| { |
| "best_metric": 90.62626262626262, |
| "best_model_checkpoint": "./whisperEnsembleModel/model_1_conservative/checkpoint-942", |
| "epoch": 4.997347480106101, |
| "eval_steps": 500, |
| "global_step": 942, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.26525198938992045, |
| "grad_norm": 4.3444318771362305, |
| "learning_rate": 1.5000000000000002e-07, |
| "loss": 2.093, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.5305039787798409, |
| "grad_norm": 0.913415253162384, |
| "learning_rate": 3.0000000000000004e-07, |
| "loss": 2.0792, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.7957559681697612, |
| "grad_norm": 2.48246169090271, |
| "learning_rate": 4.5e-07, |
| "loss": 2.0802, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.9973474801061007, |
| "eval_loss": 2.097313404083252, |
| "eval_norm_levenshtein_similarity": 89.27272727272727, |
| "eval_runtime": 195.7596, |
| "eval_samples_per_second": 1.711, |
| "eval_steps_per_second": 0.429, |
| "eval_wer": 14.3646408839779, |
| "step": 188 |
| }, |
| { |
| "epoch": 1.0610079575596818, |
| "grad_norm": 1.0793434381484985, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 2.0774, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.3262599469496021, |
| "grad_norm": 1.7866408824920654, |
| "learning_rate": 7.5e-07, |
| "loss": 2.0719, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.5915119363395225, |
| "grad_norm": 1.1607919931411743, |
| "learning_rate": 9e-07, |
| "loss": 2.0694, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.8567639257294428, |
| "grad_norm": 0.9631378650665283, |
| "learning_rate": 1.05e-06, |
| "loss": 2.0688, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 2.093317747116089, |
| "eval_norm_levenshtein_similarity": 89.56565656565655, |
| "eval_runtime": 194.7642, |
| "eval_samples_per_second": 1.72, |
| "eval_steps_per_second": 0.431, |
| "eval_wer": 13.591160220994475, |
| "step": 377 |
| }, |
| { |
| "epoch": 2.1220159151193636, |
| "grad_norm": 1.052522897720337, |
| "learning_rate": 1.2000000000000002e-06, |
| "loss": 2.0674, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.387267904509284, |
| "grad_norm": 0.5794100165367126, |
| "learning_rate": 1.35e-06, |
| "loss": 2.0651, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.6525198938992043, |
| "grad_norm": 1.1435893774032593, |
| "learning_rate": 1.5e-06, |
| "loss": 2.062, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.9177718832891246, |
| "grad_norm": 1.112978219985962, |
| "learning_rate": 1.65e-06, |
| "loss": 2.063, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.9973474801061006, |
| "eval_loss": 2.0927727222442627, |
| "eval_norm_levenshtein_similarity": 89.75757575757576, |
| "eval_runtime": 197.154, |
| "eval_samples_per_second": 1.699, |
| "eval_steps_per_second": 0.426, |
| "eval_wer": 13.701657458563535, |
| "step": 565 |
| }, |
| { |
| "epoch": 3.183023872679045, |
| "grad_norm": 0.38901379704475403, |
| "learning_rate": 1.8e-06, |
| "loss": 2.0592, |
| "step": 600 |
| }, |
| { |
| "epoch": 3.4482758620689653, |
| "grad_norm": 0.27914485335350037, |
| "learning_rate": 1.95e-06, |
| "loss": 2.0582, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.713527851458886, |
| "grad_norm": 0.308776319026947, |
| "learning_rate": 2.1e-06, |
| "loss": 2.0579, |
| "step": 700 |
| }, |
| { |
| "epoch": 3.9787798408488064, |
| "grad_norm": 0.6666736602783203, |
| "learning_rate": 2.25e-06, |
| "loss": 2.0581, |
| "step": 750 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 2.0944669246673584, |
| "eval_norm_levenshtein_similarity": 89.8080808080808, |
| "eval_runtime": 197.8784, |
| "eval_samples_per_second": 1.693, |
| "eval_steps_per_second": 0.425, |
| "eval_wer": 13.756906077348066, |
| "step": 754 |
| }, |
| { |
| "epoch": 4.244031830238727, |
| "grad_norm": 1.3886348009109497, |
| "learning_rate": 2.4000000000000003e-06, |
| "loss": 2.0558, |
| "step": 800 |
| }, |
| { |
| "epoch": 4.5092838196286475, |
| "grad_norm": 0.22981444001197815, |
| "learning_rate": 2.55e-06, |
| "loss": 2.0558, |
| "step": 850 |
| }, |
| { |
| "epoch": 4.774535809018568, |
| "grad_norm": 0.3557794988155365, |
| "learning_rate": 2.7e-06, |
| "loss": 2.0554, |
| "step": 900 |
| }, |
| { |
| "epoch": 4.997347480106101, |
| "eval_loss": 2.096412181854248, |
| "eval_norm_levenshtein_similarity": 90.62626262626262, |
| "eval_runtime": 196.2659, |
| "eval_samples_per_second": 1.707, |
| "eval_steps_per_second": 0.428, |
| "eval_wer": 12.154696132596685, |
| "step": 942 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 2632, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 14, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.350424909824e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
