| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9995835068721366, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 11.1022, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 10.5874, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.3e-06, | |
| "loss": 9.359, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.8e-06, | |
| "loss": 7.4948, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.3e-06, | |
| "loss": 5.7244, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 4.1675, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.3e-06, | |
| "loss": 3.2211, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3.8e-06, | |
| "loss": 2.4147, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.2999999999999995e-06, | |
| "loss": 1.7997, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 1.4401, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5.3e-06, | |
| "loss": 1.2247, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5.8e-06, | |
| "loss": 1.0594, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 6.300000000000001e-06, | |
| "loss": 0.9899, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 0.8842, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 7.2999999999999996e-06, | |
| "loss": 0.8798, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 7.8e-06, | |
| "loss": 0.8872, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 8.3e-06, | |
| "loss": 0.8889, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.9344, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 9.3e-06, | |
| "loss": 0.9867, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 0.8925, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.03e-05, | |
| "loss": 0.7869, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.08e-05, | |
| "loss": 0.8847, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.13e-05, | |
| "loss": 0.8221, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.18e-05, | |
| "loss": 0.8611, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.23e-05, | |
| "loss": 0.8544, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 0.8061, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3300000000000001e-05, | |
| "loss": 0.7984, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.3800000000000002e-05, | |
| "loss": 0.7396, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.43e-05, | |
| "loss": 0.8653, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.48e-05, | |
| "loss": 0.8675, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.8561133146286011, | |
| "eval_runtime": 33.8169, | |
| "eval_samples_per_second": 47.343, | |
| "eval_steps_per_second": 7.895, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 3763911720960000.0, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |