{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.8596961572832886, "eval_steps": 200, "global_step": 1600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.18, "learning_rate": 5e-05, "loss": 0.0467, "step": 100 }, { "epoch": 0.36, "learning_rate": 4.994863481875841e-05, "loss": 0.0433, "step": 200 }, { "epoch": 0.36, "eval_f1_score": 0.6251468860164512, "eval_label_f1": 0.8319623971797884, "eval_loss": 0.05231842026114464, "eval_runtime": 457.2641, "eval_samples_per_second": 2.187, "eval_steps_per_second": 0.547, "eval_wer": 0.10434244136543076, "step": 200 }, { "epoch": 0.54, "learning_rate": 4.979475034558115e-05, "loss": 0.0402, "step": 300 }, { "epoch": 0.71, "learning_rate": 4.9538978924776634e-05, "loss": 0.0391, "step": 400 }, { "epoch": 0.71, "eval_f1_score": 0.6206896551724138, "eval_label_f1": 0.8346394984326019, "eval_loss": 0.050447478890419006, "eval_runtime": 457.7184, "eval_samples_per_second": 2.185, "eval_steps_per_second": 0.546, "eval_wer": 0.10472946822509482, "step": 400 }, { "epoch": 0.89, "learning_rate": 4.9182371575975736e-05, "loss": 0.037, "step": 500 }, { "epoch": 1.07, "learning_rate": 4.8726393675266716e-05, "loss": 0.0381, "step": 600 }, { "epoch": 1.07, "eval_f1_score": 0.6142467886337096, "eval_label_f1": 0.8322304398598676, "eval_loss": 0.049576789140701294, "eval_runtime": 458.238, "eval_samples_per_second": 2.182, "eval_steps_per_second": 0.546, "eval_wer": 0.10654849446551591, "step": 600 }, { "epoch": 1.25, "learning_rate": 4.817291893365055e-05, "loss": 0.037, "step": 700 }, { "epoch": 1.43, "learning_rate": 4.752422169756048e-05, "loss": 0.0374, "step": 800 }, { "epoch": 1.43, "eval_f1_score": 0.6158081006685017, "eval_label_f1": 0.8360204482894219, "eval_loss": 0.04838084056973457, "eval_runtime": 456.7741, "eval_samples_per_second": 2.189, "eval_steps_per_second": 0.547, "eval_wer": 0.10705162938307919, "step": 800 }, { "epoch": 1.61, "learning_rate": 4.678296760308474e-05, "loss": 0.0375, "step": 900 }, { "epoch": 1.79, "learning_rate": 4.595220262229601e-05, "loss": 0.0374, "step": 1000 }, { "epoch": 1.79, "eval_f1_score": 0.6155063291139241, "eval_label_f1": 0.8370253164556962, "eval_loss": 0.047435563057661057, "eval_runtime": 463.3137, "eval_samples_per_second": 2.158, "eval_steps_per_second": 0.54, "eval_wer": 0.10693552132517997, "step": 1000 }, { "epoch": 1.97, "learning_rate": 4.503534054669892e-05, "loss": 0.0379, "step": 1100 }, { "epoch": 2.14, "learning_rate": 4.4036148959228365e-05, "loss": 0.0342, "step": 1200 }, { "epoch": 2.14, "eval_f1_score": 0.6118110236220473, "eval_label_f1": 0.8362204724409448, "eval_loss": 0.04737536609172821, "eval_runtime": 458.8739, "eval_samples_per_second": 2.179, "eval_steps_per_second": 0.545, "eval_wer": 0.1077095750445081, "step": 1200 }, { "epoch": 2.32, "learning_rate": 4.2958733752443195e-05, "loss": 0.0345, "step": 1300 }, { "epoch": 2.5, "learning_rate": 4.180752225653292e-05, "loss": 0.0362, "step": 1400 }, { "epoch": 2.5, "eval_f1_score": 0.6138147566718996, "eval_label_f1": 0.8375196232339089, "eval_loss": 0.04683598503470421, "eval_runtime": 458.5195, "eval_samples_per_second": 2.181, "eval_steps_per_second": 0.545, "eval_wer": 0.10790308847434012, "step": 1400 }, { "epoch": 2.68, "learning_rate": 4.058724504646834e-05, "loss": 0.0343, "step": 1500 }, { "epoch": 2.86, "learning_rate": 3.9302916503054246e-05, "loss": 0.0351, "step": 1600 }, { "epoch": 2.86, "eval_f1_score": 0.6101960784313726, "eval_label_f1": 0.836078431372549, "eval_loss": 0.04606114700436592, "eval_runtime": 460.3927, "eval_samples_per_second": 2.172, "eval_steps_per_second": 0.543, "eval_wer": 0.10821270996207137, "step": 1600 } ], "logging_steps": 100, "max_steps": 5000, "num_input_tokens_seen": 0, "num_train_epochs": 9, "save_steps": 200, "total_flos": 6.253717078962143e+19, "train_batch_size": 4, "trial_name": null, "trial_params": null }