| { | |
| "best_metric": 0.6864851430040411, | |
| "best_model_checkpoint": "robertuito-finetuned-class/run-4/checkpoint-1391", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 6955, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1014, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0918, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.707395498392283, | |
| "eval_f1": 0.6864851430040411, | |
| "eval_loss": 2.517336130142212, | |
| "eval_recall": 0.691174006920775, | |
| "eval_runtime": 2.732, | |
| "eval_samples_per_second": 455.345, | |
| "eval_steps_per_second": 14.275, | |
| "step": 1391 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0849, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0346, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0558, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6511254019292605, | |
| "eval_f1": 0.6550770166821124, | |
| "eval_loss": 3.2577097415924072, | |
| "eval_recall": 0.6993844873584972, | |
| "eval_runtime": 2.7559, | |
| "eval_samples_per_second": 451.394, | |
| "eval_steps_per_second": 14.151, | |
| "step": 2782 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1157, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1222, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5e-05, | |
| "loss": 0.12, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6816720257234726, | |
| "eval_f1": 0.6615872422776482, | |
| "eval_loss": 2.9731457233428955, | |
| "eval_recall": 0.6862419316028912, | |
| "eval_runtime": 2.7786, | |
| "eval_samples_per_second": 447.706, | |
| "eval_steps_per_second": 14.036, | |
| "step": 4173 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0977, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0781, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0563, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7114147909967846, | |
| "eval_f1": 0.6766621875546068, | |
| "eval_loss": 2.691765785217285, | |
| "eval_recall": 0.6623561625967606, | |
| "eval_runtime": 2.7561, | |
| "eval_samples_per_second": 451.37, | |
| "eval_steps_per_second": 14.151, | |
| "step": 5564 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0404, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0386, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6286173633440515, | |
| "eval_f1": 0.6317495399061165, | |
| "eval_loss": 2.9288928508758545, | |
| "eval_recall": 0.6781366429769011, | |
| "eval_runtime": 2.7411, | |
| "eval_samples_per_second": 453.837, | |
| "eval_steps_per_second": 14.228, | |
| "step": 6955 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6955, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 831511918018560.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 4.1843184321253836e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 9 | |
| } | |
| } | |