| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 13100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002969690721649484, | |
| "loss": 1.0671, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 0.3079470694065094, | |
| "eval_runtime": 237.7627, | |
| "eval_samples_per_second": 35.073, | |
| "eval_steps_per_second": 4.387, | |
| "eval_wer": 0.2752062328139322, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 0.00029078350515463917, | |
| "loss": 0.6433, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "eval_loss": 0.27281925082206726, | |
| "eval_runtime": 238.3892, | |
| "eval_samples_per_second": 34.981, | |
| "eval_steps_per_second": 4.375, | |
| "eval_wer": 0.2847616865261228, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.0002845979381443299, | |
| "loss": 0.5687, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "eval_loss": 0.28817421197891235, | |
| "eval_runtime": 238.0657, | |
| "eval_samples_per_second": 35.028, | |
| "eval_steps_per_second": 4.381, | |
| "eval_wer": 0.3036205316223648, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 0.00027841237113402056, | |
| "loss": 0.5355, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "eval_loss": 0.27776163816452026, | |
| "eval_runtime": 239.8353, | |
| "eval_samples_per_second": 34.77, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.29200274977085244, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 0.00027222680412371134, | |
| "loss": 0.5116, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "eval_loss": 0.2905969023704529, | |
| "eval_runtime": 240.1935, | |
| "eval_samples_per_second": 34.718, | |
| "eval_steps_per_second": 4.342, | |
| "eval_wer": 0.3013978001833181, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 0.00027468461538461536, | |
| "loss": 0.5313, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "eval_loss": 0.2984345555305481, | |
| "eval_runtime": 234.3486, | |
| "eval_samples_per_second": 35.584, | |
| "eval_steps_per_second": 4.451, | |
| "eval_wer": 0.327314390467461, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 0.00027006923076923077, | |
| "loss": 0.4996, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "eval_loss": 0.3169882297515869, | |
| "eval_runtime": 237.8663, | |
| "eval_samples_per_second": 35.058, | |
| "eval_steps_per_second": 4.385, | |
| "eval_wer": 0.3344179651695692, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.0002654538461538461, | |
| "loss": 0.4845, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "eval_loss": 0.32016345858573914, | |
| "eval_runtime": 236.9291, | |
| "eval_samples_per_second": 35.196, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.36338221814848765, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 0.00026086153846153847, | |
| "loss": 0.5092, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "eval_loss": 0.3166552186012268, | |
| "eval_runtime": 236.2482, | |
| "eval_samples_per_second": 35.298, | |
| "eval_steps_per_second": 4.415, | |
| "eval_wer": 0.3373052245646196, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 0.0002562461538461538, | |
| "loss": 0.4777, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "eval_loss": 0.32921522855758667, | |
| "eval_runtime": 235.517, | |
| "eval_samples_per_second": 35.407, | |
| "eval_steps_per_second": 4.429, | |
| "eval_wer": 0.3385655362053162, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 0.0002516307692307692, | |
| "loss": 0.4651, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "eval_loss": 0.3070097863674164, | |
| "eval_runtime": 238.1366, | |
| "eval_samples_per_second": 35.018, | |
| "eval_steps_per_second": 4.38, | |
| "eval_wer": 0.34271310724106324, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 18.32, | |
| "learning_rate": 0.0002470153846153846, | |
| "loss": 0.461, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.32, | |
| "eval_loss": 0.3148922026157379, | |
| "eval_runtime": 237.1783, | |
| "eval_samples_per_second": 35.159, | |
| "eval_steps_per_second": 4.398, | |
| "eval_wer": 0.35609532538955085, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 0.00024239999999999998, | |
| "loss": 0.4481, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "eval_loss": 0.32919421792030334, | |
| "eval_runtime": 236.0044, | |
| "eval_samples_per_second": 35.334, | |
| "eval_steps_per_second": 4.419, | |
| "eval_wer": 0.34411090742438133, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 21.37, | |
| "learning_rate": 0.00023778461538461536, | |
| "loss": 0.4479, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.37, | |
| "eval_loss": 0.3142247200012207, | |
| "eval_runtime": 234.6363, | |
| "eval_samples_per_second": 35.54, | |
| "eval_steps_per_second": 4.445, | |
| "eval_wer": 0.32089825847846015, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 22.9, | |
| "learning_rate": 0.00023316923076923077, | |
| "loss": 0.4305, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.9, | |
| "eval_loss": 0.3525453507900238, | |
| "eval_runtime": 236.821, | |
| "eval_samples_per_second": 35.212, | |
| "eval_steps_per_second": 4.404, | |
| "eval_wer": 0.35467461044912924, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.43, | |
| "learning_rate": 0.00022855384615384612, | |
| "loss": 0.4254, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 24.43, | |
| "eval_loss": 0.34136688709259033, | |
| "eval_runtime": 235.4909, | |
| "eval_samples_per_second": 35.411, | |
| "eval_steps_per_second": 4.429, | |
| "eval_wer": 0.3400091659028414, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 25.95, | |
| "learning_rate": 0.00022393846153846153, | |
| "loss": 0.4066, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 25.95, | |
| "eval_loss": 0.3118491470813751, | |
| "eval_runtime": 236.1268, | |
| "eval_samples_per_second": 35.316, | |
| "eval_steps_per_second": 4.417, | |
| "eval_wer": 0.3207378551787351, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 27.48, | |
| "learning_rate": 0.0002193230769230769, | |
| "loss": 0.4043, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 27.48, | |
| "eval_loss": 0.34181562066078186, | |
| "eval_runtime": 235.2934, | |
| "eval_samples_per_second": 35.441, | |
| "eval_steps_per_second": 4.433, | |
| "eval_wer": 0.3482584784601283, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 0.0002147076923076923, | |
| "loss": 0.3985, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "eval_loss": 0.32544735074043274, | |
| "eval_runtime": 236.4374, | |
| "eval_samples_per_second": 35.269, | |
| "eval_steps_per_second": 4.411, | |
| "eval_wer": 0.31663611365719524, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 30.53, | |
| "learning_rate": 0.00021009230769230766, | |
| "loss": 0.3982, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 30.53, | |
| "eval_loss": 0.33056947588920593, | |
| "eval_runtime": 239.8848, | |
| "eval_samples_per_second": 34.763, | |
| "eval_steps_per_second": 4.348, | |
| "eval_wer": 0.3452795600366636, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 32.06, | |
| "learning_rate": 0.00020547692307692307, | |
| "loss": 0.3929, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 32.06, | |
| "eval_loss": 0.3262433409690857, | |
| "eval_runtime": 238.1544, | |
| "eval_samples_per_second": 35.015, | |
| "eval_steps_per_second": 4.38, | |
| "eval_wer": 0.3228689275893675, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 33.59, | |
| "learning_rate": 0.00020086153846153845, | |
| "loss": 0.378, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 33.59, | |
| "eval_loss": 0.3545904755592346, | |
| "eval_runtime": 234.4177, | |
| "eval_samples_per_second": 35.573, | |
| "eval_steps_per_second": 4.449, | |
| "eval_wer": 0.33357011915673696, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 35.11, | |
| "learning_rate": 0.00019624615384615385, | |
| "loss": 0.4062, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 35.11, | |
| "eval_loss": 0.3174082636833191, | |
| "eval_runtime": 235.3658, | |
| "eval_samples_per_second": 35.43, | |
| "eval_steps_per_second": 4.431, | |
| "eval_wer": 0.34566911090742436, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 36.64, | |
| "learning_rate": 0.0001916307692307692, | |
| "loss": 0.3648, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 36.64, | |
| "eval_loss": 0.3376729190349579, | |
| "eval_runtime": 239.2202, | |
| "eval_samples_per_second": 34.859, | |
| "eval_steps_per_second": 4.36, | |
| "eval_wer": 0.33572410632447297, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 38.17, | |
| "learning_rate": 0.0001870153846153846, | |
| "loss": 0.3609, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 38.17, | |
| "eval_loss": 0.33455467224121094, | |
| "eval_runtime": 236.9279, | |
| "eval_samples_per_second": 35.196, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.351993583868011, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 39.69, | |
| "learning_rate": 0.0001824, | |
| "loss": 0.3483, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 39.69, | |
| "eval_loss": 0.3349843919277191, | |
| "eval_runtime": 235.432, | |
| "eval_samples_per_second": 35.42, | |
| "eval_steps_per_second": 4.43, | |
| "eval_wer": 0.3525893675527039, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 41.22, | |
| "learning_rate": 0.0001777846153846154, | |
| "loss": 0.3548, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 41.22, | |
| "eval_loss": 0.33302220702171326, | |
| "eval_runtime": 237.3922, | |
| "eval_samples_per_second": 35.128, | |
| "eval_steps_per_second": 4.394, | |
| "eval_wer": 0.34062786434463793, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 42.75, | |
| "learning_rate": 0.00017316923076923075, | |
| "loss": 0.3446, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 42.75, | |
| "eval_loss": 0.3398281931877136, | |
| "eval_runtime": 236.2326, | |
| "eval_samples_per_second": 35.3, | |
| "eval_steps_per_second": 4.415, | |
| "eval_wer": 0.3372135655362053, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 44.27, | |
| "learning_rate": 0.00016855384615384615, | |
| "loss": 0.3346, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 44.27, | |
| "eval_loss": 0.34486979246139526, | |
| "eval_runtime": 236.4721, | |
| "eval_samples_per_second": 35.264, | |
| "eval_steps_per_second": 4.411, | |
| "eval_wer": 0.3287809349220898, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 45.8, | |
| "learning_rate": 0.00016393846153846153, | |
| "loss": 0.3309, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 45.8, | |
| "eval_loss": 0.3319507837295532, | |
| "eval_runtime": 236.6263, | |
| "eval_samples_per_second": 35.241, | |
| "eval_steps_per_second": 4.408, | |
| "eval_wer": 0.31439046746104493, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 47.33, | |
| "learning_rate": 0.0001593230769230769, | |
| "loss": 0.326, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 47.33, | |
| "eval_loss": 0.3399747312068939, | |
| "eval_runtime": 236.0765, | |
| "eval_samples_per_second": 35.323, | |
| "eval_steps_per_second": 4.418, | |
| "eval_wer": 0.32786434463794684, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 48.85, | |
| "learning_rate": 0.0001547076923076923, | |
| "loss": 0.3189, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 48.85, | |
| "eval_loss": 0.3399554491043091, | |
| "eval_runtime": 236.2028, | |
| "eval_samples_per_second": 35.304, | |
| "eval_steps_per_second": 4.416, | |
| "eval_wer": 0.31500916590284145, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 50.38, | |
| "learning_rate": 0.0001500923076923077, | |
| "loss": 0.3165, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 50.38, | |
| "eval_loss": 0.3359447121620178, | |
| "eval_runtime": 235.0525, | |
| "eval_samples_per_second": 35.477, | |
| "eval_steps_per_second": 4.437, | |
| "eval_wer": 0.29945004582951423, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 51.91, | |
| "learning_rate": 0.00014547692307692305, | |
| "loss": 0.3132, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 51.91, | |
| "eval_loss": 0.3342697024345398, | |
| "eval_runtime": 235.4774, | |
| "eval_samples_per_second": 35.413, | |
| "eval_steps_per_second": 4.429, | |
| "eval_wer": 0.3095554537121907, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 53.44, | |
| "learning_rate": 0.00014086153846153845, | |
| "loss": 0.3092, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 53.44, | |
| "eval_loss": 0.3224042057991028, | |
| "eval_runtime": 237.4291, | |
| "eval_samples_per_second": 35.122, | |
| "eval_steps_per_second": 4.393, | |
| "eval_wer": 0.302910174152154, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 54.96, | |
| "learning_rate": 0.00013624615384615383, | |
| "loss": 0.2995, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 54.96, | |
| "eval_loss": 0.3204595744609833, | |
| "eval_runtime": 236.2646, | |
| "eval_samples_per_second": 35.295, | |
| "eval_steps_per_second": 4.415, | |
| "eval_wer": 0.29851054078826766, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 56.49, | |
| "learning_rate": 0.0001316307692307692, | |
| "loss": 0.304, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 56.49, | |
| "eval_loss": 0.35227909684181213, | |
| "eval_runtime": 236.3256, | |
| "eval_samples_per_second": 35.286, | |
| "eval_steps_per_second": 4.413, | |
| "eval_wer": 0.30339138405132904, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 58.02, | |
| "learning_rate": 0.0001270153846153846, | |
| "loss": 0.2952, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 58.02, | |
| "eval_loss": 0.3288583755493164, | |
| "eval_runtime": 238.857, | |
| "eval_samples_per_second": 34.912, | |
| "eval_steps_per_second": 4.367, | |
| "eval_wer": 0.2934005499541705, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 59.54, | |
| "learning_rate": 0.0001224, | |
| "loss": 0.2875, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 59.54, | |
| "eval_loss": 0.335005521774292, | |
| "eval_runtime": 236.8029, | |
| "eval_samples_per_second": 35.215, | |
| "eval_steps_per_second": 4.405, | |
| "eval_wer": 0.3008020164986251, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 61.07, | |
| "learning_rate": 0.00011778461538461537, | |
| "loss": 0.2868, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 61.07, | |
| "eval_loss": 0.3537150025367737, | |
| "eval_runtime": 236.5754, | |
| "eval_samples_per_second": 35.249, | |
| "eval_steps_per_second": 4.409, | |
| "eval_wer": 0.3227314390467461, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 62.6, | |
| "learning_rate": 0.00011316923076923076, | |
| "loss": 0.2875, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 62.6, | |
| "eval_loss": 0.3389277756214142, | |
| "eval_runtime": 234.9228, | |
| "eval_samples_per_second": 35.497, | |
| "eval_steps_per_second": 4.44, | |
| "eval_wer": 0.29704399633363887, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 64.12, | |
| "learning_rate": 0.00010855384615384616, | |
| "loss": 0.2778, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 64.12, | |
| "eval_loss": 0.33703726530075073, | |
| "eval_runtime": 234.6165, | |
| "eval_samples_per_second": 35.543, | |
| "eval_steps_per_second": 4.446, | |
| "eval_wer": 0.29596700274977084, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 65.65, | |
| "learning_rate": 0.00010393846153846154, | |
| "loss": 0.2706, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 65.65, | |
| "eval_loss": 0.32503727078437805, | |
| "eval_runtime": 234.7831, | |
| "eval_samples_per_second": 35.518, | |
| "eval_steps_per_second": 4.442, | |
| "eval_wer": 0.28015582034830433, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 67.18, | |
| "learning_rate": 9.934615384615383e-05, | |
| "loss": 0.2669, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 67.18, | |
| "eval_loss": 0.335059255361557, | |
| "eval_runtime": 234.788, | |
| "eval_samples_per_second": 35.517, | |
| "eval_steps_per_second": 4.442, | |
| "eval_wer": 0.2902841429880843, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 68.7, | |
| "learning_rate": 9.473076923076922e-05, | |
| "loss": 0.2615, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 68.7, | |
| "eval_loss": 0.3381515145301819, | |
| "eval_runtime": 236.5391, | |
| "eval_samples_per_second": 35.254, | |
| "eval_steps_per_second": 4.409, | |
| "eval_wer": 0.29885426214482125, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 70.23, | |
| "learning_rate": 9.01153846153846e-05, | |
| "loss": 0.2563, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 70.23, | |
| "eval_loss": 0.3312215507030487, | |
| "eval_runtime": 236.7043, | |
| "eval_samples_per_second": 35.23, | |
| "eval_steps_per_second": 4.406, | |
| "eval_wer": 0.2974793767186068, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 71.76, | |
| "learning_rate": 8.549999999999999e-05, | |
| "loss": 0.2546, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 71.76, | |
| "eval_loss": 0.3212486505508423, | |
| "eval_runtime": 238.4241, | |
| "eval_samples_per_second": 34.975, | |
| "eval_steps_per_second": 4.375, | |
| "eval_wer": 0.30034372135655363, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 73.28, | |
| "learning_rate": 8.088461538461537e-05, | |
| "loss": 0.2482, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 73.28, | |
| "eval_loss": 0.3337170481681824, | |
| "eval_runtime": 237.0835, | |
| "eval_samples_per_second": 35.173, | |
| "eval_steps_per_second": 4.399, | |
| "eval_wer": 0.30907424381301557, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 74.81, | |
| "learning_rate": 7.626923076923075e-05, | |
| "loss": 0.2504, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 74.81, | |
| "eval_loss": 0.33080631494522095, | |
| "eval_runtime": 236.549, | |
| "eval_samples_per_second": 35.253, | |
| "eval_steps_per_second": 4.409, | |
| "eval_wer": 0.3109761686526123, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 76.34, | |
| "learning_rate": 7.165384615384615e-05, | |
| "loss": 0.2456, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 76.34, | |
| "eval_loss": 0.31574100255966187, | |
| "eval_runtime": 235.6572, | |
| "eval_samples_per_second": 35.386, | |
| "eval_steps_per_second": 4.426, | |
| "eval_wer": 0.3117781851512374, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 77.86, | |
| "learning_rate": 6.703846153846153e-05, | |
| "loss": 0.2363, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 77.86, | |
| "eval_loss": 0.3251018524169922, | |
| "eval_runtime": 236.3835, | |
| "eval_samples_per_second": 35.277, | |
| "eval_steps_per_second": 4.412, | |
| "eval_wer": 0.31439046746104493, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 79.39, | |
| "learning_rate": 6.242307692307691e-05, | |
| "loss": 0.2319, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 79.39, | |
| "eval_loss": 0.32527828216552734, | |
| "eval_runtime": 236.1364, | |
| "eval_samples_per_second": 35.314, | |
| "eval_steps_per_second": 4.417, | |
| "eval_wer": 0.30382676443629697, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 80.92, | |
| "learning_rate": 5.7807692307692304e-05, | |
| "loss": 0.2266, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 80.92, | |
| "eval_loss": 0.3374153673648834, | |
| "eval_runtime": 236.6995, | |
| "eval_samples_per_second": 35.23, | |
| "eval_steps_per_second": 4.406, | |
| "eval_wer": 0.30382676443629697, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 82.44, | |
| "learning_rate": 5.321538461538461e-05, | |
| "loss": 0.2279, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 82.44, | |
| "eval_loss": 0.32676786184310913, | |
| "eval_runtime": 235.2934, | |
| "eval_samples_per_second": 35.441, | |
| "eval_steps_per_second": 4.433, | |
| "eval_wer": 0.29640238313473877, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 83.97, | |
| "learning_rate": 4.8599999999999995e-05, | |
| "loss": 0.2231, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 83.97, | |
| "eval_loss": 0.32775917649269104, | |
| "eval_runtime": 237.2828, | |
| "eval_samples_per_second": 35.144, | |
| "eval_steps_per_second": 4.396, | |
| "eval_wer": 0.29502749770852427, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 85.5, | |
| "learning_rate": 4.398461538461538e-05, | |
| "loss": 0.2185, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 85.5, | |
| "eval_loss": 0.3461829721927643, | |
| "eval_runtime": 234.8207, | |
| "eval_samples_per_second": 35.512, | |
| "eval_steps_per_second": 4.442, | |
| "eval_wer": 0.29814390467461044, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 87.02, | |
| "learning_rate": 3.9369230769230767e-05, | |
| "loss": 0.2245, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 87.02, | |
| "eval_loss": 0.3311368525028229, | |
| "eval_runtime": 234.6868, | |
| "eval_samples_per_second": 35.532, | |
| "eval_steps_per_second": 4.444, | |
| "eval_wer": 0.2894821264894592, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 88.55, | |
| "learning_rate": 3.475384615384615e-05, | |
| "loss": 0.223, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 88.55, | |
| "eval_loss": 0.3325417935848236, | |
| "eval_runtime": 234.9788, | |
| "eval_samples_per_second": 35.488, | |
| "eval_steps_per_second": 4.439, | |
| "eval_wer": 0.2876947754353804, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 90.08, | |
| "learning_rate": 3.0138461538461538e-05, | |
| "loss": 0.2121, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 90.08, | |
| "eval_loss": 0.333699494600296, | |
| "eval_runtime": 237.3406, | |
| "eval_samples_per_second": 35.135, | |
| "eval_steps_per_second": 4.395, | |
| "eval_wer": 0.282813932172319, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 91.6, | |
| "learning_rate": 2.552307692307692e-05, | |
| "loss": 0.2126, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 91.6, | |
| "eval_loss": 0.3324645757675171, | |
| "eval_runtime": 234.7938, | |
| "eval_samples_per_second": 35.516, | |
| "eval_steps_per_second": 4.442, | |
| "eval_wer": 0.28077451879010085, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 93.13, | |
| "learning_rate": 2.0907692307692305e-05, | |
| "loss": 0.2027, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 93.13, | |
| "eval_loss": 0.3276507258415222, | |
| "eval_runtime": 235.1021, | |
| "eval_samples_per_second": 35.47, | |
| "eval_steps_per_second": 4.436, | |
| "eval_wer": 0.2819660861594867, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 94.66, | |
| "learning_rate": 1.629230769230769e-05, | |
| "loss": 0.2058, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 94.66, | |
| "eval_loss": 0.33078595995903015, | |
| "eval_runtime": 236.8707, | |
| "eval_samples_per_second": 35.205, | |
| "eval_steps_per_second": 4.403, | |
| "eval_wer": 0.2827451879010083, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 96.18, | |
| "learning_rate": 1.1676923076923075e-05, | |
| "loss": 0.1991, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 96.18, | |
| "eval_loss": 0.3278521001338959, | |
| "eval_runtime": 235.6753, | |
| "eval_samples_per_second": 35.383, | |
| "eval_steps_per_second": 4.426, | |
| "eval_wer": 0.2820119156736939, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 97.71, | |
| "learning_rate": 7.061538461538461e-06, | |
| "loss": 0.1991, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 97.71, | |
| "eval_loss": 0.3299693167209625, | |
| "eval_runtime": 236.8407, | |
| "eval_samples_per_second": 35.209, | |
| "eval_steps_per_second": 4.404, | |
| "eval_wer": 0.28221814848762605, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 99.24, | |
| "learning_rate": 2.446153846153846e-06, | |
| "loss": 0.1986, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 99.24, | |
| "eval_loss": 0.3284846246242523, | |
| "eval_runtime": 236.4429, | |
| "eval_samples_per_second": 35.269, | |
| "eval_steps_per_second": 4.411, | |
| "eval_wer": 0.2834555453712191, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 13100, | |
| "total_flos": 8.36308166572502e+19, | |
| "train_loss": 0.29656382378731067, | |
| "train_runtime": 73649.3567, | |
| "train_samples_per_second": 34.023, | |
| "train_steps_per_second": 0.178 | |
| } | |
| ], | |
| "max_steps": 13100, | |
| "num_train_epochs": 100, | |
| "total_flos": 8.36308166572502e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |