| { | |
| "best_metric": 0.8085028357077502, | |
| "best_model_checkpoint": "./priority_model\\checkpoint-70", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 4.813197135925293, | |
| "learning_rate": 1.761904761904762e-05, | |
| "loss": 1.0447, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6981132075471698, | |
| "eval_f1": 0.6908805031446541, | |
| "eval_loss": 0.8730468153953552, | |
| "eval_runtime": 0.073, | |
| "eval_samples_per_second": 726.021, | |
| "eval_steps_per_second": 54.794, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 5.208207130432129, | |
| "learning_rate": 1.5476190476190476e-05, | |
| "loss": 0.9076, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7735849056603774, | |
| "eval_f1": 0.7699889219082452, | |
| "eval_loss": 0.6862769722938538, | |
| "eval_runtime": 0.063, | |
| "eval_samples_per_second": 841.277, | |
| "eval_steps_per_second": 63.493, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 4.731895923614502, | |
| "learning_rate": 1.3095238095238096e-05, | |
| "loss": 0.6144, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 3.9036455154418945, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 0.5519, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7547169811320755, | |
| "eval_f1": 0.7487870619946091, | |
| "eval_loss": 0.5854054093360901, | |
| "eval_runtime": 0.066, | |
| "eval_samples_per_second": 803.028, | |
| "eval_steps_per_second": 60.606, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 3.5843570232391357, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.4169, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7924528301886793, | |
| "eval_f1": 0.788810373716034, | |
| "eval_loss": 0.5518418550491333, | |
| "eval_runtime": 0.065, | |
| "eval_samples_per_second": 815.375, | |
| "eval_steps_per_second": 61.538, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 4.46596097946167, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.4426, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.991420269012451, | |
| "learning_rate": 3.5714285714285718e-06, | |
| "loss": 0.3359, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8113207547169812, | |
| "eval_f1": 0.8085028357077502, | |
| "eval_loss": 0.5408324599266052, | |
| "eval_runtime": 0.048, | |
| "eval_samples_per_second": 1104.181, | |
| "eval_steps_per_second": 83.334, | |
| "step": 70 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 84, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 48331673705700.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |