| { | |
| "best_metric": 0.07661789655685425, | |
| "best_model_checkpoint": "./teapotllm/checkpoint-2050", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2050, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.300591230392456, | |
| "learning_rate": 4e-05, | |
| "loss": 6.6074, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.08907566964626312, | |
| "eval_runtime": 4.8368, | |
| "eval_samples_per_second": 45.484, | |
| "eval_steps_per_second": 5.789, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.17182353138923645, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0881, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.07914955914020538, | |
| "eval_runtime": 4.8507, | |
| "eval_samples_per_second": 45.354, | |
| "eval_steps_per_second": 5.772, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.3242274522781372, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0722, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.07681724429130554, | |
| "eval_runtime": 4.8568, | |
| "eval_samples_per_second": 45.297, | |
| "eval_steps_per_second": 5.765, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.10380151122808456, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0637, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.07662593573331833, | |
| "eval_runtime": 4.8543, | |
| "eval_samples_per_second": 45.321, | |
| "eval_steps_per_second": 5.768, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.12985558807849884, | |
| "learning_rate": 0.0, | |
| "loss": 0.0595, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.07661789655685425, | |
| "eval_runtime": 4.8496, | |
| "eval_samples_per_second": 45.364, | |
| "eval_steps_per_second": 5.774, | |
| "step": 2050 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2050, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2807505145036800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |