| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.999384615384617, | |
| "global_step": 8120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.16827097535133362, | |
| "eval_mse": 0.1682709902524948, | |
| "eval_runtime": 6.9267, | |
| "eval_samples_per_second": 72.185, | |
| "eval_steps_per_second": 9.095, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.6705897491197778e-05, | |
| "loss": 0.473, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.12349746376276016, | |
| "eval_mse": 0.12349746376276016, | |
| "eval_runtime": 6.9169, | |
| "eval_samples_per_second": 72.287, | |
| "eval_steps_per_second": 9.108, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.5609709991775352e-05, | |
| "loss": 0.1309, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.12400069087743759, | |
| "eval_mse": 0.12400069087743759, | |
| "eval_runtime": 6.8578, | |
| "eval_samples_per_second": 72.91, | |
| "eval_steps_per_second": 9.187, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.4513522492352923e-05, | |
| "loss": 0.0981, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.099566251039505, | |
| "eval_mse": 0.099566251039505, | |
| "eval_runtime": 6.8428, | |
| "eval_samples_per_second": 73.07, | |
| "eval_steps_per_second": 9.207, | |
| "step": 1624 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.3417334992930497e-05, | |
| "loss": 0.0777, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.09356459975242615, | |
| "eval_mse": 0.09356459975242615, | |
| "eval_runtime": 6.8878, | |
| "eval_samples_per_second": 72.593, | |
| "eval_steps_per_second": 9.147, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.08973243087530136, | |
| "eval_mse": 0.08973242342472076, | |
| "eval_runtime": 6.8512, | |
| "eval_samples_per_second": 72.98, | |
| "eval_steps_per_second": 9.196, | |
| "step": 2436 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 1.2321147493508071e-05, | |
| "loss": 0.0627, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.09244009107351303, | |
| "eval_mse": 0.09244009107351303, | |
| "eval_runtime": 6.9051, | |
| "eval_samples_per_second": 72.41, | |
| "eval_steps_per_second": 9.124, | |
| "step": 2842 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 1.1224959994085646e-05, | |
| "loss": 0.0543, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.08615754544734955, | |
| "eval_mse": 0.08615754544734955, | |
| "eval_runtime": 6.8404, | |
| "eval_samples_per_second": 73.095, | |
| "eval_steps_per_second": 9.21, | |
| "step": 3248 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 1.012877249466322e-05, | |
| "loss": 0.0452, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.08058410882949829, | |
| "eval_mse": 0.08058410882949829, | |
| "eval_runtime": 6.8551, | |
| "eval_samples_per_second": 72.938, | |
| "eval_steps_per_second": 9.19, | |
| "step": 3654 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 9.032584995240794e-06, | |
| "loss": 0.0656, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.17249491810798645, | |
| "eval_mse": 0.17249493300914764, | |
| "eval_runtime": 6.8293, | |
| "eval_samples_per_second": 73.214, | |
| "eval_steps_per_second": 9.225, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.12468425184488297, | |
| "eval_mse": 0.12468423694372177, | |
| "eval_runtime": 6.9207, | |
| "eval_samples_per_second": 72.247, | |
| "eval_steps_per_second": 9.103, | |
| "step": 4466 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 7.936397495818367e-06, | |
| "loss": 0.0482, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.11367713660001755, | |
| "eval_mse": 0.11367714405059814, | |
| "eval_runtime": 6.9089, | |
| "eval_samples_per_second": 72.371, | |
| "eval_steps_per_second": 9.119, | |
| "step": 4872 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 6.84020999639594e-06, | |
| "loss": 0.0397, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.08477988839149475, | |
| "eval_mse": 0.08477988839149475, | |
| "eval_runtime": 6.9087, | |
| "eval_samples_per_second": 72.372, | |
| "eval_steps_per_second": 9.119, | |
| "step": 5278 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 5.744022496973514e-06, | |
| "loss": 0.0338, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.099708691239357, | |
| "eval_mse": 0.099708691239357, | |
| "eval_runtime": 6.9177, | |
| "eval_samples_per_second": 72.278, | |
| "eval_steps_per_second": 9.107, | |
| "step": 5684 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 4.647834997551088e-06, | |
| "loss": 0.0306, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.08894743025302887, | |
| "eval_mse": 0.08894742280244827, | |
| "eval_runtime": 6.8363, | |
| "eval_samples_per_second": 73.139, | |
| "eval_steps_per_second": 9.215, | |
| "step": 6090 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.09760861098766327, | |
| "eval_mse": 0.09760861098766327, | |
| "eval_runtime": 6.9072, | |
| "eval_samples_per_second": 72.388, | |
| "eval_steps_per_second": 9.121, | |
| "step": 6496 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 3.5516474981286615e-06, | |
| "loss": 0.0278, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.07102830708026886, | |
| "eval_mse": 0.07102830708026886, | |
| "eval_runtime": 6.8361, | |
| "eval_samples_per_second": 73.141, | |
| "eval_steps_per_second": 9.216, | |
| "step": 6902 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.455459998706235e-06, | |
| "loss": 0.0256, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.08258407562971115, | |
| "eval_mse": 0.08258406072854996, | |
| "eval_runtime": 6.9004, | |
| "eval_samples_per_second": 72.459, | |
| "eval_steps_per_second": 9.13, | |
| "step": 7308 | |
| }, | |
| { | |
| "epoch": 18.47, | |
| "learning_rate": 1.3592724992838087e-06, | |
| "loss": 0.0238, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.07976492494344711, | |
| "eval_mse": 0.07976492494344711, | |
| "eval_runtime": 6.8326, | |
| "eval_samples_per_second": 73.179, | |
| "eval_steps_per_second": 9.221, | |
| "step": 7714 | |
| }, | |
| { | |
| "epoch": 19.7, | |
| "learning_rate": 2.630849998613823e-07, | |
| "loss": 0.0223, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.07929830998182297, | |
| "eval_mse": 0.07929830998182297, | |
| "eval_runtime": 6.798, | |
| "eval_samples_per_second": 73.551, | |
| "eval_steps_per_second": 9.267, | |
| "step": 8120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 8120, | |
| "total_flos": 3.028673504394547e+16, | |
| "train_loss": 0.07788722984896505, | |
| "train_runtime": 7794.4594, | |
| "train_samples_per_second": 16.679, | |
| "train_steps_per_second": 1.042 | |
| } | |
| ], | |
| "max_steps": 8120, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.028673504394547e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |