| { | |
| "best_global_step": 20, | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "/home/jupyter/wav2vec2-xls-r-300m-bfloat16-trained-test/checkpoints/checkpoint-20", | |
| "epoch": 4.2, | |
| "eval_steps": 4, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 26.8810977935791, | |
| "learning_rate": 0.0, | |
| "loss": 23.4044, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 13.647429466247559, | |
| "learning_rate": 6.666666666666667e-08, | |
| "loss": 15.7161, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 9.556838035583496, | |
| "learning_rate": 1.3333333333333334e-07, | |
| "loss": 13.7542, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 23.727840423583984, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 23.3498, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 11.243566513061523, | |
| "eval_runtime": 5.1037, | |
| "eval_samples_per_second": 12.54, | |
| "eval_steps_per_second": 3.135, | |
| "eval_wer": 1.0029940119760479, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 27.989702224731445, | |
| "learning_rate": 2.6666666666666667e-07, | |
| "loss": 28.0155, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 11.38569450378418, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 14.0151, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 10.933781623840332, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 13.2631, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 13.671080589294434, | |
| "learning_rate": 4.666666666666667e-07, | |
| "loss": 16.3699, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 11.2361421585083, | |
| "eval_runtime": 5.3019, | |
| "eval_samples_per_second": 12.071, | |
| "eval_steps_per_second": 3.018, | |
| "eval_wer": 1.0029940119760479, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 12.261064529418945, | |
| "learning_rate": 5.333333333333333e-07, | |
| "loss": 13.7451, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 14.111751556396484, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 16.4979, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 15.027251243591309, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 15.1573, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 11.976419448852539, | |
| "learning_rate": 7.333333333333333e-07, | |
| "loss": 13.3081, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_loss": 11.223762512207031, | |
| "eval_runtime": 5.1899, | |
| "eval_samples_per_second": 12.332, | |
| "eval_steps_per_second": 3.083, | |
| "eval_wer": 1.0029940119760479, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 10.190653800964355, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 14.3285, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 10.525323867797852, | |
| "learning_rate": 8.666666666666667e-07, | |
| "loss": 14.0348, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 8.292677879333496, | |
| "learning_rate": 9.333333333333334e-07, | |
| "loss": 11.2145, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 9.087738990783691, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 14.0511, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 11.206008911132812, | |
| "eval_runtime": 5.1678, | |
| "eval_samples_per_second": 12.384, | |
| "eval_steps_per_second": 3.096, | |
| "eval_wer": 1.0029940119760479, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 11.053936958312988, | |
| "learning_rate": 1.0666666666666667e-06, | |
| "loss": 13.6394, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 12.242379188537598, | |
| "learning_rate": 1.1333333333333334e-06, | |
| "loss": 14.9119, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "grad_norm": 10.784377098083496, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 13.2801, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 12.148858070373535, | |
| "learning_rate": 1.2666666666666667e-06, | |
| "loss": 13.7862, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_loss": 11.182676315307617, | |
| "eval_runtime": 4.9755, | |
| "eval_samples_per_second": 12.863, | |
| "eval_steps_per_second": 3.216, | |
| "eval_wer": 1.0, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.224933794681436e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |