| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.567398119122257, | |
| "eval_steps": 50, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2.3869346733668342e-05, | |
| "loss": 0.7793, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.27101635932922363, | |
| "eval_runtime": 141.469, | |
| "eval_samples_per_second": 5.047, | |
| "eval_steps_per_second": 0.636, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.2613065326633167e-05, | |
| "loss": 0.2453, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.21987898647785187, | |
| "eval_runtime": 142.8952, | |
| "eval_samples_per_second": 4.997, | |
| "eval_steps_per_second": 0.63, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.135678391959799e-05, | |
| "loss": 0.2079, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.19277334213256836, | |
| "eval_runtime": 142.7101, | |
| "eval_samples_per_second": 5.003, | |
| "eval_steps_per_second": 0.631, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 2.0100502512562815e-05, | |
| "loss": 0.1836, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.179254949092865, | |
| "eval_runtime": 143.1949, | |
| "eval_samples_per_second": 4.986, | |
| "eval_steps_per_second": 0.629, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.884422110552764e-05, | |
| "loss": 0.1762, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.1739482283592224, | |
| "eval_runtime": 142.441, | |
| "eval_samples_per_second": 5.013, | |
| "eval_steps_per_second": 0.632, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.7587939698492464e-05, | |
| "loss": 0.1692, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.16870561242103577, | |
| "eval_runtime": 141.3005, | |
| "eval_samples_per_second": 5.053, | |
| "eval_steps_per_second": 0.637, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.6331658291457288e-05, | |
| "loss": 0.1657, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.1669251173734665, | |
| "eval_runtime": 141.02, | |
| "eval_samples_per_second": 5.063, | |
| "eval_steps_per_second": 0.638, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.507537688442211e-05, | |
| "loss": 0.1681, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.1648036390542984, | |
| "eval_runtime": 140.8855, | |
| "eval_samples_per_second": 5.068, | |
| "eval_steps_per_second": 0.639, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.3819095477386935e-05, | |
| "loss": 0.1576, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.16245362162590027, | |
| "eval_runtime": 141.4143, | |
| "eval_samples_per_second": 5.049, | |
| "eval_steps_per_second": 0.636, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.2562814070351759e-05, | |
| "loss": 0.1656, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 0.1608215868473053, | |
| "eval_runtime": 141.5918, | |
| "eval_samples_per_second": 5.043, | |
| "eval_steps_per_second": 0.636, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.1306532663316583e-05, | |
| "loss": 0.1517, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 0.1596228927373886, | |
| "eval_runtime": 142.7115, | |
| "eval_samples_per_second": 5.003, | |
| "eval_steps_per_second": 0.631, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.0050251256281408e-05, | |
| "loss": 0.1561, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 0.15880218148231506, | |
| "eval_runtime": 143.3712, | |
| "eval_samples_per_second": 4.98, | |
| "eval_steps_per_second": 0.628, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.793969849246232e-06, | |
| "loss": 0.1473, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.15745262801647186, | |
| "eval_runtime": 143.6929, | |
| "eval_samples_per_second": 4.969, | |
| "eval_steps_per_second": 0.626, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 7.537688442211055e-06, | |
| "loss": 0.1488, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.1574247181415558, | |
| "eval_runtime": 142.725, | |
| "eval_samples_per_second": 5.003, | |
| "eval_steps_per_second": 0.631, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.2814070351758795e-06, | |
| "loss": 0.1468, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 0.1565464437007904, | |
| "eval_runtime": 141.7942, | |
| "eval_samples_per_second": 5.035, | |
| "eval_steps_per_second": 0.635, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.025125628140704e-06, | |
| "loss": 0.1379, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 0.1558983027935028, | |
| "eval_runtime": 141.5337, | |
| "eval_samples_per_second": 5.045, | |
| "eval_steps_per_second": 0.636, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 3.7688442211055276e-06, | |
| "loss": 0.1414, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.15601229667663574, | |
| "eval_runtime": 141.5594, | |
| "eval_samples_per_second": 5.044, | |
| "eval_steps_per_second": 0.636, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.512562814070352e-06, | |
| "loss": 0.1377, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 0.15548652410507202, | |
| "eval_runtime": 141.6005, | |
| "eval_samples_per_second": 5.042, | |
| "eval_steps_per_second": 0.636, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.256281407035176e-06, | |
| "loss": 0.1463, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_loss": 0.1545386165380478, | |
| "eval_runtime": 141.6136, | |
| "eval_samples_per_second": 5.042, | |
| "eval_steps_per_second": 0.636, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0, | |
| "loss": 0.1441, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 0.15440967679023743, | |
| "eval_runtime": 142.3851, | |
| "eval_samples_per_second": 5.015, | |
| "eval_steps_per_second": 0.632, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 50, | |
| "total_flos": 1.7525216609776435e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |