| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6054838709677419, | |
| "eval_loss": 0.21967020630836487, | |
| "eval_runtime": 1.4167, | |
| "eval_samples_per_second": 2188.185, | |
| "eval_steps_per_second": 45.881, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3526, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8393548387096774, | |
| "eval_loss": 0.10301334410905838, | |
| "eval_runtime": 1.5626, | |
| "eval_samples_per_second": 1983.839, | |
| "eval_steps_per_second": 41.597, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8845161290322581, | |
| "eval_loss": 0.06688232719898224, | |
| "eval_runtime": 1.4727, | |
| "eval_samples_per_second": 2105.045, | |
| "eval_steps_per_second": 44.138, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1192, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8993548387096775, | |
| "eval_loss": 0.049738675355911255, | |
| "eval_runtime": 1.4561, | |
| "eval_samples_per_second": 2129.004, | |
| "eval_steps_per_second": 44.64, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0738, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9132258064516129, | |
| "eval_loss": 0.04058445617556572, | |
| "eval_runtime": 1.4513, | |
| "eval_samples_per_second": 2135.993, | |
| "eval_steps_per_second": 44.787, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9206451612903226, | |
| "eval_loss": 0.03555791452527046, | |
| "eval_runtime": 1.4469, | |
| "eval_samples_per_second": 2142.51, | |
| "eval_steps_per_second": 44.924, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0577, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 520587680240208.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.016710650067674027, | |
| "num_train_epochs": 9, | |
| "temperature": 6 | |
| } | |
| } | |