| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.361930294906166, | |
| "eval_steps": 200, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.9598, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.3436, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.8791261911392212, | |
| "eval_runtime": 1327.8534, | |
| "eval_samples_per_second": 0.753, | |
| "eval_steps_per_second": 0.047, | |
| "eval_wer": 0.8870904221802143, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 6e-06, | |
| "loss": 1.5788, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.1682, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 1.0307379961013794, | |
| "eval_runtime": 1238.5009, | |
| "eval_samples_per_second": 0.807, | |
| "eval_steps_per_second": 0.051, | |
| "eval_wer": 0.5047652804032766, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9354, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.7321, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.6299881935119629, | |
| "eval_runtime": 830.7332, | |
| "eval_samples_per_second": 1.204, | |
| "eval_steps_per_second": 0.076, | |
| "eval_wer": 0.36645400126023947, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.5397, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.4564, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 0.438092440366745, | |
| "eval_runtime": 723.2192, | |
| "eval_samples_per_second": 1.383, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.35148865784499056, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.4291, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.4095, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_loss": 0.40272918343544006, | |
| "eval_runtime": 636.5063, | |
| "eval_samples_per_second": 1.571, | |
| "eval_steps_per_second": 0.099, | |
| "eval_wer": 0.33297889098928796, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.3992, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.3813, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_loss": 0.3847169280052185, | |
| "eval_runtime": 616.7138, | |
| "eval_samples_per_second": 1.621, | |
| "eval_steps_per_second": 0.102, | |
| "eval_wer": 0.3359719596723377, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 0.3711, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.3667, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.37336310744285583, | |
| "eval_runtime": 614.3217, | |
| "eval_samples_per_second": 1.628, | |
| "eval_steps_per_second": 0.103, | |
| "eval_wer": 0.33916194076874606, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.3654, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.3583, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_loss": 0.3648846447467804, | |
| "eval_runtime": 617.5084, | |
| "eval_samples_per_second": 1.619, | |
| "eval_steps_per_second": 0.102, | |
| "eval_wer": 0.34904694391934465, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 0.353, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.3454, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.35879915952682495, | |
| "eval_runtime": 620.7465, | |
| "eval_samples_per_second": 1.611, | |
| "eval_steps_per_second": 0.101, | |
| "eval_wer": 0.35715973534971646, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 0.3399, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.3422, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_loss": 0.3536700904369354, | |
| "eval_runtime": 625.9316, | |
| "eval_samples_per_second": 1.598, | |
| "eval_steps_per_second": 0.101, | |
| "eval_wer": 0.3704710144927536, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 0.3355, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.3371, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "eval_loss": 0.3503468930721283, | |
| "eval_runtime": 634.2679, | |
| "eval_samples_per_second": 1.577, | |
| "eval_steps_per_second": 0.099, | |
| "eval_wer": 0.3811436672967864, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.3311, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.3291, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "eval_loss": 0.347513347864151, | |
| "eval_runtime": 615.3576, | |
| "eval_samples_per_second": 1.625, | |
| "eval_steps_per_second": 0.102, | |
| "eval_wer": 0.3677930056710775, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.3273, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.324, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "eval_loss": 0.3451197147369385, | |
| "eval_runtime": 628.249, | |
| "eval_samples_per_second": 1.592, | |
| "eval_steps_per_second": 0.1, | |
| "eval_wer": 0.36700535601764334, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 0.3219, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.3262, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.34306031465530396, | |
| "eval_runtime": 613.0254, | |
| "eval_samples_per_second": 1.631, | |
| "eval_steps_per_second": 0.103, | |
| "eval_wer": 0.3710223692501575, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 0.3203, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.3168, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "eval_loss": 0.34187400341033936, | |
| "eval_runtime": 628.864, | |
| "eval_samples_per_second": 1.59, | |
| "eval_steps_per_second": 0.1, | |
| "eval_wer": 0.3847274732199118, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 200, | |
| "total_flos": 1.118753215008768e+20, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |