| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.999725902238465, | |
| "eval_steps": 2000, | |
| "global_step": 16416, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.8477095516569206e-05, | |
| "loss": 1.4175, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.69541910331384e-05, | |
| "loss": 1.3652, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.54312865497076e-05, | |
| "loss": 1.3497, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.39083820662768e-05, | |
| "loss": 1.3443, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 1.2274980545043945, | |
| "eval_runtime": 4.5329, | |
| "eval_samples_per_second": 22.061, | |
| "eval_steps_per_second": 5.515, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.238547758284601e-05, | |
| "loss": 1.3359, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.086257309941521e-05, | |
| "loss": 1.3317, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.933966861598441e-05, | |
| "loss": 1.322, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3.7816764132553604e-05, | |
| "loss": 1.3148, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.2038463354110718, | |
| "eval_runtime": 4.5544, | |
| "eval_samples_per_second": 21.957, | |
| "eval_steps_per_second": 5.489, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.629385964912281e-05, | |
| "loss": 1.3261, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 3.477095516569201e-05, | |
| "loss": 1.3198, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 3.3248050682261215e-05, | |
| "loss": 1.3166, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.172514619883041e-05, | |
| "loss": 1.2259, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 1.1959730386734009, | |
| "eval_runtime": 4.5591, | |
| "eval_samples_per_second": 21.934, | |
| "eval_steps_per_second": 5.484, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.0202241715399608e-05, | |
| "loss": 1.2181, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.8679337231968812e-05, | |
| "loss": 1.2247, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 2.7156432748538012e-05, | |
| "loss": 1.2228, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.5633528265107216e-05, | |
| "loss": 1.2271, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.1902621984481812, | |
| "eval_runtime": 4.5576, | |
| "eval_samples_per_second": 21.941, | |
| "eval_steps_per_second": 5.485, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.4110623781676416e-05, | |
| "loss": 1.2324, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.2587719298245616e-05, | |
| "loss": 1.219, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.1064814814814816e-05, | |
| "loss": 1.2294, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.9541910331384016e-05, | |
| "loss": 1.2224, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 1.1794958114624023, | |
| "eval_runtime": 4.5594, | |
| "eval_samples_per_second": 21.933, | |
| "eval_steps_per_second": 5.483, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.8019005847953216e-05, | |
| "loss": 1.2206, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.6496101364522417e-05, | |
| "loss": 1.2081, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.497319688109162e-05, | |
| "loss": 1.165, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.3450292397660819e-05, | |
| "loss": 1.166, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.1834660768508911, | |
| "eval_runtime": 4.5566, | |
| "eval_samples_per_second": 21.946, | |
| "eval_steps_per_second": 5.487, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.192738791423002e-05, | |
| "loss": 1.1471, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.040448343079922e-05, | |
| "loss": 1.1654, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 8.881578947368421e-06, | |
| "loss": 1.1517, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.358674463937622e-06, | |
| "loss": 1.1599, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 1.1791805028915405, | |
| "eval_runtime": 4.5397, | |
| "eval_samples_per_second": 22.028, | |
| "eval_steps_per_second": 5.507, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 5.835769980506823e-06, | |
| "loss": 1.1621, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 4.312865497076023e-06, | |
| "loss": 1.1725, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 2.7899610136452243e-06, | |
| "loss": 1.1607, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.2670565302144249e-06, | |
| "loss": 1.1644, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 1.1775761842727661, | |
| "eval_runtime": 4.5403, | |
| "eval_samples_per_second": 22.025, | |
| "eval_steps_per_second": 5.506, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 16416, | |
| "total_flos": 6.325107751849329e+17, | |
| "train_loss": 1.2418343523557192, | |
| "train_runtime": 36070.0373, | |
| "train_samples_per_second": 7.282, | |
| "train_steps_per_second": 0.455 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16416, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6.325107751849329e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |