| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.862254025044723, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.057245080500894455, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 14103756.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.057245080500894455, | |
| "eval_loss": NaN, | |
| "eval_runtime": 3.9235, | |
| "eval_samples_per_second": 12.744, | |
| "eval_steps_per_second": 3.313, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11449016100178891, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.17173524150268335, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.990365154573717e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.22898032200357782, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.961501876182148e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.28622540250447226, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3434704830053667, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.846666218300807e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.4007155635062612, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.761185582727977e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.45796064400715564, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5152057245080501, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.535927336897098e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5724508050089445, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.397114317029975e-05, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.629695885509839, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6869409660107334, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.070090031310558e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.7441860465116279, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.883279133655399e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.8014311270125224, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.8586762075134168, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.467056167950311e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.9159212880143113, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.239426430539243e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.9731663685152058, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0304114490161003, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0876565295169947, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.490299105985507e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.144901610017889, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2021466905187834, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.946477593864228e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.259391771019678, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.664685702961344e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.3166368515205724, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.373881932021467, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.087367864990233e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.4311270125223614, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.794314081535644e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.4311270125223614, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.19, | |
| "eval_samples_per_second": 11.933, | |
| "eval_steps_per_second": 3.103, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.4883720930232558, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.5456171735241502, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.205685918464356e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.6028622540250446, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.912632135009769e-05, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.6601073345259392, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.7173524150268338, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.3353142970386564e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7745974955277282, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.053522406135775e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.8318425760286225, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.889087656529517, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.509700894014496e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.9463327370304113, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.250000000000001e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0035778175313057, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.0608228980322005, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.760573569460757e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.118067978533095, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.53294383204969e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.1753130590339893, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.2325581395348837, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.1167208663446025e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.289803220035778, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9299099686894423e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.3470483005366725, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.404293381037567, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.602885682970026e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.464072663102903e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.518783542039356, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.5760286225402504, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.2388144172720251e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.633273703041145, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1533337816991932e-05, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.690518783542039, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.747763864042934, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0384981238178534e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.8050089445438284, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0096348454262845e-05, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.862254025044723, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.862254025044723, | |
| "eval_loss": NaN, | |
| "eval_runtime": 3.9994, | |
| "eval_samples_per_second": 12.502, | |
| "eval_steps_per_second": 3.251, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.998666393657016e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |