Training checkpoint at step 10000
Browse files- trainer_state.json +365 -5
trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 2.
|
| 4 |
"best_model_checkpoint": "../runs/karpathy/fineweb-edu-100b-shuffle/meta-llama/Llama-3.2-1B/linear_adamw_wd1e-03_8x1024_mem32_bs64_hf_armt_dmem64/run_21/checkpoint-9000",
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -3248,6 +3248,366 @@
|
|
| 3248 |
"eval_samples_per_second": 3.168,
|
| 3249 |
"eval_steps_per_second": 1.584,
|
| 3250 |
"step": 9000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3251 |
}
|
| 3252 |
],
|
| 3253 |
"logging_steps": 25,
|
|
@@ -3267,7 +3627,7 @@
|
|
| 3267 |
"attributes": {}
|
| 3268 |
}
|
| 3269 |
},
|
| 3270 |
-
"total_flos":
|
| 3271 |
"train_batch_size": 1,
|
| 3272 |
"trial_name": null,
|
| 3273 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 9800,
|
| 3 |
+
"best_metric": 2.4076178073883057,
|
| 4 |
"best_model_checkpoint": "../runs/karpathy/fineweb-edu-100b-shuffle/meta-llama/Llama-3.2-1B/linear_adamw_wd1e-03_8x1024_mem32_bs64_hf_armt_dmem64/run_21/checkpoint-9000",
|
| 5 |
+
"epoch": 0.2,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 10000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3248 |
"eval_samples_per_second": 3.168,
|
| 3249 |
"eval_steps_per_second": 1.584,
|
| 3250 |
"step": 9000
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 0.1805,
|
| 3254 |
+
"grad_norm": 0.5908878679870805,
|
| 3255 |
+
"learning_rate": 9.105777777777779e-06,
|
| 3256 |
+
"loss": 2.3938,
|
| 3257 |
+
"step": 9025
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 0.181,
|
| 3261 |
+
"grad_norm": 0.5496267273049,
|
| 3262 |
+
"learning_rate": 9.100222222222223e-06,
|
| 3263 |
+
"loss": 2.3961,
|
| 3264 |
+
"step": 9050
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 0.1815,
|
| 3268 |
+
"grad_norm": 0.5979695738071065,
|
| 3269 |
+
"learning_rate": 9.094666666666668e-06,
|
| 3270 |
+
"loss": 2.3858,
|
| 3271 |
+
"step": 9075
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 0.182,
|
| 3275 |
+
"grad_norm": 0.5938166893318079,
|
| 3276 |
+
"learning_rate": 9.089111111111111e-06,
|
| 3277 |
+
"loss": 2.3862,
|
| 3278 |
+
"step": 9100
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 0.182,
|
| 3282 |
+
"eval_loss": 2.410053253173828,
|
| 3283 |
+
"eval_runtime": 32.2577,
|
| 3284 |
+
"eval_samples_per_second": 3.162,
|
| 3285 |
+
"eval_steps_per_second": 1.581,
|
| 3286 |
+
"step": 9100
|
| 3287 |
+
},
|
| 3288 |
+
{
|
| 3289 |
+
"epoch": 0.1825,
|
| 3290 |
+
"grad_norm": 0.5958942390294701,
|
| 3291 |
+
"learning_rate": 9.083555555555557e-06,
|
| 3292 |
+
"loss": 2.3928,
|
| 3293 |
+
"step": 9125
|
| 3294 |
+
},
|
| 3295 |
+
{
|
| 3296 |
+
"epoch": 0.183,
|
| 3297 |
+
"grad_norm": 0.5859164810125311,
|
| 3298 |
+
"learning_rate": 9.078000000000002e-06,
|
| 3299 |
+
"loss": 2.4022,
|
| 3300 |
+
"step": 9150
|
| 3301 |
+
},
|
| 3302 |
+
{
|
| 3303 |
+
"epoch": 0.1835,
|
| 3304 |
+
"grad_norm": 0.5798241289951321,
|
| 3305 |
+
"learning_rate": 9.072444444444445e-06,
|
| 3306 |
+
"loss": 2.3928,
|
| 3307 |
+
"step": 9175
|
| 3308 |
+
},
|
| 3309 |
+
{
|
| 3310 |
+
"epoch": 0.184,
|
| 3311 |
+
"grad_norm": 0.5882407091400851,
|
| 3312 |
+
"learning_rate": 9.066888888888889e-06,
|
| 3313 |
+
"loss": 2.3973,
|
| 3314 |
+
"step": 9200
|
| 3315 |
+
},
|
| 3316 |
+
{
|
| 3317 |
+
"epoch": 0.184,
|
| 3318 |
+
"eval_loss": 2.409634590148926,
|
| 3319 |
+
"eval_runtime": 32.249,
|
| 3320 |
+
"eval_samples_per_second": 3.163,
|
| 3321 |
+
"eval_steps_per_second": 1.581,
|
| 3322 |
+
"step": 9200
|
| 3323 |
+
},
|
| 3324 |
+
{
|
| 3325 |
+
"epoch": 0.1845,
|
| 3326 |
+
"grad_norm": 0.5903772748051019,
|
| 3327 |
+
"learning_rate": 9.061333333333334e-06,
|
| 3328 |
+
"loss": 2.3831,
|
| 3329 |
+
"step": 9225
|
| 3330 |
+
},
|
| 3331 |
+
{
|
| 3332 |
+
"epoch": 0.185,
|
| 3333 |
+
"grad_norm": 0.6211646089814673,
|
| 3334 |
+
"learning_rate": 9.05577777777778e-06,
|
| 3335 |
+
"loss": 2.3983,
|
| 3336 |
+
"step": 9250
|
| 3337 |
+
},
|
| 3338 |
+
{
|
| 3339 |
+
"epoch": 0.1855,
|
| 3340 |
+
"grad_norm": 0.6172378815389531,
|
| 3341 |
+
"learning_rate": 9.050222222222223e-06,
|
| 3342 |
+
"loss": 2.3961,
|
| 3343 |
+
"step": 9275
|
| 3344 |
+
},
|
| 3345 |
+
{
|
| 3346 |
+
"epoch": 0.186,
|
| 3347 |
+
"grad_norm": 0.6117693503941964,
|
| 3348 |
+
"learning_rate": 9.044666666666667e-06,
|
| 3349 |
+
"loss": 2.3991,
|
| 3350 |
+
"step": 9300
|
| 3351 |
+
},
|
| 3352 |
+
{
|
| 3353 |
+
"epoch": 0.186,
|
| 3354 |
+
"eval_loss": 2.4100780487060547,
|
| 3355 |
+
"eval_runtime": 31.6698,
|
| 3356 |
+
"eval_samples_per_second": 3.221,
|
| 3357 |
+
"eval_steps_per_second": 1.61,
|
| 3358 |
+
"step": 9300
|
| 3359 |
+
},
|
| 3360 |
+
{
|
| 3361 |
+
"epoch": 0.1865,
|
| 3362 |
+
"grad_norm": 0.5955035334939845,
|
| 3363 |
+
"learning_rate": 9.039111111111112e-06,
|
| 3364 |
+
"loss": 2.4013,
|
| 3365 |
+
"step": 9325
|
| 3366 |
+
},
|
| 3367 |
+
{
|
| 3368 |
+
"epoch": 0.187,
|
| 3369 |
+
"grad_norm": 0.6304889803867978,
|
| 3370 |
+
"learning_rate": 9.033555555555557e-06,
|
| 3371 |
+
"loss": 2.4045,
|
| 3372 |
+
"step": 9350
|
| 3373 |
+
},
|
| 3374 |
+
{
|
| 3375 |
+
"epoch": 0.1875,
|
| 3376 |
+
"grad_norm": 0.5650857479280212,
|
| 3377 |
+
"learning_rate": 9.028e-06,
|
| 3378 |
+
"loss": 2.3993,
|
| 3379 |
+
"step": 9375
|
| 3380 |
+
},
|
| 3381 |
+
{
|
| 3382 |
+
"epoch": 0.188,
|
| 3383 |
+
"grad_norm": 0.6102368092141387,
|
| 3384 |
+
"learning_rate": 9.022444444444444e-06,
|
| 3385 |
+
"loss": 2.3969,
|
| 3386 |
+
"step": 9400
|
| 3387 |
+
},
|
| 3388 |
+
{
|
| 3389 |
+
"epoch": 0.188,
|
| 3390 |
+
"eval_loss": 2.4091312885284424,
|
| 3391 |
+
"eval_runtime": 31.7427,
|
| 3392 |
+
"eval_samples_per_second": 3.213,
|
| 3393 |
+
"eval_steps_per_second": 1.607,
|
| 3394 |
+
"step": 9400
|
| 3395 |
+
},
|
| 3396 |
+
{
|
| 3397 |
+
"epoch": 0.1885,
|
| 3398 |
+
"grad_norm": 0.5647006274355659,
|
| 3399 |
+
"learning_rate": 9.01688888888889e-06,
|
| 3400 |
+
"loss": 2.3962,
|
| 3401 |
+
"step": 9425
|
| 3402 |
+
},
|
| 3403 |
+
{
|
| 3404 |
+
"epoch": 0.189,
|
| 3405 |
+
"grad_norm": 0.639478683787589,
|
| 3406 |
+
"learning_rate": 9.011333333333335e-06,
|
| 3407 |
+
"loss": 2.3957,
|
| 3408 |
+
"step": 9450
|
| 3409 |
+
},
|
| 3410 |
+
{
|
| 3411 |
+
"epoch": 0.1895,
|
| 3412 |
+
"grad_norm": 0.5788568545073746,
|
| 3413 |
+
"learning_rate": 9.005777777777778e-06,
|
| 3414 |
+
"loss": 2.3914,
|
| 3415 |
+
"step": 9475
|
| 3416 |
+
},
|
| 3417 |
+
{
|
| 3418 |
+
"epoch": 0.19,
|
| 3419 |
+
"grad_norm": 0.7290164754099147,
|
| 3420 |
+
"learning_rate": 9.000222222222222e-06,
|
| 3421 |
+
"loss": 2.386,
|
| 3422 |
+
"step": 9500
|
| 3423 |
+
},
|
| 3424 |
+
{
|
| 3425 |
+
"epoch": 0.19,
|
| 3426 |
+
"eval_loss": 2.4086694717407227,
|
| 3427 |
+
"eval_runtime": 31.8061,
|
| 3428 |
+
"eval_samples_per_second": 3.207,
|
| 3429 |
+
"eval_steps_per_second": 1.603,
|
| 3430 |
+
"step": 9500
|
| 3431 |
+
},
|
| 3432 |
+
{
|
| 3433 |
+
"epoch": 0.1905,
|
| 3434 |
+
"grad_norm": 0.5817637514180484,
|
| 3435 |
+
"learning_rate": 8.994666666666667e-06,
|
| 3436 |
+
"loss": 2.4006,
|
| 3437 |
+
"step": 9525
|
| 3438 |
+
},
|
| 3439 |
+
{
|
| 3440 |
+
"epoch": 0.191,
|
| 3441 |
+
"grad_norm": 0.5697879107784812,
|
| 3442 |
+
"learning_rate": 8.989111111111112e-06,
|
| 3443 |
+
"loss": 2.3899,
|
| 3444 |
+
"step": 9550
|
| 3445 |
+
},
|
| 3446 |
+
{
|
| 3447 |
+
"epoch": 0.1915,
|
| 3448 |
+
"grad_norm": 0.584610269954786,
|
| 3449 |
+
"learning_rate": 8.983555555555556e-06,
|
| 3450 |
+
"loss": 2.3944,
|
| 3451 |
+
"step": 9575
|
| 3452 |
+
},
|
| 3453 |
+
{
|
| 3454 |
+
"epoch": 0.192,
|
| 3455 |
+
"grad_norm": 0.608795413325502,
|
| 3456 |
+
"learning_rate": 8.978000000000001e-06,
|
| 3457 |
+
"loss": 2.398,
|
| 3458 |
+
"step": 9600
|
| 3459 |
+
},
|
| 3460 |
+
{
|
| 3461 |
+
"epoch": 0.192,
|
| 3462 |
+
"eval_loss": 2.408263683319092,
|
| 3463 |
+
"eval_runtime": 31.6859,
|
| 3464 |
+
"eval_samples_per_second": 3.219,
|
| 3465 |
+
"eval_steps_per_second": 1.61,
|
| 3466 |
+
"step": 9600
|
| 3467 |
+
},
|
| 3468 |
+
{
|
| 3469 |
+
"epoch": 0.1925,
|
| 3470 |
+
"grad_norm": 0.5915130204472873,
|
| 3471 |
+
"learning_rate": 8.972444444444445e-06,
|
| 3472 |
+
"loss": 2.407,
|
| 3473 |
+
"step": 9625
|
| 3474 |
+
},
|
| 3475 |
+
{
|
| 3476 |
+
"epoch": 0.193,
|
| 3477 |
+
"grad_norm": 0.59521034646126,
|
| 3478 |
+
"learning_rate": 8.96688888888889e-06,
|
| 3479 |
+
"loss": 2.3924,
|
| 3480 |
+
"step": 9650
|
| 3481 |
+
},
|
| 3482 |
+
{
|
| 3483 |
+
"epoch": 0.1935,
|
| 3484 |
+
"grad_norm": 0.6050238690396914,
|
| 3485 |
+
"learning_rate": 8.961333333333333e-06,
|
| 3486 |
+
"loss": 2.3869,
|
| 3487 |
+
"step": 9675
|
| 3488 |
+
},
|
| 3489 |
+
{
|
| 3490 |
+
"epoch": 0.194,
|
| 3491 |
+
"grad_norm": 0.5691067223521449,
|
| 3492 |
+
"learning_rate": 8.955777777777779e-06,
|
| 3493 |
+
"loss": 2.3874,
|
| 3494 |
+
"step": 9700
|
| 3495 |
+
},
|
| 3496 |
+
{
|
| 3497 |
+
"epoch": 0.194,
|
| 3498 |
+
"eval_loss": 2.408264398574829,
|
| 3499 |
+
"eval_runtime": 31.8579,
|
| 3500 |
+
"eval_samples_per_second": 3.202,
|
| 3501 |
+
"eval_steps_per_second": 1.601,
|
| 3502 |
+
"step": 9700
|
| 3503 |
+
},
|
| 3504 |
+
{
|
| 3505 |
+
"epoch": 0.1945,
|
| 3506 |
+
"grad_norm": 0.5753054034666798,
|
| 3507 |
+
"learning_rate": 8.950222222222224e-06,
|
| 3508 |
+
"loss": 2.4027,
|
| 3509 |
+
"step": 9725
|
| 3510 |
+
},
|
| 3511 |
+
{
|
| 3512 |
+
"epoch": 0.195,
|
| 3513 |
+
"grad_norm": 0.5864767839913545,
|
| 3514 |
+
"learning_rate": 8.944666666666668e-06,
|
| 3515 |
+
"loss": 2.3924,
|
| 3516 |
+
"step": 9750
|
| 3517 |
+
},
|
| 3518 |
+
{
|
| 3519 |
+
"epoch": 0.1955,
|
| 3520 |
+
"grad_norm": 0.6642807256080032,
|
| 3521 |
+
"learning_rate": 8.939111111111111e-06,
|
| 3522 |
+
"loss": 2.3709,
|
| 3523 |
+
"step": 9775
|
| 3524 |
+
},
|
| 3525 |
+
{
|
| 3526 |
+
"epoch": 0.196,
|
| 3527 |
+
"grad_norm": 0.6084139101409156,
|
| 3528 |
+
"learning_rate": 8.933555555555556e-06,
|
| 3529 |
+
"loss": 2.3958,
|
| 3530 |
+
"step": 9800
|
| 3531 |
+
},
|
| 3532 |
+
{
|
| 3533 |
+
"epoch": 0.196,
|
| 3534 |
+
"eval_loss": 2.4076178073883057,
|
| 3535 |
+
"eval_runtime": 31.7733,
|
| 3536 |
+
"eval_samples_per_second": 3.21,
|
| 3537 |
+
"eval_steps_per_second": 1.605,
|
| 3538 |
+
"step": 9800
|
| 3539 |
+
},
|
| 3540 |
+
{
|
| 3541 |
+
"epoch": 0.1965,
|
| 3542 |
+
"grad_norm": 0.592206064244208,
|
| 3543 |
+
"learning_rate": 8.928000000000002e-06,
|
| 3544 |
+
"loss": 2.3922,
|
| 3545 |
+
"step": 9825
|
| 3546 |
+
},
|
| 3547 |
+
{
|
| 3548 |
+
"epoch": 0.197,
|
| 3549 |
+
"grad_norm": 0.5685236067589632,
|
| 3550 |
+
"learning_rate": 8.922444444444445e-06,
|
| 3551 |
+
"loss": 2.3908,
|
| 3552 |
+
"step": 9850
|
| 3553 |
+
},
|
| 3554 |
+
{
|
| 3555 |
+
"epoch": 0.1975,
|
| 3556 |
+
"grad_norm": 0.6034821273699428,
|
| 3557 |
+
"learning_rate": 8.916888888888889e-06,
|
| 3558 |
+
"loss": 2.3903,
|
| 3559 |
+
"step": 9875
|
| 3560 |
+
},
|
| 3561 |
+
{
|
| 3562 |
+
"epoch": 0.198,
|
| 3563 |
+
"grad_norm": 0.5910198540350765,
|
| 3564 |
+
"learning_rate": 8.911333333333334e-06,
|
| 3565 |
+
"loss": 2.3767,
|
| 3566 |
+
"step": 9900
|
| 3567 |
+
},
|
| 3568 |
+
{
|
| 3569 |
+
"epoch": 0.198,
|
| 3570 |
+
"eval_loss": 2.407928705215454,
|
| 3571 |
+
"eval_runtime": 31.7033,
|
| 3572 |
+
"eval_samples_per_second": 3.217,
|
| 3573 |
+
"eval_steps_per_second": 1.609,
|
| 3574 |
+
"step": 9900
|
| 3575 |
+
},
|
| 3576 |
+
{
|
| 3577 |
+
"epoch": 0.1985,
|
| 3578 |
+
"grad_norm": 0.5849079897115376,
|
| 3579 |
+
"learning_rate": 8.90577777777778e-06,
|
| 3580 |
+
"loss": 2.3956,
|
| 3581 |
+
"step": 9925
|
| 3582 |
+
},
|
| 3583 |
+
{
|
| 3584 |
+
"epoch": 0.199,
|
| 3585 |
+
"grad_norm": 0.5683901924605945,
|
| 3586 |
+
"learning_rate": 8.900222222222223e-06,
|
| 3587 |
+
"loss": 2.3884,
|
| 3588 |
+
"step": 9950
|
| 3589 |
+
},
|
| 3590 |
+
{
|
| 3591 |
+
"epoch": 0.1995,
|
| 3592 |
+
"grad_norm": 0.6037241225699064,
|
| 3593 |
+
"learning_rate": 8.894666666666666e-06,
|
| 3594 |
+
"loss": 2.3934,
|
| 3595 |
+
"step": 9975
|
| 3596 |
+
},
|
| 3597 |
+
{
|
| 3598 |
+
"epoch": 0.2,
|
| 3599 |
+
"grad_norm": 0.5807810374364664,
|
| 3600 |
+
"learning_rate": 8.889111111111112e-06,
|
| 3601 |
+
"loss": 2.3999,
|
| 3602 |
+
"step": 10000
|
| 3603 |
+
},
|
| 3604 |
+
{
|
| 3605 |
+
"epoch": 0.2,
|
| 3606 |
+
"eval_loss": 2.40779447555542,
|
| 3607 |
+
"eval_runtime": 31.7288,
|
| 3608 |
+
"eval_samples_per_second": 3.215,
|
| 3609 |
+
"eval_steps_per_second": 1.607,
|
| 3610 |
+
"step": 10000
|
| 3611 |
}
|
| 3612 |
],
|
| 3613 |
"logging_steps": 25,
|
|
|
|
| 3627 |
"attributes": {}
|
| 3628 |
}
|
| 3629 |
},
|
| 3630 |
+
"total_flos": 3.183202298327204e+19,
|
| 3631 |
"train_batch_size": 1,
|
| 3632 |
"trial_name": null,
|
| 3633 |
"trial_params": null
|