Commit
·
88f2987
1
Parent(s):
f3e4eb3
Update fine-tuned model
Browse files- adapter_config.json +5 -5
- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +855 -155
- training_args.bin +1 -1
adapter_config.json
CHANGED
|
@@ -23,13 +23,13 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"
|
| 27 |
"down_proj",
|
| 28 |
-
"k_proj",
|
| 29 |
-
"v_proj",
|
| 30 |
-
"gate_proj",
|
| 31 |
"o_proj",
|
| 32 |
-
"
|
|
|
|
|
|
|
|
|
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
+
"up_proj",
|
| 27 |
"down_proj",
|
|
|
|
|
|
|
|
|
|
| 28 |
"o_proj",
|
| 29 |
+
"q_proj",
|
| 30 |
+
"v_proj",
|
| 31 |
+
"k_proj",
|
| 32 |
+
"gate_proj"
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 159967880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59dc526f6ba5a23db117f52f37b6930ade017f47b32b850573a201885b9eab70
|
| 3 |
size 159967880
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 852876198
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:65243942704c294c30f9416b50d01ef513eff9fd305acff7984c40dfc22e2b6b
|
| 3 |
size 852876198
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3d76304632bc4c4d4a6fd42f1de24e1df571ce5e7eed001edd03ac5fdbddf9f
|
| 3 |
size 14244
|
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 sha256:194456d3c9e165255d5406a0f3f62973b0bede79d91784f72431350783e27ae7
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,368 +1,1068 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.04,
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.06,
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.08,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.1,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.12,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.14,
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.16,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.18,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.2,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.22,
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
-
"learning_rate": 0.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.24,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.26,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.28,
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.3,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.32,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.34,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.36,
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.38,
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.4,
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.42,
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.44,
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
-
"learning_rate": 0.
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.46,
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
-
"learning_rate": 0.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.48,
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.5,
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 0.
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.52,
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 0.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.54,
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
-
"learning_rate":
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.56,
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate":
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.58,
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.6,
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.62,
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.64,
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.66,
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.68,
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.7,
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.72,
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.74,
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
-
"learning_rate":
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.76,
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.78,
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.8,
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
-
"learning_rate":
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.82,
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.84,
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.86,
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
-
"learning_rate":
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.88,
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.9,
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.92,
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.94,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.96,
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.98,
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 1.0,
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate": 0.
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 360 |
}
|
| 361 |
],
|
| 362 |
"logging_steps": 1,
|
| 363 |
-
"max_steps":
|
| 364 |
"num_input_tokens_seen": 0,
|
| 365 |
-
"num_train_epochs":
|
| 366 |
"save_steps": 500,
|
| 367 |
"stateful_callbacks": {
|
| 368 |
"TrainerControl": {
|
|
@@ -376,7 +1076,7 @@
|
|
| 376 |
"attributes": {}
|
| 377 |
}
|
| 378 |
},
|
| 379 |
-
"total_flos":
|
| 380 |
"train_batch_size": 1,
|
| 381 |
"trial_name": null,
|
| 382 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 150,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 0.4784928858280182,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 0.3585,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.04,
|
| 20 |
+
"grad_norm": 0.4862207770347595,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 0.3655,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.06,
|
| 27 |
+
"grad_norm": 0.4952860474586487,
|
| 28 |
+
"learning_rate": 7.500000000000001e-05,
|
| 29 |
+
"loss": 0.3235,
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.08,
|
| 34 |
+
"grad_norm": 0.4571821093559265,
|
| 35 |
+
"learning_rate": 0.0001,
|
| 36 |
+
"loss": 0.2881,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.1,
|
| 41 |
+
"grad_norm": 0.571702778339386,
|
| 42 |
+
"learning_rate": 0.000125,
|
| 43 |
+
"loss": 0.3014,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.12,
|
| 48 |
+
"grad_norm": 0.9100359678268433,
|
| 49 |
+
"learning_rate": 0.00015000000000000001,
|
| 50 |
+
"loss": 0.3376,
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.14,
|
| 55 |
+
"grad_norm": 0.881874144077301,
|
| 56 |
+
"learning_rate": 0.000175,
|
| 57 |
+
"loss": 0.3476,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.16,
|
| 62 |
+
"grad_norm": 0.898166298866272,
|
| 63 |
+
"learning_rate": 0.0002,
|
| 64 |
+
"loss": 0.3728,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.18,
|
| 69 |
+
"grad_norm": 0.7788503766059875,
|
| 70 |
+
"learning_rate": 0.00019997552766852432,
|
| 71 |
+
"loss": 0.3902,
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.2,
|
| 76 |
+
"grad_norm": 0.740749716758728,
|
| 77 |
+
"learning_rate": 0.00019990212265199738,
|
| 78 |
+
"loss": 0.4114,
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.22,
|
| 83 |
+
"grad_norm": 0.6231009364128113,
|
| 84 |
+
"learning_rate": 0.00019977982087825713,
|
| 85 |
+
"loss": 0.3626,
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.24,
|
| 90 |
+
"grad_norm": 0.5968067646026611,
|
| 91 |
+
"learning_rate": 0.00019960868220749448,
|
| 92 |
+
"loss": 0.4036,
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.26,
|
| 97 |
+
"grad_norm": 0.524666428565979,
|
| 98 |
+
"learning_rate": 0.00019938879040295508,
|
| 99 |
+
"loss": 0.4501,
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.28,
|
| 104 |
+
"grad_norm": 0.4869762659072876,
|
| 105 |
+
"learning_rate": 0.00019912025308994148,
|
| 106 |
+
"loss": 0.4329,
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.3,
|
| 111 |
+
"grad_norm": 0.5021470785140991,
|
| 112 |
+
"learning_rate": 0.0001988032017031364,
|
| 113 |
+
"loss": 0.4473,
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.32,
|
| 118 |
+
"grad_norm": 0.48024389147758484,
|
| 119 |
+
"learning_rate": 0.00019843779142227256,
|
| 120 |
+
"loss": 0.4451,
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.34,
|
| 125 |
+
"grad_norm": 0.4583094120025635,
|
| 126 |
+
"learning_rate": 0.0001980242010961803,
|
| 127 |
+
"loss": 0.4397,
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.36,
|
| 132 |
+
"grad_norm": 0.44530826807022095,
|
| 133 |
+
"learning_rate": 0.0001975626331552507,
|
| 134 |
+
"loss": 0.4625,
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.38,
|
| 139 |
+
"grad_norm": 0.4327012300491333,
|
| 140 |
+
"learning_rate": 0.00019705331351235674,
|
| 141 |
+
"loss": 0.4582,
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.4,
|
| 146 |
+
"grad_norm": 0.4469444453716278,
|
| 147 |
+
"learning_rate": 0.00019649649145228102,
|
| 148 |
+
"loss": 0.4426,
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.42,
|
| 153 |
+
"grad_norm": 0.4537307620048523,
|
| 154 |
+
"learning_rate": 0.00019589243950970402,
|
| 155 |
+
"loss": 0.4862,
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.44,
|
| 160 |
+
"grad_norm": 0.4526830315589905,
|
| 161 |
+
"learning_rate": 0.00019524145333581317,
|
| 162 |
+
"loss": 0.4991,
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.46,
|
| 167 |
+
"grad_norm": 0.4435572922229767,
|
| 168 |
+
"learning_rate": 0.00019454385155359702,
|
| 169 |
+
"loss": 0.5121,
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.48,
|
| 174 |
+
"grad_norm": 0.431131511926651,
|
| 175 |
+
"learning_rate": 0.00019379997560189675,
|
| 176 |
+
"loss": 0.4951,
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.5,
|
| 181 |
+
"grad_norm": 0.4179699122905731,
|
| 182 |
+
"learning_rate": 0.00019301018956828964,
|
| 183 |
+
"loss": 0.4739,
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.52,
|
| 188 |
+
"grad_norm": 0.42843425273895264,
|
| 189 |
+
"learning_rate": 0.00019217488001088784,
|
| 190 |
+
"loss": 0.5042,
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.54,
|
| 195 |
+
"grad_norm": 0.42004117369651794,
|
| 196 |
+
"learning_rate": 0.00019129445576913888,
|
| 197 |
+
"loss": 0.5147,
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.56,
|
| 202 |
+
"grad_norm": 0.4233074486255646,
|
| 203 |
+
"learning_rate": 0.0001903693477637204,
|
| 204 |
+
"loss": 0.4727,
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.58,
|
| 209 |
+
"grad_norm": 0.4170146882534027,
|
| 210 |
+
"learning_rate": 0.00018940000878562758,
|
| 211 |
+
"loss": 0.5034,
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.6,
|
| 216 |
+
"grad_norm": 0.4132724404335022,
|
| 217 |
+
"learning_rate": 0.0001883869132745561,
|
| 218 |
+
"loss": 0.47,
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.62,
|
| 223 |
+
"grad_norm": 0.392947793006897,
|
| 224 |
+
"learning_rate": 0.00018733055708668926,
|
| 225 |
+
"loss": 0.4673,
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.64,
|
| 230 |
+
"grad_norm": 0.42713072896003723,
|
| 231 |
+
"learning_rate": 0.00018623145725200278,
|
| 232 |
+
"loss": 0.5235,
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.66,
|
| 237 |
+
"grad_norm": 0.435976505279541,
|
| 238 |
+
"learning_rate": 0.00018509015172120621,
|
| 239 |
+
"loss": 0.5313,
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.68,
|
| 244 |
+
"grad_norm": 0.38459959626197815,
|
| 245 |
+
"learning_rate": 0.00018390719910244487,
|
| 246 |
+
"loss": 0.5082,
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.7,
|
| 251 |
+
"grad_norm": 0.3872632086277008,
|
| 252 |
+
"learning_rate": 0.00018268317838789088,
|
| 253 |
+
"loss": 0.4809,
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.72,
|
| 258 |
+
"grad_norm": 0.3977827727794647,
|
| 259 |
+
"learning_rate": 0.00018141868867035745,
|
| 260 |
+
"loss": 0.513,
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.74,
|
| 265 |
+
"grad_norm": 0.3733290731906891,
|
| 266 |
+
"learning_rate": 0.00018011434885007482,
|
| 267 |
+
"loss": 0.4954,
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.76,
|
| 272 |
+
"grad_norm": 0.35552307963371277,
|
| 273 |
+
"learning_rate": 0.00017877079733177184,
|
| 274 |
+
"loss": 0.4978,
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.78,
|
| 279 |
+
"grad_norm": 0.36296719312667847,
|
| 280 |
+
"learning_rate": 0.00017738869171221068,
|
| 281 |
+
"loss": 0.5218,
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.8,
|
| 286 |
+
"grad_norm": 0.37389761209487915,
|
| 287 |
+
"learning_rate": 0.0001759687084583285,
|
| 288 |
+
"loss": 0.5305,
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.82,
|
| 293 |
+
"grad_norm": 0.36759287118911743,
|
| 294 |
+
"learning_rate": 0.00017451154257614287,
|
| 295 |
+
"loss": 0.5028,
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.84,
|
| 300 |
+
"grad_norm": 0.3449918031692505,
|
| 301 |
+
"learning_rate": 0.00017301790727058345,
|
| 302 |
+
"loss": 0.4813,
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.86,
|
| 307 |
+
"grad_norm": 0.3552907705307007,
|
| 308 |
+
"learning_rate": 0.00017148853359641626,
|
| 309 |
+
"loss": 0.5117,
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.88,
|
| 314 |
+
"grad_norm": 0.3388005197048187,
|
| 315 |
+
"learning_rate": 0.00016992417010043142,
|
| 316 |
+
"loss": 0.5042,
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.9,
|
| 321 |
+
"grad_norm": 0.34015601873397827,
|
| 322 |
+
"learning_rate": 0.00016832558245506935,
|
| 323 |
+
"loss": 0.5185,
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.92,
|
| 328 |
+
"grad_norm": 0.3483610153198242,
|
| 329 |
+
"learning_rate": 0.0001666935530836651,
|
| 330 |
+
"loss": 0.5137,
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.94,
|
| 335 |
+
"grad_norm": 0.35940223932266235,
|
| 336 |
+
"learning_rate": 0.0001650288807774937,
|
| 337 |
+
"loss": 0.5542,
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.96,
|
| 342 |
+
"grad_norm": 0.3474995791912079,
|
| 343 |
+
"learning_rate": 0.0001633323803048047,
|
| 344 |
+
"loss": 0.5111,
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.98,
|
| 349 |
+
"grad_norm": 0.33267632126808167,
|
| 350 |
+
"learning_rate": 0.00016160488201203644,
|
| 351 |
+
"loss": 0.4997,
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 1.0,
|
| 356 |
+
"grad_norm": 0.33997905254364014,
|
| 357 |
+
"learning_rate": 0.00015984723141740576,
|
| 358 |
+
"loss": 0.5178,
|
| 359 |
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.02,
|
| 363 |
+
"grad_norm": 0.38659659028053284,
|
| 364 |
+
"learning_rate": 0.0001580602887970721,
|
| 365 |
+
"loss": 0.3993,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.04,
|
| 370 |
+
"grad_norm": 0.40527498722076416,
|
| 371 |
+
"learning_rate": 0.0001562449287640781,
|
| 372 |
+
"loss": 0.373,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.06,
|
| 377 |
+
"grad_norm": 0.3660850524902344,
|
| 378 |
+
"learning_rate": 0.00015440203984027324,
|
| 379 |
+
"loss": 0.3658,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.08,
|
| 384 |
+
"grad_norm": 0.3947189450263977,
|
| 385 |
+
"learning_rate": 0.00015253252402142988,
|
| 386 |
+
"loss": 0.395,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.1,
|
| 391 |
+
"grad_norm": 0.40227362513542175,
|
| 392 |
+
"learning_rate": 0.0001506372963357644,
|
| 393 |
+
"loss": 0.3826,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.12,
|
| 398 |
+
"grad_norm": 0.36494743824005127,
|
| 399 |
+
"learning_rate": 0.00014871728439607966,
|
| 400 |
+
"loss": 0.3372,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.1400000000000001,
|
| 405 |
+
"grad_norm": 0.39481741189956665,
|
| 406 |
+
"learning_rate": 0.00014677342794574817,
|
| 407 |
+
"loss": 0.325,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.16,
|
| 412 |
+
"grad_norm": 0.4070734977722168,
|
| 413 |
+
"learning_rate": 0.00014480667839875786,
|
| 414 |
+
"loss": 0.3674,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.18,
|
| 419 |
+
"grad_norm": 0.4173199534416199,
|
| 420 |
+
"learning_rate": 0.00014281799837404552,
|
| 421 |
+
"loss": 0.3447,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.2,
|
| 426 |
+
"grad_norm": 0.4301806390285492,
|
| 427 |
+
"learning_rate": 0.0001408083612243465,
|
| 428 |
+
"loss": 0.3192,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.22,
|
| 433 |
+
"grad_norm": 0.46535176038742065,
|
| 434 |
+
"learning_rate": 0.00013877875055979023,
|
| 435 |
+
"loss": 0.3729,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.24,
|
| 440 |
+
"grad_norm": 0.43888285756111145,
|
| 441 |
+
"learning_rate": 0.00013673015976647568,
|
| 442 |
+
"loss": 0.3285,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.26,
|
| 447 |
+
"grad_norm": 0.4748714566230774,
|
| 448 |
+
"learning_rate": 0.00013466359152026195,
|
| 449 |
+
"loss": 0.3497,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.28,
|
| 454 |
+
"grad_norm": 0.4692974090576172,
|
| 455 |
+
"learning_rate": 0.00013258005729601177,
|
| 456 |
+
"loss": 0.3596,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.3,
|
| 461 |
+
"grad_norm": 0.5036594271659851,
|
| 462 |
+
"learning_rate": 0.00013048057687252865,
|
| 463 |
+
"loss": 0.3248,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.32,
|
| 468 |
+
"grad_norm": 0.4522168040275574,
|
| 469 |
+
"learning_rate": 0.0001283661778334297,
|
| 470 |
+
"loss": 0.3417,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.34,
|
| 475 |
+
"grad_norm": 0.4366334080696106,
|
| 476 |
+
"learning_rate": 0.0001262378950641979,
|
| 477 |
+
"loss": 0.371,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.3599999999999999,
|
| 482 |
+
"grad_norm": 0.42119064927101135,
|
| 483 |
+
"learning_rate": 0.00012409677024566144,
|
| 484 |
+
"loss": 0.3203,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.38,
|
| 489 |
+
"grad_norm": 0.42640116810798645,
|
| 490 |
+
"learning_rate": 0.00012194385134414608,
|
| 491 |
+
"loss": 0.3507,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.4,
|
| 496 |
+
"grad_norm": 0.4241952896118164,
|
| 497 |
+
"learning_rate": 0.00011978019209855174,
|
| 498 |
+
"loss": 0.3497,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.42,
|
| 503 |
+
"grad_norm": 0.4114883840084076,
|
| 504 |
+
"learning_rate": 0.00011760685150460362,
|
| 505 |
+
"loss": 0.3333,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.44,
|
| 510 |
+
"grad_norm": 0.42480409145355225,
|
| 511 |
+
"learning_rate": 0.00011542489329653024,
|
| 512 |
+
"loss": 0.3768,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.46,
|
| 517 |
+
"grad_norm": 0.40450602769851685,
|
| 518 |
+
"learning_rate": 0.00011323538542642227,
|
| 519 |
+
"loss": 0.3378,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.48,
|
| 524 |
+
"grad_norm": 0.39203861355781555,
|
| 525 |
+
"learning_rate": 0.000111039399541527,
|
| 526 |
+
"loss": 0.3475,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.5,
|
| 531 |
+
"grad_norm": 0.38859793543815613,
|
| 532 |
+
"learning_rate": 0.00010883801045973425,
|
| 533 |
+
"loss": 0.3244,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.52,
|
| 538 |
+
"grad_norm": 0.3974066972732544,
|
| 539 |
+
"learning_rate": 0.00010663229564351041,
|
| 540 |
+
"loss": 0.3774,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.54,
|
| 545 |
+
"grad_norm": 0.4268834888935089,
|
| 546 |
+
"learning_rate": 0.00010442333467253789,
|
| 547 |
+
"loss": 0.3494,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.56,
|
| 552 |
+
"grad_norm": 0.4215947985649109,
|
| 553 |
+
"learning_rate": 0.00010221220871531869,
|
| 554 |
+
"loss": 0.35,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.58,
|
| 559 |
+
"grad_norm": 0.40650272369384766,
|
| 560 |
+
"learning_rate": 0.0001,
|
| 561 |
+
"loss": 0.3442,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.6,
|
| 566 |
+
"grad_norm": 0.40281081199645996,
|
| 567 |
+
"learning_rate": 9.778779128468132e-05,
|
| 568 |
+
"loss": 0.3307,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.62,
|
| 573 |
+
"grad_norm": 0.4170817732810974,
|
| 574 |
+
"learning_rate": 9.557666532746213e-05,
|
| 575 |
+
"loss": 0.3424,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.6400000000000001,
|
| 580 |
+
"grad_norm": 0.4091716408729553,
|
| 581 |
+
"learning_rate": 9.336770435648964e-05,
|
| 582 |
+
"loss": 0.3256,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.6600000000000001,
|
| 587 |
+
"grad_norm": 0.4192478656768799,
|
| 588 |
+
"learning_rate": 9.116198954026577e-05,
|
| 589 |
+
"loss": 0.3378,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.6800000000000002,
|
| 594 |
+
"grad_norm": 0.43887075781822205,
|
| 595 |
+
"learning_rate": 8.896060045847304e-05,
|
| 596 |
+
"loss": 0.3392,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.7,
|
| 601 |
+
"grad_norm": 0.42365938425064087,
|
| 602 |
+
"learning_rate": 8.676461457357776e-05,
|
| 603 |
+
"loss": 0.3347,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.72,
|
| 608 |
+
"grad_norm": 0.41718992590904236,
|
| 609 |
+
"learning_rate": 8.457510670346976e-05,
|
| 610 |
+
"loss": 0.3316,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.74,
|
| 615 |
+
"grad_norm": 0.4348912239074707,
|
| 616 |
+
"learning_rate": 8.239314849539638e-05,
|
| 617 |
+
"loss": 0.3139,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.76,
|
| 622 |
+
"grad_norm": 0.41789305210113525,
|
| 623 |
+
"learning_rate": 8.021980790144827e-05,
|
| 624 |
+
"loss": 0.3066,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.78,
|
| 629 |
+
"grad_norm": 0.4338313043117523,
|
| 630 |
+
"learning_rate": 7.805614865585396e-05,
|
| 631 |
+
"loss": 0.3497,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.8,
|
| 636 |
+
"grad_norm": 0.413273423910141,
|
| 637 |
+
"learning_rate": 7.590322975433857e-05,
|
| 638 |
+
"loss": 0.3372,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.8199999999999998,
|
| 643 |
+
"grad_norm": 0.43203985691070557,
|
| 644 |
+
"learning_rate": 7.376210493580212e-05,
|
| 645 |
+
"loss": 0.3601,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.8399999999999999,
|
| 650 |
+
"grad_norm": 0.4087103605270386,
|
| 651 |
+
"learning_rate": 7.163382216657034e-05,
|
| 652 |
+
"loss": 0.3533,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.8599999999999999,
|
| 657 |
+
"grad_norm": 0.41923317313194275,
|
| 658 |
+
"learning_rate": 6.951942312747134e-05,
|
| 659 |
+
"loss": 0.3547,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.88,
|
| 664 |
+
"grad_norm": 0.41816338896751404,
|
| 665 |
+
"learning_rate": 6.741994270398826e-05,
|
| 666 |
+
"loss": 0.3497,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.9,
|
| 671 |
+
"grad_norm": 0.4215528666973114,
|
| 672 |
+
"learning_rate": 6.533640847973808e-05,
|
| 673 |
+
"loss": 0.3396,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.92,
|
| 678 |
+
"grad_norm": 0.3927628695964813,
|
| 679 |
+
"learning_rate": 6.326984023352435e-05,
|
| 680 |
+
"loss": 0.3313,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.94,
|
| 685 |
+
"grad_norm": 0.42098793387413025,
|
| 686 |
+
"learning_rate": 6.122124944020977e-05,
|
| 687 |
+
"loss": 0.3501,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.96,
|
| 692 |
+
"grad_norm": 0.4087188243865967,
|
| 693 |
+
"learning_rate": 5.91916387756535e-05,
|
| 694 |
+
"loss": 0.3311,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.98,
|
| 699 |
+
"grad_norm": 0.42781955003738403,
|
| 700 |
+
"learning_rate": 5.718200162595449e-05,
|
| 701 |
+
"loss": 0.3149,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.0,
|
| 706 |
+
"grad_norm": 0.42511481046676636,
|
| 707 |
+
"learning_rate": 5.5193321601242156e-05,
|
| 708 |
+
"loss": 0.3716,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.02,
|
| 713 |
+
"grad_norm": 0.3746088743209839,
|
| 714 |
+
"learning_rate": 5.322657205425183e-05,
|
| 715 |
+
"loss": 0.2761,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.04,
|
| 720 |
+
"grad_norm": 0.37996241450309753,
|
| 721 |
+
"learning_rate": 5.1282715603920374e-05,
|
| 722 |
+
"loss": 0.2756,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.06,
|
| 727 |
+
"grad_norm": 0.38573023676872253,
|
| 728 |
+
"learning_rate": 4.936270366423563e-05,
|
| 729 |
+
"loss": 0.2501,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.08,
|
| 734 |
+
"grad_norm": 0.3644517660140991,
|
| 735 |
+
"learning_rate": 4.746747597857014e-05,
|
| 736 |
+
"loss": 0.2632,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.1,
|
| 741 |
+
"grad_norm": 0.3908427953720093,
|
| 742 |
+
"learning_rate": 4.559796015972677e-05,
|
| 743 |
+
"loss": 0.2799,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.12,
|
| 748 |
+
"grad_norm": 0.40510475635528564,
|
| 749 |
+
"learning_rate": 4.375507123592194e-05,
|
| 750 |
+
"loss": 0.2581,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.14,
|
| 755 |
+
"grad_norm": 0.38553154468536377,
|
| 756 |
+
"learning_rate": 4.1939711202927936e-05,
|
| 757 |
+
"loss": 0.2495,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.16,
|
| 762 |
+
"grad_norm": 0.396541565656662,
|
| 763 |
+
"learning_rate": 4.015276858259427e-05,
|
| 764 |
+
"loss": 0.2728,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.18,
|
| 769 |
+
"grad_norm": 0.3897378444671631,
|
| 770 |
+
"learning_rate": 3.839511798796357e-05,
|
| 771 |
+
"loss": 0.2551,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.2,
|
| 776 |
+
"grad_norm": 0.3955518305301666,
|
| 777 |
+
"learning_rate": 3.6667619695195285e-05,
|
| 778 |
+
"loss": 0.2704,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.22,
|
| 783 |
+
"grad_norm": 0.4070671498775482,
|
| 784 |
+
"learning_rate": 3.49711192225063e-05,
|
| 785 |
+
"loss": 0.2701,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.24,
|
| 790 |
+
"grad_norm": 0.3995331823825836,
|
| 791 |
+
"learning_rate": 3.330644691633492e-05,
|
| 792 |
+
"loss": 0.2753,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.26,
|
| 797 |
+
"grad_norm": 0.4080023169517517,
|
| 798 |
+
"learning_rate": 3.167441754493066e-05,
|
| 799 |
+
"loss": 0.2483,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.2800000000000002,
|
| 804 |
+
"grad_norm": 0.40179961919784546,
|
| 805 |
+
"learning_rate": 3.0075829899568597e-05,
|
| 806 |
+
"loss": 0.2369,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.3,
|
| 811 |
+
"grad_norm": 0.43834614753723145,
|
| 812 |
+
"learning_rate": 2.8511466403583766e-05,
|
| 813 |
+
"loss": 0.2474,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.32,
|
| 818 |
+
"grad_norm": 0.4417284429073334,
|
| 819 |
+
"learning_rate": 2.6982092729416587e-05,
|
| 820 |
+
"loss": 0.2482,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.34,
|
| 825 |
+
"grad_norm": 0.4265671670436859,
|
| 826 |
+
"learning_rate": 2.548845742385717e-05,
|
| 827 |
+
"loss": 0.2458,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.36,
|
| 832 |
+
"grad_norm": 0.4640134274959564,
|
| 833 |
+
"learning_rate": 2.403129154167153e-05,
|
| 834 |
+
"loss": 0.2555,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.38,
|
| 839 |
+
"grad_norm": 0.42158064246177673,
|
| 840 |
+
"learning_rate": 2.2611308287789344e-05,
|
| 841 |
+
"loss": 0.2485,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.4,
|
| 846 |
+
"grad_norm": 0.42740365862846375,
|
| 847 |
+
"learning_rate": 2.1229202668228197e-05,
|
| 848 |
+
"loss": 0.2681,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.42,
|
| 853 |
+
"grad_norm": 0.4214995801448822,
|
| 854 |
+
"learning_rate": 1.988565114992519e-05,
|
| 855 |
+
"loss": 0.2595,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.44,
|
| 860 |
+
"grad_norm": 0.44696784019470215,
|
| 861 |
+
"learning_rate": 1.858131132964259e-05,
|
| 862 |
+
"loss": 0.2695,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.46,
|
| 867 |
+
"grad_norm": 0.43291181325912476,
|
| 868 |
+
"learning_rate": 1.7316821612109136e-05,
|
| 869 |
+
"loss": 0.2329,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.48,
|
| 874 |
+
"grad_norm": 0.4297809600830078,
|
| 875 |
+
"learning_rate": 1.609280089755515e-05,
|
| 876 |
+
"loss": 0.2521,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.5,
|
| 881 |
+
"grad_norm": 0.42929723858833313,
|
| 882 |
+
"learning_rate": 1.4909848278793782e-05,
|
| 883 |
+
"loss": 0.2447,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.52,
|
| 888 |
+
"grad_norm": 0.414578914642334,
|
| 889 |
+
"learning_rate": 1.3768542747997215e-05,
|
| 890 |
+
"loss": 0.2597,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.54,
|
| 895 |
+
"grad_norm": 0.41869866847991943,
|
| 896 |
+
"learning_rate": 1.2669442913310725e-05,
|
| 897 |
+
"loss": 0.2621,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.56,
|
| 902 |
+
"grad_norm": 0.4209769666194916,
|
| 903 |
+
"learning_rate": 1.161308672544389e-05,
|
| 904 |
+
"loss": 0.25,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.58,
|
| 909 |
+
"grad_norm": 0.40297356247901917,
|
| 910 |
+
"learning_rate": 1.059999121437244e-05,
|
| 911 |
+
"loss": 0.239,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.6,
|
| 916 |
+
"grad_norm": 0.4161122739315033,
|
| 917 |
+
"learning_rate": 9.630652236279625e-06,
|
| 918 |
+
"loss": 0.2621,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.62,
|
| 923 |
+
"grad_norm": 0.43759092688560486,
|
| 924 |
+
"learning_rate": 8.70554423086114e-06,
|
| 925 |
+
"loss": 0.2681,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.64,
|
| 930 |
+
"grad_norm": 0.41500815749168396,
|
| 931 |
+
"learning_rate": 7.825119989112173e-06,
|
| 932 |
+
"loss": 0.2547,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.66,
|
| 937 |
+
"grad_norm": 0.40593695640563965,
|
| 938 |
+
"learning_rate": 6.989810431710375e-06,
|
| 939 |
+
"loss": 0.2482,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.68,
|
| 944 |
+
"grad_norm": 0.40096405148506165,
|
| 945 |
+
"learning_rate": 6.200024398103255e-06,
|
| 946 |
+
"loss": 0.2402,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.7,
|
| 951 |
+
"grad_norm": 0.39252418279647827,
|
| 952 |
+
"learning_rate": 5.456148446402976e-06,
|
| 953 |
+
"loss": 0.2396,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.7199999999999998,
|
| 958 |
+
"grad_norm": 0.41394537687301636,
|
| 959 |
+
"learning_rate": 4.758546664186869e-06,
|
| 960 |
+
"loss": 0.2729,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.74,
|
| 965 |
+
"grad_norm": 0.40165257453918457,
|
| 966 |
+
"learning_rate": 4.107560490295992e-06,
|
| 967 |
+
"loss": 0.2497,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.76,
|
| 972 |
+
"grad_norm": 0.42329922318458557,
|
| 973 |
+
"learning_rate": 3.5035085477190143e-06,
|
| 974 |
+
"loss": 0.2444,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.7800000000000002,
|
| 979 |
+
"grad_norm": 0.4126729965209961,
|
| 980 |
+
"learning_rate": 2.94668648764328e-06,
|
| 981 |
+
"loss": 0.256,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.8,
|
| 986 |
+
"grad_norm": 0.4390856623649597,
|
| 987 |
+
"learning_rate": 2.4373668447493224e-06,
|
| 988 |
+
"loss": 0.2591,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.82,
|
| 993 |
+
"grad_norm": 0.4362524151802063,
|
| 994 |
+
"learning_rate": 1.9757989038197146e-06,
|
| 995 |
+
"loss": 0.255,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.84,
|
| 1000 |
+
"grad_norm": 0.43163925409317017,
|
| 1001 |
+
"learning_rate": 1.562208577727442e-06,
|
| 1002 |
+
"loss": 0.2581,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.86,
|
| 1007 |
+
"grad_norm": 0.418253093957901,
|
| 1008 |
+
"learning_rate": 1.1967982968635993e-06,
|
| 1009 |
+
"loss": 0.2535,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.88,
|
| 1014 |
+
"grad_norm": 0.43555358052253723,
|
| 1015 |
+
"learning_rate": 8.797469100585431e-07,
|
| 1016 |
+
"loss": 0.272,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.9,
|
| 1021 |
+
"grad_norm": 0.412000834941864,
|
| 1022 |
+
"learning_rate": 6.11209597044926e-07,
|
| 1023 |
+
"loss": 0.2344,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.92,
|
| 1028 |
+
"grad_norm": 0.41050902009010315,
|
| 1029 |
+
"learning_rate": 3.913177925055189e-07,
|
| 1030 |
+
"loss": 0.2403,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.94,
|
| 1035 |
+
"grad_norm": 0.40769293904304504,
|
| 1036 |
+
"learning_rate": 2.201791217428917e-07,
|
| 1037 |
+
"loss": 0.2416,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.96,
|
| 1042 |
+
"grad_norm": 0.4063026010990143,
|
| 1043 |
+
"learning_rate": 9.78773480026396e-08,
|
| 1044 |
+
"loss": 0.2491,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.98,
|
| 1049 |
+
"grad_norm": 0.4020632207393646,
|
| 1050 |
+
"learning_rate": 2.447233147570005e-08,
|
| 1051 |
+
"loss": 0.2472,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.0,
|
| 1056 |
+
"grad_norm": 0.42892733216285706,
|
| 1057 |
+
"learning_rate": 0.0,
|
| 1058 |
+
"loss": 0.2704,
|
| 1059 |
+
"step": 150
|
| 1060 |
}
|
| 1061 |
],
|
| 1062 |
"logging_steps": 1,
|
| 1063 |
+
"max_steps": 150,
|
| 1064 |
"num_input_tokens_seen": 0,
|
| 1065 |
+
"num_train_epochs": 3,
|
| 1066 |
"save_steps": 500,
|
| 1067 |
"stateful_callbacks": {
|
| 1068 |
"TrainerControl": {
|
|
|
|
| 1076 |
"attributes": {}
|
| 1077 |
}
|
| 1078 |
},
|
| 1079 |
+
"total_flos": 1.817534143122309e+17,
|
| 1080 |
"train_batch_size": 1,
|
| 1081 |
"trial_name": null,
|
| 1082 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6cb8544d6924ca831de7a25711524b3181e61dc7c16799e3b28a736e7c76e00c
|
| 3 |
size 5240
|