| { | |
| "best_metric": 0.6297643780708313, | |
| "best_model_checkpoint": "/models/hfhub/DewiBrynJones/wav2vec2-xlsr-53-ft-btb-cy/checkpoint-1000", | |
| "epoch": 5.657708628005658, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.1315417256011315, | |
| "grad_norm": 1.0976253747940063, | |
| "learning_rate": 0.000147375, | |
| "loss": 4.9658, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "eval_loss": 1.1673808097839355, | |
| "eval_runtime": 174.6298, | |
| "eval_samples_per_second": 32.389, | |
| "eval_steps_per_second": 1.014, | |
| "eval_wer": 0.802891945242413, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.263083451202263, | |
| "grad_norm": 2.727048397064209, | |
| "learning_rate": 0.00029587499999999997, | |
| "loss": 1.2376, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "eval_loss": 0.6297643780708313, | |
| "eval_runtime": 181.3206, | |
| "eval_samples_per_second": 31.193, | |
| "eval_steps_per_second": 0.976, | |
| "eval_wer": 0.4413827414100239, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.3946251768033946, | |
| "grad_norm": 4.206020832061768, | |
| "learning_rate": 0.0002723571428571428, | |
| "loss": 0.8262, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.243281471004243, | |
| "eval_loss": 1.9106061458587646, | |
| "eval_runtime": 170.5081, | |
| "eval_samples_per_second": 33.171, | |
| "eval_steps_per_second": 1.038, | |
| "eval_wer": 0.8612444030748985, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.526166902404526, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0002462142857142857, | |
| "loss": 1.653, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.657708628005658, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00021792857142857142, | |
| "loss": 1.9941, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.657708628005658, | |
| "eval_loss": 1.9103525876998901, | |
| "eval_runtime": 170.4356, | |
| "eval_samples_per_second": 33.186, | |
| "eval_steps_per_second": 1.039, | |
| "eval_wer": 0.8610999662980854, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.657708628005658, | |
| "step": 2000, | |
| "total_flos": 1.5333464503060175e+19, | |
| "train_loss": 2.1353194427490236, | |
| "train_runtime": 5639.522, | |
| "train_samples_per_second": 56.742, | |
| "train_steps_per_second": 0.887 | |
| } | |
| ], | |
| "logging_steps": 400, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 1.5333464503060175e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |