| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 45.45340050377834, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 0.00039436231366470835, | |
| "loss": 1.5121, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_loss": 0.186998650431633, | |
| "eval_runtime": 714.138, | |
| "eval_samples_per_second": 7.242, | |
| "eval_wer": 0.39502654080485317, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "learning_rate": 0.00037776708973098473, | |
| "loss": 0.1108, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "eval_loss": 0.16734005510807037, | |
| "eval_runtime": 670.8576, | |
| "eval_samples_per_second": 7.71, | |
| "eval_wer": 0.33428339826896425, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 0.0003511499148708517, | |
| "loss": 0.0666, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "eval_loss": 0.17631034553050995, | |
| "eval_runtime": 758.94, | |
| "eval_samples_per_second": 6.815, | |
| "eval_wer": 0.3243469393091546, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "learning_rate": 0.00031601138191423965, | |
| "loss": 0.0459, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "eval_loss": 0.2086898684501648, | |
| "eval_runtime": 682.0381, | |
| "eval_samples_per_second": 7.583, | |
| "eval_wer": 0.32364092775148395, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "learning_rate": 0.0002743324911320655, | |
| "loss": 0.0358, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "eval_loss": 0.1979219615459442, | |
| "eval_runtime": 700.5817, | |
| "eval_samples_per_second": 7.382, | |
| "eval_wer": 0.3092069136835499, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 0.00022846296765465705, | |
| "loss": 0.0285, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "eval_loss": 0.19415870308876038, | |
| "eval_runtime": 1037.6087, | |
| "eval_samples_per_second": 4.985, | |
| "eval_wer": 0.3047093585754256, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.51, | |
| "learning_rate": 0.0001809887913391635, | |
| "loss": 0.0237, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 26.51, | |
| "eval_loss": 0.1929541826248169, | |
| "eval_runtime": 764.504, | |
| "eval_samples_per_second": 6.765, | |
| "eval_wer": 0.30026410061972125, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 30.3, | |
| "learning_rate": 0.0001345864073365157, | |
| "loss": 0.0205, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 30.3, | |
| "eval_loss": 0.2030186504125595, | |
| "eval_runtime": 1540.1942, | |
| "eval_samples_per_second": 3.358, | |
| "eval_wer": 0.29338702507648456, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "learning_rate": 9.187183650888056e-05, | |
| "loss": 0.0174, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "eval_loss": 0.20308123528957367, | |
| "eval_runtime": 964.2383, | |
| "eval_samples_per_second": 5.364, | |
| "eval_wer": 0.29469445388698584, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 37.88, | |
| "learning_rate": 5.5253192378985966e-05, | |
| "loss": 0.0151, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 37.88, | |
| "eval_loss": 0.20086798071861267, | |
| "eval_runtime": 715.6106, | |
| "eval_samples_per_second": 7.227, | |
| "eval_wer": 0.28739900112438876, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 41.66, | |
| "learning_rate": 2.679491924311226e-05, | |
| "loss": 0.0138, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 41.66, | |
| "eval_loss": 0.20132462680339813, | |
| "eval_runtime": 773.3872, | |
| "eval_samples_per_second": 6.687, | |
| "eval_wer": 0.2842350234029757, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "learning_rate": 8.10140527710055e-06, | |
| "loss": 0.0122, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "eval_loss": 0.20383873581886292, | |
| "eval_runtime": 957.0431, | |
| "eval_samples_per_second": 5.404, | |
| "eval_wer": 0.28167246293439324, | |
| "step": 6000 | |
| } | |
| ], | |
| "max_steps": 6600, | |
| "num_train_epochs": 50, | |
| "total_flos": 9.394750356471834e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |