|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 1197, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"grad_norm": 1.2109375, |
|
"learning_rate": 1.1111111111111113e-05, |
|
"loss": 0.594, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 0.8125, |
|
"learning_rate": 1.9999414238190722e-05, |
|
"loss": 0.3314, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 0.765625, |
|
"learning_rate": 1.9978919779420427e-05, |
|
"loss": 0.2863, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 0.7578125, |
|
"learning_rate": 1.9929205817437334e-05, |
|
"loss": 0.2732, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 0.74609375, |
|
"learning_rate": 1.9850417920830427e-05, |
|
"loss": 0.2624, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 0.77734375, |
|
"learning_rate": 1.9742786790241548e-05, |
|
"loss": 0.2557, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 0.70703125, |
|
"learning_rate": 1.9606627582845587e-05, |
|
"loss": 0.2508, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.7578125, |
|
"learning_rate": 1.944233898953152e-05, |
|
"loss": 0.2474, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 0.71484375, |
|
"learning_rate": 1.9250402067486523e-05, |
|
"loss": 0.2438, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"grad_norm": 0.75, |
|
"learning_rate": 1.9031378831601425e-05, |
|
"loss": 0.2405, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 0.734375, |
|
"learning_rate": 1.878591060882203e-05, |
|
"loss": 0.2359, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 0.74609375, |
|
"learning_rate": 1.851471616026505e-05, |
|
"loss": 0.2347, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.8218589576597203e-05, |
|
"loss": 0.2305, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 0.75, |
|
"learning_rate": 1.789839795284022e-05, |
|
"loss": 0.2324, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.7555078849410044e-05, |
|
"loss": 0.23, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.6796875, |
|
"learning_rate": 1.7189637546824752e-05, |
|
"loss": 0.2235, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"grad_norm": 0.70703125, |
|
"learning_rate": 1.680314410211967e-05, |
|
"loss": 0.2236, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"grad_norm": 0.7578125, |
|
"learning_rate": 1.6396730215588913e-05, |
|
"loss": 0.2236, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"grad_norm": 0.72265625, |
|
"learning_rate": 1.5971585917027864e-05, |
|
"loss": 0.2225, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.552895608117971e-05, |
|
"loss": 0.2212, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"grad_norm": 0.7265625, |
|
"learning_rate": 1.5070136782589236e-05, |
|
"loss": 0.1921, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"grad_norm": 0.75, |
|
"learning_rate": 1.4596471500537334e-05, |
|
"loss": 0.1902, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"grad_norm": 0.7265625, |
|
"learning_rate": 1.4109347185168577e-05, |
|
"loss": 0.19, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.3610190196330777e-05, |
|
"loss": 0.1869, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"grad_norm": 0.76171875, |
|
"learning_rate": 1.3100462127018108e-05, |
|
"loss": 0.1902, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"grad_norm": 0.74609375, |
|
"learning_rate": 1.2581655523647234e-05, |
|
"loss": 0.1888, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"grad_norm": 0.75390625, |
|
"learning_rate": 1.2055289515698008e-05, |
|
"loss": 0.1882, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"grad_norm": 0.74609375, |
|
"learning_rate": 1.1522905367515705e-05, |
|
"loss": 0.1881, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.0986061965299646e-05, |
|
"loss": 0.1855, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"grad_norm": 0.73046875, |
|
"learning_rate": 1.0446331252492864e-05, |
|
"loss": 0.1892, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.55, |
|
"grad_norm": 0.71484375, |
|
"learning_rate": 9.905293626938564e-06, |
|
"loss": 0.1856, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 0.7890625, |
|
"learning_rate": 9.364533313281056e-06, |
|
"loss": 0.1848, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 1.65, |
|
"grad_norm": 0.78125, |
|
"learning_rate": 8.825633724161336e-06, |
|
"loss": 0.1856, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"grad_norm": 0.73828125, |
|
"learning_rate": 8.290172823790337e-06, |
|
"loss": 0.1837, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"grad_norm": 0.75, |
|
"learning_rate": 7.759718507475822e-06, |
|
"loss": 0.1841, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"grad_norm": 0.72265625, |
|
"learning_rate": 7.235824010632284e-06, |
|
"loss": 0.1856, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"grad_norm": 0.77734375, |
|
"learning_rate": 6.720023360716818e-06, |
|
"loss": 0.1849, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"grad_norm": 0.76953125, |
|
"learning_rate": 6.2138268854082094e-06, |
|
"loss": 0.1812, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"grad_norm": 0.74609375, |
|
"learning_rate": 5.7187167901819665e-06, |
|
"loss": 0.1853, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"grad_norm": 0.72265625, |
|
"learning_rate": 5.236142818230518e-06, |
|
"loss": 0.1817, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"grad_norm": 0.76171875, |
|
"learning_rate": 4.767518005437073e-06, |
|
"loss": 0.1694, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"grad_norm": 0.76171875, |
|
"learning_rate": 4.314214542832889e-06, |
|
"loss": 0.1679, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"grad_norm": 0.73828125, |
|
"learning_rate": 3.877559758653358e-06, |
|
"loss": 0.1724, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"grad_norm": 0.77734375, |
|
"learning_rate": 3.4588322317578395e-06, |
|
"loss": 0.1685, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"grad_norm": 0.71875, |
|
"learning_rate": 3.0592580477936606e-06, |
|
"loss": 0.1674, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"grad_norm": 0.76171875, |
|
"learning_rate": 2.6800072090667085e-06, |
|
"loss": 0.1687, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 2.36, |
|
"grad_norm": 0.80078125, |
|
"learning_rate": 2.3221902086308943e-06, |
|
"loss": 0.1669, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"grad_norm": 0.7265625, |
|
"learning_rate": 1.986854778628067e-06, |
|
"loss": 0.1693, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 2.46, |
|
"grad_norm": 0.8046875, |
|
"learning_rate": 1.67498282239956e-06, |
|
"loss": 0.169, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 2.51, |
|
"grad_norm": 0.73828125, |
|
"learning_rate": 1.3874875393525722e-06, |
|
"loss": 0.1668, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.1252107510000843e-06, |
|
"loss": 0.1684, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 2.61, |
|
"grad_norm": 0.7734375, |
|
"learning_rate": 8.889204360040471e-07, |
|
"loss": 0.1663, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 6.793084814394945e-07, |
|
"loss": 0.1685, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 2.71, |
|
"grad_norm": 0.75, |
|
"learning_rate": 4.969886568641757e-07, |
|
"loss": 0.1669, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"grad_norm": 0.765625, |
|
"learning_rate": 3.424948171258824e-07, |
|
"loss": 0.1691, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 2.81, |
|
"grad_norm": 0.7890625, |
|
"learning_rate": 2.16279339169847e-07, |
|
"loss": 0.1718, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"grad_norm": 0.78125, |
|
"learning_rate": 1.1871179742348416e-07, |
|
"loss": 0.1648, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"grad_norm": 0.73828125, |
|
"learning_rate": 5.007788163703842e-08, |
|
"loss": 0.1706, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"grad_norm": 0.81640625, |
|
"learning_rate": 1.0578560348905031e-08, |
|
"loss": 0.1694, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 1197, |
|
"total_flos": 8.742938292204339e+17, |
|
"train_loss": 0.20622981182216304, |
|
"train_runtime": 3123.6217, |
|
"train_samples_per_second": 55.182, |
|
"train_steps_per_second": 0.383 |
|
} |
|
], |
|
"logging_steps": 20, |
|
"max_steps": 1197, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 10000, |
|
"total_flos": 8.742938292204339e+17, |
|
"train_batch_size": 36, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|