| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0012104940522840316, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.103830903292389e-05, | |
| "eval_loss": 3.7120916843414307, | |
| "eval_runtime": 1355.8503, | |
| "eval_samples_per_second": 10.006, | |
| "eval_steps_per_second": 5.003, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00015519154516461942, | |
| "grad_norm": 2.1552507877349854, | |
| "learning_rate": 5e-05, | |
| "loss": 3.9595, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00031038309032923884, | |
| "grad_norm": 3.6430537700653076, | |
| "learning_rate": 0.0001, | |
| "loss": 3.5018, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00031038309032923884, | |
| "eval_loss": 3.1252193450927734, | |
| "eval_runtime": 1356.105, | |
| "eval_samples_per_second": 10.004, | |
| "eval_steps_per_second": 5.002, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0004655746354938583, | |
| "grad_norm": 2.4750373363494873, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.5722, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006207661806584777, | |
| "grad_norm": 1.5241283178329468, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.5678, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006207661806584777, | |
| "eval_loss": 1.1251658201217651, | |
| "eval_runtime": 1355.943, | |
| "eval_samples_per_second": 10.005, | |
| "eval_steps_per_second": 5.002, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0007759577258230971, | |
| "grad_norm": 1.527565360069275, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.1063, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0009311492709877166, | |
| "grad_norm": 1.7839757204055786, | |
| "learning_rate": 5e-05, | |
| "loss": 0.799, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0009311492709877166, | |
| "eval_loss": 0.6185202598571777, | |
| "eval_runtime": 1357.0146, | |
| "eval_samples_per_second": 9.997, | |
| "eval_steps_per_second": 4.998, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0010863408161523361, | |
| "grad_norm": 1.172727346420288, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.4837, | |
| "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": 1.1971306928996352e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |