| { | |
| "best_metric": 0.8666666746139526, | |
| "best_model_checkpoint": "/content/best_model/checkpoint-3000", | |
| "epoch": 20.0, | |
| "global_step": 7020, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.8575498575498575e-05, | |
| "loss": 0.5673, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_accuracy": 0.8437275886535645, | |
| "eval_loss": 0.43583494424819946, | |
| "eval_runtime": 13.3841, | |
| "eval_samples_per_second": 104.228, | |
| "eval_steps_per_second": 3.287, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.7150997150997152e-05, | |
| "loss": 0.2898, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.8523297309875488, | |
| "eval_loss": 0.4844760596752167, | |
| "eval_runtime": 13.3926, | |
| "eval_samples_per_second": 104.162, | |
| "eval_steps_per_second": 3.285, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 1.5726495726495726e-05, | |
| "loss": 0.1669, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_accuracy": 0.8573476672172546, | |
| "eval_loss": 0.6232547760009766, | |
| "eval_runtime": 13.3901, | |
| "eval_samples_per_second": 104.181, | |
| "eval_steps_per_second": 3.286, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 1.4301994301994305e-05, | |
| "loss": 0.1087, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "eval_accuracy": 0.8573476672172546, | |
| "eval_loss": 0.7262628078460693, | |
| "eval_runtime": 13.3968, | |
| "eval_samples_per_second": 104.129, | |
| "eval_steps_per_second": 3.284, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 1.2877492877492879e-05, | |
| "loss": 0.0728, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_accuracy": 0.8637992739677429, | |
| "eval_loss": 0.8840554356575012, | |
| "eval_runtime": 13.3781, | |
| "eval_samples_per_second": 104.275, | |
| "eval_steps_per_second": 3.289, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 1.1452991452991454e-05, | |
| "loss": 0.0512, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "eval_accuracy": 0.8666666746139526, | |
| "eval_loss": 0.9500740170478821, | |
| "eval_runtime": 13.4339, | |
| "eval_samples_per_second": 103.842, | |
| "eval_steps_per_second": 3.275, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "learning_rate": 1.002849002849003e-05, | |
| "loss": 0.0372, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "eval_accuracy": 0.856630802154541, | |
| "eval_loss": 1.0440385341644287, | |
| "eval_runtime": 13.391, | |
| "eval_samples_per_second": 104.175, | |
| "eval_steps_per_second": 3.286, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 8.603988603988605e-06, | |
| "loss": 0.0262, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "eval_accuracy": 0.8609318733215332, | |
| "eval_loss": 1.0770107507705688, | |
| "eval_runtime": 13.3894, | |
| "eval_samples_per_second": 104.187, | |
| "eval_steps_per_second": 3.286, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 0.0243, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "eval_accuracy": 0.8616487383842468, | |
| "eval_loss": 1.0931395292282104, | |
| "eval_runtime": 13.3865, | |
| "eval_samples_per_second": 104.209, | |
| "eval_steps_per_second": 3.287, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 5.7549857549857555e-06, | |
| "loss": 0.023, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "eval_accuracy": 0.8630824089050293, | |
| "eval_loss": 1.1087766885757446, | |
| "eval_runtime": 13.3932, | |
| "eval_samples_per_second": 104.157, | |
| "eval_steps_per_second": 3.285, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.67, | |
| "learning_rate": 4.330484330484331e-06, | |
| "loss": 0.0163, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.67, | |
| "eval_accuracy": 0.8580645322799683, | |
| "eval_loss": 1.1263514757156372, | |
| "eval_runtime": 13.3902, | |
| "eval_samples_per_second": 104.18, | |
| "eval_steps_per_second": 3.286, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 0.0111, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "eval_accuracy": 0.8616487383842468, | |
| "eval_loss": 1.154114842414856, | |
| "eval_runtime": 13.3822, | |
| "eval_samples_per_second": 104.243, | |
| "eval_steps_per_second": 3.288, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 1.4814814814814815e-06, | |
| "loss": 0.0098, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "eval_accuracy": 0.8630824089050293, | |
| "eval_loss": 1.1541603803634644, | |
| "eval_runtime": 13.3814, | |
| "eval_samples_per_second": 104.25, | |
| "eval_steps_per_second": 3.288, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 19.94, | |
| "learning_rate": 5.6980056980056986e-08, | |
| "loss": 0.0074, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 19.94, | |
| "eval_accuracy": 0.8637992739677429, | |
| "eval_loss": 1.1652880907058716, | |
| "eval_runtime": 13.3866, | |
| "eval_samples_per_second": 104.209, | |
| "eval_steps_per_second": 3.287, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 7020, | |
| "total_flos": 3.4632303608448e+16, | |
| "train_loss": 0.10057472327884626, | |
| "train_runtime": 6733.6679, | |
| "train_samples_per_second": 33.361, | |
| "train_steps_per_second": 1.043 | |
| } | |
| ], | |
| "max_steps": 7020, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.4632303608448e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |