Training in progress, step 1700, checkpoint
Browse files
checkpoint-1700/adapter_config.json
CHANGED
|
@@ -23,13 +23,13 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"
|
| 27 |
-
"o_proj",
|
| 28 |
-
"v_proj",
|
| 29 |
-
"q_proj",
|
| 30 |
"down_proj",
|
|
|
|
| 31 |
"k_proj",
|
| 32 |
-
"
|
|
|
|
|
|
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
+
"gate_proj",
|
|
|
|
|
|
|
|
|
|
| 27 |
"down_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
"k_proj",
|
| 30 |
+
"v_proj",
|
| 31 |
+
"o_proj",
|
| 32 |
+
"up_proj"
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
checkpoint-1700/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 167832240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7357333e3e91412df976df3d250390e0fd30236c33b7eaf3bb51283a15a6eb2a
|
| 3 |
size 167832240
|
checkpoint-1700/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 335922386
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e309d64c0bb627b59fd4d34f925d0843fde35efeaf0e7fc7f2df0dd52d4df7f
|
| 3 |
size 335922386
|
checkpoint-1700/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 sha256:117282b20d380d2e100b4d897d919c7db6f76b0af924bed1116f73baccdd725d
|
| 3 |
size 14244
|
checkpoint-1700/trainer_state.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.7913593256059008,
|
| 5 |
-
"eval_steps":
|
| 6 |
"global_step": 1700,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
|
@@ -10,1464 +10,2552 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01053740779768177,
|
| 13 |
-
"grad_norm": 3.
|
| 14 |
"learning_rate": 4.982437653670531e-05,
|
| 15 |
-
"loss": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.02107481559536354,
|
| 20 |
-
"grad_norm": 1.
|
| 21 |
"learning_rate": 4.964875307341061e-05,
|
| 22 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03161222339304531,
|
| 27 |
-
"grad_norm": 1.
|
| 28 |
"learning_rate": 4.947312961011591e-05,
|
| 29 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.04214963119072708,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
"learning_rate": 4.929750614682122e-05,
|
| 36 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.05268703898840885,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
"learning_rate": 4.9121882683526524e-05,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.05268703898840885,
|
| 48 |
-
"eval_loss": 0.
|
| 49 |
-
"eval_runtime": 13.
|
| 50 |
-
"eval_samples_per_second": 35.
|
| 51 |
-
"eval_steps_per_second": 2.
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.06322444678609063,
|
| 56 |
-
"grad_norm": 0.
|
| 57 |
"learning_rate": 4.894625922023183e-05,
|
| 58 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
"step": 60
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.0737618545837724,
|
| 63 |
-
"grad_norm": 0.
|
| 64 |
"learning_rate": 4.877063575693713e-05,
|
| 65 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
"step": 70
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.08429926238145416,
|
| 70 |
-
"grad_norm": 0.
|
| 71 |
"learning_rate": 4.8595012293642434e-05,
|
| 72 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
"step": 80
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.09483667017913593,
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
"learning_rate": 4.841938883034774e-05,
|
| 79 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
"step": 90
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.1053740779768177,
|
| 84 |
-
"grad_norm": 0.
|
| 85 |
"learning_rate": 4.824376536705304e-05,
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.1053740779768177,
|
| 91 |
-
"eval_loss": 0.
|
| 92 |
-
"eval_runtime": 13.
|
| 93 |
-
"eval_samples_per_second": 35.
|
| 94 |
-
"eval_steps_per_second": 2.
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.11591148577449947,
|
| 99 |
-
"grad_norm": 0.
|
| 100 |
"learning_rate": 4.8068141903758344e-05,
|
| 101 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
"step": 110
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.12644889357218125,
|
| 106 |
-
"grad_norm": 0.
|
| 107 |
"learning_rate": 4.789251844046364e-05,
|
| 108 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 109 |
"step": 120
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.136986301369863,
|
| 113 |
-
"grad_norm": 0.
|
| 114 |
"learning_rate": 4.7716894977168955e-05,
|
| 115 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
"step": 130
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.1475237091675448,
|
| 120 |
-
"grad_norm": 0.
|
| 121 |
"learning_rate": 4.754127151387426e-05,
|
| 122 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
"step": 140
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.15806111696522657,
|
| 127 |
-
"grad_norm": 0.
|
| 128 |
"learning_rate": 4.736564805057956e-05,
|
| 129 |
"loss": 0.317,
|
| 130 |
"step": 150
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.15806111696522657,
|
| 134 |
-
"eval_loss": 0.
|
| 135 |
-
"eval_runtime": 13.
|
| 136 |
-
"eval_samples_per_second": 35.
|
| 137 |
-
"eval_steps_per_second": 2.
|
| 138 |
"step": 150
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 0.16859852476290832,
|
| 142 |
-
"grad_norm": 0.
|
| 143 |
"learning_rate": 4.7190024587284866e-05,
|
| 144 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 145 |
"step": 160
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 0.1791359325605901,
|
| 149 |
-
"grad_norm": 0.
|
| 150 |
"learning_rate": 4.7014401123990165e-05,
|
| 151 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 152 |
"step": 170
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 0.18967334035827185,
|
| 156 |
-
"grad_norm": 0.
|
| 157 |
"learning_rate": 4.683877766069547e-05,
|
| 158 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 159 |
"step": 180
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 0.20021074815595363,
|
| 163 |
-
"grad_norm": 0.
|
| 164 |
"learning_rate": 4.6663154197400776e-05,
|
| 165 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 166 |
"step": 190
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 0.2107481559536354,
|
| 170 |
-
"grad_norm": 0.
|
| 171 |
"learning_rate": 4.6487530734106075e-05,
|
| 172 |
-
"loss": 0.
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.2107481559536354,
|
| 177 |
-
"eval_loss": 0.
|
| 178 |
-
"eval_runtime": 13.
|
| 179 |
-
"eval_samples_per_second": 35.
|
| 180 |
-
"eval_steps_per_second": 2.
|
| 181 |
"step": 200
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.22128556375131717,
|
| 185 |
-
"grad_norm": 0.
|
| 186 |
"learning_rate": 4.631190727081138e-05,
|
| 187 |
"loss": 0.2924,
|
| 188 |
"step": 210
|
| 189 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 190 |
{
|
| 191 |
"epoch": 0.23182297154899895,
|
| 192 |
-
"grad_norm": 0.
|
| 193 |
"learning_rate": 4.6136283807516686e-05,
|
| 194 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 195 |
"step": 220
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 0.24236037934668073,
|
| 199 |
-
"grad_norm": 0.
|
| 200 |
"learning_rate": 4.596066034422199e-05,
|
| 201 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
"step": 230
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 0.2528977871443625,
|
| 206 |
-
"grad_norm": 0.
|
| 207 |
"learning_rate": 4.57850368809273e-05,
|
| 208 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 209 |
"step": 240
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.26343519494204426,
|
| 213 |
-
"grad_norm": 0.
|
| 214 |
"learning_rate": 4.5609413417632596e-05,
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 250
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 0.26343519494204426,
|
| 220 |
-
"eval_loss": 0.
|
| 221 |
-
"eval_runtime": 13.
|
| 222 |
-
"eval_samples_per_second": 35.
|
| 223 |
-
"eval_steps_per_second": 2.
|
| 224 |
"step": 250
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.273972602739726,
|
| 228 |
-
"grad_norm": 0.
|
| 229 |
"learning_rate": 4.54337899543379e-05,
|
| 230 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 231 |
"step": 260
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 0.2845100105374078,
|
| 235 |
-
"grad_norm": 0.
|
| 236 |
"learning_rate": 4.525816649104321e-05,
|
| 237 |
"loss": 0.3112,
|
| 238 |
"step": 270
|
| 239 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 240 |
{
|
| 241 |
"epoch": 0.2950474183350896,
|
| 242 |
-
"grad_norm": 0.
|
| 243 |
"learning_rate": 4.5082543027748506e-05,
|
| 244 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 245 |
"step": 280
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 0.3055848261327713,
|
| 249 |
-
"grad_norm": 0.
|
| 250 |
"learning_rate": 4.490691956445381e-05,
|
| 251 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 252 |
"step": 290
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 0.31612223393045313,
|
| 256 |
-
"grad_norm": 0.
|
| 257 |
"learning_rate": 4.473129610115912e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 300
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 0.31612223393045313,
|
| 263 |
-
"eval_loss": 0.
|
| 264 |
-
"eval_runtime": 13.
|
| 265 |
-
"eval_samples_per_second": 35.
|
| 266 |
-
"eval_steps_per_second": 2.
|
| 267 |
"step": 300
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 0.3266596417281349,
|
| 271 |
-
"grad_norm": 0.
|
| 272 |
"learning_rate": 4.455567263786442e-05,
|
| 273 |
"loss": 0.3084,
|
| 274 |
"step": 310
|
| 275 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 276 |
{
|
| 277 |
"epoch": 0.33719704952581664,
|
| 278 |
-
"grad_norm": 0.
|
| 279 |
"learning_rate": 4.438004917456973e-05,
|
| 280 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 281 |
"step": 320
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 0.34773445732349845,
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
"learning_rate": 4.420442571127503e-05,
|
| 287 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 288 |
"step": 330
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 0.3582718651211802,
|
| 292 |
-
"grad_norm": 0.
|
| 293 |
"learning_rate": 4.4028802247980333e-05,
|
| 294 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 295 |
"step": 340
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 0.36880927291886195,
|
| 299 |
-
"grad_norm": 0.
|
| 300 |
"learning_rate": 4.385317878468563e-05,
|
| 301 |
"loss": 0.3084,
|
| 302 |
"step": 350
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 0.36880927291886195,
|
| 306 |
-
"eval_loss": 0.
|
| 307 |
-
"eval_runtime": 13.
|
| 308 |
-
"eval_samples_per_second": 35.
|
| 309 |
-
"eval_steps_per_second": 2.
|
| 310 |
"step": 350
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.3793466807165437,
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
"learning_rate": 4.367755532139094e-05,
|
| 316 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 317 |
"step": 360
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.3898840885142255,
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
"learning_rate": 4.3501931858096244e-05,
|
| 323 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
"step": 370
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.40042149631190727,
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
"learning_rate": 4.332630839480154e-05,
|
| 330 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 331 |
"step": 380
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.410958904109589,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 4.3150684931506855e-05,
|
| 337 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 338 |
"step": 390
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.4214963119072708,
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
"learning_rate": 4.297506146821216e-05,
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 400
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.4214963119072708,
|
| 349 |
-
"eval_loss": 0.
|
| 350 |
-
"eval_runtime": 13.
|
| 351 |
-
"eval_samples_per_second": 35.
|
| 352 |
-
"eval_steps_per_second": 2.
|
| 353 |
"step": 400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.4320337197049526,
|
| 357 |
-
"grad_norm": 0.
|
| 358 |
"learning_rate": 4.279943800491746e-05,
|
| 359 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 360 |
"step": 410
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 0.44257112750263433,
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
"learning_rate": 4.2623814541622765e-05,
|
| 366 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 367 |
"step": 420
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 0.45310853530031614,
|
| 371 |
-
"grad_norm": 0.
|
| 372 |
"learning_rate": 4.2448191078328064e-05,
|
| 373 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 374 |
"step": 430
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.4636459430979979,
|
| 378 |
-
"grad_norm": 0.
|
| 379 |
"learning_rate": 4.227256761503337e-05,
|
| 380 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 381 |
"step": 440
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 0.47418335089567965,
|
| 385 |
-
"grad_norm": 0.
|
| 386 |
"learning_rate": 4.2096944151738675e-05,
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 450
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 0.47418335089567965,
|
| 392 |
-
"eval_loss": 0.
|
| 393 |
-
"eval_runtime": 13.
|
| 394 |
-
"eval_samples_per_second": 35.
|
| 395 |
-
"eval_steps_per_second": 2.
|
| 396 |
"step": 450
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 0.48472075869336145,
|
| 400 |
-
"grad_norm": 0.
|
| 401 |
"learning_rate": 4.1921320688443974e-05,
|
| 402 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 403 |
"step": 460
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 0.4952581664910432,
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
"learning_rate": 4.174569722514928e-05,
|
| 409 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 410 |
"step": 470
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 0.505795574288725,
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
"learning_rate": 4.1570073761854585e-05,
|
| 416 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 417 |
"step": 480
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.5163329820864068,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
"learning_rate": 4.139445029855989e-05,
|
| 423 |
"loss": 0.2911,
|
| 424 |
"step": 490
|
| 425 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 426 |
{
|
| 427 |
"epoch": 0.5268703898840885,
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
"learning_rate": 4.12188268352652e-05,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 500
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.5268703898840885,
|
| 435 |
-
"eval_loss": 0.
|
| 436 |
-
"eval_runtime": 13.
|
| 437 |
-
"eval_samples_per_second": 35.
|
| 438 |
-
"eval_steps_per_second": 2.
|
| 439 |
"step": 500
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 0.5374077976817703,
|
| 443 |
-
"grad_norm": 0.
|
| 444 |
"learning_rate": 4.1043203371970496e-05,
|
| 445 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 446 |
"step": 510
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 0.547945205479452,
|
| 450 |
-
"grad_norm": 0.
|
| 451 |
"learning_rate": 4.08675799086758e-05,
|
| 452 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 453 |
"step": 520
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 0.5584826132771338,
|
| 457 |
-
"grad_norm": 0.
|
| 458 |
"learning_rate": 4.069195644538111e-05,
|
| 459 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 460 |
"step": 530
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 0.5690200210748156,
|
| 464 |
-
"grad_norm": 0.
|
| 465 |
"learning_rate": 4.0516332982086406e-05,
|
| 466 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 467 |
"step": 540
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 0.5795574288724974,
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
"learning_rate": 4.034070951879171e-05,
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 550
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 0.5795574288724974,
|
| 478 |
-
"eval_loss": 0.
|
| 479 |
-
"eval_runtime": 13.
|
| 480 |
-
"eval_samples_per_second": 35.
|
| 481 |
-
"eval_steps_per_second": 2.
|
| 482 |
"step": 550
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 0.5900948366701791,
|
| 486 |
-
"grad_norm": 0.
|
| 487 |
"learning_rate": 4.016508605549702e-05,
|
| 488 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 489 |
"step": 560
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 0.6006322444678609,
|
| 493 |
-
"grad_norm": 0.
|
| 494 |
"learning_rate": 3.998946259220232e-05,
|
| 495 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 496 |
"step": 570
|
| 497 |
},
|
| 498 |
{
|
| 499 |
"epoch": 0.6111696522655427,
|
| 500 |
-
"grad_norm": 0.
|
| 501 |
"learning_rate": 3.981383912890763e-05,
|
| 502 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 503 |
"step": 580
|
| 504 |
},
|
| 505 |
{
|
| 506 |
"epoch": 0.6217070600632244,
|
| 507 |
-
"grad_norm": 0.
|
| 508 |
"learning_rate": 3.963821566561293e-05,
|
| 509 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 510 |
"step": 590
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 0.6322444678609063,
|
| 514 |
-
"grad_norm": 0.
|
| 515 |
"learning_rate": 3.946259220231823e-05,
|
| 516 |
-
"loss": 0.
|
| 517 |
"step": 600
|
| 518 |
},
|
| 519 |
{
|
| 520 |
"epoch": 0.6322444678609063,
|
| 521 |
-
"eval_loss": 0.
|
| 522 |
-
"eval_runtime": 13.
|
| 523 |
-
"eval_samples_per_second": 35.
|
| 524 |
-
"eval_steps_per_second": 2.
|
| 525 |
"step": 600
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 0.642781875658588,
|
| 529 |
-
"grad_norm": 0.
|
| 530 |
"learning_rate": 3.928696873902353e-05,
|
| 531 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 532 |
"step": 610
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 0.6533192834562698,
|
| 536 |
-
"grad_norm": 0.
|
| 537 |
"learning_rate": 3.911134527572884e-05,
|
| 538 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 539 |
"step": 620
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 0.6638566912539515,
|
| 543 |
-
"grad_norm": 0.
|
| 544 |
"learning_rate": 3.893572181243414e-05,
|
| 545 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 546 |
"step": 630
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 0.6743940990516333,
|
| 550 |
-
"grad_norm": 0.
|
| 551 |
"learning_rate": 3.876009834913945e-05,
|
| 552 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 553 |
"step": 640
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 0.684931506849315,
|
| 557 |
-
"grad_norm": 0.
|
| 558 |
"learning_rate": 3.8584474885844754e-05,
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 650
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 0.684931506849315,
|
| 564 |
-
"eval_loss": 0.
|
| 565 |
-
"eval_runtime": 13.
|
| 566 |
-
"eval_samples_per_second": 35.
|
| 567 |
-
"eval_steps_per_second": 2.
|
| 568 |
"step": 650
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 0.6954689146469969,
|
| 572 |
-
"grad_norm": 0.
|
| 573 |
"learning_rate": 3.840885142255005e-05,
|
| 574 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 575 |
"step": 660
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 0.7060063224446786,
|
| 579 |
-
"grad_norm": 0.
|
| 580 |
"learning_rate": 3.823322795925536e-05,
|
| 581 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 582 |
"step": 670
|
| 583 |
},
|
| 584 |
{
|
| 585 |
"epoch": 0.7165437302423604,
|
| 586 |
-
"grad_norm": 0.
|
| 587 |
"learning_rate": 3.8057604495960664e-05,
|
| 588 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 589 |
"step": 680
|
| 590 |
},
|
| 591 |
{
|
| 592 |
"epoch": 0.7270811380400422,
|
| 593 |
-
"grad_norm": 0.
|
| 594 |
"learning_rate": 3.788198103266596e-05,
|
| 595 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 596 |
"step": 690
|
| 597 |
},
|
| 598 |
{
|
| 599 |
"epoch": 0.7376185458377239,
|
| 600 |
-
"grad_norm": 0.
|
| 601 |
"learning_rate": 3.770635756937127e-05,
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 700
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 0.7376185458377239,
|
| 607 |
-
"eval_loss": 0.
|
| 608 |
-
"eval_runtime": 13.
|
| 609 |
-
"eval_samples_per_second": 35.
|
| 610 |
-
"eval_steps_per_second": 2.
|
| 611 |
"step": 700
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 0.7481559536354057,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
"learning_rate": 3.7530734106076575e-05,
|
| 617 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 618 |
"step": 710
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 0.7586933614330874,
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
"learning_rate": 3.7355110642781874e-05,
|
| 624 |
"loss": 0.2325,
|
| 625 |
"step": 720
|
| 626 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 627 |
{
|
| 628 |
"epoch": 0.7692307692307693,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
"learning_rate": 3.717948717948718e-05,
|
| 631 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 632 |
"step": 730
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 0.779768177028451,
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
"learning_rate": 3.7003863716192485e-05,
|
| 638 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 639 |
"step": 740
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 0.7903055848261328,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
"learning_rate": 3.682824025289779e-05,
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 750
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 0.7903055848261328,
|
| 650 |
-
"eval_loss": 0.
|
| 651 |
-
"eval_runtime": 13.
|
| 652 |
-
"eval_samples_per_second": 35.
|
| 653 |
-
"eval_steps_per_second": 2.
|
| 654 |
"step": 750
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 0.8008429926238145,
|
| 658 |
-
"grad_norm": 0.
|
| 659 |
"learning_rate": 3.6652616789603096e-05,
|
| 660 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 661 |
"step": 760
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 0.8113804004214963,
|
| 665 |
-
"grad_norm": 0.
|
| 666 |
"learning_rate": 3.6476993326308395e-05,
|
| 667 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 668 |
"step": 770
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 0.821917808219178,
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
"learning_rate": 3.63013698630137e-05,
|
| 674 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 675 |
"step": 780
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 0.8324552160168599,
|
| 679 |
-
"grad_norm": 0.
|
| 680 |
"learning_rate": 3.6125746399719e-05,
|
| 681 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 682 |
"step": 790
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 0.8429926238145417,
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
"learning_rate": 3.5950122936424305e-05,
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 0.8429926238145417,
|
| 693 |
-
"eval_loss": 0.
|
| 694 |
-
"eval_runtime": 13.
|
| 695 |
-
"eval_samples_per_second": 35.
|
| 696 |
-
"eval_steps_per_second": 2.
|
| 697 |
"step": 800
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 0.8535300316122234,
|
| 701 |
-
"grad_norm": 0.
|
| 702 |
"learning_rate": 3.577449947312961e-05,
|
| 703 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 704 |
"step": 810
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 0.8640674394099052,
|
| 708 |
-
"grad_norm": 0.
|
| 709 |
"learning_rate": 3.5598876009834916e-05,
|
| 710 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 711 |
"step": 820
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 0.8746048472075869,
|
| 715 |
-
"grad_norm": 0.
|
| 716 |
"learning_rate": 3.542325254654022e-05,
|
| 717 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 718 |
"step": 830
|
| 719 |
},
|
| 720 |
{
|
| 721 |
"epoch": 0.8851422550052687,
|
| 722 |
-
"grad_norm": 0.
|
| 723 |
"learning_rate": 3.524762908324553e-05,
|
| 724 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 725 |
"step": 840
|
| 726 |
},
|
| 727 |
{
|
| 728 |
"epoch": 0.8956796628029505,
|
| 729 |
-
"grad_norm": 0.
|
| 730 |
"learning_rate": 3.507200561995083e-05,
|
| 731 |
-
"loss": 0.
|
| 732 |
"step": 850
|
| 733 |
},
|
| 734 |
{
|
| 735 |
"epoch": 0.8956796628029505,
|
| 736 |
-
"eval_loss": 0.
|
| 737 |
-
"eval_runtime": 13.
|
| 738 |
-
"eval_samples_per_second": 35.
|
| 739 |
-
"eval_steps_per_second": 2.
|
| 740 |
"step": 850
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 0.9062170706006323,
|
| 744 |
-
"grad_norm": 0.
|
| 745 |
"learning_rate": 3.489638215665613e-05,
|
| 746 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 747 |
"step": 860
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 0.916754478398314,
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
"learning_rate": 3.472075869336143e-05,
|
| 753 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 754 |
"step": 870
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 0.9272918861959958,
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
"learning_rate": 3.454513523006674e-05,
|
| 760 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 761 |
"step": 880
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.9378292939936775,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
"learning_rate": 3.436951176677204e-05,
|
| 767 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 768 |
"step": 890
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 0.9483667017913593,
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
"learning_rate": 3.419388830347735e-05,
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 900
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 0.9483667017913593,
|
| 779 |
-
"eval_loss": 0.
|
| 780 |
-
"eval_runtime": 13.
|
| 781 |
-
"eval_samples_per_second": 35.
|
| 782 |
-
"eval_steps_per_second": 2.
|
| 783 |
"step": 900
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 0.958904109589041,
|
| 787 |
-
"grad_norm": 0.
|
| 788 |
"learning_rate": 3.4018264840182654e-05,
|
| 789 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 790 |
"step": 910
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 0.9694415173867229,
|
| 794 |
-
"grad_norm": 0.
|
| 795 |
"learning_rate": 3.384264137688795e-05,
|
| 796 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 797 |
"step": 920
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 0.9799789251844047,
|
| 801 |
-
"grad_norm": 0.
|
| 802 |
"learning_rate": 3.366701791359326e-05,
|
| 803 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 804 |
"step": 930
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 0.9905163329820864,
|
| 808 |
-
"grad_norm": 0.
|
| 809 |
"learning_rate": 3.3491394450298564e-05,
|
| 810 |
"loss": 0.2401,
|
| 811 |
"step": 940
|
| 812 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 813 |
{
|
| 814 |
"epoch": 1.0010537407797682,
|
| 815 |
-
"grad_norm": 0.
|
| 816 |
"learning_rate": 3.331577098700386e-05,
|
| 817 |
-
"loss": 0.
|
| 818 |
"step": 950
|
| 819 |
},
|
| 820 |
{
|
| 821 |
"epoch": 1.0010537407797682,
|
| 822 |
-
"eval_loss": 0.
|
| 823 |
-
"eval_runtime": 13.
|
| 824 |
-
"eval_samples_per_second": 35.
|
| 825 |
-
"eval_steps_per_second": 2.
|
| 826 |
"step": 950
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 1.01159114857745,
|
| 830 |
-
"grad_norm": 0.
|
| 831 |
"learning_rate": 3.314014752370917e-05,
|
| 832 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 833 |
"step": 960
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 1.0221285563751317,
|
| 837 |
-
"grad_norm": 0.
|
| 838 |
"learning_rate": 3.2964524060414474e-05,
|
| 839 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 840 |
"step": 970
|
| 841 |
},
|
| 842 |
{
|
| 843 |
"epoch": 1.0326659641728135,
|
| 844 |
-
"grad_norm": 0.
|
| 845 |
"learning_rate": 3.278890059711977e-05,
|
| 846 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 847 |
"step": 980
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 1.0432033719704952,
|
| 851 |
-
"grad_norm": 0.
|
| 852 |
"learning_rate": 3.261327713382508e-05,
|
| 853 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 854 |
"step": 990
|
| 855 |
},
|
| 856 |
{
|
| 857 |
"epoch": 1.053740779768177,
|
| 858 |
-
"grad_norm": 0.
|
| 859 |
"learning_rate": 3.2437653670530384e-05,
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 1000
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 1.053740779768177,
|
| 865 |
-
"eval_loss": 0.
|
| 866 |
-
"eval_runtime": 13.
|
| 867 |
-
"eval_samples_per_second": 35.
|
| 868 |
-
"eval_steps_per_second": 2.
|
| 869 |
"step": 1000
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 1.064278187565859,
|
| 873 |
-
"grad_norm": 0.
|
| 874 |
"learning_rate": 3.226203020723569e-05,
|
| 875 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 876 |
"step": 1010
|
| 877 |
},
|
| 878 |
{
|
| 879 |
"epoch": 1.0748155953635405,
|
| 880 |
-
"grad_norm": 0.
|
| 881 |
"learning_rate": 3.2086406743940996e-05,
|
| 882 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 883 |
"step": 1020
|
| 884 |
},
|
| 885 |
{
|
| 886 |
"epoch": 1.0853530031612224,
|
| 887 |
-
"grad_norm": 0.
|
| 888 |
"learning_rate": 3.1910783280646294e-05,
|
| 889 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 890 |
"step": 1030
|
| 891 |
},
|
| 892 |
{
|
| 893 |
"epoch": 1.095890410958904,
|
| 894 |
-
"grad_norm": 0.
|
| 895 |
"learning_rate": 3.17351598173516e-05,
|
| 896 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 897 |
"step": 1040
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 1.106427818756586,
|
| 901 |
-
"grad_norm": 0.
|
| 902 |
"learning_rate": 3.15595363540569e-05,
|
| 903 |
-
"loss": 0.
|
| 904 |
"step": 1050
|
| 905 |
},
|
| 906 |
{
|
| 907 |
"epoch": 1.106427818756586,
|
| 908 |
-
"eval_loss": 0.
|
| 909 |
-
"eval_runtime": 13.
|
| 910 |
-
"eval_samples_per_second": 35.
|
| 911 |
-
"eval_steps_per_second": 2.
|
| 912 |
"step": 1050
|
| 913 |
},
|
| 914 |
{
|
| 915 |
"epoch": 1.1169652265542676,
|
| 916 |
-
"grad_norm": 0.
|
| 917 |
"learning_rate": 3.1383912890762205e-05,
|
| 918 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 919 |
"step": 1060
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 1.1275026343519494,
|
| 923 |
-
"grad_norm": 0.
|
| 924 |
"learning_rate": 3.120828942746751e-05,
|
| 925 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 926 |
"step": 1070
|
| 927 |
},
|
| 928 |
{
|
| 929 |
"epoch": 1.1380400421496313,
|
| 930 |
-
"grad_norm": 0.
|
| 931 |
"learning_rate": 3.1032665964172816e-05,
|
| 932 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 933 |
"step": 1080
|
| 934 |
},
|
| 935 |
{
|
| 936 |
"epoch": 1.148577449947313,
|
| 937 |
-
"grad_norm": 0.
|
| 938 |
"learning_rate": 3.085704250087812e-05,
|
| 939 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 940 |
"step": 1090
|
| 941 |
},
|
| 942 |
{
|
| 943 |
"epoch": 1.1591148577449948,
|
| 944 |
-
"grad_norm": 0.
|
| 945 |
"learning_rate": 3.068141903758342e-05,
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 1100
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 1.1591148577449948,
|
| 951 |
-
"eval_loss": 0.
|
| 952 |
-
"eval_runtime": 13.
|
| 953 |
-
"eval_samples_per_second": 35.
|
| 954 |
-
"eval_steps_per_second": 2.
|
| 955 |
"step": 1100
|
| 956 |
},
|
| 957 |
{
|
| 958 |
"epoch": 1.1696522655426764,
|
| 959 |
-
"grad_norm": 0.
|
| 960 |
"learning_rate": 3.0505795574288726e-05,
|
| 961 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 962 |
"step": 1110
|
| 963 |
},
|
| 964 |
{
|
| 965 |
"epoch": 1.1801896733403583,
|
| 966 |
-
"grad_norm": 0.
|
| 967 |
"learning_rate": 3.0330172110994032e-05,
|
| 968 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 969 |
"step": 1120
|
| 970 |
},
|
| 971 |
{
|
| 972 |
"epoch": 1.1907270811380402,
|
| 973 |
-
"grad_norm": 0.
|
| 974 |
"learning_rate": 3.015454864769933e-05,
|
| 975 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 976 |
"step": 1130
|
| 977 |
},
|
| 978 |
{
|
| 979 |
"epoch": 1.2012644889357218,
|
| 980 |
-
"grad_norm": 0.
|
| 981 |
"learning_rate": 2.997892518440464e-05,
|
| 982 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 983 |
"step": 1140
|
| 984 |
},
|
| 985 |
{
|
| 986 |
"epoch": 1.2118018967334037,
|
| 987 |
-
"grad_norm": 0.
|
| 988 |
"learning_rate": 2.9803301721109945e-05,
|
| 989 |
-
"loss": 0.
|
| 990 |
"step": 1150
|
| 991 |
},
|
| 992 |
{
|
| 993 |
"epoch": 1.2118018967334037,
|
| 994 |
-
"eval_loss": 0.
|
| 995 |
-
"eval_runtime": 13.
|
| 996 |
-
"eval_samples_per_second": 35.
|
| 997 |
-
"eval_steps_per_second": 2.
|
| 998 |
"step": 1150
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
"epoch": 1.2223393045310853,
|
| 1002 |
-
"grad_norm": 0.
|
| 1003 |
"learning_rate": 2.9627678257815244e-05,
|
| 1004 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1005 |
"step": 1160
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
"epoch": 1.2328767123287672,
|
| 1009 |
-
"grad_norm": 0.
|
| 1010 |
"learning_rate": 2.945205479452055e-05,
|
| 1011 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1012 |
"step": 1170
|
| 1013 |
},
|
| 1014 |
{
|
| 1015 |
"epoch": 1.2434141201264488,
|
| 1016 |
-
"grad_norm": 0.
|
| 1017 |
"learning_rate": 2.9276431331225852e-05,
|
| 1018 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1019 |
"step": 1180
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
"epoch": 1.2539515279241307,
|
| 1023 |
-
"grad_norm": 0.
|
| 1024 |
"learning_rate": 2.9100807867931158e-05,
|
| 1025 |
"loss": 0.2077,
|
| 1026 |
"step": 1190
|
| 1027 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1028 |
{
|
| 1029 |
"epoch": 1.2644889357218125,
|
| 1030 |
-
"grad_norm": 0.
|
| 1031 |
"learning_rate": 2.8925184404636463e-05,
|
| 1032 |
-
"loss": 0.
|
| 1033 |
"step": 1200
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 1.2644889357218125,
|
| 1037 |
-
"eval_loss": 0.
|
| 1038 |
-
"eval_runtime": 13.
|
| 1039 |
-
"eval_samples_per_second":
|
| 1040 |
-
"eval_steps_per_second": 2.
|
| 1041 |
"step": 1200
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
"epoch": 1.2750263435194942,
|
| 1045 |
-
"grad_norm": 0.
|
| 1046 |
"learning_rate": 2.8749560941341762e-05,
|
| 1047 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1048 |
"step": 1210
|
| 1049 |
},
|
| 1050 |
{
|
| 1051 |
"epoch": 1.285563751317176,
|
| 1052 |
-
"grad_norm": 0.
|
| 1053 |
"learning_rate": 2.8573937478047068e-05,
|
| 1054 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1055 |
"step": 1220
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
"epoch": 1.2961011591148577,
|
| 1059 |
-
"grad_norm": 0.
|
| 1060 |
"learning_rate": 2.839831401475237e-05,
|
| 1061 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1062 |
"step": 1230
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
"epoch": 1.3066385669125395,
|
| 1066 |
-
"grad_norm": 0.
|
| 1067 |
"learning_rate": 2.8222690551457676e-05,
|
| 1068 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1069 |
"step": 1240
|
| 1070 |
},
|
| 1071 |
{
|
| 1072 |
"epoch": 1.3171759747102212,
|
| 1073 |
-
"grad_norm": 0.
|
| 1074 |
"learning_rate": 2.804706708816298e-05,
|
| 1075 |
-
"loss": 0.
|
| 1076 |
"step": 1250
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
"epoch": 1.3171759747102212,
|
| 1080 |
-
"eval_loss": 0.
|
| 1081 |
-
"eval_runtime": 13.
|
| 1082 |
-
"eval_samples_per_second": 35.
|
| 1083 |
-
"eval_steps_per_second": 2.
|
| 1084 |
"step": 1250
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
"epoch": 1.327713382507903,
|
| 1088 |
-
"grad_norm": 0.
|
| 1089 |
"learning_rate": 2.7871443624868284e-05,
|
| 1090 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1091 |
"step": 1260
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 1.338250790305585,
|
| 1095 |
-
"grad_norm": 0.
|
| 1096 |
"learning_rate": 2.769582016157359e-05,
|
| 1097 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1098 |
"step": 1270
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
"epoch": 1.3487881981032666,
|
| 1102 |
-
"grad_norm": 0.
|
| 1103 |
"learning_rate": 2.7520196698278895e-05,
|
| 1104 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1105 |
"step": 1280
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 1.3593256059009484,
|
| 1109 |
-
"grad_norm": 0.
|
| 1110 |
"learning_rate": 2.7344573234984194e-05,
|
| 1111 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1112 |
"step": 1290
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
"epoch": 1.36986301369863,
|
| 1116 |
-
"grad_norm": 0.
|
| 1117 |
"learning_rate": 2.71689497716895e-05,
|
| 1118 |
-
"loss": 0.
|
| 1119 |
"step": 1300
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 1.36986301369863,
|
| 1123 |
-
"eval_loss": 0.
|
| 1124 |
-
"eval_runtime": 13.
|
| 1125 |
-
"eval_samples_per_second": 35.
|
| 1126 |
-
"eval_steps_per_second": 2.
|
| 1127 |
"step": 1300
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
"epoch": 1.380400421496312,
|
| 1131 |
-
"grad_norm": 0.
|
| 1132 |
"learning_rate": 2.6993326308394802e-05,
|
| 1133 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1134 |
"step": 1310
|
| 1135 |
},
|
| 1136 |
{
|
| 1137 |
"epoch": 1.3909378292939936,
|
| 1138 |
-
"grad_norm": 0.
|
| 1139 |
"learning_rate": 2.6817702845100107e-05,
|
| 1140 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1141 |
"step": 1320
|
| 1142 |
},
|
| 1143 |
{
|
| 1144 |
"epoch": 1.4014752370916754,
|
| 1145 |
-
"grad_norm": 0.
|
| 1146 |
"learning_rate": 2.6642079381805413e-05,
|
| 1147 |
"loss": 0.1889,
|
| 1148 |
"step": 1330
|
| 1149 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1150 |
{
|
| 1151 |
"epoch": 1.4120126448893573,
|
| 1152 |
-
"grad_norm": 0.
|
| 1153 |
"learning_rate": 2.6466455918510712e-05,
|
| 1154 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1155 |
"step": 1340
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
"epoch": 1.422550052687039,
|
| 1159 |
-
"grad_norm": 0.
|
| 1160 |
"learning_rate": 2.6290832455216018e-05,
|
| 1161 |
-
"loss": 0.
|
| 1162 |
"step": 1350
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
"epoch": 1.422550052687039,
|
| 1166 |
-
"eval_loss": 0.
|
| 1167 |
-
"eval_runtime": 13.
|
| 1168 |
-
"eval_samples_per_second": 35.
|
| 1169 |
-
"eval_steps_per_second": 2.
|
| 1170 |
"step": 1350
|
| 1171 |
},
|
| 1172 |
{
|
| 1173 |
"epoch": 1.4330874604847208,
|
| 1174 |
-
"grad_norm": 0.
|
| 1175 |
"learning_rate": 2.611520899192132e-05,
|
| 1176 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1177 |
"step": 1360
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
"epoch": 1.4436248682824027,
|
| 1181 |
-
"grad_norm": 0.
|
| 1182 |
"learning_rate": 2.5939585528626625e-05,
|
| 1183 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1184 |
"step": 1370
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
"epoch": 1.4541622760800843,
|
| 1188 |
-
"grad_norm": 0.
|
| 1189 |
"learning_rate": 2.576396206533193e-05,
|
| 1190 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1191 |
"step": 1380
|
| 1192 |
},
|
| 1193 |
{
|
| 1194 |
"epoch": 1.464699683877766,
|
| 1195 |
-
"grad_norm": 0.
|
| 1196 |
"learning_rate": 2.5588338602037233e-05,
|
| 1197 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1198 |
"step": 1390
|
| 1199 |
},
|
| 1200 |
{
|
| 1201 |
"epoch": 1.4752370916754478,
|
| 1202 |
-
"grad_norm": 0.
|
| 1203 |
"learning_rate": 2.541271513874254e-05,
|
| 1204 |
"loss": 0.1853,
|
| 1205 |
"step": 1400
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
"epoch": 1.4752370916754478,
|
| 1209 |
-
"eval_loss": 0.
|
| 1210 |
-
"eval_runtime": 13.
|
| 1211 |
-
"eval_samples_per_second": 35.
|
| 1212 |
-
"eval_steps_per_second": 2.
|
| 1213 |
"step": 1400
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 1.4857744994731297,
|
| 1217 |
-
"grad_norm": 0.
|
| 1218 |
"learning_rate": 2.5237091675447845e-05,
|
| 1219 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1220 |
"step": 1410
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 1.4963119072708113,
|
| 1224 |
-
"grad_norm": 0.
|
| 1225 |
"learning_rate": 2.5061468212153144e-05,
|
| 1226 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1227 |
"step": 1420
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 1.5068493150684932,
|
| 1231 |
-
"grad_norm": 0.
|
| 1232 |
"learning_rate": 2.4885844748858446e-05,
|
| 1233 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1234 |
"step": 1430
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
"epoch": 1.517386722866175,
|
| 1238 |
-
"grad_norm": 0.
|
| 1239 |
"learning_rate": 2.4710221285563755e-05,
|
| 1240 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1241 |
"step": 1440
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 1.5279241306638567,
|
| 1245 |
-
"grad_norm": 0.
|
| 1246 |
"learning_rate": 2.4534597822269057e-05,
|
| 1247 |
-
"loss": 0.
|
| 1248 |
"step": 1450
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
"epoch": 1.5279241306638567,
|
| 1252 |
-
"eval_loss": 0.
|
| 1253 |
-
"eval_runtime": 13.
|
| 1254 |
-
"eval_samples_per_second": 35.
|
| 1255 |
-
"eval_steps_per_second": 2.
|
| 1256 |
"step": 1450
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
"epoch": 1.5384615384615383,
|
| 1260 |
-
"grad_norm": 0.
|
| 1261 |
"learning_rate": 2.435897435897436e-05,
|
| 1262 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1263 |
"step": 1460
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
"epoch": 1.5489989462592202,
|
| 1267 |
-
"grad_norm": 0.
|
| 1268 |
"learning_rate": 2.418335089567966e-05,
|
| 1269 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1270 |
"step": 1470
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
"epoch": 1.559536354056902,
|
| 1274 |
-
"grad_norm": 0.
|
| 1275 |
"learning_rate": 2.4007727432384967e-05,
|
| 1276 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1277 |
"step": 1480
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
"epoch": 1.5700737618545837,
|
| 1281 |
-
"grad_norm": 0.
|
| 1282 |
"learning_rate": 2.3832103969090273e-05,
|
| 1283 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1284 |
"step": 1490
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
"epoch": 1.5806111696522656,
|
| 1288 |
-
"grad_norm": 0.
|
| 1289 |
"learning_rate": 2.3656480505795575e-05,
|
| 1290 |
-
"loss": 0.
|
| 1291 |
"step": 1500
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
"epoch": 1.5806111696522656,
|
| 1295 |
-
"eval_loss": 0.
|
| 1296 |
-
"eval_runtime": 13.
|
| 1297 |
-
"eval_samples_per_second": 35.
|
| 1298 |
-
"eval_steps_per_second": 2.
|
| 1299 |
"step": 1500
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
"epoch": 1.5911485774499474,
|
| 1303 |
-
"grad_norm": 0.
|
| 1304 |
"learning_rate": 2.3480857042500877e-05,
|
| 1305 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1306 |
"step": 1510
|
| 1307 |
},
|
| 1308 |
{
|
| 1309 |
"epoch": 1.601685985247629,
|
| 1310 |
-
"grad_norm": 0.
|
| 1311 |
"learning_rate": 2.3305233579206183e-05,
|
| 1312 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1313 |
"step": 1520
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
"epoch": 1.6122233930453107,
|
| 1317 |
-
"grad_norm": 0.
|
| 1318 |
"learning_rate": 2.312961011591149e-05,
|
| 1319 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1320 |
"step": 1530
|
| 1321 |
},
|
| 1322 |
{
|
| 1323 |
"epoch": 1.6227608008429928,
|
| 1324 |
-
"grad_norm": 0.
|
| 1325 |
"learning_rate": 2.295398665261679e-05,
|
| 1326 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1327 |
"step": 1540
|
| 1328 |
},
|
| 1329 |
{
|
| 1330 |
"epoch": 1.6332982086406744,
|
| 1331 |
-
"grad_norm": 0.
|
| 1332 |
"learning_rate": 2.2778363189322093e-05,
|
| 1333 |
-
"loss": 0.
|
| 1334 |
"step": 1550
|
| 1335 |
},
|
| 1336 |
{
|
| 1337 |
"epoch": 1.6332982086406744,
|
| 1338 |
-
"eval_loss": 0.
|
| 1339 |
-
"eval_runtime": 13.
|
| 1340 |
-
"eval_samples_per_second": 35.
|
| 1341 |
-
"eval_steps_per_second": 2.
|
| 1342 |
"step": 1550
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
"epoch": 1.643835616438356,
|
| 1346 |
-
"grad_norm": 0.
|
| 1347 |
"learning_rate": 2.2602739726027396e-05,
|
| 1348 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1349 |
"step": 1560
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
"epoch": 1.654373024236038,
|
| 1353 |
-
"grad_norm": 0.
|
| 1354 |
"learning_rate": 2.2427116262732705e-05,
|
| 1355 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1356 |
"step": 1570
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
"epoch": 1.6649104320337198,
|
| 1360 |
-
"grad_norm": 0.
|
| 1361 |
"learning_rate": 2.2251492799438007e-05,
|
| 1362 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1363 |
"step": 1580
|
| 1364 |
},
|
| 1365 |
{
|
| 1366 |
"epoch": 1.6754478398314014,
|
| 1367 |
-
"grad_norm": 0.
|
| 1368 |
"learning_rate": 2.207586933614331e-05,
|
| 1369 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1370 |
"step": 1590
|
| 1371 |
},
|
| 1372 |
{
|
| 1373 |
"epoch": 1.685985247629083,
|
| 1374 |
-
"grad_norm": 0.
|
| 1375 |
"learning_rate": 2.190024587284861e-05,
|
| 1376 |
-
"loss": 0.
|
| 1377 |
"step": 1600
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
"epoch": 1.685985247629083,
|
| 1381 |
-
"eval_loss": 0.
|
| 1382 |
-
"eval_runtime": 13.
|
| 1383 |
-
"eval_samples_per_second": 35.
|
| 1384 |
-
"eval_steps_per_second": 2.
|
| 1385 |
"step": 1600
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
"epoch": 1.6965226554267652,
|
| 1389 |
-
"grad_norm": 0.
|
| 1390 |
"learning_rate": 2.1724622409553917e-05,
|
| 1391 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1392 |
"step": 1610
|
| 1393 |
},
|
| 1394 |
{
|
| 1395 |
"epoch": 1.7070600632244468,
|
| 1396 |
-
"grad_norm": 0.
|
| 1397 |
"learning_rate": 2.1548998946259223e-05,
|
| 1398 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1399 |
"step": 1620
|
| 1400 |
},
|
| 1401 |
{
|
| 1402 |
"epoch": 1.7175974710221285,
|
| 1403 |
-
"grad_norm": 0.
|
| 1404 |
"learning_rate": 2.1373375482964525e-05,
|
| 1405 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1406 |
"step": 1630
|
| 1407 |
},
|
| 1408 |
{
|
| 1409 |
"epoch": 1.7281348788198103,
|
| 1410 |
-
"grad_norm": 0.
|
| 1411 |
"learning_rate": 2.1197752019669827e-05,
|
| 1412 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1413 |
"step": 1640
|
| 1414 |
},
|
| 1415 |
{
|
| 1416 |
"epoch": 1.7386722866174922,
|
| 1417 |
-
"grad_norm": 0.
|
| 1418 |
"learning_rate": 2.1022128556375133e-05,
|
| 1419 |
-
"loss": 0.
|
| 1420 |
"step": 1650
|
| 1421 |
},
|
| 1422 |
{
|
| 1423 |
"epoch": 1.7386722866174922,
|
| 1424 |
-
"eval_loss": 0.
|
| 1425 |
-
"eval_runtime": 13.
|
| 1426 |
-
"eval_samples_per_second": 35.
|
| 1427 |
-
"eval_steps_per_second": 2.
|
| 1428 |
"step": 1650
|
| 1429 |
},
|
| 1430 |
{
|
| 1431 |
"epoch": 1.7492096944151738,
|
| 1432 |
-
"grad_norm": 0.
|
| 1433 |
"learning_rate": 2.084650509308044e-05,
|
| 1434 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1435 |
"step": 1660
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
"epoch": 1.7597471022128557,
|
| 1439 |
-
"grad_norm": 0.
|
| 1440 |
"learning_rate": 2.067088162978574e-05,
|
| 1441 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1442 |
"step": 1670
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
"epoch": 1.7702845100105375,
|
| 1446 |
-
"grad_norm": 0.
|
| 1447 |
"learning_rate": 2.0495258166491043e-05,
|
| 1448 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1449 |
"step": 1680
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
"epoch": 1.7808219178082192,
|
| 1453 |
-
"grad_norm": 0.
|
| 1454 |
"learning_rate": 2.0319634703196345e-05,
|
| 1455 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1456 |
"step": 1690
|
| 1457 |
},
|
| 1458 |
{
|
| 1459 |
"epoch": 1.7913593256059008,
|
| 1460 |
-
"grad_norm": 0.
|
| 1461 |
"learning_rate": 2.0144011239901654e-05,
|
| 1462 |
-
"loss": 0.
|
| 1463 |
"step": 1700
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
"epoch": 1.7913593256059008,
|
| 1467 |
-
"eval_loss": 0.
|
| 1468 |
-
"eval_runtime": 13.
|
| 1469 |
-
"eval_samples_per_second": 35.
|
| 1470 |
-
"eval_steps_per_second": 2.
|
| 1471 |
"step": 1700
|
| 1472 |
}
|
| 1473 |
],
|
|
@@ -1475,7 +2563,7 @@
|
|
| 1475 |
"max_steps": 2847,
|
| 1476 |
"num_input_tokens_seen": 0,
|
| 1477 |
"num_train_epochs": 3,
|
| 1478 |
-
"save_steps":
|
| 1479 |
"stateful_callbacks": {
|
| 1480 |
"TrainerControl": {
|
| 1481 |
"args": {
|
|
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.7913593256059008,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
"global_step": 1700,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01053740779768177,
|
| 13 |
+
"grad_norm": 3.3616116046905518,
|
| 14 |
"learning_rate": 4.982437653670531e-05,
|
| 15 |
+
"loss": 1.7881,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01053740779768177,
|
| 20 |
+
"eval_loss": 0.6833231449127197,
|
| 21 |
+
"eval_runtime": 13.3064,
|
| 22 |
+
"eval_samples_per_second": 36.073,
|
| 23 |
+
"eval_steps_per_second": 2.255,
|
| 24 |
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.02107481559536354,
|
| 28 |
+
"grad_norm": 1.3658663034439087,
|
| 29 |
"learning_rate": 4.964875307341061e-05,
|
| 30 |
+
"loss": 0.5571,
|
| 31 |
+
"step": 20
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.02107481559536354,
|
| 35 |
+
"eval_loss": 0.5293903350830078,
|
| 36 |
+
"eval_runtime": 13.4286,
|
| 37 |
+
"eval_samples_per_second": 35.745,
|
| 38 |
+
"eval_steps_per_second": 2.234,
|
| 39 |
"step": 20
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.03161222339304531,
|
| 43 |
+
"grad_norm": 1.0748717784881592,
|
| 44 |
"learning_rate": 4.947312961011591e-05,
|
| 45 |
+
"loss": 0.4482,
|
| 46 |
+
"step": 30
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.03161222339304531,
|
| 50 |
+
"eval_loss": 0.47912800312042236,
|
| 51 |
+
"eval_runtime": 13.4918,
|
| 52 |
+
"eval_samples_per_second": 35.577,
|
| 53 |
+
"eval_steps_per_second": 2.224,
|
| 54 |
"step": 30
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 0.04214963119072708,
|
| 58 |
+
"grad_norm": 0.890529215335846,
|
| 59 |
"learning_rate": 4.929750614682122e-05,
|
| 60 |
+
"loss": 0.3722,
|
| 61 |
+
"step": 40
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.04214963119072708,
|
| 65 |
+
"eval_loss": 0.44374603033065796,
|
| 66 |
+
"eval_runtime": 13.5807,
|
| 67 |
+
"eval_samples_per_second": 35.344,
|
| 68 |
+
"eval_steps_per_second": 2.209,
|
| 69 |
"step": 40
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.05268703898840885,
|
| 73 |
+
"grad_norm": 0.8233364224433899,
|
| 74 |
"learning_rate": 4.9121882683526524e-05,
|
| 75 |
+
"loss": 0.3534,
|
| 76 |
"step": 50
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 0.05268703898840885,
|
| 80 |
+
"eval_loss": 0.4318523108959198,
|
| 81 |
+
"eval_runtime": 13.6096,
|
| 82 |
+
"eval_samples_per_second": 35.269,
|
| 83 |
+
"eval_steps_per_second": 2.204,
|
| 84 |
"step": 50
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 0.06322444678609063,
|
| 88 |
+
"grad_norm": 0.7144497036933899,
|
| 89 |
"learning_rate": 4.894625922023183e-05,
|
| 90 |
+
"loss": 0.3512,
|
| 91 |
+
"step": 60
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.06322444678609063,
|
| 95 |
+
"eval_loss": 0.41558387875556946,
|
| 96 |
+
"eval_runtime": 13.6488,
|
| 97 |
+
"eval_samples_per_second": 35.168,
|
| 98 |
+
"eval_steps_per_second": 2.198,
|
| 99 |
"step": 60
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.0737618545837724,
|
| 103 |
+
"grad_norm": 0.6796255707740784,
|
| 104 |
"learning_rate": 4.877063575693713e-05,
|
| 105 |
+
"loss": 0.3588,
|
| 106 |
+
"step": 70
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.0737618545837724,
|
| 110 |
+
"eval_loss": 0.4085235297679901,
|
| 111 |
+
"eval_runtime": 13.6512,
|
| 112 |
+
"eval_samples_per_second": 35.162,
|
| 113 |
+
"eval_steps_per_second": 2.198,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.08429926238145416,
|
| 118 |
+
"grad_norm": 0.6846384406089783,
|
| 119 |
"learning_rate": 4.8595012293642434e-05,
|
| 120 |
+
"loss": 0.3504,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08429926238145416,
|
| 125 |
+
"eval_loss": 0.3993188142776489,
|
| 126 |
+
"eval_runtime": 13.6377,
|
| 127 |
+
"eval_samples_per_second": 35.197,
|
| 128 |
+
"eval_steps_per_second": 2.2,
|
| 129 |
"step": 80
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.09483667017913593,
|
| 133 |
+
"grad_norm": 0.7391332983970642,
|
| 134 |
"learning_rate": 4.841938883034774e-05,
|
| 135 |
+
"loss": 0.3299,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.09483667017913593,
|
| 140 |
+
"eval_loss": 0.3922707140445709,
|
| 141 |
+
"eval_runtime": 13.6059,
|
| 142 |
+
"eval_samples_per_second": 35.279,
|
| 143 |
+
"eval_steps_per_second": 2.205,
|
| 144 |
"step": 90
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 0.1053740779768177,
|
| 148 |
+
"grad_norm": 0.6717132925987244,
|
| 149 |
"learning_rate": 4.824376536705304e-05,
|
| 150 |
+
"loss": 0.3358,
|
| 151 |
"step": 100
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.1053740779768177,
|
| 155 |
+
"eval_loss": 0.390600323677063,
|
| 156 |
+
"eval_runtime": 13.6022,
|
| 157 |
+
"eval_samples_per_second": 35.288,
|
| 158 |
+
"eval_steps_per_second": 2.206,
|
| 159 |
"step": 100
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 0.11591148577449947,
|
| 163 |
+
"grad_norm": 0.6522348523139954,
|
| 164 |
"learning_rate": 4.8068141903758344e-05,
|
| 165 |
+
"loss": 0.311,
|
| 166 |
+
"step": 110
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.11591148577449947,
|
| 170 |
+
"eval_loss": 0.3844388425350189,
|
| 171 |
+
"eval_runtime": 13.5921,
|
| 172 |
+
"eval_samples_per_second": 35.315,
|
| 173 |
+
"eval_steps_per_second": 2.207,
|
| 174 |
"step": 110
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 0.12644889357218125,
|
| 178 |
+
"grad_norm": 0.5558998584747314,
|
| 179 |
"learning_rate": 4.789251844046364e-05,
|
| 180 |
+
"loss": 0.3161,
|
| 181 |
+
"step": 120
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.12644889357218125,
|
| 185 |
+
"eval_loss": 0.3852270245552063,
|
| 186 |
+
"eval_runtime": 13.5961,
|
| 187 |
+
"eval_samples_per_second": 35.304,
|
| 188 |
+
"eval_steps_per_second": 2.207,
|
| 189 |
"step": 120
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 0.136986301369863,
|
| 193 |
+
"grad_norm": 0.639998197555542,
|
| 194 |
"learning_rate": 4.7716894977168955e-05,
|
| 195 |
+
"loss": 0.3037,
|
| 196 |
+
"step": 130
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.136986301369863,
|
| 200 |
+
"eval_loss": 0.3841981887817383,
|
| 201 |
+
"eval_runtime": 13.6057,
|
| 202 |
+
"eval_samples_per_second": 35.279,
|
| 203 |
+
"eval_steps_per_second": 2.205,
|
| 204 |
"step": 130
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 0.1475237091675448,
|
| 208 |
+
"grad_norm": 0.583459198474884,
|
| 209 |
"learning_rate": 4.754127151387426e-05,
|
| 210 |
+
"loss": 0.3232,
|
| 211 |
+
"step": 140
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.1475237091675448,
|
| 215 |
+
"eval_loss": 0.3742731213569641,
|
| 216 |
+
"eval_runtime": 13.5955,
|
| 217 |
+
"eval_samples_per_second": 35.306,
|
| 218 |
+
"eval_steps_per_second": 2.207,
|
| 219 |
"step": 140
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.15806111696522657,
|
| 223 |
+
"grad_norm": 0.5834677219390869,
|
| 224 |
"learning_rate": 4.736564805057956e-05,
|
| 225 |
"loss": 0.317,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.15806111696522657,
|
| 230 |
+
"eval_loss": 0.3765258193016052,
|
| 231 |
+
"eval_runtime": 13.5995,
|
| 232 |
+
"eval_samples_per_second": 35.295,
|
| 233 |
+
"eval_steps_per_second": 2.206,
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.16859852476290832,
|
| 238 |
+
"grad_norm": 0.5975239872932434,
|
| 239 |
"learning_rate": 4.7190024587284866e-05,
|
| 240 |
+
"loss": 0.2987,
|
| 241 |
+
"step": 160
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.16859852476290832,
|
| 245 |
+
"eval_loss": 0.37060925364494324,
|
| 246 |
+
"eval_runtime": 13.5971,
|
| 247 |
+
"eval_samples_per_second": 35.302,
|
| 248 |
+
"eval_steps_per_second": 2.206,
|
| 249 |
"step": 160
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 0.1791359325605901,
|
| 253 |
+
"grad_norm": 0.6860706806182861,
|
| 254 |
"learning_rate": 4.7014401123990165e-05,
|
| 255 |
+
"loss": 0.3166,
|
| 256 |
+
"step": 170
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.1791359325605901,
|
| 260 |
+
"eval_loss": 0.37034350633621216,
|
| 261 |
+
"eval_runtime": 13.5923,
|
| 262 |
+
"eval_samples_per_second": 35.314,
|
| 263 |
+
"eval_steps_per_second": 2.207,
|
| 264 |
"step": 170
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 0.18967334035827185,
|
| 268 |
+
"grad_norm": 0.7161134481430054,
|
| 269 |
"learning_rate": 4.683877766069547e-05,
|
| 270 |
+
"loss": 0.3036,
|
| 271 |
+
"step": 180
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.18967334035827185,
|
| 275 |
+
"eval_loss": 0.3633898198604584,
|
| 276 |
+
"eval_runtime": 13.6104,
|
| 277 |
+
"eval_samples_per_second": 35.267,
|
| 278 |
+
"eval_steps_per_second": 2.204,
|
| 279 |
"step": 180
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 0.20021074815595363,
|
| 283 |
+
"grad_norm": 0.8212491273880005,
|
| 284 |
"learning_rate": 4.6663154197400776e-05,
|
| 285 |
+
"loss": 0.3023,
|
| 286 |
+
"step": 190
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.20021074815595363,
|
| 290 |
+
"eval_loss": 0.36356136202812195,
|
| 291 |
+
"eval_runtime": 13.5967,
|
| 292 |
+
"eval_samples_per_second": 35.303,
|
| 293 |
+
"eval_steps_per_second": 2.206,
|
| 294 |
"step": 190
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 0.2107481559536354,
|
| 298 |
+
"grad_norm": 0.5965659022331238,
|
| 299 |
"learning_rate": 4.6487530734106075e-05,
|
| 300 |
+
"loss": 0.2999,
|
| 301 |
"step": 200
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 0.2107481559536354,
|
| 305 |
+
"eval_loss": 0.3622604310512543,
|
| 306 |
+
"eval_runtime": 13.5997,
|
| 307 |
+
"eval_samples_per_second": 35.295,
|
| 308 |
+
"eval_steps_per_second": 2.206,
|
| 309 |
"step": 200
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 0.22128556375131717,
|
| 313 |
+
"grad_norm": 0.6638055443763733,
|
| 314 |
"learning_rate": 4.631190727081138e-05,
|
| 315 |
"loss": 0.2924,
|
| 316 |
"step": 210
|
| 317 |
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.22128556375131717,
|
| 320 |
+
"eval_loss": 0.36028483510017395,
|
| 321 |
+
"eval_runtime": 13.6089,
|
| 322 |
+
"eval_samples_per_second": 35.271,
|
| 323 |
+
"eval_steps_per_second": 2.204,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
{
|
| 327 |
"epoch": 0.23182297154899895,
|
| 328 |
+
"grad_norm": 0.4681139886379242,
|
| 329 |
"learning_rate": 4.6136283807516686e-05,
|
| 330 |
+
"loss": 0.3012,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.23182297154899895,
|
| 335 |
+
"eval_loss": 0.3585042357444763,
|
| 336 |
+
"eval_runtime": 13.6099,
|
| 337 |
+
"eval_samples_per_second": 35.269,
|
| 338 |
+
"eval_steps_per_second": 2.204,
|
| 339 |
"step": 220
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.24236037934668073,
|
| 343 |
+
"grad_norm": 0.6495727300643921,
|
| 344 |
"learning_rate": 4.596066034422199e-05,
|
| 345 |
+
"loss": 0.2884,
|
| 346 |
+
"step": 230
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.24236037934668073,
|
| 350 |
+
"eval_loss": 0.36549657583236694,
|
| 351 |
+
"eval_runtime": 13.6085,
|
| 352 |
+
"eval_samples_per_second": 35.272,
|
| 353 |
+
"eval_steps_per_second": 2.205,
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 0.2528977871443625,
|
| 358 |
+
"grad_norm": 0.4928750991821289,
|
| 359 |
"learning_rate": 4.57850368809273e-05,
|
| 360 |
+
"loss": 0.2838,
|
| 361 |
+
"step": 240
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.2528977871443625,
|
| 365 |
+
"eval_loss": 0.35502439737319946,
|
| 366 |
+
"eval_runtime": 13.6002,
|
| 367 |
+
"eval_samples_per_second": 35.293,
|
| 368 |
+
"eval_steps_per_second": 2.206,
|
| 369 |
"step": 240
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 0.26343519494204426,
|
| 373 |
+
"grad_norm": 0.6839736700057983,
|
| 374 |
"learning_rate": 4.5609413417632596e-05,
|
| 375 |
+
"loss": 0.2861,
|
| 376 |
"step": 250
|
| 377 |
},
|
| 378 |
{
|
| 379 |
"epoch": 0.26343519494204426,
|
| 380 |
+
"eval_loss": 0.3560730516910553,
|
| 381 |
+
"eval_runtime": 13.6056,
|
| 382 |
+
"eval_samples_per_second": 35.28,
|
| 383 |
+
"eval_steps_per_second": 2.205,
|
| 384 |
"step": 250
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 0.273972602739726,
|
| 388 |
+
"grad_norm": 0.5583881735801697,
|
| 389 |
"learning_rate": 4.54337899543379e-05,
|
| 390 |
+
"loss": 0.3019,
|
| 391 |
+
"step": 260
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.273972602739726,
|
| 395 |
+
"eval_loss": 0.3566754162311554,
|
| 396 |
+
"eval_runtime": 13.5986,
|
| 397 |
+
"eval_samples_per_second": 35.298,
|
| 398 |
+
"eval_steps_per_second": 2.206,
|
| 399 |
"step": 260
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 0.2845100105374078,
|
| 403 |
+
"grad_norm": 0.5884770750999451,
|
| 404 |
"learning_rate": 4.525816649104321e-05,
|
| 405 |
"loss": 0.3112,
|
| 406 |
"step": 270
|
| 407 |
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.2845100105374078,
|
| 410 |
+
"eval_loss": 0.3552027940750122,
|
| 411 |
+
"eval_runtime": 13.6456,
|
| 412 |
+
"eval_samples_per_second": 35.176,
|
| 413 |
+
"eval_steps_per_second": 2.199,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
{
|
| 417 |
"epoch": 0.2950474183350896,
|
| 418 |
+
"grad_norm": 0.540367603302002,
|
| 419 |
"learning_rate": 4.5082543027748506e-05,
|
| 420 |
+
"loss": 0.2805,
|
| 421 |
+
"step": 280
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.2950474183350896,
|
| 425 |
+
"eval_loss": 0.3551066815853119,
|
| 426 |
+
"eval_runtime": 13.59,
|
| 427 |
+
"eval_samples_per_second": 35.32,
|
| 428 |
+
"eval_steps_per_second": 2.208,
|
| 429 |
"step": 280
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.3055848261327713,
|
| 433 |
+
"grad_norm": 0.4952965974807739,
|
| 434 |
"learning_rate": 4.490691956445381e-05,
|
| 435 |
+
"loss": 0.318,
|
| 436 |
+
"step": 290
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3055848261327713,
|
| 440 |
+
"eval_loss": 0.35322481393814087,
|
| 441 |
+
"eval_runtime": 13.5781,
|
| 442 |
+
"eval_samples_per_second": 35.351,
|
| 443 |
+
"eval_steps_per_second": 2.209,
|
| 444 |
"step": 290
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 0.31612223393045313,
|
| 448 |
+
"grad_norm": 0.48861801624298096,
|
| 449 |
"learning_rate": 4.473129610115912e-05,
|
| 450 |
+
"loss": 0.2878,
|
| 451 |
"step": 300
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 0.31612223393045313,
|
| 455 |
+
"eval_loss": 0.3530054986476898,
|
| 456 |
+
"eval_runtime": 13.5721,
|
| 457 |
+
"eval_samples_per_second": 35.367,
|
| 458 |
+
"eval_steps_per_second": 2.21,
|
| 459 |
"step": 300
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 0.3266596417281349,
|
| 463 |
+
"grad_norm": 0.523873507976532,
|
| 464 |
"learning_rate": 4.455567263786442e-05,
|
| 465 |
"loss": 0.3084,
|
| 466 |
"step": 310
|
| 467 |
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.3266596417281349,
|
| 470 |
+
"eval_loss": 0.35032400488853455,
|
| 471 |
+
"eval_runtime": 13.5686,
|
| 472 |
+
"eval_samples_per_second": 35.376,
|
| 473 |
+
"eval_steps_per_second": 2.211,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
{
|
| 477 |
"epoch": 0.33719704952581664,
|
| 478 |
+
"grad_norm": 0.6253566741943359,
|
| 479 |
"learning_rate": 4.438004917456973e-05,
|
| 480 |
+
"loss": 0.2947,
|
| 481 |
+
"step": 320
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.33719704952581664,
|
| 485 |
+
"eval_loss": 0.3499966263771057,
|
| 486 |
+
"eval_runtime": 13.5767,
|
| 487 |
+
"eval_samples_per_second": 35.355,
|
| 488 |
+
"eval_steps_per_second": 2.21,
|
| 489 |
"step": 320
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 0.34773445732349845,
|
| 493 |
+
"grad_norm": 0.5352979898452759,
|
| 494 |
"learning_rate": 4.420442571127503e-05,
|
| 495 |
+
"loss": 0.2884,
|
| 496 |
+
"step": 330
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.34773445732349845,
|
| 500 |
+
"eval_loss": 0.34622183442115784,
|
| 501 |
+
"eval_runtime": 13.5616,
|
| 502 |
+
"eval_samples_per_second": 35.394,
|
| 503 |
+
"eval_steps_per_second": 2.212,
|
| 504 |
"step": 330
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 0.3582718651211802,
|
| 508 |
+
"grad_norm": 0.4931572675704956,
|
| 509 |
"learning_rate": 4.4028802247980333e-05,
|
| 510 |
+
"loss": 0.3085,
|
| 511 |
+
"step": 340
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.3582718651211802,
|
| 515 |
+
"eval_loss": 0.3468419909477234,
|
| 516 |
+
"eval_runtime": 13.5639,
|
| 517 |
+
"eval_samples_per_second": 35.388,
|
| 518 |
+
"eval_steps_per_second": 2.212,
|
| 519 |
"step": 340
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 0.36880927291886195,
|
| 523 |
+
"grad_norm": 0.5033659934997559,
|
| 524 |
"learning_rate": 4.385317878468563e-05,
|
| 525 |
"loss": 0.3084,
|
| 526 |
"step": 350
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 0.36880927291886195,
|
| 530 |
+
"eval_loss": 0.34584841132164,
|
| 531 |
+
"eval_runtime": 13.5544,
|
| 532 |
+
"eval_samples_per_second": 35.413,
|
| 533 |
+
"eval_steps_per_second": 2.213,
|
| 534 |
"step": 350
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 0.3793466807165437,
|
| 538 |
+
"grad_norm": 0.5806519389152527,
|
| 539 |
"learning_rate": 4.367755532139094e-05,
|
| 540 |
+
"loss": 0.2761,
|
| 541 |
+
"step": 360
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3793466807165437,
|
| 545 |
+
"eval_loss": 0.34370240569114685,
|
| 546 |
+
"eval_runtime": 13.5617,
|
| 547 |
+
"eval_samples_per_second": 35.394,
|
| 548 |
+
"eval_steps_per_second": 2.212,
|
| 549 |
"step": 360
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 0.3898840885142255,
|
| 553 |
+
"grad_norm": 0.6204981803894043,
|
| 554 |
"learning_rate": 4.3501931858096244e-05,
|
| 555 |
+
"loss": 0.3024,
|
| 556 |
+
"step": 370
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3898840885142255,
|
| 560 |
+
"eval_loss": 0.3421412408351898,
|
| 561 |
+
"eval_runtime": 13.5312,
|
| 562 |
+
"eval_samples_per_second": 35.474,
|
| 563 |
+
"eval_steps_per_second": 2.217,
|
| 564 |
"step": 370
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 0.40042149631190727,
|
| 568 |
+
"grad_norm": 0.48026618361473083,
|
| 569 |
"learning_rate": 4.332630839480154e-05,
|
| 570 |
+
"loss": 0.2788,
|
| 571 |
+
"step": 380
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.40042149631190727,
|
| 575 |
+
"eval_loss": 0.3414471745491028,
|
| 576 |
+
"eval_runtime": 13.5435,
|
| 577 |
+
"eval_samples_per_second": 35.441,
|
| 578 |
+
"eval_steps_per_second": 2.215,
|
| 579 |
"step": 380
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 0.410958904109589,
|
| 583 |
+
"grad_norm": 0.48742201924324036,
|
| 584 |
"learning_rate": 4.3150684931506855e-05,
|
| 585 |
+
"loss": 0.2708,
|
| 586 |
+
"step": 390
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.410958904109589,
|
| 590 |
+
"eval_loss": 0.3414710462093353,
|
| 591 |
+
"eval_runtime": 13.5593,
|
| 592 |
+
"eval_samples_per_second": 35.4,
|
| 593 |
+
"eval_steps_per_second": 2.213,
|
| 594 |
"step": 390
|
| 595 |
},
|
| 596 |
{
|
| 597 |
"epoch": 0.4214963119072708,
|
| 598 |
+
"grad_norm": 0.5737293362617493,
|
| 599 |
"learning_rate": 4.297506146821216e-05,
|
| 600 |
+
"loss": 0.2735,
|
| 601 |
"step": 400
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 0.4214963119072708,
|
| 605 |
+
"eval_loss": 0.34133002161979675,
|
| 606 |
+
"eval_runtime": 13.5644,
|
| 607 |
+
"eval_samples_per_second": 35.387,
|
| 608 |
+
"eval_steps_per_second": 2.212,
|
| 609 |
"step": 400
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 0.4320337197049526,
|
| 613 |
+
"grad_norm": 0.5055080652236938,
|
| 614 |
"learning_rate": 4.279943800491746e-05,
|
| 615 |
+
"loss": 0.2846,
|
| 616 |
+
"step": 410
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.4320337197049526,
|
| 620 |
+
"eval_loss": 0.3391459584236145,
|
| 621 |
+
"eval_runtime": 13.5574,
|
| 622 |
+
"eval_samples_per_second": 35.405,
|
| 623 |
+
"eval_steps_per_second": 2.213,
|
| 624 |
"step": 410
|
| 625 |
},
|
| 626 |
{
|
| 627 |
"epoch": 0.44257112750263433,
|
| 628 |
+
"grad_norm": 0.5293213129043579,
|
| 629 |
"learning_rate": 4.2623814541622765e-05,
|
| 630 |
+
"loss": 0.2757,
|
| 631 |
+
"step": 420
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.44257112750263433,
|
| 635 |
+
"eval_loss": 0.3393440246582031,
|
| 636 |
+
"eval_runtime": 13.5643,
|
| 637 |
+
"eval_samples_per_second": 35.387,
|
| 638 |
+
"eval_steps_per_second": 2.212,
|
| 639 |
"step": 420
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 0.45310853530031614,
|
| 643 |
+
"grad_norm": 0.6627737283706665,
|
| 644 |
"learning_rate": 4.2448191078328064e-05,
|
| 645 |
+
"loss": 0.2829,
|
| 646 |
+
"step": 430
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.45310853530031614,
|
| 650 |
+
"eval_loss": 0.34091198444366455,
|
| 651 |
+
"eval_runtime": 13.5863,
|
| 652 |
+
"eval_samples_per_second": 35.33,
|
| 653 |
+
"eval_steps_per_second": 2.208,
|
| 654 |
"step": 430
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 0.4636459430979979,
|
| 658 |
+
"grad_norm": 0.5125611424446106,
|
| 659 |
"learning_rate": 4.227256761503337e-05,
|
| 660 |
+
"loss": 0.2632,
|
| 661 |
+
"step": 440
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.4636459430979979,
|
| 665 |
+
"eval_loss": 0.3409743010997772,
|
| 666 |
+
"eval_runtime": 13.572,
|
| 667 |
+
"eval_samples_per_second": 35.367,
|
| 668 |
+
"eval_steps_per_second": 2.21,
|
| 669 |
"step": 440
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 0.47418335089567965,
|
| 673 |
+
"grad_norm": 0.613191545009613,
|
| 674 |
"learning_rate": 4.2096944151738675e-05,
|
| 675 |
+
"loss": 0.2709,
|
| 676 |
"step": 450
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 0.47418335089567965,
|
| 680 |
+
"eval_loss": 0.3384413421154022,
|
| 681 |
+
"eval_runtime": 13.5818,
|
| 682 |
+
"eval_samples_per_second": 35.341,
|
| 683 |
+
"eval_steps_per_second": 2.209,
|
| 684 |
"step": 450
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 0.48472075869336145,
|
| 688 |
+
"grad_norm": 0.4908188581466675,
|
| 689 |
"learning_rate": 4.1921320688443974e-05,
|
| 690 |
+
"loss": 0.2609,
|
| 691 |
+
"step": 460
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.48472075869336145,
|
| 695 |
+
"eval_loss": 0.33855584263801575,
|
| 696 |
+
"eval_runtime": 13.5848,
|
| 697 |
+
"eval_samples_per_second": 35.334,
|
| 698 |
+
"eval_steps_per_second": 2.208,
|
| 699 |
"step": 460
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 0.4952581664910432,
|
| 703 |
+
"grad_norm": 0.5873326659202576,
|
| 704 |
"learning_rate": 4.174569722514928e-05,
|
| 705 |
+
"loss": 0.2709,
|
| 706 |
+
"step": 470
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.4952581664910432,
|
| 710 |
+
"eval_loss": 0.3383730947971344,
|
| 711 |
+
"eval_runtime": 13.591,
|
| 712 |
+
"eval_samples_per_second": 35.317,
|
| 713 |
+
"eval_steps_per_second": 2.207,
|
| 714 |
"step": 470
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 0.505795574288725,
|
| 718 |
+
"grad_norm": 0.5365350246429443,
|
| 719 |
"learning_rate": 4.1570073761854585e-05,
|
| 720 |
+
"loss": 0.2848,
|
| 721 |
+
"step": 480
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.505795574288725,
|
| 725 |
+
"eval_loss": 0.3378843367099762,
|
| 726 |
+
"eval_runtime": 13.5963,
|
| 727 |
+
"eval_samples_per_second": 35.304,
|
| 728 |
+
"eval_steps_per_second": 2.206,
|
| 729 |
"step": 480
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 0.5163329820864068,
|
| 733 |
+
"grad_norm": 0.5642485022544861,
|
| 734 |
"learning_rate": 4.139445029855989e-05,
|
| 735 |
"loss": 0.2911,
|
| 736 |
"step": 490
|
| 737 |
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.5163329820864068,
|
| 740 |
+
"eval_loss": 0.3355303406715393,
|
| 741 |
+
"eval_runtime": 13.6025,
|
| 742 |
+
"eval_samples_per_second": 35.288,
|
| 743 |
+
"eval_steps_per_second": 2.205,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
{
|
| 747 |
"epoch": 0.5268703898840885,
|
| 748 |
+
"grad_norm": 0.6058154702186584,
|
| 749 |
"learning_rate": 4.12188268352652e-05,
|
| 750 |
+
"loss": 0.262,
|
| 751 |
"step": 500
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 0.5268703898840885,
|
| 755 |
+
"eval_loss": 0.3364686667919159,
|
| 756 |
+
"eval_runtime": 13.6033,
|
| 757 |
+
"eval_samples_per_second": 35.286,
|
| 758 |
+
"eval_steps_per_second": 2.205,
|
| 759 |
"step": 500
|
| 760 |
},
|
| 761 |
{
|
| 762 |
"epoch": 0.5374077976817703,
|
| 763 |
+
"grad_norm": 0.666671633720398,
|
| 764 |
"learning_rate": 4.1043203371970496e-05,
|
| 765 |
+
"loss": 0.2797,
|
| 766 |
+
"step": 510
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.5374077976817703,
|
| 770 |
+
"eval_loss": 0.33817631006240845,
|
| 771 |
+
"eval_runtime": 13.6129,
|
| 772 |
+
"eval_samples_per_second": 35.261,
|
| 773 |
+
"eval_steps_per_second": 2.204,
|
| 774 |
"step": 510
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 0.547945205479452,
|
| 778 |
+
"grad_norm": 0.5990138053894043,
|
| 779 |
"learning_rate": 4.08675799086758e-05,
|
| 780 |
+
"loss": 0.2606,
|
| 781 |
+
"step": 520
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.547945205479452,
|
| 785 |
+
"eval_loss": 0.3355501890182495,
|
| 786 |
+
"eval_runtime": 13.6034,
|
| 787 |
+
"eval_samples_per_second": 35.285,
|
| 788 |
+
"eval_steps_per_second": 2.205,
|
| 789 |
"step": 520
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 0.5584826132771338,
|
| 793 |
+
"grad_norm": 0.5905662775039673,
|
| 794 |
"learning_rate": 4.069195644538111e-05,
|
| 795 |
+
"loss": 0.2687,
|
| 796 |
+
"step": 530
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.5584826132771338,
|
| 800 |
+
"eval_loss": 0.3357114791870117,
|
| 801 |
+
"eval_runtime": 13.5907,
|
| 802 |
+
"eval_samples_per_second": 35.318,
|
| 803 |
+
"eval_steps_per_second": 2.207,
|
| 804 |
"step": 530
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 0.5690200210748156,
|
| 808 |
+
"grad_norm": 0.6308056116104126,
|
| 809 |
"learning_rate": 4.0516332982086406e-05,
|
| 810 |
+
"loss": 0.2964,
|
| 811 |
+
"step": 540
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.5690200210748156,
|
| 815 |
+
"eval_loss": 0.3342372477054596,
|
| 816 |
+
"eval_runtime": 13.5897,
|
| 817 |
+
"eval_samples_per_second": 35.321,
|
| 818 |
+
"eval_steps_per_second": 2.208,
|
| 819 |
"step": 540
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 0.5795574288724974,
|
| 823 |
+
"grad_norm": 0.5483158230781555,
|
| 824 |
"learning_rate": 4.034070951879171e-05,
|
| 825 |
+
"loss": 0.262,
|
| 826 |
"step": 550
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 0.5795574288724974,
|
| 830 |
+
"eval_loss": 0.33271318674087524,
|
| 831 |
+
"eval_runtime": 13.5802,
|
| 832 |
+
"eval_samples_per_second": 35.346,
|
| 833 |
+
"eval_steps_per_second": 2.209,
|
| 834 |
"step": 550
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 0.5900948366701791,
|
| 838 |
+
"grad_norm": 0.5433160662651062,
|
| 839 |
"learning_rate": 4.016508605549702e-05,
|
| 840 |
+
"loss": 0.2575,
|
| 841 |
+
"step": 560
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.5900948366701791,
|
| 845 |
+
"eval_loss": 0.3352707326412201,
|
| 846 |
+
"eval_runtime": 13.5772,
|
| 847 |
+
"eval_samples_per_second": 35.353,
|
| 848 |
+
"eval_steps_per_second": 2.21,
|
| 849 |
"step": 560
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 0.6006322444678609,
|
| 853 |
+
"grad_norm": 0.7090951800346375,
|
| 854 |
"learning_rate": 3.998946259220232e-05,
|
| 855 |
+
"loss": 0.2624,
|
| 856 |
+
"step": 570
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6006322444678609,
|
| 860 |
+
"eval_loss": 0.33547133207321167,
|
| 861 |
+
"eval_runtime": 13.5717,
|
| 862 |
+
"eval_samples_per_second": 35.368,
|
| 863 |
+
"eval_steps_per_second": 2.21,
|
| 864 |
"step": 570
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 0.6111696522655427,
|
| 868 |
+
"grad_norm": 0.7720335721969604,
|
| 869 |
"learning_rate": 3.981383912890763e-05,
|
| 870 |
+
"loss": 0.2628,
|
| 871 |
+
"step": 580
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.6111696522655427,
|
| 875 |
+
"eval_loss": 0.3337227702140808,
|
| 876 |
+
"eval_runtime": 13.5578,
|
| 877 |
+
"eval_samples_per_second": 35.404,
|
| 878 |
+
"eval_steps_per_second": 2.213,
|
| 879 |
"step": 580
|
| 880 |
},
|
| 881 |
{
|
| 882 |
"epoch": 0.6217070600632244,
|
| 883 |
+
"grad_norm": 0.5641891360282898,
|
| 884 |
"learning_rate": 3.963821566561293e-05,
|
| 885 |
+
"loss": 0.2617,
|
| 886 |
+
"step": 590
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.6217070600632244,
|
| 890 |
+
"eval_loss": 0.3328714668750763,
|
| 891 |
+
"eval_runtime": 13.5205,
|
| 892 |
+
"eval_samples_per_second": 35.502,
|
| 893 |
+
"eval_steps_per_second": 2.219,
|
| 894 |
"step": 590
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 0.6322444678609063,
|
| 898 |
+
"grad_norm": 0.5985394716262817,
|
| 899 |
"learning_rate": 3.946259220231823e-05,
|
| 900 |
+
"loss": 0.2543,
|
| 901 |
"step": 600
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 0.6322444678609063,
|
| 905 |
+
"eval_loss": 0.3321264386177063,
|
| 906 |
+
"eval_runtime": 13.5572,
|
| 907 |
+
"eval_samples_per_second": 35.406,
|
| 908 |
+
"eval_steps_per_second": 2.213,
|
| 909 |
"step": 600
|
| 910 |
},
|
| 911 |
{
|
| 912 |
"epoch": 0.642781875658588,
|
| 913 |
+
"grad_norm": 0.6638147234916687,
|
| 914 |
"learning_rate": 3.928696873902353e-05,
|
| 915 |
+
"loss": 0.2772,
|
| 916 |
+
"step": 610
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.642781875658588,
|
| 920 |
+
"eval_loss": 0.33292463421821594,
|
| 921 |
+
"eval_runtime": 13.5485,
|
| 922 |
+
"eval_samples_per_second": 35.428,
|
| 923 |
+
"eval_steps_per_second": 2.214,
|
| 924 |
"step": 610
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 0.6533192834562698,
|
| 928 |
+
"grad_norm": 0.6401410102844238,
|
| 929 |
"learning_rate": 3.911134527572884e-05,
|
| 930 |
+
"loss": 0.2725,
|
| 931 |
+
"step": 620
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.6533192834562698,
|
| 935 |
+
"eval_loss": 0.3313460648059845,
|
| 936 |
+
"eval_runtime": 13.5546,
|
| 937 |
+
"eval_samples_per_second": 35.412,
|
| 938 |
+
"eval_steps_per_second": 2.213,
|
| 939 |
"step": 620
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 0.6638566912539515,
|
| 943 |
+
"grad_norm": 0.5968373417854309,
|
| 944 |
"learning_rate": 3.893572181243414e-05,
|
| 945 |
+
"loss": 0.2626,
|
| 946 |
+
"step": 630
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.6638566912539515,
|
| 950 |
+
"eval_loss": 0.3308354616165161,
|
| 951 |
+
"eval_runtime": 13.5566,
|
| 952 |
+
"eval_samples_per_second": 35.407,
|
| 953 |
+
"eval_steps_per_second": 2.213,
|
| 954 |
"step": 630
|
| 955 |
},
|
| 956 |
{
|
| 957 |
"epoch": 0.6743940990516333,
|
| 958 |
+
"grad_norm": 0.4671958088874817,
|
| 959 |
"learning_rate": 3.876009834913945e-05,
|
| 960 |
+
"loss": 0.2667,
|
| 961 |
+
"step": 640
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6743940990516333,
|
| 965 |
+
"eval_loss": 0.33001866936683655,
|
| 966 |
+
"eval_runtime": 13.5739,
|
| 967 |
+
"eval_samples_per_second": 35.362,
|
| 968 |
+
"eval_steps_per_second": 2.21,
|
| 969 |
"step": 640
|
| 970 |
},
|
| 971 |
{
|
| 972 |
"epoch": 0.684931506849315,
|
| 973 |
+
"grad_norm": 0.5390825867652893,
|
| 974 |
"learning_rate": 3.8584474885844754e-05,
|
| 975 |
+
"loss": 0.2599,
|
| 976 |
"step": 650
|
| 977 |
},
|
| 978 |
{
|
| 979 |
"epoch": 0.684931506849315,
|
| 980 |
+
"eval_loss": 0.3301464915275574,
|
| 981 |
+
"eval_runtime": 13.5678,
|
| 982 |
+
"eval_samples_per_second": 35.378,
|
| 983 |
+
"eval_steps_per_second": 2.211,
|
| 984 |
"step": 650
|
| 985 |
},
|
| 986 |
{
|
| 987 |
"epoch": 0.6954689146469969,
|
| 988 |
+
"grad_norm": 0.5784376263618469,
|
| 989 |
"learning_rate": 3.840885142255005e-05,
|
| 990 |
+
"loss": 0.2452,
|
| 991 |
+
"step": 660
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.6954689146469969,
|
| 995 |
+
"eval_loss": 0.33094659447669983,
|
| 996 |
+
"eval_runtime": 13.5767,
|
| 997 |
+
"eval_samples_per_second": 35.355,
|
| 998 |
+
"eval_steps_per_second": 2.21,
|
| 999 |
"step": 660
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
"epoch": 0.7060063224446786,
|
| 1003 |
+
"grad_norm": 0.5976417660713196,
|
| 1004 |
"learning_rate": 3.823322795925536e-05,
|
| 1005 |
+
"loss": 0.2688,
|
| 1006 |
+
"step": 670
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.7060063224446786,
|
| 1010 |
+
"eval_loss": 0.3312263488769531,
|
| 1011 |
+
"eval_runtime": 13.5986,
|
| 1012 |
+
"eval_samples_per_second": 35.298,
|
| 1013 |
+
"eval_steps_per_second": 2.206,
|
| 1014 |
"step": 670
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
"epoch": 0.7165437302423604,
|
| 1018 |
+
"grad_norm": 0.5902723073959351,
|
| 1019 |
"learning_rate": 3.8057604495960664e-05,
|
| 1020 |
+
"loss": 0.2596,
|
| 1021 |
+
"step": 680
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.7165437302423604,
|
| 1025 |
+
"eval_loss": 0.32888591289520264,
|
| 1026 |
+
"eval_runtime": 13.597,
|
| 1027 |
+
"eval_samples_per_second": 35.302,
|
| 1028 |
+
"eval_steps_per_second": 2.206,
|
| 1029 |
"step": 680
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
"epoch": 0.7270811380400422,
|
| 1033 |
+
"grad_norm": 0.6102134585380554,
|
| 1034 |
"learning_rate": 3.788198103266596e-05,
|
| 1035 |
+
"loss": 0.2623,
|
| 1036 |
+
"step": 690
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.7270811380400422,
|
| 1040 |
+
"eval_loss": 0.3287771940231323,
|
| 1041 |
+
"eval_runtime": 13.6069,
|
| 1042 |
+
"eval_samples_per_second": 35.276,
|
| 1043 |
+
"eval_steps_per_second": 2.205,
|
| 1044 |
"step": 690
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
"epoch": 0.7376185458377239,
|
| 1048 |
+
"grad_norm": 0.5212683081626892,
|
| 1049 |
"learning_rate": 3.770635756937127e-05,
|
| 1050 |
+
"loss": 0.2497,
|
| 1051 |
"step": 700
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 0.7376185458377239,
|
| 1055 |
+
"eval_loss": 0.330812007188797,
|
| 1056 |
+
"eval_runtime": 13.6008,
|
| 1057 |
+
"eval_samples_per_second": 35.292,
|
| 1058 |
+
"eval_steps_per_second": 2.206,
|
| 1059 |
"step": 700
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
"epoch": 0.7481559536354057,
|
| 1063 |
+
"grad_norm": 0.614651620388031,
|
| 1064 |
"learning_rate": 3.7530734106076575e-05,
|
| 1065 |
+
"loss": 0.2568,
|
| 1066 |
+
"step": 710
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7481559536354057,
|
| 1070 |
+
"eval_loss": 0.3295523524284363,
|
| 1071 |
+
"eval_runtime": 13.6076,
|
| 1072 |
+
"eval_samples_per_second": 35.274,
|
| 1073 |
+
"eval_steps_per_second": 2.205,
|
| 1074 |
"step": 710
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
"epoch": 0.7586933614330874,
|
| 1078 |
+
"grad_norm": 0.4703381061553955,
|
| 1079 |
"learning_rate": 3.7355110642781874e-05,
|
| 1080 |
"loss": 0.2325,
|
| 1081 |
"step": 720
|
| 1082 |
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.7586933614330874,
|
| 1085 |
+
"eval_loss": 0.32799476385116577,
|
| 1086 |
+
"eval_runtime": 13.6031,
|
| 1087 |
+
"eval_samples_per_second": 35.286,
|
| 1088 |
+
"eval_steps_per_second": 2.205,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
{
|
| 1092 |
"epoch": 0.7692307692307693,
|
| 1093 |
+
"grad_norm": 0.5316283702850342,
|
| 1094 |
"learning_rate": 3.717948717948718e-05,
|
| 1095 |
+
"loss": 0.2427,
|
| 1096 |
+
"step": 730
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.7692307692307693,
|
| 1100 |
+
"eval_loss": 0.32939499616622925,
|
| 1101 |
+
"eval_runtime": 13.602,
|
| 1102 |
+
"eval_samples_per_second": 35.289,
|
| 1103 |
+
"eval_steps_per_second": 2.206,
|
| 1104 |
"step": 730
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
"epoch": 0.779768177028451,
|
| 1108 |
+
"grad_norm": 0.5144335627555847,
|
| 1109 |
"learning_rate": 3.7003863716192485e-05,
|
| 1110 |
+
"loss": 0.2563,
|
| 1111 |
+
"step": 740
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.779768177028451,
|
| 1115 |
+
"eval_loss": 0.3298059403896332,
|
| 1116 |
+
"eval_runtime": 13.5958,
|
| 1117 |
+
"eval_samples_per_second": 35.305,
|
| 1118 |
+
"eval_steps_per_second": 2.207,
|
| 1119 |
"step": 740
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 0.7903055848261328,
|
| 1123 |
+
"grad_norm": 0.5247990489006042,
|
| 1124 |
"learning_rate": 3.682824025289779e-05,
|
| 1125 |
+
"loss": 0.2358,
|
| 1126 |
"step": 750
|
| 1127 |
},
|
| 1128 |
{
|
| 1129 |
"epoch": 0.7903055848261328,
|
| 1130 |
+
"eval_loss": 0.32794949412345886,
|
| 1131 |
+
"eval_runtime": 13.6027,
|
| 1132 |
+
"eval_samples_per_second": 35.287,
|
| 1133 |
+
"eval_steps_per_second": 2.205,
|
| 1134 |
"step": 750
|
| 1135 |
},
|
| 1136 |
{
|
| 1137 |
"epoch": 0.8008429926238145,
|
| 1138 |
+
"grad_norm": 0.5819652676582336,
|
| 1139 |
"learning_rate": 3.6652616789603096e-05,
|
| 1140 |
+
"loss": 0.2423,
|
| 1141 |
+
"step": 760
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.8008429926238145,
|
| 1145 |
+
"eval_loss": 0.3263894319534302,
|
| 1146 |
+
"eval_runtime": 13.588,
|
| 1147 |
+
"eval_samples_per_second": 35.325,
|
| 1148 |
+
"eval_steps_per_second": 2.208,
|
| 1149 |
"step": 760
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
"epoch": 0.8113804004214963,
|
| 1153 |
+
"grad_norm": 0.5112130045890808,
|
| 1154 |
"learning_rate": 3.6476993326308395e-05,
|
| 1155 |
+
"loss": 0.2638,
|
| 1156 |
+
"step": 770
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.8113804004214963,
|
| 1160 |
+
"eval_loss": 0.3270197808742523,
|
| 1161 |
+
"eval_runtime": 13.5821,
|
| 1162 |
+
"eval_samples_per_second": 35.341,
|
| 1163 |
+
"eval_steps_per_second": 2.209,
|
| 1164 |
"step": 770
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 0.821917808219178,
|
| 1168 |
+
"grad_norm": 0.5146061778068542,
|
| 1169 |
"learning_rate": 3.63013698630137e-05,
|
| 1170 |
+
"loss": 0.2375,
|
| 1171 |
+
"step": 780
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.821917808219178,
|
| 1175 |
+
"eval_loss": 0.3265083432197571,
|
| 1176 |
+
"eval_runtime": 13.5816,
|
| 1177 |
+
"eval_samples_per_second": 35.342,
|
| 1178 |
+
"eval_steps_per_second": 2.209,
|
| 1179 |
"step": 780
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
"epoch": 0.8324552160168599,
|
| 1183 |
+
"grad_norm": 0.5121347904205322,
|
| 1184 |
"learning_rate": 3.6125746399719e-05,
|
| 1185 |
+
"loss": 0.2468,
|
| 1186 |
+
"step": 790
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.8324552160168599,
|
| 1190 |
+
"eval_loss": 0.32619577646255493,
|
| 1191 |
+
"eval_runtime": 13.5874,
|
| 1192 |
+
"eval_samples_per_second": 35.327,
|
| 1193 |
+
"eval_steps_per_second": 2.208,
|
| 1194 |
"step": 790
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
"epoch": 0.8429926238145417,
|
| 1198 |
+
"grad_norm": 0.5837659239768982,
|
| 1199 |
"learning_rate": 3.5950122936424305e-05,
|
| 1200 |
+
"loss": 0.2392,
|
| 1201 |
"step": 800
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
"epoch": 0.8429926238145417,
|
| 1205 |
+
"eval_loss": 0.3283963203430176,
|
| 1206 |
+
"eval_runtime": 13.5978,
|
| 1207 |
+
"eval_samples_per_second": 35.3,
|
| 1208 |
+
"eval_steps_per_second": 2.206,
|
| 1209 |
"step": 800
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
"epoch": 0.8535300316122234,
|
| 1213 |
+
"grad_norm": 0.6293447613716125,
|
| 1214 |
"learning_rate": 3.577449947312961e-05,
|
| 1215 |
+
"loss": 0.2415,
|
| 1216 |
+
"step": 810
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.8535300316122234,
|
| 1220 |
+
"eval_loss": 0.32890576124191284,
|
| 1221 |
+
"eval_runtime": 13.5797,
|
| 1222 |
+
"eval_samples_per_second": 35.347,
|
| 1223 |
+
"eval_steps_per_second": 2.209,
|
| 1224 |
"step": 810
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
"epoch": 0.8640674394099052,
|
| 1228 |
+
"grad_norm": 0.5239388942718506,
|
| 1229 |
"learning_rate": 3.5598876009834916e-05,
|
| 1230 |
+
"loss": 0.2459,
|
| 1231 |
+
"step": 820
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.8640674394099052,
|
| 1235 |
+
"eval_loss": 0.3269563317298889,
|
| 1236 |
+
"eval_runtime": 13.5785,
|
| 1237 |
+
"eval_samples_per_second": 35.35,
|
| 1238 |
+
"eval_steps_per_second": 2.209,
|
| 1239 |
"step": 820
|
| 1240 |
},
|
| 1241 |
{
|
| 1242 |
"epoch": 0.8746048472075869,
|
| 1243 |
+
"grad_norm": 0.5803388953208923,
|
| 1244 |
"learning_rate": 3.542325254654022e-05,
|
| 1245 |
+
"loss": 0.2561,
|
| 1246 |
+
"step": 830
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.8746048472075869,
|
| 1250 |
+
"eval_loss": 0.32730814814567566,
|
| 1251 |
+
"eval_runtime": 13.5623,
|
| 1252 |
+
"eval_samples_per_second": 35.392,
|
| 1253 |
+
"eval_steps_per_second": 2.212,
|
| 1254 |
"step": 830
|
| 1255 |
},
|
| 1256 |
{
|
| 1257 |
"epoch": 0.8851422550052687,
|
| 1258 |
+
"grad_norm": 0.541553795337677,
|
| 1259 |
"learning_rate": 3.524762908324553e-05,
|
| 1260 |
+
"loss": 0.2407,
|
| 1261 |
+
"step": 840
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.8851422550052687,
|
| 1265 |
+
"eval_loss": 0.32772594690322876,
|
| 1266 |
+
"eval_runtime": 13.5543,
|
| 1267 |
+
"eval_samples_per_second": 35.413,
|
| 1268 |
+
"eval_steps_per_second": 2.213,
|
| 1269 |
"step": 840
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 0.8956796628029505,
|
| 1273 |
+
"grad_norm": 0.547488272190094,
|
| 1274 |
"learning_rate": 3.507200561995083e-05,
|
| 1275 |
+
"loss": 0.2512,
|
| 1276 |
"step": 850
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
"epoch": 0.8956796628029505,
|
| 1280 |
+
"eval_loss": 0.32760006189346313,
|
| 1281 |
+
"eval_runtime": 13.5629,
|
| 1282 |
+
"eval_samples_per_second": 35.391,
|
| 1283 |
+
"eval_steps_per_second": 2.212,
|
| 1284 |
"step": 850
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
"epoch": 0.9062170706006323,
|
| 1288 |
+
"grad_norm": 0.5485065579414368,
|
| 1289 |
"learning_rate": 3.489638215665613e-05,
|
| 1290 |
+
"loss": 0.2341,
|
| 1291 |
+
"step": 860
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.9062170706006323,
|
| 1295 |
+
"eval_loss": 0.327595055103302,
|
| 1296 |
+
"eval_runtime": 13.5606,
|
| 1297 |
+
"eval_samples_per_second": 35.397,
|
| 1298 |
+
"eval_steps_per_second": 2.212,
|
| 1299 |
"step": 860
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
"epoch": 0.916754478398314,
|
| 1303 |
+
"grad_norm": 0.5831624865531921,
|
| 1304 |
"learning_rate": 3.472075869336143e-05,
|
| 1305 |
+
"loss": 0.2465,
|
| 1306 |
+
"step": 870
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.916754478398314,
|
| 1310 |
+
"eval_loss": 0.325825572013855,
|
| 1311 |
+
"eval_runtime": 13.5618,
|
| 1312 |
+
"eval_samples_per_second": 35.394,
|
| 1313 |
+
"eval_steps_per_second": 2.212,
|
| 1314 |
"step": 870
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
"epoch": 0.9272918861959958,
|
| 1318 |
+
"grad_norm": 0.528672456741333,
|
| 1319 |
"learning_rate": 3.454513523006674e-05,
|
| 1320 |
+
"loss": 0.2355,
|
| 1321 |
+
"step": 880
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.9272918861959958,
|
| 1325 |
+
"eval_loss": 0.32486769556999207,
|
| 1326 |
+
"eval_runtime": 13.553,
|
| 1327 |
+
"eval_samples_per_second": 35.417,
|
| 1328 |
+
"eval_steps_per_second": 2.214,
|
| 1329 |
"step": 880
|
| 1330 |
},
|
| 1331 |
{
|
| 1332 |
"epoch": 0.9378292939936775,
|
| 1333 |
+
"grad_norm": 0.5654621124267578,
|
| 1334 |
"learning_rate": 3.436951176677204e-05,
|
| 1335 |
+
"loss": 0.244,
|
| 1336 |
+
"step": 890
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.9378292939936775,
|
| 1340 |
+
"eval_loss": 0.32602888345718384,
|
| 1341 |
+
"eval_runtime": 13.5581,
|
| 1342 |
+
"eval_samples_per_second": 35.403,
|
| 1343 |
+
"eval_steps_per_second": 2.213,
|
| 1344 |
"step": 890
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
"epoch": 0.9483667017913593,
|
| 1348 |
+
"grad_norm": 0.5009713172912598,
|
| 1349 |
"learning_rate": 3.419388830347735e-05,
|
| 1350 |
+
"loss": 0.2299,
|
| 1351 |
"step": 900
|
| 1352 |
},
|
| 1353 |
{
|
| 1354 |
"epoch": 0.9483667017913593,
|
| 1355 |
+
"eval_loss": 0.32582351565361023,
|
| 1356 |
+
"eval_runtime": 13.5546,
|
| 1357 |
+
"eval_samples_per_second": 35.412,
|
| 1358 |
+
"eval_steps_per_second": 2.213,
|
| 1359 |
"step": 900
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
"epoch": 0.958904109589041,
|
| 1363 |
+
"grad_norm": 0.5322644114494324,
|
| 1364 |
"learning_rate": 3.4018264840182654e-05,
|
| 1365 |
+
"loss": 0.2382,
|
| 1366 |
+
"step": 910
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.958904109589041,
|
| 1370 |
+
"eval_loss": 0.32392945885658264,
|
| 1371 |
+
"eval_runtime": 13.5582,
|
| 1372 |
+
"eval_samples_per_second": 35.403,
|
| 1373 |
+
"eval_steps_per_second": 2.213,
|
| 1374 |
"step": 910
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
"epoch": 0.9694415173867229,
|
| 1378 |
+
"grad_norm": 0.5522642731666565,
|
| 1379 |
"learning_rate": 3.384264137688795e-05,
|
| 1380 |
+
"loss": 0.2478,
|
| 1381 |
+
"step": 920
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.9694415173867229,
|
| 1385 |
+
"eval_loss": 0.3241512179374695,
|
| 1386 |
+
"eval_runtime": 13.5666,
|
| 1387 |
+
"eval_samples_per_second": 35.381,
|
| 1388 |
+
"eval_steps_per_second": 2.211,
|
| 1389 |
"step": 920
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 0.9799789251844047,
|
| 1393 |
+
"grad_norm": 0.5779546499252319,
|
| 1394 |
"learning_rate": 3.366701791359326e-05,
|
| 1395 |
+
"loss": 0.2534,
|
| 1396 |
+
"step": 930
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.9799789251844047,
|
| 1400 |
+
"eval_loss": 0.3243790864944458,
|
| 1401 |
+
"eval_runtime": 13.524,
|
| 1402 |
+
"eval_samples_per_second": 35.492,
|
| 1403 |
+
"eval_steps_per_second": 2.218,
|
| 1404 |
"step": 930
|
| 1405 |
},
|
| 1406 |
{
|
| 1407 |
"epoch": 0.9905163329820864,
|
| 1408 |
+
"grad_norm": 0.6141674518585205,
|
| 1409 |
"learning_rate": 3.3491394450298564e-05,
|
| 1410 |
"loss": 0.2401,
|
| 1411 |
"step": 940
|
| 1412 |
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.9905163329820864,
|
| 1415 |
+
"eval_loss": 0.3262033462524414,
|
| 1416 |
+
"eval_runtime": 13.5269,
|
| 1417 |
+
"eval_samples_per_second": 35.485,
|
| 1418 |
+
"eval_steps_per_second": 2.218,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
{
|
| 1422 |
"epoch": 1.0010537407797682,
|
| 1423 |
+
"grad_norm": 0.42756763100624084,
|
| 1424 |
"learning_rate": 3.331577098700386e-05,
|
| 1425 |
+
"loss": 0.2338,
|
| 1426 |
"step": 950
|
| 1427 |
},
|
| 1428 |
{
|
| 1429 |
"epoch": 1.0010537407797682,
|
| 1430 |
+
"eval_loss": 0.3266028165817261,
|
| 1431 |
+
"eval_runtime": 13.5168,
|
| 1432 |
+
"eval_samples_per_second": 35.511,
|
| 1433 |
+
"eval_steps_per_second": 2.219,
|
| 1434 |
"step": 950
|
| 1435 |
},
|
| 1436 |
{
|
| 1437 |
"epoch": 1.01159114857745,
|
| 1438 |
+
"grad_norm": 0.5351940989494324,
|
| 1439 |
"learning_rate": 3.314014752370917e-05,
|
| 1440 |
+
"loss": 0.1934,
|
| 1441 |
+
"step": 960
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.01159114857745,
|
| 1445 |
+
"eval_loss": 0.3327268064022064,
|
| 1446 |
+
"eval_runtime": 13.5225,
|
| 1447 |
+
"eval_samples_per_second": 35.497,
|
| 1448 |
+
"eval_steps_per_second": 2.219,
|
| 1449 |
"step": 960
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
"epoch": 1.0221285563751317,
|
| 1453 |
+
"grad_norm": 0.5101184248924255,
|
| 1454 |
"learning_rate": 3.2964524060414474e-05,
|
| 1455 |
+
"loss": 0.208,
|
| 1456 |
+
"step": 970
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.0221285563751317,
|
| 1460 |
+
"eval_loss": 0.3375646471977234,
|
| 1461 |
+
"eval_runtime": 13.5136,
|
| 1462 |
+
"eval_samples_per_second": 35.52,
|
| 1463 |
+
"eval_steps_per_second": 2.22,
|
| 1464 |
"step": 970
|
| 1465 |
},
|
| 1466 |
{
|
| 1467 |
"epoch": 1.0326659641728135,
|
| 1468 |
+
"grad_norm": 0.6187830567359924,
|
| 1469 |
"learning_rate": 3.278890059711977e-05,
|
| 1470 |
+
"loss": 0.2145,
|
| 1471 |
+
"step": 980
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 1.0326659641728135,
|
| 1475 |
+
"eval_loss": 0.3343217074871063,
|
| 1476 |
+
"eval_runtime": 13.5266,
|
| 1477 |
+
"eval_samples_per_second": 35.486,
|
| 1478 |
+
"eval_steps_per_second": 2.218,
|
| 1479 |
"step": 980
|
| 1480 |
},
|
| 1481 |
{
|
| 1482 |
"epoch": 1.0432033719704952,
|
| 1483 |
+
"grad_norm": 0.669201672077179,
|
| 1484 |
"learning_rate": 3.261327713382508e-05,
|
| 1485 |
+
"loss": 0.2141,
|
| 1486 |
+
"step": 990
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 1.0432033719704952,
|
| 1490 |
+
"eval_loss": 0.33267149329185486,
|
| 1491 |
+
"eval_runtime": 13.5144,
|
| 1492 |
+
"eval_samples_per_second": 35.518,
|
| 1493 |
+
"eval_steps_per_second": 2.22,
|
| 1494 |
"step": 990
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
"epoch": 1.053740779768177,
|
| 1498 |
+
"grad_norm": 0.5544168949127197,
|
| 1499 |
"learning_rate": 3.2437653670530384e-05,
|
| 1500 |
+
"loss": 0.2074,
|
| 1501 |
"step": 1000
|
| 1502 |
},
|
| 1503 |
{
|
| 1504 |
"epoch": 1.053740779768177,
|
| 1505 |
+
"eval_loss": 0.33638790249824524,
|
| 1506 |
+
"eval_runtime": 13.5127,
|
| 1507 |
+
"eval_samples_per_second": 35.522,
|
| 1508 |
+
"eval_steps_per_second": 2.22,
|
| 1509 |
"step": 1000
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
"epoch": 1.064278187565859,
|
| 1513 |
+
"grad_norm": 0.6323490738868713,
|
| 1514 |
"learning_rate": 3.226203020723569e-05,
|
| 1515 |
+
"loss": 0.2012,
|
| 1516 |
+
"step": 1010
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 1.064278187565859,
|
| 1520 |
+
"eval_loss": 0.3400089144706726,
|
| 1521 |
+
"eval_runtime": 13.5093,
|
| 1522 |
+
"eval_samples_per_second": 35.531,
|
| 1523 |
+
"eval_steps_per_second": 2.221,
|
| 1524 |
"step": 1010
|
| 1525 |
},
|
| 1526 |
{
|
| 1527 |
"epoch": 1.0748155953635405,
|
| 1528 |
+
"grad_norm": 0.5988993048667908,
|
| 1529 |
"learning_rate": 3.2086406743940996e-05,
|
| 1530 |
+
"loss": 0.1954,
|
| 1531 |
+
"step": 1020
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 1.0748155953635405,
|
| 1535 |
+
"eval_loss": 0.3393259048461914,
|
| 1536 |
+
"eval_runtime": 13.5107,
|
| 1537 |
+
"eval_samples_per_second": 35.527,
|
| 1538 |
+
"eval_steps_per_second": 2.22,
|
| 1539 |
"step": 1020
|
| 1540 |
},
|
| 1541 |
{
|
| 1542 |
"epoch": 1.0853530031612224,
|
| 1543 |
+
"grad_norm": 0.6308508515357971,
|
| 1544 |
"learning_rate": 3.1910783280646294e-05,
|
| 1545 |
+
"loss": 0.2066,
|
| 1546 |
+
"step": 1030
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 1.0853530031612224,
|
| 1550 |
+
"eval_loss": 0.337782084941864,
|
| 1551 |
+
"eval_runtime": 13.5075,
|
| 1552 |
+
"eval_samples_per_second": 35.536,
|
| 1553 |
+
"eval_steps_per_second": 2.221,
|
| 1554 |
"step": 1030
|
| 1555 |
},
|
| 1556 |
{
|
| 1557 |
"epoch": 1.095890410958904,
|
| 1558 |
+
"grad_norm": 0.5316015481948853,
|
| 1559 |
"learning_rate": 3.17351598173516e-05,
|
| 1560 |
+
"loss": 0.2074,
|
| 1561 |
+
"step": 1040
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.095890410958904,
|
| 1565 |
+
"eval_loss": 0.3380493223667145,
|
| 1566 |
+
"eval_runtime": 13.531,
|
| 1567 |
+
"eval_samples_per_second": 35.474,
|
| 1568 |
+
"eval_steps_per_second": 2.217,
|
| 1569 |
"step": 1040
|
| 1570 |
},
|
| 1571 |
{
|
| 1572 |
"epoch": 1.106427818756586,
|
| 1573 |
+
"grad_norm": 0.6207646131515503,
|
| 1574 |
"learning_rate": 3.15595363540569e-05,
|
| 1575 |
+
"loss": 0.2088,
|
| 1576 |
"step": 1050
|
| 1577 |
},
|
| 1578 |
{
|
| 1579 |
"epoch": 1.106427818756586,
|
| 1580 |
+
"eval_loss": 0.340431809425354,
|
| 1581 |
+
"eval_runtime": 13.5161,
|
| 1582 |
+
"eval_samples_per_second": 35.513,
|
| 1583 |
+
"eval_steps_per_second": 2.22,
|
| 1584 |
"step": 1050
|
| 1585 |
},
|
| 1586 |
{
|
| 1587 |
"epoch": 1.1169652265542676,
|
| 1588 |
+
"grad_norm": 0.5499827265739441,
|
| 1589 |
"learning_rate": 3.1383912890762205e-05,
|
| 1590 |
+
"loss": 0.2049,
|
| 1591 |
+
"step": 1060
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 1.1169652265542676,
|
| 1595 |
+
"eval_loss": 0.34178677201271057,
|
| 1596 |
+
"eval_runtime": 13.5149,
|
| 1597 |
+
"eval_samples_per_second": 35.516,
|
| 1598 |
+
"eval_steps_per_second": 2.22,
|
| 1599 |
"step": 1060
|
| 1600 |
},
|
| 1601 |
{
|
| 1602 |
"epoch": 1.1275026343519494,
|
| 1603 |
+
"grad_norm": 0.5874269604682922,
|
| 1604 |
"learning_rate": 3.120828942746751e-05,
|
| 1605 |
+
"loss": 0.2027,
|
| 1606 |
+
"step": 1070
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.1275026343519494,
|
| 1610 |
+
"eval_loss": 0.3401419222354889,
|
| 1611 |
+
"eval_runtime": 13.5078,
|
| 1612 |
+
"eval_samples_per_second": 35.535,
|
| 1613 |
+
"eval_steps_per_second": 2.221,
|
| 1614 |
"step": 1070
|
| 1615 |
},
|
| 1616 |
{
|
| 1617 |
"epoch": 1.1380400421496313,
|
| 1618 |
+
"grad_norm": 0.5852048397064209,
|
| 1619 |
"learning_rate": 3.1032665964172816e-05,
|
| 1620 |
+
"loss": 0.1999,
|
| 1621 |
+
"step": 1080
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 1.1380400421496313,
|
| 1625 |
+
"eval_loss": 0.3392138183116913,
|
| 1626 |
+
"eval_runtime": 13.5107,
|
| 1627 |
+
"eval_samples_per_second": 35.527,
|
| 1628 |
+
"eval_steps_per_second": 2.22,
|
| 1629 |
"step": 1080
|
| 1630 |
},
|
| 1631 |
{
|
| 1632 |
"epoch": 1.148577449947313,
|
| 1633 |
+
"grad_norm": 0.6016051769256592,
|
| 1634 |
"learning_rate": 3.085704250087812e-05,
|
| 1635 |
+
"loss": 0.2044,
|
| 1636 |
+
"step": 1090
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 1.148577449947313,
|
| 1640 |
+
"eval_loss": 0.34129372239112854,
|
| 1641 |
+
"eval_runtime": 13.5118,
|
| 1642 |
+
"eval_samples_per_second": 35.525,
|
| 1643 |
+
"eval_steps_per_second": 2.22,
|
| 1644 |
"step": 1090
|
| 1645 |
},
|
| 1646 |
{
|
| 1647 |
"epoch": 1.1591148577449948,
|
| 1648 |
+
"grad_norm": 0.5797489285469055,
|
| 1649 |
"learning_rate": 3.068141903758342e-05,
|
| 1650 |
+
"loss": 0.1974,
|
| 1651 |
"step": 1100
|
| 1652 |
},
|
| 1653 |
{
|
| 1654 |
"epoch": 1.1591148577449948,
|
| 1655 |
+
"eval_loss": 0.34007471799850464,
|
| 1656 |
+
"eval_runtime": 13.5066,
|
| 1657 |
+
"eval_samples_per_second": 35.538,
|
| 1658 |
+
"eval_steps_per_second": 2.221,
|
| 1659 |
"step": 1100
|
| 1660 |
},
|
| 1661 |
{
|
| 1662 |
"epoch": 1.1696522655426764,
|
| 1663 |
+
"grad_norm": 0.6308450102806091,
|
| 1664 |
"learning_rate": 3.0505795574288726e-05,
|
| 1665 |
+
"loss": 0.2037,
|
| 1666 |
+
"step": 1110
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 1.1696522655426764,
|
| 1670 |
+
"eval_loss": 0.3387734293937683,
|
| 1671 |
+
"eval_runtime": 13.5225,
|
| 1672 |
+
"eval_samples_per_second": 35.496,
|
| 1673 |
+
"eval_steps_per_second": 2.219,
|
| 1674 |
"step": 1110
|
| 1675 |
},
|
| 1676 |
{
|
| 1677 |
"epoch": 1.1801896733403583,
|
| 1678 |
+
"grad_norm": 0.6952560544013977,
|
| 1679 |
"learning_rate": 3.0330172110994032e-05,
|
| 1680 |
+
"loss": 0.2009,
|
| 1681 |
+
"step": 1120
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 1.1801896733403583,
|
| 1685 |
+
"eval_loss": 0.3363548815250397,
|
| 1686 |
+
"eval_runtime": 13.515,
|
| 1687 |
+
"eval_samples_per_second": 35.516,
|
| 1688 |
+
"eval_steps_per_second": 2.22,
|
| 1689 |
"step": 1120
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
"epoch": 1.1907270811380402,
|
| 1693 |
+
"grad_norm": 0.6670855283737183,
|
| 1694 |
"learning_rate": 3.015454864769933e-05,
|
| 1695 |
+
"loss": 0.2046,
|
| 1696 |
+
"step": 1130
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 1.1907270811380402,
|
| 1700 |
+
"eval_loss": 0.3387907147407532,
|
| 1701 |
+
"eval_runtime": 13.5274,
|
| 1702 |
+
"eval_samples_per_second": 35.484,
|
| 1703 |
+
"eval_steps_per_second": 2.218,
|
| 1704 |
"step": 1130
|
| 1705 |
},
|
| 1706 |
{
|
| 1707 |
"epoch": 1.2012644889357218,
|
| 1708 |
+
"grad_norm": 0.7167879343032837,
|
| 1709 |
"learning_rate": 2.997892518440464e-05,
|
| 1710 |
+
"loss": 0.2039,
|
| 1711 |
+
"step": 1140
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.2012644889357218,
|
| 1715 |
+
"eval_loss": 0.33721429109573364,
|
| 1716 |
+
"eval_runtime": 13.5247,
|
| 1717 |
+
"eval_samples_per_second": 35.491,
|
| 1718 |
+
"eval_steps_per_second": 2.218,
|
| 1719 |
"step": 1140
|
| 1720 |
},
|
| 1721 |
{
|
| 1722 |
"epoch": 1.2118018967334037,
|
| 1723 |
+
"grad_norm": 0.5917449593544006,
|
| 1724 |
"learning_rate": 2.9803301721109945e-05,
|
| 1725 |
+
"loss": 0.1997,
|
| 1726 |
"step": 1150
|
| 1727 |
},
|
| 1728 |
{
|
| 1729 |
"epoch": 1.2118018967334037,
|
| 1730 |
+
"eval_loss": 0.33842217922210693,
|
| 1731 |
+
"eval_runtime": 13.5215,
|
| 1732 |
+
"eval_samples_per_second": 35.499,
|
| 1733 |
+
"eval_steps_per_second": 2.219,
|
| 1734 |
"step": 1150
|
| 1735 |
},
|
| 1736 |
{
|
| 1737 |
"epoch": 1.2223393045310853,
|
| 1738 |
+
"grad_norm": 0.6303601264953613,
|
| 1739 |
"learning_rate": 2.9627678257815244e-05,
|
| 1740 |
+
"loss": 0.2063,
|
| 1741 |
+
"step": 1160
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 1.2223393045310853,
|
| 1745 |
+
"eval_loss": 0.3402339816093445,
|
| 1746 |
+
"eval_runtime": 13.5168,
|
| 1747 |
+
"eval_samples_per_second": 35.511,
|
| 1748 |
+
"eval_steps_per_second": 2.219,
|
| 1749 |
"step": 1160
|
| 1750 |
},
|
| 1751 |
{
|
| 1752 |
"epoch": 1.2328767123287672,
|
| 1753 |
+
"grad_norm": 0.7068540453910828,
|
| 1754 |
"learning_rate": 2.945205479452055e-05,
|
| 1755 |
+
"loss": 0.1959,
|
| 1756 |
+
"step": 1170
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 1.2328767123287672,
|
| 1760 |
+
"eval_loss": 0.3410193622112274,
|
| 1761 |
+
"eval_runtime": 13.5262,
|
| 1762 |
+
"eval_samples_per_second": 35.487,
|
| 1763 |
+
"eval_steps_per_second": 2.218,
|
| 1764 |
"step": 1170
|
| 1765 |
},
|
| 1766 |
{
|
| 1767 |
"epoch": 1.2434141201264488,
|
| 1768 |
+
"grad_norm": 0.6145871877670288,
|
| 1769 |
"learning_rate": 2.9276431331225852e-05,
|
| 1770 |
+
"loss": 0.1979,
|
| 1771 |
+
"step": 1180
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 1.2434141201264488,
|
| 1775 |
+
"eval_loss": 0.34399738907814026,
|
| 1776 |
+
"eval_runtime": 13.5178,
|
| 1777 |
+
"eval_samples_per_second": 35.509,
|
| 1778 |
+
"eval_steps_per_second": 2.219,
|
| 1779 |
"step": 1180
|
| 1780 |
},
|
| 1781 |
{
|
| 1782 |
"epoch": 1.2539515279241307,
|
| 1783 |
+
"grad_norm": 0.6426234841346741,
|
| 1784 |
"learning_rate": 2.9100807867931158e-05,
|
| 1785 |
"loss": 0.2077,
|
| 1786 |
"step": 1190
|
| 1787 |
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 1.2539515279241307,
|
| 1790 |
+
"eval_loss": 0.339756041765213,
|
| 1791 |
+
"eval_runtime": 13.5275,
|
| 1792 |
+
"eval_samples_per_second": 35.483,
|
| 1793 |
+
"eval_steps_per_second": 2.218,
|
| 1794 |
+
"step": 1190
|
| 1795 |
+
},
|
| 1796 |
{
|
| 1797 |
"epoch": 1.2644889357218125,
|
| 1798 |
+
"grad_norm": 0.6372554302215576,
|
| 1799 |
"learning_rate": 2.8925184404636463e-05,
|
| 1800 |
+
"loss": 0.1938,
|
| 1801 |
"step": 1200
|
| 1802 |
},
|
| 1803 |
{
|
| 1804 |
"epoch": 1.2644889357218125,
|
| 1805 |
+
"eval_loss": 0.3394372761249542,
|
| 1806 |
+
"eval_runtime": 13.5262,
|
| 1807 |
+
"eval_samples_per_second": 35.487,
|
| 1808 |
+
"eval_steps_per_second": 2.218,
|
| 1809 |
"step": 1200
|
| 1810 |
},
|
| 1811 |
{
|
| 1812 |
"epoch": 1.2750263435194942,
|
| 1813 |
+
"grad_norm": 0.49700266122817993,
|
| 1814 |
"learning_rate": 2.8749560941341762e-05,
|
| 1815 |
+
"loss": 0.1973,
|
| 1816 |
+
"step": 1210
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.2750263435194942,
|
| 1820 |
+
"eval_loss": 0.34174108505249023,
|
| 1821 |
+
"eval_runtime": 13.5375,
|
| 1822 |
+
"eval_samples_per_second": 35.457,
|
| 1823 |
+
"eval_steps_per_second": 2.216,
|
| 1824 |
"step": 1210
|
| 1825 |
},
|
| 1826 |
{
|
| 1827 |
"epoch": 1.285563751317176,
|
| 1828 |
+
"grad_norm": 0.6628106832504272,
|
| 1829 |
"learning_rate": 2.8573937478047068e-05,
|
| 1830 |
+
"loss": 0.1892,
|
| 1831 |
+
"step": 1220
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 1.285563751317176,
|
| 1835 |
+
"eval_loss": 0.3427627980709076,
|
| 1836 |
+
"eval_runtime": 13.5801,
|
| 1837 |
+
"eval_samples_per_second": 35.346,
|
| 1838 |
+
"eval_steps_per_second": 2.209,
|
| 1839 |
"step": 1220
|
| 1840 |
},
|
| 1841 |
{
|
| 1842 |
"epoch": 1.2961011591148577,
|
| 1843 |
+
"grad_norm": 0.6381083130836487,
|
| 1844 |
"learning_rate": 2.839831401475237e-05,
|
| 1845 |
+
"loss": 0.2023,
|
| 1846 |
+
"step": 1230
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 1.2961011591148577,
|
| 1850 |
+
"eval_loss": 0.3431331515312195,
|
| 1851 |
+
"eval_runtime": 13.5304,
|
| 1852 |
+
"eval_samples_per_second": 35.476,
|
| 1853 |
+
"eval_steps_per_second": 2.217,
|
| 1854 |
"step": 1230
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
"epoch": 1.3066385669125395,
|
| 1858 |
+
"grad_norm": 0.6037564873695374,
|
| 1859 |
"learning_rate": 2.8222690551457676e-05,
|
| 1860 |
+
"loss": 0.1925,
|
| 1861 |
+
"step": 1240
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 1.3066385669125395,
|
| 1865 |
+
"eval_loss": 0.3404982388019562,
|
| 1866 |
+
"eval_runtime": 13.5282,
|
| 1867 |
+
"eval_samples_per_second": 35.481,
|
| 1868 |
+
"eval_steps_per_second": 2.218,
|
| 1869 |
"step": 1240
|
| 1870 |
},
|
| 1871 |
{
|
| 1872 |
"epoch": 1.3171759747102212,
|
| 1873 |
+
"grad_norm": 0.6601694822311401,
|
| 1874 |
"learning_rate": 2.804706708816298e-05,
|
| 1875 |
+
"loss": 0.1926,
|
| 1876 |
"step": 1250
|
| 1877 |
},
|
| 1878 |
{
|
| 1879 |
"epoch": 1.3171759747102212,
|
| 1880 |
+
"eval_loss": 0.34181320667266846,
|
| 1881 |
+
"eval_runtime": 13.5621,
|
| 1882 |
+
"eval_samples_per_second": 35.393,
|
| 1883 |
+
"eval_steps_per_second": 2.212,
|
| 1884 |
"step": 1250
|
| 1885 |
},
|
| 1886 |
{
|
| 1887 |
"epoch": 1.327713382507903,
|
| 1888 |
+
"grad_norm": 0.5349809527397156,
|
| 1889 |
"learning_rate": 2.7871443624868284e-05,
|
| 1890 |
+
"loss": 0.1936,
|
| 1891 |
+
"step": 1260
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 1.327713382507903,
|
| 1895 |
+
"eval_loss": 0.34073132276535034,
|
| 1896 |
+
"eval_runtime": 13.6477,
|
| 1897 |
+
"eval_samples_per_second": 35.171,
|
| 1898 |
+
"eval_steps_per_second": 2.198,
|
| 1899 |
"step": 1260
|
| 1900 |
},
|
| 1901 |
{
|
| 1902 |
"epoch": 1.338250790305585,
|
| 1903 |
+
"grad_norm": 0.6491771340370178,
|
| 1904 |
"learning_rate": 2.769582016157359e-05,
|
| 1905 |
+
"loss": 0.1993,
|
| 1906 |
+
"step": 1270
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 1.338250790305585,
|
| 1910 |
+
"eval_loss": 0.34483978152275085,
|
| 1911 |
+
"eval_runtime": 13.5329,
|
| 1912 |
+
"eval_samples_per_second": 35.469,
|
| 1913 |
+
"eval_steps_per_second": 2.217,
|
| 1914 |
"step": 1270
|
| 1915 |
},
|
| 1916 |
{
|
| 1917 |
"epoch": 1.3487881981032666,
|
| 1918 |
+
"grad_norm": 0.6270915865898132,
|
| 1919 |
"learning_rate": 2.7520196698278895e-05,
|
| 1920 |
+
"loss": 0.2007,
|
| 1921 |
+
"step": 1280
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.3487881981032666,
|
| 1925 |
+
"eval_loss": 0.34615617990493774,
|
| 1926 |
+
"eval_runtime": 13.5185,
|
| 1927 |
+
"eval_samples_per_second": 35.507,
|
| 1928 |
+
"eval_steps_per_second": 2.219,
|
| 1929 |
"step": 1280
|
| 1930 |
},
|
| 1931 |
{
|
| 1932 |
"epoch": 1.3593256059009484,
|
| 1933 |
+
"grad_norm": 0.6853300333023071,
|
| 1934 |
"learning_rate": 2.7344573234984194e-05,
|
| 1935 |
+
"loss": 0.2131,
|
| 1936 |
+
"step": 1290
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 1.3593256059009484,
|
| 1940 |
+
"eval_loss": 0.3406980335712433,
|
| 1941 |
+
"eval_runtime": 13.5247,
|
| 1942 |
+
"eval_samples_per_second": 35.491,
|
| 1943 |
+
"eval_steps_per_second": 2.218,
|
| 1944 |
"step": 1290
|
| 1945 |
},
|
| 1946 |
{
|
| 1947 |
"epoch": 1.36986301369863,
|
| 1948 |
+
"grad_norm": 0.5863353610038757,
|
| 1949 |
"learning_rate": 2.71689497716895e-05,
|
| 1950 |
+
"loss": 0.1846,
|
| 1951 |
"step": 1300
|
| 1952 |
},
|
| 1953 |
{
|
| 1954 |
"epoch": 1.36986301369863,
|
| 1955 |
+
"eval_loss": 0.34115493297576904,
|
| 1956 |
+
"eval_runtime": 13.5104,
|
| 1957 |
+
"eval_samples_per_second": 35.528,
|
| 1958 |
+
"eval_steps_per_second": 2.221,
|
| 1959 |
"step": 1300
|
| 1960 |
},
|
| 1961 |
{
|
| 1962 |
"epoch": 1.380400421496312,
|
| 1963 |
+
"grad_norm": 0.7511035203933716,
|
| 1964 |
"learning_rate": 2.6993326308394802e-05,
|
| 1965 |
+
"loss": 0.1911,
|
| 1966 |
+
"step": 1310
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 1.380400421496312,
|
| 1970 |
+
"eval_loss": 0.34829092025756836,
|
| 1971 |
+
"eval_runtime": 13.5081,
|
| 1972 |
+
"eval_samples_per_second": 35.534,
|
| 1973 |
+
"eval_steps_per_second": 2.221,
|
| 1974 |
"step": 1310
|
| 1975 |
},
|
| 1976 |
{
|
| 1977 |
"epoch": 1.3909378292939936,
|
| 1978 |
+
"grad_norm": 0.6482399106025696,
|
| 1979 |
"learning_rate": 2.6817702845100107e-05,
|
| 1980 |
+
"loss": 0.2013,
|
| 1981 |
+
"step": 1320
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 1.3909378292939936,
|
| 1985 |
+
"eval_loss": 0.3465197682380676,
|
| 1986 |
+
"eval_runtime": 13.5316,
|
| 1987 |
+
"eval_samples_per_second": 35.473,
|
| 1988 |
+
"eval_steps_per_second": 2.217,
|
| 1989 |
"step": 1320
|
| 1990 |
},
|
| 1991 |
{
|
| 1992 |
"epoch": 1.4014752370916754,
|
| 1993 |
+
"grad_norm": 0.570846438407898,
|
| 1994 |
"learning_rate": 2.6642079381805413e-05,
|
| 1995 |
"loss": 0.1889,
|
| 1996 |
"step": 1330
|
| 1997 |
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 1.4014752370916754,
|
| 2000 |
+
"eval_loss": 0.34688571095466614,
|
| 2001 |
+
"eval_runtime": 13.5157,
|
| 2002 |
+
"eval_samples_per_second": 35.514,
|
| 2003 |
+
"eval_steps_per_second": 2.22,
|
| 2004 |
+
"step": 1330
|
| 2005 |
+
},
|
| 2006 |
{
|
| 2007 |
"epoch": 1.4120126448893573,
|
| 2008 |
+
"grad_norm": 0.6336379051208496,
|
| 2009 |
"learning_rate": 2.6466455918510712e-05,
|
| 2010 |
+
"loss": 0.1994,
|
| 2011 |
+
"step": 1340
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 1.4120126448893573,
|
| 2015 |
+
"eval_loss": 0.3466794788837433,
|
| 2016 |
+
"eval_runtime": 13.5071,
|
| 2017 |
+
"eval_samples_per_second": 35.537,
|
| 2018 |
+
"eval_steps_per_second": 2.221,
|
| 2019 |
"step": 1340
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
"epoch": 1.422550052687039,
|
| 2023 |
+
"grad_norm": 0.653731644153595,
|
| 2024 |
"learning_rate": 2.6290832455216018e-05,
|
| 2025 |
+
"loss": 0.1894,
|
| 2026 |
"step": 1350
|
| 2027 |
},
|
| 2028 |
{
|
| 2029 |
"epoch": 1.422550052687039,
|
| 2030 |
+
"eval_loss": 0.34787434339523315,
|
| 2031 |
+
"eval_runtime": 13.5062,
|
| 2032 |
+
"eval_samples_per_second": 35.539,
|
| 2033 |
+
"eval_steps_per_second": 2.221,
|
| 2034 |
"step": 1350
|
| 2035 |
},
|
| 2036 |
{
|
| 2037 |
"epoch": 1.4330874604847208,
|
| 2038 |
+
"grad_norm": 0.6545577049255371,
|
| 2039 |
"learning_rate": 2.611520899192132e-05,
|
| 2040 |
+
"loss": 0.1981,
|
| 2041 |
+
"step": 1360
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 1.4330874604847208,
|
| 2045 |
+
"eval_loss": 0.34745165705680847,
|
| 2046 |
+
"eval_runtime": 13.4977,
|
| 2047 |
+
"eval_samples_per_second": 35.562,
|
| 2048 |
+
"eval_steps_per_second": 2.223,
|
| 2049 |
"step": 1360
|
| 2050 |
},
|
| 2051 |
{
|
| 2052 |
"epoch": 1.4436248682824027,
|
| 2053 |
+
"grad_norm": 0.7751778364181519,
|
| 2054 |
"learning_rate": 2.5939585528626625e-05,
|
| 2055 |
+
"loss": 0.183,
|
| 2056 |
+
"step": 1370
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 1.4436248682824027,
|
| 2060 |
+
"eval_loss": 0.3481937348842621,
|
| 2061 |
+
"eval_runtime": 13.5303,
|
| 2062 |
+
"eval_samples_per_second": 35.476,
|
| 2063 |
+
"eval_steps_per_second": 2.217,
|
| 2064 |
"step": 1370
|
| 2065 |
},
|
| 2066 |
{
|
| 2067 |
"epoch": 1.4541622760800843,
|
| 2068 |
+
"grad_norm": 0.7329134345054626,
|
| 2069 |
"learning_rate": 2.576396206533193e-05,
|
| 2070 |
+
"loss": 0.1906,
|
| 2071 |
+
"step": 1380
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 1.4541622760800843,
|
| 2075 |
+
"eval_loss": 0.3493328094482422,
|
| 2076 |
+
"eval_runtime": 13.5155,
|
| 2077 |
+
"eval_samples_per_second": 35.515,
|
| 2078 |
+
"eval_steps_per_second": 2.22,
|
| 2079 |
"step": 1380
|
| 2080 |
},
|
| 2081 |
{
|
| 2082 |
"epoch": 1.464699683877766,
|
| 2083 |
+
"grad_norm": 0.7945630550384521,
|
| 2084 |
"learning_rate": 2.5588338602037233e-05,
|
| 2085 |
+
"loss": 0.1913,
|
| 2086 |
+
"step": 1390
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 1.464699683877766,
|
| 2090 |
+
"eval_loss": 0.3451732397079468,
|
| 2091 |
+
"eval_runtime": 13.5276,
|
| 2092 |
+
"eval_samples_per_second": 35.483,
|
| 2093 |
+
"eval_steps_per_second": 2.218,
|
| 2094 |
"step": 1390
|
| 2095 |
},
|
| 2096 |
{
|
| 2097 |
"epoch": 1.4752370916754478,
|
| 2098 |
+
"grad_norm": 0.7154361009597778,
|
| 2099 |
"learning_rate": 2.541271513874254e-05,
|
| 2100 |
"loss": 0.1853,
|
| 2101 |
"step": 1400
|
| 2102 |
},
|
| 2103 |
{
|
| 2104 |
"epoch": 1.4752370916754478,
|
| 2105 |
+
"eval_loss": 0.3447299301624298,
|
| 2106 |
+
"eval_runtime": 13.5329,
|
| 2107 |
+
"eval_samples_per_second": 35.469,
|
| 2108 |
+
"eval_steps_per_second": 2.217,
|
| 2109 |
"step": 1400
|
| 2110 |
},
|
| 2111 |
{
|
| 2112 |
"epoch": 1.4857744994731297,
|
| 2113 |
+
"grad_norm": 0.6370890140533447,
|
| 2114 |
"learning_rate": 2.5237091675447845e-05,
|
| 2115 |
+
"loss": 0.1822,
|
| 2116 |
+
"step": 1410
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 1.4857744994731297,
|
| 2120 |
+
"eval_loss": 0.34814396500587463,
|
| 2121 |
+
"eval_runtime": 13.5296,
|
| 2122 |
+
"eval_samples_per_second": 35.478,
|
| 2123 |
+
"eval_steps_per_second": 2.217,
|
| 2124 |
"step": 1410
|
| 2125 |
},
|
| 2126 |
{
|
| 2127 |
"epoch": 1.4963119072708113,
|
| 2128 |
+
"grad_norm": 0.7020454406738281,
|
| 2129 |
"learning_rate": 2.5061468212153144e-05,
|
| 2130 |
+
"loss": 0.1918,
|
| 2131 |
+
"step": 1420
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.4963119072708113,
|
| 2135 |
+
"eval_loss": 0.35007527470588684,
|
| 2136 |
+
"eval_runtime": 13.5235,
|
| 2137 |
+
"eval_samples_per_second": 35.494,
|
| 2138 |
+
"eval_steps_per_second": 2.218,
|
| 2139 |
"step": 1420
|
| 2140 |
},
|
| 2141 |
{
|
| 2142 |
"epoch": 1.5068493150684932,
|
| 2143 |
+
"grad_norm": 0.7104793787002563,
|
| 2144 |
"learning_rate": 2.4885844748858446e-05,
|
| 2145 |
+
"loss": 0.2008,
|
| 2146 |
+
"step": 1430
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 1.5068493150684932,
|
| 2150 |
+
"eval_loss": 0.3479911684989929,
|
| 2151 |
+
"eval_runtime": 13.5547,
|
| 2152 |
+
"eval_samples_per_second": 35.412,
|
| 2153 |
+
"eval_steps_per_second": 2.213,
|
| 2154 |
"step": 1430
|
| 2155 |
},
|
| 2156 |
{
|
| 2157 |
"epoch": 1.517386722866175,
|
| 2158 |
+
"grad_norm": 0.8473650217056274,
|
| 2159 |
"learning_rate": 2.4710221285563755e-05,
|
| 2160 |
+
"loss": 0.2042,
|
| 2161 |
+
"step": 1440
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 1.517386722866175,
|
| 2165 |
+
"eval_loss": 0.34348005056381226,
|
| 2166 |
+
"eval_runtime": 13.5204,
|
| 2167 |
+
"eval_samples_per_second": 35.502,
|
| 2168 |
+
"eval_steps_per_second": 2.219,
|
| 2169 |
"step": 1440
|
| 2170 |
},
|
| 2171 |
{
|
| 2172 |
"epoch": 1.5279241306638567,
|
| 2173 |
+
"grad_norm": 0.6656286716461182,
|
| 2174 |
"learning_rate": 2.4534597822269057e-05,
|
| 2175 |
+
"loss": 0.1968,
|
| 2176 |
"step": 1450
|
| 2177 |
},
|
| 2178 |
{
|
| 2179 |
"epoch": 1.5279241306638567,
|
| 2180 |
+
"eval_loss": 0.3460903763771057,
|
| 2181 |
+
"eval_runtime": 13.5325,
|
| 2182 |
+
"eval_samples_per_second": 35.47,
|
| 2183 |
+
"eval_steps_per_second": 2.217,
|
| 2184 |
"step": 1450
|
| 2185 |
},
|
| 2186 |
{
|
| 2187 |
"epoch": 1.5384615384615383,
|
| 2188 |
+
"grad_norm": 0.6165732145309448,
|
| 2189 |
"learning_rate": 2.435897435897436e-05,
|
| 2190 |
+
"loss": 0.1948,
|
| 2191 |
+
"step": 1460
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 1.5384615384615383,
|
| 2195 |
+
"eval_loss": 0.3462531864643097,
|
| 2196 |
+
"eval_runtime": 13.5278,
|
| 2197 |
+
"eval_samples_per_second": 35.483,
|
| 2198 |
+
"eval_steps_per_second": 2.218,
|
| 2199 |
"step": 1460
|
| 2200 |
},
|
| 2201 |
{
|
| 2202 |
"epoch": 1.5489989462592202,
|
| 2203 |
+
"grad_norm": 0.8803408741950989,
|
| 2204 |
"learning_rate": 2.418335089567966e-05,
|
| 2205 |
+
"loss": 0.1814,
|
| 2206 |
+
"step": 1470
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 1.5489989462592202,
|
| 2210 |
+
"eval_loss": 0.3515132665634155,
|
| 2211 |
+
"eval_runtime": 13.5309,
|
| 2212 |
+
"eval_samples_per_second": 35.474,
|
| 2213 |
+
"eval_steps_per_second": 2.217,
|
| 2214 |
"step": 1470
|
| 2215 |
},
|
| 2216 |
{
|
| 2217 |
"epoch": 1.559536354056902,
|
| 2218 |
+
"grad_norm": 0.628629744052887,
|
| 2219 |
"learning_rate": 2.4007727432384967e-05,
|
| 2220 |
+
"loss": 0.1894,
|
| 2221 |
+
"step": 1480
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 1.559536354056902,
|
| 2225 |
+
"eval_loss": 0.3480067551136017,
|
| 2226 |
+
"eval_runtime": 13.5314,
|
| 2227 |
+
"eval_samples_per_second": 35.473,
|
| 2228 |
+
"eval_steps_per_second": 2.217,
|
| 2229 |
"step": 1480
|
| 2230 |
},
|
| 2231 |
{
|
| 2232 |
"epoch": 1.5700737618545837,
|
| 2233 |
+
"grad_norm": 0.6435084939002991,
|
| 2234 |
"learning_rate": 2.3832103969090273e-05,
|
| 2235 |
+
"loss": 0.1898,
|
| 2236 |
+
"step": 1490
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.5700737618545837,
|
| 2240 |
+
"eval_loss": 0.3466918170452118,
|
| 2241 |
+
"eval_runtime": 13.5167,
|
| 2242 |
+
"eval_samples_per_second": 35.512,
|
| 2243 |
+
"eval_steps_per_second": 2.219,
|
| 2244 |
"step": 1490
|
| 2245 |
},
|
| 2246 |
{
|
| 2247 |
"epoch": 1.5806111696522656,
|
| 2248 |
+
"grad_norm": 0.5330148935317993,
|
| 2249 |
"learning_rate": 2.3656480505795575e-05,
|
| 2250 |
+
"loss": 0.188,
|
| 2251 |
"step": 1500
|
| 2252 |
},
|
| 2253 |
{
|
| 2254 |
"epoch": 1.5806111696522656,
|
| 2255 |
+
"eval_loss": 0.3500882387161255,
|
| 2256 |
+
"eval_runtime": 13.516,
|
| 2257 |
+
"eval_samples_per_second": 35.514,
|
| 2258 |
+
"eval_steps_per_second": 2.22,
|
| 2259 |
"step": 1500
|
| 2260 |
},
|
| 2261 |
{
|
| 2262 |
"epoch": 1.5911485774499474,
|
| 2263 |
+
"grad_norm": 0.5372351408004761,
|
| 2264 |
"learning_rate": 2.3480857042500877e-05,
|
| 2265 |
+
"loss": 0.1963,
|
| 2266 |
+
"step": 1510
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 1.5911485774499474,
|
| 2270 |
+
"eval_loss": 0.35102787613868713,
|
| 2271 |
+
"eval_runtime": 13.5059,
|
| 2272 |
+
"eval_samples_per_second": 35.54,
|
| 2273 |
+
"eval_steps_per_second": 2.221,
|
| 2274 |
"step": 1510
|
| 2275 |
},
|
| 2276 |
{
|
| 2277 |
"epoch": 1.601685985247629,
|
| 2278 |
+
"grad_norm": 0.6970928907394409,
|
| 2279 |
"learning_rate": 2.3305233579206183e-05,
|
| 2280 |
+
"loss": 0.194,
|
| 2281 |
+
"step": 1520
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 1.601685985247629,
|
| 2285 |
+
"eval_loss": 0.3498727083206177,
|
| 2286 |
+
"eval_runtime": 13.5044,
|
| 2287 |
+
"eval_samples_per_second": 35.544,
|
| 2288 |
+
"eval_steps_per_second": 2.221,
|
| 2289 |
"step": 1520
|
| 2290 |
},
|
| 2291 |
{
|
| 2292 |
"epoch": 1.6122233930453107,
|
| 2293 |
+
"grad_norm": 0.6487205028533936,
|
| 2294 |
"learning_rate": 2.312961011591149e-05,
|
| 2295 |
+
"loss": 0.1926,
|
| 2296 |
+
"step": 1530
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 1.6122233930453107,
|
| 2300 |
+
"eval_loss": 0.3481922745704651,
|
| 2301 |
+
"eval_runtime": 13.5029,
|
| 2302 |
+
"eval_samples_per_second": 35.548,
|
| 2303 |
+
"eval_steps_per_second": 2.222,
|
| 2304 |
"step": 1530
|
| 2305 |
},
|
| 2306 |
{
|
| 2307 |
"epoch": 1.6227608008429928,
|
| 2308 |
+
"grad_norm": 0.6962392330169678,
|
| 2309 |
"learning_rate": 2.295398665261679e-05,
|
| 2310 |
+
"loss": 0.1957,
|
| 2311 |
+
"step": 1540
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 1.6227608008429928,
|
| 2315 |
+
"eval_loss": 0.3489096462726593,
|
| 2316 |
+
"eval_runtime": 13.5,
|
| 2317 |
+
"eval_samples_per_second": 35.555,
|
| 2318 |
+
"eval_steps_per_second": 2.222,
|
| 2319 |
"step": 1540
|
| 2320 |
},
|
| 2321 |
{
|
| 2322 |
"epoch": 1.6332982086406744,
|
| 2323 |
+
"grad_norm": 0.9060853719711304,
|
| 2324 |
"learning_rate": 2.2778363189322093e-05,
|
| 2325 |
+
"loss": 0.2036,
|
| 2326 |
"step": 1550
|
| 2327 |
},
|
| 2328 |
{
|
| 2329 |
"epoch": 1.6332982086406744,
|
| 2330 |
+
"eval_loss": 0.34907764196395874,
|
| 2331 |
+
"eval_runtime": 13.4978,
|
| 2332 |
+
"eval_samples_per_second": 35.561,
|
| 2333 |
+
"eval_steps_per_second": 2.223,
|
| 2334 |
"step": 1550
|
| 2335 |
},
|
| 2336 |
{
|
| 2337 |
"epoch": 1.643835616438356,
|
| 2338 |
+
"grad_norm": 0.7215871810913086,
|
| 2339 |
"learning_rate": 2.2602739726027396e-05,
|
| 2340 |
+
"loss": 0.1932,
|
| 2341 |
+
"step": 1560
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.643835616438356,
|
| 2345 |
+
"eval_loss": 0.34920182824134827,
|
| 2346 |
+
"eval_runtime": 13.4995,
|
| 2347 |
+
"eval_samples_per_second": 35.557,
|
| 2348 |
+
"eval_steps_per_second": 2.222,
|
| 2349 |
"step": 1560
|
| 2350 |
},
|
| 2351 |
{
|
| 2352 |
"epoch": 1.654373024236038,
|
| 2353 |
+
"grad_norm": 0.9014811515808105,
|
| 2354 |
"learning_rate": 2.2427116262732705e-05,
|
| 2355 |
+
"loss": 0.1852,
|
| 2356 |
+
"step": 1570
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 1.654373024236038,
|
| 2360 |
+
"eval_loss": 0.3518800139427185,
|
| 2361 |
+
"eval_runtime": 13.4922,
|
| 2362 |
+
"eval_samples_per_second": 35.576,
|
| 2363 |
+
"eval_steps_per_second": 2.224,
|
| 2364 |
"step": 1570
|
| 2365 |
},
|
| 2366 |
{
|
| 2367 |
"epoch": 1.6649104320337198,
|
| 2368 |
+
"grad_norm": 0.8547218441963196,
|
| 2369 |
"learning_rate": 2.2251492799438007e-05,
|
| 2370 |
+
"loss": 0.179,
|
| 2371 |
+
"step": 1580
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 1.6649104320337198,
|
| 2375 |
+
"eval_loss": 0.3540193438529968,
|
| 2376 |
+
"eval_runtime": 13.4989,
|
| 2377 |
+
"eval_samples_per_second": 35.558,
|
| 2378 |
+
"eval_steps_per_second": 2.222,
|
| 2379 |
"step": 1580
|
| 2380 |
},
|
| 2381 |
{
|
| 2382 |
"epoch": 1.6754478398314014,
|
| 2383 |
+
"grad_norm": 0.7223723530769348,
|
| 2384 |
"learning_rate": 2.207586933614331e-05,
|
| 2385 |
+
"loss": 0.1871,
|
| 2386 |
+
"step": 1590
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 1.6754478398314014,
|
| 2390 |
+
"eval_loss": 0.3551488220691681,
|
| 2391 |
+
"eval_runtime": 13.5204,
|
| 2392 |
+
"eval_samples_per_second": 35.502,
|
| 2393 |
+
"eval_steps_per_second": 2.219,
|
| 2394 |
"step": 1590
|
| 2395 |
},
|
| 2396 |
{
|
| 2397 |
"epoch": 1.685985247629083,
|
| 2398 |
+
"grad_norm": 0.7338292002677917,
|
| 2399 |
"learning_rate": 2.190024587284861e-05,
|
| 2400 |
+
"loss": 0.1865,
|
| 2401 |
"step": 1600
|
| 2402 |
},
|
| 2403 |
{
|
| 2404 |
"epoch": 1.685985247629083,
|
| 2405 |
+
"eval_loss": 0.3537011742591858,
|
| 2406 |
+
"eval_runtime": 13.5153,
|
| 2407 |
+
"eval_samples_per_second": 35.515,
|
| 2408 |
+
"eval_steps_per_second": 2.22,
|
| 2409 |
"step": 1600
|
| 2410 |
},
|
| 2411 |
{
|
| 2412 |
"epoch": 1.6965226554267652,
|
| 2413 |
+
"grad_norm": 0.6816204190254211,
|
| 2414 |
"learning_rate": 2.1724622409553917e-05,
|
| 2415 |
+
"loss": 0.1859,
|
| 2416 |
+
"step": 1610
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 1.6965226554267652,
|
| 2420 |
+
"eval_loss": 0.3547250032424927,
|
| 2421 |
+
"eval_runtime": 13.5235,
|
| 2422 |
+
"eval_samples_per_second": 35.494,
|
| 2423 |
+
"eval_steps_per_second": 2.218,
|
| 2424 |
"step": 1610
|
| 2425 |
},
|
| 2426 |
{
|
| 2427 |
"epoch": 1.7070600632244468,
|
| 2428 |
+
"grad_norm": 0.6339895725250244,
|
| 2429 |
"learning_rate": 2.1548998946259223e-05,
|
| 2430 |
+
"loss": 0.1925,
|
| 2431 |
+
"step": 1620
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 1.7070600632244468,
|
| 2435 |
+
"eval_loss": 0.35097014904022217,
|
| 2436 |
+
"eval_runtime": 13.5126,
|
| 2437 |
+
"eval_samples_per_second": 35.522,
|
| 2438 |
+
"eval_steps_per_second": 2.22,
|
| 2439 |
"step": 1620
|
| 2440 |
},
|
| 2441 |
{
|
| 2442 |
"epoch": 1.7175974710221285,
|
| 2443 |
+
"grad_norm": 0.9739067554473877,
|
| 2444 |
"learning_rate": 2.1373375482964525e-05,
|
| 2445 |
+
"loss": 0.184,
|
| 2446 |
+
"step": 1630
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.7175974710221285,
|
| 2450 |
+
"eval_loss": 0.3519115746021271,
|
| 2451 |
+
"eval_runtime": 13.5322,
|
| 2452 |
+
"eval_samples_per_second": 35.471,
|
| 2453 |
+
"eval_steps_per_second": 2.217,
|
| 2454 |
"step": 1630
|
| 2455 |
},
|
| 2456 |
{
|
| 2457 |
"epoch": 1.7281348788198103,
|
| 2458 |
+
"grad_norm": 0.6770708560943604,
|
| 2459 |
"learning_rate": 2.1197752019669827e-05,
|
| 2460 |
+
"loss": 0.1799,
|
| 2461 |
+
"step": 1640
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 1.7281348788198103,
|
| 2465 |
+
"eval_loss": 0.35324668884277344,
|
| 2466 |
+
"eval_runtime": 13.526,
|
| 2467 |
+
"eval_samples_per_second": 35.487,
|
| 2468 |
+
"eval_steps_per_second": 2.218,
|
| 2469 |
"step": 1640
|
| 2470 |
},
|
| 2471 |
{
|
| 2472 |
"epoch": 1.7386722866174922,
|
| 2473 |
+
"grad_norm": 0.6459254622459412,
|
| 2474 |
"learning_rate": 2.1022128556375133e-05,
|
| 2475 |
+
"loss": 0.1805,
|
| 2476 |
"step": 1650
|
| 2477 |
},
|
| 2478 |
{
|
| 2479 |
"epoch": 1.7386722866174922,
|
| 2480 |
+
"eval_loss": 0.3586582839488983,
|
| 2481 |
+
"eval_runtime": 13.5461,
|
| 2482 |
+
"eval_samples_per_second": 35.434,
|
| 2483 |
+
"eval_steps_per_second": 2.215,
|
| 2484 |
"step": 1650
|
| 2485 |
},
|
| 2486 |
{
|
| 2487 |
"epoch": 1.7492096944151738,
|
| 2488 |
+
"grad_norm": 0.8696146607398987,
|
| 2489 |
"learning_rate": 2.084650509308044e-05,
|
| 2490 |
+
"loss": 0.1797,
|
| 2491 |
+
"step": 1660
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 1.7492096944151738,
|
| 2495 |
+
"eval_loss": 0.36139217019081116,
|
| 2496 |
+
"eval_runtime": 13.541,
|
| 2497 |
+
"eval_samples_per_second": 35.448,
|
| 2498 |
+
"eval_steps_per_second": 2.215,
|
| 2499 |
"step": 1660
|
| 2500 |
},
|
| 2501 |
{
|
| 2502 |
"epoch": 1.7597471022128557,
|
| 2503 |
+
"grad_norm": 0.7976329326629639,
|
| 2504 |
"learning_rate": 2.067088162978574e-05,
|
| 2505 |
+
"loss": 0.1837,
|
| 2506 |
+
"step": 1670
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 1.7597471022128557,
|
| 2510 |
+
"eval_loss": 0.35478851199150085,
|
| 2511 |
+
"eval_runtime": 13.5263,
|
| 2512 |
+
"eval_samples_per_second": 35.486,
|
| 2513 |
+
"eval_steps_per_second": 2.218,
|
| 2514 |
"step": 1670
|
| 2515 |
},
|
| 2516 |
{
|
| 2517 |
"epoch": 1.7702845100105375,
|
| 2518 |
+
"grad_norm": 0.6949226260185242,
|
| 2519 |
"learning_rate": 2.0495258166491043e-05,
|
| 2520 |
+
"loss": 0.1737,
|
| 2521 |
+
"step": 1680
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 1.7702845100105375,
|
| 2525 |
+
"eval_loss": 0.35503342747688293,
|
| 2526 |
+
"eval_runtime": 13.5265,
|
| 2527 |
+
"eval_samples_per_second": 35.486,
|
| 2528 |
+
"eval_steps_per_second": 2.218,
|
| 2529 |
"step": 1680
|
| 2530 |
},
|
| 2531 |
{
|
| 2532 |
"epoch": 1.7808219178082192,
|
| 2533 |
+
"grad_norm": 0.6455687284469604,
|
| 2534 |
"learning_rate": 2.0319634703196345e-05,
|
| 2535 |
+
"loss": 0.1901,
|
| 2536 |
+
"step": 1690
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 1.7808219178082192,
|
| 2540 |
+
"eval_loss": 0.35389164090156555,
|
| 2541 |
+
"eval_runtime": 13.5419,
|
| 2542 |
+
"eval_samples_per_second": 35.445,
|
| 2543 |
+
"eval_steps_per_second": 2.215,
|
| 2544 |
"step": 1690
|
| 2545 |
},
|
| 2546 |
{
|
| 2547 |
"epoch": 1.7913593256059008,
|
| 2548 |
+
"grad_norm": 0.7231824398040771,
|
| 2549 |
"learning_rate": 2.0144011239901654e-05,
|
| 2550 |
+
"loss": 0.1747,
|
| 2551 |
"step": 1700
|
| 2552 |
},
|
| 2553 |
{
|
| 2554 |
"epoch": 1.7913593256059008,
|
| 2555 |
+
"eval_loss": 0.3568112254142761,
|
| 2556 |
+
"eval_runtime": 13.5287,
|
| 2557 |
+
"eval_samples_per_second": 35.48,
|
| 2558 |
+
"eval_steps_per_second": 2.218,
|
| 2559 |
"step": 1700
|
| 2560 |
}
|
| 2561 |
],
|
|
|
|
| 2563 |
"max_steps": 2847,
|
| 2564 |
"num_input_tokens_seen": 0,
|
| 2565 |
"num_train_epochs": 3,
|
| 2566 |
+
"save_steps": 10,
|
| 2567 |
"stateful_callbacks": {
|
| 2568 |
"TrainerControl": {
|
| 2569 |
"args": {
|
checkpoint-1700/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b0de208437fea36dff15ca77d3c3e30091c141220f1de003345ef564472f4ff
|
| 3 |
size 5880
|