{ "best_metric": 25.88532883642496, "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-40", "epoch": 6.349206349206349, "eval_steps": 40, "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.6349206349206349, "grad_norm": 5.399875164031982, "learning_rate": 3.75e-05, "loss": 0.3295, "step": 20 }, { "epoch": 1.2698412698412698, "grad_norm": 3.3471007347106934, "learning_rate": 3.75e-05, "loss": 0.2449, "step": 40 }, { "epoch": 1.2698412698412698, "eval_loss": 0.15969730913639069, "eval_runtime": 56.6547, "eval_samples_per_second": 1.765, "eval_steps_per_second": 1.765, "eval_wer": 25.88532883642496, "step": 40 }, { "epoch": 1.9047619047619047, "grad_norm": 3.7166926860809326, "learning_rate": 3.75e-05, "loss": 0.1628, "step": 60 }, { "epoch": 2.5396825396825395, "grad_norm": 2.8016231060028076, "learning_rate": 3.75e-05, "loss": 0.1002, "step": 80 }, { "epoch": 2.5396825396825395, "eval_loss": 0.16435778141021729, "eval_runtime": 56.8979, "eval_samples_per_second": 1.758, "eval_steps_per_second": 1.758, "eval_wer": 26.644182124789207, "step": 80 }, { "epoch": 3.1746031746031744, "grad_norm": 2.115607976913452, "learning_rate": 3.75e-05, "loss": 0.0833, "step": 100 }, { "epoch": 3.8095238095238093, "grad_norm": 2.412848711013794, "learning_rate": 3.75e-05, "loss": 0.0492, "step": 120 }, { "epoch": 3.8095238095238093, "eval_loss": 0.1730262041091919, "eval_runtime": 56.4319, "eval_samples_per_second": 1.772, "eval_steps_per_second": 1.772, "eval_wer": 27.065767284991573, "step": 120 }, { "epoch": 4.444444444444445, "grad_norm": 1.6271615028381348, "learning_rate": 3.75e-05, "loss": 0.0323, "step": 140 }, { "epoch": 5.079365079365079, "grad_norm": 1.2262483835220337, "learning_rate": 3.75e-05, "loss": 0.0269, "step": 160 }, { "epoch": 5.079365079365079, "eval_loss": 0.1800723522901535, "eval_runtime": 57.7815, "eval_samples_per_second": 1.731, "eval_steps_per_second": 1.731, "eval_wer": 26.981450252951095, "step": 160 }, { "epoch": 5.714285714285714, "grad_norm": 1.4037569761276245, "learning_rate": 3.75e-05, "loss": 0.0162, "step": 180 }, { "epoch": 6.349206349206349, "grad_norm": 1.1472747325897217, "learning_rate": 3.75e-05, "loss": 0.0156, "step": 200 }, { "epoch": 6.349206349206349, "eval_loss": 0.20425428450107574, "eval_runtime": 56.7961, "eval_samples_per_second": 1.761, "eval_steps_per_second": 1.761, "eval_wer": 27.65598650927487, "step": 200 }, { "epoch": 6.349206349206349, "step": 200, "total_flos": 1.5637915828224e+17, "train_loss": 0.10607707306742668, "train_runtime": 1147.3926, "train_samples_per_second": 5.578, "train_steps_per_second": 0.174 } ], "logging_steps": 20, "max_steps": 200, "num_input_tokens_seen": 0, "num_train_epochs": 7, "save_steps": 40, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.5637915828224e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }