| { | |
| "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": 0.7797, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.2723180055618286, | |
| "eval_runtime": 135.6616, | |
| "eval_samples_per_second": 5.263, | |
| "eval_steps_per_second": 0.663, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.2613065326633167e-05, | |
| "loss": 0.2457, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.22004182636737823, | |
| "eval_runtime": 136.2348, | |
| "eval_samples_per_second": 5.241, | |
| "eval_steps_per_second": 0.661, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.135678391959799e-05, | |
| "loss": 0.2088, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.19266444444656372, | |
| "eval_runtime": 136.6465, | |
| "eval_samples_per_second": 5.225, | |
| "eval_steps_per_second": 0.659, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 2.0100502512562815e-05, | |
| "loss": 0.1832, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.17922177910804749, | |
| "eval_runtime": 136.7121, | |
| "eval_samples_per_second": 5.223, | |
| "eval_steps_per_second": 0.658, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.884422110552764e-05, | |
| "loss": 0.1754, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.17311859130859375, | |
| "eval_runtime": 136.3058, | |
| "eval_samples_per_second": 5.238, | |
| "eval_steps_per_second": 0.66, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.7587939698492464e-05, | |
| "loss": 0.169, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.16897280514240265, | |
| "eval_runtime": 136.923, | |
| "eval_samples_per_second": 5.215, | |
| "eval_steps_per_second": 0.657, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.6331658291457288e-05, | |
| "loss": 0.166, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.1663457602262497, | |
| "eval_runtime": 136.6033, | |
| "eval_samples_per_second": 5.227, | |
| "eval_steps_per_second": 0.659, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.507537688442211e-05, | |
| "loss": 0.1682, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.16482460498809814, | |
| "eval_runtime": 136.5801, | |
| "eval_samples_per_second": 5.228, | |
| "eval_steps_per_second": 0.659, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.3819095477386935e-05, | |
| "loss": 0.1576, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.16245244443416595, | |
| "eval_runtime": 136.7662, | |
| "eval_samples_per_second": 5.221, | |
| "eval_steps_per_second": 0.658, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.2562814070351759e-05, | |
| "loss": 0.165, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 0.16068558394908905, | |
| "eval_runtime": 136.6019, | |
| "eval_samples_per_second": 5.227, | |
| "eval_steps_per_second": 0.659, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.1306532663316583e-05, | |
| "loss": 0.152, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 0.15984833240509033, | |
| "eval_runtime": 136.8975, | |
| "eval_samples_per_second": 5.216, | |
| "eval_steps_per_second": 0.657, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.0050251256281408e-05, | |
| "loss": 0.1563, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 0.15865428745746613, | |
| "eval_runtime": 136.9521, | |
| "eval_samples_per_second": 5.214, | |
| "eval_steps_per_second": 0.657, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.793969849246232e-06, | |
| "loss": 0.1477, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.1577940434217453, | |
| "eval_runtime": 136.5669, | |
| "eval_samples_per_second": 5.228, | |
| "eval_steps_per_second": 0.659, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 7.537688442211055e-06, | |
| "loss": 0.1491, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.157754048705101, | |
| "eval_runtime": 136.107, | |
| "eval_samples_per_second": 5.246, | |
| "eval_steps_per_second": 0.661, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.2814070351758795e-06, | |
| "loss": 0.1466, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 0.1569654941558838, | |
| "eval_runtime": 137.1916, | |
| "eval_samples_per_second": 5.204, | |
| "eval_steps_per_second": 0.656, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.025125628140704e-06, | |
| "loss": 0.1383, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 0.15617845952510834, | |
| "eval_runtime": 136.7366, | |
| "eval_samples_per_second": 5.222, | |
| "eval_steps_per_second": 0.658, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 3.7688442211055276e-06, | |
| "loss": 0.1417, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.15615858137607574, | |
| "eval_runtime": 136.2828, | |
| "eval_samples_per_second": 5.239, | |
| "eval_steps_per_second": 0.66, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.512562814070352e-06, | |
| "loss": 0.1374, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 0.155540332198143, | |
| "eval_runtime": 137.0904, | |
| "eval_samples_per_second": 5.208, | |
| "eval_steps_per_second": 0.657, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.256281407035176e-06, | |
| "loss": 0.147, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_loss": 0.15468443930149078, | |
| "eval_runtime": 136.9218, | |
| "eval_samples_per_second": 5.215, | |
| "eval_steps_per_second": 0.657, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0, | |
| "loss": 0.1448, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 0.15455935895442963, | |
| "eval_runtime": 136.7415, | |
| "eval_samples_per_second": 5.222, | |
| "eval_steps_per_second": 0.658, | |
| "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 | |
| } | |