{ "best_metric": null, "best_model_checkpoint": null, "epoch": 25.0, "eval_steps": 100, "global_step": 800, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 3.12, "learning_rate": 8.958333333333334e-07, "loss": 0.258, "step": 100 }, { "epoch": 3.12, "eval_accuracy": 0.9438073394495413, "eval_f1": 0.9539906103286386, "eval_loss": 0.15450334548950195, "eval_precision": 0.9442379182156134, "eval_recall": 0.9639468690702088, "eval_runtime": 8.5544, "eval_samples_per_second": 101.936, "eval_steps_per_second": 1.637, "step": 100 }, { "epoch": 6.25, "learning_rate": 7.916666666666666e-07, "loss": 0.167, "step": 200 }, { "epoch": 6.25, "eval_accuracy": 0.9461009174311926, "eval_f1": 0.955785512699906, "eval_loss": 0.14454717934131622, "eval_precision": 0.9477611940298507, "eval_recall": 0.9639468690702088, "eval_runtime": 7.7646, "eval_samples_per_second": 112.305, "eval_steps_per_second": 1.803, "step": 200 }, { "epoch": 9.38, "learning_rate": 6.875e-07, "loss": 0.1256, "step": 300 }, { "epoch": 9.38, "eval_accuracy": 0.948394495412844, "eval_f1": 0.9576669802445908, "eval_loss": 0.1421377956867218, "eval_precision": 0.9496268656716418, "eval_recall": 0.9658444022770398, "eval_runtime": 7.7937, "eval_samples_per_second": 111.886, "eval_steps_per_second": 1.796, "step": 300 }, { "epoch": 12.5, "learning_rate": 5.833333333333334e-07, "loss": 0.1077, "step": 400 }, { "epoch": 12.5, "eval_accuracy": 0.9461009174311926, "eval_f1": 0.9559512652296157, "eval_loss": 0.15086643397808075, "eval_precision": 0.9444444444444444, "eval_recall": 0.967741935483871, "eval_runtime": 8.6557, "eval_samples_per_second": 100.743, "eval_steps_per_second": 1.617, "step": 400 }, { "epoch": 15.62, "learning_rate": 4.791666666666667e-07, "loss": 0.0938, "step": 500 }, { "epoch": 15.62, "eval_accuracy": 0.9472477064220184, "eval_f1": 0.9567669172932332, "eval_loss": 0.14971153438091278, "eval_precision": 0.9478584729981379, "eval_recall": 0.9658444022770398, "eval_runtime": 8.1439, "eval_samples_per_second": 107.074, "eval_steps_per_second": 1.719, "step": 500 }, { "epoch": 18.75, "learning_rate": 3.75e-07, "loss": 0.0902, "step": 600 }, { "epoch": 18.75, "eval_accuracy": 0.9472477064220184, "eval_f1": 0.9567669172932332, "eval_loss": 0.1525813788175583, "eval_precision": 0.9478584729981379, "eval_recall": 0.9658444022770398, "eval_runtime": 7.7724, "eval_samples_per_second": 112.192, "eval_steps_per_second": 1.801, "step": 600 }, { "epoch": 21.88, "learning_rate": 2.708333333333333e-07, "loss": 0.0808, "step": 700 }, { "epoch": 21.88, "eval_accuracy": 0.9472477064220184, "eval_f1": 0.9568480300187617, "eval_loss": 0.15572945773601532, "eval_precision": 0.9461966604823747, "eval_recall": 0.967741935483871, "eval_runtime": 8.1645, "eval_samples_per_second": 106.804, "eval_steps_per_second": 1.715, "step": 700 }, { "epoch": 25.0, "learning_rate": 1.6666666666666665e-07, "loss": 0.0757, "step": 800 }, { "epoch": 25.0, "eval_accuracy": 0.9495412844036697, "eval_f1": 0.9586466165413534, "eval_loss": 0.1543181836605072, "eval_precision": 0.9497206703910615, "eval_recall": 0.967741935483871, "eval_runtime": 7.7426, "eval_samples_per_second": 112.624, "eval_steps_per_second": 1.808, "step": 800 } ], "logging_steps": 100, "max_steps": 960, "num_train_epochs": 30, "save_steps": 100, "total_flos": 3343154847168000.0, "trial_name": null, "trial_params": null }