Training in progress, step 50, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 250490408
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c18b4cf96983818fb58b4a5e336f7b4a778ac8603896037ce0c1a673994e7518
|
| 3 |
size 250490408
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7fb381715a6f4a8adab9f7d14d2d354ecb9694c12411525b7da9969b9055c04
|
| 3 |
+
size 255265850
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5893d9851ea92670a3d4726f43f20d99c50539fb42caed0b0dc2086e89cfc271
|
| 3 |
size 14512
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:683aee245ed54365847ef67e55fa73019e878211c9d91ace134780fe9514381e
|
| 3 |
size 14512
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b5bd1b09d58ee3e65553e6f9772dc7c5ca98238b1cade33cccd500df2328864
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -11,341 +11,60 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.08333333333333333,
|
| 13 |
"eval_loss": 3.261073350906372,
|
| 14 |
-
"eval_runtime": 10.
|
| 15 |
-
"eval_samples_per_second": 141.
|
| 16 |
-
"eval_steps_per_second": 2.
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.8333333333333334,
|
| 21 |
-
"grad_norm":
|
| 22 |
"learning_rate": 6.666666666666667e-05,
|
| 23 |
-
"loss": 6.
|
| 24 |
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 1.6666666666666665,
|
| 28 |
-
"grad_norm":
|
| 29 |
"learning_rate": 0.00013333333333333334,
|
| 30 |
-
"loss": 5.
|
| 31 |
"step": 20
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 2.5,
|
| 35 |
-
"grad_norm": 4.
|
| 36 |
"learning_rate": 0.0002,
|
| 37 |
-
"loss": 5.
|
| 38 |
"step": 30
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 3.3333333333333335,
|
| 42 |
-
"grad_norm": 4.
|
| 43 |
"learning_rate": 0.00019984815164333163,
|
| 44 |
-
"loss": 5.
|
| 45 |
-
"step": 40
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 3.3333333333333335,
|
| 49 |
-
"eval_loss": 3.1405746936798096,
|
| 50 |
-
"eval_runtime": 10.6366,
|
| 51 |
-
"eval_samples_per_second": 141.117,
|
| 52 |
-
"eval_steps_per_second": 2.256,
|
| 53 |
"step": 40
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 4.166666666666667,
|
| 57 |
-
"grad_norm":
|
| 58 |
"learning_rate": 0.00019939306773179497,
|
| 59 |
-
"loss": 5.
|
| 60 |
"step": 50
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch":
|
| 64 |
-
"
|
| 65 |
-
"
|
| 66 |
-
"
|
| 67 |
-
"
|
| 68 |
-
|
| 69 |
-
{
|
| 70 |
-
"epoch": 5.833333333333333,
|
| 71 |
-
"grad_norm": 4.59375,
|
| 72 |
-
"learning_rate": 0.00019757963826274357,
|
| 73 |
-
"loss": 4.6862,
|
| 74 |
-
"step": 70
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 6.666666666666667,
|
| 78 |
-
"grad_norm": 4.3125,
|
| 79 |
-
"learning_rate": 0.00019622680003092503,
|
| 80 |
-
"loss": 4.4865,
|
| 81 |
-
"step": 80
|
| 82 |
-
},
|
| 83 |
-
{
|
| 84 |
-
"epoch": 6.666666666666667,
|
| 85 |
-
"eval_loss": 3.095829725265503,
|
| 86 |
-
"eval_runtime": 10.2886,
|
| 87 |
-
"eval_samples_per_second": 145.89,
|
| 88 |
-
"eval_steps_per_second": 2.333,
|
| 89 |
-
"step": 80
|
| 90 |
-
},
|
| 91 |
-
{
|
| 92 |
-
"epoch": 7.5,
|
| 93 |
-
"grad_norm": 3.875,
|
| 94 |
-
"learning_rate": 0.00019458172417006347,
|
| 95 |
-
"loss": 4.3738,
|
| 96 |
-
"step": 90
|
| 97 |
-
},
|
| 98 |
-
{
|
| 99 |
-
"epoch": 8.333333333333334,
|
| 100 |
-
"grad_norm": 4.0625,
|
| 101 |
-
"learning_rate": 0.00019264940672148018,
|
| 102 |
-
"loss": 4.2197,
|
| 103 |
-
"step": 100
|
| 104 |
-
},
|
| 105 |
-
{
|
| 106 |
-
"epoch": 9.166666666666666,
|
| 107 |
-
"grad_norm": 4.625,
|
| 108 |
-
"learning_rate": 0.00019043571606975777,
|
| 109 |
-
"loss": 4.1157,
|
| 110 |
-
"step": 110
|
| 111 |
-
},
|
| 112 |
-
{
|
| 113 |
-
"epoch": 10.0,
|
| 114 |
-
"grad_norm": 4.40625,
|
| 115 |
-
"learning_rate": 0.0001879473751206489,
|
| 116 |
-
"loss": 4.015,
|
| 117 |
-
"step": 120
|
| 118 |
-
},
|
| 119 |
-
{
|
| 120 |
-
"epoch": 10.0,
|
| 121 |
-
"eval_loss": 3.147190570831299,
|
| 122 |
-
"eval_runtime": 10.4825,
|
| 123 |
-
"eval_samples_per_second": 143.191,
|
| 124 |
-
"eval_steps_per_second": 2.29,
|
| 125 |
-
"step": 120
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 10.833333333333334,
|
| 129 |
-
"grad_norm": 3.4375,
|
| 130 |
-
"learning_rate": 0.00018519194088383273,
|
| 131 |
-
"loss": 3.913,
|
| 132 |
-
"step": 130
|
| 133 |
-
},
|
| 134 |
-
{
|
| 135 |
-
"epoch": 11.666666666666666,
|
| 136 |
-
"grad_norm": 3.859375,
|
| 137 |
-
"learning_rate": 0.0001821777815225245,
|
| 138 |
-
"loss": 3.8009,
|
| 139 |
-
"step": 140
|
| 140 |
-
},
|
| 141 |
-
{
|
| 142 |
-
"epoch": 12.5,
|
| 143 |
-
"grad_norm": 4.375,
|
| 144 |
-
"learning_rate": 0.00017891405093963938,
|
| 145 |
-
"loss": 3.7176,
|
| 146 |
-
"step": 150
|
| 147 |
-
},
|
| 148 |
-
{
|
| 149 |
-
"epoch": 13.333333333333334,
|
| 150 |
-
"grad_norm": 4.25,
|
| 151 |
-
"learning_rate": 0.00017541066097768963,
|
| 152 |
-
"loss": 3.6337,
|
| 153 |
-
"step": 160
|
| 154 |
-
},
|
| 155 |
-
{
|
| 156 |
-
"epoch": 13.333333333333334,
|
| 157 |
-
"eval_loss": 3.2374608516693115,
|
| 158 |
-
"eval_runtime": 10.482,
|
| 159 |
-
"eval_samples_per_second": 143.198,
|
| 160 |
-
"eval_steps_per_second": 2.29,
|
| 161 |
-
"step": 160
|
| 162 |
-
},
|
| 163 |
-
{
|
| 164 |
-
"epoch": 14.166666666666666,
|
| 165 |
-
"grad_norm": 4.5,
|
| 166 |
-
"learning_rate": 0.00017167825131684513,
|
| 167 |
-
"loss": 3.5521,
|
| 168 |
-
"step": 170
|
| 169 |
-
},
|
| 170 |
-
{
|
| 171 |
-
"epoch": 15.0,
|
| 172 |
-
"grad_norm": 4.90625,
|
| 173 |
-
"learning_rate": 0.00016772815716257412,
|
| 174 |
-
"loss": 3.4769,
|
| 175 |
-
"step": 180
|
| 176 |
-
},
|
| 177 |
-
{
|
| 178 |
-
"epoch": 15.833333333333334,
|
| 179 |
-
"grad_norm": 5.6875,
|
| 180 |
-
"learning_rate": 0.00016357237482099684,
|
| 181 |
-
"loss": 3.3954,
|
| 182 |
-
"step": 190
|
| 183 |
-
},
|
| 184 |
-
{
|
| 185 |
-
"epoch": 16.666666666666668,
|
| 186 |
-
"grad_norm": 3.921875,
|
| 187 |
-
"learning_rate": 0.00015922352526649803,
|
| 188 |
-
"loss": 3.3223,
|
| 189 |
-
"step": 200
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"epoch": 16.666666666666668,
|
| 193 |
-
"eval_loss": 3.2679085731506348,
|
| 194 |
-
"eval_runtime": 10.4994,
|
| 195 |
-
"eval_samples_per_second": 142.96,
|
| 196 |
-
"eval_steps_per_second": 2.286,
|
| 197 |
-
"step": 200
|
| 198 |
-
},
|
| 199 |
-
{
|
| 200 |
-
"epoch": 17.5,
|
| 201 |
-
"grad_norm": 4.1875,
|
| 202 |
-
"learning_rate": 0.00015469481581224272,
|
| 203 |
-
"loss": 3.2339,
|
| 204 |
-
"step": 210
|
| 205 |
-
},
|
| 206 |
-
{
|
| 207 |
-
"epoch": 18.333333333333332,
|
| 208 |
-
"grad_norm": 4.46875,
|
| 209 |
-
"learning_rate": 0.00015000000000000001,
|
| 210 |
-
"loss": 3.1968,
|
| 211 |
-
"step": 220
|
| 212 |
-
},
|
| 213 |
-
{
|
| 214 |
-
"epoch": 19.166666666666668,
|
| 215 |
-
"grad_norm": 4.03125,
|
| 216 |
-
"learning_rate": 0.00014515333583108896,
|
| 217 |
-
"loss": 3.1189,
|
| 218 |
-
"step": 230
|
| 219 |
-
},
|
| 220 |
-
{
|
| 221 |
-
"epoch": 20.0,
|
| 222 |
-
"grad_norm": 3.75,
|
| 223 |
-
"learning_rate": 0.00014016954246529696,
|
| 224 |
-
"loss": 3.0702,
|
| 225 |
-
"step": 240
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 20.0,
|
| 229 |
-
"eval_loss": 3.0596697330474854,
|
| 230 |
-
"eval_runtime": 10.3849,
|
| 231 |
-
"eval_samples_per_second": 144.537,
|
| 232 |
-
"eval_steps_per_second": 2.311,
|
| 233 |
-
"step": 240
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 20.833333333333332,
|
| 237 |
-
"grad_norm": 3.890625,
|
| 238 |
-
"learning_rate": 0.00013506375551927547,
|
| 239 |
-
"loss": 3.0297,
|
| 240 |
-
"step": 250
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 21.666666666666668,
|
| 244 |
-
"grad_norm": 3.96875,
|
| 245 |
-
"learning_rate": 0.00012985148110016947,
|
| 246 |
-
"loss": 2.9882,
|
| 247 |
-
"step": 260
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 22.5,
|
| 251 |
-
"grad_norm": 3.859375,
|
| 252 |
-
"learning_rate": 0.00012454854871407994,
|
| 253 |
-
"loss": 2.931,
|
| 254 |
-
"step": 270
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 23.333333333333332,
|
| 258 |
-
"grad_norm": 4.0625,
|
| 259 |
-
"learning_rate": 0.00011917106319237386,
|
| 260 |
-
"loss": 2.8933,
|
| 261 |
-
"step": 280
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 23.333333333333332,
|
| 265 |
-
"eval_loss": 3.0961873531341553,
|
| 266 |
-
"eval_runtime": 10.406,
|
| 267 |
-
"eval_samples_per_second": 144.243,
|
| 268 |
-
"eval_steps_per_second": 2.306,
|
| 269 |
-
"step": 280
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 24.166666666666668,
|
| 273 |
-
"grad_norm": 3.84375,
|
| 274 |
-
"learning_rate": 0.00011373535578184082,
|
| 275 |
-
"loss": 2.8435,
|
| 276 |
-
"step": 290
|
| 277 |
-
},
|
| 278 |
-
{
|
| 279 |
-
"epoch": 25.0,
|
| 280 |
-
"grad_norm": 3.53125,
|
| 281 |
-
"learning_rate": 0.00010825793454723325,
|
| 282 |
-
"loss": 2.8095,
|
| 283 |
-
"step": 300
|
| 284 |
-
},
|
| 285 |
-
{
|
| 286 |
-
"epoch": 25.833333333333332,
|
| 287 |
-
"grad_norm": 3.59375,
|
| 288 |
-
"learning_rate": 0.00010275543423681621,
|
| 289 |
-
"loss": 2.7729,
|
| 290 |
-
"step": 310
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 26.666666666666668,
|
| 294 |
-
"grad_norm": 4.125,
|
| 295 |
-
"learning_rate": 9.724456576318381e-05,
|
| 296 |
-
"loss": 2.7436,
|
| 297 |
-
"step": 320
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 26.666666666666668,
|
| 301 |
-
"eval_loss": 3.106285810470581,
|
| 302 |
-
"eval_runtime": 10.2808,
|
| 303 |
-
"eval_samples_per_second": 146.0,
|
| 304 |
-
"eval_steps_per_second": 2.334,
|
| 305 |
-
"step": 320
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 27.5,
|
| 309 |
-
"grad_norm": 3.15625,
|
| 310 |
-
"learning_rate": 9.174206545276677e-05,
|
| 311 |
-
"loss": 2.703,
|
| 312 |
-
"step": 330
|
| 313 |
-
},
|
| 314 |
-
{
|
| 315 |
-
"epoch": 28.333333333333332,
|
| 316 |
-
"grad_norm": 4.15625,
|
| 317 |
-
"learning_rate": 8.626464421815919e-05,
|
| 318 |
-
"loss": 2.692,
|
| 319 |
-
"step": 340
|
| 320 |
-
},
|
| 321 |
-
{
|
| 322 |
-
"epoch": 29.166666666666668,
|
| 323 |
-
"grad_norm": 3.765625,
|
| 324 |
-
"learning_rate": 8.082893680762619e-05,
|
| 325 |
-
"loss": 2.6466,
|
| 326 |
-
"step": 350
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 30.0,
|
| 330 |
-
"grad_norm": 3.71875,
|
| 331 |
-
"learning_rate": 7.54514512859201e-05,
|
| 332 |
-
"loss": 2.6371,
|
| 333 |
-
"step": 360
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 30.0,
|
| 337 |
-
"eval_loss": 3.0690042972564697,
|
| 338 |
-
"eval_runtime": 10.5907,
|
| 339 |
-
"eval_samples_per_second": 141.728,
|
| 340 |
-
"eval_steps_per_second": 2.266,
|
| 341 |
-
"step": 360
|
| 342 |
}
|
| 343 |
],
|
| 344 |
"logging_steps": 10,
|
| 345 |
"max_steps": 600,
|
| 346 |
"num_input_tokens_seen": 0,
|
| 347 |
"num_train_epochs": 50,
|
| 348 |
-
"save_steps":
|
| 349 |
"stateful_callbacks": {
|
| 350 |
"TrainerControl": {
|
| 351 |
"args": {
|
|
@@ -358,7 +77,7 @@
|
|
| 358 |
"attributes": {}
|
| 359 |
}
|
| 360 |
},
|
| 361 |
-
"total_flos":
|
| 362 |
"train_batch_size": 32,
|
| 363 |
"trial_name": null,
|
| 364 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.166666666666667,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 50,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.08333333333333333,
|
| 13 |
"eval_loss": 3.261073350906372,
|
| 14 |
+
"eval_runtime": 10.6205,
|
| 15 |
+
"eval_samples_per_second": 141.33,
|
| 16 |
+
"eval_steps_per_second": 2.26,
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.8333333333333334,
|
| 21 |
+
"grad_norm": 8.0625,
|
| 22 |
"learning_rate": 6.666666666666667e-05,
|
| 23 |
+
"loss": 6.5691,
|
| 24 |
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 1.6666666666666665,
|
| 28 |
+
"grad_norm": 3.890625,
|
| 29 |
"learning_rate": 0.00013333333333333334,
|
| 30 |
+
"loss": 5.8025,
|
| 31 |
"step": 20
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 2.5,
|
| 35 |
+
"grad_norm": 4.375,
|
| 36 |
"learning_rate": 0.0002,
|
| 37 |
+
"loss": 5.5103,
|
| 38 |
"step": 30
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 3.3333333333333335,
|
| 42 |
+
"grad_norm": 4.78125,
|
| 43 |
"learning_rate": 0.00019984815164333163,
|
| 44 |
+
"loss": 5.3092,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
"step": 40
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 4.166666666666667,
|
| 49 |
+
"grad_norm": 4.1875,
|
| 50 |
"learning_rate": 0.00019939306773179497,
|
| 51 |
+
"loss": 5.0693,
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 4.166666666666667,
|
| 56 |
+
"eval_loss": 3.1816465854644775,
|
| 57 |
+
"eval_runtime": 10.8479,
|
| 58 |
+
"eval_samples_per_second": 138.368,
|
| 59 |
+
"eval_steps_per_second": 2.212,
|
| 60 |
+
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
}
|
| 62 |
],
|
| 63 |
"logging_steps": 10,
|
| 64 |
"max_steps": 600,
|
| 65 |
"num_input_tokens_seen": 0,
|
| 66 |
"num_train_epochs": 50,
|
| 67 |
+
"save_steps": 50,
|
| 68 |
"stateful_callbacks": {
|
| 69 |
"TrainerControl": {
|
| 70 |
"args": {
|
|
|
|
| 77 |
"attributes": {}
|
| 78 |
}
|
| 79 |
},
|
| 80 |
+
"total_flos": 3242111495307264.0,
|
| 81 |
"train_batch_size": 32,
|
| 82 |
"trial_name": null,
|
| 83 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6840
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:701b8fdcf7ee9a32ee2820bf97b3ab785ffbf4aa2b41bcd64bfa018085f95834
|
| 3 |
size 6840
|