| { | |
| "best_metric": 0.31921109557151794, | |
| "best_model_checkpoint": "/scratch/skscla001/speech/results/xls-r-1b-bemgen-male-model/checkpoint-1100", | |
| "epoch": 5.396396396396397, | |
| "eval_steps": 100, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.36036036036036034, | |
| "eval_loss": 3.5018131732940674, | |
| "eval_runtime": 20.1337, | |
| "eval_samples_per_second": 13.212, | |
| "eval_steps_per_second": 3.328, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7207207207207207, | |
| "eval_loss": 2.861574172973633, | |
| "eval_runtime": 19.7549, | |
| "eval_samples_per_second": 13.465, | |
| "eval_steps_per_second": 3.392, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0792792792792794, | |
| "eval_loss": 1.1005113124847412, | |
| "eval_runtime": 19.8914, | |
| "eval_samples_per_second": 13.373, | |
| "eval_steps_per_second": 3.368, | |
| "eval_wer": 0.9874489200233508, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4396396396396396, | |
| "eval_loss": 0.6284775137901306, | |
| "eval_runtime": 19.8994, | |
| "eval_samples_per_second": 13.367, | |
| "eval_steps_per_second": 3.367, | |
| "eval_wer": 0.835960303561004, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 4.3291192054748535, | |
| "learning_rate": 2.976e-05, | |
| "loss": 5.4329, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 0.5177034139633179, | |
| "eval_runtime": 20.1072, | |
| "eval_samples_per_second": 13.229, | |
| "eval_steps_per_second": 3.332, | |
| "eval_wer": 0.7589025102159953, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.1585585585585587, | |
| "eval_loss": 0.4120151698589325, | |
| "eval_runtime": 19.9558, | |
| "eval_samples_per_second": 13.329, | |
| "eval_steps_per_second": 3.357, | |
| "eval_wer": 0.5989492119089317, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.518918918918919, | |
| "eval_loss": 0.3997887372970581, | |
| "eval_runtime": 19.8674, | |
| "eval_samples_per_second": 13.389, | |
| "eval_steps_per_second": 3.372, | |
| "eval_wer": 0.549620548744892, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.879279279279279, | |
| "eval_loss": 0.37145501375198364, | |
| "eval_runtime": 19.9372, | |
| "eval_samples_per_second": 13.342, | |
| "eval_steps_per_second": 3.361, | |
| "eval_wer": 0.5653823701109165, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.237837837837838, | |
| "eval_loss": 0.335114449262619, | |
| "eval_runtime": 20.0684, | |
| "eval_samples_per_second": 13.255, | |
| "eval_steps_per_second": 3.339, | |
| "eval_wer": 0.4871570344424985, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.5981981981981983, | |
| "grad_norm": 2.873544692993164, | |
| "learning_rate": 2.8094750320102435e-05, | |
| "loss": 0.644, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.5981981981981983, | |
| "eval_loss": 0.33343327045440674, | |
| "eval_runtime": 19.9311, | |
| "eval_samples_per_second": 13.346, | |
| "eval_steps_per_second": 3.362, | |
| "eval_wer": 0.5014594279042616, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.9585585585585585, | |
| "eval_loss": 0.31921109557151794, | |
| "eval_runtime": 19.8773, | |
| "eval_samples_per_second": 13.382, | |
| "eval_steps_per_second": 3.371, | |
| "eval_wer": 0.4795680093403386, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.317117117117117, | |
| "eval_loss": 0.3245755434036255, | |
| "eval_runtime": 19.8844, | |
| "eval_samples_per_second": 13.377, | |
| "eval_steps_per_second": 3.369, | |
| "eval_wer": 0.45942790426152946, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.677477477477478, | |
| "eval_loss": 0.3242240250110626, | |
| "eval_runtime": 19.9276, | |
| "eval_samples_per_second": 13.348, | |
| "eval_steps_per_second": 3.362, | |
| "eval_wer": 0.46117921774664333, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.036036036036036, | |
| "eval_loss": 0.3202609419822693, | |
| "eval_runtime": 20.0808, | |
| "eval_samples_per_second": 13.246, | |
| "eval_steps_per_second": 3.337, | |
| "eval_wer": 0.44424985405720957, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.396396396396397, | |
| "grad_norm": 1.6697975397109985, | |
| "learning_rate": 2.61741357234315e-05, | |
| "loss": 0.3852, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.396396396396397, | |
| "eval_loss": 0.3452622890472412, | |
| "eval_runtime": 19.9999, | |
| "eval_samples_per_second": 13.3, | |
| "eval_steps_per_second": 3.35, | |
| "eval_wer": 0.44162288382953885, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.396396396396397, | |
| "step": 1500, | |
| "total_flos": 9.503204377194754e+18, | |
| "train_loss": 2.1540360310872395, | |
| "train_runtime": 2027.2043, | |
| "train_samples_per_second": 32.809, | |
| "train_steps_per_second": 4.099 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8310, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 4, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 1 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.503204377194754e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |