| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0055793193230425885, | |
| "eval_steps": 2, | |
| "global_step": 15, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0003719546215361726, | |
| "grad_norm": 28.77135467529297, | |
| "learning_rate": 2e-05, | |
| "loss": 45.9604, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003719546215361726, | |
| "eval_loss": 11.40648078918457, | |
| "eval_runtime": 137.0275, | |
| "eval_samples_per_second": 8.261, | |
| "eval_steps_per_second": 4.131, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007439092430723452, | |
| "grad_norm": 23.410951614379883, | |
| "learning_rate": 4e-05, | |
| "loss": 46.8527, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0007439092430723452, | |
| "eval_loss": 11.40257453918457, | |
| "eval_runtime": 137.8996, | |
| "eval_samples_per_second": 8.209, | |
| "eval_steps_per_second": 4.104, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0011158638646085177, | |
| "grad_norm": 33.30792999267578, | |
| "learning_rate": 6e-05, | |
| "loss": 44.8008, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0014878184861446904, | |
| "grad_norm": 27.510290145874023, | |
| "learning_rate": 8e-05, | |
| "loss": 48.3234, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0014878184861446904, | |
| "eval_loss": 11.253341674804688, | |
| "eval_runtime": 138.0011, | |
| "eval_samples_per_second": 8.203, | |
| "eval_steps_per_second": 4.101, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.001859773107680863, | |
| "grad_norm": 23.642324447631836, | |
| "learning_rate": 0.0001, | |
| "loss": 42.2976, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0022317277292170355, | |
| "grad_norm": 37.700042724609375, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 42.5346, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0022317277292170355, | |
| "eval_loss": 10.256876945495605, | |
| "eval_runtime": 138.1284, | |
| "eval_samples_per_second": 8.195, | |
| "eval_steps_per_second": 4.098, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002603682350753208, | |
| "grad_norm": 46.39228057861328, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 39.644, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0029756369722893808, | |
| "grad_norm": 47.501041412353516, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 35.7466, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0029756369722893808, | |
| "eval_loss": 7.803372859954834, | |
| "eval_runtime": 138.0673, | |
| "eval_samples_per_second": 8.199, | |
| "eval_steps_per_second": 4.099, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0033475915938255534, | |
| "grad_norm": 45.90718460083008, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 31.2201, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003719546215361726, | |
| "grad_norm": 51.53364181518555, | |
| "learning_rate": 5e-05, | |
| "loss": 23.9333, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.003719546215361726, | |
| "eval_loss": 5.604811191558838, | |
| "eval_runtime": 138.2369, | |
| "eval_samples_per_second": 8.189, | |
| "eval_steps_per_second": 4.094, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004091500836897899, | |
| "grad_norm": 53.22500991821289, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 27.342, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.004463455458434071, | |
| "grad_norm": 59.24814224243164, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 18.2533, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004463455458434071, | |
| "eval_loss": 4.373373985290527, | |
| "eval_runtime": 138.3748, | |
| "eval_samples_per_second": 8.181, | |
| "eval_steps_per_second": 4.09, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004835410079970244, | |
| "grad_norm": 45.4249153137207, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 7.9498, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.005207364701506416, | |
| "grad_norm": 51.06169509887695, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 16.3781, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.005207364701506416, | |
| "eval_loss": 4.014023780822754, | |
| "eval_runtime": 138.2266, | |
| "eval_samples_per_second": 8.189, | |
| "eval_steps_per_second": 4.095, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0055793193230425885, | |
| "grad_norm": 36.74103927612305, | |
| "learning_rate": 0.0, | |
| "loss": 14.7442, | |
| "step": 15 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 15, | |
| "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": 9498612518092800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |