| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5897082090377808, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3566, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5390245914459229, | |
| "learning_rate": 0.0001, | |
| "loss": 2.2503, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5155873015873016, | |
| "eval_loss": 2.3237216472625732, | |
| "eval_runtime": 6.3164, | |
| "eval_samples_per_second": 79.159, | |
| "eval_steps_per_second": 9.974, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 17.2, | |
| "eval_f1": 25.586581359816655, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.7468857169151306, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1798, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.9109599590301514, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1094, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8458774089813232, | |
| "learning_rate": 0.0001, | |
| "loss": 2.087, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5163809523809524, | |
| "eval_loss": 2.3309242725372314, | |
| "eval_runtime": 5.1896, | |
| "eval_samples_per_second": 96.347, | |
| "eval_steps_per_second": 12.14, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 18.6, | |
| "eval_f1": 26.043196248196264, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.171212077140808, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8413, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.1294909715652466, | |
| "learning_rate": 0.0001, | |
| "loss": 1.849, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5144761904761905, | |
| "eval_loss": 2.401942253112793, | |
| "eval_runtime": 5.5775, | |
| "eval_samples_per_second": 89.646, | |
| "eval_steps_per_second": 11.295, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 18.4, | |
| "eval_f1": 26.966370002546494, | |
| "step": 750 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.292008841170125e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |