| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 100.0, |
| "eval_steps": 20, |
| "global_step": 100, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 10.0, |
| "grad_norm": 40.950557708740234, |
| "learning_rate": 4.55e-05, |
| "loss": 18.8161, |
| "step": 10 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 35.285560607910156, |
| "learning_rate": 4.05e-05, |
| "loss": 14.196, |
| "step": 20 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 6.419579029083252, |
| "eval_runtime": 0.0679, |
| "eval_samples_per_second": 14.729, |
| "eval_steps_per_second": 14.729, |
| "step": 20 |
| }, |
| { |
| "epoch": 30.0, |
| "grad_norm": 29.6579647064209, |
| "learning_rate": 3.55e-05, |
| "loss": 11.7946, |
| "step": 30 |
| }, |
| { |
| "epoch": 40.0, |
| "grad_norm": 30.451223373413086, |
| "learning_rate": 3.05e-05, |
| "loss": 10.2795, |
| "step": 40 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_loss": 5.401611804962158, |
| "eval_runtime": 0.0671, |
| "eval_samples_per_second": 14.913, |
| "eval_steps_per_second": 14.913, |
| "step": 40 |
| }, |
| { |
| "epoch": 50.0, |
| "grad_norm": 28.49588394165039, |
| "learning_rate": 2.5500000000000003e-05, |
| "loss": 9.029, |
| "step": 50 |
| }, |
| { |
| "epoch": 60.0, |
| "grad_norm": 26.983068466186523, |
| "learning_rate": 2.05e-05, |
| "loss": 8.0003, |
| "step": 60 |
| }, |
| { |
| "epoch": 60.0, |
| "eval_loss": 4.586005687713623, |
| "eval_runtime": 0.0293, |
| "eval_samples_per_second": 34.155, |
| "eval_steps_per_second": 34.155, |
| "step": 60 |
| }, |
| { |
| "epoch": 70.0, |
| "grad_norm": 25.09807014465332, |
| "learning_rate": 1.55e-05, |
| "loss": 7.1517, |
| "step": 70 |
| }, |
| { |
| "epoch": 80.0, |
| "grad_norm": 25.080015182495117, |
| "learning_rate": 1.05e-05, |
| "loss": 6.2983, |
| "step": 80 |
| }, |
| { |
| "epoch": 80.0, |
| "eval_loss": 3.923098087310791, |
| "eval_runtime": 0.0442, |
| "eval_samples_per_second": 22.645, |
| "eval_steps_per_second": 22.645, |
| "step": 80 |
| }, |
| { |
| "epoch": 90.0, |
| "grad_norm": 22.326396942138672, |
| "learning_rate": 5.500000000000001e-06, |
| "loss": 5.9046, |
| "step": 90 |
| }, |
| { |
| "epoch": 100.0, |
| "grad_norm": 18.527441024780273, |
| "learning_rate": 5.000000000000001e-07, |
| "loss": 5.7314, |
| "step": 100 |
| }, |
| { |
| "epoch": 100.0, |
| "eval_loss": 3.8312597274780273, |
| "eval_runtime": 0.0315, |
| "eval_samples_per_second": 31.709, |
| "eval_steps_per_second": 31.709, |
| "step": 100 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 100, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 100, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 37178100940800.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
