| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 14, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 7e-05, | |
| "loss": 2.6914, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.44772838299951145, | |
| "eval_loss": 2.689453125, | |
| "eval_runtime": 2.0744, | |
| "eval_samples_per_second": 5.785, | |
| "eval_steps_per_second": 0.964, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7e-05, | |
| "loss": 2.6897, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.44772838299951145, | |
| "eval_loss": 2.689453125, | |
| "eval_runtime": 1.2329, | |
| "eval_samples_per_second": 9.733, | |
| "eval_steps_per_second": 1.622, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 6.5e-05, | |
| "loss": 2.668, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.44031916625956685, | |
| "eval_loss": 2.703125, | |
| "eval_runtime": 1.3532, | |
| "eval_samples_per_second": 8.868, | |
| "eval_steps_per_second": 1.478, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 2.7434, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.45330565054551375, | |
| "eval_loss": 2.591796875, | |
| "eval_runtime": 1.1891, | |
| "eval_samples_per_second": 10.092, | |
| "eval_steps_per_second": 1.682, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 5.4999999999999995e-05, | |
| "loss": 2.6265, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.4617733268197362, | |
| "eval_loss": 2.541015625, | |
| "eval_runtime": 1.3466, | |
| "eval_samples_per_second": 8.912, | |
| "eval_steps_per_second": 1.485, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.9999999999999996e-05, | |
| "loss": 2.5259, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.46405308581664223, | |
| "eval_loss": 2.515625, | |
| "eval_runtime": 1.3043, | |
| "eval_samples_per_second": 9.2, | |
| "eval_steps_per_second": 1.533, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 2.5566, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.46665852467024915, | |
| "eval_loss": 2.490234375, | |
| "eval_runtime": 1.3073, | |
| "eval_samples_per_second": 9.179, | |
| "eval_steps_per_second": 1.53, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 2.2317, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.4706888128969223, | |
| "eval_loss": 2.4765625, | |
| "eval_runtime": 2.2111, | |
| "eval_samples_per_second": 5.427, | |
| "eval_steps_per_second": 0.905, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.2397, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.4705259729685719, | |
| "eval_loss": 2.47265625, | |
| "eval_runtime": 1.2075, | |
| "eval_samples_per_second": 9.938, | |
| "eval_steps_per_second": 1.656, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 2.0162, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.4689789936492428, | |
| "eval_loss": 2.4765625, | |
| "eval_runtime": 1.6528, | |
| "eval_samples_per_second": 7.26, | |
| "eval_steps_per_second": 1.21, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 2.4999999999999998e-05, | |
| "loss": 2.0537, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_accuracy": 0.4706888128969223, | |
| "eval_loss": 2.48046875, | |
| "eval_runtime": 1.6588, | |
| "eval_samples_per_second": 7.234, | |
| "eval_steps_per_second": 1.206, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 2.1432, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.4714215925744993, | |
| "eval_loss": 2.470703125, | |
| "eval_runtime": 2.2121, | |
| "eval_samples_per_second": 5.425, | |
| "eval_steps_per_second": 0.904, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.4999999999999999e-05, | |
| "loss": 2.0822, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_accuracy": 0.47235792216251427, | |
| "eval_loss": 2.45703125, | |
| "eval_runtime": 1.2064, | |
| "eval_samples_per_second": 9.947, | |
| "eval_steps_per_second": 1.658, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 1.9056, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.47410845139228136, | |
| "eval_loss": 2.451171875, | |
| "eval_runtime": 2.1018, | |
| "eval_samples_per_second": 5.709, | |
| "eval_steps_per_second": 0.952, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 14, | |
| "total_flos": 821815148544.0, | |
| "train_loss": 2.3695591517857144, | |
| "train_runtime": 69.0598, | |
| "train_samples_per_second": 1.448, | |
| "train_steps_per_second": 0.203 | |
| } | |
| ], | |
| "max_steps": 14, | |
| "num_train_epochs": 2, | |
| "total_flos": 821815148544.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |