| { | |
| "best_metric": 0.2201576977968216, | |
| "best_model_checkpoint": "/scratch/skscla001/speech/results/xls-r-1b-bemgen-female-model/checkpoint-1300", | |
| "epoch": 5.987654320987654, | |
| "eval_steps": 100, | |
| "global_step": 1700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3527336860670194, | |
| "eval_loss": 4.0290117263793945, | |
| "eval_runtime": 20.5733, | |
| "eval_samples_per_second": 13.367, | |
| "eval_steps_per_second": 3.354, | |
| "eval_wer": 1.0089621277825962, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7054673721340388, | |
| "eval_loss": 2.847311019897461, | |
| "eval_runtime": 20.3386, | |
| "eval_samples_per_second": 13.521, | |
| "eval_steps_per_second": 3.393, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.056437389770723, | |
| "eval_loss": 0.6011506915092468, | |
| "eval_runtime": 20.4093, | |
| "eval_samples_per_second": 13.474, | |
| "eval_steps_per_second": 3.381, | |
| "eval_wer": 0.9268574732581671, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4091710758377425, | |
| "eval_loss": 0.41906026005744934, | |
| "eval_runtime": 20.4439, | |
| "eval_samples_per_second": 13.451, | |
| "eval_steps_per_second": 3.375, | |
| "eval_wer": 0.8378143972246314, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.7619047619047619, | |
| "grad_norm": 2.2041492462158203, | |
| "learning_rate": 2.976e-05, | |
| "loss": 4.9128, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.7619047619047619, | |
| "eval_loss": 0.33197441697120667, | |
| "eval_runtime": 20.5028, | |
| "eval_samples_per_second": 13.413, | |
| "eval_steps_per_second": 3.365, | |
| "eval_wer": 0.6423821913847932, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.112874779541446, | |
| "eval_loss": 0.2715909779071808, | |
| "eval_runtime": 20.5575, | |
| "eval_samples_per_second": 13.377, | |
| "eval_steps_per_second": 3.356, | |
| "eval_wer": 0.5177797051170858, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.4656084656084656, | |
| "eval_loss": 0.27242884039878845, | |
| "eval_runtime": 20.5374, | |
| "eval_samples_per_second": 13.39, | |
| "eval_steps_per_second": 3.36, | |
| "eval_wer": 0.4929170280427869, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.818342151675485, | |
| "eval_loss": 0.2515610158443451, | |
| "eval_runtime": 20.5916, | |
| "eval_samples_per_second": 13.355, | |
| "eval_steps_per_second": 3.351, | |
| "eval_wer": 0.4787510841283608, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.1693121693121693, | |
| "eval_loss": 0.2385055273771286, | |
| "eval_runtime": 20.5752, | |
| "eval_samples_per_second": 13.366, | |
| "eval_steps_per_second": 3.354, | |
| "eval_wer": 0.44376987568661463, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.5220458553791887, | |
| "grad_norm": 2.694538116455078, | |
| "learning_rate": 2.8137672090112644e-05, | |
| "loss": 0.4407, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.5220458553791887, | |
| "eval_loss": 0.2374495565891266, | |
| "eval_runtime": 20.582, | |
| "eval_samples_per_second": 13.361, | |
| "eval_steps_per_second": 3.352, | |
| "eval_wer": 0.43451864700780574, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.874779541446208, | |
| "eval_loss": 0.2354438155889511, | |
| "eval_runtime": 20.5688, | |
| "eval_samples_per_second": 13.37, | |
| "eval_steps_per_second": 3.355, | |
| "eval_wer": 0.4096559699335068, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.225749559082892, | |
| "eval_loss": 0.22052845358848572, | |
| "eval_runtime": 20.5553, | |
| "eval_samples_per_second": 13.379, | |
| "eval_steps_per_second": 3.357, | |
| "eval_wer": 0.3960682278115062, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.578483245149911, | |
| "eval_loss": 0.2201576977968216, | |
| "eval_runtime": 20.6429, | |
| "eval_samples_per_second": 13.322, | |
| "eval_steps_per_second": 3.343, | |
| "eval_wer": 0.3897080080948251, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.931216931216931, | |
| "eval_loss": 0.22464053332805634, | |
| "eval_runtime": 20.524, | |
| "eval_samples_per_second": 13.399, | |
| "eval_steps_per_second": 3.362, | |
| "eval_wer": 0.3897080080948251, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.2821869488536155, | |
| "grad_norm": 1.8704134225845337, | |
| "learning_rate": 2.626032540675845e-05, | |
| "loss": 0.2698, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.2821869488536155, | |
| "eval_loss": 0.2338586151599884, | |
| "eval_runtime": 20.5858, | |
| "eval_samples_per_second": 13.359, | |
| "eval_steps_per_second": 3.352, | |
| "eval_wer": 0.36657993639780284, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.634920634920634, | |
| "eval_loss": 0.2358291894197464, | |
| "eval_runtime": 20.5183, | |
| "eval_samples_per_second": 13.403, | |
| "eval_steps_per_second": 3.363, | |
| "eval_wer": 0.3735183579069095, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.987654320987654, | |
| "eval_loss": 0.22457349300384521, | |
| "eval_runtime": 20.6137, | |
| "eval_samples_per_second": 13.341, | |
| "eval_steps_per_second": 3.347, | |
| "eval_wer": 0.38074588031222895, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.987654320987654, | |
| "step": 1700, | |
| "total_flos": 1.0310954059792316e+19, | |
| "train_loss": 1.6788882132137524, | |
| "train_runtime": 2349.413, | |
| "train_samples_per_second": 28.922, | |
| "train_steps_per_second": 3.614 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8490, | |
| "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": 3 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0310954059792316e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |