| { | |
| "best_metric": 54.37442075996293, | |
| "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-dsw-tiny-ml/checkpoint-600", | |
| "epoch": 24.04, | |
| "eval_steps": 100, | |
| "global_step": 1000, | |
| "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": 1.982, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.160712527409633e-06, | |
| "loss": 1.4663, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.85912902234906e-06, | |
| "loss": 1.2536, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 7.344547104469332e-06, | |
| "loss": 1.1736, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_loss": 1.1669921875, | |
| "eval_runtime": 744.9075, | |
| "eval_samples_per_second": 0.89, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 99.7775718257646, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.716963756434345e-06, | |
| "loss": 1.0945, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.019180844200955e-06, | |
| "loss": 1.058, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 8.27351214279797e-06, | |
| "loss": 0.9986, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 8.49307723936858e-06, | |
| "loss": 0.9647, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_loss": 1.0048828125, | |
| "eval_runtime": 741.5505, | |
| "eval_samples_per_second": 0.894, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 95.48656163113995, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 8.686247975778677e-06, | |
| "loss": 0.8708, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 8.858694625217149e-06, | |
| "loss": 0.7678, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 9.014436199608479e-06, | |
| "loss": 0.6767, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 9.156425255148058e-06, | |
| "loss": 0.5311, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_loss": 0.6806640625, | |
| "eval_runtime": 738.6797, | |
| "eval_samples_per_second": 0.898, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 74.55977757182578, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 9.28689473531776e-06, | |
| "loss": 0.4551, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 9.407574351377137e-06, | |
| "loss": 0.3868, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 9.519831289296397e-06, | |
| "loss": 0.3221, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 9.624764935335318e-06, | |
| "loss": 0.3036, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "eval_loss": 0.541015625, | |
| "eval_runtime": 740.1324, | |
| "eval_samples_per_second": 0.896, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 61.575532900834105, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 9.723272550712454e-06, | |
| "loss": 0.2357, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 9.816095971633122e-06, | |
| "loss": 0.2188, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 9.90385555539545e-06, | |
| "loss": 0.1986, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 9.987075336738768e-06, | |
| "loss": 0.1672, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "eval_loss": 0.5146484375, | |
| "eval_runtime": 740.3934, | |
| "eval_samples_per_second": 0.895, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 56.570898980537535, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.148, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 9.100000000000001e-06, | |
| "loss": 0.1238, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.1068, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 8.1e-06, | |
| "loss": 0.1006, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "eval_loss": 0.55029296875, | |
| "eval_runtime": 738.8532, | |
| "eval_samples_per_second": 0.897, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 54.37442075996293, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.0787, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 7.100000000000001e-06, | |
| "loss": 0.0677, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 0.0644, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 6.1e-06, | |
| "loss": 0.0484, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "eval_loss": 0.5859375, | |
| "eval_runtime": 738.8622, | |
| "eval_samples_per_second": 0.897, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 54.50417052826692, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.0474, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 18.03, | |
| "learning_rate": 5.1e-06, | |
| "loss": 0.037, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 0.0315, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 4.1e-06, | |
| "loss": 0.0305, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "eval_loss": 0.65625, | |
| "eval_runtime": 739.3222, | |
| "eval_samples_per_second": 0.897, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 55.41241890639481, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.0211, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 3.1000000000000004e-06, | |
| "loss": 0.0205, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 21.04, | |
| "learning_rate": 2.6e-06, | |
| "loss": 0.0181, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 2.1000000000000002e-06, | |
| "loss": 0.0147, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "eval_loss": 0.70947265625, | |
| "eval_runtime": 740.6818, | |
| "eval_samples_per_second": 0.895, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 54.87488415199259, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.0145, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 23.03, | |
| "learning_rate": 1.1e-06, | |
| "loss": 0.0121, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.0105, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 24.04, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 0.0116, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 24.04, | |
| "eval_loss": 0.73828125, | |
| "eval_runtime": 740.4643, | |
| "eval_samples_per_second": 0.895, | |
| "eval_steps_per_second": 0.057, | |
| "eval_wer": 55.097312326227986, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 24.04, | |
| "step": 1000, | |
| "total_flos": 7.730329851579597e+17, | |
| "train_loss": 0.0, | |
| "train_runtime": 26.9074, | |
| "train_samples_per_second": 2378.53, | |
| "train_steps_per_second": 74.329 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 100, | |
| "total_flos": 7.730329851579597e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |