| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0021108179419525065, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.443271767810026e-05, | |
| "grad_norm": 11.129352569580078, | |
| "learning_rate": 2e-05, | |
| "loss": 7.1042, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.443271767810026e-05, | |
| "eval_loss": 7.151242256164551, | |
| "eval_runtime": 420.6162, | |
| "eval_samples_per_second": 11.856, | |
| "eval_steps_per_second": 5.929, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016886543535620053, | |
| "grad_norm": 11.819311141967773, | |
| "learning_rate": 4e-05, | |
| "loss": 6.2728, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002532981530343008, | |
| "grad_norm": 12.868980407714844, | |
| "learning_rate": 6e-05, | |
| "loss": 7.5328, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002532981530343008, | |
| "eval_loss": 7.0206756591796875, | |
| "eval_runtime": 422.3489, | |
| "eval_samples_per_second": 11.808, | |
| "eval_steps_per_second": 5.905, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00033773087071240106, | |
| "grad_norm": 9.709155082702637, | |
| "learning_rate": 8e-05, | |
| "loss": 6.0845, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00042216358839050134, | |
| "grad_norm": 12.9670991897583, | |
| "learning_rate": 0.0001, | |
| "loss": 6.0418, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005065963060686016, | |
| "grad_norm": 13.192237854003906, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 6.0895, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005065963060686016, | |
| "eval_loss": 4.537143230438232, | |
| "eval_runtime": 422.4237, | |
| "eval_samples_per_second": 11.806, | |
| "eval_steps_per_second": 5.904, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005910290237467018, | |
| "grad_norm": 22.602550506591797, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 4.155, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0006754617414248021, | |
| "grad_norm": 15.708683967590332, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 3.0346, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0007598944591029024, | |
| "grad_norm": 17.4925594329834, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.8198, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0007598944591029024, | |
| "eval_loss": 1.1246894598007202, | |
| "eval_runtime": 422.3964, | |
| "eval_samples_per_second": 11.806, | |
| "eval_steps_per_second": 5.904, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008443271767810027, | |
| "grad_norm": 8.983549118041992, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.2177, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000928759894459103, | |
| "grad_norm": 7.7590765953063965, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.1052, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0010131926121372031, | |
| "grad_norm": 7.711614608764648, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.3309, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0010131926121372031, | |
| "eval_loss": 0.4180523753166199, | |
| "eval_runtime": 422.3609, | |
| "eval_samples_per_second": 11.807, | |
| "eval_steps_per_second": 5.905, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0010976253298153034, | |
| "grad_norm": 5.040955066680908, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.2055, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0011820580474934037, | |
| "grad_norm": 10.291000366210938, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.4698, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.001266490765171504, | |
| "grad_norm": 9.118146896362305, | |
| "learning_rate": 5e-05, | |
| "loss": 0.231, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001266490765171504, | |
| "eval_loss": 0.28742024302482605, | |
| "eval_runtime": 422.1719, | |
| "eval_samples_per_second": 11.813, | |
| "eval_steps_per_second": 5.908, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0013509234828496042, | |
| "grad_norm": 9.136611938476562, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.2849, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0014353562005277045, | |
| "grad_norm": 7.953118324279785, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.1662, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0015197889182058048, | |
| "grad_norm": 2.821458578109741, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0864, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0015197889182058048, | |
| "eval_loss": 0.24295015633106232, | |
| "eval_runtime": 422.3791, | |
| "eval_samples_per_second": 11.807, | |
| "eval_steps_per_second": 5.905, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001604221635883905, | |
| "grad_norm": 1.723188042640686, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.044, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0016886543535620053, | |
| "grad_norm": 6.9622321128845215, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1891, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0017730870712401056, | |
| "grad_norm": 5.619901180267334, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.1357, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0017730870712401056, | |
| "eval_loss": 0.21679861843585968, | |
| "eval_runtime": 422.402, | |
| "eval_samples_per_second": 11.806, | |
| "eval_steps_per_second": 5.904, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.001857519788918206, | |
| "grad_norm": 8.403288841247559, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.2989, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.001941952506596306, | |
| "grad_norm": 2.9916210174560547, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0602, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0020263852242744062, | |
| "grad_norm": 4.566269874572754, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.1002, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0020263852242744062, | |
| "eval_loss": 0.20803606510162354, | |
| "eval_runtime": 422.4259, | |
| "eval_samples_per_second": 11.806, | |
| "eval_steps_per_second": 5.904, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0021108179419525065, | |
| "grad_norm": 3.1905064582824707, | |
| "learning_rate": 0.0, | |
| "loss": 0.0737, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9247821240729600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |