| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.953846153846154, | |
| "eval_steps": 500, | |
| "global_step": 144, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4578, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.9936215093023884e-05, | |
| "loss": 0.5148, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.974567407496712e-05, | |
| "loss": 0.3974, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.9430807674052092e-05, | |
| "loss": 0.3542, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.899563263509725e-05, | |
| "loss": 0.3317, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.8445700477978207e-05, | |
| "loss": 0.3224, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.778802667699196e-05, | |
| "loss": 0.3048, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.7031001164581828e-05, | |
| "loss": 0.2939, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.618428130112533e-05, | |
| "loss": 0.2771, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.5258668676167548e-05, | |
| "loss": 0.2667, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1.4265971312744252e-05, | |
| "loss": 0.2237, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.3218853032651719e-05, | |
| "loss": 0.2101, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.2130671904307692e-05, | |
| "loss": 0.2087, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.1015309834121083e-05, | |
| "loss": 0.2125, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 9.886995475270205e-06, | |
| "loss": 0.2062, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 8.76012271303888e-06, | |
| "loss": 0.2046, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 7.649067042289681e-06, | |
| "loss": 0.2009, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 6.568002179543409e-06, | |
| "loss": 0.2015, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 5.530719249141148e-06, | |
| "loss": 0.1915, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.550450850127626e-06, | |
| "loss": 0.1746, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 3.6397022482313804e-06, | |
| "loss": 0.1439, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 2.8100918464225304e-06, | |
| "loss": 0.1338, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.072202969162234e-06, | |
| "loss": 0.1347, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.4354488511294418e-06, | |
| "loss": 0.1338, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.079525527612321e-07, | |
| "loss": 0.1316, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4.964433345219354e-07, | |
| "loss": 0.1303, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 2.0617081185259512e-07, | |
| "loss": 0.1291, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 4.083798592444899e-08, | |
| "loss": 0.1243, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "step": 144, | |
| "total_flos": 3.741283197648896e+17, | |
| "train_loss": 0.2684229459199641, | |
| "train_runtime": 6412.5843, | |
| "train_samples_per_second": 1.459, | |
| "train_steps_per_second": 0.022 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 144, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3.741283197648896e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |