| { | |
| "best_metric": 33.92029657089898, | |
| "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-listen-dsw-small-ml/checkpoint-700", | |
| "epoch": 8.068, | |
| "eval_steps": 100, | |
| "global_step": 700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.898977360288234e-06, | |
| "loss": 1.5105, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.160712527409633e-06, | |
| "loss": 1.0831, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.85912902234906e-06, | |
| "loss": 0.7569, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.344547104469332e-06, | |
| "loss": 0.4577, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.482421875, | |
| "eval_runtime": 4089.3252, | |
| "eval_samples_per_second": 0.162, | |
| "eval_steps_per_second": 0.02, | |
| "eval_wer": 58.38739573679332, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.716963756434345e-06, | |
| "loss": 0.364, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 8.019180844200955e-06, | |
| "loss": 0.2875, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.27351214279797e-06, | |
| "loss": 0.2145, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.49307723936858e-06, | |
| "loss": 0.1781, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 0.306640625, | |
| "eval_runtime": 4087.5604, | |
| "eval_samples_per_second": 0.162, | |
| "eval_steps_per_second": 0.02, | |
| "eval_wer": 41.06580166821131, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.686247975778677e-06, | |
| "loss": 0.1593, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.858694625217149e-06, | |
| "loss": 0.1224, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 9.014436199608479e-06, | |
| "loss": 0.1031, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 9.156425255148058e-06, | |
| "loss": 0.0935, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "eval_loss": 0.290283203125, | |
| "eval_runtime": 4061.9776, | |
| "eval_samples_per_second": 0.163, | |
| "eval_steps_per_second": 0.02, | |
| "eval_wer": 35.644114921223355, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 9.28689473531776e-06, | |
| "loss": 0.0784, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 9.407574351377137e-06, | |
| "loss": 0.0624, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 9.519831289296397e-06, | |
| "loss": 0.0557, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 9.624764935335318e-06, | |
| "loss": 0.057, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.328857421875, | |
| "eval_runtime": 4014.2783, | |
| "eval_samples_per_second": 0.165, | |
| "eval_steps_per_second": 0.021, | |
| "eval_wer": 36.61723818350324, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 9.723272550712454e-06, | |
| "loss": 0.0364, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 9.816095971633122e-06, | |
| "loss": 0.0388, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 9.90385555539545e-06, | |
| "loss": 0.0392, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 9.987075336738768e-06, | |
| "loss": 0.0285, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_loss": 0.342529296875, | |
| "eval_runtime": 4051.7069, | |
| "eval_samples_per_second": 0.164, | |
| "eval_steps_per_second": 0.02, | |
| "eval_wer": 35.31974050046339, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.0279, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 9.100000000000001e-06, | |
| "loss": 0.0287, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.0199, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 8.1e-06, | |
| "loss": 0.0203, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_loss": 0.36474609375, | |
| "eval_runtime": 4154.7446, | |
| "eval_samples_per_second": 0.16, | |
| "eval_steps_per_second": 0.02, | |
| "eval_wer": 34.38368860055607, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.0184, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 7.100000000000001e-06, | |
| "loss": 0.0117, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 0.0098, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 6.1e-06, | |
| "loss": 0.0103, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "eval_loss": 0.382568359375, | |
| "eval_runtime": 4054.0209, | |
| "eval_samples_per_second": 0.164, | |
| "eval_steps_per_second": 0.02, | |
| "eval_wer": 33.92029657089898, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "step": 700, | |
| "total_flos": 3.213687048481276e+18, | |
| "train_loss": 0.0, | |
| "train_runtime": 19.7929, | |
| "train_samples_per_second": 808.369, | |
| "train_steps_per_second": 50.523 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 100, | |
| "total_flos": 3.213687048481276e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |