| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.444444444444445, |
| "eval_steps": 500, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 0.45093148946762085, |
| "learning_rate": 2.45e-05, |
| "loss": 0.3659, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 0.3078092932701111, |
| "learning_rate": 4.9500000000000004e-05, |
| "loss": 0.2591, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 0.2012849748134613, |
| "learning_rate": 4.727777777777778e-05, |
| "loss": 0.1304, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 0.15963983535766602, |
| "learning_rate": 4.4500000000000004e-05, |
| "loss": 0.0813, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.16948182880878448, |
| "learning_rate": 4.172222222222222e-05, |
| "loss": 0.0622, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.17727190256118774, |
| "learning_rate": 3.894444444444444e-05, |
| "loss": 0.0557, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 0.13764996826648712, |
| "learning_rate": 3.6166666666666674e-05, |
| "loss": 0.0473, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.17411553859710693, |
| "learning_rate": 3.338888888888889e-05, |
| "loss": 0.0396, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.1547096222639084, |
| "learning_rate": 3.061111111111111e-05, |
| "loss": 0.0375, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 0.12937426567077637, |
| "learning_rate": 2.7833333333333333e-05, |
| "loss": 0.0369, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 0.14638905227184296, |
| "learning_rate": 2.5055555555555555e-05, |
| "loss": 0.0332, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.14906126260757446, |
| "learning_rate": 2.2277777777777778e-05, |
| "loss": 0.0319, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.12663030624389648, |
| "learning_rate": 1.9500000000000003e-05, |
| "loss": 0.031, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.17405924201011658, |
| "learning_rate": 1.6722222222222222e-05, |
| "loss": 0.0294, |
| "step": 700 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.102853924036026, |
| "learning_rate": 1.3944444444444446e-05, |
| "loss": 0.0304, |
| "step": 750 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.34207192063331604, |
| "learning_rate": 1.1166666666666668e-05, |
| "loss": 0.0281, |
| "step": 800 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 0.1295415461063385, |
| "learning_rate": 8.38888888888889e-06, |
| "loss": 0.0266, |
| "step": 850 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.12049657851457596, |
| "learning_rate": 5.611111111111112e-06, |
| "loss": 0.0281, |
| "step": 900 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.1062990203499794, |
| "learning_rate": 2.8333333333333335e-06, |
| "loss": 0.0273, |
| "step": 950 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.125990629196167, |
| "learning_rate": 5.555555555555556e-08, |
| "loss": 0.0266, |
| "step": 1000 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "step": 1000, |
| "total_flos": 1.099960418304e+16, |
| "train_loss": 0.07043375110626221, |
| "train_runtime": 1202.1987, |
| "train_samples_per_second": 13.309, |
| "train_steps_per_second": 0.832 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.099960418304e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
