| { | |
| "best_metric": 0.2851985559566787, | |
| "best_model_checkpoint": "results/checkpoint-7944", | |
| "epoch": 4.0, | |
| "global_step": 7944, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.96012084592145e-05, | |
| "loss": 2.8915, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9202416918429007e-05, | |
| "loss": 2.4239, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.8803625377643506e-05, | |
| "loss": 2.2481, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.840483383685801e-05, | |
| "loss": 2.1996, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.800604229607251e-05, | |
| "loss": 2.1824, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.760725075528701e-05, | |
| "loss": 2.0517, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.7208459214501513e-05, | |
| "loss": 2.0375, | |
| "step": 1386 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.6809667673716012e-05, | |
| "loss": 2.0427, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.6410876132930515e-05, | |
| "loss": 1.967, | |
| "step": 1782 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.6012084592145018e-05, | |
| "loss": 1.982, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.2740715098165274, | |
| "eval_loss": 2.498654365539551, | |
| "eval_runtime": 49.553, | |
| "eval_samples_per_second": 408.068, | |
| "eval_steps_per_second": 40.825, | |
| "step": 1986 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.5613293051359517e-05, | |
| "loss": 1.8129, | |
| "step": 2178 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.521450151057402e-05, | |
| "loss": 1.7615, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.481570996978852e-05, | |
| "loss": 1.82, | |
| "step": 2574 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.4416918429003021e-05, | |
| "loss": 1.7927, | |
| "step": 2772 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.4018126888217525e-05, | |
| "loss": 1.7828, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.3619335347432026e-05, | |
| "loss": 1.7809, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.3220543806646526e-05, | |
| "loss": 1.7006, | |
| "step": 3366 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.2821752265861027e-05, | |
| "loss": 1.7561, | |
| "step": 3564 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.242296072507553e-05, | |
| "loss": 1.7157, | |
| "step": 3762 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.2024169184290032e-05, | |
| "loss": 1.7302, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.27822560704218385, | |
| "eval_loss": 2.486177444458008, | |
| "eval_runtime": 49.6405, | |
| "eval_samples_per_second": 407.349, | |
| "eval_steps_per_second": 40.753, | |
| "step": 3972 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.1625377643504531e-05, | |
| "loss": 1.5066, | |
| "step": 4158 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.1226586102719035e-05, | |
| "loss": 1.5125, | |
| "step": 4356 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.0827794561933536e-05, | |
| "loss": 1.5064, | |
| "step": 4554 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.0429003021148037e-05, | |
| "loss": 1.5471, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1.003021148036254e-05, | |
| "loss": 1.5293, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 9.63141993957704e-06, | |
| "loss": 1.5216, | |
| "step": 5148 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 9.232628398791542e-06, | |
| "loss": 1.6028, | |
| "step": 5346 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 8.833836858006042e-06, | |
| "loss": 1.5475, | |
| "step": 5544 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.435045317220545e-06, | |
| "loss": 1.506, | |
| "step": 5742 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 8.036253776435046e-06, | |
| "loss": 1.5518, | |
| "step": 5940 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.2739231492013254, | |
| "eval_loss": 2.57331919670105, | |
| "eval_runtime": 49.6582, | |
| "eval_samples_per_second": 407.204, | |
| "eval_steps_per_second": 40.739, | |
| "step": 5958 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 7.637462235649547e-06, | |
| "loss": 1.3584, | |
| "step": 6138 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 7.238670694864049e-06, | |
| "loss": 1.3197, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 6.83987915407855e-06, | |
| "loss": 1.3561, | |
| "step": 6534 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 6.441087613293052e-06, | |
| "loss": 1.3575, | |
| "step": 6732 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 6.042296072507553e-06, | |
| "loss": 1.3798, | |
| "step": 6930 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.643504531722055e-06, | |
| "loss": 1.355, | |
| "step": 7128 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 5.2447129909365566e-06, | |
| "loss": 1.3312, | |
| "step": 7326 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 4.8459214501510575e-06, | |
| "loss": 1.4011, | |
| "step": 7524 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 4.447129909365559e-06, | |
| "loss": 1.4124, | |
| "step": 7722 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 4.048338368580061e-06, | |
| "loss": 1.3597, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.2851985559566787, | |
| "eval_loss": 2.571429491043091, | |
| "eval_runtime": 49.4476, | |
| "eval_samples_per_second": 408.938, | |
| "eval_steps_per_second": 40.912, | |
| "step": 7944 | |
| } | |
| ], | |
| "max_steps": 9930, | |
| "num_train_epochs": 5, | |
| "total_flos": 2302111312251120.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |