| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.34557235421166305, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011519078473722102, | |
| "grad_norm": 19.9857120513916, | |
| "learning_rate": 1e-05, | |
| "loss": 13.0002, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011519078473722102, | |
| "eval_loss": 15.875571250915527, | |
| "eval_runtime": 43.2844, | |
| "eval_samples_per_second": 3.396, | |
| "eval_steps_per_second": 1.71, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.023038156947444204, | |
| "grad_norm": 18.35199737548828, | |
| "learning_rate": 2e-05, | |
| "loss": 13.3377, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03455723542116631, | |
| "grad_norm": 20.499919891357422, | |
| "learning_rate": 3e-05, | |
| "loss": 11.844, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04607631389488841, | |
| "grad_norm": 21.232027053833008, | |
| "learning_rate": 4e-05, | |
| "loss": 12.9647, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05759539236861051, | |
| "grad_norm": 19.538623809814453, | |
| "learning_rate": 5e-05, | |
| "loss": 11.7034, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06911447084233262, | |
| "grad_norm": 20.755231857299805, | |
| "learning_rate": 6e-05, | |
| "loss": 11.0206, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.08063354931605471, | |
| "grad_norm": 29.196537017822266, | |
| "learning_rate": 7e-05, | |
| "loss": 10.6785, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.09215262778977681, | |
| "grad_norm": 24.786623001098633, | |
| "learning_rate": 8e-05, | |
| "loss": 10.1987, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09215262778977681, | |
| "eval_loss": NaN, | |
| "eval_runtime": 26.4418, | |
| "eval_samples_per_second": 5.559, | |
| "eval_steps_per_second": 2.799, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.10367170626349892, | |
| "grad_norm": 18.314897537231445, | |
| "learning_rate": 9e-05, | |
| "loss": 8.648, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.11519078473722102, | |
| "grad_norm": 15.454906463623047, | |
| "learning_rate": 0.0001, | |
| "loss": 6.5857, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12670986321094313, | |
| "grad_norm": 26.08247184753418, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 6.7318, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.13822894168466524, | |
| "grad_norm": 45.0203857421875, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 6.6639, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.14974802015838734, | |
| "grad_norm": 44.75263595581055, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 5.3623, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.16126709863210942, | |
| "grad_norm": 48.91502380371094, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 4.9463, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.17278617710583152, | |
| "grad_norm": 26.311792373657227, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.9532, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.18430525557955363, | |
| "grad_norm": 45.528385162353516, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 4.0457, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.18430525557955363, | |
| "eval_loss": NaN, | |
| "eval_runtime": 26.5549, | |
| "eval_samples_per_second": 5.536, | |
| "eval_steps_per_second": 2.787, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.19582433405327573, | |
| "grad_norm": 42.59834671020508, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 4.3579, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.20734341252699784, | |
| "grad_norm": 18.206153869628906, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 3.8252, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.21886249100071994, | |
| "grad_norm": 23.28106689453125, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 3.7844, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.23038156947444205, | |
| "grad_norm": 26.331771850585938, | |
| "learning_rate": 5e-05, | |
| "loss": 3.4637, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24190064794816415, | |
| "grad_norm": 24.80760955810547, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 3.3158, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.25341972642188626, | |
| "grad_norm": 11.706987380981445, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.8568, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.26493880489560834, | |
| "grad_norm": 13.360960960388184, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 3.0235, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.27645788336933047, | |
| "grad_norm": 14.578153610229492, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.4507, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.27645788336933047, | |
| "eval_loss": NaN, | |
| "eval_runtime": 26.5699, | |
| "eval_samples_per_second": 5.533, | |
| "eval_steps_per_second": 2.785, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.28797696184305255, | |
| "grad_norm": 13.395209312438965, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 2.4717, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2994960403167747, | |
| "grad_norm": 11.832671165466309, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.4415, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.31101511879049676, | |
| "grad_norm": 13.087400436401367, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.4317, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.32253419726421884, | |
| "grad_norm": 13.915994644165039, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 2.6166, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.33405327573794097, | |
| "grad_norm": 14.689959526062012, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.4773, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.34557235421166305, | |
| "grad_norm": 13.360550880432129, | |
| "learning_rate": 0.0, | |
| "loss": 2.2851, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.8405464677023744e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |