{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.011894230554791517,
"eval_steps": 13,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00023788461109583033,
"grad_norm": 11.802818298339844,
"learning_rate": 5e-05,
"loss": 4.7672,
"step": 1
},
{
"epoch": 0.00023788461109583033,
"eval_loss": 5.097123622894287,
"eval_runtime": 710.4802,
"eval_samples_per_second": 39.86,
"eval_steps_per_second": 9.965,
"step": 1
},
{
"epoch": 0.00047576922219166065,
"grad_norm": 11.809027671813965,
"learning_rate": 0.0001,
"loss": 5.0036,
"step": 2
},
{
"epoch": 0.000713653833287491,
"grad_norm": 11.069571495056152,
"learning_rate": 9.989294616193017e-05,
"loss": 4.856,
"step": 3
},
{
"epoch": 0.0009515384443833213,
"grad_norm": 9.365679740905762,
"learning_rate": 9.957224306869053e-05,
"loss": 3.8019,
"step": 4
},
{
"epoch": 0.0011894230554791517,
"grad_norm": 7.9849162101745605,
"learning_rate": 9.903926402016153e-05,
"loss": 3.2773,
"step": 5
},
{
"epoch": 0.001427307666574982,
"grad_norm": 7.107789039611816,
"learning_rate": 9.829629131445342e-05,
"loss": 2.4892,
"step": 6
},
{
"epoch": 0.0016651922776708124,
"grad_norm": 8.650176048278809,
"learning_rate": 9.73465064747553e-05,
"loss": 1.9592,
"step": 7
},
{
"epoch": 0.0019030768887666426,
"grad_norm": 9.36210823059082,
"learning_rate": 9.619397662556435e-05,
"loss": 1.5217,
"step": 8
},
{
"epoch": 0.002140961499862473,
"grad_norm": 6.943857192993164,
"learning_rate": 9.484363707663442e-05,
"loss": 0.9687,
"step": 9
},
{
"epoch": 0.0023788461109583033,
"grad_norm": 5.760282039642334,
"learning_rate": 9.330127018922194e-05,
"loss": 0.5684,
"step": 10
},
{
"epoch": 0.0026167307220541338,
"grad_norm": 5.303816318511963,
"learning_rate": 9.157348061512727e-05,
"loss": 0.4237,
"step": 11
},
{
"epoch": 0.002854615333149964,
"grad_norm": 2.8232803344726562,
"learning_rate": 8.966766701456177e-05,
"loss": 0.1988,
"step": 12
},
{
"epoch": 0.0030924999442457943,
"grad_norm": 5.23857307434082,
"learning_rate": 8.759199037394887e-05,
"loss": 0.1468,
"step": 13
},
{
"epoch": 0.0030924999442457943,
"eval_loss": 0.12124715000391006,
"eval_runtime": 711.0925,
"eval_samples_per_second": 39.826,
"eval_steps_per_second": 9.957,
"step": 13
},
{
"epoch": 0.0033303845553416247,
"grad_norm": 5.04780387878418,
"learning_rate": 8.535533905932738e-05,
"loss": 0.131,
"step": 14
},
{
"epoch": 0.0035682691664374547,
"grad_norm": 1.841731309890747,
"learning_rate": 8.296729075500344e-05,
"loss": 0.0827,
"step": 15
},
{
"epoch": 0.003806153777533285,
"grad_norm": 1.6053825616836548,
"learning_rate": 8.043807145043604e-05,
"loss": 0.0752,
"step": 16
},
{
"epoch": 0.004044038388629116,
"grad_norm": 1.511683702468872,
"learning_rate": 7.777851165098012e-05,
"loss": 0.0741,
"step": 17
},
{
"epoch": 0.004281922999724946,
"grad_norm": 5.841466903686523,
"learning_rate": 7.500000000000001e-05,
"loss": 0.0867,
"step": 18
},
{
"epoch": 0.004519807610820777,
"grad_norm": 5.441026210784912,
"learning_rate": 7.211443451095007e-05,
"loss": 0.0663,
"step": 19
},
{
"epoch": 0.004757692221916607,
"grad_norm": 1.9340760707855225,
"learning_rate": 6.91341716182545e-05,
"loss": 0.0498,
"step": 20
},
{
"epoch": 0.004995576833012437,
"grad_norm": 1.7755597829818726,
"learning_rate": 6.607197326515808e-05,
"loss": 0.0431,
"step": 21
},
{
"epoch": 0.0052334614441082675,
"grad_norm": 0.8990055918693542,
"learning_rate": 6.294095225512603e-05,
"loss": 0.0285,
"step": 22
},
{
"epoch": 0.0054713460552040976,
"grad_norm": 1.0627079010009766,
"learning_rate": 5.9754516100806423e-05,
"loss": 0.0427,
"step": 23
},
{
"epoch": 0.005709230666299928,
"grad_norm": 0.7508691549301147,
"learning_rate": 5.6526309611002594e-05,
"loss": 0.0114,
"step": 24
},
{
"epoch": 0.0059471152773957585,
"grad_norm": 0.787846028804779,
"learning_rate": 5.327015646150716e-05,
"loss": 0.0305,
"step": 25
},
{
"epoch": 0.0061849998884915885,
"grad_norm": 1.2530877590179443,
"learning_rate": 5e-05,
"loss": 0.0366,
"step": 26
},
{
"epoch": 0.0061849998884915885,
"eval_loss": 0.02799982577562332,
"eval_runtime": 711.2213,
"eval_samples_per_second": 39.819,
"eval_steps_per_second": 9.955,
"step": 26
},
{
"epoch": 0.0064228844995874185,
"grad_norm": 0.5535579323768616,
"learning_rate": 4.6729843538492847e-05,
"loss": 0.0174,
"step": 27
},
{
"epoch": 0.0066607691106832494,
"grad_norm": 0.6739106774330139,
"learning_rate": 4.347369038899744e-05,
"loss": 0.0377,
"step": 28
},
{
"epoch": 0.0068986537217790795,
"grad_norm": 0.37905821204185486,
"learning_rate": 4.0245483899193595e-05,
"loss": 0.0143,
"step": 29
},
{
"epoch": 0.0071365383328749095,
"grad_norm": 0.8597642779350281,
"learning_rate": 3.705904774487396e-05,
"loss": 0.0382,
"step": 30
},
{
"epoch": 0.00737442294397074,
"grad_norm": 0.3920349180698395,
"learning_rate": 3.392802673484193e-05,
"loss": 0.0144,
"step": 31
},
{
"epoch": 0.00761230755506657,
"grad_norm": 0.33046308159828186,
"learning_rate": 3.086582838174551e-05,
"loss": 0.0061,
"step": 32
},
{
"epoch": 0.0078501921661624,
"grad_norm": 0.44863665103912354,
"learning_rate": 2.7885565489049946e-05,
"loss": 0.0412,
"step": 33
},
{
"epoch": 0.008088076777258231,
"grad_norm": 0.2641896605491638,
"learning_rate": 2.500000000000001e-05,
"loss": 0.0157,
"step": 34
},
{
"epoch": 0.008325961388354062,
"grad_norm": 0.35212084650993347,
"learning_rate": 2.2221488349019903e-05,
"loss": 0.0171,
"step": 35
},
{
"epoch": 0.008563845999449891,
"grad_norm": 0.19091854989528656,
"learning_rate": 1.9561928549563968e-05,
"loss": 0.0145,
"step": 36
},
{
"epoch": 0.008801730610545722,
"grad_norm": 0.22365093231201172,
"learning_rate": 1.703270924499656e-05,
"loss": 0.0156,
"step": 37
},
{
"epoch": 0.009039615221641553,
"grad_norm": 0.5021849274635315,
"learning_rate": 1.4644660940672627e-05,
"loss": 0.0359,
"step": 38
},
{
"epoch": 0.009277499832737382,
"grad_norm": 0.8826900720596313,
"learning_rate": 1.2408009626051137e-05,
"loss": 0.0497,
"step": 39
},
{
"epoch": 0.009277499832737382,
"eval_loss": 0.02232297882437706,
"eval_runtime": 711.2232,
"eval_samples_per_second": 39.819,
"eval_steps_per_second": 9.955,
"step": 39
},
{
"epoch": 0.009515384443833213,
"grad_norm": 0.6363677382469177,
"learning_rate": 1.0332332985438248e-05,
"loss": 0.0181,
"step": 40
},
{
"epoch": 0.009753269054929044,
"grad_norm": 0.5468412637710571,
"learning_rate": 8.426519384872733e-06,
"loss": 0.0182,
"step": 41
},
{
"epoch": 0.009991153666024873,
"grad_norm": 0.6606681942939758,
"learning_rate": 6.698729810778065e-06,
"loss": 0.0205,
"step": 42
},
{
"epoch": 0.010229038277120704,
"grad_norm": 0.5074648261070251,
"learning_rate": 5.156362923365588e-06,
"loss": 0.0163,
"step": 43
},
{
"epoch": 0.010466922888216535,
"grad_norm": 0.27615487575531006,
"learning_rate": 3.8060233744356633e-06,
"loss": 0.0218,
"step": 44
},
{
"epoch": 0.010704807499312364,
"grad_norm": 0.4286714196205139,
"learning_rate": 2.653493525244721e-06,
"loss": 0.0221,
"step": 45
},
{
"epoch": 0.010942692110408195,
"grad_norm": 0.49691057205200195,
"learning_rate": 1.70370868554659e-06,
"loss": 0.0504,
"step": 46
},
{
"epoch": 0.011180576721504026,
"grad_norm": 0.4235456585884094,
"learning_rate": 9.607359798384785e-07,
"loss": 0.0221,
"step": 47
},
{
"epoch": 0.011418461332599855,
"grad_norm": 0.6265489459037781,
"learning_rate": 4.277569313094809e-07,
"loss": 0.0074,
"step": 48
},
{
"epoch": 0.011656345943695686,
"grad_norm": 0.4294980466365814,
"learning_rate": 1.0705383806982606e-07,
"loss": 0.0147,
"step": 49
},
{
"epoch": 0.011894230554791517,
"grad_norm": 0.5574062466621399,
"learning_rate": 0.0,
"loss": 0.0383,
"step": 50
}
],
"logging_steps": 1,
"max_steps": 50,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 13,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.548356111302656e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}