| { | |
| "best_metric": 0.3674071431159973, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-85", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 85, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 0.9983, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9323308270676691, | |
| "eval_loss": 0.8032251596450806, | |
| "eval_runtime": 1.4706, | |
| "eval_samples_per_second": 90.439, | |
| "eval_steps_per_second": 2.04, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.5294117647058822e-05, | |
| "loss": 0.8323, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.2941176470588238e-05, | |
| "loss": 0.6984, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9548872180451128, | |
| "eval_loss": 0.5943161845207214, | |
| "eval_runtime": 1.3971, | |
| "eval_samples_per_second": 95.195, | |
| "eval_steps_per_second": 2.147, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.0588235294117648e-05, | |
| "loss": 0.6093, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 8.23529411764706e-06, | |
| "loss": 0.5056, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.4566296339035034, | |
| "eval_runtime": 1.3208, | |
| "eval_samples_per_second": 100.695, | |
| "eval_steps_per_second": 2.271, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.4601, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.3891778290271759, | |
| "eval_runtime": 1.5607, | |
| "eval_samples_per_second": 85.218, | |
| "eval_steps_per_second": 1.922, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 3.529411764705883e-06, | |
| "loss": 0.4161, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 0.3883, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.3674071431159973, | |
| "eval_runtime": 1.3973, | |
| "eval_samples_per_second": 95.18, | |
| "eval_steps_per_second": 2.147, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 85, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_loss": 0.6000023210749906, | |
| "train_runtime": 107.5648, | |
| "train_samples_per_second": 48.064, | |
| "train_steps_per_second": 0.79 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 85, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 4.006371770595533e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |