| { |
| "best_metric": 2.9821596145629883, |
| "best_model_checkpoint": "output/ghost/checkpoint-76", |
| "epoch": 4.0, |
| "global_step": 76, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.26, |
| "learning_rate": 0.00011506151581352585, |
| "loss": 3.8815, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.53, |
| "learning_rate": 6.293505690059801e-05, |
| "loss": 3.9129, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.79, |
| "learning_rate": 1.4464961055407408e-05, |
| "loss": 3.5855, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.072023868560791, |
| "eval_runtime": 1.4457, |
| "eval_samples_per_second": 21.444, |
| "eval_steps_per_second": 2.767, |
| "step": 19 |
| }, |
| { |
| "epoch": 1.05, |
| "learning_rate": 9.35614586573241e-07, |
| "loss": 3.6273, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.32, |
| "learning_rate": 3.1079356352801514e-05, |
| "loss": 3.2682, |
| "step": 25 |
| }, |
| { |
| "epoch": 1.58, |
| "learning_rate": 8.54403044178588e-05, |
| "loss": 3.8264, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.84, |
| "learning_rate": 0.00012893189933276512, |
| "loss": 3.2622, |
| "step": 35 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 3.008462905883789, |
| "eval_runtime": 1.4442, |
| "eval_samples_per_second": 21.466, |
| "eval_steps_per_second": 2.77, |
| "step": 38 |
| }, |
| { |
| "epoch": 2.11, |
| "learning_rate": 0.00013348306278066356, |
| "loss": 3.1918, |
| "step": 40 |
| }, |
| { |
| "epoch": 2.37, |
| "learning_rate": 9.615630613119375e-05, |
| "loss": 3.2537, |
| "step": 45 |
| }, |
| { |
| "epoch": 2.63, |
| "learning_rate": 4.1043693868806304e-05, |
| "loss": 3.0971, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.89, |
| "learning_rate": 3.7169372193364486e-06, |
| "loss": 2.9811, |
| "step": 55 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 2.9955391883850098, |
| "eval_runtime": 1.4389, |
| "eval_samples_per_second": 21.544, |
| "eval_steps_per_second": 2.78, |
| "step": 57 |
| }, |
| { |
| "epoch": 3.16, |
| "learning_rate": 8.268100667234827e-06, |
| "loss": 2.811, |
| "step": 60 |
| }, |
| { |
| "epoch": 3.42, |
| "learning_rate": 5.175969558214116e-05, |
| "loss": 3.1649, |
| "step": 65 |
| }, |
| { |
| "epoch": 3.68, |
| "learning_rate": 0.0001061206436471984, |
| "loss": 2.7835, |
| "step": 70 |
| }, |
| { |
| "epoch": 3.95, |
| "learning_rate": 0.00013626438541342674, |
| "loss": 2.9919, |
| "step": 75 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 2.9821596145629883, |
| "eval_runtime": 1.4488, |
| "eval_samples_per_second": 21.397, |
| "eval_steps_per_second": 2.761, |
| "step": 76 |
| } |
| ], |
| "max_steps": 380, |
| "num_train_epochs": 20, |
| "total_flos": 76297273344000.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
