| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.1611291291820123, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.175, | |
| "mean_token_accuracy": 0.7521877646446228, | |
| "num_tokens": 2465164.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.6622829891502945, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.0551, | |
| "mean_token_accuracy": 0.7775114715099335, | |
| "num_tokens": 4604565.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.3321063533662744, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.974, | |
| "mean_token_accuracy": 0.7821209251880645, | |
| "num_tokens": 7112937.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.17372883404582676, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.9079, | |
| "mean_token_accuracy": 0.797761058807373, | |
| "num_tokens": 9211540.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.3250721867237713, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.8587, | |
| "mean_token_accuracy": 0.80385662317276, | |
| "num_tokens": 11676501.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.12878896880550417, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8491, | |
| "mean_token_accuracy": 0.8127783119678498, | |
| "num_tokens": 13823724.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.12156986655948128, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.8216, | |
| "mean_token_accuracy": 0.8109423339366912, | |
| "num_tokens": 16296257.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.11634538375626934, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.7842, | |
| "mean_token_accuracy": 0.8188613414764404, | |
| "num_tokens": 18424879.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 0.10458399531482464, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.7492, | |
| "mean_token_accuracy": 0.8257159292697906, | |
| "num_tokens": 20909688.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.10973027739730677, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.7812, | |
| "mean_token_accuracy": 0.8185243546962738, | |
| "num_tokens": 23043160.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 0.11172428707492313, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.7663, | |
| "mean_token_accuracy": 0.8235744714736939, | |
| "num_tokens": 25563345.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.1207429261911167, | |
| "learning_rate": 0.0, | |
| "loss": 0.7442, | |
| "mean_token_accuracy": 0.8240064084529877, | |
| "num_tokens": 27655450.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 60, | |
| "total_flos": 18782965727232.0, | |
| "train_loss": 0.8722121318181356, | |
| "train_runtime": 466.0776, | |
| "train_samples_per_second": 7.853, | |
| "train_steps_per_second": 0.129 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 60, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "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": 18782965727232.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |