| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 375, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 9.75, | |
| "learning_rate": 1.9491669984901377e-05, | |
| "loss": 0.6663, | |
| "mean_token_accuracy": 0.8282047653198242, | |
| "num_input_tokens_seen": 2327984, | |
| "num_tokens": 2327984.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 17.25, | |
| "learning_rate": 1.72967916579403e-05, | |
| "loss": 0.3542, | |
| "mean_token_accuracy": 0.9049889421463013, | |
| "num_input_tokens_seen": 5204816, | |
| "num_tokens": 5204816.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.25, | |
| "learning_rate": 1.3756762552443555e-05, | |
| "loss": 0.0478, | |
| "mean_token_accuracy": 0.9875464010238647, | |
| "num_input_tokens_seen": 7271696, | |
| "num_tokens": 7271696.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 9.524180841762577e-06, | |
| "loss": 0.014, | |
| "mean_token_accuracy": 0.9964023971557617, | |
| "num_input_tokens_seen": 9599680, | |
| "num_tokens": 9599680.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.00946044921875, | |
| "learning_rate": 5.379315560596038e-06, | |
| "loss": 0.0023, | |
| "mean_token_accuracy": 0.9995168459415436, | |
| "num_input_tokens_seen": 12476512, | |
| "num_tokens": 12476512.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.169921875, | |
| "learning_rate": 2.0862653732958914e-06, | |
| "loss": 0.0002, | |
| "mean_token_accuracy": 0.9999828958511352, | |
| "num_input_tokens_seen": 14543392, | |
| "num_tokens": 14543392.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.0380859375, | |
| "learning_rate": 2.520983216615047e-07, | |
| "loss": 0.0001, | |
| "mean_token_accuracy": 1.0, | |
| "num_input_tokens_seen": 16871376, | |
| "num_tokens": 16871376.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "mean_token_accuracy": 1.0, | |
| "num_input_tokens_seen": 18179240, | |
| "num_tokens": 18179240.0, | |
| "step": 375, | |
| "total_flos": 1.0562823256014848e+17, | |
| "train_loss": 0.14466818872839213, | |
| "train_runtime": 6016.1372, | |
| "train_samples_per_second": 0.499, | |
| "train_steps_per_second": 0.062, | |
| "train_tokens_per_second": 176.962 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 375, | |
| "num_input_tokens_seen": 18179240, | |
| "num_train_epochs": 5, | |
| "save_steps": 0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0562823256014848e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |