aadityap's picture
Model save
4d67577 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 39,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07692307692307693,
"grad_norm": 0.012279312698118647,
"learning_rate": 2e-05,
"loss": 0.1813,
"step": 1
},
{
"epoch": 0.15384615384615385,
"grad_norm": 0.013334805575189963,
"learning_rate": 4e-05,
"loss": 0.1857,
"step": 2
},
{
"epoch": 0.23076923076923078,
"grad_norm": 0.015612095744898443,
"learning_rate": 6.000000000000001e-05,
"loss": 0.1729,
"step": 3
},
{
"epoch": 0.3076923076923077,
"grad_norm": 0.01398827814833297,
"learning_rate": 8e-05,
"loss": 0.1855,
"step": 4
},
{
"epoch": 0.38461538461538464,
"grad_norm": 0.01718202733092743,
"learning_rate": 7.983897175980957e-05,
"loss": 0.1841,
"step": 5
},
{
"epoch": 0.46153846153846156,
"grad_norm": 0.02026760713177952,
"learning_rate": 7.93571835439452e-05,
"loss": 0.1942,
"step": 6
},
{
"epoch": 0.5384615384615384,
"grad_norm": 0.021116358424805134,
"learning_rate": 7.855851442783414e-05,
"loss": 0.1636,
"step": 7
},
{
"epoch": 0.6153846153846154,
"grad_norm": 0.019659753277959224,
"learning_rate": 7.74493948255895e-05,
"loss": 0.1789,
"step": 8
},
{
"epoch": 0.6923076923076923,
"grad_norm": 0.024195499403827956,
"learning_rate": 7.603875471609677e-05,
"loss": 0.1542,
"step": 9
},
{
"epoch": 0.7692307692307693,
"grad_norm": 0.022282939008047377,
"learning_rate": 7.433795174407465e-05,
"loss": 0.1487,
"step": 10
},
{
"epoch": 0.8461538461538461,
"grad_norm": 0.01809434218165359,
"learning_rate": 7.236067977499791e-05,
"loss": 0.1424,
"step": 11
},
{
"epoch": 0.9230769230769231,
"grad_norm": 0.01498779327659144,
"learning_rate": 7.012285864014445e-05,
"loss": 0.1386,
"step": 12
},
{
"epoch": 1.0,
"grad_norm": 0.014528843485793582,
"learning_rate": 6.76425059594746e-05,
"loss": 0.1369,
"step": 13
},
{
"epoch": 1.0769230769230769,
"grad_norm": 0.01496419138280413,
"learning_rate": 6.493959207434934e-05,
"loss": 0.1303,
"step": 14
},
{
"epoch": 1.1538461538461537,
"grad_norm": 0.016070989448498302,
"learning_rate": 6.20358792580841e-05,
"loss": 0.1286,
"step": 15
},
{
"epoch": 1.2307692307692308,
"grad_norm": 0.011433097296470634,
"learning_rate": 5.895474649891995e-05,
"loss": 0.121,
"step": 16
},
{
"epoch": 1.3076923076923077,
"grad_norm": 0.012070129386737415,
"learning_rate": 5.572100126615695e-05,
"loss": 0.1193,
"step": 17
},
{
"epoch": 1.3846153846153846,
"grad_norm": 0.012168329527924309,
"learning_rate": 5.23606797749979e-05,
"loss": 0.1204,
"step": 18
},
{
"epoch": 1.4615384615384617,
"grad_norm": 0.011013350274673917,
"learning_rate": 4.890083735825258e-05,
"loss": 0.1097,
"step": 19
},
{
"epoch": 1.5384615384615383,
"grad_norm": 0.010443793995378611,
"learning_rate": 4.5369330632706223e-05,
"loss": 0.1061,
"step": 20
},
{
"epoch": 1.6153846153846154,
"grad_norm": 0.01083313899885156,
"learning_rate": 4.17945932140206e-05,
"loss": 0.105,
"step": 21
},
{
"epoch": 1.6923076923076923,
"grad_norm": 0.009173732807179515,
"learning_rate": 3.820540678597942e-05,
"loss": 0.1112,
"step": 22
},
{
"epoch": 1.7692307692307692,
"grad_norm": 0.008949126805738214,
"learning_rate": 3.4630669367293797e-05,
"loss": 0.1237,
"step": 23
},
{
"epoch": 1.8461538461538463,
"grad_norm": 0.00892395968966195,
"learning_rate": 3.109916264174743e-05,
"loss": 0.1166,
"step": 24
},
{
"epoch": 1.9230769230769231,
"grad_norm": 0.008376249238064166,
"learning_rate": 2.7639320225002108e-05,
"loss": 0.1001,
"step": 25
},
{
"epoch": 2.0,
"grad_norm": 0.008886958404047452,
"learning_rate": 2.427899873384306e-05,
"loss": 0.0944,
"step": 26
},
{
"epoch": 2.076923076923077,
"grad_norm": 0.009891413110134516,
"learning_rate": 2.1045253501080058e-05,
"loss": 0.0918,
"step": 27
},
{
"epoch": 2.1538461538461537,
"grad_norm": 0.009524750918219289,
"learning_rate": 1.7964120741915905e-05,
"loss": 0.1005,
"step": 28
},
{
"epoch": 2.230769230769231,
"grad_norm": 0.010234607627096642,
"learning_rate": 1.5060407925650662e-05,
"loss": 0.0831,
"step": 29
},
{
"epoch": 2.3076923076923075,
"grad_norm": 0.009243983236580544,
"learning_rate": 1.2357494040525416e-05,
"loss": 0.113,
"step": 30
},
{
"epoch": 2.3846153846153846,
"grad_norm": 0.0095813183402306,
"learning_rate": 9.877141359855567e-06,
"loss": 0.1059,
"step": 31
},
{
"epoch": 2.4615384615384617,
"grad_norm": 0.009290469368289603,
"learning_rate": 7.639320225002106e-06,
"loss": 0.0894,
"step": 32
},
{
"epoch": 2.5384615384615383,
"grad_norm": 0.009152847350819013,
"learning_rate": 5.662048255925357e-06,
"loss": 0.0855,
"step": 33
},
{
"epoch": 2.6153846153846154,
"grad_norm": 0.009326281921041311,
"learning_rate": 3.961245283903239e-06,
"loss": 0.0887,
"step": 34
},
{
"epoch": 2.6923076923076925,
"grad_norm": 0.009455056184031627,
"learning_rate": 2.550605174410512e-06,
"loss": 0.0896,
"step": 35
},
{
"epoch": 2.769230769230769,
"grad_norm": 0.008940021878816214,
"learning_rate": 1.4414855721658705e-06,
"loss": 0.0864,
"step": 36
},
{
"epoch": 2.8461538461538463,
"grad_norm": 0.008976796883489532,
"learning_rate": 6.428164560548134e-07,
"loss": 0.0799,
"step": 37
},
{
"epoch": 2.9230769230769234,
"grad_norm": 0.009569723870099887,
"learning_rate": 1.6102824019043728e-07,
"loss": 0.0851,
"step": 38
},
{
"epoch": 3.0,
"grad_norm": 0.008668054116823958,
"learning_rate": 0.0,
"loss": 0.0857,
"step": 39
},
{
"epoch": 3.0,
"step": 39,
"total_flos": 75361378631680.0,
"train_loss": 0.12404328622879127,
"train_runtime": 1240.5674,
"train_samples_per_second": 0.484,
"train_steps_per_second": 0.031
}
],
"logging_steps": 1,
"max_steps": 39,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 75361378631680.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}