| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.997802197802198, | |
| "global_step": 6810, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 11.6291, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 3.028644323348999, | |
| "eval_runtime": 643.9816, | |
| "eval_samples_per_second": 8.087, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0002952456418383518, | |
| "loss": 3.0796, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 2.9614171981811523, | |
| "eval_runtime": 650.4972, | |
| "eval_samples_per_second": 8.006, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 0.0002809825673534073, | |
| "loss": 1.6756, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_loss": 0.6643654108047485, | |
| "eval_runtime": 668.5248, | |
| "eval_samples_per_second": 7.79, | |
| "eval_wer": 0.5919852387016294, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.00026671949286846274, | |
| "loss": 0.5169, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "eval_loss": 0.48998841643333435, | |
| "eval_runtime": 670.8355, | |
| "eval_samples_per_second": 7.763, | |
| "eval_wer": 0.4740247927440942, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.0002524564183835182, | |
| "loss": 0.3643, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "eval_loss": 0.4465296268463135, | |
| "eval_runtime": 685.664, | |
| "eval_samples_per_second": 7.596, | |
| "eval_wer": 0.43660178279061307, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 0.00023819334389857367, | |
| "loss": 0.277, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "eval_loss": 0.4225097894668579, | |
| "eval_runtime": 675.9712, | |
| "eval_samples_per_second": 7.704, | |
| "eval_wer": 0.4157592453026326, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.00022393026941362913, | |
| "loss": 0.2271, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "eval_loss": 0.4406384825706482, | |
| "eval_runtime": 685.3056, | |
| "eval_samples_per_second": 7.6, | |
| "eval_wer": 0.39260375789391616, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 10.57, | |
| "learning_rate": 0.0002096671949286846, | |
| "loss": 0.1895, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 10.57, | |
| "eval_loss": 0.4478096067905426, | |
| "eval_runtime": 689.1958, | |
| "eval_samples_per_second": 7.557, | |
| "eval_wer": 0.3899529613555446, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 0.0001954041204437401, | |
| "loss": 0.1653, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "eval_loss": 0.4374055862426758, | |
| "eval_runtime": 693.8794, | |
| "eval_samples_per_second": 7.506, | |
| "eval_wer": 0.3792977987993451, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "learning_rate": 0.00018114104595879556, | |
| "loss": 0.1366, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "eval_loss": 0.4575331211090088, | |
| "eval_runtime": 703.7067, | |
| "eval_samples_per_second": 7.401, | |
| "eval_wer": 0.37355440629954, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.54, | |
| "learning_rate": 0.000166877971473851, | |
| "loss": 0.1219, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 14.54, | |
| "eval_loss": 0.4847738444805145, | |
| "eval_runtime": 705.6195, | |
| "eval_samples_per_second": 7.381, | |
| "eval_wer": 0.36983809350554847, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 0.00015261489698890646, | |
| "loss": 0.1102, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "eval_loss": 0.4509291350841522, | |
| "eval_runtime": 713.2815, | |
| "eval_samples_per_second": 7.301, | |
| "eval_wer": 0.35738974505574467, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 17.18, | |
| "learning_rate": 0.00013835182250396196, | |
| "loss": 0.1017, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 17.18, | |
| "eval_loss": 0.45042145252227783, | |
| "eval_runtime": 714.3616, | |
| "eval_samples_per_second": 7.29, | |
| "eval_wer": 0.34715039372125056, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 0.00012408874801901742, | |
| "loss": 0.0937, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "eval_loss": 0.4522610008716583, | |
| "eval_runtime": 719.9965, | |
| "eval_samples_per_second": 7.233, | |
| "eval_wer": 0.3469164999090413, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 19.82, | |
| "learning_rate": 0.00010982567353407289, | |
| "loss": 0.0829, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 19.82, | |
| "eval_loss": 0.4488724172115326, | |
| "eval_runtime": 720.1162, | |
| "eval_samples_per_second": 7.232, | |
| "eval_wer": 0.3452012786195067, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "learning_rate": 9.556259904912835e-05, | |
| "loss": 0.0752, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "eval_loss": 0.4762090742588043, | |
| "eval_runtime": 730.1963, | |
| "eval_samples_per_second": 7.132, | |
| "eval_wer": 0.33800254684373293, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 22.47, | |
| "learning_rate": 8.129952456418383e-05, | |
| "loss": 0.0698, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 22.47, | |
| "eval_loss": 0.480989009141922, | |
| "eval_runtime": 727.7194, | |
| "eval_samples_per_second": 7.157, | |
| "eval_wer": 0.33597546713791937, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 23.79, | |
| "learning_rate": 6.70364500792393e-05, | |
| "loss": 0.0642, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 23.79, | |
| "eval_loss": 0.49236759543418884, | |
| "eval_runtime": 737.7861, | |
| "eval_samples_per_second": 7.059, | |
| "eval_wer": 0.3343382104524546, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 25.11, | |
| "learning_rate": 5.2773375594294765e-05, | |
| "loss": 0.058, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 25.11, | |
| "eval_loss": 0.4960506558418274, | |
| "eval_runtime": 740.7836, | |
| "eval_samples_per_second": 7.03, | |
| "eval_wer": 0.3302840510408275, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 26.43, | |
| "learning_rate": 3.851030110935023e-05, | |
| "loss": 0.0553, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 26.43, | |
| "eval_loss": 0.4915299415588379, | |
| "eval_runtime": 748.2447, | |
| "eval_samples_per_second": 6.96, | |
| "eval_wer": 0.32506042256815404, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 27.75, | |
| "learning_rate": 2.4247226624405704e-05, | |
| "loss": 0.0545, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 27.75, | |
| "eval_loss": 0.5011223554611206, | |
| "eval_runtime": 762.4791, | |
| "eval_samples_per_second": 6.83, | |
| "eval_wer": 0.3210062631565269, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 29.07, | |
| "learning_rate": 9.984152139461172e-06, | |
| "loss": 0.0523, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 29.07, | |
| "eval_loss": 0.49885958433151245, | |
| "eval_runtime": 743.9824, | |
| "eval_samples_per_second": 7.0, | |
| "eval_wer": 0.3229034018555576, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 6810, | |
| "total_flos": 4.998078272539758e+19, | |
| "train_runtime": 61300.1017, | |
| "train_samples_per_second": 0.111 | |
| } | |
| ], | |
| "max_steps": 6810, | |
| "num_train_epochs": 30, | |
| "total_flos": 4.998078272539758e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |