Upload folder using huggingface_hub
Browse files- config.json +1 -1
- model-00001-of-00002.safetensors +2 -2
- model-00002-of-00002.safetensors +1 -1
- model.safetensors.index.json +2 -2
- optimizer.pt +2 -2
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +55 -600
- training_args.bin +1 -1
config.json
CHANGED
|
@@ -31,5 +31,5 @@
|
|
| 31 |
"tie_word_embeddings": true,
|
| 32 |
"transformers_version": "4.56.0",
|
| 33 |
"use_cache": true,
|
| 34 |
-
"vocab_size":
|
| 35 |
}
|
|
|
|
| 31 |
"tie_word_embeddings": true,
|
| 32 |
"transformers_version": "4.56.0",
|
| 33 |
"use_cache": true,
|
| 34 |
+
"vocab_size": 156940
|
| 35 |
}
|
model-00001-of-00002.safetensors
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 sha256:33a77d12adc3c90a6d03d30b2ebb684027e4f21c9ca19ce87864192c1cbdd7b8
|
| 3 |
+
size 4991037968
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1610725592
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16b197f53c7eb5f953acb0917626be62d881b989a1678589016fddbdd2e620b8
|
| 3 |
size 1610725592
|
model.safetensors.index.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"metadata": {
|
| 3 |
-
"total_parameters":
|
| 4 |
-
"total_size":
|
| 5 |
},
|
| 6 |
"weight_map": {
|
| 7 |
"model.embed_tokens.weight": "model-00001-of-00002.safetensors",
|
|
|
|
| 1 |
{
|
| 2 |
"metadata": {
|
| 3 |
+
"total_parameters": 3300867072,
|
| 4 |
+
"total_size": 6601734144
|
| 5 |
},
|
| 6 |
"weight_map": {
|
| 7 |
"model.embed_tokens.weight": "model-00001-of-00002.safetensors",
|
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 sha256:709540f11f094ab7fcb18f525b097ef780a66646213b0f225b0cec2172f4c781
|
| 3 |
+
size 13203690391
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8e2011629d8bed3ef560fa11175cac55684c4e12a72634bb24abf767b6c7399
|
| 3 |
size 14645
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3159b8255e3ba63ecfbf9ad9882d37c8b55d7643e07b70fee54fef23e5ee0ce
|
| 3 |
size 1465
|
trainer_state.json
CHANGED
|
@@ -1,642 +1,97 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric":
|
| 4 |
-
"best_model_checkpoint":
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"epoch": 0.
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate":
|
| 16 |
-
"loss":
|
| 17 |
"step": 50
|
| 18 |
},
|
| 19 |
{
|
| 20 |
-
"epoch": 0.
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss":
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"loss":
|
| 31 |
"step": 150
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 0.
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"grad_norm":
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss":
|
| 45 |
"step": 250
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch":
|
| 49 |
-
"grad_norm":
|
| 50 |
-
"learning_rate":
|
| 51 |
-
"loss": 5.
|
| 52 |
"step": 300
|
| 53 |
},
|
| 54 |
{
|
| 55 |
-
"epoch":
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss":
|
| 59 |
"step": 350
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch":
|
| 63 |
-
"grad_norm":
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss":
|
| 66 |
"step": 400
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch":
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss":
|
| 73 |
"step": 450
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch":
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss":
|
| 80 |
"step": 500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch":
|
| 84 |
-
"
|
| 85 |
-
"
|
| 86 |
-
"
|
| 87 |
-
"
|
| 88 |
-
|
| 89 |
-
{
|
| 90 |
-
"epoch": 1.3452914798206277,
|
| 91 |
-
"grad_norm": 5.15625,
|
| 92 |
-
"learning_rate": 4.923476455971e-05,
|
| 93 |
-
"loss": 4.7371,
|
| 94 |
-
"step": 600
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 1.4573991031390134,
|
| 98 |
-
"grad_norm": 7.3125,
|
| 99 |
-
"learning_rate": 4.898780414293411e-05,
|
| 100 |
-
"loss": 4.7189,
|
| 101 |
-
"step": 650
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 1.5695067264573992,
|
| 105 |
-
"grad_norm": 4.21875,
|
| 106 |
-
"learning_rate": 4.870716645577244e-05,
|
| 107 |
-
"loss": 4.7196,
|
| 108 |
-
"step": 700
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 1.6816143497757847,
|
| 112 |
-
"grad_norm": 4.96875,
|
| 113 |
-
"learning_rate": 4.839324549474148e-05,
|
| 114 |
-
"loss": 4.7285,
|
| 115 |
-
"step": 750
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 1.7937219730941703,
|
| 119 |
-
"grad_norm": 4.125,
|
| 120 |
-
"learning_rate": 4.804648198384507e-05,
|
| 121 |
-
"loss": 4.7366,
|
| 122 |
-
"step": 800
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 1.905829596412556,
|
| 126 |
-
"grad_norm": 4.21875,
|
| 127 |
-
"learning_rate": 4.7667362755827306e-05,
|
| 128 |
-
"loss": 4.712,
|
| 129 |
-
"step": 850
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 2.0179372197309418,
|
| 133 |
-
"grad_norm": 4.09375,
|
| 134 |
-
"learning_rate": 4.725642006869207e-05,
|
| 135 |
-
"loss": 4.6238,
|
| 136 |
-
"step": 900
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 2.1300448430493275,
|
| 140 |
-
"grad_norm": 4.3125,
|
| 141 |
-
"learning_rate": 4.68142308584484e-05,
|
| 142 |
-
"loss": 4.3582,
|
| 143 |
-
"step": 950
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 2.242152466367713,
|
| 147 |
-
"grad_norm": 6.625,
|
| 148 |
-
"learning_rate": 4.634141592913097e-05,
|
| 149 |
-
"loss": 4.3645,
|
| 150 |
-
"step": 1000
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 2.3542600896860986,
|
| 154 |
-
"grad_norm": 4.15625,
|
| 155 |
-
"learning_rate": 4.583863908123282e-05,
|
| 156 |
-
"loss": 4.413,
|
| 157 |
-
"step": 1050
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 2.4663677130044843,
|
| 161 |
-
"grad_norm": 4.21875,
|
| 162 |
-
"learning_rate": 4.530660617977393e-05,
|
| 163 |
-
"loss": 4.3569,
|
| 164 |
-
"step": 1100
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 2.57847533632287,
|
| 168 |
-
"grad_norm": 4.125,
|
| 169 |
-
"learning_rate": 4.474606416331397e-05,
|
| 170 |
-
"loss": 4.3704,
|
| 171 |
-
"step": 1150
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 2.6905829596412554,
|
| 175 |
-
"grad_norm": 4.15625,
|
| 176 |
-
"learning_rate": 4.415779999530064e-05,
|
| 177 |
-
"loss": 4.406,
|
| 178 |
-
"step": 1200
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 2.802690582959641,
|
| 182 |
-
"grad_norm": 3.921875,
|
| 183 |
-
"learning_rate": 4.354263955922568e-05,
|
| 184 |
-
"loss": 4.3779,
|
| 185 |
-
"step": 1250
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 2.914798206278027,
|
| 189 |
-
"grad_norm": 4.21875,
|
| 190 |
-
"learning_rate": 4.290144649913973e-05,
|
| 191 |
-
"loss": 4.3843,
|
| 192 |
-
"step": 1300
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 3.0269058295964126,
|
| 196 |
-
"grad_norm": 5.65625,
|
| 197 |
-
"learning_rate": 4.2235121007153975e-05,
|
| 198 |
-
"loss": 4.246,
|
| 199 |
-
"step": 1350
|
| 200 |
-
},
|
| 201 |
-
{
|
| 202 |
-
"epoch": 3.1390134529147984,
|
| 203 |
-
"grad_norm": 5.3125,
|
| 204 |
-
"learning_rate": 4.1544598559630694e-05,
|
| 205 |
-
"loss": 3.891,
|
| 206 |
-
"step": 1400
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 3.2511210762331837,
|
| 210 |
-
"grad_norm": 5.9375,
|
| 211 |
-
"learning_rate": 4.083084860383708e-05,
|
| 212 |
-
"loss": 3.9163,
|
| 213 |
-
"step": 1450
|
| 214 |
-
},
|
| 215 |
-
{
|
| 216 |
-
"epoch": 3.3632286995515694,
|
| 217 |
-
"grad_norm": 4.96875,
|
| 218 |
-
"learning_rate": 4.009487319690626e-05,
|
| 219 |
-
"loss": 3.9105,
|
| 220 |
-
"step": 1500
|
| 221 |
-
},
|
| 222 |
-
{
|
| 223 |
-
"epoch": 3.475336322869955,
|
| 224 |
-
"grad_norm": 4.90625,
|
| 225 |
-
"learning_rate": 3.9337705599016145e-05,
|
| 226 |
-
"loss": 3.9388,
|
| 227 |
-
"step": 1550
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 3.587443946188341,
|
| 231 |
-
"grad_norm": 4.78125,
|
| 232 |
-
"learning_rate": 3.856040882276136e-05,
|
| 233 |
-
"loss": 3.9035,
|
| 234 |
-
"step": 1600
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 3.6995515695067267,
|
| 238 |
-
"grad_norm": 4.6875,
|
| 239 |
-
"learning_rate": 3.776407414075477e-05,
|
| 240 |
-
"loss": 3.9022,
|
| 241 |
-
"step": 1650
|
| 242 |
-
},
|
| 243 |
-
{
|
| 244 |
-
"epoch": 3.811659192825112,
|
| 245 |
-
"grad_norm": 5.625,
|
| 246 |
-
"learning_rate": 3.6949819553553814e-05,
|
| 247 |
-
"loss": 3.9107,
|
| 248 |
-
"step": 1700
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 3.9237668161434978,
|
| 252 |
-
"grad_norm": 4.71875,
|
| 253 |
-
"learning_rate": 3.611878822006261e-05,
|
| 254 |
-
"loss": 3.9135,
|
| 255 |
-
"step": 1750
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 4.0358744394618835,
|
| 259 |
-
"grad_norm": 7.25,
|
| 260 |
-
"learning_rate": 3.527214685261339e-05,
|
| 261 |
-
"loss": 3.7243,
|
| 262 |
-
"step": 1800
|
| 263 |
-
},
|
| 264 |
-
{
|
| 265 |
-
"epoch": 4.147982062780269,
|
| 266 |
-
"grad_norm": 6.21875,
|
| 267 |
-
"learning_rate": 3.4411084078980574e-05,
|
| 268 |
-
"loss": 3.3222,
|
| 269 |
-
"step": 1850
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 4.260089686098655,
|
| 273 |
-
"grad_norm": 6.65625,
|
| 274 |
-
"learning_rate": 3.353680877362694e-05,
|
| 275 |
-
"loss": 3.3072,
|
| 276 |
-
"step": 1900
|
| 277 |
-
},
|
| 278 |
-
{
|
| 279 |
-
"epoch": 4.37219730941704,
|
| 280 |
-
"grad_norm": 6.75,
|
| 281 |
-
"learning_rate": 3.265054836052488e-05,
|
| 282 |
-
"loss": 3.2928,
|
| 283 |
-
"step": 1950
|
| 284 |
-
},
|
| 285 |
-
{
|
| 286 |
-
"epoch": 4.484304932735426,
|
| 287 |
-
"grad_norm": 7.15625,
|
| 288 |
-
"learning_rate": 3.1753547089935345e-05,
|
| 289 |
-
"loss": 3.2997,
|
| 290 |
-
"step": 2000
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 4.596412556053812,
|
| 294 |
-
"grad_norm": 6.71875,
|
| 295 |
-
"learning_rate": 3.084706429156379e-05,
|
| 296 |
-
"loss": 3.3155,
|
| 297 |
-
"step": 2050
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 4.708520179372197,
|
| 301 |
-
"grad_norm": 5.5625,
|
| 302 |
-
"learning_rate": 2.9932372606545638e-05,
|
| 303 |
-
"loss": 3.3295,
|
| 304 |
-
"step": 2100
|
| 305 |
-
},
|
| 306 |
-
{
|
| 307 |
-
"epoch": 4.820627802690583,
|
| 308 |
-
"grad_norm": 6.65625,
|
| 309 |
-
"learning_rate": 2.9010756200743363e-05,
|
| 310 |
-
"loss": 3.3681,
|
| 311 |
-
"step": 2150
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 4.932735426008969,
|
| 315 |
-
"grad_norm": 6.40625,
|
| 316 |
-
"learning_rate": 2.808350896186362e-05,
|
| 317 |
-
"loss": 3.3639,
|
| 318 |
-
"step": 2200
|
| 319 |
-
},
|
| 320 |
-
{
|
| 321 |
-
"epoch": 5.044843049327354,
|
| 322 |
-
"grad_norm": 7.875,
|
| 323 |
-
"learning_rate": 2.7151932682925563e-05,
|
| 324 |
-
"loss": 3.115,
|
| 325 |
-
"step": 2250
|
| 326 |
-
},
|
| 327 |
-
{
|
| 328 |
-
"epoch": 5.15695067264574,
|
| 329 |
-
"grad_norm": 10.5,
|
| 330 |
-
"learning_rate": 2.6217335234630654e-05,
|
| 331 |
-
"loss": 2.8213,
|
| 332 |
-
"step": 2300
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 5.2690582959641254,
|
| 336 |
-
"grad_norm": 7.84375,
|
| 337 |
-
"learning_rate": 2.5281028729199775e-05,
|
| 338 |
-
"loss": 2.7705,
|
| 339 |
-
"step": 2350
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 5.381165919282511,
|
| 343 |
-
"grad_norm": 9.6875,
|
| 344 |
-
"learning_rate": 2.4344327678255555e-05,
|
| 345 |
-
"loss": 2.7746,
|
| 346 |
-
"step": 2400
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 5.493273542600897,
|
| 350 |
-
"grad_norm": 6.8125,
|
| 351 |
-
"learning_rate": 2.34085471473361e-05,
|
| 352 |
-
"loss": 2.7827,
|
| 353 |
-
"step": 2450
|
| 354 |
-
},
|
| 355 |
-
{
|
| 356 |
-
"epoch": 5.605381165919282,
|
| 357 |
-
"grad_norm": 8.8125,
|
| 358 |
-
"learning_rate": 2.2475000909631068e-05,
|
| 359 |
-
"loss": 2.8412,
|
| 360 |
-
"step": 2500
|
| 361 |
-
},
|
| 362 |
-
{
|
| 363 |
-
"epoch": 5.7174887892376685,
|
| 364 |
-
"grad_norm": 7.78125,
|
| 365 |
-
"learning_rate": 2.1544999601532084e-05,
|
| 366 |
-
"loss": 2.7982,
|
| 367 |
-
"step": 2550
|
| 368 |
-
},
|
| 369 |
-
{
|
| 370 |
-
"epoch": 5.829596412556054,
|
| 371 |
-
"grad_norm": 8.1875,
|
| 372 |
-
"learning_rate": 2.0619848882587013e-05,
|
| 373 |
-
"loss": 2.8419,
|
| 374 |
-
"step": 2600
|
| 375 |
-
},
|
| 376 |
-
{
|
| 377 |
-
"epoch": 5.941704035874439,
|
| 378 |
-
"grad_norm": 8.4375,
|
| 379 |
-
"learning_rate": 1.9700847602441465e-05,
|
| 380 |
-
"loss": 2.7266,
|
| 381 |
-
"step": 2650
|
| 382 |
-
},
|
| 383 |
-
{
|
| 384 |
-
"epoch": 6.053811659192825,
|
| 385 |
-
"grad_norm": 8.25,
|
| 386 |
-
"learning_rate": 1.878928597734082e-05,
|
| 387 |
-
"loss": 2.6429,
|
| 388 |
-
"step": 2700
|
| 389 |
-
},
|
| 390 |
-
{
|
| 391 |
-
"epoch": 6.165919282511211,
|
| 392 |
-
"grad_norm": 8.25,
|
| 393 |
-
"learning_rate": 1.7886443778753052e-05,
|
| 394 |
-
"loss": 2.4437,
|
| 395 |
-
"step": 2750
|
| 396 |
-
},
|
| 397 |
-
{
|
| 398 |
-
"epoch": 6.278026905829597,
|
| 399 |
-
"grad_norm": 9.75,
|
| 400 |
-
"learning_rate": 1.699358853665535e-05,
|
| 401 |
-
"loss": 2.3789,
|
| 402 |
-
"step": 2800
|
| 403 |
-
},
|
| 404 |
-
{
|
| 405 |
-
"epoch": 6.390134529147982,
|
| 406 |
-
"grad_norm": 8.75,
|
| 407 |
-
"learning_rate": 1.6111973760006838e-05,
|
| 408 |
-
"loss": 2.4337,
|
| 409 |
-
"step": 2850
|
| 410 |
-
},
|
| 411 |
-
{
|
| 412 |
-
"epoch": 6.502242152466367,
|
| 413 |
-
"grad_norm": 7.6875,
|
| 414 |
-
"learning_rate": 1.5242837176906038e-05,
|
| 415 |
-
"loss": 2.4162,
|
| 416 |
-
"step": 2900
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 6.614349775784754,
|
| 420 |
-
"grad_norm": 9.875,
|
| 421 |
-
"learning_rate": 1.4387398996903488e-05,
|
| 422 |
-
"loss": 2.4613,
|
| 423 |
-
"step": 2950
|
| 424 |
-
},
|
| 425 |
-
{
|
| 426 |
-
"epoch": 6.726457399103139,
|
| 427 |
-
"grad_norm": 9.5,
|
| 428 |
-
"learning_rate": 1.3546860197909212e-05,
|
| 429 |
-
"loss": 2.4606,
|
| 430 |
-
"step": 3000
|
| 431 |
-
},
|
| 432 |
-
{
|
| 433 |
-
"epoch": 6.838565022421525,
|
| 434 |
-
"grad_norm": 9.875,
|
| 435 |
-
"learning_rate": 1.2722400840100257e-05,
|
| 436 |
-
"loss": 2.4374,
|
| 437 |
-
"step": 3050
|
| 438 |
-
},
|
| 439 |
-
{
|
| 440 |
-
"epoch": 6.95067264573991,
|
| 441 |
-
"grad_norm": 9.8125,
|
| 442 |
-
"learning_rate": 1.1915178409195171e-05,
|
| 443 |
-
"loss": 2.4009,
|
| 444 |
-
"step": 3100
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 7.062780269058296,
|
| 448 |
-
"grad_norm": 8.875,
|
| 449 |
-
"learning_rate": 1.1126326191421625e-05,
|
| 450 |
-
"loss": 2.3286,
|
| 451 |
-
"step": 3150
|
| 452 |
-
},
|
| 453 |
-
{
|
| 454 |
-
"epoch": 7.174887892376682,
|
| 455 |
-
"grad_norm": 8.25,
|
| 456 |
-
"learning_rate": 1.035695168245843e-05,
|
| 457 |
-
"loss": 2.2594,
|
| 458 |
-
"step": 3200
|
| 459 |
-
},
|
| 460 |
-
{
|
| 461 |
-
"epoch": 7.286995515695067,
|
| 462 |
-
"grad_norm": 8.1875,
|
| 463 |
-
"learning_rate": 9.608135032585758e-06,
|
| 464 |
-
"loss": 2.2772,
|
| 465 |
-
"step": 3250
|
| 466 |
-
},
|
| 467 |
-
{
|
| 468 |
-
"epoch": 7.3991031390134525,
|
| 469 |
-
"grad_norm": 8.25,
|
| 470 |
-
"learning_rate": 8.88092753022657e-06,
|
| 471 |
-
"loss": 2.2576,
|
| 472 |
-
"step": 3300
|
| 473 |
-
},
|
| 474 |
-
{
|
| 475 |
-
"epoch": 7.511210762331839,
|
| 476 |
-
"grad_norm": 8.1875,
|
| 477 |
-
"learning_rate": 8.176350126008015e-06,
|
| 478 |
-
"loss": 2.2233,
|
| 479 |
-
"step": 3350
|
| 480 |
-
},
|
| 481 |
-
{
|
| 482 |
-
"epoch": 7.623318385650224,
|
| 483 |
-
"grad_norm": 10.4375,
|
| 484 |
-
"learning_rate": 7.49539199941511e-06,
|
| 485 |
-
"loss": 2.2273,
|
| 486 |
-
"step": 3400
|
| 487 |
-
},
|
| 488 |
-
{
|
| 489 |
-
"epoch": 7.73542600896861,
|
| 490 |
-
"grad_norm": 9.625,
|
| 491 |
-
"learning_rate": 6.839009170049096e-06,
|
| 492 |
-
"loss": 2.2692,
|
| 493 |
-
"step": 3450
|
| 494 |
-
},
|
| 495 |
-
{
|
| 496 |
-
"epoch": 7.8475336322869955,
|
| 497 |
-
"grad_norm": 9.0,
|
| 498 |
-
"learning_rate": 6.208123155439854e-06,
|
| 499 |
-
"loss": 2.2847,
|
| 500 |
-
"step": 3500
|
| 501 |
-
},
|
| 502 |
-
{
|
| 503 |
-
"epoch": 7.959641255605381,
|
| 504 |
-
"grad_norm": 8.875,
|
| 505 |
-
"learning_rate": 5.603619677297028e-06,
|
| 506 |
-
"loss": 2.251,
|
| 507 |
-
"step": 3550
|
| 508 |
-
},
|
| 509 |
-
{
|
| 510 |
-
"epoch": 8.071748878923767,
|
| 511 |
-
"grad_norm": 8.6875,
|
| 512 |
-
"learning_rate": 5.026347418016134e-06,
|
| 513 |
-
"loss": 2.2404,
|
| 514 |
-
"step": 3600
|
| 515 |
-
},
|
| 516 |
-
{
|
| 517 |
-
"epoch": 8.183856502242152,
|
| 518 |
-
"grad_norm": 8.4375,
|
| 519 |
-
"learning_rate": 4.477116829185235e-06,
|
| 520 |
-
"loss": 2.2465,
|
| 521 |
-
"step": 3650
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"epoch": 8.295964125560538,
|
| 525 |
-
"grad_norm": 8.9375,
|
| 526 |
-
"learning_rate": 3.956698993765226e-06,
|
| 527 |
-
"loss": 2.2267,
|
| 528 |
-
"step": 3700
|
| 529 |
-
},
|
| 530 |
-
{
|
| 531 |
-
"epoch": 8.408071748878923,
|
| 532 |
-
"grad_norm": 8.625,
|
| 533 |
-
"learning_rate": 3.4658245435410407e-06,
|
| 534 |
-
"loss": 2.1912,
|
| 535 |
-
"step": 3750
|
| 536 |
-
},
|
| 537 |
-
{
|
| 538 |
-
"epoch": 8.52017937219731,
|
| 539 |
-
"grad_norm": 9.5,
|
| 540 |
-
"learning_rate": 3.0051826333634818e-06,
|
| 541 |
-
"loss": 2.2076,
|
| 542 |
-
"step": 3800
|
| 543 |
-
},
|
| 544 |
-
{
|
| 545 |
-
"epoch": 8.632286995515695,
|
| 546 |
-
"grad_norm": 8.5,
|
| 547 |
-
"learning_rate": 2.5754199736220312e-06,
|
| 548 |
-
"loss": 2.2122,
|
| 549 |
-
"step": 3850
|
| 550 |
-
},
|
| 551 |
-
{
|
| 552 |
-
"epoch": 8.74439461883408,
|
| 553 |
-
"grad_norm": 9.3125,
|
| 554 |
-
"learning_rate": 2.177139922306773e-06,
|
| 555 |
-
"loss": 2.1805,
|
| 556 |
-
"step": 3900
|
| 557 |
-
},
|
| 558 |
-
{
|
| 559 |
-
"epoch": 8.856502242152466,
|
| 560 |
-
"grad_norm": 9.125,
|
| 561 |
-
"learning_rate": 1.810901637934137e-06,
|
| 562 |
-
"loss": 2.1735,
|
| 563 |
-
"step": 3950
|
| 564 |
-
},
|
| 565 |
-
{
|
| 566 |
-
"epoch": 8.968609865470851,
|
| 567 |
-
"grad_norm": 9.125,
|
| 568 |
-
"learning_rate": 1.4772192945258528e-06,
|
| 569 |
-
"loss": 2.186,
|
| 570 |
-
"step": 4000
|
| 571 |
-
},
|
| 572 |
-
{
|
| 573 |
-
"epoch": 9.080717488789238,
|
| 574 |
-
"grad_norm": 9.0,
|
| 575 |
-
"learning_rate": 1.1765613597430309e-06,
|
| 576 |
-
"loss": 2.1942,
|
| 577 |
-
"step": 4050
|
| 578 |
-
},
|
| 579 |
-
{
|
| 580 |
-
"epoch": 9.192825112107624,
|
| 581 |
-
"grad_norm": 8.4375,
|
| 582 |
-
"learning_rate": 9.093499371889385e-07,
|
| 583 |
-
"loss": 2.2041,
|
| 584 |
-
"step": 4100
|
| 585 |
-
},
|
| 586 |
-
{
|
| 587 |
-
"epoch": 9.304932735426009,
|
| 588 |
-
"grad_norm": 8.75,
|
| 589 |
-
"learning_rate": 6.75960173803819e-07,
|
| 590 |
-
"loss": 2.1515,
|
| 591 |
-
"step": 4150
|
| 592 |
-
},
|
| 593 |
-
{
|
| 594 |
-
"epoch": 9.417040358744394,
|
| 595 |
-
"grad_norm": 8.6875,
|
| 596 |
-
"learning_rate": 4.767197331837364e-07,
|
| 597 |
-
"loss": 2.1865,
|
| 598 |
-
"step": 4200
|
| 599 |
-
},
|
| 600 |
-
{
|
| 601 |
-
"epoch": 9.52914798206278,
|
| 602 |
-
"grad_norm": 8.875,
|
| 603 |
-
"learning_rate": 3.1190833556278276e-07,
|
| 604 |
-
"loss": 2.1974,
|
| 605 |
-
"step": 4250
|
| 606 |
-
},
|
| 607 |
-
{
|
| 608 |
-
"epoch": 9.641255605381167,
|
| 609 |
-
"grad_norm": 9.0625,
|
| 610 |
-
"learning_rate": 1.8175736510465114e-07,
|
| 611 |
-
"loss": 2.1827,
|
| 612 |
-
"step": 4300
|
| 613 |
-
},
|
| 614 |
-
{
|
| 615 |
-
"epoch": 9.753363228699552,
|
| 616 |
-
"grad_norm": 8.875,
|
| 617 |
-
"learning_rate": 8.644954505474812e-08,
|
| 618 |
-
"loss": 2.1896,
|
| 619 |
-
"step": 4350
|
| 620 |
-
},
|
| 621 |
-
{
|
| 622 |
-
"epoch": 9.865470852017937,
|
| 623 |
-
"grad_norm": 8.6875,
|
| 624 |
-
"learning_rate": 2.611868120898919e-08,
|
| 625 |
-
"loss": 2.2084,
|
| 626 |
-
"step": 4400
|
| 627 |
-
},
|
| 628 |
-
{
|
| 629 |
-
"epoch": 9.977578475336323,
|
| 630 |
-
"grad_norm": 8.875,
|
| 631 |
-
"learning_rate": 8.494740594333639e-10,
|
| 632 |
-
"loss": 2.2342,
|
| 633 |
-
"step": 4450
|
| 634 |
}
|
| 635 |
],
|
| 636 |
"logging_steps": 50,
|
| 637 |
-
"max_steps":
|
| 638 |
"num_input_tokens_seen": 0,
|
| 639 |
-
"num_train_epochs":
|
| 640 |
"save_steps": 500,
|
| 641 |
"stateful_callbacks": {
|
| 642 |
"TrainerControl": {
|
|
@@ -645,12 +100,12 @@
|
|
| 645 |
"should_evaluate": false,
|
| 646 |
"should_log": false,
|
| 647 |
"should_save": true,
|
| 648 |
-
"should_training_stop":
|
| 649 |
},
|
| 650 |
"attributes": {}
|
| 651 |
}
|
| 652 |
},
|
| 653 |
-
"total_flos": 1.
|
| 654 |
"train_batch_size": 1,
|
| 655 |
"trial_name": null,
|
| 656 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 500,
|
| 3 |
+
"best_metric": 1.3274219036102295,
|
| 4 |
+
"best_model_checkpoint": "./orpheus-turkish-emotion-finetune/checkpoint-500",
|
| 5 |
+
"epoch": 2.4884735202492214,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"epoch": 0.24922118380062305,
|
| 14 |
+
"grad_norm": 100.0,
|
| 15 |
+
"learning_rate": 1.218905472636816e-06,
|
| 16 |
+
"loss": 9.7783,
|
| 17 |
"step": 50
|
| 18 |
},
|
| 19 |
{
|
| 20 |
+
"epoch": 0.4984423676012461,
|
| 21 |
+
"grad_norm": 74.5,
|
| 22 |
+
"learning_rate": 2.4626865671641794e-06,
|
| 23 |
+
"loss": 9.5155,
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 0.7476635514018691,
|
| 28 |
+
"grad_norm": 74.0,
|
| 29 |
+
"learning_rate": 3.706467661691542e-06,
|
| 30 |
+
"loss": 9.0351,
|
| 31 |
"step": 150
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"epoch": 0.9968847352024922,
|
| 35 |
+
"grad_norm": 80.5,
|
| 36 |
+
"learning_rate": 4.950248756218906e-06,
|
| 37 |
+
"loss": 8.4023,
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 1.2442367601246107,
|
| 42 |
+
"grad_norm": 80.0,
|
| 43 |
+
"learning_rate": 6.194029850746269e-06,
|
| 44 |
+
"loss": 7.3784,
|
| 45 |
"step": 250
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 1.4934579439252336,
|
| 49 |
+
"grad_norm": 134.0,
|
| 50 |
+
"learning_rate": 7.437810945273633e-06,
|
| 51 |
+
"loss": 5.7507,
|
| 52 |
"step": 300
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 1.7426791277258566,
|
| 56 |
+
"grad_norm": 65.0,
|
| 57 |
+
"learning_rate": 8.681592039800995e-06,
|
| 58 |
+
"loss": 3.8008,
|
| 59 |
"step": 350
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 1.9919003115264797,
|
| 63 |
+
"grad_norm": 17.0,
|
| 64 |
+
"learning_rate": 9.925373134328359e-06,
|
| 65 |
+
"loss": 2.103,
|
| 66 |
"step": 400
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 2.2392523364485983,
|
| 70 |
+
"grad_norm": 1.65625,
|
| 71 |
+
"learning_rate": 9.995836696556696e-06,
|
| 72 |
+
"loss": 1.4184,
|
| 73 |
"step": 450
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 2.4884735202492214,
|
| 77 |
+
"grad_norm": 0.84765625,
|
| 78 |
+
"learning_rate": 9.982274873915892e-06,
|
| 79 |
+
"loss": 1.2978,
|
| 80 |
"step": 500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 2.4884735202492214,
|
| 84 |
+
"eval_loss": 1.3274219036102295,
|
| 85 |
+
"eval_runtime": 15.5223,
|
| 86 |
+
"eval_samples_per_second": 11.532,
|
| 87 |
+
"eval_steps_per_second": 1.482,
|
| 88 |
+
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
}
|
| 90 |
],
|
| 91 |
"logging_steps": 50,
|
| 92 |
+
"max_steps": 4020,
|
| 93 |
"num_input_tokens_seen": 0,
|
| 94 |
+
"num_train_epochs": 20,
|
| 95 |
"save_steps": 500,
|
| 96 |
"stateful_callbacks": {
|
| 97 |
"TrainerControl": {
|
|
|
|
| 100 |
"should_evaluate": false,
|
| 101 |
"should_log": false,
|
| 102 |
"should_save": true,
|
| 103 |
+
"should_training_stop": false
|
| 104 |
},
|
| 105 |
"attributes": {}
|
| 106 |
}
|
| 107 |
},
|
| 108 |
+
"total_flos": 1.3833925120386662e+17,
|
| 109 |
"train_batch_size": 1,
|
| 110 |
"trial_name": null,
|
| 111 |
"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 5777
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e91f58a9a988419f219f097fff3f0e1762f623e4b2b1a8cf942cacee3271dc13
|
| 3 |
size 5777
|