| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 32, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 0.03922202742278759, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1908, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 0.03600727500414045, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1872, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 0.041458677460302705, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1789, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.042477239921338006, | |
| "learning_rate": 5e-05, | |
| "loss": 0.191, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.04699963432811002, | |
| "learning_rate": 4.984280524733107e-05, | |
| "loss": 0.1832, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.04547329936830524, | |
| "learning_rate": 4.937319780454559e-05, | |
| "loss": 0.1806, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 0.04367593014445298, | |
| "learning_rate": 4.8597083257709194e-05, | |
| "loss": 0.169, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.03429260122917999, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.1523, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 0.025555649134774323, | |
| "learning_rate": 4.6168104980707107e-05, | |
| "loss": 0.1608, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.035622860476729544, | |
| "learning_rate": 4.454578706170075e-05, | |
| "loss": 0.1505, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 0.0459250933312251, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.1613, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.044900129344464294, | |
| "learning_rate": 4.058724504646834e-05, | |
| "loss": 0.1597, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 0.03285936274496835, | |
| "learning_rate": 3.830080191288342e-05, | |
| "loss": 0.1295, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.03104349497374127, | |
| "learning_rate": 3.5847093477938956e-05, | |
| "loss": 0.1582, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.024404726332163458, | |
| "learning_rate": 3.3256976548879184e-05, | |
| "loss": 0.137, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.022789243213683145, | |
| "learning_rate": 3.056302334890786e-05, | |
| "loss": 0.1367, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 0.017586552406833603, | |
| "learning_rate": 2.7799111902582696e-05, | |
| "loss": 0.1292, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.016792838194229127, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1372, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1875, | |
| "grad_norm": 0.019142947911569803, | |
| "learning_rate": 2.2200888097417307e-05, | |
| "loss": 0.1274, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.019220489564682752, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 0.1303, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3125, | |
| "grad_norm": 0.02264252406960103, | |
| "learning_rate": 1.6743023451120832e-05, | |
| "loss": 0.1362, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 0.020883245611579733, | |
| "learning_rate": 1.4152906522061048e-05, | |
| "loss": 0.1352, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 0.021254829840046948, | |
| "learning_rate": 1.1699198087116589e-05, | |
| "loss": 0.1193, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.022396775591741604, | |
| "learning_rate": 9.412754953531663e-06, | |
| "loss": 0.1193, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.01930808848159821, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.1366, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.018929254567235987, | |
| "learning_rate": 5.454212938299255e-06, | |
| "loss": 0.1257, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6875, | |
| "grad_norm": 0.02122756340453363, | |
| "learning_rate": 3.831895019292897e-06, | |
| "loss": 0.1256, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.017855075491123307, | |
| "learning_rate": 2.475778302439524e-06, | |
| "loss": 0.1273, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 0.0181714053917743, | |
| "learning_rate": 1.4029167422908107e-06, | |
| "loss": 0.1231, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.01977447479487821, | |
| "learning_rate": 6.268021954544096e-07, | |
| "loss": 0.1352, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9375, | |
| "grad_norm": 0.022144566953258688, | |
| "learning_rate": 1.571947526689349e-07, | |
| "loss": 0.1342, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.020972476400823264, | |
| "learning_rate": 0.0, | |
| "loss": 0.117, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 32, | |
| "total_flos": 600196704632832.0, | |
| "train_loss": 0.14642456639558077, | |
| "train_runtime": 1046.7446, | |
| "train_samples_per_second": 0.478, | |
| "train_steps_per_second": 0.031 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 32, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 600196704632832.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |