| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.487562189054726, | |
| "eval_steps": 25, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.43993993993994e-05, | |
| "loss": 0.8106, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.2529616057872772, | |
| "eval_runtime": 9.2377, | |
| "eval_samples_per_second": 9.634, | |
| "eval_steps_per_second": 1.299, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.3773773773773775e-05, | |
| "loss": 0.1936, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.1928737312555313, | |
| "eval_runtime": 9.3374, | |
| "eval_samples_per_second": 9.532, | |
| "eval_steps_per_second": 1.285, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.1423, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.18249212205410004, | |
| "eval_runtime": 9.3738, | |
| "eval_samples_per_second": 9.495, | |
| "eval_steps_per_second": 1.28, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.2522522522522523e-05, | |
| "loss": 0.1499, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.17995011806488037, | |
| "eval_runtime": 9.4057, | |
| "eval_samples_per_second": 9.462, | |
| "eval_steps_per_second": 1.276, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 2.18968968968969e-05, | |
| "loss": 0.1177, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.174322709441185, | |
| "eval_runtime": 9.4044, | |
| "eval_samples_per_second": 9.464, | |
| "eval_steps_per_second": 1.276, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.1271271271271275e-05, | |
| "loss": 0.128, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.1779523640871048, | |
| "eval_runtime": 9.4208, | |
| "eval_samples_per_second": 9.447, | |
| "eval_steps_per_second": 1.274, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 2.0645645645645647e-05, | |
| "loss": 0.1023, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.1758423000574112, | |
| "eval_runtime": 9.3522, | |
| "eval_samples_per_second": 9.517, | |
| "eval_steps_per_second": 1.283, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.0020020020020023e-05, | |
| "loss": 0.1145, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.17488867044448853, | |
| "eval_runtime": 9.397, | |
| "eval_samples_per_second": 9.471, | |
| "eval_steps_per_second": 1.277, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.9394394394394395e-05, | |
| "loss": 0.0945, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.1771743893623352, | |
| "eval_runtime": 9.2878, | |
| "eval_samples_per_second": 9.583, | |
| "eval_steps_per_second": 1.292, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.8768768768768768e-05, | |
| "loss": 0.0813, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 0.21932578086853027, | |
| "eval_runtime": 9.3376, | |
| "eval_samples_per_second": 9.531, | |
| "eval_steps_per_second": 1.285, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.8143143143143144e-05, | |
| "loss": 0.0898, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.17662294209003448, | |
| "eval_runtime": 9.3638, | |
| "eval_samples_per_second": 9.505, | |
| "eval_steps_per_second": 1.282, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.7517517517517516e-05, | |
| "loss": 0.0846, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.1968599408864975, | |
| "eval_runtime": 9.3136, | |
| "eval_samples_per_second": 9.556, | |
| "eval_steps_per_second": 1.288, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.6891891891891892e-05, | |
| "loss": 0.0743, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 0.19715578854084015, | |
| "eval_runtime": 9.4533, | |
| "eval_samples_per_second": 9.415, | |
| "eval_steps_per_second": 1.269, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.6266266266266268e-05, | |
| "loss": 0.0763, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 0.20438912510871887, | |
| "eval_runtime": 9.3395, | |
| "eval_samples_per_second": 9.529, | |
| "eval_steps_per_second": 1.285, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.564064064064064e-05, | |
| "loss": 0.0742, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.19478727877140045, | |
| "eval_runtime": 9.3251, | |
| "eval_samples_per_second": 9.544, | |
| "eval_steps_per_second": 1.287, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.5015015015015016e-05, | |
| "loss": 0.0642, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.21925467252731323, | |
| "eval_runtime": 9.3222, | |
| "eval_samples_per_second": 9.547, | |
| "eval_steps_per_second": 1.287, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.438938938938939e-05, | |
| "loss": 0.0595, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.2161593735218048, | |
| "eval_runtime": 9.317, | |
| "eval_samples_per_second": 9.552, | |
| "eval_steps_per_second": 1.288, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.3763763763763765e-05, | |
| "loss": 0.0549, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.216683030128479, | |
| "eval_runtime": 9.247, | |
| "eval_samples_per_second": 9.625, | |
| "eval_steps_per_second": 1.298, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.3138138138138139e-05, | |
| "loss": 0.051, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 0.2373329997062683, | |
| "eval_runtime": 8.9608, | |
| "eval_samples_per_second": 9.932, | |
| "eval_steps_per_second": 1.339, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.2512512512512515e-05, | |
| "loss": 0.0552, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 0.22366206347942352, | |
| "eval_runtime": 9.3515, | |
| "eval_samples_per_second": 9.517, | |
| "eval_steps_per_second": 1.283, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.1886886886886887e-05, | |
| "loss": 0.0551, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 0.20705640316009521, | |
| "eval_runtime": 9.3684, | |
| "eval_samples_per_second": 9.5, | |
| "eval_steps_per_second": 1.281, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.1261261261261261e-05, | |
| "loss": 0.0558, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.22326058149337769, | |
| "eval_runtime": 9.355, | |
| "eval_samples_per_second": 9.514, | |
| "eval_steps_per_second": 1.283, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.0635635635635637e-05, | |
| "loss": 0.0576, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 0.21783779561519623, | |
| "eval_runtime": 9.1966, | |
| "eval_samples_per_second": 9.677, | |
| "eval_steps_per_second": 1.305, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.0010010010010011e-05, | |
| "loss": 0.0534, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_loss": 0.2279473841190338, | |
| "eval_runtime": 9.1641, | |
| "eval_samples_per_second": 9.712, | |
| "eval_steps_per_second": 1.309, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 9.384384384384384e-06, | |
| "loss": 0.0531, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 0.230553537607193, | |
| "eval_runtime": 9.2393, | |
| "eval_samples_per_second": 9.633, | |
| "eval_steps_per_second": 1.299, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 8.758758758758758e-06, | |
| "loss": 0.0507, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 0.2528133690357208, | |
| "eval_runtime": 9.2368, | |
| "eval_samples_per_second": 9.635, | |
| "eval_steps_per_second": 1.299, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.133133133133134e-06, | |
| "loss": 0.0496, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 0.27711576223373413, | |
| "eval_runtime": 9.2578, | |
| "eval_samples_per_second": 9.614, | |
| "eval_steps_per_second": 1.296, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 7.507507507507508e-06, | |
| "loss": 0.0525, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_loss": 0.24864882230758667, | |
| "eval_runtime": 9.3261, | |
| "eval_samples_per_second": 9.543, | |
| "eval_steps_per_second": 1.287, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 6.881881881881882e-06, | |
| "loss": 0.0477, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 0.26756495237350464, | |
| "eval_runtime": 9.1971, | |
| "eval_samples_per_second": 9.677, | |
| "eval_steps_per_second": 1.305, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 6.256256256256257e-06, | |
| "loss": 0.0505, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 0.2570458650588989, | |
| "eval_runtime": 9.2232, | |
| "eval_samples_per_second": 9.65, | |
| "eval_steps_per_second": 1.301, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 5.630630630630631e-06, | |
| "loss": 0.0483, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.25569072365760803, | |
| "eval_runtime": 9.2668, | |
| "eval_samples_per_second": 9.604, | |
| "eval_steps_per_second": 1.295, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 5.005005005005006e-06, | |
| "loss": 0.0499, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_loss": 0.26184016466140747, | |
| "eval_runtime": 9.3556, | |
| "eval_samples_per_second": 9.513, | |
| "eval_steps_per_second": 1.283, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.379379379379379e-06, | |
| "loss": 0.0438, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.2792048454284668, | |
| "eval_runtime": 9.3539, | |
| "eval_samples_per_second": 9.515, | |
| "eval_steps_per_second": 1.283, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.753753753753754e-06, | |
| "loss": 0.0433, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 0.2763405740261078, | |
| "eval_runtime": 9.3987, | |
| "eval_samples_per_second": 9.469, | |
| "eval_steps_per_second": 1.277, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 3.1281281281281287e-06, | |
| "loss": 0.0438, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 0.28726398944854736, | |
| "eval_runtime": 9.3415, | |
| "eval_samples_per_second": 9.527, | |
| "eval_steps_per_second": 1.285, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.502502502502503e-06, | |
| "loss": 0.0439, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.29283300042152405, | |
| "eval_runtime": 9.3228, | |
| "eval_samples_per_second": 9.547, | |
| "eval_steps_per_second": 1.287, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.876876876876877e-06, | |
| "loss": 0.0443, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_loss": 0.28899624943733215, | |
| "eval_runtime": 9.3231, | |
| "eval_samples_per_second": 9.546, | |
| "eval_steps_per_second": 1.287, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1.2512512512512514e-06, | |
| "loss": 0.0458, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_loss": 0.28502869606018066, | |
| "eval_runtime": 9.3924, | |
| "eval_samples_per_second": 9.476, | |
| "eval_steps_per_second": 1.278, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 6.256256256256257e-07, | |
| "loss": 0.0431, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "eval_loss": 0.2905334532260895, | |
| "eval_runtime": 9.3998, | |
| "eval_samples_per_second": 9.468, | |
| "eval_steps_per_second": 1.277, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.0, | |
| "loss": 0.0427, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_loss": 0.2926941215991974, | |
| "eval_runtime": 9.3458, | |
| "eval_samples_per_second": 9.523, | |
| "eval_steps_per_second": 1.284, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 25, | |
| "total_flos": 2.1565516640256e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |