| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.424778761061947, | |
| "eval_steps": 500, | |
| "global_step": 35, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1415929203539823, | |
| "grad_norm": 23.903297652464886, | |
| "learning_rate": 4.989935734988098e-06, | |
| "loss": 0.9826, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 3.331359502922686, | |
| "learning_rate": 4.752422169756048e-06, | |
| "loss": 0.4239, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.2831858407079646, | |
| "grad_norm": 1.2251403720606269, | |
| "learning_rate": 4.058724504646834e-06, | |
| "loss": 0.2485, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.991150442477876, | |
| "grad_norm": 1.02948929057659, | |
| "learning_rate": 3.056302334890786e-06, | |
| "loss": 0.2028, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.566371681415929, | |
| "grad_norm": 0.830474450051415, | |
| "learning_rate": 1.9436976651092143e-06, | |
| "loss": 0.1656, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.1415929203539825, | |
| "grad_norm": 2.49966326639145, | |
| "learning_rate": 9.412754953531664e-07, | |
| "loss": 0.1383, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.849557522123894, | |
| "grad_norm": 0.7327188705845169, | |
| "learning_rate": 2.4757783024395244e-07, | |
| "loss": 0.1194, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.424778761061947, | |
| "grad_norm": 0.6691628690742748, | |
| "learning_rate": 0.0, | |
| "loss": 0.1139, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.424778761061947, | |
| "step": 35, | |
| "total_flos": 2.3667830138339328e+17, | |
| "train_loss": 0.21773582696914673, | |
| "train_runtime": 2998.0685, | |
| "train_samples_per_second": 1.499, | |
| "train_steps_per_second": 0.012 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 35, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3667830138339328e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |