Upload folder using huggingface_hub
Browse files- all_results.json +5 -5
- model-00001-of-00007.safetensors +1 -1
- model-00002-of-00007.safetensors +1 -1
- model-00003-of-00007.safetensors +1 -1
- model-00004-of-00007.safetensors +1 -1
- model-00005-of-00007.safetensors +1 -1
- model-00006-of-00007.safetensors +1 -1
- model-00007-of-00007.safetensors +1 -1
- runs/Feb25_15-17-17_pc3-1FECDDA/events.out.tfevents.1740493039.pc3-1FECDDA.51860.0 +3 -0
- runs/Feb25_15-17-17_pc3-1FECDDA/events.out.tfevents.1740499439.pc3-1FECDDA.51860.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +306 -306
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.6134969325153374,
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
+
"train_loss": 1.7183861404657363,
|
| 5 |
+
"train_runtime": 6171.3384,
|
| 6 |
+
"train_samples_per_second": 0.016,
|
| 7 |
+
"train_steps_per_second": 0.016
|
| 8 |
}
|
model-00001-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4886498936
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54f8d2f34e5d9b2e70a5906cdf814d6a09295ca229896c2270bd61bd8907f750
|
| 3 |
size 4886498936
|
model-00002-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4832007448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3be61331596bfd8573fec430b264a4b7669a2a582283df1243c0e525bacfdd58
|
| 3 |
size 4832007448
|
model-00003-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999813112
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce060f810fd79516ccdbe8223c38ac915921afe050eb5b4b7d18c34f9aef389d
|
| 3 |
size 4999813112
|
model-00004-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999813128
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b12509d15b2de31c96e6dd57fa9b0c98029e5aeffb5511d99c0a6635813f7927
|
| 3 |
size 4999813128
|
model-00005-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4832007496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c39e5e56ea5b8c17fe8ef96510c14392ae8cb5eb2d10b1b658fc77f8174ecfb7
|
| 3 |
size 4832007496
|
model-00006-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999813120
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca737a761c3135ce22c2a3d55fa9bcdaaa69ca035234e18112e59c1063dda438
|
| 3 |
size 4999813120
|
model-00007-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2571190952
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7b7fc2ec856d887761446b2921c064c0b13990342385b1124a4a64617e401b9
|
| 3 |
size 2571190952
|
runs/Feb25_15-17-17_pc3-1FECDDA/events.out.tfevents.1740493039.pc3-1FECDDA.51860.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf3651646ec14f3bf5234a318ca1535a5eaf09ceddf1ca178ea85286c0ba9cb1
|
| 3 |
+
size 26323
|
runs/Feb25_15-17-17_pc3-1FECDDA/events.out.tfevents.1740499439.pc3-1FECDDA.51860.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09b5b959b3c30a6c26f37407e2ecef8ac558f28e674cc2e875de2c00be599c2e
|
| 3 |
+
size 354
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.6134969325153374,
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
+
"train_loss": 1.7183861404657363,
|
| 5 |
+
"train_runtime": 6171.3384,
|
| 6 |
+
"train_samples_per_second": 0.016,
|
| 7 |
+
"train_steps_per_second": 0.016
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,713 +9,713 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 5e-06,
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 1e-05,
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 9.89795918367347e-06,
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 9.795918367346939e-06,
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 9.693877551020408e-06,
|
| 43 |
-
"loss":
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 9.591836734693878e-06,
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 9.489795918367348e-06,
|
| 57 |
-
"loss":
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 9.387755102040818e-06,
|
| 64 |
-
"loss":
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 9.285714285714288e-06,
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 9.183673469387756e-06,
|
| 78 |
-
"loss":
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
"learning_rate": 9.081632653061225e-06,
|
| 85 |
-
"loss":
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
"learning_rate": 8.979591836734695e-06,
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
"learning_rate": 8.877551020408163e-06,
|
| 99 |
-
"loss":
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 8.775510204081633e-06,
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
"learning_rate": 8.673469387755103e-06,
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
"learning_rate": 8.571428571428571e-06,
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
"learning_rate": 8.469387755102042e-06,
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
"learning_rate": 8.36734693877551e-06,
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
"learning_rate": 8.26530612244898e-06,
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
"learning_rate": 8.16326530612245e-06,
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
"learning_rate": 8.06122448979592e-06,
|
| 155 |
-
"loss": 1.
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
"learning_rate": 7.959183673469388e-06,
|
| 162 |
-
"loss": 1.
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
"learning_rate": 7.857142857142858e-06,
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm":
|
| 175 |
"learning_rate": 7.755102040816327e-06,
|
| 176 |
-
"loss": 2.
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
"learning_rate": 7.653061224489796e-06,
|
| 183 |
-
"loss":
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
"learning_rate": 7.551020408163265e-06,
|
| 190 |
-
"loss":
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm":
|
| 196 |
"learning_rate": 7.448979591836736e-06,
|
| 197 |
-
"loss":
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm":
|
| 203 |
"learning_rate": 7.346938775510205e-06,
|
| 204 |
-
"loss": 1.
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
"learning_rate": 7.244897959183675e-06,
|
| 211 |
-
"loss":
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm":
|
| 217 |
"learning_rate": 7.1428571428571436e-06,
|
| 218 |
-
"loss":
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
"learning_rate": 7.0408163265306125e-06,
|
| 225 |
-
"loss":
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
"learning_rate": 6.938775510204082e-06,
|
| 232 |
-
"loss":
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
"learning_rate": 6.836734693877551e-06,
|
| 239 |
-
"loss": 1.
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
"learning_rate": 6.734693877551021e-06,
|
| 246 |
-
"loss": 1.
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm":
|
| 252 |
"learning_rate": 6.63265306122449e-06,
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
"learning_rate": 6.530612244897959e-06,
|
| 260 |
-
"loss":
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
"learning_rate": 6.4285714285714295e-06,
|
| 267 |
-
"loss": 1.
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
"learning_rate": 6.326530612244899e-06,
|
| 274 |
-
"loss":
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm":
|
| 280 |
"learning_rate": 6.224489795918368e-06,
|
| 281 |
-
"loss":
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
"learning_rate": 6.122448979591837e-06,
|
| 288 |
-
"loss":
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm":
|
| 294 |
"learning_rate": 6.020408163265307e-06,
|
| 295 |
-
"loss": 1.
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm":
|
| 301 |
"learning_rate": 5.918367346938776e-06,
|
| 302 |
-
"loss":
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm": 19.
|
| 308 |
"learning_rate": 5.816326530612246e-06,
|
| 309 |
-
"loss": 1.
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm":
|
| 315 |
"learning_rate": 5.7142857142857145e-06,
|
| 316 |
-
"loss": 2.
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm":
|
| 322 |
"learning_rate": 5.6122448979591834e-06,
|
| 323 |
-
"loss":
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm":
|
| 329 |
"learning_rate": 5.510204081632653e-06,
|
| 330 |
-
"loss": 2.
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm":
|
| 336 |
"learning_rate": 5.408163265306123e-06,
|
| 337 |
-
"loss":
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm":
|
| 343 |
"learning_rate": 5.306122448979593e-06,
|
| 344 |
-
"loss": 1.
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm":
|
| 350 |
"learning_rate": 5.204081632653062e-06,
|
| 351 |
-
"loss":
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm":
|
| 357 |
"learning_rate": 5.1020408163265315e-06,
|
| 358 |
-
"loss":
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm":
|
| 364 |
"learning_rate": 5e-06,
|
| 365 |
-
"loss": 1.
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm":
|
| 371 |
"learning_rate": 4.897959183673469e-06,
|
| 372 |
-
"loss":
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm":
|
| 378 |
"learning_rate": 4.795918367346939e-06,
|
| 379 |
-
"loss":
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm":
|
| 385 |
"learning_rate": 4.693877551020409e-06,
|
| 386 |
-
"loss": 1.
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm":
|
| 392 |
"learning_rate": 4.591836734693878e-06,
|
| 393 |
-
"loss": 1.
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm": 29.
|
| 399 |
"learning_rate": 4.489795918367348e-06,
|
| 400 |
-
"loss": 1.
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm":
|
| 406 |
"learning_rate": 4.3877551020408165e-06,
|
| 407 |
-
"loss": 1.
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm":
|
| 413 |
"learning_rate": 4.2857142857142855e-06,
|
| 414 |
-
"loss": 1.
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm":
|
| 420 |
"learning_rate": 4.183673469387755e-06,
|
| 421 |
-
"loss": 1.
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm":
|
| 427 |
"learning_rate": 4.081632653061225e-06,
|
| 428 |
-
"loss": 1.
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
"learning_rate": 3.979591836734694e-06,
|
| 435 |
-
"loss": 1.
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm":
|
| 441 |
"learning_rate": 3.877551020408164e-06,
|
| 442 |
-
"loss": 1.
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm":
|
| 448 |
"learning_rate": 3.7755102040816327e-06,
|
| 449 |
-
"loss": 1.
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm":
|
| 455 |
"learning_rate": 3.6734693877551024e-06,
|
| 456 |
-
"loss": 1.
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm":
|
| 462 |
"learning_rate": 3.5714285714285718e-06,
|
| 463 |
-
"loss":
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm":
|
| 469 |
"learning_rate": 3.469387755102041e-06,
|
| 470 |
-
"loss": 1.
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm":
|
| 476 |
"learning_rate": 3.3673469387755105e-06,
|
| 477 |
-
"loss": 1.
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm":
|
| 483 |
"learning_rate": 3.2653061224489794e-06,
|
| 484 |
-
"loss":
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm": 27.
|
| 490 |
"learning_rate": 3.1632653061224496e-06,
|
| 491 |
-
"loss": 1.
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm":
|
| 497 |
"learning_rate": 3.0612244897959185e-06,
|
| 498 |
-
"loss": 1.
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm":
|
| 504 |
"learning_rate": 2.959183673469388e-06,
|
| 505 |
-
"loss": 1.
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm":
|
| 511 |
"learning_rate": 2.8571428571428573e-06,
|
| 512 |
-
"loss": 1.
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm":
|
| 518 |
"learning_rate": 2.7551020408163266e-06,
|
| 519 |
-
"loss":
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm":
|
| 525 |
"learning_rate": 2.6530612244897964e-06,
|
| 526 |
-
"loss": 1.
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm":
|
| 532 |
"learning_rate": 2.5510204081632657e-06,
|
| 533 |
-
"loss": 1.
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm":
|
| 539 |
"learning_rate": 2.4489795918367347e-06,
|
| 540 |
-
"loss":
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm":
|
| 546 |
"learning_rate": 2.3469387755102044e-06,
|
| 547 |
-
"loss": 1.
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm":
|
| 553 |
"learning_rate": 2.244897959183674e-06,
|
| 554 |
-
"loss": 1.
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"grad_norm":
|
| 560 |
"learning_rate": 2.1428571428571427e-06,
|
| 561 |
-
"loss": 1.
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"grad_norm":
|
| 567 |
"learning_rate": 2.0408163265306125e-06,
|
| 568 |
-
"loss":
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"grad_norm":
|
| 574 |
"learning_rate": 1.938775510204082e-06,
|
| 575 |
-
"loss":
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 0.
|
| 580 |
-
"grad_norm":
|
| 581 |
"learning_rate": 1.8367346938775512e-06,
|
| 582 |
-
"loss": 1.
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"grad_norm":
|
| 588 |
"learning_rate": 1.7346938775510206e-06,
|
| 589 |
-
"loss": 1.
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"grad_norm":
|
| 595 |
"learning_rate": 1.6326530612244897e-06,
|
| 596 |
-
"loss": 1.
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 0.
|
| 601 |
-
"grad_norm":
|
| 602 |
"learning_rate": 1.5306122448979593e-06,
|
| 603 |
-
"loss": 1.
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 0.
|
| 608 |
-
"grad_norm": 21.
|
| 609 |
"learning_rate": 1.4285714285714286e-06,
|
| 610 |
-
"loss":
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 0.
|
| 615 |
-
"grad_norm": 23.
|
| 616 |
"learning_rate": 1.3265306122448982e-06,
|
| 617 |
-
"loss": 1.
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 0.
|
| 622 |
-
"grad_norm":
|
| 623 |
"learning_rate": 1.2244897959183673e-06,
|
| 624 |
-
"loss": 1.
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 0.
|
| 629 |
-
"grad_norm":
|
| 630 |
"learning_rate": 1.122448979591837e-06,
|
| 631 |
-
"loss": 1.
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"grad_norm":
|
| 637 |
"learning_rate": 1.0204081632653063e-06,
|
| 638 |
-
"loss": 1.
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"grad_norm": 25.
|
| 644 |
"learning_rate": 9.183673469387756e-07,
|
| 645 |
-
"loss": 1.
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 0.
|
| 650 |
-
"grad_norm":
|
| 651 |
"learning_rate": 8.163265306122449e-07,
|
| 652 |
-
"loss": 1.
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 0.
|
| 657 |
-
"grad_norm":
|
| 658 |
"learning_rate": 7.142857142857143e-07,
|
| 659 |
-
"loss": 1.
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 0.
|
| 664 |
-
"grad_norm":
|
| 665 |
"learning_rate": 6.122448979591837e-07,
|
| 666 |
-
"loss":
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 0.
|
| 671 |
-
"grad_norm":
|
| 672 |
"learning_rate": 5.102040816326531e-07,
|
| 673 |
-
"loss":
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"grad_norm":
|
| 679 |
"learning_rate": 4.0816326530612243e-07,
|
| 680 |
-
"loss": 1.
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 0.
|
| 685 |
-
"grad_norm":
|
| 686 |
"learning_rate": 3.0612244897959183e-07,
|
| 687 |
-
"loss": 1.
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 0.
|
| 692 |
-
"grad_norm":
|
| 693 |
"learning_rate": 2.0408163265306121e-07,
|
| 694 |
-
"loss":
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 0.
|
| 699 |
-
"grad_norm":
|
| 700 |
"learning_rate": 1.0204081632653061e-07,
|
| 701 |
-
"loss":
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 0.
|
| 706 |
-
"grad_norm":
|
| 707 |
"learning_rate": 0.0,
|
| 708 |
-
"loss":
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 0.
|
| 713 |
"step": 100,
|
| 714 |
"total_flos": 2305515375820800.0,
|
| 715 |
-
"train_loss": 1.
|
| 716 |
-
"train_runtime":
|
| 717 |
-
"train_samples_per_second": 0.
|
| 718 |
-
"train_steps_per_second": 0.
|
| 719 |
}
|
| 720 |
],
|
| 721 |
"logging_steps": 1,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.6134969325153374,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.006134969325153374,
|
| 13 |
+
"grad_norm": 88.35982513427734,
|
| 14 |
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 2.5293,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.012269938650306749,
|
| 20 |
+
"grad_norm": 71.40210723876953,
|
| 21 |
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 2.6329,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.018404907975460124,
|
| 27 |
+
"grad_norm": 82.39716339111328,
|
| 28 |
"learning_rate": 9.89795918367347e-06,
|
| 29 |
+
"loss": 2.7434,
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.024539877300613498,
|
| 34 |
+
"grad_norm": 35.27679443359375,
|
| 35 |
"learning_rate": 9.795918367346939e-06,
|
| 36 |
+
"loss": 1.7051,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.03067484662576687,
|
| 41 |
+
"grad_norm": 44.54578399658203,
|
| 42 |
"learning_rate": 9.693877551020408e-06,
|
| 43 |
+
"loss": 1.8552,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.03680981595092025,
|
| 48 |
+
"grad_norm": 66.71598052978516,
|
| 49 |
"learning_rate": 9.591836734693878e-06,
|
| 50 |
+
"loss": 2.3346,
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.04294478527607362,
|
| 55 |
+
"grad_norm": 59.80192184448242,
|
| 56 |
"learning_rate": 9.489795918367348e-06,
|
| 57 |
+
"loss": 2.1256,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.049079754601226995,
|
| 62 |
+
"grad_norm": 48.737754821777344,
|
| 63 |
"learning_rate": 9.387755102040818e-06,
|
| 64 |
+
"loss": 1.7364,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.05521472392638037,
|
| 69 |
+
"grad_norm": 28.469051361083984,
|
| 70 |
"learning_rate": 9.285714285714288e-06,
|
| 71 |
+
"loss": 1.9437,
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.06134969325153374,
|
| 76 |
+
"grad_norm": 54.786128997802734,
|
| 77 |
"learning_rate": 9.183673469387756e-06,
|
| 78 |
+
"loss": 3.0007,
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.06748466257668712,
|
| 83 |
+
"grad_norm": 64.78273010253906,
|
| 84 |
"learning_rate": 9.081632653061225e-06,
|
| 85 |
+
"loss": 1.8778,
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.0736196319018405,
|
| 90 |
+
"grad_norm": 50.23028564453125,
|
| 91 |
"learning_rate": 8.979591836734695e-06,
|
| 92 |
+
"loss": 1.8922,
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.07975460122699386,
|
| 97 |
+
"grad_norm": 64.22184753417969,
|
| 98 |
"learning_rate": 8.877551020408163e-06,
|
| 99 |
+
"loss": 2.0061,
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.08588957055214724,
|
| 104 |
+
"grad_norm": 30.814348220825195,
|
| 105 |
"learning_rate": 8.775510204081633e-06,
|
| 106 |
+
"loss": 1.6799,
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.09202453987730061,
|
| 111 |
+
"grad_norm": 33.7660026550293,
|
| 112 |
"learning_rate": 8.673469387755103e-06,
|
| 113 |
+
"loss": 1.8202,
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.09815950920245399,
|
| 118 |
+
"grad_norm": 19.061237335205078,
|
| 119 |
"learning_rate": 8.571428571428571e-06,
|
| 120 |
+
"loss": 1.5798,
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.10429447852760736,
|
| 125 |
+
"grad_norm": 34.668060302734375,
|
| 126 |
"learning_rate": 8.469387755102042e-06,
|
| 127 |
+
"loss": 1.5905,
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.11042944785276074,
|
| 132 |
+
"grad_norm": 32.6378059387207,
|
| 133 |
"learning_rate": 8.36734693877551e-06,
|
| 134 |
+
"loss": 1.6961,
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.1165644171779141,
|
| 139 |
+
"grad_norm": 37.90195083618164,
|
| 140 |
"learning_rate": 8.26530612244898e-06,
|
| 141 |
+
"loss": 1.7675,
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.12269938650306748,
|
| 146 |
+
"grad_norm": 50.06489562988281,
|
| 147 |
"learning_rate": 8.16326530612245e-06,
|
| 148 |
+
"loss": 1.4142,
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.12883435582822086,
|
| 153 |
+
"grad_norm": 50.70490264892578,
|
| 154 |
"learning_rate": 8.06122448979592e-06,
|
| 155 |
+
"loss": 1.4977,
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.13496932515337423,
|
| 160 |
+
"grad_norm": 32.71562576293945,
|
| 161 |
"learning_rate": 7.959183673469388e-06,
|
| 162 |
+
"loss": 1.3493,
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.1411042944785276,
|
| 167 |
+
"grad_norm": 36.24839782714844,
|
| 168 |
"learning_rate": 7.857142857142858e-06,
|
| 169 |
+
"loss": 1.5274,
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.147239263803681,
|
| 174 |
+
"grad_norm": 40.0679817199707,
|
| 175 |
"learning_rate": 7.755102040816327e-06,
|
| 176 |
+
"loss": 2.0856,
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.15337423312883436,
|
| 181 |
+
"grad_norm": 64.05142211914062,
|
| 182 |
"learning_rate": 7.653061224489796e-06,
|
| 183 |
+
"loss": 2.0254,
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.15950920245398773,
|
| 188 |
+
"grad_norm": 35.79407501220703,
|
| 189 |
"learning_rate": 7.551020408163265e-06,
|
| 190 |
+
"loss": 1.8358,
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.1656441717791411,
|
| 195 |
+
"grad_norm": 40.65586471557617,
|
| 196 |
"learning_rate": 7.448979591836736e-06,
|
| 197 |
+
"loss": 2.43,
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.17177914110429449,
|
| 202 |
+
"grad_norm": 45.51654052734375,
|
| 203 |
"learning_rate": 7.346938775510205e-06,
|
| 204 |
+
"loss": 1.5213,
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.17791411042944785,
|
| 209 |
+
"grad_norm": 18.54694175720215,
|
| 210 |
"learning_rate": 7.244897959183675e-06,
|
| 211 |
+
"loss": 1.7857,
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.18404907975460122,
|
| 216 |
+
"grad_norm": 14.598164558410645,
|
| 217 |
"learning_rate": 7.1428571428571436e-06,
|
| 218 |
+
"loss": 0.6952,
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.1901840490797546,
|
| 223 |
+
"grad_norm": 32.71424865722656,
|
| 224 |
"learning_rate": 7.0408163265306125e-06,
|
| 225 |
+
"loss": 2.0721,
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.19631901840490798,
|
| 230 |
+
"grad_norm": 32.908966064453125,
|
| 231 |
"learning_rate": 6.938775510204082e-06,
|
| 232 |
+
"loss": 2.0366,
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.20245398773006135,
|
| 237 |
+
"grad_norm": 20.541730880737305,
|
| 238 |
"learning_rate": 6.836734693877551e-06,
|
| 239 |
+
"loss": 1.9455,
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.2085889570552147,
|
| 244 |
+
"grad_norm": 33.4826545715332,
|
| 245 |
"learning_rate": 6.734693877551021e-06,
|
| 246 |
+
"loss": 1.875,
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.2147239263803681,
|
| 251 |
+
"grad_norm": 27.954130172729492,
|
| 252 |
"learning_rate": 6.63265306122449e-06,
|
| 253 |
+
"loss": 1.5669,
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.22085889570552147,
|
| 258 |
+
"grad_norm": 28.17574119567871,
|
| 259 |
"learning_rate": 6.530612244897959e-06,
|
| 260 |
+
"loss": 0.7674,
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.22699386503067484,
|
| 265 |
+
"grad_norm": 27.853933334350586,
|
| 266 |
"learning_rate": 6.4285714285714295e-06,
|
| 267 |
+
"loss": 1.5963,
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.2331288343558282,
|
| 272 |
+
"grad_norm": 19.903644561767578,
|
| 273 |
"learning_rate": 6.326530612244899e-06,
|
| 274 |
+
"loss": 0.5877,
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.2392638036809816,
|
| 279 |
+
"grad_norm": 35.6341552734375,
|
| 280 |
"learning_rate": 6.224489795918368e-06,
|
| 281 |
+
"loss": 2.0134,
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.24539877300613497,
|
| 286 |
+
"grad_norm": 35.47877502441406,
|
| 287 |
"learning_rate": 6.122448979591837e-06,
|
| 288 |
+
"loss": 1.9894,
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.25153374233128833,
|
| 293 |
+
"grad_norm": 28.387544631958008,
|
| 294 |
"learning_rate": 6.020408163265307e-06,
|
| 295 |
+
"loss": 1.5612,
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.25766871165644173,
|
| 300 |
+
"grad_norm": 16.537662506103516,
|
| 301 |
"learning_rate": 5.918367346938776e-06,
|
| 302 |
+
"loss": 0.5503,
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.26380368098159507,
|
| 307 |
+
"grad_norm": 19.7191162109375,
|
| 308 |
"learning_rate": 5.816326530612246e-06,
|
| 309 |
+
"loss": 1.5554,
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.26993865030674846,
|
| 314 |
+
"grad_norm": 28.564775466918945,
|
| 315 |
"learning_rate": 5.7142857142857145e-06,
|
| 316 |
+
"loss": 2.0555,
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.27607361963190186,
|
| 321 |
+
"grad_norm": 37.1331672668457,
|
| 322 |
"learning_rate": 5.6122448979591834e-06,
|
| 323 |
+
"loss": 2.2492,
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.2822085889570552,
|
| 328 |
+
"grad_norm": 40.33460998535156,
|
| 329 |
"learning_rate": 5.510204081632653e-06,
|
| 330 |
+
"loss": 2.8251,
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.2883435582822086,
|
| 335 |
+
"grad_norm": 36.697486877441406,
|
| 336 |
"learning_rate": 5.408163265306123e-06,
|
| 337 |
+
"loss": 0.6507,
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.294478527607362,
|
| 342 |
+
"grad_norm": 38.03018569946289,
|
| 343 |
"learning_rate": 5.306122448979593e-06,
|
| 344 |
+
"loss": 1.335,
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.3006134969325153,
|
| 349 |
+
"grad_norm": 34.11387634277344,
|
| 350 |
"learning_rate": 5.204081632653062e-06,
|
| 351 |
+
"loss": 2.0543,
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.3067484662576687,
|
| 356 |
+
"grad_norm": 33.1131706237793,
|
| 357 |
"learning_rate": 5.1020408163265315e-06,
|
| 358 |
+
"loss": 2.1334,
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.3128834355828221,
|
| 363 |
+
"grad_norm": 27.1938419342041,
|
| 364 |
"learning_rate": 5e-06,
|
| 365 |
+
"loss": 1.4209,
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.31901840490797545,
|
| 370 |
+
"grad_norm": 27.597105026245117,
|
| 371 |
"learning_rate": 4.897959183673469e-06,
|
| 372 |
+
"loss": 1.5301,
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.32515337423312884,
|
| 377 |
+
"grad_norm": 48.28635787963867,
|
| 378 |
"learning_rate": 4.795918367346939e-06,
|
| 379 |
+
"loss": 1.8728,
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.3312883435582822,
|
| 384 |
+
"grad_norm": 29.6666316986084,
|
| 385 |
"learning_rate": 4.693877551020409e-06,
|
| 386 |
+
"loss": 1.7786,
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.3374233128834356,
|
| 391 |
+
"grad_norm": 24.084613800048828,
|
| 392 |
"learning_rate": 4.591836734693878e-06,
|
| 393 |
+
"loss": 1.3896,
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.34355828220858897,
|
| 398 |
+
"grad_norm": 29.24120330810547,
|
| 399 |
"learning_rate": 4.489795918367348e-06,
|
| 400 |
+
"loss": 1.2668,
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.3496932515337423,
|
| 405 |
+
"grad_norm": 20.895090103149414,
|
| 406 |
"learning_rate": 4.3877551020408165e-06,
|
| 407 |
+
"loss": 1.8259,
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.3558282208588957,
|
| 412 |
+
"grad_norm": 28.952999114990234,
|
| 413 |
"learning_rate": 4.2857142857142855e-06,
|
| 414 |
+
"loss": 1.7763,
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.3619631901840491,
|
| 419 |
+
"grad_norm": 19.181821823120117,
|
| 420 |
"learning_rate": 4.183673469387755e-06,
|
| 421 |
+
"loss": 1.9193,
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.36809815950920244,
|
| 426 |
+
"grad_norm": 22.564476013183594,
|
| 427 |
"learning_rate": 4.081632653061225e-06,
|
| 428 |
+
"loss": 1.1478,
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.37423312883435583,
|
| 433 |
+
"grad_norm": 27.566129684448242,
|
| 434 |
"learning_rate": 3.979591836734694e-06,
|
| 435 |
+
"loss": 1.5727,
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.3803680981595092,
|
| 440 |
+
"grad_norm": 29.382600784301758,
|
| 441 |
"learning_rate": 3.877551020408164e-06,
|
| 442 |
+
"loss": 1.7253,
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.38650306748466257,
|
| 447 |
+
"grad_norm": 18.100589752197266,
|
| 448 |
"learning_rate": 3.7755102040816327e-06,
|
| 449 |
+
"loss": 1.8991,
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.39263803680981596,
|
| 454 |
+
"grad_norm": 28.909286499023438,
|
| 455 |
"learning_rate": 3.6734693877551024e-06,
|
| 456 |
+
"loss": 1.3451,
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.3987730061349693,
|
| 461 |
+
"grad_norm": 22.60663414001465,
|
| 462 |
"learning_rate": 3.5714285714285718e-06,
|
| 463 |
+
"loss": 1.4336,
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.4049079754601227,
|
| 468 |
+
"grad_norm": 18.791940689086914,
|
| 469 |
"learning_rate": 3.469387755102041e-06,
|
| 470 |
+
"loss": 1.056,
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.4110429447852761,
|
| 475 |
+
"grad_norm": 27.81426239013672,
|
| 476 |
"learning_rate": 3.3673469387755105e-06,
|
| 477 |
+
"loss": 1.6418,
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.4171779141104294,
|
| 482 |
+
"grad_norm": 48.35829162597656,
|
| 483 |
"learning_rate": 3.2653061224489794e-06,
|
| 484 |
+
"loss": 2.4501,
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.4233128834355828,
|
| 489 |
+
"grad_norm": 27.157045364379883,
|
| 490 |
"learning_rate": 3.1632653061224496e-06,
|
| 491 |
+
"loss": 1.5875,
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.4294478527607362,
|
| 496 |
+
"grad_norm": 30.503982543945312,
|
| 497 |
"learning_rate": 3.0612244897959185e-06,
|
| 498 |
+
"loss": 1.3503,
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.43558282208588955,
|
| 503 |
+
"grad_norm": 23.582530975341797,
|
| 504 |
"learning_rate": 2.959183673469388e-06,
|
| 505 |
+
"loss": 1.3834,
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.44171779141104295,
|
| 510 |
+
"grad_norm": 28.846086502075195,
|
| 511 |
"learning_rate": 2.8571428571428573e-06,
|
| 512 |
+
"loss": 1.7887,
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.44785276073619634,
|
| 517 |
+
"grad_norm": 14.605477333068848,
|
| 518 |
"learning_rate": 2.7551020408163266e-06,
|
| 519 |
+
"loss": 0.979,
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.4539877300613497,
|
| 524 |
+
"grad_norm": 25.760046005249023,
|
| 525 |
"learning_rate": 2.6530612244897964e-06,
|
| 526 |
+
"loss": 1.5663,
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.4601226993865031,
|
| 531 |
+
"grad_norm": 18.653833389282227,
|
| 532 |
"learning_rate": 2.5510204081632657e-06,
|
| 533 |
+
"loss": 1.7759,
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.4662576687116564,
|
| 538 |
+
"grad_norm": 27.419715881347656,
|
| 539 |
"learning_rate": 2.4489795918367347e-06,
|
| 540 |
+
"loss": 2.4035,
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.4723926380368098,
|
| 545 |
+
"grad_norm": 27.352413177490234,
|
| 546 |
"learning_rate": 2.3469387755102044e-06,
|
| 547 |
+
"loss": 1.1448,
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.4785276073619632,
|
| 552 |
+
"grad_norm": 40.72161102294922,
|
| 553 |
"learning_rate": 2.244897959183674e-06,
|
| 554 |
+
"loss": 1.8509,
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.48466257668711654,
|
| 559 |
+
"grad_norm": 33.176212310791016,
|
| 560 |
"learning_rate": 2.1428571428571427e-06,
|
| 561 |
+
"loss": 1.8893,
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.49079754601226994,
|
| 566 |
+
"grad_norm": 29.56044578552246,
|
| 567 |
"learning_rate": 2.0408163265306125e-06,
|
| 568 |
+
"loss": 2.131,
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.49693251533742333,
|
| 573 |
+
"grad_norm": 23.454936981201172,
|
| 574 |
"learning_rate": 1.938775510204082e-06,
|
| 575 |
+
"loss": 1.4972,
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.5030674846625767,
|
| 580 |
+
"grad_norm": 36.35444259643555,
|
| 581 |
"learning_rate": 1.8367346938775512e-06,
|
| 582 |
+
"loss": 1.6202,
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.50920245398773,
|
| 587 |
+
"grad_norm": 27.361839294433594,
|
| 588 |
"learning_rate": 1.7346938775510206e-06,
|
| 589 |
+
"loss": 1.7515,
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.5153374233128835,
|
| 594 |
+
"grad_norm": 40.22319793701172,
|
| 595 |
"learning_rate": 1.6326530612244897e-06,
|
| 596 |
+
"loss": 1.7665,
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.5214723926380368,
|
| 601 |
+
"grad_norm": 24.91071891784668,
|
| 602 |
"learning_rate": 1.5306122448979593e-06,
|
| 603 |
+
"loss": 1.7337,
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 0.5276073619631901,
|
| 608 |
+
"grad_norm": 21.369565963745117,
|
| 609 |
"learning_rate": 1.4285714285714286e-06,
|
| 610 |
+
"loss": 2.279,
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.5337423312883436,
|
| 615 |
+
"grad_norm": 23.668081283569336,
|
| 616 |
"learning_rate": 1.3265306122448982e-06,
|
| 617 |
+
"loss": 1.07,
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.5398773006134969,
|
| 622 |
+
"grad_norm": 25.3243408203125,
|
| 623 |
"learning_rate": 1.2244897959183673e-06,
|
| 624 |
+
"loss": 1.7643,
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.5460122699386503,
|
| 629 |
+
"grad_norm": 24.15099334716797,
|
| 630 |
"learning_rate": 1.122448979591837e-06,
|
| 631 |
+
"loss": 1.4947,
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.5521472392638037,
|
| 636 |
+
"grad_norm": 16.557153701782227,
|
| 637 |
"learning_rate": 1.0204081632653063e-06,
|
| 638 |
+
"loss": 1.5758,
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.558282208588957,
|
| 643 |
+
"grad_norm": 25.761274337768555,
|
| 644 |
"learning_rate": 9.183673469387756e-07,
|
| 645 |
+
"loss": 1.8561,
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 0.5644171779141104,
|
| 650 |
+
"grad_norm": 26.440319061279297,
|
| 651 |
"learning_rate": 8.163265306122449e-07,
|
| 652 |
+
"loss": 1.6105,
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.5705521472392638,
|
| 657 |
+
"grad_norm": 22.7464542388916,
|
| 658 |
"learning_rate": 7.142857142857143e-07,
|
| 659 |
+
"loss": 1.1884,
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 0.5766871165644172,
|
| 664 |
+
"grad_norm": 34.55144500732422,
|
| 665 |
"learning_rate": 6.122448979591837e-07,
|
| 666 |
+
"loss": 1.7501,
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 0.5828220858895705,
|
| 671 |
+
"grad_norm": 35.58631896972656,
|
| 672 |
"learning_rate": 5.102040816326531e-07,
|
| 673 |
+
"loss": 1.3714,
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.588957055214724,
|
| 678 |
+
"grad_norm": 23.65045738220215,
|
| 679 |
"learning_rate": 4.0816326530612243e-07,
|
| 680 |
+
"loss": 1.3916,
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 0.5950920245398773,
|
| 685 |
+
"grad_norm": 38.99002456665039,
|
| 686 |
"learning_rate": 3.0612244897959183e-07,
|
| 687 |
+
"loss": 1.7839,
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 0.6012269938650306,
|
| 692 |
+
"grad_norm": 21.398700714111328,
|
| 693 |
"learning_rate": 2.0408163265306121e-07,
|
| 694 |
+
"loss": 0.9391,
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 0.6073619631901841,
|
| 699 |
+
"grad_norm": 21.622156143188477,
|
| 700 |
"learning_rate": 1.0204081632653061e-07,
|
| 701 |
+
"loss": 1.4566,
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 0.6134969325153374,
|
| 706 |
+
"grad_norm": 47.138668060302734,
|
| 707 |
"learning_rate": 0.0,
|
| 708 |
+
"loss": 2.9306,
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 0.6134969325153374,
|
| 713 |
"step": 100,
|
| 714 |
"total_flos": 2305515375820800.0,
|
| 715 |
+
"train_loss": 1.7183861404657363,
|
| 716 |
+
"train_runtime": 6171.3384,
|
| 717 |
+
"train_samples_per_second": 0.016,
|
| 718 |
+
"train_steps_per_second": 0.016
|
| 719 |
}
|
| 720 |
],
|
| 721 |
"logging_steps": 1,
|