| { | |
| "best_metric": 0.9938686490058899, | |
| "best_model_checkpoint": "model_result/01_01_llama3.1_8B_SFT/checkpoint-4595", | |
| "epoch": 4.999320005439957, | |
| "eval_steps": 919, | |
| "global_step": 4595, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9998640010879913, | |
| "grad_norm": 0.5976972579956055, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.1291, | |
| "step": 919 | |
| }, | |
| { | |
| "epoch": 0.9998640010879913, | |
| "eval_loss": 1.044102430343628, | |
| "eval_runtime": 358.9496, | |
| "eval_samples_per_second": 2.613, | |
| "eval_steps_per_second": 0.655, | |
| "step": 919 | |
| }, | |
| { | |
| "epoch": 1.9997280021759827, | |
| "grad_norm": 0.47506141662597656, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.0116, | |
| "step": 1838 | |
| }, | |
| { | |
| "epoch": 1.9997280021759827, | |
| "eval_loss": 1.0129140615463257, | |
| "eval_runtime": 358.903, | |
| "eval_samples_per_second": 2.614, | |
| "eval_steps_per_second": 0.655, | |
| "step": 1838 | |
| }, | |
| { | |
| "epoch": 2.999592003263974, | |
| "grad_norm": 0.5260523557662964, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9776, | |
| "step": 2757 | |
| }, | |
| { | |
| "epoch": 2.999592003263974, | |
| "eval_loss": 1.0010725259780884, | |
| "eval_runtime": 358.7054, | |
| "eval_samples_per_second": 2.615, | |
| "eval_steps_per_second": 0.655, | |
| "step": 2757 | |
| }, | |
| { | |
| "epoch": 3.9994560043519654, | |
| "grad_norm": 0.5811532139778137, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9537, | |
| "step": 3676 | |
| }, | |
| { | |
| "epoch": 3.9994560043519654, | |
| "eval_loss": 0.9950182437896729, | |
| "eval_runtime": 358.8496, | |
| "eval_samples_per_second": 2.614, | |
| "eval_steps_per_second": 0.655, | |
| "step": 3676 | |
| }, | |
| { | |
| "epoch": 4.999320005439957, | |
| "grad_norm": 0.5733142495155334, | |
| "learning_rate": 0.0, | |
| "loss": 0.936, | |
| "step": 4595 | |
| }, | |
| { | |
| "epoch": 4.999320005439957, | |
| "eval_loss": 0.9938686490058899, | |
| "eval_runtime": 358.1374, | |
| "eval_samples_per_second": 2.619, | |
| "eval_steps_per_second": 0.656, | |
| "step": 4595 | |
| } | |
| ], | |
| "logging_steps": 919, | |
| "max_steps": 4595, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 919, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3192927949748437e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |