648985147,
|
| 1046 |
"epoch": 1.245022970903522,
|
| 1047 |
+
"grad_norm": 0.859375,
|
| 1048 |
"learning_rate": 7.682926829268293e-05,
|
| 1049 |
+
"loss": 0.044949762523174286,
|
| 1050 |
+
"mean_token_accuracy": 0.987742405384779,
|
| 1051 |
+
"num_tokens": 687496.0,
|
| 1052 |
"step": 102
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
+
"entropy": 0.21969830617308617,
|
| 1056 |
"epoch": 1.2572741194486983,
|
| 1057 |
+
"grad_norm": 0.8671875,
|
| 1058 |
"learning_rate": 7.560975609756099e-05,
|
| 1059 |
+
"loss": 0.036048222333192825,
|
| 1060 |
+
"mean_token_accuracy": 0.98578891903162,
|
| 1061 |
+
"num_tokens": 694818.0,
|
| 1062 |
"step": 103
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
+
"entropy": 0.228535583242774,
|
| 1066 |
"epoch": 1.2695252679938744,
|
| 1067 |
+
"grad_norm": 1.7109375,
|
| 1068 |
"learning_rate": 7.439024390243903e-05,
|
| 1069 |
+
"loss": 0.050321951508522034,
|
| 1070 |
+
"mean_token_accuracy": 0.9846261814236641,
|
| 1071 |
+
"num_tokens": 701351.0,
|
| 1072 |
"step": 104
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
+
"entropy": 0.21918219700455666,
|
| 1076 |
"epoch": 1.2817764165390506,
|
| 1077 |
+
"grad_norm": 0.57421875,
|
| 1078 |
"learning_rate": 7.317073170731707e-05,
|
| 1079 |
+
"loss": 0.03220512717962265,
|
| 1080 |
+
"mean_token_accuracy": 0.9897662363946438,
|
| 1081 |
+
"num_tokens": 707212.0,
|
| 1082 |
"step": 105
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
+
"entropy": 0.21648676693439484,
|
| 1086 |
"epoch": 1.2940275650842268,
|
| 1087 |
+
"grad_norm": 0.921875,
|
| 1088 |
"learning_rate": 7.195121951219513e-05,
|
| 1089 |
+
"loss": 0.031827542930841446,
|
| 1090 |
+
"mean_token_accuracy": 0.9904872179031372,
|
| 1091 |
+
"num_tokens": 714524.0,
|
| 1092 |
"step": 106
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
+
"entropy": 0.20004846714437008,
|
| 1096 |
"epoch": 1.3062787136294027,
|
| 1097 |
+
"grad_norm": 1.0234375,
|
| 1098 |
"learning_rate": 7.073170731707317e-05,
|
| 1099 |
+
"loss": 0.03981270268559456,
|
| 1100 |
+
"mean_token_accuracy": 0.9861926138401031,
|
| 1101 |
+
"num_tokens": 722033.0,
|
| 1102 |
"step": 107
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
+
"entropy": 0.21497153211385012,
|
| 1106 |
"epoch": 1.318529862174579,
|
| 1107 |
+
"grad_norm": 0.53515625,
|
| 1108 |
"learning_rate": 6.951219512195122e-05,
|
| 1109 |
+
"loss": 0.03612194582819939,
|
| 1110 |
+
"mean_token_accuracy": 0.9883794784545898,
|
| 1111 |
+
"num_tokens": 728835.0,
|
| 1112 |
"step": 108
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
+
"entropy": 0.22441515233367682,
|
| 1116 |
"epoch": 1.3307810107197549,
|
| 1117 |
+
"grad_norm": 0.66796875,
|
| 1118 |
"learning_rate": 6.829268292682928e-05,
|
| 1119 |
+
"loss": 0.037204962223768234,
|
| 1120 |
+
"mean_token_accuracy": 0.9865190424025059,
|
| 1121 |
+
"num_tokens": 735463.0,
|
| 1122 |
"step": 109
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
+
"entropy": 0.21172351390123367,
|
| 1126 |
"epoch": 1.343032159264931,
|
| 1127 |
+
"grad_norm": 0.314453125,
|
| 1128 |
"learning_rate": 6.707317073170732e-05,
|
| 1129 |
+
"loss": 0.03260833024978638,
|
| 1130 |
+
"mean_token_accuracy": 0.9877017810940742,
|
| 1131 |
+
"num_tokens": 742536.0,
|
| 1132 |
"step": 110
|
| 1133 |
},
|
| 1134 |
{
|
| 1135 |
+
"entropy": 0.19597876677289605,
|
| 1136 |
"epoch": 1.3552833078101072,
|
| 1137 |
+
"grad_norm": 0.419921875,
|
| 1138 |
"learning_rate": 6.585365853658538e-05,
|
| 1139 |
+
"loss": 0.0339697040617466,
|
| 1140 |
+
"mean_token_accuracy": 0.990579642355442,
|
| 1141 |
+
"num_tokens": 749606.0,
|
| 1142 |
"step": 111
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
+
"entropy": 0.21933963894844055,
|
| 1146 |
"epoch": 1.3675344563552834,
|
| 1147 |
+
"grad_norm": 0.53515625,
|
| 1148 |
"learning_rate": 6.463414634146342e-05,
|
| 1149 |
+
"loss": 0.028515402227640152,
|
| 1150 |
+
"mean_token_accuracy": 0.9883383698761463,
|
| 1151 |
+
"num_tokens": 755287.0,
|
| 1152 |
"step": 112
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
+
"entropy": 0.21494697034358978,
|
| 1156 |
"epoch": 1.3797856049004595,
|
| 1157 |
+
"grad_norm": 0.37890625,
|
| 1158 |
"learning_rate": 6.341463414634146e-05,
|
| 1159 |
+
"loss": 0.03924579173326492,
|
| 1160 |
+
"mean_token_accuracy": 0.9876385144889355,
|
| 1161 |
+
"num_tokens": 763515.0,
|
| 1162 |
"step": 113
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
+
"entropy": 0.22842750838026404,
|
| 1166 |
"epoch": 1.3920367534456355,
|
| 1167 |
+
"grad_norm": 1.1484375,
|
| 1168 |
"learning_rate": 6.219512195121952e-05,
|
| 1169 |
+
"loss": 0.0367334708571434,
|
| 1170 |
+
"mean_token_accuracy": 0.9872251562774181,
|
| 1171 |
+
"num_tokens": 769660.0,
|
| 1172 |
"step": 114
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"entropy": 0.2147415135987103,
|
| 1176 |
"epoch": 1.4042879019908117,
|
| 1177 |
+
"grad_norm": 0.921875,
|
| 1178 |
"learning_rate": 6.097560975609756e-05,
|
| 1179 |
+
"loss": 0.030023006722331047,
|
| 1180 |
+
"mean_token_accuracy": 0.9890519753098488,
|
| 1181 |
+
"num_tokens": 777068.0,
|
| 1182 |
"step": 115
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
+
"entropy": 0.2247378919273615,
|
| 1186 |
"epoch": 1.4165390505359878,
|
| 1187 |
+
"grad_norm": 0.9375,
|
| 1188 |
"learning_rate": 5.975609756097561e-05,
|
| 1189 |
+
"loss": 0.03915408253669739,
|
| 1190 |
+
"mean_token_accuracy": 0.9883266240358353,
|
| 1191 |
+
"num_tokens": 783422.0,
|
| 1192 |
"step": 116
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"entropy": 0.19090860895812511,
|
| 1196 |
"epoch": 1.4287901990811638,
|
| 1197 |
+
"grad_norm": 0.765625,
|
| 1198 |
"learning_rate": 5.853658536585366e-05,
|
| 1199 |
+
"loss": 0.037202730774879456,
|
| 1200 |
+
"mean_token_accuracy": 0.9874398410320282,
|
| 1201 |
+
"num_tokens": 790851.0,
|
| 1202 |
"step": 117
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
+
"entropy": 0.2285028137266636,
|
| 1206 |
"epoch": 1.44104134762634,
|
| 1207 |
+
"grad_norm": 1.9140625,
|
| 1208 |
"learning_rate": 5.731707317073171e-05,
|
| 1209 |
+
"loss": 0.043229859322309494,
|
| 1210 |
+
"mean_token_accuracy": 0.9905107729136944,
|
| 1211 |
+
"num_tokens": 797801.0,
|
| 1212 |
"step": 118
|
| 1213 |
},
|
| 1214 |
{
|
| 1215 |
+
"entropy": 0.2443255502730608,
|
| 1216 |
"epoch": 1.4532924961715161,
|
| 1217 |
+
"grad_norm": 0.365234375,
|
| 1218 |
"learning_rate": 5.6097560975609764e-05,
|
| 1219 |
+
"loss": 0.04100143164396286,
|
| 1220 |
+
"mean_token_accuracy": 0.9880562499165535,
|
| 1221 |
+
"num_tokens": 804371.0,
|
| 1222 |
"step": 119
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
+
"entropy": 0.19626039918512106,
|
| 1226 |
"epoch": 1.4655436447166923,
|
| 1227 |
+
"grad_norm": 0.83984375,
|
| 1228 |
"learning_rate": 5.487804878048781e-05,
|
| 1229 |
+
"loss": 0.038516998291015625,
|
| 1230 |
+
"mean_token_accuracy": 0.988171175122261,
|
| 1231 |
+
"num_tokens": 812335.0,
|
| 1232 |
"step": 120
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
+
"entropy": 0.2181866616010666,
|
| 1236 |
"epoch": 1.4777947932618682,
|
| 1237 |
+
"grad_norm": 0.53515625,
|
| 1238 |
"learning_rate": 5.365853658536586e-05,
|
| 1239 |
+
"loss": 0.02816646918654442,
|
| 1240 |
+
"mean_token_accuracy": 0.9916124008595943,
|
| 1241 |
+
"num_tokens": 818577.0,
|
| 1242 |
"step": 121
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"entropy": 0.20635052677243948,
|
| 1246 |
"epoch": 1.4900459418070444,
|
| 1247 |
+
"grad_norm": 0.74609375,
|
| 1248 |
"learning_rate": 5.2439024390243904e-05,
|
| 1249 |
+
"loss": 0.04106622561812401,
|
| 1250 |
+
"mean_token_accuracy": 0.9839451834559441,
|
| 1251 |
+
"num_tokens": 825535.0,
|
| 1252 |
"step": 122
|
| 1253 |
},
|
| 1254 |
{
|
| 1255 |
+
"entropy": 0.21835408825427294,
|
| 1256 |
"epoch": 1.5022970903522204,
|
| 1257 |
+
"grad_norm": 0.427734375,
|
| 1258 |
"learning_rate": 5.121951219512195e-05,
|
| 1259 |
+
"loss": 0.026341412216424942,
|
| 1260 |
+
"mean_token_accuracy": 0.9940293915569782,
|
| 1261 |
+
"num_tokens": 831505.0,
|
| 1262 |
"step": 123
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
+
"entropy": 0.21729151718318462,
|
| 1266 |
"epoch": 1.5145482388973965,
|
| 1267 |
+
"grad_norm": 0.455078125,
|
| 1268 |
"learning_rate": 5e-05,
|
| 1269 |
+
"loss": 0.028432821854948997,
|
| 1270 |
+
"mean_token_accuracy": 0.9925089627504349,
|
| 1271 |
+
"num_tokens": 838385.0,
|
| 1272 |
"step": 124
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
+
"entropy": 0.23625962156802416,
|
| 1276 |
"epoch": 1.5267993874425727,
|
| 1277 |
+
"grad_norm": 0.72265625,
|
| 1278 |
"learning_rate": 4.878048780487805e-05,
|
| 1279 |
+
"loss": 0.03885198384523392,
|
| 1280 |
+
"mean_token_accuracy": 0.9883155077695847,
|
| 1281 |
+
"num_tokens": 845433.0,
|
| 1282 |
"step": 125
|
| 1283 |
},
|
| 1284 |
{
|
| 1285 |
+
"entropy": 0.21153692342340946,
|
| 1286 |
"epoch": 1.5390505359877489,
|
| 1287 |
+
"grad_norm": 0.66796875,
|
| 1288 |
"learning_rate": 4.75609756097561e-05,
|
| 1289 |
+
"loss": 0.03570759296417236,
|
| 1290 |
+
"mean_token_accuracy": 0.9910184219479561,
|
| 1291 |
+
"num_tokens": 852471.0,
|
| 1292 |
"step": 126
|
| 1293 |
},
|
| 1294 |
{
|
| 1295 |
+
"entropy": 0.23752436228096485,
|
| 1296 |
"epoch": 1.551301684532925,
|
| 1297 |
+
"grad_norm": 0.640625,
|
| 1298 |
"learning_rate": 4.634146341463415e-05,
|
| 1299 |
+
"loss": 0.028638798743486404,
|
| 1300 |
+
"mean_token_accuracy": 0.9928638078272343,
|
| 1301 |
+
"num_tokens": 858702.0,
|
| 1302 |
"step": 127
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
+
"entropy": 0.2128417994827032,
|
| 1306 |
"epoch": 1.5635528330781012,
|
| 1307 |
+
"grad_norm": 0.8828125,
|
| 1308 |
"learning_rate": 4.51219512195122e-05,
|
| 1309 |
+
"loss": 0.029636576771736145,
|
| 1310 |
+
"mean_token_accuracy": 0.9910452663898468,
|
| 1311 |
+
"num_tokens": 865325.0,
|
| 1312 |
"step": 128
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
+
"entropy": 0.216589767485857,
|
| 1316 |
"epoch": 1.5758039816232772,
|
| 1317 |
+
"grad_norm": 0.453125,
|
| 1318 |
"learning_rate": 4.390243902439025e-05,
|
| 1319 |
+
"loss": 0.03238631784915924,
|
| 1320 |
+
"mean_token_accuracy": 0.9904623441398144,
|
| 1321 |
+
"num_tokens": 871341.0,
|
| 1322 |
"step": 129
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
+
"entropy": 0.19242106284946203,
|
| 1326 |
"epoch": 1.5880551301684533,
|
| 1327 |
+
"grad_norm": 0.392578125,
|
| 1328 |
"learning_rate": 4.26829268292683e-05,
|
| 1329 |
+
"loss": 0.0261989776045084,
|
| 1330 |
+
"mean_token_accuracy": 0.9925210140645504,
|
| 1331 |
+
"num_tokens": 878973.0,
|
| 1332 |
"step": 130
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
+
"entropy": 0.22208478767424822,
|
| 1336 |
"epoch": 1.6003062787136293,
|
| 1337 |
+
"grad_norm": 0.328125,
|
| 1338 |
"learning_rate": 4.146341463414634e-05,
|
| 1339 |
+
"loss": 0.029643766582012177,
|
| 1340 |
+
"mean_token_accuracy": 0.9926025420427322,
|
| 1341 |
+
"num_tokens": 885517.0,
|
| 1342 |
"step": 131
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
+
"entropy": 0.19283092580735683,
|
| 1346 |
"epoch": 1.6125574272588055,
|
| 1347 |
+
"grad_norm": 0.423828125,
|
| 1348 |
"learning_rate": 4.0243902439024395e-05,
|
| 1349 |
+
"loss": 0.03948017954826355,
|
| 1350 |
+
"mean_token_accuracy": 0.9875317811965942,
|
| 1351 |
+
"num_tokens": 893273.0,
|
| 1352 |
"step": 132
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
+
"entropy": 0.18790056556463242,
|
| 1356 |
"epoch": 1.6248085758039816,
|
| 1357 |
+
"grad_norm": 0.625,
|
| 1358 |
"learning_rate": 3.9024390243902444e-05,
|
| 1359 |
+
"loss": 0.025747017934918404,
|
| 1360 |
+
"mean_token_accuracy": 0.9934940375387669,
|
| 1361 |
+
"num_tokens": 900019.0,
|
| 1362 |
"step": 133
|
| 1363 |
},
|
| 1364 |
{
|
| 1365 |
+
"entropy": 0.20814241049811244,
|
| 1366 |
"epoch": 1.6370597243491578,
|
| 1367 |
+
"grad_norm": 0.376953125,
|
| 1368 |
"learning_rate": 3.780487804878049e-05,
|
| 1369 |
+
"loss": 0.03998865559697151,
|
| 1370 |
+
"mean_token_accuracy": 0.9876968078315258,
|
| 1371 |
+
"num_tokens": 906633.0,
|
| 1372 |
"step": 134
|
| 1373 |
},
|
| 1374 |
{
|
| 1375 |
+
"entropy": 0.1975369704887271,
|
| 1376 |
"epoch": 1.649310872894334,
|
| 1377 |
+
"grad_norm": 0.3203125,
|
| 1378 |
"learning_rate": 3.6585365853658535e-05,
|
| 1379 |
+
"loss": 0.031131668016314507,
|
| 1380 |
+
"mean_token_accuracy": 0.9915927015244961,
|
| 1381 |
+
"num_tokens": 913990.0,
|
| 1382 |
"step": 135
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
+
"entropy": 0.23459685500711203,
|
| 1386 |
"epoch": 1.66156202143951,
|
| 1387 |
+
"grad_norm": 0.76171875,
|
| 1388 |
"learning_rate": 3.5365853658536584e-05,
|
| 1389 |
+
"loss": 0.03373259678483009,
|
| 1390 |
+
"mean_token_accuracy": 0.9898596629500389,
|
| 1391 |
+
"num_tokens": 919248.0,
|
| 1392 |
"step": 136
|
| 1393 |
},
|
| 1394 |
{
|
| 1395 |
+
"entropy": 0.1909911371767521,
|
| 1396 |
"epoch": 1.673813169984686,
|
| 1397 |
+
"grad_norm": 0.60546875,
|
| 1398 |
"learning_rate": 3.414634146341464e-05,
|
| 1399 |
+
"loss": 0.037791188806295395,
|
| 1400 |
+
"mean_token_accuracy": 0.9897548258304596,
|
| 1401 |
+
"num_tokens": 926248.0,
|
| 1402 |
"step": 137
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
+
"entropy": 0.2332595670595765,
|
| 1406 |
"epoch": 1.686064318529862,
|
| 1407 |
+
"grad_norm": 0.89453125,
|
| 1408 |
"learning_rate": 3.292682926829269e-05,
|
| 1409 |
+
"loss": 0.03799242898821831,
|
| 1410 |
+
"mean_token_accuracy": 0.9867184162139893,
|
| 1411 |
+
"num_tokens": 932490.0,
|
| 1412 |
"step": 138
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
+
"entropy": 0.22243124432861805,
|
| 1416 |
"epoch": 1.6983154670750382,
|
| 1417 |
+
"grad_norm": 0.61328125,
|
| 1418 |
"learning_rate": 3.170731707317073e-05,
|
| 1419 |
+
"loss": 0.04291514679789543,
|
| 1420 |
+
"mean_token_accuracy": 0.9877815246582031,
|
| 1421 |
+
"num_tokens": 938756.0,
|
| 1422 |
"step": 139
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
+
"entropy": 0.20778016652911901,
|
| 1426 |
"epoch": 1.7105666156202144,
|
| 1427 |
+
"grad_norm": 0.41796875,
|
| 1428 |
"learning_rate": 3.048780487804878e-05,
|
| 1429 |
+
"loss": 0.023588458076119423,
|
| 1430 |
+
"mean_token_accuracy": 0.9942950084805489,
|
| 1431 |
+
"num_tokens": 945866.0,
|
| 1432 |
"step": 140
|
| 1433 |
},
|
| 1434 |
{
|
| 1435 |
+
"entropy": 0.18776059616357088,
|
| 1436 |
"epoch": 1.7228177641653906,
|
| 1437 |
+
"grad_norm": 0.41796875,
|
| 1438 |
"learning_rate": 2.926829268292683e-05,
|
| 1439 |
+
"loss": 0.03229852020740509,
|
| 1440 |
+
"mean_token_accuracy": 0.9909596405923367,
|
| 1441 |
+
"num_tokens": 952865.0,
|
| 1442 |
"step": 141
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
+
"entropy": 0.18707702960819006,
|
| 1446 |
"epoch": 1.7350689127105667,
|
| 1447 |
+
"grad_norm": 0.609375,
|
| 1448 |
"learning_rate": 2.8048780487804882e-05,
|
| 1449 |
+
"loss": 0.03691868111491203,
|
| 1450 |
+
"mean_token_accuracy": 0.9900590926408768,
|
| 1451 |
+
"num_tokens": 959190.0,
|
| 1452 |
"step": 142
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
+
"entropy": 0.1914756903424859,
|
| 1456 |
"epoch": 1.7473200612557427,
|
| 1457 |
+
"grad_norm": 0.408203125,
|
| 1458 |
"learning_rate": 2.682926829268293e-05,
|
| 1459 |
+
"loss": 0.03487441688776016,
|
| 1460 |
+
"mean_token_accuracy": 0.9909356310963631,
|
| 1461 |
+
"num_tokens": 966059.0,
|
| 1462 |
"step": 143
|
| 1463 |
},
|
| 1464 |
{
|
| 1465 |
+
"entropy": 0.20852853963151574,
|
| 1466 |
"epoch": 1.7595712098009189,
|
| 1467 |
+
"grad_norm": 0.380859375,
|
| 1468 |
"learning_rate": 2.5609756097560977e-05,
|
| 1469 |
+
"loss": 0.03023841790854931,
|
| 1470 |
+
"mean_token_accuracy": 0.9922478385269642,
|
| 1471 |
+
"num_tokens": 973553.0,
|
| 1472 |
"step": 144
|
| 1473 |
},
|
| 1474 |
{
|
| 1475 |
+
"entropy": 0.18278094567358494,
|
| 1476 |
"epoch": 1.7718223583460948,
|
| 1477 |
+
"grad_norm": 0.65625,
|
| 1478 |
"learning_rate": 2.4390243902439026e-05,
|
| 1479 |
+
"loss": 0.03335383161902428,
|
| 1480 |
+
"mean_token_accuracy": 0.9902437664568424,
|
| 1481 |
+
"num_tokens": 980748.0,
|
| 1482 |
"step": 145
|
| 1483 |
},
|
| 1484 |
{
|
| 1485 |
+
"entropy": 0.2156418706290424,
|
| 1486 |
"epoch": 1.784073506891271,
|
| 1487 |
+
"grad_norm": 0.58984375,
|
| 1488 |
"learning_rate": 2.3170731707317075e-05,
|
| 1489 |
+
"loss": 0.026211977005004883,
|
| 1490 |
+
"mean_token_accuracy": 0.9913386814296246,
|
| 1491 |
+
"num_tokens": 987018.0,
|
| 1492 |
"step": 146
|
| 1493 |
},
|
| 1494 |
{
|
| 1495 |
+
"entropy": 0.2084086169488728,
|
| 1496 |
"epoch": 1.7963246554364471,
|
| 1497 |
+
"grad_norm": 0.37890625,
|
| 1498 |
"learning_rate": 2.1951219512195124e-05,
|
| 1499 |
+
"loss": 0.029074503108859062,
|
| 1500 |
+
"mean_token_accuracy": 0.9920879267156124,
|
| 1501 |
+
"num_tokens": 993841.0,
|
| 1502 |
"step": 147
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
+
"entropy": 0.2162067350000143,
|
| 1506 |
"epoch": 1.8085758039816233,
|
| 1507 |
+
"grad_norm": 0.38671875,
|
| 1508 |
"learning_rate": 2.073170731707317e-05,
|
| 1509 |
+
"loss": 0.027591165155172348,
|
| 1510 |
+
"mean_token_accuracy": 0.9916894063353539,
|
| 1511 |
+
"num_tokens": 1000318.0,
|
| 1512 |
"step": 148
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
+
"entropy": 0.22895692195743322,
|
| 1516 |
"epoch": 1.8208269525267995,
|
| 1517 |
+
"grad_norm": 1.421875,
|
| 1518 |
"learning_rate": 1.9512195121951222e-05,
|
| 1519 |
+
"loss": 0.034101299941539764,
|
| 1520 |
+
"mean_token_accuracy": 0.9889856353402138,
|
| 1521 |
+
"num_tokens": 1005747.0,
|
| 1522 |
"step": 149
|
| 1523 |
},
|
| 1524 |
{
|
| 1525 |
+
"entropy": 0.21029841899871826,
|
| 1526 |
"epoch": 1.8330781010719757,
|
| 1527 |
+
"grad_norm": 0.59375,
|
| 1528 |
"learning_rate": 1.8292682926829268e-05,
|
| 1529 |
+
"loss": 0.04408642649650574,
|
| 1530 |
+
"mean_token_accuracy": 0.988445583730936,
|
| 1531 |
+
"num_tokens": 1013365.0,
|
| 1532 |
"step": 150
|
| 1533 |
},
|
| 1534 |
{
|
| 1535 |
"epoch": 1.8330781010719757,
|
| 1536 |
+
"eval_entropy": 0.21028992522885834,
|
| 1537 |
+
"eval_loss": 0.06481878459453583,
|
| 1538 |
+
"eval_mean_token_accuracy": 0.9753203677094501,
|
| 1539 |
+
"eval_num_tokens": 1013365.0,
|
| 1540 |
+
"eval_runtime": 64.1224,
|
| 1541 |
+
"eval_samples_per_second": 1.076,
|
| 1542 |
+
"eval_steps_per_second": 1.076,
|
| 1543 |
"step": 150
|
| 1544 |
},
|
| 1545 |
{
|
| 1546 |
+
"entropy": 0.23085341975092888,
|
| 1547 |
"epoch": 1.8453292496171516,
|
| 1548 |
+
"grad_norm": 0.375,
|
| 1549 |
"learning_rate": 1.707317073170732e-05,
|
| 1550 |
+
"loss": 0.03518415987491608,
|
| 1551 |
+
"mean_token_accuracy": 0.9906566366553307,
|
| 1552 |
+
"num_tokens": 1018758.0,
|
| 1553 |
"step": 151
|
| 1554 |
},
|
| 1555 |
{
|
| 1556 |
+
"entropy": 0.21574621414765716,
|
| 1557 |
"epoch": 1.8575803981623276,
|
| 1558 |
+
"grad_norm": 0.51953125,
|
| 1559 |
"learning_rate": 1.5853658536585366e-05,
|
| 1560 |
+
"loss": 0.028541577979922295,
|
| 1561 |
+
"mean_token_accuracy": 0.9907862320542336,
|
| 1562 |
+
"num_tokens": 1026224.0,
|
| 1563 |
"step": 152
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
+
"entropy": 0.18795023765414953,
|
| 1567 |
"epoch": 1.8698315467075037,
|
| 1568 |
+
"grad_norm": 0.5234375,
|
| 1569 |
"learning_rate": 1.4634146341463415e-05,
|
| 1570 |
+
"loss": 0.02955229952931404,
|
| 1571 |
+
"mean_token_accuracy": 0.9923531115055084,
|
| 1572 |
+
"num_tokens": 1033949.0,
|
| 1573 |
"step": 153
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
+
"entropy": 0.21928654983639717,
|
| 1577 |
"epoch": 1.88208269525268,
|
| 1578 |
+
"grad_norm": 0.2421875,
|
| 1579 |
"learning_rate": 1.3414634146341466e-05,
|
| 1580 |
+
"loss": 0.02629980631172657,
|
| 1581 |
+
"mean_token_accuracy": 0.9936561770737171,
|
| 1582 |
+
"num_tokens": 1041050.0,
|
| 1583 |
"step": 154
|
| 1584 |
},
|
| 1585 |
{
|
| 1586 |
+
"entropy": 0.22051549516618252,
|
| 1587 |
"epoch": 1.894333843797856,
|
| 1588 |
+
"grad_norm": 0.419921875,
|
| 1589 |
"learning_rate": 1.2195121951219513e-05,
|
| 1590 |
+
"loss": 0.03377772495150566,
|
| 1591 |
+
"mean_token_accuracy": 0.9897954650223255,
|
| 1592 |
+
"num_tokens": 1047621.0,
|
| 1593 |
"step": 155
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
+
"entropy": 0.220434432849288,
|
| 1597 |
"epoch": 1.9065849923430322,
|
| 1598 |
+
"grad_norm": 0.640625,
|
| 1599 |
"learning_rate": 1.0975609756097562e-05,
|
| 1600 |
+
"loss": 0.03985638543963432,
|
| 1601 |
+
"mean_token_accuracy": 0.9877757839858532,
|
| 1602 |
+
"num_tokens": 1053963.0,
|
| 1603 |
"step": 156
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
+
"entropy": 0.20754134468734264,
|
| 1607 |
"epoch": 1.9188361408882084,
|
| 1608 |
+
"grad_norm": 0.34765625,
|
| 1609 |
"learning_rate": 9.756097560975611e-06,
|
| 1610 |
+
"loss": 0.03090263158082962,
|
| 1611 |
+
"mean_token_accuracy": 0.9914594441652298,
|
| 1612 |
+
"num_tokens": 1061378.0,
|
| 1613 |
"step": 157
|
| 1614 |
},
|
| 1615 |
{
|
| 1616 |
+
"entropy": 0.21179267205297947,
|
| 1617 |
"epoch": 1.9310872894333844,
|
| 1618 |
+
"grad_norm": 0.5625,
|
| 1619 |
"learning_rate": 8.53658536585366e-06,
|
| 1620 |
+
"loss": 0.028556756675243378,
|
| 1621 |
+
"mean_token_accuracy": 0.9921947717666626,
|
| 1622 |
+
"num_tokens": 1067649.0,
|
| 1623 |
"step": 158
|
| 1624 |
},
|
| 1625 |
{
|
| 1626 |
+
"entropy": 0.20413818582892418,
|
| 1627 |
"epoch": 1.9433384379785605,
|
| 1628 |
+
"grad_norm": 0.54296875,
|
| 1629 |
"learning_rate": 7.317073170731707e-06,
|
| 1630 |
+
"loss": 0.029063822701573372,
|
| 1631 |
+
"mean_token_accuracy": 0.9915811195969582,
|
| 1632 |
+
"num_tokens": 1073729.0,
|
| 1633 |
"step": 159
|
| 1634 |
},
|
| 1635 |
{
|
| 1636 |
+
"entropy": 0.19627654599025846,
|
| 1637 |
"epoch": 1.9555895865237365,
|
| 1638 |
+
"grad_norm": 0.3125,
|
| 1639 |
"learning_rate": 6.0975609756097564e-06,
|
| 1640 |
+
"loss": 0.03230921924114227,
|
| 1641 |
+
"mean_token_accuracy": 0.9897297285497189,
|
| 1642 |
+
"num_tokens": 1080422.0,
|
| 1643 |
"step": 160
|
| 1644 |
},
|
| 1645 |
{
|
| 1646 |
+
"entropy": 0.20519507862627506,
|
| 1647 |
"epoch": 1.9678407350689127,
|
| 1648 |
+
"grad_norm": 0.306640625,
|
| 1649 |
"learning_rate": 4.8780487804878055e-06,
|
| 1650 |
+
"loss": 0.026061467826366425,
|
| 1651 |
+
"mean_token_accuracy": 0.9930403046309948,
|
| 1652 |
+
"num_tokens": 1087517.0,
|
| 1653 |
"step": 161
|
| 1654 |
},
|
| 1655 |
{
|
| 1656 |
+
"entropy": 0.21147159207612276,
|
| 1657 |
"epoch": 1.9800918836140888,
|
| 1658 |
+
"grad_norm": 0.458984375,
|
| 1659 |
"learning_rate": 3.6585365853658537e-06,
|
| 1660 |
+
"loss": 0.03128533065319061,
|
| 1661 |
+
"mean_token_accuracy": 0.9902758039534092,
|
| 1662 |
+
"num_tokens": 1093989.0,
|
| 1663 |
"step": 162
|
| 1664 |
},
|
| 1665 |
{
|
| 1666 |
+
"entropy": 0.2088263975456357,
|
| 1667 |
"epoch": 1.992343032159265,
|
| 1668 |
+
"grad_norm": 0.5390625,
|
| 1669 |
"learning_rate": 2.4390243902439027e-06,
|
| 1670 |
+
"loss": 0.047114044427871704,
|
| 1671 |
+
"mean_token_accuracy": 0.9882928691804409,
|
| 1672 |
+
"num_tokens": 1101351.0,
|
| 1673 |
"step": 163
|
| 1674 |
},
|
| 1675 |
{
|
| 1676 |
+
"entropy": 0.2400845021009445,
|
| 1677 |
"epoch": 2.0,
|
| 1678 |
+
"grad_norm": 0.61328125,
|
| 1679 |
"learning_rate": 1.2195121951219514e-06,
|
| 1680 |
+
"loss": 0.024358952417969704,
|
| 1681 |
+
"mean_token_accuracy": 0.9906890511512756,
|
| 1682 |
+
"num_tokens": 1105216.0,
|
| 1683 |
"step": 164
|
| 1684 |
}
|
| 1685 |
],
|
|
|
|
| 1700 |
"attributes": {}
|
| 1701 |
}
|
| 1702 |
},
|
| 1703 |
+
"total_flos": 5.004551367111475e+16,
|
| 1704 |
"train_batch_size": 1,
|
| 1705 |
"trial_name": null,
|
| 1706 |
"trial_params": null
|