| { | |
| "best_metric": 0.6896090724966321, | |
| "best_model_checkpoint": "robertuito-finetuned-class/run-3/checkpoint-4173", | |
| "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": 0.0973, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1564, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6937299035369775, | |
| "eval_f1": 0.6874258776859443, | |
| "eval_loss": 2.4691991806030273, | |
| "eval_recall": 0.7043517878228231, | |
| "eval_runtime": 2.7026, | |
| "eval_samples_per_second": 460.298, | |
| "eval_steps_per_second": 14.431, | |
| "step": 1391 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1692, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0836, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1061, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7057877813504824, | |
| "eval_f1": 0.684732277475961, | |
| "eval_loss": 2.516491413116455, | |
| "eval_recall": 0.6915848416105334, | |
| "eval_runtime": 2.7425, | |
| "eval_samples_per_second": 453.594, | |
| "eval_steps_per_second": 14.22, | |
| "step": 2782 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0878, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0875, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0944, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7017684887459807, | |
| "eval_f1": 0.6896090724966321, | |
| "eval_loss": 2.5273776054382324, | |
| "eval_recall": 0.7008374270116728, | |
| "eval_runtime": 2.6801, | |
| "eval_samples_per_second": 464.16, | |
| "eval_steps_per_second": 14.552, | |
| "step": 4173 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1055, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0663, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1577, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6889067524115756, | |
| "eval_f1": 0.6774362773843833, | |
| "eval_loss": 2.4060730934143066, | |
| "eval_recall": 0.7025269104985858, | |
| "eval_runtime": 2.7132, | |
| "eval_samples_per_second": 458.492, | |
| "eval_steps_per_second": 14.374, | |
| "step": 5564 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 5e-05, | |
| "loss": 0.097, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0509, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7001607717041801, | |
| "eval_f1": 0.6789899894443611, | |
| "eval_loss": 2.549386978149414, | |
| "eval_recall": 0.6900844757396032, | |
| "eval_runtime": 2.6922, | |
| "eval_samples_per_second": 462.08, | |
| "eval_steps_per_second": 14.486, | |
| "step": 6955 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6955, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 830114619195392.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 8.108485051435389e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 34 | |
| } | |
| } | |