| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 36.7816091954023, | |
| "global_step": 6400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": NaN, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 3.315488815307617, | |
| "eval_runtime": 135.9943, | |
| "eval_samples_per_second": 11.192, | |
| "eval_wer": 0.9802160088826083, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.00025054945054945053, | |
| "loss": NaN, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "eval_loss": 1.239622712135315, | |
| "eval_runtime": 136.5249, | |
| "eval_samples_per_second": 11.148, | |
| "eval_wer": 0.8971434339356011, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 0.0001846153846153846, | |
| "loss": NaN, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.5218076705932617, | |
| "eval_runtime": 136.039, | |
| "eval_samples_per_second": 11.188, | |
| "eval_wer": 0.5810033309780963, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 0.00011868131868131866, | |
| "loss": NaN, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.45067664980888367, | |
| "eval_runtime": 136.2412, | |
| "eval_samples_per_second": 11.171, | |
| "eval_wer": 0.5031795700010094, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 5.274725274725275e-05, | |
| "loss": NaN, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 0.4806972146034241, | |
| "eval_runtime": 135.9031, | |
| "eval_samples_per_second": 11.199, | |
| "eval_wer": 0.48622186332895934, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 0.00021176470588235295, | |
| "loss": 1.706, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.4303271174430847, | |
| "eval_runtime": 143.5562, | |
| "eval_samples_per_second": 10.602, | |
| "eval_wer": 0.4874331280912486, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 0.00019318885448916406, | |
| "loss": 0.2803, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "eval_loss": 0.39102795720100403, | |
| "eval_runtime": 144.4788, | |
| "eval_samples_per_second": 10.534, | |
| "eval_wer": 0.4415060058544463, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 0.0001746130030959752, | |
| "loss": 0.2162, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "eval_loss": 0.3744983971118927, | |
| "eval_runtime": 146.0666, | |
| "eval_samples_per_second": 10.42, | |
| "eval_wer": 0.41728071060866057, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 20.69, | |
| "learning_rate": 0.00015603715170278636, | |
| "loss": 0.1768, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.69, | |
| "eval_loss": 0.4224047064781189, | |
| "eval_runtime": 145.7761, | |
| "eval_samples_per_second": 10.441, | |
| "eval_wer": 0.4194004239426668, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "learning_rate": 0.00013746130030959752, | |
| "loss": 0.1502, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "eval_loss": 0.37262818217277527, | |
| "eval_runtime": 145.6131, | |
| "eval_samples_per_second": 10.452, | |
| "eval_wer": 0.4009286363177551, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 25.29, | |
| "learning_rate": 0.00011888544891640867, | |
| "loss": 0.1214, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 25.29, | |
| "eval_loss": 0.3383936583995819, | |
| "eval_runtime": 146.7077, | |
| "eval_samples_per_second": 10.374, | |
| "eval_wer": 0.3776117896436863, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 27.59, | |
| "learning_rate": 0.00010030959752321979, | |
| "loss": 0.105, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 27.59, | |
| "eval_loss": 0.3606471121311188, | |
| "eval_runtime": 146.6063, | |
| "eval_samples_per_second": 10.382, | |
| "eval_wer": 0.3755930150398708, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 29.89, | |
| "learning_rate": 8.173374613003094e-05, | |
| "loss": 0.0917, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 29.89, | |
| "eval_loss": 0.35728928446769714, | |
| "eval_runtime": 147.1176, | |
| "eval_samples_per_second": 10.345, | |
| "eval_wer": 0.3658019582113657, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 32.18, | |
| "learning_rate": 6.315789473684209e-05, | |
| "loss": 0.0782, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 32.18, | |
| "eval_loss": 0.36559608578681946, | |
| "eval_runtime": 148.3984, | |
| "eval_samples_per_second": 10.256, | |
| "eval_wer": 0.3630766124962148, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 34.48, | |
| "learning_rate": 4.4582043343653246e-05, | |
| "loss": 0.0713, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 34.48, | |
| "eval_loss": 0.3449805676937103, | |
| "eval_runtime": 147.7614, | |
| "eval_samples_per_second": 10.3, | |
| "eval_wer": 0.35611184011305136, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 36.78, | |
| "learning_rate": 2.6006191950464395e-05, | |
| "loss": 0.0666, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 36.78, | |
| "eval_loss": 0.3429848849773407, | |
| "eval_runtime": 147.6625, | |
| "eval_samples_per_second": 10.307, | |
| "eval_wer": 0.3475320480468356, | |
| "step": 6400 | |
| } | |
| ], | |
| "max_steps": 6960, | |
| "num_train_epochs": 40, | |
| "total_flos": 3.238027552454268e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |