| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8955223880597015, | |
| "grad_norm": 0.900656714392764, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.932, | |
| "mean_token_accuracy": 0.8076159665981929, | |
| "num_tokens": 9406298.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.716417910447761, | |
| "grad_norm": 0.4764308318290881, | |
| "learning_rate": 1.9428774454610845e-05, | |
| "loss": 0.7128, | |
| "mean_token_accuracy": 0.8432981382716785, | |
| "num_tokens": 17339132.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.5373134328358207, | |
| "grad_norm": 0.29201950249687686, | |
| "learning_rate": 1.7219560939545246e-05, | |
| "loss": 0.6113, | |
| "mean_token_accuracy": 0.864370275627483, | |
| "num_tokens": 25334534.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.3582089552238807, | |
| "grad_norm": 0.20142965457472725, | |
| "learning_rate": 1.3728564777803089e-05, | |
| "loss": 0.538, | |
| "mean_token_accuracy": 0.8771601741964167, | |
| "num_tokens": 33275501.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.17910447761194, | |
| "grad_norm": 0.17437828908394032, | |
| "learning_rate": 9.57558796803852e-06, | |
| "loss": 0.5454, | |
| "mean_token_accuracy": 0.8782916686751626, | |
| "num_tokens": 41225062.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.22117380793963837, | |
| "learning_rate": 5.497962551823266e-06, | |
| "loss": 0.527, | |
| "mean_token_accuracy": 0.8798674431714144, | |
| "num_tokens": 49232938.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.895522388059701, | |
| "grad_norm": 0.14468869535451614, | |
| "learning_rate": 2.2196424568156073e-06, | |
| "loss": 0.5209, | |
| "mean_token_accuracy": 0.8800060609976451, | |
| "num_tokens": 58665081.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 6.7164179104477615, | |
| "grad_norm": 0.12750932530621908, | |
| "learning_rate": 3.226705306650113e-07, | |
| "loss": 0.4861, | |
| "mean_token_accuracy": 0.8856360175392844, | |
| "num_tokens": 66624412.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "mean_token_accuracy": 0.8826957188154522, | |
| "num_tokens": 68894483.0, | |
| "step": 42, | |
| "total_flos": 106682420035584.0, | |
| "train_loss": 0.6038688605739957, | |
| "train_runtime": 1070.0148, | |
| "train_samples_per_second": 3.474, | |
| "train_steps_per_second": 0.039 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 106682420035584.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |