| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3908045977011494, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.418604651162792e-05, | |
| "loss": 1.9238, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.1818181872367859, | |
| "eval_loss": 1.9564208984375, | |
| "eval_runtime": 19.1351, | |
| "eval_samples_per_second": 4.599, | |
| "eval_steps_per_second": 1.15, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 8.837209302325582e-05, | |
| "loss": 1.9589, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.15909090638160706, | |
| "eval_loss": 1.9498623609542847, | |
| "eval_runtime": 21.5957, | |
| "eval_samples_per_second": 4.075, | |
| "eval_steps_per_second": 1.019, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 8.255813953488373e-05, | |
| "loss": 1.9677, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_accuracy": 0.17045454680919647, | |
| "eval_loss": 1.8762873411178589, | |
| "eval_runtime": 19.1489, | |
| "eval_samples_per_second": 4.596, | |
| "eval_steps_per_second": 1.149, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 7.674418604651163e-05, | |
| "loss": 1.9357, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.21590909361839294, | |
| "eval_loss": 1.8796330690383911, | |
| "eval_runtime": 21.6027, | |
| "eval_samples_per_second": 4.074, | |
| "eval_steps_per_second": 1.018, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 7.093023255813955e-05, | |
| "loss": 1.9529, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_accuracy": 0.22727273404598236, | |
| "eval_loss": 1.841020107269287, | |
| "eval_runtime": 22.0824, | |
| "eval_samples_per_second": 3.985, | |
| "eval_steps_per_second": 0.996, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.511627906976745e-05, | |
| "loss": 1.8197, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.35227271914482117, | |
| "eval_loss": 1.6947021484375, | |
| "eval_runtime": 21.8177, | |
| "eval_samples_per_second": 4.033, | |
| "eval_steps_per_second": 1.008, | |
| "step": 60 | |
| } | |
| ], | |
| "max_steps": 172, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.7581701250290854e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |