| { | |
| "best_metric": 0.21721933782100677, | |
| "best_model_checkpoint": "./xls-r-1b-bem-genbed-all/checkpoint-3000", | |
| "epoch": 4.9966953073364175, | |
| "eval_steps": 200, | |
| "global_step": 3780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26437541308658297, | |
| "grad_norm": 3.2960169315338135, | |
| "learning_rate": 1.97e-05, | |
| "loss": 4.6827, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26437541308658297, | |
| "eval_loss": 2.834711790084839, | |
| "eval_runtime": 123.0061, | |
| "eval_samples_per_second": 16.406, | |
| "eval_steps_per_second": 4.105, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5287508261731659, | |
| "grad_norm": 5.020152568817139, | |
| "learning_rate": 3.97e-05, | |
| "loss": 1.0401, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5287508261731659, | |
| "eval_loss": 0.5636318922042847, | |
| "eval_runtime": 121.385, | |
| "eval_samples_per_second": 16.625, | |
| "eval_steps_per_second": 4.16, | |
| "eval_wer": 0.9410307234886026, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7931262392597488, | |
| "grad_norm": 4.118492126464844, | |
| "learning_rate": 4.852134146341464e-05, | |
| "loss": 0.4289, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7931262392597488, | |
| "eval_loss": 0.40182533860206604, | |
| "eval_runtime": 122.3487, | |
| "eval_samples_per_second": 16.494, | |
| "eval_steps_per_second": 4.128, | |
| "eval_wer": 0.9028741328047571, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0575016523463319, | |
| "grad_norm": 1.0666348934173584, | |
| "learning_rate": 4.5472560975609756e-05, | |
| "loss": 0.3449, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0575016523463319, | |
| "eval_loss": 0.3604024052619934, | |
| "eval_runtime": 123.3539, | |
| "eval_samples_per_second": 16.359, | |
| "eval_steps_per_second": 4.094, | |
| "eval_wer": 0.8771060455896927, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3218770654329148, | |
| "grad_norm": 3.203847885131836, | |
| "learning_rate": 4.242378048780488e-05, | |
| "loss": 0.2954, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3218770654329148, | |
| "eval_loss": 0.33885934948921204, | |
| "eval_runtime": 124.083, | |
| "eval_samples_per_second": 16.263, | |
| "eval_steps_per_second": 4.07, | |
| "eval_wer": 0.8741328047571854, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5862524785194978, | |
| "grad_norm": 2.469949722290039, | |
| "learning_rate": 3.9375e-05, | |
| "loss": 0.2719, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5862524785194978, | |
| "eval_loss": 0.29620760679244995, | |
| "eval_runtime": 124.1198, | |
| "eval_samples_per_second": 16.258, | |
| "eval_steps_per_second": 4.069, | |
| "eval_wer": 0.8439048562933598, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8506278916060808, | |
| "grad_norm": 1.500631332397461, | |
| "learning_rate": 3.632621951219513e-05, | |
| "loss": 0.2472, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.8506278916060808, | |
| "eval_loss": 0.2701094448566437, | |
| "eval_runtime": 123.0962, | |
| "eval_samples_per_second": 16.394, | |
| "eval_steps_per_second": 4.102, | |
| "eval_wer": 0.8052527254707631, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.1150033046926637, | |
| "grad_norm": 0.6989238262176514, | |
| "learning_rate": 3.327743902439025e-05, | |
| "loss": 0.2093, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.1150033046926637, | |
| "eval_loss": 0.2598518133163452, | |
| "eval_runtime": 123.5685, | |
| "eval_samples_per_second": 16.331, | |
| "eval_steps_per_second": 4.087, | |
| "eval_wer": 0.8285431119920713, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.3793787177792467, | |
| "grad_norm": 0.9642230868339539, | |
| "learning_rate": 3.022865853658537e-05, | |
| "loss": 0.1725, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.3793787177792467, | |
| "eval_loss": 0.2533758580684662, | |
| "eval_runtime": 124.2174, | |
| "eval_samples_per_second": 16.246, | |
| "eval_steps_per_second": 4.065, | |
| "eval_wer": 0.8374628344895937, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.6437541308658297, | |
| "grad_norm": 0.6002829670906067, | |
| "learning_rate": 2.717987804878049e-05, | |
| "loss": 0.1675, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.6437541308658297, | |
| "eval_loss": 0.240593820810318, | |
| "eval_runtime": 125.6029, | |
| "eval_samples_per_second": 16.067, | |
| "eval_steps_per_second": 4.021, | |
| "eval_wer": 0.7690782953419227, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.9081295439524126, | |
| "grad_norm": 1.0664541721343994, | |
| "learning_rate": 2.413109756097561e-05, | |
| "loss": 0.1632, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.9081295439524126, | |
| "eval_loss": 0.2308683693408966, | |
| "eval_runtime": 124.4, | |
| "eval_samples_per_second": 16.222, | |
| "eval_steps_per_second": 4.059, | |
| "eval_wer": 0.7616451932606542, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.1725049570389956, | |
| "grad_norm": 0.6247605681419373, | |
| "learning_rate": 2.108231707317073e-05, | |
| "loss": 0.1295, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.1725049570389956, | |
| "eval_loss": 0.23871323466300964, | |
| "eval_runtime": 124.4212, | |
| "eval_samples_per_second": 16.219, | |
| "eval_steps_per_second": 4.059, | |
| "eval_wer": 0.7556987115956393, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.4368803701255786, | |
| "grad_norm": 0.5814207792282104, | |
| "learning_rate": 1.8033536585365853e-05, | |
| "loss": 0.1082, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.4368803701255786, | |
| "eval_loss": 0.2275388240814209, | |
| "eval_runtime": 124.2026, | |
| "eval_samples_per_second": 16.248, | |
| "eval_steps_per_second": 4.066, | |
| "eval_wer": 0.7329038652130823, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.7012557832121615, | |
| "grad_norm": 0.5891350507736206, | |
| "learning_rate": 1.4984756097560976e-05, | |
| "loss": 0.1059, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.7012557832121615, | |
| "eval_loss": 0.223988875746727, | |
| "eval_runtime": 124.3149, | |
| "eval_samples_per_second": 16.233, | |
| "eval_steps_per_second": 4.062, | |
| "eval_wer": 0.7329038652130823, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.965631196298744, | |
| "grad_norm": 1.1988484859466553, | |
| "learning_rate": 1.1935975609756097e-05, | |
| "loss": 0.1049, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.965631196298744, | |
| "eval_loss": 0.21721933782100677, | |
| "eval_runtime": 124.3819, | |
| "eval_samples_per_second": 16.224, | |
| "eval_steps_per_second": 4.06, | |
| "eval_wer": 0.7294350842418236, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.2300066093853275, | |
| "grad_norm": 0.32344383001327515, | |
| "learning_rate": 8.88719512195122e-06, | |
| "loss": 0.0657, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.2300066093853275, | |
| "eval_loss": 0.23203983902931213, | |
| "eval_runtime": 125.8598, | |
| "eval_samples_per_second": 16.034, | |
| "eval_steps_per_second": 4.012, | |
| "eval_wer": 0.722001982160555, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.49438202247191, | |
| "grad_norm": 0.8019347190856934, | |
| "learning_rate": 5.838414634146342e-06, | |
| "loss": 0.059, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.49438202247191, | |
| "eval_loss": 0.23412850499153137, | |
| "eval_runtime": 124.439, | |
| "eval_samples_per_second": 16.217, | |
| "eval_steps_per_second": 4.058, | |
| "eval_wer": 0.7215064420218038, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.758757435558493, | |
| "grad_norm": 0.9007149934768677, | |
| "learning_rate": 2.7896341463414635e-06, | |
| "loss": 0.0582, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.758757435558493, | |
| "eval_loss": 0.2315901815891266, | |
| "eval_runtime": 124.1565, | |
| "eval_samples_per_second": 16.254, | |
| "eval_steps_per_second": 4.067, | |
| "eval_wer": 0.711595639246779, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.9966953073364175, | |
| "step": 3780, | |
| "total_flos": 1.7289369341985591e+19, | |
| "train_loss": 0.4604275799302197, | |
| "train_runtime": 6372.0716, | |
| "train_samples_per_second": 4.746, | |
| "train_steps_per_second": 0.593 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 3780, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7289369341985591e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |