| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03439911797133407, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008820286659316428, | |
| "eval_loss": 4.544261932373047, | |
| "eval_runtime": 38.3483, | |
| "eval_samples_per_second": 12.465, | |
| "eval_steps_per_second": 6.232, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004410143329658214, | |
| "grad_norm": 3.595167875289917, | |
| "learning_rate": 5e-05, | |
| "loss": 4.2546, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008820286659316428, | |
| "grad_norm": 3.887547016143799, | |
| "learning_rate": 0.0001, | |
| "loss": 4.0028, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008820286659316428, | |
| "eval_loss": 3.5908195972442627, | |
| "eval_runtime": 38.5295, | |
| "eval_samples_per_second": 12.406, | |
| "eval_steps_per_second": 6.203, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013230429988974642, | |
| "grad_norm": 10.134819984436035, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.3156, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.017640573318632856, | |
| "grad_norm": 5.363346099853516, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.8045, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.017640573318632856, | |
| "eval_loss": 2.6722819805145264, | |
| "eval_runtime": 38.5022, | |
| "eval_samples_per_second": 12.415, | |
| "eval_steps_per_second": 6.207, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.022050716648291068, | |
| "grad_norm": 6.477142810821533, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.4782, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.026460859977949284, | |
| "grad_norm": 5.2271623611450195, | |
| "learning_rate": 5e-05, | |
| "loss": 2.5298, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.026460859977949284, | |
| "eval_loss": 2.3998324871063232, | |
| "eval_runtime": 38.7979, | |
| "eval_samples_per_second": 12.32, | |
| "eval_steps_per_second": 6.16, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.030871003307607496, | |
| "grad_norm": 4.6522016525268555, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.3057, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7453256523448320.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |