| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 3180, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.9422912001609802, | |
| "learning_rate": 0.00044485348269869936, | |
| "loss": 0.2476, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7641935483870967, | |
| "eval_loss": 0.10522185266017914, | |
| "eval_runtime": 5.3982, | |
| "eval_samples_per_second": 574.263, | |
| "eval_steps_per_second": 12.041, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.2760656476020813, | |
| "learning_rate": 0.0003955979626094617, | |
| "loss": 0.0756, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.854516129032258, | |
| "eval_loss": 0.06962967664003372, | |
| "eval_runtime": 5.4277, | |
| "eval_samples_per_second": 571.144, | |
| "eval_steps_per_second": 11.976, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 1.5241812467575073, | |
| "learning_rate": 0.0003463424425202239, | |
| "loss": 0.0447, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8735483870967742, | |
| "eval_loss": 0.05896682292222977, | |
| "eval_runtime": 5.3747, | |
| "eval_samples_per_second": 576.777, | |
| "eval_steps_per_second": 12.094, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.3112484812736511, | |
| "learning_rate": 0.0002970869224309861, | |
| "loss": 0.0337, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9064516129032258, | |
| "eval_loss": 0.04424069821834564, | |
| "eval_runtime": 5.449, | |
| "eval_samples_per_second": 568.916, | |
| "eval_steps_per_second": 11.929, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.06658962368965149, | |
| "learning_rate": 0.00024783140234174836, | |
| "loss": 0.0262, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9090322580645162, | |
| "eval_loss": 0.04311687871813774, | |
| "eval_runtime": 5.4423, | |
| "eval_samples_per_second": 569.617, | |
| "eval_steps_per_second": 11.944, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.15654200315475464, | |
| "learning_rate": 0.0001985758822525106, | |
| "loss": 0.0223, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9141935483870968, | |
| "eval_loss": 0.03970247507095337, | |
| "eval_runtime": 5.399, | |
| "eval_samples_per_second": 574.182, | |
| "eval_steps_per_second": 12.039, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.06671025604009628, | |
| "learning_rate": 0.00014932036216327283, | |
| "loss": 0.0194, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.917741935483871, | |
| "eval_loss": 0.03794796019792557, | |
| "eval_runtime": 5.3981, | |
| "eval_samples_per_second": 574.279, | |
| "eval_steps_per_second": 12.041, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.06658264994621277, | |
| "learning_rate": 0.00010006484207403508, | |
| "loss": 0.0179, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9232258064516129, | |
| "eval_loss": 0.03362081199884415, | |
| "eval_runtime": 5.4142, | |
| "eval_samples_per_second": 572.564, | |
| "eval_steps_per_second": 12.005, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.06578544527292252, | |
| "learning_rate": 5.080932198479731e-05, | |
| "loss": 0.0164, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.92, | |
| "eval_loss": 0.032231684774160385, | |
| "eval_runtime": 5.4161, | |
| "eval_samples_per_second": 572.368, | |
| "eval_steps_per_second": 12.001, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.968553459119496, | |
| "grad_norm": 0.0489405132830143, | |
| "learning_rate": 1.5538018955595509e-06, | |
| "loss": 0.0152, | |
| "step": 3170 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 825254092458012.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7440020908825002, | |
| "learning_rate": 0.0004941090027879372, | |
| "lr_scheduler_type": "linear", | |
| "num_train_epochs": 10, | |
| "temperature": 13.346731795373474, | |
| "weight_decay": 0.2513528053510888 | |
| } | |
| } | |