| { | |
| "best_metric": 0.2754187285900116, | |
| "best_model_checkpoint": "/models/hfhub/DewiBrynJones/wav2vec2-xlsr-53-ft-ccv-en-cy/checkpoint-9000", | |
| "epoch": 4.524886877828054, | |
| "eval_steps": 500, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.8404507637023926, | |
| "learning_rate": 0.000147375, | |
| "loss": 6.0574, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 2.0296876430511475, | |
| "eval_runtime": 1138.2592, | |
| "eval_samples_per_second": 23.042, | |
| "eval_steps_per_second": 0.72, | |
| "eval_wer": 0.9990915399659327, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.7007296085357666, | |
| "learning_rate": 0.00029737499999999995, | |
| "loss": 1.224, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.5368403792381287, | |
| "eval_runtime": 1086.8898, | |
| "eval_samples_per_second": 24.131, | |
| "eval_steps_per_second": 0.754, | |
| "eval_wer": 0.4342018379492356, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.096679449081421, | |
| "learning_rate": 0.0002856219512195122, | |
| "loss": 0.434, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.48607054352760315, | |
| "eval_runtime": 1099.4965, | |
| "eval_samples_per_second": 23.855, | |
| "eval_steps_per_second": 0.746, | |
| "eval_wer": 0.3890522154227914, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.0235676765441895, | |
| "learning_rate": 0.00027098780487804874, | |
| "loss": 0.3643, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.7186101078987122, | |
| "learning_rate": 0.00025635365853658536, | |
| "loss": 0.3295, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.4301180839538574, | |
| "eval_runtime": 1082.408, | |
| "eval_samples_per_second": 24.231, | |
| "eval_steps_per_second": 0.758, | |
| "eval_wer": 0.34114777195970813, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 1.1324162483215332, | |
| "learning_rate": 0.00024171951219512195, | |
| "loss": 0.2739, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 0.381789892911911, | |
| "eval_runtime": 1080.4732, | |
| "eval_samples_per_second": 24.275, | |
| "eval_steps_per_second": 0.759, | |
| "eval_wer": 0.30533089394990853, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 0.6504969596862793, | |
| "learning_rate": 0.0002270853658536585, | |
| "loss": 0.2619, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 0.38941365480422974, | |
| "eval_runtime": 1085.3029, | |
| "eval_samples_per_second": 24.167, | |
| "eval_steps_per_second": 0.756, | |
| "eval_wer": 0.30601644480895, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 0.8715857863426208, | |
| "learning_rate": 0.0002124512195121951, | |
| "loss": 0.2517, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.3497116267681122, | |
| "eval_runtime": 1075.6705, | |
| "eval_samples_per_second": 24.383, | |
| "eval_steps_per_second": 0.762, | |
| "eval_wer": 0.2801926271738902, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 0.5574731826782227, | |
| "learning_rate": 0.0001978170731707317, | |
| "loss": 0.2346, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 1.0655726194381714, | |
| "learning_rate": 0.00018318292682926828, | |
| "loss": 0.2244, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.35193705558776855, | |
| "eval_runtime": 1078.0396, | |
| "eval_samples_per_second": 24.329, | |
| "eval_steps_per_second": 0.761, | |
| "eval_wer": 0.2792000504700019, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.5482127070426941, | |
| "learning_rate": 0.0001685487804878049, | |
| "loss": 0.1854, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 0.33764052391052246, | |
| "eval_runtime": 1076.667, | |
| "eval_samples_per_second": 24.36, | |
| "eval_steps_per_second": 0.762, | |
| "eval_wer": 0.2718104010262234, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 0.6764945387840271, | |
| "learning_rate": 0.00015391463414634145, | |
| "loss": 0.1779, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 0.32059213519096375, | |
| "eval_runtime": 1088.6325, | |
| "eval_samples_per_second": 24.093, | |
| "eval_steps_per_second": 0.753, | |
| "eval_wer": 0.25195466111496645, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 1.0232322216033936, | |
| "learning_rate": 0.00013928048780487804, | |
| "loss": 0.1749, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 0.31690067052841187, | |
| "eval_runtime": 1101.7225, | |
| "eval_samples_per_second": 23.806, | |
| "eval_steps_per_second": 0.744, | |
| "eval_wer": 0.2534687611717452, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 0.6185225248336792, | |
| "learning_rate": 0.00012464634146341463, | |
| "loss": 0.172, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 0.9978949427604675, | |
| "learning_rate": 0.0001100121951219512, | |
| "loss": 0.1636, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 0.3122297525405884, | |
| "eval_runtime": 1102.4428, | |
| "eval_samples_per_second": 23.791, | |
| "eval_steps_per_second": 0.744, | |
| "eval_wer": 0.24648707757659874, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 0.9038313031196594, | |
| "learning_rate": 9.53780487804878e-05, | |
| "loss": 0.137, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "eval_loss": 0.30537155270576477, | |
| "eval_runtime": 1108.9259, | |
| "eval_samples_per_second": 23.652, | |
| "eval_steps_per_second": 0.739, | |
| "eval_wer": 0.23823523226715454, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 0.9038735032081604, | |
| "learning_rate": 8.074390243902438e-05, | |
| "loss": 0.1311, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_loss": 0.29557299613952637, | |
| "eval_runtime": 1090.6163, | |
| "eval_samples_per_second": 24.049, | |
| "eval_steps_per_second": 0.752, | |
| "eval_wer": 0.22798141021596954, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "grad_norm": 0.523389995098114, | |
| "learning_rate": 6.610975609756097e-05, | |
| "loss": 0.1261, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 0.28975382447242737, | |
| "eval_runtime": 1108.1062, | |
| "eval_samples_per_second": 23.669, | |
| "eval_steps_per_second": 0.74, | |
| "eval_wer": 0.22361575505225748, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 0.8056386709213257, | |
| "learning_rate": 5.147560975609756e-05, | |
| "loss": 0.1242, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 1.5024261474609375, | |
| "learning_rate": 3.684146341463414e-05, | |
| "loss": 0.1187, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_loss": 0.28465163707733154, | |
| "eval_runtime": 1092.1505, | |
| "eval_samples_per_second": 24.015, | |
| "eval_steps_per_second": 0.751, | |
| "eval_wer": 0.21763085399449036, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 1.5841491222381592, | |
| "learning_rate": 2.2207317073170727e-05, | |
| "loss": 0.1011, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_loss": 0.2762569189071655, | |
| "eval_runtime": 1102.2478, | |
| "eval_samples_per_second": 23.795, | |
| "eval_steps_per_second": 0.744, | |
| "eval_wer": 0.21240720879860367, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "grad_norm": 0.9246074557304382, | |
| "learning_rate": 7.573170731707317e-06, | |
| "loss": 0.0981, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "eval_loss": 0.2754187285900116, | |
| "eval_runtime": 1092.5524, | |
| "eval_samples_per_second": 24.006, | |
| "eval_steps_per_second": 0.751, | |
| "eval_wer": 0.2115155720985006, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "step": 9000, | |
| "total_flos": 8.933861078537978e+19, | |
| "train_loss": 0.5072881503634983, | |
| "train_runtime": 52839.7574, | |
| "train_samples_per_second": 10.901, | |
| "train_steps_per_second": 0.17 | |
| } | |
| ], | |
| "logging_steps": 400, | |
| "max_steps": 9000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 8.933861078537978e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |