| { | |
| "best_metric": 4.375102519989014, | |
| "best_model_checkpoint": "nrshoudi/hubert-large-ll60k_arabic/checkpoint-16370", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 32740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9816737935247405e-05, | |
| "loss": 15.0257, | |
| "step": 1637 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 4.074129104614258, | |
| "eval_per": 1.0, | |
| "eval_runtime": 81.2424, | |
| "eval_samples_per_second": 10.007, | |
| "eval_steps_per_second": 5.01, | |
| "eval_wer": 1.0, | |
| "step": 1637 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.981673793524741e-05, | |
| "loss": 4.0101, | |
| "step": 3274 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 4.04423189163208, | |
| "eval_per": 1.0, | |
| "eval_runtime": 83.0026, | |
| "eval_samples_per_second": 9.795, | |
| "eval_steps_per_second": 4.903, | |
| "eval_wer": 1.0, | |
| "step": 3274 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.446480689608362e-05, | |
| "loss": 3.9446, | |
| "step": 4911 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 4.0444488525390625, | |
| "eval_per": 1.0, | |
| "eval_runtime": 84.4804, | |
| "eval_samples_per_second": 9.624, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 1.0, | |
| "step": 4911 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.890925134052806e-05, | |
| "loss": 3.9302, | |
| "step": 6548 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 4.044007778167725, | |
| "eval_per": 1.0, | |
| "eval_runtime": 86.94, | |
| "eval_samples_per_second": 9.351, | |
| "eval_steps_per_second": 4.681, | |
| "eval_wer": 1.0, | |
| "step": 6548 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.335369578497252e-05, | |
| "loss": 3.9456, | |
| "step": 8185 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 4.045283317565918, | |
| "eval_per": 1.0, | |
| "eval_runtime": 86.4841, | |
| "eval_samples_per_second": 9.401, | |
| "eval_steps_per_second": 4.706, | |
| "eval_wer": 1.0, | |
| "step": 8185 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.779814022941696e-05, | |
| "loss": 4.0388, | |
| "step": 9822 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 4.041508197784424, | |
| "eval_per": 1.0, | |
| "eval_runtime": 88.0752, | |
| "eval_samples_per_second": 9.231, | |
| "eval_steps_per_second": 4.621, | |
| "eval_wer": 1.0, | |
| "step": 9822 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.224597841580127e-05, | |
| "loss": 3.9297, | |
| "step": 11459 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 4.044224739074707, | |
| "eval_per": 1.0, | |
| "eval_runtime": 88.7389, | |
| "eval_samples_per_second": 9.162, | |
| "eval_steps_per_second": 4.586, | |
| "eval_wer": 1.0, | |
| "step": 11459 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.669042286024571e-05, | |
| "loss": 3.9443, | |
| "step": 13096 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 4.039473533630371, | |
| "eval_per": 1.0, | |
| "eval_runtime": 88.5683, | |
| "eval_samples_per_second": 9.179, | |
| "eval_steps_per_second": 4.595, | |
| "eval_wer": 1.0, | |
| "step": 13096 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.113486730469016e-05, | |
| "loss": 3.9362, | |
| "step": 14733 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 4.064728260040283, | |
| "eval_per": 0.9999764439837935, | |
| "eval_runtime": 90.2258, | |
| "eval_samples_per_second": 9.011, | |
| "eval_steps_per_second": 4.511, | |
| "eval_wer": 0.9999533386216228, | |
| "step": 14733 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5.558270549107446e-05, | |
| "loss": 3.9785, | |
| "step": 16370 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 4.375102519989014, | |
| "eval_per": 1.0, | |
| "eval_runtime": 89.0532, | |
| "eval_samples_per_second": 9.129, | |
| "eval_steps_per_second": 4.57, | |
| "eval_wer": 1.0, | |
| "step": 16370 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5.003733116133849e-05, | |
| "loss": 3.9934, | |
| "step": 18007 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 4.029526233673096, | |
| "eval_per": 1.0, | |
| "eval_runtime": 93.2708, | |
| "eval_samples_per_second": 8.717, | |
| "eval_steps_per_second": 4.364, | |
| "eval_wer": 1.0, | |
| "step": 18007 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.448177560578294e-05, | |
| "loss": 3.9361, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 4.027557373046875, | |
| "eval_per": 1.0, | |
| "eval_runtime": 94.3966, | |
| "eval_samples_per_second": 8.613, | |
| "eval_steps_per_second": 4.312, | |
| "eval_wer": 1.0, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.893300753410711e-05, | |
| "loss": 3.913, | |
| "step": 21281 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 4.023890972137451, | |
| "eval_per": 1.0, | |
| "eval_runtime": 96.1974, | |
| "eval_samples_per_second": 8.451, | |
| "eval_steps_per_second": 4.231, | |
| "eval_wer": 1.0, | |
| "step": 21281 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.337745197855155e-05, | |
| "loss": 3.9023, | |
| "step": 22918 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 4.022657871246338, | |
| "eval_per": 1.0, | |
| "eval_runtime": 97.2278, | |
| "eval_samples_per_second": 8.362, | |
| "eval_steps_per_second": 4.186, | |
| "eval_wer": 1.0, | |
| "step": 22918 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.7821896422995997e-05, | |
| "loss": 3.8957, | |
| "step": 24555 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 4.022843360900879, | |
| "eval_per": 1.0, | |
| "eval_runtime": 97.5296, | |
| "eval_samples_per_second": 8.336, | |
| "eval_steps_per_second": 4.173, | |
| "eval_wer": 1.0, | |
| "step": 24555 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.226634086744044e-05, | |
| "loss": 3.9021, | |
| "step": 26192 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 4.02068567276001, | |
| "eval_per": 1.0, | |
| "eval_runtime": 96.5808, | |
| "eval_samples_per_second": 8.418, | |
| "eval_steps_per_second": 4.214, | |
| "eval_wer": 1.0, | |
| "step": 26192 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.6714179053824745e-05, | |
| "loss": 3.8897, | |
| "step": 27829 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 4.02132511138916, | |
| "eval_per": 1.0, | |
| "eval_runtime": 97.8428, | |
| "eval_samples_per_second": 8.309, | |
| "eval_steps_per_second": 4.16, | |
| "eval_wer": 1.0, | |
| "step": 27829 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.1158623498269193e-05, | |
| "loss": 3.8939, | |
| "step": 29466 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 4.0213165283203125, | |
| "eval_per": 1.0, | |
| "eval_runtime": 97.7024, | |
| "eval_samples_per_second": 8.321, | |
| "eval_steps_per_second": 4.166, | |
| "eval_wer": 1.0, | |
| "step": 29466 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5.609855426593362e-06, | |
| "loss": 3.8949, | |
| "step": 31103 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 4.020359992980957, | |
| "eval_per": 1.0, | |
| "eval_runtime": 96.6299, | |
| "eval_samples_per_second": 8.414, | |
| "eval_steps_per_second": 4.212, | |
| "eval_wer": 1.0, | |
| "step": 31103 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.429987103780629e-08, | |
| "loss": 3.8987, | |
| "step": 32740 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 4.019679069519043, | |
| "eval_per": 1.0, | |
| "eval_runtime": 98.2061, | |
| "eval_samples_per_second": 8.279, | |
| "eval_steps_per_second": 4.144, | |
| "eval_wer": 1.0, | |
| "step": 32740 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 32740, | |
| "total_flos": 1.0578414234417158e+19, | |
| "train_loss": 4.4901714383208615, | |
| "train_runtime": 11782.1292, | |
| "train_samples_per_second": 5.556, | |
| "train_steps_per_second": 2.779 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 32740, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.0578414234417158e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |