Training in progress, step 1700, checkpoint
Browse files
checkpoint-1700/adapter_config.json
CHANGED
|
@@ -25,11 +25,11 @@
|
|
| 25 |
"target_modules": [
|
| 26 |
"k_proj",
|
| 27 |
"q_proj",
|
|
|
|
| 28 |
"v_proj",
|
|
|
|
| 29 |
"up_proj",
|
| 30 |
-
"
|
| 31 |
-
"down_proj",
|
| 32 |
-
"o_proj"
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
|
|
|
| 25 |
"target_modules": [
|
| 26 |
"k_proj",
|
| 27 |
"q_proj",
|
| 28 |
+
"gate_proj",
|
| 29 |
"v_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
"up_proj",
|
| 32 |
+
"down_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 161533192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d51845d44410d8eb271f42c6aa5278a13039068d622376b00a71773fca66eb00
|
| 3 |
size 161533192
|
checkpoint-1700/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 323290986
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6117269567de848622d534dc019a600bb4958f250719752c4fac4e8a32a1118
|
| 3 |
size 323290986
|
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:a99dfa80a09e2a75a5b00784377e971d15bcac342659bc343a6d6527ca8bbf57
|
| 3 |
size 14244
|
checkpoint-1700/trainer_state.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.8027571580063628,
|
| 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": 0.010604453870625663,
|
| 13 |
-
"grad_norm": 2.
|
| 14 |
"learning_rate": 4.982325910215624e-05,
|
| 15 |
-
"loss": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.021208907741251327,
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
"learning_rate": 4.964651820431248e-05,
|
| 22 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.031813361611876985,
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
"learning_rate": 4.946977730646872e-05,
|
| 29 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.042417815482502653,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
"learning_rate": 4.929303640862496e-05,
|
| 36 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.053022269353128315,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
"learning_rate": 4.91162955107812e-05,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.053022269353128315,
|
| 48 |
-
"eval_loss": 0.
|
| 49 |
-
"eval_runtime":
|
| 50 |
-
"eval_samples_per_second":
|
| 51 |
-
"eval_steps_per_second": 2.
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.06362672322375397,
|
| 56 |
-
"grad_norm": 0.
|
| 57 |
"learning_rate": 4.893955461293744e-05,
|
| 58 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
"step": 60
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.07423117709437964,
|
| 63 |
-
"grad_norm": 0.
|
| 64 |
"learning_rate": 4.876281371509368e-05,
|
| 65 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
"step": 70
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.08483563096500531,
|
| 70 |
-
"grad_norm": 0.
|
| 71 |
"learning_rate": 4.858607281724991e-05,
|
| 72 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
"step": 80
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.09544008483563096,
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
"learning_rate": 4.8409331919406155e-05,
|
| 79 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
"step": 90
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.10604453870625663,
|
| 84 |
-
"grad_norm": 0.
|
| 85 |
"learning_rate": 4.823259102156239e-05,
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.10604453870625663,
|
| 91 |
-
"eval_loss": 0.
|
| 92 |
-
"eval_runtime": 12.
|
| 93 |
-
"eval_samples_per_second":
|
| 94 |
-
"eval_steps_per_second": 2.
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.11664899257688228,
|
| 99 |
-
"grad_norm": 0.
|
| 100 |
"learning_rate": 4.8055850123718633e-05,
|
| 101 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
"step": 110
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.12725344644750794,
|
| 106 |
-
"grad_norm": 0.
|
| 107 |
"learning_rate": 4.7879109225874866e-05,
|
| 108 |
"loss": 0.539,
|
| 109 |
"step": 120
|
| 110 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
{
|
| 112 |
"epoch": 0.1378579003181336,
|
| 113 |
-
"grad_norm": 0.
|
| 114 |
"learning_rate": 4.770236832803111e-05,
|
| 115 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
"step": 130
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.14846235418875928,
|
| 120 |
-
"grad_norm": 0.
|
| 121 |
"learning_rate": 4.7525627430187344e-05,
|
| 122 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
"step": 140
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.15906680805938495,
|
| 127 |
-
"grad_norm": 0.
|
| 128 |
"learning_rate": 4.734888653234359e-05,
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 150
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.15906680805938495,
|
| 134 |
-
"eval_loss": 0.
|
| 135 |
-
"eval_runtime": 12.
|
| 136 |
-
"eval_samples_per_second":
|
| 137 |
-
"eval_steps_per_second": 2.
|
| 138 |
"step": 150
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 0.16967126193001061,
|
| 142 |
-
"grad_norm": 0.
|
| 143 |
"learning_rate": 4.717214563449982e-05,
|
| 144 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 145 |
"step": 160
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 0.18027571580063625,
|
| 149 |
-
"grad_norm": 0.
|
| 150 |
"learning_rate": 4.699540473665606e-05,
|
| 151 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 152 |
"step": 170
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 0.19088016967126192,
|
| 156 |
-
"grad_norm": 0.
|
| 157 |
"learning_rate": 4.681866383881231e-05,
|
| 158 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 159 |
"step": 180
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 0.2014846235418876,
|
| 163 |
-
"grad_norm": 0.
|
| 164 |
"learning_rate": 4.664192294096854e-05,
|
| 165 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 166 |
"step": 190
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 0.21208907741251326,
|
| 170 |
-
"grad_norm": 0.
|
| 171 |
"learning_rate": 4.6465182043124786e-05,
|
| 172 |
-
"loss": 0.
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.21208907741251326,
|
| 177 |
-
"eval_loss": 0.
|
| 178 |
-
"eval_runtime": 12.
|
| 179 |
-
"eval_samples_per_second":
|
| 180 |
-
"eval_steps_per_second": 2.
|
| 181 |
"step": 200
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.22269353128313893,
|
| 185 |
-
"grad_norm": 0.
|
| 186 |
"learning_rate": 4.628844114528102e-05,
|
| 187 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 188 |
"step": 210
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 0.23329798515376457,
|
| 192 |
-
"grad_norm": 0.
|
| 193 |
"learning_rate": 4.6111700247437265e-05,
|
| 194 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 195 |
"step": 220
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 0.24390243902439024,
|
| 199 |
-
"grad_norm": 0.
|
| 200 |
"learning_rate": 4.59349593495935e-05,
|
| 201 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
"step": 230
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 0.2545068928950159,
|
| 206 |
-
"grad_norm": 0.
|
| 207 |
"learning_rate": 4.5758218451749736e-05,
|
| 208 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 209 |
"step": 240
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.2651113467656416,
|
| 213 |
-
"grad_norm": 0.
|
| 214 |
"learning_rate": 4.5581477553905975e-05,
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 250
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 0.2651113467656416,
|
| 220 |
-
"eval_loss": 0.
|
| 221 |
-
"eval_runtime": 12.
|
| 222 |
-
"eval_samples_per_second":
|
| 223 |
-
"eval_steps_per_second": 2.
|
| 224 |
"step": 250
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.2757158006362672,
|
| 228 |
-
"grad_norm": 0.
|
| 229 |
"learning_rate": 4.5404736656062215e-05,
|
| 230 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 231 |
"step": 260
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 0.2863202545068929,
|
| 235 |
-
"grad_norm": 0.
|
| 236 |
"learning_rate": 4.5227995758218454e-05,
|
| 237 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 238 |
"step": 270
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 0.29692470837751855,
|
| 242 |
-
"grad_norm": 0.
|
| 243 |
"learning_rate": 4.505125486037469e-05,
|
| 244 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 245 |
"step": 280
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 0.3075291622481442,
|
| 249 |
-
"grad_norm": 0.
|
| 250 |
"learning_rate": 4.487451396253093e-05,
|
| 251 |
"loss": 0.4891,
|
| 252 |
"step": 290
|
| 253 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 254 |
{
|
| 255 |
"epoch": 0.3181336161187699,
|
| 256 |
-
"grad_norm": 0.
|
| 257 |
"learning_rate": 4.469777306468717e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 300
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 0.3181336161187699,
|
| 263 |
-
"eval_loss": 0.
|
| 264 |
-
"eval_runtime": 12.
|
| 265 |
-
"eval_samples_per_second":
|
| 266 |
-
"eval_steps_per_second": 2.
|
| 267 |
"step": 300
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 0.32873806998939553,
|
| 271 |
-
"grad_norm": 0.
|
| 272 |
"learning_rate": 4.452103216684341e-05,
|
| 273 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 274 |
"step": 310
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 0.33934252386002123,
|
| 278 |
-
"grad_norm": 0.
|
| 279 |
"learning_rate": 4.434429126899965e-05,
|
| 280 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 281 |
"step": 320
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 0.34994697773064687,
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
"learning_rate": 4.416755037115589e-05,
|
| 287 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 288 |
"step": 330
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 0.3605514316012725,
|
| 292 |
-
"grad_norm": 0.
|
| 293 |
"learning_rate": 4.399080947331213e-05,
|
| 294 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 295 |
"step": 340
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 0.3711558854718982,
|
| 299 |
-
"grad_norm": 0.
|
| 300 |
"learning_rate": 4.381406857546837e-05,
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 350
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 0.3711558854718982,
|
| 306 |
-
"eval_loss": 0.
|
| 307 |
-
"eval_runtime": 12.
|
| 308 |
-
"eval_samples_per_second":
|
| 309 |
-
"eval_steps_per_second": 2.
|
| 310 |
"step": 350
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.38176033934252385,
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
"learning_rate": 4.3637327677624606e-05,
|
| 316 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 317 |
"step": 360
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.39236479321314954,
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
"learning_rate": 4.3460586779780846e-05,
|
| 323 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
"step": 370
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.4029692470837752,
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
"learning_rate": 4.328384588193708e-05,
|
| 330 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 331 |
"step": 380
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.4135737009544008,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 4.3107104984093324e-05,
|
| 337 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 338 |
"step": 390
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.4241781548250265,
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
"learning_rate": 4.2930364086249556e-05,
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 400
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.4241781548250265,
|
| 349 |
-
"eval_loss": 0.
|
| 350 |
-
"eval_runtime": 12.
|
| 351 |
-
"eval_samples_per_second":
|
| 352 |
-
"eval_steps_per_second": 2.
|
| 353 |
"step": 400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.43478260869565216,
|
| 357 |
-
"grad_norm": 0.
|
| 358 |
"learning_rate": 4.27536231884058e-05,
|
| 359 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 360 |
"step": 410
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 0.44538706256627786,
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
"learning_rate": 4.2576882290562035e-05,
|
| 366 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 367 |
"step": 420
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 0.4559915164369035,
|
| 371 |
-
"grad_norm": 0.
|
| 372 |
"learning_rate": 4.240014139271828e-05,
|
| 373 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 374 |
"step": 430
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.46659597030752914,
|
| 378 |
-
"grad_norm": 0.
|
| 379 |
"learning_rate": 4.222340049487451e-05,
|
| 380 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 381 |
"step": 440
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 0.47720042417815484,
|
| 385 |
-
"grad_norm": 0.
|
| 386 |
"learning_rate": 4.204665959703076e-05,
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 450
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 0.47720042417815484,
|
| 392 |
-
"eval_loss": 0.
|
| 393 |
-
"eval_runtime": 12.
|
| 394 |
-
"eval_samples_per_second": 39.
|
| 395 |
-
"eval_steps_per_second": 2.
|
| 396 |
"step": 450
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 0.4878048780487805,
|
| 400 |
-
"grad_norm": 0.
|
| 401 |
"learning_rate": 4.186991869918699e-05,
|
| 402 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 403 |
"step": 460
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 0.4984093319194062,
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
"learning_rate": 4.169317780134323e-05,
|
| 409 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 410 |
"step": 470
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 0.5090137857900318,
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
"learning_rate": 4.151643690349947e-05,
|
| 416 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 417 |
"step": 480
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.5196182396606575,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
"learning_rate": 4.133969600565571e-05,
|
| 423 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 424 |
"step": 490
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 0.5302226935312832,
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
"learning_rate": 4.116295510781195e-05,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 500
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.5302226935312832,
|
| 435 |
-
"eval_loss": 0.
|
| 436 |
-
"eval_runtime": 12.
|
| 437 |
-
"eval_samples_per_second": 39.
|
| 438 |
-
"eval_steps_per_second": 2.
|
| 439 |
"step": 500
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 0.5408271474019088,
|
| 443 |
-
"grad_norm": 0.
|
| 444 |
"learning_rate": 4.098621420996819e-05,
|
| 445 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 446 |
"step": 510
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 0.5514316012725344,
|
| 450 |
-
"grad_norm": 0.
|
| 451 |
"learning_rate": 4.080947331212443e-05,
|
| 452 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 453 |
"step": 520
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 0.5620360551431601,
|
| 457 |
-
"grad_norm": 0.
|
| 458 |
"learning_rate": 4.0632732414280666e-05,
|
| 459 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 460 |
"step": 530
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 0.5726405090137858,
|
| 464 |
-
"grad_norm": 0.
|
| 465 |
"learning_rate": 4.0455991516436905e-05,
|
| 466 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 467 |
"step": 540
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 0.5832449628844114,
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
"learning_rate": 4.0279250618593144e-05,
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 550
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 0.5832449628844114,
|
| 478 |
-
"eval_loss": 0.
|
| 479 |
-
"eval_runtime": 12.
|
| 480 |
-
"eval_samples_per_second": 39.
|
| 481 |
-
"eval_steps_per_second": 2.
|
| 482 |
"step": 550
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 0.5938494167550371,
|
| 486 |
-
"grad_norm": 0.
|
| 487 |
"learning_rate": 4.0102509720749383e-05,
|
| 488 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 489 |
"step": 560
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 0.6044538706256628,
|
| 493 |
-
"grad_norm": 0.
|
| 494 |
"learning_rate": 3.992576882290562e-05,
|
| 495 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 496 |
"step": 570
|
| 497 |
},
|
| 498 |
{
|
| 499 |
"epoch": 0.6150583244962884,
|
| 500 |
-
"grad_norm": 0.
|
| 501 |
"learning_rate": 3.974902792506186e-05,
|
| 502 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 503 |
"step": 580
|
| 504 |
},
|
| 505 |
{
|
| 506 |
"epoch": 0.6256627783669141,
|
| 507 |
-
"grad_norm": 0.
|
| 508 |
"learning_rate": 3.95722870272181e-05,
|
| 509 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 510 |
"step": 590
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 0.6362672322375398,
|
| 514 |
-
"grad_norm": 0.
|
| 515 |
"learning_rate": 3.939554612937434e-05,
|
| 516 |
-
"loss": 0.
|
| 517 |
"step": 600
|
| 518 |
},
|
| 519 |
{
|
| 520 |
"epoch": 0.6362672322375398,
|
| 521 |
-
"eval_loss": 0.
|
| 522 |
-
"eval_runtime": 12.
|
| 523 |
-
"eval_samples_per_second": 39.
|
| 524 |
-
"eval_steps_per_second": 2.
|
| 525 |
"step": 600
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 0.6468716861081655,
|
| 529 |
-
"grad_norm": 0.
|
| 530 |
"learning_rate": 3.921880523153057e-05,
|
| 531 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 532 |
"step": 610
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 0.6574761399787911,
|
| 536 |
-
"grad_norm": 0.
|
| 537 |
"learning_rate": 3.904206433368682e-05,
|
| 538 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 539 |
"step": 620
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 0.6680805938494168,
|
| 543 |
-
"grad_norm": 0.
|
| 544 |
"learning_rate": 3.886532343584305e-05,
|
| 545 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 546 |
"step": 630
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 0.6786850477200425,
|
| 550 |
-
"grad_norm": 0.
|
| 551 |
"learning_rate": 3.86885825379993e-05,
|
| 552 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 553 |
"step": 640
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 0.689289501590668,
|
| 557 |
-
"grad_norm": 0.
|
| 558 |
"learning_rate": 3.851184164015553e-05,
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 650
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 0.689289501590668,
|
| 564 |
-
"eval_loss": 0.
|
| 565 |
-
"eval_runtime": 12.
|
| 566 |
-
"eval_samples_per_second": 39.
|
| 567 |
-
"eval_steps_per_second": 2.
|
| 568 |
"step": 650
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 0.6998939554612937,
|
| 572 |
-
"grad_norm": 0.
|
| 573 |
"learning_rate": 3.8335100742311775e-05,
|
| 574 |
"loss": 0.4426,
|
| 575 |
"step": 660
|
| 576 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 577 |
{
|
| 578 |
"epoch": 0.7104984093319194,
|
| 579 |
-
"grad_norm": 0.
|
| 580 |
"learning_rate": 3.8158359844468015e-05,
|
| 581 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 582 |
"step": 670
|
| 583 |
},
|
| 584 |
{
|
| 585 |
"epoch": 0.721102863202545,
|
| 586 |
-
"grad_norm": 0.
|
| 587 |
"learning_rate": 3.7981618946624254e-05,
|
| 588 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 589 |
"step": 680
|
| 590 |
},
|
| 591 |
{
|
| 592 |
"epoch": 0.7317073170731707,
|
| 593 |
-
"grad_norm": 0.
|
| 594 |
"learning_rate": 3.780487804878049e-05,
|
| 595 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 596 |
"step": 690
|
| 597 |
},
|
| 598 |
{
|
| 599 |
"epoch": 0.7423117709437964,
|
| 600 |
-
"grad_norm": 0.
|
| 601 |
"learning_rate": 3.7628137150936725e-05,
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 700
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 0.7423117709437964,
|
| 607 |
-
"eval_loss": 0.
|
| 608 |
-
"eval_runtime": 12.
|
| 609 |
-
"eval_samples_per_second":
|
| 610 |
-
"eval_steps_per_second": 2.
|
| 611 |
"step": 700
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 0.7529162248144221,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
"learning_rate": 3.745139625309297e-05,
|
| 617 |
"loss": 0.4418,
|
| 618 |
"step": 710
|
| 619 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 620 |
{
|
| 621 |
"epoch": 0.7635206786850477,
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
"learning_rate": 3.7274655355249204e-05,
|
| 624 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 625 |
"step": 720
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 0.7741251325556734,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
"learning_rate": 3.709791445740545e-05,
|
| 631 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 632 |
"step": 730
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 0.7847295864262991,
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
"learning_rate": 3.692117355956168e-05,
|
| 638 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 639 |
"step": 740
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 0.7953340402969247,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
"learning_rate": 3.674443266171793e-05,
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 750
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 0.7953340402969247,
|
| 650 |
-
"eval_loss": 0.
|
| 651 |
-
"eval_runtime": 12.
|
| 652 |
-
"eval_samples_per_second":
|
| 653 |
-
"eval_steps_per_second": 2.
|
| 654 |
"step": 750
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 0.8059384941675504,
|
| 658 |
-
"grad_norm": 0.
|
| 659 |
"learning_rate": 3.656769176387416e-05,
|
| 660 |
"loss": 0.4087,
|
| 661 |
"step": 760
|
| 662 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 663 |
{
|
| 664 |
"epoch": 0.8165429480381761,
|
| 665 |
-
"grad_norm": 0.
|
| 666 |
"learning_rate": 3.63909508660304e-05,
|
| 667 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 668 |
"step": 770
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 0.8271474019088016,
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
"learning_rate": 3.621420996818664e-05,
|
| 674 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 675 |
"step": 780
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 0.8377518557794273,
|
| 679 |
-
"grad_norm": 0.
|
| 680 |
"learning_rate": 3.603746907034288e-05,
|
| 681 |
-
"loss": 0.
|
| 682 |
"step": 790
|
| 683 |
},
|
| 684 |
{
|
| 685 |
-
"epoch": 0.
|
| 686 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 687 |
"learning_rate": 3.586072817249912e-05,
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 0.848356309650053,
|
| 693 |
-
"eval_loss": 0.
|
| 694 |
-
"eval_runtime": 12.
|
| 695 |
-
"eval_samples_per_second":
|
| 696 |
-
"eval_steps_per_second": 2.
|
| 697 |
"step": 800
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 0.8589607635206787,
|
| 701 |
-
"grad_norm": 0.
|
| 702 |
"learning_rate": 3.5683987274655356e-05,
|
| 703 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 704 |
"step": 810
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 0.8695652173913043,
|
| 708 |
-
"grad_norm": 0.
|
| 709 |
"learning_rate": 3.5507246376811596e-05,
|
| 710 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 711 |
"step": 820
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 0.88016967126193,
|
| 715 |
-
"grad_norm": 0.
|
| 716 |
"learning_rate": 3.5330505478967835e-05,
|
| 717 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 718 |
"step": 830
|
| 719 |
},
|
| 720 |
{
|
| 721 |
"epoch": 0.8907741251325557,
|
| 722 |
-
"grad_norm": 0.
|
| 723 |
"learning_rate": 3.5153764581124074e-05,
|
| 724 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 725 |
"step": 840
|
| 726 |
},
|
| 727 |
{
|
| 728 |
"epoch": 0.9013785790031813,
|
| 729 |
-
"grad_norm": 0.
|
| 730 |
"learning_rate": 3.497702368328031e-05,
|
| 731 |
-
"loss": 0.
|
| 732 |
"step": 850
|
| 733 |
},
|
| 734 |
{
|
| 735 |
"epoch": 0.9013785790031813,
|
| 736 |
-
"eval_loss": 0.
|
| 737 |
-
"eval_runtime": 12.
|
| 738 |
-
"eval_samples_per_second":
|
| 739 |
-
"eval_steps_per_second": 2.
|
| 740 |
"step": 850
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 0.911983032873807,
|
| 744 |
-
"grad_norm": 0.
|
| 745 |
"learning_rate": 3.480028278543655e-05,
|
| 746 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 747 |
"step": 860
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 0.9225874867444327,
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
"learning_rate": 3.462354188759279e-05,
|
| 753 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 754 |
"step": 870
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 0.9331919406150583,
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
"learning_rate": 3.444680098974903e-05,
|
| 760 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 761 |
"step": 880
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.943796394485684,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
"learning_rate": 3.427006009190527e-05,
|
| 767 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 768 |
"step": 890
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 0.9544008483563097,
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
"learning_rate": 3.409331919406151e-05,
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 900
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 0.9544008483563097,
|
| 779 |
-
"eval_loss": 0.
|
| 780 |
-
"eval_runtime": 12.
|
| 781 |
-
"eval_samples_per_second":
|
| 782 |
-
"eval_steps_per_second": 2.
|
| 783 |
"step": 900
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 0.9650053022269353,
|
| 787 |
-
"grad_norm": 0.
|
| 788 |
"learning_rate": 3.391657829621775e-05,
|
| 789 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 790 |
"step": 910
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 0.975609756097561,
|
| 794 |
-
"grad_norm": 0.
|
| 795 |
"learning_rate": 3.373983739837399e-05,
|
| 796 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 797 |
"step": 920
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 0.9862142099681867,
|
| 801 |
-
"grad_norm": 0.
|
| 802 |
"learning_rate": 3.356309650053022e-05,
|
| 803 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 804 |
"step": 930
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 0.9968186638388123,
|
| 808 |
-
"grad_norm": 0.
|
| 809 |
"learning_rate": 3.3386355602686466e-05,
|
| 810 |
"loss": 0.4208,
|
| 811 |
"step": 940
|
| 812 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 813 |
{
|
| 814 |
"epoch": 1.007423117709438,
|
| 815 |
-
"grad_norm": 0.
|
| 816 |
"learning_rate": 3.32096147048427e-05,
|
| 817 |
-
"loss": 0.
|
| 818 |
"step": 950
|
| 819 |
},
|
| 820 |
{
|
| 821 |
"epoch": 1.007423117709438,
|
| 822 |
-
"eval_loss": 0.
|
| 823 |
-
"eval_runtime": 12.
|
| 824 |
-
"eval_samples_per_second": 39.
|
| 825 |
-
"eval_steps_per_second": 2.
|
| 826 |
"step": 950
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 1.0180275715800637,
|
| 830 |
-
"grad_norm": 0.
|
| 831 |
"learning_rate": 3.3032873806998944e-05,
|
| 832 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 833 |
"step": 960
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 1.0286320254506893,
|
| 837 |
-
"grad_norm": 0.
|
| 838 |
"learning_rate": 3.285613290915518e-05,
|
| 839 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 840 |
"step": 970
|
| 841 |
},
|
| 842 |
{
|
| 843 |
"epoch": 1.039236479321315,
|
| 844 |
-
"grad_norm": 0.
|
| 845 |
"learning_rate": 3.267939201131142e-05,
|
| 846 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 847 |
"step": 980
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 1.0498409331919407,
|
| 851 |
-
"grad_norm": 0.
|
| 852 |
"learning_rate": 3.2502651113467655e-05,
|
| 853 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 854 |
"step": 990
|
| 855 |
},
|
| 856 |
{
|
| 857 |
"epoch": 1.0604453870625663,
|
| 858 |
-
"grad_norm": 0.
|
| 859 |
"learning_rate": 3.2325910215623894e-05,
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 1000
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 1.0604453870625663,
|
| 865 |
-
"eval_loss": 0.
|
| 866 |
-
"eval_runtime": 12.
|
| 867 |
-
"eval_samples_per_second": 39.
|
| 868 |
-
"eval_steps_per_second": 2.
|
| 869 |
"step": 1000
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 1.0710498409331919,
|
| 873 |
-
"grad_norm": 0.
|
| 874 |
"learning_rate": 3.2149169317780133e-05,
|
| 875 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 876 |
"step": 1010
|
| 877 |
},
|
| 878 |
{
|
| 879 |
"epoch": 1.0816542948038177,
|
| 880 |
-
"grad_norm": 0.
|
| 881 |
"learning_rate": 3.197242841993637e-05,
|
| 882 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 883 |
"step": 1020
|
| 884 |
},
|
| 885 |
{
|
| 886 |
"epoch": 1.0922587486744433,
|
| 887 |
-
"grad_norm": 0.
|
| 888 |
"learning_rate": 3.179568752209262e-05,
|
| 889 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 890 |
"step": 1030
|
| 891 |
},
|
| 892 |
{
|
| 893 |
"epoch": 1.1028632025450689,
|
| 894 |
-
"grad_norm": 0.
|
| 895 |
"learning_rate": 3.161894662424885e-05,
|
| 896 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 897 |
"step": 1040
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 1.1134676564156947,
|
| 901 |
-
"grad_norm": 0.
|
| 902 |
"learning_rate": 3.14422057264051e-05,
|
| 903 |
-
"loss": 0.
|
| 904 |
"step": 1050
|
| 905 |
},
|
| 906 |
{
|
| 907 |
"epoch": 1.1134676564156947,
|
| 908 |
-
"eval_loss": 0.
|
| 909 |
-
"eval_runtime": 12.
|
| 910 |
-
"eval_samples_per_second": 39.
|
| 911 |
-
"eval_steps_per_second": 2.
|
| 912 |
"step": 1050
|
| 913 |
},
|
| 914 |
{
|
| 915 |
"epoch": 1.1240721102863203,
|
| 916 |
-
"grad_norm": 0.
|
| 917 |
"learning_rate": 3.126546482856133e-05,
|
| 918 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 919 |
"step": 1060
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 1.1346765641569458,
|
| 923 |
-
"grad_norm": 0.
|
| 924 |
"learning_rate": 3.108872393071757e-05,
|
| 925 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 926 |
"step": 1070
|
| 927 |
},
|
| 928 |
{
|
| 929 |
"epoch": 1.1452810180275717,
|
| 930 |
-
"grad_norm": 0.
|
| 931 |
"learning_rate": 3.091198303287381e-05,
|
| 932 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 933 |
"step": 1080
|
| 934 |
},
|
| 935 |
{
|
| 936 |
"epoch": 1.1558854718981972,
|
| 937 |
-
"grad_norm": 0.
|
| 938 |
"learning_rate": 3.073524213503005e-05,
|
| 939 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 940 |
"step": 1090
|
| 941 |
},
|
| 942 |
{
|
| 943 |
"epoch": 1.1664899257688228,
|
| 944 |
-
"grad_norm": 0.
|
| 945 |
"learning_rate": 3.0558501237186286e-05,
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 1100
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 1.1664899257688228,
|
| 951 |
-
"eval_loss": 0.
|
| 952 |
-
"eval_runtime": 12.
|
| 953 |
-
"eval_samples_per_second": 39.
|
| 954 |
-
"eval_steps_per_second": 2.
|
| 955 |
"step": 1100
|
| 956 |
},
|
| 957 |
{
|
| 958 |
"epoch": 1.1770943796394486,
|
| 959 |
-
"grad_norm": 0.
|
| 960 |
"learning_rate": 3.0381760339342525e-05,
|
| 961 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 962 |
"step": 1110
|
| 963 |
},
|
| 964 |
{
|
| 965 |
"epoch": 1.1876988335100742,
|
| 966 |
-
"grad_norm": 0.
|
| 967 |
"learning_rate": 3.0205019441498768e-05,
|
| 968 |
"loss": 0.3706,
|
| 969 |
"step": 1120
|
| 970 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 971 |
{
|
| 972 |
"epoch": 1.1983032873806998,
|
| 973 |
-
"grad_norm": 0.
|
| 974 |
"learning_rate": 3.0028278543655004e-05,
|
| 975 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 976 |
"step": 1130
|
| 977 |
},
|
| 978 |
{
|
| 979 |
"epoch": 1.2089077412513256,
|
| 980 |
-
"grad_norm": 0.
|
| 981 |
"learning_rate": 2.985153764581124e-05,
|
| 982 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 983 |
"step": 1140
|
| 984 |
},
|
| 985 |
{
|
| 986 |
"epoch": 1.2195121951219512,
|
| 987 |
-
"grad_norm": 0.
|
| 988 |
"learning_rate": 2.9674796747967482e-05,
|
| 989 |
-
"loss": 0.
|
| 990 |
"step": 1150
|
| 991 |
},
|
| 992 |
{
|
| 993 |
"epoch": 1.2195121951219512,
|
| 994 |
-
"eval_loss": 0.
|
| 995 |
-
"eval_runtime": 12.
|
| 996 |
-
"eval_samples_per_second": 39.
|
| 997 |
-
"eval_steps_per_second": 2.
|
| 998 |
"step": 1150
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
"epoch": 1.2301166489925768,
|
| 1002 |
-
"grad_norm": 0.
|
| 1003 |
"learning_rate": 2.9498055850123718e-05,
|
| 1004 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1005 |
"step": 1160
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
"epoch": 1.2407211028632026,
|
| 1009 |
-
"grad_norm": 1.
|
| 1010 |
"learning_rate": 2.932131495227996e-05,
|
| 1011 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1012 |
"step": 1170
|
| 1013 |
},
|
| 1014 |
{
|
| 1015 |
"epoch": 1.2513255567338282,
|
| 1016 |
-
"grad_norm": 1.
|
| 1017 |
"learning_rate": 2.9144574054436196e-05,
|
| 1018 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1019 |
"step": 1180
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
"epoch": 1.2619300106044538,
|
| 1023 |
-
"grad_norm": 1.
|
| 1024 |
"learning_rate": 2.896783315659244e-05,
|
| 1025 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1026 |
"step": 1190
|
| 1027 |
},
|
| 1028 |
{
|
| 1029 |
"epoch": 1.2725344644750796,
|
| 1030 |
-
"grad_norm": 0.
|
| 1031 |
"learning_rate": 2.8791092258748675e-05,
|
| 1032 |
-
"loss": 0.
|
| 1033 |
"step": 1200
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 1.2725344644750796,
|
| 1037 |
-
"eval_loss": 0.
|
| 1038 |
-
"eval_runtime": 12.
|
| 1039 |
-
"eval_samples_per_second":
|
| 1040 |
-
"eval_steps_per_second": 2.
|
| 1041 |
"step": 1200
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
"epoch": 1.2831389183457051,
|
| 1045 |
-
"grad_norm": 0.
|
| 1046 |
"learning_rate": 2.8614351360904917e-05,
|
| 1047 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1048 |
"step": 1210
|
| 1049 |
},
|
| 1050 |
{
|
| 1051 |
"epoch": 1.2937433722163307,
|
| 1052 |
-
"grad_norm": 0.
|
| 1053 |
"learning_rate": 2.8437610463061153e-05,
|
| 1054 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1055 |
"step": 1220
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
"epoch": 1.3043478260869565,
|
| 1059 |
-
"grad_norm": 0.
|
| 1060 |
"learning_rate": 2.826086956521739e-05,
|
| 1061 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1062 |
"step": 1230
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
"epoch": 1.3149522799575821,
|
| 1066 |
-
"grad_norm": 1.
|
| 1067 |
"learning_rate": 2.808412866737363e-05,
|
| 1068 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1069 |
"step": 1240
|
| 1070 |
},
|
| 1071 |
{
|
| 1072 |
"epoch": 1.3255567338282077,
|
| 1073 |
-
"grad_norm": 1.
|
| 1074 |
"learning_rate": 2.790738776952987e-05,
|
| 1075 |
-
"loss": 0.
|
| 1076 |
"step": 1250
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
"epoch": 1.3255567338282077,
|
| 1080 |
-
"eval_loss": 0.
|
| 1081 |
-
"eval_runtime": 12.
|
| 1082 |
-
"eval_samples_per_second":
|
| 1083 |
-
"eval_steps_per_second": 2.
|
| 1084 |
"step": 1250
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
"epoch": 1.3361611876988335,
|
| 1088 |
-
"grad_norm": 0.
|
| 1089 |
"learning_rate": 2.773064687168611e-05,
|
| 1090 |
"loss": 0.3554,
|
| 1091 |
"step": 1260
|
| 1092 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1093 |
{
|
| 1094 |
"epoch": 1.346765641569459,
|
| 1095 |
-
"grad_norm": 0.
|
| 1096 |
"learning_rate": 2.755390597384235e-05,
|
| 1097 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1098 |
"step": 1270
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
"epoch": 1.357370095440085,
|
| 1102 |
-
"grad_norm": 1.
|
| 1103 |
"learning_rate": 2.7377165075998588e-05,
|
| 1104 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1105 |
"step": 1280
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 1.3679745493107105,
|
| 1109 |
-
"grad_norm": 0.
|
| 1110 |
"learning_rate": 2.7200424178154827e-05,
|
| 1111 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1112 |
"step": 1290
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
"epoch": 1.378579003181336,
|
| 1116 |
-
"grad_norm": 0.
|
| 1117 |
"learning_rate": 2.7023683280311063e-05,
|
| 1118 |
-
"loss": 0.
|
| 1119 |
"step": 1300
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 1.378579003181336,
|
| 1123 |
-
"eval_loss": 0.
|
| 1124 |
-
"eval_runtime": 12.
|
| 1125 |
-
"eval_samples_per_second":
|
| 1126 |
-
"eval_steps_per_second": 2.
|
| 1127 |
"step": 1300
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
"epoch": 1.389183457051962,
|
| 1131 |
-
"grad_norm": 1.
|
| 1132 |
"learning_rate": 2.6846942382467306e-05,
|
| 1133 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1134 |
"step": 1310
|
| 1135 |
},
|
| 1136 |
{
|
| 1137 |
"epoch": 1.3997879109225875,
|
| 1138 |
-
"grad_norm":
|
| 1139 |
"learning_rate": 2.667020148462354e-05,
|
| 1140 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1141 |
"step": 1320
|
| 1142 |
},
|
| 1143 |
{
|
| 1144 |
"epoch": 1.4103923647932133,
|
| 1145 |
-
"grad_norm": 1.
|
| 1146 |
"learning_rate": 2.6493460586779784e-05,
|
| 1147 |
"loss": 0.3619,
|
| 1148 |
"step": 1330
|
| 1149 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1150 |
{
|
| 1151 |
"epoch": 1.4209968186638389,
|
| 1152 |
-
"grad_norm": 0.
|
| 1153 |
"learning_rate": 2.631671968893602e-05,
|
| 1154 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1155 |
"step": 1340
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
"epoch": 1.4316012725344645,
|
| 1159 |
-
"grad_norm": 1.
|
| 1160 |
"learning_rate": 2.6139978791092262e-05,
|
| 1161 |
-
"loss": 0.
|
| 1162 |
"step": 1350
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
"epoch": 1.4316012725344645,
|
| 1166 |
-
"eval_loss": 0.
|
| 1167 |
-
"eval_runtime": 12.
|
| 1168 |
-
"eval_samples_per_second":
|
| 1169 |
-
"eval_steps_per_second": 2.
|
| 1170 |
"step": 1350
|
| 1171 |
},
|
| 1172 |
{
|
| 1173 |
"epoch": 1.4422057264050903,
|
| 1174 |
-
"grad_norm": 1.
|
| 1175 |
"learning_rate": 2.5963237893248498e-05,
|
| 1176 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1177 |
"step": 1360
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
"epoch": 1.4528101802757158,
|
| 1181 |
-
"grad_norm": 0.
|
| 1182 |
"learning_rate": 2.5786496995404734e-05,
|
| 1183 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1184 |
"step": 1370
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
"epoch": 1.4634146341463414,
|
| 1188 |
-
"grad_norm": 0.
|
| 1189 |
"learning_rate": 2.5609756097560977e-05,
|
| 1190 |
"loss": 0.3786,
|
| 1191 |
"step": 1380
|
| 1192 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1193 |
{
|
| 1194 |
"epoch": 1.4740190880169672,
|
| 1195 |
-
"grad_norm": 1.
|
| 1196 |
"learning_rate": 2.5433015199717212e-05,
|
| 1197 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1198 |
"step": 1390
|
| 1199 |
},
|
| 1200 |
{
|
| 1201 |
"epoch": 1.4846235418875928,
|
| 1202 |
-
"grad_norm": 1.
|
| 1203 |
"learning_rate": 2.5256274301873455e-05,
|
| 1204 |
-
"loss": 0.
|
| 1205 |
"step": 1400
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
"epoch": 1.4846235418875928,
|
| 1209 |
-
"eval_loss": 0.
|
| 1210 |
-
"eval_runtime": 12.
|
| 1211 |
-
"eval_samples_per_second":
|
| 1212 |
-
"eval_steps_per_second": 2.
|
| 1213 |
"step": 1400
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 1.4952279957582184,
|
| 1217 |
-
"grad_norm": 1.
|
| 1218 |
"learning_rate": 2.507953340402969e-05,
|
| 1219 |
"loss": 0.3491,
|
| 1220 |
"step": 1410
|
| 1221 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1222 |
{
|
| 1223 |
"epoch": 1.5058324496288442,
|
| 1224 |
-
"grad_norm": 1.
|
| 1225 |
"learning_rate": 2.4902792506185933e-05,
|
| 1226 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1227 |
"step": 1420
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 1.5164369034994698,
|
| 1231 |
-
"grad_norm": 1.
|
| 1232 |
"learning_rate": 2.4726051608342173e-05,
|
| 1233 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1234 |
"step": 1430
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
"epoch": 1.5270413573700954,
|
| 1238 |
-
"grad_norm": 1.
|
| 1239 |
"learning_rate": 2.4549310710498412e-05,
|
| 1240 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1241 |
"step": 1440
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 1.5376458112407212,
|
| 1245 |
-
"grad_norm": 0.
|
| 1246 |
"learning_rate": 2.437256981265465e-05,
|
| 1247 |
-
"loss": 0.
|
| 1248 |
"step": 1450
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
"epoch": 1.5376458112407212,
|
| 1252 |
-
"eval_loss": 0.
|
| 1253 |
-
"eval_runtime": 12.
|
| 1254 |
-
"eval_samples_per_second":
|
| 1255 |
-
"eval_steps_per_second": 2.
|
| 1256 |
"step": 1450
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
"epoch": 1.5482502651113468,
|
| 1260 |
-
"grad_norm": 1.
|
| 1261 |
"learning_rate": 2.419582891481089e-05,
|
| 1262 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1263 |
"step": 1460
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
"epoch": 1.5588547189819724,
|
| 1267 |
-
"grad_norm": 1.
|
| 1268 |
"learning_rate": 2.401908801696713e-05,
|
| 1269 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1270 |
"step": 1470
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
"epoch": 1.5694591728525982,
|
| 1274 |
-
"grad_norm": 0.
|
| 1275 |
"learning_rate": 2.3842347119123365e-05,
|
| 1276 |
"loss": 0.3645,
|
| 1277 |
"step": 1480
|
| 1278 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1279 |
{
|
| 1280 |
"epoch": 1.5800636267232238,
|
| 1281 |
-
"grad_norm": 0.
|
| 1282 |
"learning_rate": 2.3665606221279604e-05,
|
| 1283 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1284 |
"step": 1490
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
"epoch": 1.5906680805938493,
|
| 1288 |
-
"grad_norm": 1.
|
| 1289 |
"learning_rate": 2.3488865323435844e-05,
|
| 1290 |
-
"loss": 0.
|
| 1291 |
"step": 1500
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
"epoch": 1.5906680805938493,
|
| 1295 |
-
"eval_loss": 0.
|
| 1296 |
-
"eval_runtime": 12.
|
| 1297 |
-
"eval_samples_per_second": 39.
|
| 1298 |
-
"eval_steps_per_second": 2.
|
| 1299 |
"step": 1500
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
"epoch": 1.6012725344644752,
|
| 1303 |
-
"grad_norm": 1.
|
| 1304 |
"learning_rate": 2.3312124425592083e-05,
|
| 1305 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1306 |
"step": 1510
|
| 1307 |
},
|
| 1308 |
{
|
| 1309 |
"epoch": 1.6118769883351007,
|
| 1310 |
-
"grad_norm": 0.
|
| 1311 |
"learning_rate": 2.3135383527748322e-05,
|
| 1312 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1313 |
"step": 1520
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
"epoch": 1.6224814422057263,
|
| 1317 |
-
"grad_norm": 1.
|
| 1318 |
"learning_rate": 2.295864262990456e-05,
|
| 1319 |
"loss": 0.3522,
|
| 1320 |
"step": 1530
|
| 1321 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1322 |
{
|
| 1323 |
"epoch": 1.6330858960763521,
|
| 1324 |
-
"grad_norm": 1.
|
| 1325 |
"learning_rate": 2.27819017320608e-05,
|
| 1326 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1327 |
"step": 1540
|
| 1328 |
},
|
| 1329 |
{
|
| 1330 |
"epoch": 1.6436903499469777,
|
| 1331 |
-
"grad_norm": 0.
|
| 1332 |
"learning_rate": 2.260516083421704e-05,
|
| 1333 |
"loss": 0.3836,
|
| 1334 |
"step": 1550
|
| 1335 |
},
|
| 1336 |
{
|
| 1337 |
"epoch": 1.6436903499469777,
|
| 1338 |
-
"eval_loss": 0.
|
| 1339 |
-
"eval_runtime": 12.
|
| 1340 |
-
"eval_samples_per_second": 39.
|
| 1341 |
-
"eval_steps_per_second": 2.
|
| 1342 |
"step": 1550
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
"epoch": 1.6542948038176033,
|
| 1346 |
-
"grad_norm": 1.
|
| 1347 |
"learning_rate": 2.2428419936373275e-05,
|
| 1348 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1349 |
"step": 1560
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
"epoch": 1.664899257688229,
|
| 1353 |
-
"grad_norm": 0.
|
| 1354 |
"learning_rate": 2.2251679038529514e-05,
|
| 1355 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1356 |
"step": 1570
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
"epoch": 1.6755037115588547,
|
| 1360 |
-
"grad_norm": 1.
|
| 1361 |
"learning_rate": 2.2074938140685754e-05,
|
| 1362 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1363 |
"step": 1580
|
| 1364 |
},
|
| 1365 |
{
|
| 1366 |
"epoch": 1.6861081654294803,
|
| 1367 |
-
"grad_norm": 1.
|
| 1368 |
"learning_rate": 2.1898197242841996e-05,
|
| 1369 |
"loss": 0.3768,
|
| 1370 |
"step": 1590
|
| 1371 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1372 |
{
|
| 1373 |
"epoch": 1.696712619300106,
|
| 1374 |
-
"grad_norm": 1.
|
| 1375 |
"learning_rate": 2.1721456344998235e-05,
|
| 1376 |
-
"loss": 0.
|
| 1377 |
"step": 1600
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
"epoch": 1.696712619300106,
|
| 1381 |
-
"eval_loss": 0.
|
| 1382 |
-
"eval_runtime": 12.
|
| 1383 |
-
"eval_samples_per_second": 39.
|
| 1384 |
-
"eval_steps_per_second": 2.
|
| 1385 |
"step": 1600
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
"epoch": 1.7073170731707317,
|
| 1389 |
-
"grad_norm": 1.
|
| 1390 |
"learning_rate": 2.1544715447154475e-05,
|
| 1391 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1392 |
"step": 1610
|
| 1393 |
},
|
| 1394 |
{
|
| 1395 |
"epoch": 1.7179215270413573,
|
| 1396 |
-
"grad_norm": 1.
|
| 1397 |
"learning_rate": 2.1367974549310714e-05,
|
| 1398 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1399 |
"step": 1620
|
| 1400 |
},
|
| 1401 |
{
|
| 1402 |
"epoch": 1.728525980911983,
|
| 1403 |
-
"grad_norm": 1.
|
| 1404 |
"learning_rate": 2.1191233651466953e-05,
|
| 1405 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1406 |
"step": 1630
|
| 1407 |
},
|
| 1408 |
{
|
| 1409 |
"epoch": 1.7391304347826086,
|
| 1410 |
-
"grad_norm": 1.
|
| 1411 |
"learning_rate": 2.101449275362319e-05,
|
| 1412 |
"loss": 0.3966,
|
| 1413 |
"step": 1640
|
| 1414 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1415 |
{
|
| 1416 |
"epoch": 1.7497348886532342,
|
| 1417 |
-
"grad_norm": 1.
|
| 1418 |
"learning_rate": 2.0837751855779428e-05,
|
| 1419 |
-
"loss": 0.
|
| 1420 |
"step": 1650
|
| 1421 |
},
|
| 1422 |
{
|
| 1423 |
"epoch": 1.7497348886532342,
|
| 1424 |
-
"eval_loss": 0.
|
| 1425 |
-
"eval_runtime": 12.
|
| 1426 |
-
"eval_samples_per_second": 39.
|
| 1427 |
-
"eval_steps_per_second": 2.
|
| 1428 |
"step": 1650
|
| 1429 |
},
|
| 1430 |
{
|
| 1431 |
"epoch": 1.76033934252386,
|
| 1432 |
-
"grad_norm": 1.
|
| 1433 |
"learning_rate": 2.0661010957935667e-05,
|
| 1434 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1435 |
"step": 1660
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
"epoch": 1.7709437963944858,
|
| 1439 |
-
"grad_norm": 0.
|
| 1440 |
"learning_rate": 2.0484270060091906e-05,
|
| 1441 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1442 |
"step": 1670
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
"epoch": 1.7815482502651112,
|
| 1446 |
-
"grad_norm": 1.
|
| 1447 |
"learning_rate": 2.0307529162248146e-05,
|
| 1448 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1449 |
"step": 1680
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
"epoch": 1.792152704135737,
|
| 1453 |
-
"grad_norm": 1.
|
| 1454 |
"learning_rate": 2.0130788264404385e-05,
|
| 1455 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1456 |
"step": 1690
|
| 1457 |
},
|
| 1458 |
{
|
| 1459 |
"epoch": 1.8027571580063628,
|
| 1460 |
-
"grad_norm": 1.
|
| 1461 |
"learning_rate": 1.9954047366560624e-05,
|
| 1462 |
-
"loss": 0.
|
| 1463 |
"step": 1700
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
"epoch": 1.8027571580063628,
|
| 1467 |
-
"eval_loss": 0.
|
| 1468 |
-
"eval_runtime": 12.
|
| 1469 |
-
"eval_samples_per_second":
|
| 1470 |
-
"eval_steps_per_second": 2.
|
| 1471 |
"step": 1700
|
| 1472 |
}
|
| 1473 |
],
|
|
@@ -1475,7 +2563,7 @@
|
|
| 1475 |
"max_steps": 2829,
|
| 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": 1.8027571580063628,
|
| 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": 0.010604453870625663,
|
| 13 |
+
"grad_norm": 2.423347234725952,
|
| 14 |
"learning_rate": 4.982325910215624e-05,
|
| 15 |
+
"loss": 1.699,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.010604453870625663,
|
| 20 |
+
"eval_loss": 1.0601601600646973,
|
| 21 |
+
"eval_runtime": 11.7552,
|
| 22 |
+
"eval_samples_per_second": 40.833,
|
| 23 |
+
"eval_steps_per_second": 2.552,
|
| 24 |
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.021208907741251327,
|
| 28 |
+
"grad_norm": 0.707025408744812,
|
| 29 |
"learning_rate": 4.964651820431248e-05,
|
| 30 |
+
"loss": 0.8572,
|
| 31 |
+
"step": 20
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.021208907741251327,
|
| 35 |
+
"eval_loss": 0.8629258275032043,
|
| 36 |
+
"eval_runtime": 11.8364,
|
| 37 |
+
"eval_samples_per_second": 40.553,
|
| 38 |
+
"eval_steps_per_second": 2.535,
|
| 39 |
"step": 20
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.031813361611876985,
|
| 43 |
+
"grad_norm": 0.7233765125274658,
|
| 44 |
"learning_rate": 4.946977730646872e-05,
|
| 45 |
+
"loss": 0.7187,
|
| 46 |
+
"step": 30
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.031813361611876985,
|
| 50 |
+
"eval_loss": 0.8168506622314453,
|
| 51 |
+
"eval_runtime": 11.9031,
|
| 52 |
+
"eval_samples_per_second": 40.326,
|
| 53 |
+
"eval_steps_per_second": 2.52,
|
| 54 |
"step": 30
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 0.042417815482502653,
|
| 58 |
+
"grad_norm": 0.6566457748413086,
|
| 59 |
"learning_rate": 4.929303640862496e-05,
|
| 60 |
+
"loss": 0.6442,
|
| 61 |
+
"step": 40
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.042417815482502653,
|
| 65 |
+
"eval_loss": 0.7701226472854614,
|
| 66 |
+
"eval_runtime": 11.9415,
|
| 67 |
+
"eval_samples_per_second": 40.196,
|
| 68 |
+
"eval_steps_per_second": 2.512,
|
| 69 |
"step": 40
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.053022269353128315,
|
| 73 |
+
"grad_norm": 0.6793617606163025,
|
| 74 |
"learning_rate": 4.91162955107812e-05,
|
| 75 |
+
"loss": 0.6736,
|
| 76 |
"step": 50
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 0.053022269353128315,
|
| 80 |
+
"eval_loss": 0.7462028861045837,
|
| 81 |
+
"eval_runtime": 11.9825,
|
| 82 |
+
"eval_samples_per_second": 40.059,
|
| 83 |
+
"eval_steps_per_second": 2.504,
|
| 84 |
"step": 50
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 0.06362672322375397,
|
| 88 |
+
"grad_norm": 0.6813817024230957,
|
| 89 |
"learning_rate": 4.893955461293744e-05,
|
| 90 |
+
"loss": 0.6056,
|
| 91 |
+
"step": 60
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.06362672322375397,
|
| 95 |
+
"eval_loss": 0.7320327162742615,
|
| 96 |
+
"eval_runtime": 11.9898,
|
| 97 |
+
"eval_samples_per_second": 40.034,
|
| 98 |
+
"eval_steps_per_second": 2.502,
|
| 99 |
"step": 60
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.07423117709437964,
|
| 103 |
+
"grad_norm": 0.666418731212616,
|
| 104 |
"learning_rate": 4.876281371509368e-05,
|
| 105 |
+
"loss": 0.5822,
|
| 106 |
+
"step": 70
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.07423117709437964,
|
| 110 |
+
"eval_loss": 0.7147635817527771,
|
| 111 |
+
"eval_runtime": 12.0091,
|
| 112 |
+
"eval_samples_per_second": 39.97,
|
| 113 |
+
"eval_steps_per_second": 2.498,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.08483563096500531,
|
| 118 |
+
"grad_norm": 0.5959348678588867,
|
| 119 |
"learning_rate": 4.858607281724991e-05,
|
| 120 |
+
"loss": 0.5647,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08483563096500531,
|
| 125 |
+
"eval_loss": 0.6980472803115845,
|
| 126 |
+
"eval_runtime": 12.0191,
|
| 127 |
+
"eval_samples_per_second": 39.936,
|
| 128 |
+
"eval_steps_per_second": 2.496,
|
| 129 |
"step": 80
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.09544008483563096,
|
| 133 |
+
"grad_norm": 0.7531402707099915,
|
| 134 |
"learning_rate": 4.8409331919406155e-05,
|
| 135 |
+
"loss": 0.5738,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.09544008483563096,
|
| 140 |
+
"eval_loss": 0.6838169693946838,
|
| 141 |
+
"eval_runtime": 12.0197,
|
| 142 |
+
"eval_samples_per_second": 39.934,
|
| 143 |
+
"eval_steps_per_second": 2.496,
|
| 144 |
"step": 90
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 0.10604453870625663,
|
| 148 |
+
"grad_norm": 0.7243250608444214,
|
| 149 |
"learning_rate": 4.823259102156239e-05,
|
| 150 |
+
"loss": 0.5892,
|
| 151 |
"step": 100
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.10604453870625663,
|
| 155 |
+
"eval_loss": 0.6715003848075867,
|
| 156 |
+
"eval_runtime": 12.0427,
|
| 157 |
+
"eval_samples_per_second": 39.858,
|
| 158 |
+
"eval_steps_per_second": 2.491,
|
| 159 |
"step": 100
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 0.11664899257688228,
|
| 163 |
+
"grad_norm": 0.704513430595398,
|
| 164 |
"learning_rate": 4.8055850123718633e-05,
|
| 165 |
+
"loss": 0.5758,
|
| 166 |
+
"step": 110
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.11664899257688228,
|
| 170 |
+
"eval_loss": 0.669924795627594,
|
| 171 |
+
"eval_runtime": 12.0634,
|
| 172 |
+
"eval_samples_per_second": 39.79,
|
| 173 |
+
"eval_steps_per_second": 2.487,
|
| 174 |
"step": 110
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 0.12725344644750794,
|
| 178 |
+
"grad_norm": 0.7061131000518799,
|
| 179 |
"learning_rate": 4.7879109225874866e-05,
|
| 180 |
"loss": 0.539,
|
| 181 |
"step": 120
|
| 182 |
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.12725344644750794,
|
| 185 |
+
"eval_loss": 0.6550979614257812,
|
| 186 |
+
"eval_runtime": 12.0374,
|
| 187 |
+
"eval_samples_per_second": 39.876,
|
| 188 |
+
"eval_steps_per_second": 2.492,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
{
|
| 192 |
"epoch": 0.1378579003181336,
|
| 193 |
+
"grad_norm": 0.851974606513977,
|
| 194 |
"learning_rate": 4.770236832803111e-05,
|
| 195 |
+
"loss": 0.5297,
|
| 196 |
+
"step": 130
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.1378579003181336,
|
| 200 |
+
"eval_loss": 0.6511003971099854,
|
| 201 |
+
"eval_runtime": 12.0653,
|
| 202 |
+
"eval_samples_per_second": 39.784,
|
| 203 |
+
"eval_steps_per_second": 2.486,
|
| 204 |
"step": 130
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 0.14846235418875928,
|
| 208 |
+
"grad_norm": 0.739651083946228,
|
| 209 |
"learning_rate": 4.7525627430187344e-05,
|
| 210 |
+
"loss": 0.5388,
|
| 211 |
+
"step": 140
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.14846235418875928,
|
| 215 |
+
"eval_loss": 0.6424244046211243,
|
| 216 |
+
"eval_runtime": 12.0531,
|
| 217 |
+
"eval_samples_per_second": 39.824,
|
| 218 |
+
"eval_steps_per_second": 2.489,
|
| 219 |
"step": 140
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.15906680805938495,
|
| 223 |
+
"grad_norm": 0.6982805132865906,
|
| 224 |
"learning_rate": 4.734888653234359e-05,
|
| 225 |
+
"loss": 0.526,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.15906680805938495,
|
| 230 |
+
"eval_loss": 0.6392945647239685,
|
| 231 |
+
"eval_runtime": 12.0387,
|
| 232 |
+
"eval_samples_per_second": 39.871,
|
| 233 |
+
"eval_steps_per_second": 2.492,
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.16967126193001061,
|
| 238 |
+
"grad_norm": 0.7386083006858826,
|
| 239 |
"learning_rate": 4.717214563449982e-05,
|
| 240 |
+
"loss": 0.5486,
|
| 241 |
+
"step": 160
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.16967126193001061,
|
| 245 |
+
"eval_loss": 0.6340259313583374,
|
| 246 |
+
"eval_runtime": 12.0343,
|
| 247 |
+
"eval_samples_per_second": 39.886,
|
| 248 |
+
"eval_steps_per_second": 2.493,
|
| 249 |
"step": 160
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 0.18027571580063625,
|
| 253 |
+
"grad_norm": 0.7634233236312866,
|
| 254 |
"learning_rate": 4.699540473665606e-05,
|
| 255 |
+
"loss": 0.4994,
|
| 256 |
+
"step": 170
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.18027571580063625,
|
| 260 |
+
"eval_loss": 0.6314084529876709,
|
| 261 |
+
"eval_runtime": 12.0318,
|
| 262 |
+
"eval_samples_per_second": 39.894,
|
| 263 |
+
"eval_steps_per_second": 2.493,
|
| 264 |
"step": 170
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 0.19088016967126192,
|
| 268 |
+
"grad_norm": 0.7392905354499817,
|
| 269 |
"learning_rate": 4.681866383881231e-05,
|
| 270 |
+
"loss": 0.5183,
|
| 271 |
+
"step": 180
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.19088016967126192,
|
| 275 |
+
"eval_loss": 0.6283704042434692,
|
| 276 |
+
"eval_runtime": 12.0358,
|
| 277 |
+
"eval_samples_per_second": 39.881,
|
| 278 |
+
"eval_steps_per_second": 2.493,
|
| 279 |
"step": 180
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 0.2014846235418876,
|
| 283 |
+
"grad_norm": 0.7040207386016846,
|
| 284 |
"learning_rate": 4.664192294096854e-05,
|
| 285 |
+
"loss": 0.4937,
|
| 286 |
+
"step": 190
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.2014846235418876,
|
| 290 |
+
"eval_loss": 0.6267850995063782,
|
| 291 |
+
"eval_runtime": 12.0297,
|
| 292 |
+
"eval_samples_per_second": 39.901,
|
| 293 |
+
"eval_steps_per_second": 2.494,
|
| 294 |
"step": 190
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 0.21208907741251326,
|
| 298 |
+
"grad_norm": 0.7676352858543396,
|
| 299 |
"learning_rate": 4.6465182043124786e-05,
|
| 300 |
+
"loss": 0.5047,
|
| 301 |
"step": 200
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 0.21208907741251326,
|
| 305 |
+
"eval_loss": 0.6272422075271606,
|
| 306 |
+
"eval_runtime": 12.0227,
|
| 307 |
+
"eval_samples_per_second": 39.925,
|
| 308 |
+
"eval_steps_per_second": 2.495,
|
| 309 |
"step": 200
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 0.22269353128313893,
|
| 313 |
+
"grad_norm": 0.7943865656852722,
|
| 314 |
"learning_rate": 4.628844114528102e-05,
|
| 315 |
+
"loss": 0.4956,
|
| 316 |
+
"step": 210
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.22269353128313893,
|
| 320 |
+
"eval_loss": 0.6196587085723877,
|
| 321 |
+
"eval_runtime": 12.0212,
|
| 322 |
+
"eval_samples_per_second": 39.929,
|
| 323 |
+
"eval_steps_per_second": 2.496,
|
| 324 |
"step": 210
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.23329798515376457,
|
| 328 |
+
"grad_norm": 0.6889236569404602,
|
| 329 |
"learning_rate": 4.6111700247437265e-05,
|
| 330 |
+
"loss": 0.5004,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.23329798515376457,
|
| 335 |
+
"eval_loss": 0.6165774464607239,
|
| 336 |
+
"eval_runtime": 12.0128,
|
| 337 |
+
"eval_samples_per_second": 39.958,
|
| 338 |
+
"eval_steps_per_second": 2.497,
|
| 339 |
"step": 220
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.24390243902439024,
|
| 343 |
+
"grad_norm": 0.7515945434570312,
|
| 344 |
"learning_rate": 4.59349593495935e-05,
|
| 345 |
+
"loss": 0.5093,
|
| 346 |
+
"step": 230
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.24390243902439024,
|
| 350 |
+
"eval_loss": 0.6142643094062805,
|
| 351 |
+
"eval_runtime": 12.0272,
|
| 352 |
+
"eval_samples_per_second": 39.91,
|
| 353 |
+
"eval_steps_per_second": 2.494,
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 0.2545068928950159,
|
| 358 |
+
"grad_norm": 0.7514846920967102,
|
| 359 |
"learning_rate": 4.5758218451749736e-05,
|
| 360 |
+
"loss": 0.4756,
|
| 361 |
+
"step": 240
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.2545068928950159,
|
| 365 |
+
"eval_loss": 0.609127402305603,
|
| 366 |
+
"eval_runtime": 12.0191,
|
| 367 |
+
"eval_samples_per_second": 39.936,
|
| 368 |
+
"eval_steps_per_second": 2.496,
|
| 369 |
"step": 240
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 0.2651113467656416,
|
| 373 |
+
"grad_norm": 0.7803705930709839,
|
| 374 |
"learning_rate": 4.5581477553905975e-05,
|
| 375 |
+
"loss": 0.5167,
|
| 376 |
"step": 250
|
| 377 |
},
|
| 378 |
{
|
| 379 |
"epoch": 0.2651113467656416,
|
| 380 |
+
"eval_loss": 0.6084603071212769,
|
| 381 |
+
"eval_runtime": 12.0245,
|
| 382 |
+
"eval_samples_per_second": 39.918,
|
| 383 |
+
"eval_steps_per_second": 2.495,
|
| 384 |
"step": 250
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 0.2757158006362672,
|
| 388 |
+
"grad_norm": 0.7443625330924988,
|
| 389 |
"learning_rate": 4.5404736656062215e-05,
|
| 390 |
+
"loss": 0.4767,
|
| 391 |
+
"step": 260
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.2757158006362672,
|
| 395 |
+
"eval_loss": 0.6066035032272339,
|
| 396 |
+
"eval_runtime": 12.0148,
|
| 397 |
+
"eval_samples_per_second": 39.951,
|
| 398 |
+
"eval_steps_per_second": 2.497,
|
| 399 |
"step": 260
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 0.2863202545068929,
|
| 403 |
+
"grad_norm": 0.7620787024497986,
|
| 404 |
"learning_rate": 4.5227995758218454e-05,
|
| 405 |
+
"loss": 0.4914,
|
| 406 |
+
"step": 270
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.2863202545068929,
|
| 410 |
+
"eval_loss": 0.6044597029685974,
|
| 411 |
+
"eval_runtime": 12.0147,
|
| 412 |
+
"eval_samples_per_second": 39.951,
|
| 413 |
+
"eval_steps_per_second": 2.497,
|
| 414 |
"step": 270
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 0.29692470837751855,
|
| 418 |
+
"grad_norm": 0.7282053828239441,
|
| 419 |
"learning_rate": 4.505125486037469e-05,
|
| 420 |
+
"loss": 0.4854,
|
| 421 |
+
"step": 280
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.29692470837751855,
|
| 425 |
+
"eval_loss": 0.6000507473945618,
|
| 426 |
+
"eval_runtime": 12.0195,
|
| 427 |
+
"eval_samples_per_second": 39.935,
|
| 428 |
+
"eval_steps_per_second": 2.496,
|
| 429 |
"step": 280
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.3075291622481442,
|
| 433 |
+
"grad_norm": 0.7459163665771484,
|
| 434 |
"learning_rate": 4.487451396253093e-05,
|
| 435 |
"loss": 0.4891,
|
| 436 |
"step": 290
|
| 437 |
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3075291622481442,
|
| 440 |
+
"eval_loss": 0.6018843054771423,
|
| 441 |
+
"eval_runtime": 12.0272,
|
| 442 |
+
"eval_samples_per_second": 39.909,
|
| 443 |
+
"eval_steps_per_second": 2.494,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
{
|
| 447 |
"epoch": 0.3181336161187699,
|
| 448 |
+
"grad_norm": 0.7495472431182861,
|
| 449 |
"learning_rate": 4.469777306468717e-05,
|
| 450 |
+
"loss": 0.48,
|
| 451 |
"step": 300
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 0.3181336161187699,
|
| 455 |
+
"eval_loss": 0.5989978909492493,
|
| 456 |
+
"eval_runtime": 12.0374,
|
| 457 |
+
"eval_samples_per_second": 39.876,
|
| 458 |
+
"eval_steps_per_second": 2.492,
|
| 459 |
"step": 300
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 0.32873806998939553,
|
| 463 |
+
"grad_norm": 0.7306896448135376,
|
| 464 |
"learning_rate": 4.452103216684341e-05,
|
| 465 |
+
"loss": 0.514,
|
| 466 |
+
"step": 310
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.32873806998939553,
|
| 470 |
+
"eval_loss": 0.5953137874603271,
|
| 471 |
+
"eval_runtime": 12.033,
|
| 472 |
+
"eval_samples_per_second": 39.89,
|
| 473 |
+
"eval_steps_per_second": 2.493,
|
| 474 |
"step": 310
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 0.33934252386002123,
|
| 478 |
+
"grad_norm": 0.8132976293563843,
|
| 479 |
"learning_rate": 4.434429126899965e-05,
|
| 480 |
+
"loss": 0.4896,
|
| 481 |
+
"step": 320
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.33934252386002123,
|
| 485 |
+
"eval_loss": 0.5938466787338257,
|
| 486 |
+
"eval_runtime": 12.0422,
|
| 487 |
+
"eval_samples_per_second": 39.86,
|
| 488 |
+
"eval_steps_per_second": 2.491,
|
| 489 |
"step": 320
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 0.34994697773064687,
|
| 493 |
+
"grad_norm": 0.7447174191474915,
|
| 494 |
"learning_rate": 4.416755037115589e-05,
|
| 495 |
+
"loss": 0.4724,
|
| 496 |
+
"step": 330
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.34994697773064687,
|
| 500 |
+
"eval_loss": 0.5906462073326111,
|
| 501 |
+
"eval_runtime": 12.0377,
|
| 502 |
+
"eval_samples_per_second": 39.875,
|
| 503 |
+
"eval_steps_per_second": 2.492,
|
| 504 |
"step": 330
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 0.3605514316012725,
|
| 508 |
+
"grad_norm": 0.7956692576408386,
|
| 509 |
"learning_rate": 4.399080947331213e-05,
|
| 510 |
+
"loss": 0.4648,
|
| 511 |
+
"step": 340
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.3605514316012725,
|
| 515 |
+
"eval_loss": 0.5904966592788696,
|
| 516 |
+
"eval_runtime": 12.0368,
|
| 517 |
+
"eval_samples_per_second": 39.878,
|
| 518 |
+
"eval_steps_per_second": 2.492,
|
| 519 |
"step": 340
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 0.3711558854718982,
|
| 523 |
+
"grad_norm": 0.7292013168334961,
|
| 524 |
"learning_rate": 4.381406857546837e-05,
|
| 525 |
+
"loss": 0.4904,
|
| 526 |
"step": 350
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 0.3711558854718982,
|
| 530 |
+
"eval_loss": 0.5848659873008728,
|
| 531 |
+
"eval_runtime": 12.0452,
|
| 532 |
+
"eval_samples_per_second": 39.85,
|
| 533 |
+
"eval_steps_per_second": 2.491,
|
| 534 |
"step": 350
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 0.38176033934252385,
|
| 538 |
+
"grad_norm": 0.6860175132751465,
|
| 539 |
"learning_rate": 4.3637327677624606e-05,
|
| 540 |
+
"loss": 0.4731,
|
| 541 |
+
"step": 360
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.38176033934252385,
|
| 545 |
+
"eval_loss": 0.5840328335762024,
|
| 546 |
+
"eval_runtime": 12.0322,
|
| 547 |
+
"eval_samples_per_second": 39.893,
|
| 548 |
+
"eval_steps_per_second": 2.493,
|
| 549 |
"step": 360
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 0.39236479321314954,
|
| 553 |
+
"grad_norm": 0.827422559261322,
|
| 554 |
"learning_rate": 4.3460586779780846e-05,
|
| 555 |
+
"loss": 0.4571,
|
| 556 |
+
"step": 370
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.39236479321314954,
|
| 560 |
+
"eval_loss": 0.5892025828361511,
|
| 561 |
+
"eval_runtime": 12.0697,
|
| 562 |
+
"eval_samples_per_second": 39.769,
|
| 563 |
+
"eval_steps_per_second": 2.486,
|
| 564 |
"step": 370
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 0.4029692470837752,
|
| 568 |
+
"grad_norm": 0.7977253198623657,
|
| 569 |
"learning_rate": 4.328384588193708e-05,
|
| 570 |
+
"loss": 0.465,
|
| 571 |
+
"step": 380
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.4029692470837752,
|
| 575 |
+
"eval_loss": 0.5796211361885071,
|
| 576 |
+
"eval_runtime": 12.043,
|
| 577 |
+
"eval_samples_per_second": 39.857,
|
| 578 |
+
"eval_steps_per_second": 2.491,
|
| 579 |
"step": 380
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 0.4135737009544008,
|
| 583 |
+
"grad_norm": 0.7545391321182251,
|
| 584 |
"learning_rate": 4.3107104984093324e-05,
|
| 585 |
+
"loss": 0.4433,
|
| 586 |
+
"step": 390
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.4135737009544008,
|
| 590 |
+
"eval_loss": 0.5798179507255554,
|
| 591 |
+
"eval_runtime": 12.0361,
|
| 592 |
+
"eval_samples_per_second": 39.88,
|
| 593 |
+
"eval_steps_per_second": 2.492,
|
| 594 |
"step": 390
|
| 595 |
},
|
| 596 |
{
|
| 597 |
"epoch": 0.4241781548250265,
|
| 598 |
+
"grad_norm": 0.7273393273353577,
|
| 599 |
"learning_rate": 4.2930364086249556e-05,
|
| 600 |
+
"loss": 0.4662,
|
| 601 |
"step": 400
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 0.4241781548250265,
|
| 605 |
+
"eval_loss": 0.5788875818252563,
|
| 606 |
+
"eval_runtime": 12.032,
|
| 607 |
+
"eval_samples_per_second": 39.894,
|
| 608 |
+
"eval_steps_per_second": 2.493,
|
| 609 |
"step": 400
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 0.43478260869565216,
|
| 613 |
+
"grad_norm": 0.742859959602356,
|
| 614 |
"learning_rate": 4.27536231884058e-05,
|
| 615 |
+
"loss": 0.4937,
|
| 616 |
+
"step": 410
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.43478260869565216,
|
| 620 |
+
"eval_loss": 0.5762577056884766,
|
| 621 |
+
"eval_runtime": 12.0232,
|
| 622 |
+
"eval_samples_per_second": 39.923,
|
| 623 |
+
"eval_steps_per_second": 2.495,
|
| 624 |
"step": 410
|
| 625 |
},
|
| 626 |
{
|
| 627 |
"epoch": 0.44538706256627786,
|
| 628 |
+
"grad_norm": 0.8810379505157471,
|
| 629 |
"learning_rate": 4.2576882290562035e-05,
|
| 630 |
+
"loss": 0.4611,
|
| 631 |
+
"step": 420
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.44538706256627786,
|
| 635 |
+
"eval_loss": 0.5756653547286987,
|
| 636 |
+
"eval_runtime": 12.0308,
|
| 637 |
+
"eval_samples_per_second": 39.898,
|
| 638 |
+
"eval_steps_per_second": 2.494,
|
| 639 |
"step": 420
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 0.4559915164369035,
|
| 643 |
+
"grad_norm": 0.8116405010223389,
|
| 644 |
"learning_rate": 4.240014139271828e-05,
|
| 645 |
+
"loss": 0.48,
|
| 646 |
+
"step": 430
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4559915164369035,
|
| 650 |
+
"eval_loss": 0.5730367302894592,
|
| 651 |
+
"eval_runtime": 12.0291,
|
| 652 |
+
"eval_samples_per_second": 39.903,
|
| 653 |
+
"eval_steps_per_second": 2.494,
|
| 654 |
"step": 430
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 0.46659597030752914,
|
| 658 |
+
"grad_norm": 0.7950704097747803,
|
| 659 |
"learning_rate": 4.222340049487451e-05,
|
| 660 |
+
"loss": 0.4634,
|
| 661 |
+
"step": 440
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.46659597030752914,
|
| 665 |
+
"eval_loss": 0.5719767808914185,
|
| 666 |
+
"eval_runtime": 12.0229,
|
| 667 |
+
"eval_samples_per_second": 39.924,
|
| 668 |
+
"eval_steps_per_second": 2.495,
|
| 669 |
"step": 440
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 0.47720042417815484,
|
| 673 |
+
"grad_norm": 0.7839304208755493,
|
| 674 |
"learning_rate": 4.204665959703076e-05,
|
| 675 |
+
"loss": 0.4448,
|
| 676 |
"step": 450
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 0.47720042417815484,
|
| 680 |
+
"eval_loss": 0.5742294192314148,
|
| 681 |
+
"eval_runtime": 12.0181,
|
| 682 |
+
"eval_samples_per_second": 39.94,
|
| 683 |
+
"eval_steps_per_second": 2.496,
|
| 684 |
"step": 450
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 0.4878048780487805,
|
| 688 |
+
"grad_norm": 0.8098810911178589,
|
| 689 |
"learning_rate": 4.186991869918699e-05,
|
| 690 |
+
"loss": 0.4501,
|
| 691 |
+
"step": 460
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.4878048780487805,
|
| 695 |
+
"eval_loss": 0.5708529353141785,
|
| 696 |
+
"eval_runtime": 12.0037,
|
| 697 |
+
"eval_samples_per_second": 39.988,
|
| 698 |
+
"eval_steps_per_second": 2.499,
|
| 699 |
"step": 460
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 0.4984093319194062,
|
| 703 |
+
"grad_norm": 0.7419692277908325,
|
| 704 |
"learning_rate": 4.169317780134323e-05,
|
| 705 |
+
"loss": 0.4413,
|
| 706 |
+
"step": 470
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.4984093319194062,
|
| 710 |
+
"eval_loss": 0.5701093077659607,
|
| 711 |
+
"eval_runtime": 12.0178,
|
| 712 |
+
"eval_samples_per_second": 39.941,
|
| 713 |
+
"eval_steps_per_second": 2.496,
|
| 714 |
"step": 470
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 0.5090137857900318,
|
| 718 |
+
"grad_norm": 0.8493075370788574,
|
| 719 |
"learning_rate": 4.151643690349947e-05,
|
| 720 |
+
"loss": 0.4622,
|
| 721 |
+
"step": 480
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.5090137857900318,
|
| 725 |
+
"eval_loss": 0.5684540867805481,
|
| 726 |
+
"eval_runtime": 12.02,
|
| 727 |
+
"eval_samples_per_second": 39.933,
|
| 728 |
+
"eval_steps_per_second": 2.496,
|
| 729 |
"step": 480
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 0.5196182396606575,
|
| 733 |
+
"grad_norm": 0.8144774436950684,
|
| 734 |
"learning_rate": 4.133969600565571e-05,
|
| 735 |
+
"loss": 0.422,
|
| 736 |
+
"step": 490
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.5196182396606575,
|
| 740 |
+
"eval_loss": 0.56850665807724,
|
| 741 |
+
"eval_runtime": 12.0321,
|
| 742 |
+
"eval_samples_per_second": 39.893,
|
| 743 |
+
"eval_steps_per_second": 2.493,
|
| 744 |
"step": 490
|
| 745 |
},
|
| 746 |
{
|
| 747 |
"epoch": 0.5302226935312832,
|
| 748 |
+
"grad_norm": 0.790009617805481,
|
| 749 |
"learning_rate": 4.116295510781195e-05,
|
| 750 |
+
"loss": 0.4808,
|
| 751 |
"step": 500
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 0.5302226935312832,
|
| 755 |
+
"eval_loss": 0.5695191621780396,
|
| 756 |
+
"eval_runtime": 12.0343,
|
| 757 |
+
"eval_samples_per_second": 39.886,
|
| 758 |
+
"eval_steps_per_second": 2.493,
|
| 759 |
"step": 500
|
| 760 |
},
|
| 761 |
{
|
| 762 |
"epoch": 0.5408271474019088,
|
| 763 |
+
"grad_norm": 0.7809069752693176,
|
| 764 |
"learning_rate": 4.098621420996819e-05,
|
| 765 |
+
"loss": 0.4501,
|
| 766 |
+
"step": 510
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.5408271474019088,
|
| 770 |
+
"eval_loss": 0.5662057995796204,
|
| 771 |
+
"eval_runtime": 12.0285,
|
| 772 |
+
"eval_samples_per_second": 39.905,
|
| 773 |
+
"eval_steps_per_second": 2.494,
|
| 774 |
"step": 510
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 0.5514316012725344,
|
| 778 |
+
"grad_norm": 0.8309295177459717,
|
| 779 |
"learning_rate": 4.080947331212443e-05,
|
| 780 |
+
"loss": 0.4385,
|
| 781 |
+
"step": 520
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.5514316012725344,
|
| 785 |
+
"eval_loss": 0.5643609166145325,
|
| 786 |
+
"eval_runtime": 12.0404,
|
| 787 |
+
"eval_samples_per_second": 39.866,
|
| 788 |
+
"eval_steps_per_second": 2.492,
|
| 789 |
"step": 520
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 0.5620360551431601,
|
| 793 |
+
"grad_norm": 0.9043754935264587,
|
| 794 |
"learning_rate": 4.0632732414280666e-05,
|
| 795 |
+
"loss": 0.4409,
|
| 796 |
+
"step": 530
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.5620360551431601,
|
| 800 |
+
"eval_loss": 0.5642057657241821,
|
| 801 |
+
"eval_runtime": 12.0631,
|
| 802 |
+
"eval_samples_per_second": 39.791,
|
| 803 |
+
"eval_steps_per_second": 2.487,
|
| 804 |
"step": 530
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 0.5726405090137858,
|
| 808 |
+
"grad_norm": 0.8891374468803406,
|
| 809 |
"learning_rate": 4.0455991516436905e-05,
|
| 810 |
+
"loss": 0.4694,
|
| 811 |
+
"step": 540
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.5726405090137858,
|
| 815 |
+
"eval_loss": 0.56215900182724,
|
| 816 |
+
"eval_runtime": 12.1217,
|
| 817 |
+
"eval_samples_per_second": 39.598,
|
| 818 |
+
"eval_steps_per_second": 2.475,
|
| 819 |
"step": 540
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 0.5832449628844114,
|
| 823 |
+
"grad_norm": 0.8346860408782959,
|
| 824 |
"learning_rate": 4.0279250618593144e-05,
|
| 825 |
+
"loss": 0.4304,
|
| 826 |
"step": 550
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 0.5832449628844114,
|
| 830 |
+
"eval_loss": 0.5631229281425476,
|
| 831 |
+
"eval_runtime": 12.2151,
|
| 832 |
+
"eval_samples_per_second": 39.296,
|
| 833 |
+
"eval_steps_per_second": 2.456,
|
| 834 |
"step": 550
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 0.5938494167550371,
|
| 838 |
+
"grad_norm": 0.9839991927146912,
|
| 839 |
"learning_rate": 4.0102509720749383e-05,
|
| 840 |
+
"loss": 0.4705,
|
| 841 |
+
"step": 560
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.5938494167550371,
|
| 845 |
+
"eval_loss": 0.5627191066741943,
|
| 846 |
+
"eval_runtime": 12.2483,
|
| 847 |
+
"eval_samples_per_second": 39.189,
|
| 848 |
+
"eval_steps_per_second": 2.449,
|
| 849 |
"step": 560
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 0.6044538706256628,
|
| 853 |
+
"grad_norm": 0.8321048617362976,
|
| 854 |
"learning_rate": 3.992576882290562e-05,
|
| 855 |
+
"loss": 0.4424,
|
| 856 |
+
"step": 570
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6044538706256628,
|
| 860 |
+
"eval_loss": 0.5603087544441223,
|
| 861 |
+
"eval_runtime": 12.2401,
|
| 862 |
+
"eval_samples_per_second": 39.215,
|
| 863 |
+
"eval_steps_per_second": 2.451,
|
| 864 |
"step": 570
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 0.6150583244962884,
|
| 868 |
+
"grad_norm": 0.8030439615249634,
|
| 869 |
"learning_rate": 3.974902792506186e-05,
|
| 870 |
+
"loss": 0.4174,
|
| 871 |
+
"step": 580
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.6150583244962884,
|
| 875 |
+
"eval_loss": 0.5613930821418762,
|
| 876 |
+
"eval_runtime": 12.2027,
|
| 877 |
+
"eval_samples_per_second": 39.336,
|
| 878 |
+
"eval_steps_per_second": 2.458,
|
| 879 |
"step": 580
|
| 880 |
},
|
| 881 |
{
|
| 882 |
"epoch": 0.6256627783669141,
|
| 883 |
+
"grad_norm": 0.8421948552131653,
|
| 884 |
"learning_rate": 3.95722870272181e-05,
|
| 885 |
+
"loss": 0.445,
|
| 886 |
+
"step": 590
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.6256627783669141,
|
| 890 |
+
"eval_loss": 0.561574399471283,
|
| 891 |
+
"eval_runtime": 12.1733,
|
| 892 |
+
"eval_samples_per_second": 39.431,
|
| 893 |
+
"eval_steps_per_second": 2.464,
|
| 894 |
"step": 590
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 0.6362672322375398,
|
| 898 |
+
"grad_norm": 0.8296137452125549,
|
| 899 |
"learning_rate": 3.939554612937434e-05,
|
| 900 |
+
"loss": 0.4771,
|
| 901 |
"step": 600
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 0.6362672322375398,
|
| 905 |
+
"eval_loss": 0.5598013401031494,
|
| 906 |
+
"eval_runtime": 12.155,
|
| 907 |
+
"eval_samples_per_second": 39.49,
|
| 908 |
+
"eval_steps_per_second": 2.468,
|
| 909 |
"step": 600
|
| 910 |
},
|
| 911 |
{
|
| 912 |
"epoch": 0.6468716861081655,
|
| 913 |
+
"grad_norm": 0.7955836057662964,
|
| 914 |
"learning_rate": 3.921880523153057e-05,
|
| 915 |
+
"loss": 0.4379,
|
| 916 |
+
"step": 610
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.6468716861081655,
|
| 920 |
+
"eval_loss": 0.55992192029953,
|
| 921 |
+
"eval_runtime": 12.141,
|
| 922 |
+
"eval_samples_per_second": 39.536,
|
| 923 |
+
"eval_steps_per_second": 2.471,
|
| 924 |
"step": 610
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 0.6574761399787911,
|
| 928 |
+
"grad_norm": 0.8401910066604614,
|
| 929 |
"learning_rate": 3.904206433368682e-05,
|
| 930 |
+
"loss": 0.4403,
|
| 931 |
+
"step": 620
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.6574761399787911,
|
| 935 |
+
"eval_loss": 0.5604690909385681,
|
| 936 |
+
"eval_runtime": 12.1179,
|
| 937 |
+
"eval_samples_per_second": 39.611,
|
| 938 |
+
"eval_steps_per_second": 2.476,
|
| 939 |
"step": 620
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 0.6680805938494168,
|
| 943 |
+
"grad_norm": 0.7533448338508606,
|
| 944 |
"learning_rate": 3.886532343584305e-05,
|
| 945 |
+
"loss": 0.4163,
|
| 946 |
+
"step": 630
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.6680805938494168,
|
| 950 |
+
"eval_loss": 0.5586678385734558,
|
| 951 |
+
"eval_runtime": 12.1102,
|
| 952 |
+
"eval_samples_per_second": 39.636,
|
| 953 |
+
"eval_steps_per_second": 2.477,
|
| 954 |
"step": 630
|
| 955 |
},
|
| 956 |
{
|
| 957 |
"epoch": 0.6786850477200425,
|
| 958 |
+
"grad_norm": 0.8505811095237732,
|
| 959 |
"learning_rate": 3.86885825379993e-05,
|
| 960 |
+
"loss": 0.4265,
|
| 961 |
+
"step": 640
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6786850477200425,
|
| 965 |
+
"eval_loss": 0.5571100115776062,
|
| 966 |
+
"eval_runtime": 12.0969,
|
| 967 |
+
"eval_samples_per_second": 39.68,
|
| 968 |
+
"eval_steps_per_second": 2.48,
|
| 969 |
"step": 640
|
| 970 |
},
|
| 971 |
{
|
| 972 |
"epoch": 0.689289501590668,
|
| 973 |
+
"grad_norm": 0.8246090412139893,
|
| 974 |
"learning_rate": 3.851184164015553e-05,
|
| 975 |
+
"loss": 0.4359,
|
| 976 |
"step": 650
|
| 977 |
},
|
| 978 |
{
|
| 979 |
"epoch": 0.689289501590668,
|
| 980 |
+
"eval_loss": 0.557594358921051,
|
| 981 |
+
"eval_runtime": 12.0842,
|
| 982 |
+
"eval_samples_per_second": 39.721,
|
| 983 |
+
"eval_steps_per_second": 2.483,
|
| 984 |
"step": 650
|
| 985 |
},
|
| 986 |
{
|
| 987 |
"epoch": 0.6998939554612937,
|
| 988 |
+
"grad_norm": 0.8045751452445984,
|
| 989 |
"learning_rate": 3.8335100742311775e-05,
|
| 990 |
"loss": 0.4426,
|
| 991 |
"step": 660
|
| 992 |
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.6998939554612937,
|
| 995 |
+
"eval_loss": 0.5541183352470398,
|
| 996 |
+
"eval_runtime": 12.0882,
|
| 997 |
+
"eval_samples_per_second": 39.708,
|
| 998 |
+
"eval_steps_per_second": 2.482,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
{
|
| 1002 |
"epoch": 0.7104984093319194,
|
| 1003 |
+
"grad_norm": 0.8575365543365479,
|
| 1004 |
"learning_rate": 3.8158359844468015e-05,
|
| 1005 |
+
"loss": 0.4273,
|
| 1006 |
+
"step": 670
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.7104984093319194,
|
| 1010 |
+
"eval_loss": 0.5536326169967651,
|
| 1011 |
+
"eval_runtime": 12.0756,
|
| 1012 |
+
"eval_samples_per_second": 39.75,
|
| 1013 |
+
"eval_steps_per_second": 2.484,
|
| 1014 |
"step": 670
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
"epoch": 0.721102863202545,
|
| 1018 |
+
"grad_norm": 0.7538650035858154,
|
| 1019 |
"learning_rate": 3.7981618946624254e-05,
|
| 1020 |
+
"loss": 0.4257,
|
| 1021 |
+
"step": 680
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.721102863202545,
|
| 1025 |
+
"eval_loss": 0.5530498623847961,
|
| 1026 |
+
"eval_runtime": 12.0756,
|
| 1027 |
+
"eval_samples_per_second": 39.75,
|
| 1028 |
+
"eval_steps_per_second": 2.484,
|
| 1029 |
"step": 680
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
"epoch": 0.7317073170731707,
|
| 1033 |
+
"grad_norm": 0.8397036194801331,
|
| 1034 |
"learning_rate": 3.780487804878049e-05,
|
| 1035 |
+
"loss": 0.4753,
|
| 1036 |
+
"step": 690
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.7317073170731707,
|
| 1040 |
+
"eval_loss": 0.5513155460357666,
|
| 1041 |
+
"eval_runtime": 12.0794,
|
| 1042 |
+
"eval_samples_per_second": 39.737,
|
| 1043 |
+
"eval_steps_per_second": 2.484,
|
| 1044 |
"step": 690
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
"epoch": 0.7423117709437964,
|
| 1048 |
+
"grad_norm": 0.8593904376029968,
|
| 1049 |
"learning_rate": 3.7628137150936725e-05,
|
| 1050 |
+
"loss": 0.4469,
|
| 1051 |
"step": 700
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 0.7423117709437964,
|
| 1055 |
+
"eval_loss": 0.5494736433029175,
|
| 1056 |
+
"eval_runtime": 12.0462,
|
| 1057 |
+
"eval_samples_per_second": 39.847,
|
| 1058 |
+
"eval_steps_per_second": 2.49,
|
| 1059 |
"step": 700
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
"epoch": 0.7529162248144221,
|
| 1063 |
+
"grad_norm": 0.8573834896087646,
|
| 1064 |
"learning_rate": 3.745139625309297e-05,
|
| 1065 |
"loss": 0.4418,
|
| 1066 |
"step": 710
|
| 1067 |
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7529162248144221,
|
| 1070 |
+
"eval_loss": 0.550113320350647,
|
| 1071 |
+
"eval_runtime": 12.1417,
|
| 1072 |
+
"eval_samples_per_second": 39.533,
|
| 1073 |
+
"eval_steps_per_second": 2.471,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
{
|
| 1077 |
"epoch": 0.7635206786850477,
|
| 1078 |
+
"grad_norm": 0.8773534893989563,
|
| 1079 |
"learning_rate": 3.7274655355249204e-05,
|
| 1080 |
+
"loss": 0.42,
|
| 1081 |
+
"step": 720
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.7635206786850477,
|
| 1085 |
+
"eval_loss": 0.5515414476394653,
|
| 1086 |
+
"eval_runtime": 12.0672,
|
| 1087 |
+
"eval_samples_per_second": 39.777,
|
| 1088 |
+
"eval_steps_per_second": 2.486,
|
| 1089 |
"step": 720
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
"epoch": 0.7741251325556734,
|
| 1093 |
+
"grad_norm": 0.8355915546417236,
|
| 1094 |
"learning_rate": 3.709791445740545e-05,
|
| 1095 |
+
"loss": 0.4692,
|
| 1096 |
+
"step": 730
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.7741251325556734,
|
| 1100 |
+
"eval_loss": 0.5495920777320862,
|
| 1101 |
+
"eval_runtime": 12.0669,
|
| 1102 |
+
"eval_samples_per_second": 39.778,
|
| 1103 |
+
"eval_steps_per_second": 2.486,
|
| 1104 |
"step": 730
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
"epoch": 0.7847295864262991,
|
| 1108 |
+
"grad_norm": 0.8989934921264648,
|
| 1109 |
"learning_rate": 3.692117355956168e-05,
|
| 1110 |
+
"loss": 0.4463,
|
| 1111 |
+
"step": 740
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.7847295864262991,
|
| 1115 |
+
"eval_loss": 0.5498195886611938,
|
| 1116 |
+
"eval_runtime": 12.0627,
|
| 1117 |
+
"eval_samples_per_second": 39.792,
|
| 1118 |
+
"eval_steps_per_second": 2.487,
|
| 1119 |
"step": 740
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 0.7953340402969247,
|
| 1123 |
+
"grad_norm": 0.7843239307403564,
|
| 1124 |
"learning_rate": 3.674443266171793e-05,
|
| 1125 |
+
"loss": 0.4083,
|
| 1126 |
"step": 750
|
| 1127 |
},
|
| 1128 |
{
|
| 1129 |
"epoch": 0.7953340402969247,
|
| 1130 |
+
"eval_loss": 0.5499619841575623,
|
| 1131 |
+
"eval_runtime": 12.0741,
|
| 1132 |
+
"eval_samples_per_second": 39.754,
|
| 1133 |
+
"eval_steps_per_second": 2.485,
|
| 1134 |
"step": 750
|
| 1135 |
},
|
| 1136 |
{
|
| 1137 |
"epoch": 0.8059384941675504,
|
| 1138 |
+
"grad_norm": 0.8288501501083374,
|
| 1139 |
"learning_rate": 3.656769176387416e-05,
|
| 1140 |
"loss": 0.4087,
|
| 1141 |
"step": 760
|
| 1142 |
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.8059384941675504,
|
| 1145 |
+
"eval_loss": 0.5498246550559998,
|
| 1146 |
+
"eval_runtime": 12.0728,
|
| 1147 |
+
"eval_samples_per_second": 39.759,
|
| 1148 |
+
"eval_steps_per_second": 2.485,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
{
|
| 1152 |
"epoch": 0.8165429480381761,
|
| 1153 |
+
"grad_norm": 0.8535901308059692,
|
| 1154 |
"learning_rate": 3.63909508660304e-05,
|
| 1155 |
+
"loss": 0.425,
|
| 1156 |
+
"step": 770
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.8165429480381761,
|
| 1160 |
+
"eval_loss": 0.5467780828475952,
|
| 1161 |
+
"eval_runtime": 12.0926,
|
| 1162 |
+
"eval_samples_per_second": 39.694,
|
| 1163 |
+
"eval_steps_per_second": 2.481,
|
| 1164 |
"step": 770
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 0.8271474019088016,
|
| 1168 |
+
"grad_norm": 0.7982149124145508,
|
| 1169 |
"learning_rate": 3.621420996818664e-05,
|
| 1170 |
+
"loss": 0.4224,
|
| 1171 |
+
"step": 780
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8271474019088016,
|
| 1175 |
+
"eval_loss": 0.5447956323623657,
|
| 1176 |
+
"eval_runtime": 12.0968,
|
| 1177 |
+
"eval_samples_per_second": 39.68,
|
| 1178 |
+
"eval_steps_per_second": 2.48,
|
| 1179 |
"step": 780
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
"epoch": 0.8377518557794273,
|
| 1183 |
+
"grad_norm": 0.843986451625824,
|
| 1184 |
"learning_rate": 3.603746907034288e-05,
|
| 1185 |
+
"loss": 0.4519,
|
| 1186 |
"step": 790
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
+
"epoch": 0.8377518557794273,
|
| 1190 |
+
"eval_loss": 0.5447117686271667,
|
| 1191 |
+
"eval_runtime": 12.0796,
|
| 1192 |
+
"eval_samples_per_second": 39.737,
|
| 1193 |
+
"eval_steps_per_second": 2.484,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.848356309650053,
|
| 1198 |
+
"grad_norm": 0.8621006608009338,
|
| 1199 |
"learning_rate": 3.586072817249912e-05,
|
| 1200 |
+
"loss": 0.4522,
|
| 1201 |
"step": 800
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
"epoch": 0.848356309650053,
|
| 1205 |
+
"eval_loss": 0.5455495119094849,
|
| 1206 |
+
"eval_runtime": 12.0913,
|
| 1207 |
+
"eval_samples_per_second": 39.698,
|
| 1208 |
+
"eval_steps_per_second": 2.481,
|
| 1209 |
"step": 800
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
"epoch": 0.8589607635206787,
|
| 1213 |
+
"grad_norm": 0.8939923644065857,
|
| 1214 |
"learning_rate": 3.5683987274655356e-05,
|
| 1215 |
+
"loss": 0.4328,
|
| 1216 |
+
"step": 810
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.8589607635206787,
|
| 1220 |
+
"eval_loss": 0.543522298336029,
|
| 1221 |
+
"eval_runtime": 12.0966,
|
| 1222 |
+
"eval_samples_per_second": 39.68,
|
| 1223 |
+
"eval_steps_per_second": 2.48,
|
| 1224 |
"step": 810
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
"epoch": 0.8695652173913043,
|
| 1228 |
+
"grad_norm": 0.7654563784599304,
|
| 1229 |
"learning_rate": 3.5507246376811596e-05,
|
| 1230 |
+
"loss": 0.408,
|
| 1231 |
+
"step": 820
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.8695652173913043,
|
| 1235 |
+
"eval_loss": 0.5410619378089905,
|
| 1236 |
+
"eval_runtime": 12.1165,
|
| 1237 |
+
"eval_samples_per_second": 39.615,
|
| 1238 |
+
"eval_steps_per_second": 2.476,
|
| 1239 |
"step": 820
|
| 1240 |
},
|
| 1241 |
{
|
| 1242 |
"epoch": 0.88016967126193,
|
| 1243 |
+
"grad_norm": 0.8473424315452576,
|
| 1244 |
"learning_rate": 3.5330505478967835e-05,
|
| 1245 |
+
"loss": 0.4285,
|
| 1246 |
+
"step": 830
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.88016967126193,
|
| 1250 |
+
"eval_loss": 0.5404618978500366,
|
| 1251 |
+
"eval_runtime": 12.0956,
|
| 1252 |
+
"eval_samples_per_second": 39.684,
|
| 1253 |
+
"eval_steps_per_second": 2.48,
|
| 1254 |
"step": 830
|
| 1255 |
},
|
| 1256 |
{
|
| 1257 |
"epoch": 0.8907741251325557,
|
| 1258 |
+
"grad_norm": 0.9181646108627319,
|
| 1259 |
"learning_rate": 3.5153764581124074e-05,
|
| 1260 |
+
"loss": 0.438,
|
| 1261 |
+
"step": 840
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.8907741251325557,
|
| 1265 |
+
"eval_loss": 0.5407660007476807,
|
| 1266 |
+
"eval_runtime": 12.0993,
|
| 1267 |
+
"eval_samples_per_second": 39.672,
|
| 1268 |
+
"eval_steps_per_second": 2.479,
|
| 1269 |
"step": 840
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 0.9013785790031813,
|
| 1273 |
+
"grad_norm": 0.837807297706604,
|
| 1274 |
"learning_rate": 3.497702368328031e-05,
|
| 1275 |
+
"loss": 0.4143,
|
| 1276 |
"step": 850
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
"epoch": 0.9013785790031813,
|
| 1280 |
+
"eval_loss": 0.5391095280647278,
|
| 1281 |
+
"eval_runtime": 12.0981,
|
| 1282 |
+
"eval_samples_per_second": 39.676,
|
| 1283 |
+
"eval_steps_per_second": 2.48,
|
| 1284 |
"step": 850
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
"epoch": 0.911983032873807,
|
| 1288 |
+
"grad_norm": 0.8593597412109375,
|
| 1289 |
"learning_rate": 3.480028278543655e-05,
|
| 1290 |
+
"loss": 0.4265,
|
| 1291 |
+
"step": 860
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.911983032873807,
|
| 1295 |
+
"eval_loss": 0.5380513668060303,
|
| 1296 |
+
"eval_runtime": 12.106,
|
| 1297 |
+
"eval_samples_per_second": 39.65,
|
| 1298 |
+
"eval_steps_per_second": 2.478,
|
| 1299 |
"step": 860
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
"epoch": 0.9225874867444327,
|
| 1303 |
+
"grad_norm": 0.8957661986351013,
|
| 1304 |
"learning_rate": 3.462354188759279e-05,
|
| 1305 |
+
"loss": 0.4285,
|
| 1306 |
+
"step": 870
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.9225874867444327,
|
| 1310 |
+
"eval_loss": 0.5382495522499084,
|
| 1311 |
+
"eval_runtime": 12.1079,
|
| 1312 |
+
"eval_samples_per_second": 39.644,
|
| 1313 |
+
"eval_steps_per_second": 2.478,
|
| 1314 |
"step": 870
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
"epoch": 0.9331919406150583,
|
| 1318 |
+
"grad_norm": 0.9220432043075562,
|
| 1319 |
"learning_rate": 3.444680098974903e-05,
|
| 1320 |
+
"loss": 0.4302,
|
| 1321 |
+
"step": 880
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.9331919406150583,
|
| 1325 |
+
"eval_loss": 0.5378026366233826,
|
| 1326 |
+
"eval_runtime": 12.1094,
|
| 1327 |
+
"eval_samples_per_second": 39.639,
|
| 1328 |
+
"eval_steps_per_second": 2.477,
|
| 1329 |
"step": 880
|
| 1330 |
},
|
| 1331 |
{
|
| 1332 |
"epoch": 0.943796394485684,
|
| 1333 |
+
"grad_norm": 0.9168163537979126,
|
| 1334 |
"learning_rate": 3.427006009190527e-05,
|
| 1335 |
+
"loss": 0.4173,
|
| 1336 |
+
"step": 890
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.943796394485684,
|
| 1340 |
+
"eval_loss": 0.5384272933006287,
|
| 1341 |
+
"eval_runtime": 12.0854,
|
| 1342 |
+
"eval_samples_per_second": 39.717,
|
| 1343 |
+
"eval_steps_per_second": 2.482,
|
| 1344 |
"step": 890
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
"epoch": 0.9544008483563097,
|
| 1348 |
+
"grad_norm": 0.7673654556274414,
|
| 1349 |
"learning_rate": 3.409331919406151e-05,
|
| 1350 |
+
"loss": 0.4088,
|
| 1351 |
"step": 900
|
| 1352 |
},
|
| 1353 |
{
|
| 1354 |
"epoch": 0.9544008483563097,
|
| 1355 |
+
"eval_loss": 0.5370467305183411,
|
| 1356 |
+
"eval_runtime": 12.0929,
|
| 1357 |
+
"eval_samples_per_second": 39.693,
|
| 1358 |
+
"eval_steps_per_second": 2.481,
|
| 1359 |
"step": 900
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
"epoch": 0.9650053022269353,
|
| 1363 |
+
"grad_norm": 0.8615649938583374,
|
| 1364 |
"learning_rate": 3.391657829621775e-05,
|
| 1365 |
+
"loss": 0.4415,
|
| 1366 |
+
"step": 910
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.9650053022269353,
|
| 1370 |
+
"eval_loss": 0.5357659459114075,
|
| 1371 |
+
"eval_runtime": 12.1061,
|
| 1372 |
+
"eval_samples_per_second": 39.649,
|
| 1373 |
+
"eval_steps_per_second": 2.478,
|
| 1374 |
"step": 910
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
"epoch": 0.975609756097561,
|
| 1378 |
+
"grad_norm": 0.79376620054245,
|
| 1379 |
"learning_rate": 3.373983739837399e-05,
|
| 1380 |
+
"loss": 0.4139,
|
| 1381 |
+
"step": 920
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.975609756097561,
|
| 1385 |
+
"eval_loss": 0.5371720790863037,
|
| 1386 |
+
"eval_runtime": 12.0998,
|
| 1387 |
+
"eval_samples_per_second": 39.67,
|
| 1388 |
+
"eval_steps_per_second": 2.479,
|
| 1389 |
"step": 920
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 0.9862142099681867,
|
| 1393 |
+
"grad_norm": 0.8736401796340942,
|
| 1394 |
"learning_rate": 3.356309650053022e-05,
|
| 1395 |
+
"loss": 0.4027,
|
| 1396 |
+
"step": 930
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.9862142099681867,
|
| 1400 |
+
"eval_loss": 0.5382711887359619,
|
| 1401 |
+
"eval_runtime": 12.1062,
|
| 1402 |
+
"eval_samples_per_second": 39.649,
|
| 1403 |
+
"eval_steps_per_second": 2.478,
|
| 1404 |
"step": 930
|
| 1405 |
},
|
| 1406 |
{
|
| 1407 |
"epoch": 0.9968186638388123,
|
| 1408 |
+
"grad_norm": 0.8468198776245117,
|
| 1409 |
"learning_rate": 3.3386355602686466e-05,
|
| 1410 |
"loss": 0.4208,
|
| 1411 |
"step": 940
|
| 1412 |
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.9968186638388123,
|
| 1415 |
+
"eval_loss": 0.536177933216095,
|
| 1416 |
+
"eval_runtime": 12.0937,
|
| 1417 |
+
"eval_samples_per_second": 39.69,
|
| 1418 |
+
"eval_steps_per_second": 2.481,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
{
|
| 1422 |
"epoch": 1.007423117709438,
|
| 1423 |
+
"grad_norm": 0.8486480116844177,
|
| 1424 |
"learning_rate": 3.32096147048427e-05,
|
| 1425 |
+
"loss": 0.3802,
|
| 1426 |
"step": 950
|
| 1427 |
},
|
| 1428 |
{
|
| 1429 |
"epoch": 1.007423117709438,
|
| 1430 |
+
"eval_loss": 0.5357194542884827,
|
| 1431 |
+
"eval_runtime": 12.1057,
|
| 1432 |
+
"eval_samples_per_second": 39.651,
|
| 1433 |
+
"eval_steps_per_second": 2.478,
|
| 1434 |
"step": 950
|
| 1435 |
},
|
| 1436 |
{
|
| 1437 |
"epoch": 1.0180275715800637,
|
| 1438 |
+
"grad_norm": 0.901928186416626,
|
| 1439 |
"learning_rate": 3.3032873806998944e-05,
|
| 1440 |
+
"loss": 0.3791,
|
| 1441 |
+
"step": 960
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.0180275715800637,
|
| 1445 |
+
"eval_loss": 0.5389688014984131,
|
| 1446 |
+
"eval_runtime": 12.1026,
|
| 1447 |
+
"eval_samples_per_second": 39.661,
|
| 1448 |
+
"eval_steps_per_second": 2.479,
|
| 1449 |
"step": 960
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
"epoch": 1.0286320254506893,
|
| 1453 |
+
"grad_norm": 0.8977916836738586,
|
| 1454 |
"learning_rate": 3.285613290915518e-05,
|
| 1455 |
+
"loss": 0.4038,
|
| 1456 |
+
"step": 970
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.0286320254506893,
|
| 1460 |
+
"eval_loss": 0.5394784212112427,
|
| 1461 |
+
"eval_runtime": 12.1121,
|
| 1462 |
+
"eval_samples_per_second": 39.63,
|
| 1463 |
+
"eval_steps_per_second": 2.477,
|
| 1464 |
"step": 970
|
| 1465 |
},
|
| 1466 |
{
|
| 1467 |
"epoch": 1.039236479321315,
|
| 1468 |
+
"grad_norm": 0.8839123845100403,
|
| 1469 |
"learning_rate": 3.267939201131142e-05,
|
| 1470 |
+
"loss": 0.3824,
|
| 1471 |
+
"step": 980
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 1.039236479321315,
|
| 1475 |
+
"eval_loss": 0.5386835932731628,
|
| 1476 |
+
"eval_runtime": 12.1082,
|
| 1477 |
+
"eval_samples_per_second": 39.642,
|
| 1478 |
+
"eval_steps_per_second": 2.478,
|
| 1479 |
"step": 980
|
| 1480 |
},
|
| 1481 |
{
|
| 1482 |
"epoch": 1.0498409331919407,
|
| 1483 |
+
"grad_norm": 0.8673653602600098,
|
| 1484 |
"learning_rate": 3.2502651113467655e-05,
|
| 1485 |
+
"loss": 0.361,
|
| 1486 |
+
"step": 990
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 1.0498409331919407,
|
| 1490 |
+
"eval_loss": 0.5394459962844849,
|
| 1491 |
+
"eval_runtime": 12.0998,
|
| 1492 |
+
"eval_samples_per_second": 39.67,
|
| 1493 |
+
"eval_steps_per_second": 2.479,
|
| 1494 |
"step": 990
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
"epoch": 1.0604453870625663,
|
| 1498 |
+
"grad_norm": 0.8955957889556885,
|
| 1499 |
"learning_rate": 3.2325910215623894e-05,
|
| 1500 |
+
"loss": 0.3757,
|
| 1501 |
"step": 1000
|
| 1502 |
},
|
| 1503 |
{
|
| 1504 |
"epoch": 1.0604453870625663,
|
| 1505 |
+
"eval_loss": 0.539962649345398,
|
| 1506 |
+
"eval_runtime": 12.0999,
|
| 1507 |
+
"eval_samples_per_second": 39.67,
|
| 1508 |
+
"eval_steps_per_second": 2.479,
|
| 1509 |
"step": 1000
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
"epoch": 1.0710498409331919,
|
| 1513 |
+
"grad_norm": 0.927700936794281,
|
| 1514 |
"learning_rate": 3.2149169317780133e-05,
|
| 1515 |
+
"loss": 0.3693,
|
| 1516 |
+
"step": 1010
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 1.0710498409331919,
|
| 1520 |
+
"eval_loss": 0.538655698299408,
|
| 1521 |
+
"eval_runtime": 12.0888,
|
| 1522 |
+
"eval_samples_per_second": 39.706,
|
| 1523 |
+
"eval_steps_per_second": 2.482,
|
| 1524 |
"step": 1010
|
| 1525 |
},
|
| 1526 |
{
|
| 1527 |
"epoch": 1.0816542948038177,
|
| 1528 |
+
"grad_norm": 0.8936448097229004,
|
| 1529 |
"learning_rate": 3.197242841993637e-05,
|
| 1530 |
+
"loss": 0.3951,
|
| 1531 |
+
"step": 1020
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 1.0816542948038177,
|
| 1535 |
+
"eval_loss": 0.5395565629005432,
|
| 1536 |
+
"eval_runtime": 12.0892,
|
| 1537 |
+
"eval_samples_per_second": 39.705,
|
| 1538 |
+
"eval_steps_per_second": 2.482,
|
| 1539 |
"step": 1020
|
| 1540 |
},
|
| 1541 |
{
|
| 1542 |
"epoch": 1.0922587486744433,
|
| 1543 |
+
"grad_norm": 0.9137620329856873,
|
| 1544 |
"learning_rate": 3.179568752209262e-05,
|
| 1545 |
+
"loss": 0.4011,
|
| 1546 |
+
"step": 1030
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 1.0922587486744433,
|
| 1550 |
+
"eval_loss": 0.5394957661628723,
|
| 1551 |
+
"eval_runtime": 12.0863,
|
| 1552 |
+
"eval_samples_per_second": 39.714,
|
| 1553 |
+
"eval_steps_per_second": 2.482,
|
| 1554 |
"step": 1030
|
| 1555 |
},
|
| 1556 |
{
|
| 1557 |
"epoch": 1.1028632025450689,
|
| 1558 |
+
"grad_norm": 0.9682084918022156,
|
| 1559 |
"learning_rate": 3.161894662424885e-05,
|
| 1560 |
+
"loss": 0.3682,
|
| 1561 |
+
"step": 1040
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.1028632025450689,
|
| 1565 |
+
"eval_loss": 0.5405410528182983,
|
| 1566 |
+
"eval_runtime": 12.0995,
|
| 1567 |
+
"eval_samples_per_second": 39.671,
|
| 1568 |
+
"eval_steps_per_second": 2.479,
|
| 1569 |
"step": 1040
|
| 1570 |
},
|
| 1571 |
{
|
| 1572 |
"epoch": 1.1134676564156947,
|
| 1573 |
+
"grad_norm": 0.85400390625,
|
| 1574 |
"learning_rate": 3.14422057264051e-05,
|
| 1575 |
+
"loss": 0.3661,
|
| 1576 |
"step": 1050
|
| 1577 |
},
|
| 1578 |
{
|
| 1579 |
"epoch": 1.1134676564156947,
|
| 1580 |
+
"eval_loss": 0.5404468774795532,
|
| 1581 |
+
"eval_runtime": 12.0829,
|
| 1582 |
+
"eval_samples_per_second": 39.726,
|
| 1583 |
+
"eval_steps_per_second": 2.483,
|
| 1584 |
"step": 1050
|
| 1585 |
},
|
| 1586 |
{
|
| 1587 |
"epoch": 1.1240721102863203,
|
| 1588 |
+
"grad_norm": 0.9295483231544495,
|
| 1589 |
"learning_rate": 3.126546482856133e-05,
|
| 1590 |
+
"loss": 0.3723,
|
| 1591 |
+
"step": 1060
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 1.1240721102863203,
|
| 1595 |
+
"eval_loss": 0.5390296578407288,
|
| 1596 |
+
"eval_runtime": 12.0991,
|
| 1597 |
+
"eval_samples_per_second": 39.673,
|
| 1598 |
+
"eval_steps_per_second": 2.48,
|
| 1599 |
"step": 1060
|
| 1600 |
},
|
| 1601 |
{
|
| 1602 |
"epoch": 1.1346765641569458,
|
| 1603 |
+
"grad_norm": 0.8571534752845764,
|
| 1604 |
"learning_rate": 3.108872393071757e-05,
|
| 1605 |
+
"loss": 0.4045,
|
| 1606 |
+
"step": 1070
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.1346765641569458,
|
| 1610 |
+
"eval_loss": 0.5403757691383362,
|
| 1611 |
+
"eval_runtime": 12.1222,
|
| 1612 |
+
"eval_samples_per_second": 39.597,
|
| 1613 |
+
"eval_steps_per_second": 2.475,
|
| 1614 |
"step": 1070
|
| 1615 |
},
|
| 1616 |
{
|
| 1617 |
"epoch": 1.1452810180275717,
|
| 1618 |
+
"grad_norm": 0.8732081651687622,
|
| 1619 |
"learning_rate": 3.091198303287381e-05,
|
| 1620 |
+
"loss": 0.3635,
|
| 1621 |
+
"step": 1080
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 1.1452810180275717,
|
| 1625 |
+
"eval_loss": 0.5405797362327576,
|
| 1626 |
+
"eval_runtime": 12.0953,
|
| 1627 |
+
"eval_samples_per_second": 39.685,
|
| 1628 |
+
"eval_steps_per_second": 2.48,
|
| 1629 |
"step": 1080
|
| 1630 |
},
|
| 1631 |
{
|
| 1632 |
"epoch": 1.1558854718981972,
|
| 1633 |
+
"grad_norm": 0.9684129357337952,
|
| 1634 |
"learning_rate": 3.073524213503005e-05,
|
| 1635 |
+
"loss": 0.3678,
|
| 1636 |
+
"step": 1090
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 1.1558854718981972,
|
| 1640 |
+
"eval_loss": 0.5389444828033447,
|
| 1641 |
+
"eval_runtime": 12.0812,
|
| 1642 |
+
"eval_samples_per_second": 39.731,
|
| 1643 |
+
"eval_steps_per_second": 2.483,
|
| 1644 |
"step": 1090
|
| 1645 |
},
|
| 1646 |
{
|
| 1647 |
"epoch": 1.1664899257688228,
|
| 1648 |
+
"grad_norm": 0.9644559025764465,
|
| 1649 |
"learning_rate": 3.0558501237186286e-05,
|
| 1650 |
+
"loss": 0.3805,
|
| 1651 |
"step": 1100
|
| 1652 |
},
|
| 1653 |
{
|
| 1654 |
"epoch": 1.1664899257688228,
|
| 1655 |
+
"eval_loss": 0.5397114157676697,
|
| 1656 |
+
"eval_runtime": 12.099,
|
| 1657 |
+
"eval_samples_per_second": 39.673,
|
| 1658 |
+
"eval_steps_per_second": 2.48,
|
| 1659 |
"step": 1100
|
| 1660 |
},
|
| 1661 |
{
|
| 1662 |
"epoch": 1.1770943796394486,
|
| 1663 |
+
"grad_norm": 0.9264994859695435,
|
| 1664 |
"learning_rate": 3.0381760339342525e-05,
|
| 1665 |
+
"loss": 0.3735,
|
| 1666 |
+
"step": 1110
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 1.1770943796394486,
|
| 1670 |
+
"eval_loss": 0.5386097431182861,
|
| 1671 |
+
"eval_runtime": 12.0907,
|
| 1672 |
+
"eval_samples_per_second": 39.7,
|
| 1673 |
+
"eval_steps_per_second": 2.481,
|
| 1674 |
"step": 1110
|
| 1675 |
},
|
| 1676 |
{
|
| 1677 |
"epoch": 1.1876988335100742,
|
| 1678 |
+
"grad_norm": 0.9708051085472107,
|
| 1679 |
"learning_rate": 3.0205019441498768e-05,
|
| 1680 |
"loss": 0.3706,
|
| 1681 |
"step": 1120
|
| 1682 |
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 1.1876988335100742,
|
| 1685 |
+
"eval_loss": 0.5378307700157166,
|
| 1686 |
+
"eval_runtime": 12.1091,
|
| 1687 |
+
"eval_samples_per_second": 39.64,
|
| 1688 |
+
"eval_steps_per_second": 2.477,
|
| 1689 |
+
"step": 1120
|
| 1690 |
+
},
|
| 1691 |
{
|
| 1692 |
"epoch": 1.1983032873806998,
|
| 1693 |
+
"grad_norm": 0.9819815754890442,
|
| 1694 |
"learning_rate": 3.0028278543655004e-05,
|
| 1695 |
+
"loss": 0.3604,
|
| 1696 |
+
"step": 1130
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 1.1983032873806998,
|
| 1700 |
+
"eval_loss": 0.537283718585968,
|
| 1701 |
+
"eval_runtime": 12.0976,
|
| 1702 |
+
"eval_samples_per_second": 39.677,
|
| 1703 |
+
"eval_steps_per_second": 2.48,
|
| 1704 |
"step": 1130
|
| 1705 |
},
|
| 1706 |
{
|
| 1707 |
"epoch": 1.2089077412513256,
|
| 1708 |
+
"grad_norm": 0.8939089775085449,
|
| 1709 |
"learning_rate": 2.985153764581124e-05,
|
| 1710 |
+
"loss": 0.3843,
|
| 1711 |
+
"step": 1140
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.2089077412513256,
|
| 1715 |
+
"eval_loss": 0.5357462167739868,
|
| 1716 |
+
"eval_runtime": 12.1052,
|
| 1717 |
+
"eval_samples_per_second": 39.652,
|
| 1718 |
+
"eval_steps_per_second": 2.478,
|
| 1719 |
"step": 1140
|
| 1720 |
},
|
| 1721 |
{
|
| 1722 |
"epoch": 1.2195121951219512,
|
| 1723 |
+
"grad_norm": 0.9389526844024658,
|
| 1724 |
"learning_rate": 2.9674796747967482e-05,
|
| 1725 |
+
"loss": 0.3933,
|
| 1726 |
"step": 1150
|
| 1727 |
},
|
| 1728 |
{
|
| 1729 |
"epoch": 1.2195121951219512,
|
| 1730 |
+
"eval_loss": 0.5356990098953247,
|
| 1731 |
+
"eval_runtime": 12.0919,
|
| 1732 |
+
"eval_samples_per_second": 39.696,
|
| 1733 |
+
"eval_steps_per_second": 2.481,
|
| 1734 |
"step": 1150
|
| 1735 |
},
|
| 1736 |
{
|
| 1737 |
"epoch": 1.2301166489925768,
|
| 1738 |
+
"grad_norm": 0.9576265811920166,
|
| 1739 |
"learning_rate": 2.9498055850123718e-05,
|
| 1740 |
+
"loss": 0.3524,
|
| 1741 |
+
"step": 1160
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 1.2301166489925768,
|
| 1745 |
+
"eval_loss": 0.5370804071426392,
|
| 1746 |
+
"eval_runtime": 12.1005,
|
| 1747 |
+
"eval_samples_per_second": 39.668,
|
| 1748 |
+
"eval_steps_per_second": 2.479,
|
| 1749 |
"step": 1160
|
| 1750 |
},
|
| 1751 |
{
|
| 1752 |
"epoch": 1.2407211028632026,
|
| 1753 |
+
"grad_norm": 1.1763864755630493,
|
| 1754 |
"learning_rate": 2.932131495227996e-05,
|
| 1755 |
+
"loss": 0.3874,
|
| 1756 |
+
"step": 1170
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 1.2407211028632026,
|
| 1760 |
+
"eval_loss": 0.5390760898590088,
|
| 1761 |
+
"eval_runtime": 12.0929,
|
| 1762 |
+
"eval_samples_per_second": 39.693,
|
| 1763 |
+
"eval_steps_per_second": 2.481,
|
| 1764 |
"step": 1170
|
| 1765 |
},
|
| 1766 |
{
|
| 1767 |
"epoch": 1.2513255567338282,
|
| 1768 |
+
"grad_norm": 1.0538442134857178,
|
| 1769 |
"learning_rate": 2.9144574054436196e-05,
|
| 1770 |
+
"loss": 0.3749,
|
| 1771 |
+
"step": 1180
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 1.2513255567338282,
|
| 1775 |
+
"eval_loss": 0.5350478887557983,
|
| 1776 |
+
"eval_runtime": 12.0888,
|
| 1777 |
+
"eval_samples_per_second": 39.706,
|
| 1778 |
+
"eval_steps_per_second": 2.482,
|
| 1779 |
"step": 1180
|
| 1780 |
},
|
| 1781 |
{
|
| 1782 |
"epoch": 1.2619300106044538,
|
| 1783 |
+
"grad_norm": 1.0100423097610474,
|
| 1784 |
"learning_rate": 2.896783315659244e-05,
|
| 1785 |
+
"loss": 0.3661,
|
| 1786 |
+
"step": 1190
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 1.2619300106044538,
|
| 1790 |
+
"eval_loss": 0.534929633140564,
|
| 1791 |
+
"eval_runtime": 12.0936,
|
| 1792 |
+
"eval_samples_per_second": 39.691,
|
| 1793 |
+
"eval_steps_per_second": 2.481,
|
| 1794 |
"step": 1190
|
| 1795 |
},
|
| 1796 |
{
|
| 1797 |
"epoch": 1.2725344644750796,
|
| 1798 |
+
"grad_norm": 0.9935658574104309,
|
| 1799 |
"learning_rate": 2.8791092258748675e-05,
|
| 1800 |
+
"loss": 0.3734,
|
| 1801 |
"step": 1200
|
| 1802 |
},
|
| 1803 |
{
|
| 1804 |
"epoch": 1.2725344644750796,
|
| 1805 |
+
"eval_loss": 0.5362860560417175,
|
| 1806 |
+
"eval_runtime": 12.1044,
|
| 1807 |
+
"eval_samples_per_second": 39.655,
|
| 1808 |
+
"eval_steps_per_second": 2.478,
|
| 1809 |
"step": 1200
|
| 1810 |
},
|
| 1811 |
{
|
| 1812 |
"epoch": 1.2831389183457051,
|
| 1813 |
+
"grad_norm": 0.9751375317573547,
|
| 1814 |
"learning_rate": 2.8614351360904917e-05,
|
| 1815 |
+
"loss": 0.3767,
|
| 1816 |
+
"step": 1210
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.2831389183457051,
|
| 1820 |
+
"eval_loss": 0.536407470703125,
|
| 1821 |
+
"eval_runtime": 12.091,
|
| 1822 |
+
"eval_samples_per_second": 39.699,
|
| 1823 |
+
"eval_steps_per_second": 2.481,
|
| 1824 |
"step": 1210
|
| 1825 |
},
|
| 1826 |
{
|
| 1827 |
"epoch": 1.2937433722163307,
|
| 1828 |
+
"grad_norm": 0.9417859315872192,
|
| 1829 |
"learning_rate": 2.8437610463061153e-05,
|
| 1830 |
+
"loss": 0.3592,
|
| 1831 |
+
"step": 1220
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 1.2937433722163307,
|
| 1835 |
+
"eval_loss": 0.5357491374015808,
|
| 1836 |
+
"eval_runtime": 12.0892,
|
| 1837 |
+
"eval_samples_per_second": 39.705,
|
| 1838 |
+
"eval_steps_per_second": 2.482,
|
| 1839 |
"step": 1220
|
| 1840 |
},
|
| 1841 |
{
|
| 1842 |
"epoch": 1.3043478260869565,
|
| 1843 |
+
"grad_norm": 0.9397806525230408,
|
| 1844 |
"learning_rate": 2.826086956521739e-05,
|
| 1845 |
+
"loss": 0.3675,
|
| 1846 |
+
"step": 1230
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 1.3043478260869565,
|
| 1850 |
+
"eval_loss": 0.5362255573272705,
|
| 1851 |
+
"eval_runtime": 12.0809,
|
| 1852 |
+
"eval_samples_per_second": 39.732,
|
| 1853 |
+
"eval_steps_per_second": 2.483,
|
| 1854 |
"step": 1230
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
"epoch": 1.3149522799575821,
|
| 1858 |
+
"grad_norm": 1.0435277223587036,
|
| 1859 |
"learning_rate": 2.808412866737363e-05,
|
| 1860 |
+
"loss": 0.3627,
|
| 1861 |
+
"step": 1240
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 1.3149522799575821,
|
| 1865 |
+
"eval_loss": 0.5366498231887817,
|
| 1866 |
+
"eval_runtime": 12.0831,
|
| 1867 |
+
"eval_samples_per_second": 39.725,
|
| 1868 |
+
"eval_steps_per_second": 2.483,
|
| 1869 |
"step": 1240
|
| 1870 |
},
|
| 1871 |
{
|
| 1872 |
"epoch": 1.3255567338282077,
|
| 1873 |
+
"grad_norm": 1.0592031478881836,
|
| 1874 |
"learning_rate": 2.790738776952987e-05,
|
| 1875 |
+
"loss": 0.3898,
|
| 1876 |
"step": 1250
|
| 1877 |
},
|
| 1878 |
{
|
| 1879 |
"epoch": 1.3255567338282077,
|
| 1880 |
+
"eval_loss": 0.5350601673126221,
|
| 1881 |
+
"eval_runtime": 12.0837,
|
| 1882 |
+
"eval_samples_per_second": 39.723,
|
| 1883 |
+
"eval_steps_per_second": 2.483,
|
| 1884 |
"step": 1250
|
| 1885 |
},
|
| 1886 |
{
|
| 1887 |
"epoch": 1.3361611876988335,
|
| 1888 |
+
"grad_norm": 0.995845377445221,
|
| 1889 |
"learning_rate": 2.773064687168611e-05,
|
| 1890 |
"loss": 0.3554,
|
| 1891 |
"step": 1260
|
| 1892 |
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 1.3361611876988335,
|
| 1895 |
+
"eval_loss": 0.5366798639297485,
|
| 1896 |
+
"eval_runtime": 12.0969,
|
| 1897 |
+
"eval_samples_per_second": 39.68,
|
| 1898 |
+
"eval_steps_per_second": 2.48,
|
| 1899 |
+
"step": 1260
|
| 1900 |
+
},
|
| 1901 |
{
|
| 1902 |
"epoch": 1.346765641569459,
|
| 1903 |
+
"grad_norm": 0.993667721748352,
|
| 1904 |
"learning_rate": 2.755390597384235e-05,
|
| 1905 |
+
"loss": 0.3891,
|
| 1906 |
+
"step": 1270
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 1.346765641569459,
|
| 1910 |
+
"eval_loss": 0.5358466506004333,
|
| 1911 |
+
"eval_runtime": 12.0659,
|
| 1912 |
+
"eval_samples_per_second": 39.782,
|
| 1913 |
+
"eval_steps_per_second": 2.486,
|
| 1914 |
"step": 1270
|
| 1915 |
},
|
| 1916 |
{
|
| 1917 |
"epoch": 1.357370095440085,
|
| 1918 |
+
"grad_norm": 1.1085600852966309,
|
| 1919 |
"learning_rate": 2.7377165075998588e-05,
|
| 1920 |
+
"loss": 0.3809,
|
| 1921 |
+
"step": 1280
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.357370095440085,
|
| 1925 |
+
"eval_loss": 0.5354708433151245,
|
| 1926 |
+
"eval_runtime": 12.0712,
|
| 1927 |
+
"eval_samples_per_second": 39.764,
|
| 1928 |
+
"eval_steps_per_second": 2.485,
|
| 1929 |
"step": 1280
|
| 1930 |
},
|
| 1931 |
{
|
| 1932 |
"epoch": 1.3679745493107105,
|
| 1933 |
+
"grad_norm": 0.9772036075592041,
|
| 1934 |
"learning_rate": 2.7200424178154827e-05,
|
| 1935 |
+
"loss": 0.3804,
|
| 1936 |
+
"step": 1290
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 1.3679745493107105,
|
| 1940 |
+
"eval_loss": 0.5343257188796997,
|
| 1941 |
+
"eval_runtime": 12.0879,
|
| 1942 |
+
"eval_samples_per_second": 39.709,
|
| 1943 |
+
"eval_steps_per_second": 2.482,
|
| 1944 |
"step": 1290
|
| 1945 |
},
|
| 1946 |
{
|
| 1947 |
"epoch": 1.378579003181336,
|
| 1948 |
+
"grad_norm": 0.9579578042030334,
|
| 1949 |
"learning_rate": 2.7023683280311063e-05,
|
| 1950 |
+
"loss": 0.3721,
|
| 1951 |
"step": 1300
|
| 1952 |
},
|
| 1953 |
{
|
| 1954 |
"epoch": 1.378579003181336,
|
| 1955 |
+
"eval_loss": 0.5338610410690308,
|
| 1956 |
+
"eval_runtime": 12.0714,
|
| 1957 |
+
"eval_samples_per_second": 39.763,
|
| 1958 |
+
"eval_steps_per_second": 2.485,
|
| 1959 |
"step": 1300
|
| 1960 |
},
|
| 1961 |
{
|
| 1962 |
"epoch": 1.389183457051962,
|
| 1963 |
+
"grad_norm": 1.0014525651931763,
|
| 1964 |
"learning_rate": 2.6846942382467306e-05,
|
| 1965 |
+
"loss": 0.3618,
|
| 1966 |
+
"step": 1310
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 1.389183457051962,
|
| 1970 |
+
"eval_loss": 0.5343756079673767,
|
| 1971 |
+
"eval_runtime": 12.0719,
|
| 1972 |
+
"eval_samples_per_second": 39.762,
|
| 1973 |
+
"eval_steps_per_second": 2.485,
|
| 1974 |
"step": 1310
|
| 1975 |
},
|
| 1976 |
{
|
| 1977 |
"epoch": 1.3997879109225875,
|
| 1978 |
+
"grad_norm": 0.9999486804008484,
|
| 1979 |
"learning_rate": 2.667020148462354e-05,
|
| 1980 |
+
"loss": 0.3673,
|
| 1981 |
+
"step": 1320
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 1.3997879109225875,
|
| 1985 |
+
"eval_loss": 0.5333850979804993,
|
| 1986 |
+
"eval_runtime": 12.0766,
|
| 1987 |
+
"eval_samples_per_second": 39.746,
|
| 1988 |
+
"eval_steps_per_second": 2.484,
|
| 1989 |
"step": 1320
|
| 1990 |
},
|
| 1991 |
{
|
| 1992 |
"epoch": 1.4103923647932133,
|
| 1993 |
+
"grad_norm": 1.1561279296875,
|
| 1994 |
"learning_rate": 2.6493460586779784e-05,
|
| 1995 |
"loss": 0.3619,
|
| 1996 |
"step": 1330
|
| 1997 |
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 1.4103923647932133,
|
| 2000 |
+
"eval_loss": 0.5335540771484375,
|
| 2001 |
+
"eval_runtime": 12.0771,
|
| 2002 |
+
"eval_samples_per_second": 39.745,
|
| 2003 |
+
"eval_steps_per_second": 2.484,
|
| 2004 |
+
"step": 1330
|
| 2005 |
+
},
|
| 2006 |
{
|
| 2007 |
"epoch": 1.4209968186638389,
|
| 2008 |
+
"grad_norm": 0.9691271185874939,
|
| 2009 |
"learning_rate": 2.631671968893602e-05,
|
| 2010 |
+
"loss": 0.3682,
|
| 2011 |
+
"step": 1340
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 1.4209968186638389,
|
| 2015 |
+
"eval_loss": 0.5338456034660339,
|
| 2016 |
+
"eval_runtime": 12.0787,
|
| 2017 |
+
"eval_samples_per_second": 39.74,
|
| 2018 |
+
"eval_steps_per_second": 2.484,
|
| 2019 |
"step": 1340
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
"epoch": 1.4316012725344645,
|
| 2023 |
+
"grad_norm": 1.0488156080245972,
|
| 2024 |
"learning_rate": 2.6139978791092262e-05,
|
| 2025 |
+
"loss": 0.3581,
|
| 2026 |
"step": 1350
|
| 2027 |
},
|
| 2028 |
{
|
| 2029 |
"epoch": 1.4316012725344645,
|
| 2030 |
+
"eval_loss": 0.533444344997406,
|
| 2031 |
+
"eval_runtime": 12.0798,
|
| 2032 |
+
"eval_samples_per_second": 39.736,
|
| 2033 |
+
"eval_steps_per_second": 2.483,
|
| 2034 |
"step": 1350
|
| 2035 |
},
|
| 2036 |
{
|
| 2037 |
"epoch": 1.4422057264050903,
|
| 2038 |
+
"grad_norm": 1.0487726926803589,
|
| 2039 |
"learning_rate": 2.5963237893248498e-05,
|
| 2040 |
+
"loss": 0.3909,
|
| 2041 |
+
"step": 1360
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 1.4422057264050903,
|
| 2045 |
+
"eval_loss": 0.5327271223068237,
|
| 2046 |
+
"eval_runtime": 12.0828,
|
| 2047 |
+
"eval_samples_per_second": 39.726,
|
| 2048 |
+
"eval_steps_per_second": 2.483,
|
| 2049 |
"step": 1360
|
| 2050 |
},
|
| 2051 |
{
|
| 2052 |
"epoch": 1.4528101802757158,
|
| 2053 |
+
"grad_norm": 0.9647484421730042,
|
| 2054 |
"learning_rate": 2.5786496995404734e-05,
|
| 2055 |
+
"loss": 0.3605,
|
| 2056 |
+
"step": 1370
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 1.4528101802757158,
|
| 2060 |
+
"eval_loss": 0.5315544009208679,
|
| 2061 |
+
"eval_runtime": 12.0984,
|
| 2062 |
+
"eval_samples_per_second": 39.675,
|
| 2063 |
+
"eval_steps_per_second": 2.48,
|
| 2064 |
"step": 1370
|
| 2065 |
},
|
| 2066 |
{
|
| 2067 |
"epoch": 1.4634146341463414,
|
| 2068 |
+
"grad_norm": 0.9921063780784607,
|
| 2069 |
"learning_rate": 2.5609756097560977e-05,
|
| 2070 |
"loss": 0.3786,
|
| 2071 |
"step": 1380
|
| 2072 |
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 1.4634146341463414,
|
| 2075 |
+
"eval_loss": 0.5312519073486328,
|
| 2076 |
+
"eval_runtime": 12.1289,
|
| 2077 |
+
"eval_samples_per_second": 39.575,
|
| 2078 |
+
"eval_steps_per_second": 2.473,
|
| 2079 |
+
"step": 1380
|
| 2080 |
+
},
|
| 2081 |
{
|
| 2082 |
"epoch": 1.4740190880169672,
|
| 2083 |
+
"grad_norm": 1.0455048084259033,
|
| 2084 |
"learning_rate": 2.5433015199717212e-05,
|
| 2085 |
+
"loss": 0.3621,
|
| 2086 |
+
"step": 1390
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 1.4740190880169672,
|
| 2090 |
+
"eval_loss": 0.5322020649909973,
|
| 2091 |
+
"eval_runtime": 12.1191,
|
| 2092 |
+
"eval_samples_per_second": 39.607,
|
| 2093 |
+
"eval_steps_per_second": 2.475,
|
| 2094 |
"step": 1390
|
| 2095 |
},
|
| 2096 |
{
|
| 2097 |
"epoch": 1.4846235418875928,
|
| 2098 |
+
"grad_norm": 1.0896190404891968,
|
| 2099 |
"learning_rate": 2.5256274301873455e-05,
|
| 2100 |
+
"loss": 0.3618,
|
| 2101 |
"step": 1400
|
| 2102 |
},
|
| 2103 |
{
|
| 2104 |
"epoch": 1.4846235418875928,
|
| 2105 |
+
"eval_loss": 0.5336935520172119,
|
| 2106 |
+
"eval_runtime": 12.12,
|
| 2107 |
+
"eval_samples_per_second": 39.604,
|
| 2108 |
+
"eval_steps_per_second": 2.475,
|
| 2109 |
"step": 1400
|
| 2110 |
},
|
| 2111 |
{
|
| 2112 |
"epoch": 1.4952279957582184,
|
| 2113 |
+
"grad_norm": 1.031595230102539,
|
| 2114 |
"learning_rate": 2.507953340402969e-05,
|
| 2115 |
"loss": 0.3491,
|
| 2116 |
"step": 1410
|
| 2117 |
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 1.4952279957582184,
|
| 2120 |
+
"eval_loss": 0.5328320264816284,
|
| 2121 |
+
"eval_runtime": 12.1004,
|
| 2122 |
+
"eval_samples_per_second": 39.668,
|
| 2123 |
+
"eval_steps_per_second": 2.479,
|
| 2124 |
+
"step": 1410
|
| 2125 |
+
},
|
| 2126 |
{
|
| 2127 |
"epoch": 1.5058324496288442,
|
| 2128 |
+
"grad_norm": 1.0015314817428589,
|
| 2129 |
"learning_rate": 2.4902792506185933e-05,
|
| 2130 |
+
"loss": 0.3483,
|
| 2131 |
+
"step": 1420
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.5058324496288442,
|
| 2135 |
+
"eval_loss": 0.5334290266036987,
|
| 2136 |
+
"eval_runtime": 12.0914,
|
| 2137 |
+
"eval_samples_per_second": 39.698,
|
| 2138 |
+
"eval_steps_per_second": 2.481,
|
| 2139 |
"step": 1420
|
| 2140 |
},
|
| 2141 |
{
|
| 2142 |
"epoch": 1.5164369034994698,
|
| 2143 |
+
"grad_norm": 1.1234946250915527,
|
| 2144 |
"learning_rate": 2.4726051608342173e-05,
|
| 2145 |
+
"loss": 0.3711,
|
| 2146 |
+
"step": 1430
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 1.5164369034994698,
|
| 2150 |
+
"eval_loss": 0.5323832035064697,
|
| 2151 |
+
"eval_runtime": 12.0888,
|
| 2152 |
+
"eval_samples_per_second": 39.706,
|
| 2153 |
+
"eval_steps_per_second": 2.482,
|
| 2154 |
"step": 1430
|
| 2155 |
},
|
| 2156 |
{
|
| 2157 |
"epoch": 1.5270413573700954,
|
| 2158 |
+
"grad_norm": 1.07817804813385,
|
| 2159 |
"learning_rate": 2.4549310710498412e-05,
|
| 2160 |
+
"loss": 0.3598,
|
| 2161 |
+
"step": 1440
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 1.5270413573700954,
|
| 2165 |
+
"eval_loss": 0.5321078300476074,
|
| 2166 |
+
"eval_runtime": 12.0889,
|
| 2167 |
+
"eval_samples_per_second": 39.706,
|
| 2168 |
+
"eval_steps_per_second": 2.482,
|
| 2169 |
"step": 1440
|
| 2170 |
},
|
| 2171 |
{
|
| 2172 |
"epoch": 1.5376458112407212,
|
| 2173 |
+
"grad_norm": 0.9693513512611389,
|
| 2174 |
"learning_rate": 2.437256981265465e-05,
|
| 2175 |
+
"loss": 0.355,
|
| 2176 |
"step": 1450
|
| 2177 |
},
|
| 2178 |
{
|
| 2179 |
"epoch": 1.5376458112407212,
|
| 2180 |
+
"eval_loss": 0.5310765504837036,
|
| 2181 |
+
"eval_runtime": 12.0932,
|
| 2182 |
+
"eval_samples_per_second": 39.692,
|
| 2183 |
+
"eval_steps_per_second": 2.481,
|
| 2184 |
"step": 1450
|
| 2185 |
},
|
| 2186 |
{
|
| 2187 |
"epoch": 1.5482502651113468,
|
| 2188 |
+
"grad_norm": 1.121456265449524,
|
| 2189 |
"learning_rate": 2.419582891481089e-05,
|
| 2190 |
+
"loss": 0.3856,
|
| 2191 |
+
"step": 1460
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 1.5482502651113468,
|
| 2195 |
+
"eval_loss": 0.5301222205162048,
|
| 2196 |
+
"eval_runtime": 12.0879,
|
| 2197 |
+
"eval_samples_per_second": 39.709,
|
| 2198 |
+
"eval_steps_per_second": 2.482,
|
| 2199 |
"step": 1460
|
| 2200 |
},
|
| 2201 |
{
|
| 2202 |
"epoch": 1.5588547189819724,
|
| 2203 |
+
"grad_norm": 1.0205426216125488,
|
| 2204 |
"learning_rate": 2.401908801696713e-05,
|
| 2205 |
+
"loss": 0.3532,
|
| 2206 |
+
"step": 1470
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 1.5588547189819724,
|
| 2210 |
+
"eval_loss": 0.5307323336601257,
|
| 2211 |
+
"eval_runtime": 12.0727,
|
| 2212 |
+
"eval_samples_per_second": 39.759,
|
| 2213 |
+
"eval_steps_per_second": 2.485,
|
| 2214 |
"step": 1470
|
| 2215 |
},
|
| 2216 |
{
|
| 2217 |
"epoch": 1.5694591728525982,
|
| 2218 |
+
"grad_norm": 0.9773093461990356,
|
| 2219 |
"learning_rate": 2.3842347119123365e-05,
|
| 2220 |
"loss": 0.3645,
|
| 2221 |
"step": 1480
|
| 2222 |
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 1.5694591728525982,
|
| 2225 |
+
"eval_loss": 0.5309693813323975,
|
| 2226 |
+
"eval_runtime": 12.0729,
|
| 2227 |
+
"eval_samples_per_second": 39.759,
|
| 2228 |
+
"eval_steps_per_second": 2.485,
|
| 2229 |
+
"step": 1480
|
| 2230 |
+
},
|
| 2231 |
{
|
| 2232 |
"epoch": 1.5800636267232238,
|
| 2233 |
+
"grad_norm": 0.9958512783050537,
|
| 2234 |
"learning_rate": 2.3665606221279604e-05,
|
| 2235 |
+
"loss": 0.3514,
|
| 2236 |
+
"step": 1490
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.5800636267232238,
|
| 2240 |
+
"eval_loss": 0.5321856737136841,
|
| 2241 |
+
"eval_runtime": 12.0688,
|
| 2242 |
+
"eval_samples_per_second": 39.772,
|
| 2243 |
+
"eval_steps_per_second": 2.486,
|
| 2244 |
"step": 1490
|
| 2245 |
},
|
| 2246 |
{
|
| 2247 |
"epoch": 1.5906680805938493,
|
| 2248 |
+
"grad_norm": 1.0428261756896973,
|
| 2249 |
"learning_rate": 2.3488865323435844e-05,
|
| 2250 |
+
"loss": 0.3569,
|
| 2251 |
"step": 1500
|
| 2252 |
},
|
| 2253 |
{
|
| 2254 |
"epoch": 1.5906680805938493,
|
| 2255 |
+
"eval_loss": 0.5321253538131714,
|
| 2256 |
+
"eval_runtime": 12.074,
|
| 2257 |
+
"eval_samples_per_second": 39.755,
|
| 2258 |
+
"eval_steps_per_second": 2.485,
|
| 2259 |
"step": 1500
|
| 2260 |
},
|
| 2261 |
{
|
| 2262 |
"epoch": 1.6012725344644752,
|
| 2263 |
+
"grad_norm": 1.0427707433700562,
|
| 2264 |
"learning_rate": 2.3312124425592083e-05,
|
| 2265 |
+
"loss": 0.355,
|
| 2266 |
+
"step": 1510
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 1.6012725344644752,
|
| 2270 |
+
"eval_loss": 0.5326208472251892,
|
| 2271 |
+
"eval_runtime": 12.0649,
|
| 2272 |
+
"eval_samples_per_second": 39.785,
|
| 2273 |
+
"eval_steps_per_second": 2.487,
|
| 2274 |
"step": 1510
|
| 2275 |
},
|
| 2276 |
{
|
| 2277 |
"epoch": 1.6118769883351007,
|
| 2278 |
+
"grad_norm": 0.9864311814308167,
|
| 2279 |
"learning_rate": 2.3135383527748322e-05,
|
| 2280 |
+
"loss": 0.3468,
|
| 2281 |
+
"step": 1520
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 1.6118769883351007,
|
| 2285 |
+
"eval_loss": 0.5332564115524292,
|
| 2286 |
+
"eval_runtime": 12.062,
|
| 2287 |
+
"eval_samples_per_second": 39.794,
|
| 2288 |
+
"eval_steps_per_second": 2.487,
|
| 2289 |
"step": 1520
|
| 2290 |
},
|
| 2291 |
{
|
| 2292 |
"epoch": 1.6224814422057263,
|
| 2293 |
+
"grad_norm": 1.1604270935058594,
|
| 2294 |
"learning_rate": 2.295864262990456e-05,
|
| 2295 |
"loss": 0.3522,
|
| 2296 |
"step": 1530
|
| 2297 |
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 1.6224814422057263,
|
| 2300 |
+
"eval_loss": 0.5316776633262634,
|
| 2301 |
+
"eval_runtime": 12.0976,
|
| 2302 |
+
"eval_samples_per_second": 39.677,
|
| 2303 |
+
"eval_steps_per_second": 2.48,
|
| 2304 |
+
"step": 1530
|
| 2305 |
+
},
|
| 2306 |
{
|
| 2307 |
"epoch": 1.6330858960763521,
|
| 2308 |
+
"grad_norm": 1.1040781736373901,
|
| 2309 |
"learning_rate": 2.27819017320608e-05,
|
| 2310 |
+
"loss": 0.3737,
|
| 2311 |
+
"step": 1540
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 1.6330858960763521,
|
| 2315 |
+
"eval_loss": 0.5313908457756042,
|
| 2316 |
+
"eval_runtime": 12.0607,
|
| 2317 |
+
"eval_samples_per_second": 39.799,
|
| 2318 |
+
"eval_steps_per_second": 2.487,
|
| 2319 |
"step": 1540
|
| 2320 |
},
|
| 2321 |
{
|
| 2322 |
"epoch": 1.6436903499469777,
|
| 2323 |
+
"grad_norm": 0.9892606735229492,
|
| 2324 |
"learning_rate": 2.260516083421704e-05,
|
| 2325 |
"loss": 0.3836,
|
| 2326 |
"step": 1550
|
| 2327 |
},
|
| 2328 |
{
|
| 2329 |
"epoch": 1.6436903499469777,
|
| 2330 |
+
"eval_loss": 0.53000807762146,
|
| 2331 |
+
"eval_runtime": 12.0736,
|
| 2332 |
+
"eval_samples_per_second": 39.756,
|
| 2333 |
+
"eval_steps_per_second": 2.485,
|
| 2334 |
"step": 1550
|
| 2335 |
},
|
| 2336 |
{
|
| 2337 |
"epoch": 1.6542948038176033,
|
| 2338 |
+
"grad_norm": 1.1608302593231201,
|
| 2339 |
"learning_rate": 2.2428419936373275e-05,
|
| 2340 |
+
"loss": 0.392,
|
| 2341 |
+
"step": 1560
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.6542948038176033,
|
| 2345 |
+
"eval_loss": 0.5281988978385925,
|
| 2346 |
+
"eval_runtime": 12.0895,
|
| 2347 |
+
"eval_samples_per_second": 39.704,
|
| 2348 |
+
"eval_steps_per_second": 2.481,
|
| 2349 |
"step": 1560
|
| 2350 |
},
|
| 2351 |
{
|
| 2352 |
"epoch": 1.664899257688229,
|
| 2353 |
+
"grad_norm": 0.9343358278274536,
|
| 2354 |
"learning_rate": 2.2251679038529514e-05,
|
| 2355 |
+
"loss": 0.3548,
|
| 2356 |
+
"step": 1570
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 1.664899257688229,
|
| 2360 |
+
"eval_loss": 0.5271891355514526,
|
| 2361 |
+
"eval_runtime": 12.0753,
|
| 2362 |
+
"eval_samples_per_second": 39.751,
|
| 2363 |
+
"eval_steps_per_second": 2.484,
|
| 2364 |
"step": 1570
|
| 2365 |
},
|
| 2366 |
{
|
| 2367 |
"epoch": 1.6755037115588547,
|
| 2368 |
+
"grad_norm": 1.0240166187286377,
|
| 2369 |
"learning_rate": 2.2074938140685754e-05,
|
| 2370 |
+
"loss": 0.3659,
|
| 2371 |
+
"step": 1580
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 1.6755037115588547,
|
| 2375 |
+
"eval_loss": 0.5283417105674744,
|
| 2376 |
+
"eval_runtime": 12.0862,
|
| 2377 |
+
"eval_samples_per_second": 39.715,
|
| 2378 |
+
"eval_steps_per_second": 2.482,
|
| 2379 |
"step": 1580
|
| 2380 |
},
|
| 2381 |
{
|
| 2382 |
"epoch": 1.6861081654294803,
|
| 2383 |
+
"grad_norm": 1.029309868812561,
|
| 2384 |
"learning_rate": 2.1898197242841996e-05,
|
| 2385 |
"loss": 0.3768,
|
| 2386 |
"step": 1590
|
| 2387 |
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 1.6861081654294803,
|
| 2390 |
+
"eval_loss": 0.5291869640350342,
|
| 2391 |
+
"eval_runtime": 12.0977,
|
| 2392 |
+
"eval_samples_per_second": 39.677,
|
| 2393 |
+
"eval_steps_per_second": 2.48,
|
| 2394 |
+
"step": 1590
|
| 2395 |
+
},
|
| 2396 |
{
|
| 2397 |
"epoch": 1.696712619300106,
|
| 2398 |
+
"grad_norm": 1.0808477401733398,
|
| 2399 |
"learning_rate": 2.1721456344998235e-05,
|
| 2400 |
+
"loss": 0.3631,
|
| 2401 |
"step": 1600
|
| 2402 |
},
|
| 2403 |
{
|
| 2404 |
"epoch": 1.696712619300106,
|
| 2405 |
+
"eval_loss": 0.5284417867660522,
|
| 2406 |
+
"eval_runtime": 12.0899,
|
| 2407 |
+
"eval_samples_per_second": 39.703,
|
| 2408 |
+
"eval_steps_per_second": 2.481,
|
| 2409 |
"step": 1600
|
| 2410 |
},
|
| 2411 |
{
|
| 2412 |
"epoch": 1.7073170731707317,
|
| 2413 |
+
"grad_norm": 1.0758904218673706,
|
| 2414 |
"learning_rate": 2.1544715447154475e-05,
|
| 2415 |
+
"loss": 0.3749,
|
| 2416 |
+
"step": 1610
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 1.7073170731707317,
|
| 2420 |
+
"eval_loss": 0.5268154144287109,
|
| 2421 |
+
"eval_runtime": 12.0941,
|
| 2422 |
+
"eval_samples_per_second": 39.689,
|
| 2423 |
+
"eval_steps_per_second": 2.481,
|
| 2424 |
"step": 1610
|
| 2425 |
},
|
| 2426 |
{
|
| 2427 |
"epoch": 1.7179215270413573,
|
| 2428 |
+
"grad_norm": 1.0460011959075928,
|
| 2429 |
"learning_rate": 2.1367974549310714e-05,
|
| 2430 |
+
"loss": 0.3448,
|
| 2431 |
+
"step": 1620
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 1.7179215270413573,
|
| 2435 |
+
"eval_loss": 0.5249053239822388,
|
| 2436 |
+
"eval_runtime": 12.0931,
|
| 2437 |
+
"eval_samples_per_second": 39.692,
|
| 2438 |
+
"eval_steps_per_second": 2.481,
|
| 2439 |
"step": 1620
|
| 2440 |
},
|
| 2441 |
{
|
| 2442 |
"epoch": 1.728525980911983,
|
| 2443 |
+
"grad_norm": 1.2108927965164185,
|
| 2444 |
"learning_rate": 2.1191233651466953e-05,
|
| 2445 |
+
"loss": 0.3578,
|
| 2446 |
+
"step": 1630
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.728525980911983,
|
| 2450 |
+
"eval_loss": 0.5253214836120605,
|
| 2451 |
+
"eval_runtime": 12.0927,
|
| 2452 |
+
"eval_samples_per_second": 39.693,
|
| 2453 |
+
"eval_steps_per_second": 2.481,
|
| 2454 |
"step": 1630
|
| 2455 |
},
|
| 2456 |
{
|
| 2457 |
"epoch": 1.7391304347826086,
|
| 2458 |
+
"grad_norm": 1.1448472738265991,
|
| 2459 |
"learning_rate": 2.101449275362319e-05,
|
| 2460 |
"loss": 0.3966,
|
| 2461 |
"step": 1640
|
| 2462 |
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 1.7391304347826086,
|
| 2465 |
+
"eval_loss": 0.5242108106613159,
|
| 2466 |
+
"eval_runtime": 12.1019,
|
| 2467 |
+
"eval_samples_per_second": 39.663,
|
| 2468 |
+
"eval_steps_per_second": 2.479,
|
| 2469 |
+
"step": 1640
|
| 2470 |
+
},
|
| 2471 |
{
|
| 2472 |
"epoch": 1.7497348886532342,
|
| 2473 |
+
"grad_norm": 1.083829402923584,
|
| 2474 |
"learning_rate": 2.0837751855779428e-05,
|
| 2475 |
+
"loss": 0.3509,
|
| 2476 |
"step": 1650
|
| 2477 |
},
|
| 2478 |
{
|
| 2479 |
"epoch": 1.7497348886532342,
|
| 2480 |
+
"eval_loss": 0.52347731590271,
|
| 2481 |
+
"eval_runtime": 12.0917,
|
| 2482 |
+
"eval_samples_per_second": 39.697,
|
| 2483 |
+
"eval_steps_per_second": 2.481,
|
| 2484 |
"step": 1650
|
| 2485 |
},
|
| 2486 |
{
|
| 2487 |
"epoch": 1.76033934252386,
|
| 2488 |
+
"grad_norm": 1.0172953605651855,
|
| 2489 |
"learning_rate": 2.0661010957935667e-05,
|
| 2490 |
+
"loss": 0.3626,
|
| 2491 |
+
"step": 1660
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 1.76033934252386,
|
| 2495 |
+
"eval_loss": 0.5247266292572021,
|
| 2496 |
+
"eval_runtime": 12.0939,
|
| 2497 |
+
"eval_samples_per_second": 39.69,
|
| 2498 |
+
"eval_steps_per_second": 2.481,
|
| 2499 |
"step": 1660
|
| 2500 |
},
|
| 2501 |
{
|
| 2502 |
"epoch": 1.7709437963944858,
|
| 2503 |
+
"grad_norm": 0.9958562254905701,
|
| 2504 |
"learning_rate": 2.0484270060091906e-05,
|
| 2505 |
+
"loss": 0.3719,
|
| 2506 |
+
"step": 1670
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 1.7709437963944858,
|
| 2510 |
+
"eval_loss": 0.5264424085617065,
|
| 2511 |
+
"eval_runtime": 12.0974,
|
| 2512 |
+
"eval_samples_per_second": 39.678,
|
| 2513 |
+
"eval_steps_per_second": 2.48,
|
| 2514 |
"step": 1670
|
| 2515 |
},
|
| 2516 |
{
|
| 2517 |
"epoch": 1.7815482502651112,
|
| 2518 |
+
"grad_norm": 1.1131070852279663,
|
| 2519 |
"learning_rate": 2.0307529162248146e-05,
|
| 2520 |
+
"loss": 0.3708,
|
| 2521 |
+
"step": 1680
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 1.7815482502651112,
|
| 2525 |
+
"eval_loss": 0.5272196531295776,
|
| 2526 |
+
"eval_runtime": 12.0832,
|
| 2527 |
+
"eval_samples_per_second": 39.725,
|
| 2528 |
+
"eval_steps_per_second": 2.483,
|
| 2529 |
"step": 1680
|
| 2530 |
},
|
| 2531 |
{
|
| 2532 |
"epoch": 1.792152704135737,
|
| 2533 |
+
"grad_norm": 1.0758273601531982,
|
| 2534 |
"learning_rate": 2.0130788264404385e-05,
|
| 2535 |
+
"loss": 0.3751,
|
| 2536 |
+
"step": 1690
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 1.792152704135737,
|
| 2540 |
+
"eval_loss": 0.5256540179252625,
|
| 2541 |
+
"eval_runtime": 12.082,
|
| 2542 |
+
"eval_samples_per_second": 39.729,
|
| 2543 |
+
"eval_steps_per_second": 2.483,
|
| 2544 |
"step": 1690
|
| 2545 |
},
|
| 2546 |
{
|
| 2547 |
"epoch": 1.8027571580063628,
|
| 2548 |
+
"grad_norm": 1.0669254064559937,
|
| 2549 |
"learning_rate": 1.9954047366560624e-05,
|
| 2550 |
+
"loss": 0.3563,
|
| 2551 |
"step": 1700
|
| 2552 |
},
|
| 2553 |
{
|
| 2554 |
"epoch": 1.8027571580063628,
|
| 2555 |
+
"eval_loss": 0.525856614112854,
|
| 2556 |
+
"eval_runtime": 12.0902,
|
| 2557 |
+
"eval_samples_per_second": 39.702,
|
| 2558 |
+
"eval_steps_per_second": 2.481,
|
| 2559 |
"step": 1700
|
| 2560 |
}
|
| 2561 |
],
|
|
|
|
| 2563 |
"max_steps": 2829,
|
| 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:faea3967f69a90637927708e340ebfb2d331bed5de86d878fcddb572ecb69193
|
| 3 |
size 5880
|