| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 6480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 3.359571933746338, | |
| "eval_runtime": 1.8892, | |
| "eval_samples_per_second": 52.934, | |
| "eval_steps_per_second": 6.881, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 2.9279825687408447, | |
| "eval_runtime": 1.8734, | |
| "eval_samples_per_second": 53.378, | |
| "eval_steps_per_second": 6.939, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 1.5090594291687012, | |
| "eval_runtime": 1.876, | |
| "eval_samples_per_second": 53.306, | |
| "eval_steps_per_second": 6.93, | |
| "eval_wer": 0.9649737302977233, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.994255006313324, | |
| "eval_runtime": 1.8636, | |
| "eval_samples_per_second": 53.659, | |
| "eval_steps_per_second": 6.976, | |
| "eval_wer": 0.9176882661996497, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0002982, | |
| "loss": 3.1184, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 0.7590276598930359, | |
| "eval_runtime": 1.9271, | |
| "eval_samples_per_second": 51.893, | |
| "eval_steps_per_second": 6.746, | |
| "eval_wer": 0.7793345008756567, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": 0.7336117625236511, | |
| "eval_runtime": 1.8782, | |
| "eval_samples_per_second": 53.242, | |
| "eval_steps_per_second": 6.921, | |
| "eval_wer": 0.7408056042031523, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 0.7039847373962402, | |
| "eval_runtime": 1.8772, | |
| "eval_samples_per_second": 53.27, | |
| "eval_steps_per_second": 6.925, | |
| "eval_wer": 0.7618213660245184, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 0.6815319061279297, | |
| "eval_runtime": 1.8953, | |
| "eval_samples_per_second": 52.762, | |
| "eval_steps_per_second": 6.859, | |
| "eval_wer": 0.723292469352014, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 0.6456977128982544, | |
| "eval_runtime": 1.8847, | |
| "eval_samples_per_second": 53.058, | |
| "eval_steps_per_second": 6.898, | |
| "eval_wer": 0.6865148861646234, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.00027511705685618725, | |
| "loss": 0.7917, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_loss": 0.5704939365386963, | |
| "eval_runtime": 1.8802, | |
| "eval_samples_per_second": 53.187, | |
| "eval_steps_per_second": 6.914, | |
| "eval_wer": 0.681260945709282, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.5708367824554443, | |
| "eval_runtime": 1.8769, | |
| "eval_samples_per_second": 53.28, | |
| "eval_steps_per_second": 6.926, | |
| "eval_wer": 0.6619964973730298, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_loss": 0.5888168811798096, | |
| "eval_runtime": 1.8801, | |
| "eval_samples_per_second": 53.188, | |
| "eval_steps_per_second": 6.914, | |
| "eval_wer": 0.6462346760070052, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_loss": 0.6509290337562561, | |
| "eval_runtime": 1.8971, | |
| "eval_samples_per_second": 52.711, | |
| "eval_steps_per_second": 6.852, | |
| "eval_wer": 0.6970227670753065, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_loss": 0.5871497392654419, | |
| "eval_runtime": 1.8883, | |
| "eval_samples_per_second": 52.957, | |
| "eval_steps_per_second": 6.884, | |
| "eval_wer": 0.6462346760070052, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.00025008361204013375, | |
| "loss": 0.5909, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "eval_loss": 0.619912326335907, | |
| "eval_runtime": 1.8867, | |
| "eval_samples_per_second": 53.002, | |
| "eval_steps_per_second": 6.89, | |
| "eval_wer": 0.681260945709282, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 0.6230051517486572, | |
| "eval_runtime": 1.8866, | |
| "eval_samples_per_second": 53.006, | |
| "eval_steps_per_second": 6.891, | |
| "eval_wer": 0.5919439579684763, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_loss": 0.572126030921936, | |
| "eval_runtime": 1.896, | |
| "eval_samples_per_second": 52.741, | |
| "eval_steps_per_second": 6.856, | |
| "eval_wer": 0.6427320490367776, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_loss": 0.5330815315246582, | |
| "eval_runtime": 1.8961, | |
| "eval_samples_per_second": 52.741, | |
| "eval_steps_per_second": 6.856, | |
| "eval_wer": 0.5866900175131349, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "eval_loss": 0.5560976266860962, | |
| "eval_runtime": 1.9173, | |
| "eval_samples_per_second": 52.157, | |
| "eval_steps_per_second": 6.78, | |
| "eval_wer": 0.6007005253940455, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.00022510033444816051, | |
| "loss": 0.4607, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "eval_loss": 0.5414324402809143, | |
| "eval_runtime": 1.8832, | |
| "eval_samples_per_second": 53.102, | |
| "eval_steps_per_second": 6.903, | |
| "eval_wer": 0.5849387040280211, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_loss": 0.5389540791511536, | |
| "eval_runtime": 1.8801, | |
| "eval_samples_per_second": 53.19, | |
| "eval_steps_per_second": 6.915, | |
| "eval_wer": 0.5586690017513135, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "eval_loss": 0.5313354730606079, | |
| "eval_runtime": 1.8925, | |
| "eval_samples_per_second": 52.842, | |
| "eval_steps_per_second": 6.869, | |
| "eval_wer": 0.5569176882661997, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "eval_loss": 0.5893406271934509, | |
| "eval_runtime": 1.8839, | |
| "eval_samples_per_second": 53.083, | |
| "eval_steps_per_second": 6.901, | |
| "eval_wer": 0.5796847635726795, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_loss": 0.550654411315918, | |
| "eval_runtime": 1.8741, | |
| "eval_samples_per_second": 53.359, | |
| "eval_steps_per_second": 6.937, | |
| "eval_wer": 0.5954465849387041, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 0.00020006688963210701, | |
| "loss": 0.3933, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "eval_loss": 0.552050769329071, | |
| "eval_runtime": 1.8818, | |
| "eval_samples_per_second": 53.14, | |
| "eval_steps_per_second": 6.908, | |
| "eval_wer": 0.6024518388791593, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "eval_loss": 0.5662926435470581, | |
| "eval_runtime": 1.9296, | |
| "eval_samples_per_second": 51.824, | |
| "eval_steps_per_second": 6.737, | |
| "eval_wer": 0.5989492119089317, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_loss": 0.5636402368545532, | |
| "eval_runtime": 1.8763, | |
| "eval_samples_per_second": 53.297, | |
| "eval_steps_per_second": 6.929, | |
| "eval_wer": 0.5831873905429071, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "eval_loss": 0.5463794469833374, | |
| "eval_runtime": 1.8753, | |
| "eval_samples_per_second": 53.325, | |
| "eval_steps_per_second": 6.932, | |
| "eval_wer": 0.5919439579684763, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "eval_loss": 0.5623293519020081, | |
| "eval_runtime": 1.8826, | |
| "eval_samples_per_second": 53.118, | |
| "eval_steps_per_second": 6.905, | |
| "eval_wer": 0.5831873905429071, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.00017503344481605351, | |
| "loss": 0.3367, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "eval_loss": 0.5324317812919617, | |
| "eval_runtime": 1.8849, | |
| "eval_samples_per_second": 53.053, | |
| "eval_steps_per_second": 6.897, | |
| "eval_wer": 0.5691768826619965, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "eval_loss": 0.5907294750213623, | |
| "eval_runtime": 1.8884, | |
| "eval_samples_per_second": 52.954, | |
| "eval_steps_per_second": 6.884, | |
| "eval_wer": 0.5394045534150613, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 0.5653238892555237, | |
| "eval_runtime": 1.8757, | |
| "eval_samples_per_second": 53.314, | |
| "eval_steps_per_second": 6.931, | |
| "eval_wer": 0.5814360770577933, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "eval_loss": 0.5707294940948486, | |
| "eval_runtime": 1.9074, | |
| "eval_samples_per_second": 52.427, | |
| "eval_steps_per_second": 6.816, | |
| "eval_wer": 0.5814360770577933, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "eval_loss": 0.5753923654556274, | |
| "eval_runtime": 1.8899, | |
| "eval_samples_per_second": 52.912, | |
| "eval_steps_per_second": 6.879, | |
| "eval_wer": 0.542907180385289, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 0.00015, | |
| "loss": 0.2856, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_loss": 0.5953063368797302, | |
| "eval_runtime": 1.8933, | |
| "eval_samples_per_second": 52.818, | |
| "eval_steps_per_second": 6.866, | |
| "eval_wer": 0.5569176882661997, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_loss": 0.6274660229682922, | |
| "eval_runtime": 1.8873, | |
| "eval_samples_per_second": 52.985, | |
| "eval_steps_per_second": 6.888, | |
| "eval_wer": 0.5394045534150613, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "eval_loss": 0.6253136992454529, | |
| "eval_runtime": 1.8922, | |
| "eval_samples_per_second": 52.847, | |
| "eval_steps_per_second": 6.87, | |
| "eval_wer": 0.5569176882661997, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_loss": 0.5930343866348267, | |
| "eval_runtime": 1.9165, | |
| "eval_samples_per_second": 52.177, | |
| "eval_steps_per_second": 6.783, | |
| "eval_wer": 0.542907180385289, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_loss": 0.6082107424736023, | |
| "eval_runtime": 1.9067, | |
| "eval_samples_per_second": 52.448, | |
| "eval_steps_per_second": 6.818, | |
| "eval_wer": 0.521891418563923, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 0.00012496655518394646, | |
| "loss": 0.2522, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "eval_loss": 0.6026180982589722, | |
| "eval_runtime": 1.9243, | |
| "eval_samples_per_second": 51.967, | |
| "eval_steps_per_second": 6.756, | |
| "eval_wer": 0.5446584938704028, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "eval_loss": 0.6052154302597046, | |
| "eval_runtime": 1.9113, | |
| "eval_samples_per_second": 52.32, | |
| "eval_steps_per_second": 6.802, | |
| "eval_wer": 0.5271453590192644, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "eval_loss": 0.5870827436447144, | |
| "eval_runtime": 1.8877, | |
| "eval_samples_per_second": 52.973, | |
| "eval_steps_per_second": 6.887, | |
| "eval_wer": 0.521891418563923, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_loss": 0.5870257019996643, | |
| "eval_runtime": 1.8862, | |
| "eval_samples_per_second": 53.017, | |
| "eval_steps_per_second": 6.892, | |
| "eval_wer": 0.5236427320490368, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "eval_loss": 0.5880929231643677, | |
| "eval_runtime": 1.9034, | |
| "eval_samples_per_second": 52.538, | |
| "eval_steps_per_second": 6.83, | |
| "eval_wer": 0.5131348511383538, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 9.993311036789297e-05, | |
| "loss": 0.2167, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "eval_loss": 0.6122171878814697, | |
| "eval_runtime": 1.8922, | |
| "eval_samples_per_second": 52.847, | |
| "eval_steps_per_second": 6.87, | |
| "eval_wer": 0.5288966725043783, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "eval_loss": 0.612826406955719, | |
| "eval_runtime": 1.891, | |
| "eval_samples_per_second": 52.882, | |
| "eval_steps_per_second": 6.875, | |
| "eval_wer": 0.5166374781085814, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_loss": 0.6134529113769531, | |
| "eval_runtime": 1.891, | |
| "eval_samples_per_second": 52.883, | |
| "eval_steps_per_second": 6.875, | |
| "eval_wer": 0.5376532399299475, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "eval_loss": 0.6054602265357971, | |
| "eval_runtime": 1.8917, | |
| "eval_samples_per_second": 52.863, | |
| "eval_steps_per_second": 6.872, | |
| "eval_wer": 0.5183887915936952, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 11.34, | |
| "eval_loss": 0.6724901795387268, | |
| "eval_runtime": 1.8847, | |
| "eval_samples_per_second": 53.058, | |
| "eval_steps_per_second": 6.898, | |
| "eval_wer": 0.5569176882661997, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "learning_rate": 7.489966555183946e-05, | |
| "loss": 0.1965, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "eval_loss": 0.6481964588165283, | |
| "eval_runtime": 1.9092, | |
| "eval_samples_per_second": 52.378, | |
| "eval_steps_per_second": 6.809, | |
| "eval_wer": 0.542907180385289, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 11.81, | |
| "eval_loss": 0.6037153601646423, | |
| "eval_runtime": 1.8885, | |
| "eval_samples_per_second": 52.953, | |
| "eval_steps_per_second": 6.884, | |
| "eval_wer": 0.5096322241681261, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "eval_loss": 0.5931165218353271, | |
| "eval_runtime": 1.9024, | |
| "eval_samples_per_second": 52.566, | |
| "eval_steps_per_second": 6.834, | |
| "eval_wer": 0.5131348511383538, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "eval_loss": 0.5853209495544434, | |
| "eval_runtime": 1.9149, | |
| "eval_samples_per_second": 52.223, | |
| "eval_steps_per_second": 6.789, | |
| "eval_wer": 0.51138353765324, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_loss": 0.5798044800758362, | |
| "eval_runtime": 1.9229, | |
| "eval_samples_per_second": 52.006, | |
| "eval_steps_per_second": 6.761, | |
| "eval_wer": 0.521891418563923, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 4.991638795986622e-05, | |
| "loss": 0.172, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "eval_loss": 0.5774720907211304, | |
| "eval_runtime": 1.8968, | |
| "eval_samples_per_second": 52.72, | |
| "eval_steps_per_second": 6.854, | |
| "eval_wer": 0.500875656742557, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_loss": 0.5782347321510315, | |
| "eval_runtime": 1.9216, | |
| "eval_samples_per_second": 52.041, | |
| "eval_steps_per_second": 6.765, | |
| "eval_wer": 0.5043782837127846, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 13.19, | |
| "eval_loss": 0.5803666114807129, | |
| "eval_runtime": 1.9303, | |
| "eval_samples_per_second": 51.806, | |
| "eval_steps_per_second": 6.735, | |
| "eval_wer": 0.5183887915936952, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "eval_loss": 0.5976961255073547, | |
| "eval_runtime": 1.8886, | |
| "eval_samples_per_second": 52.948, | |
| "eval_steps_per_second": 6.883, | |
| "eval_wer": 0.521891418563923, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "eval_loss": 0.6069247722625732, | |
| "eval_runtime": 1.9188, | |
| "eval_samples_per_second": 52.117, | |
| "eval_steps_per_second": 6.775, | |
| "eval_wer": 0.5236427320490368, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 2.488294314381271e-05, | |
| "loss": 0.1622, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "eval_loss": 0.5849621295928955, | |
| "eval_runtime": 1.8831, | |
| "eval_samples_per_second": 53.103, | |
| "eval_steps_per_second": 6.903, | |
| "eval_wer": 0.5131348511383538, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "eval_loss": 0.5757902264595032, | |
| "eval_runtime": 1.9024, | |
| "eval_samples_per_second": 52.566, | |
| "eval_steps_per_second": 6.834, | |
| "eval_wer": 0.5096322241681261, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "eval_loss": 0.5752313137054443, | |
| "eval_runtime": 1.8854, | |
| "eval_samples_per_second": 53.038, | |
| "eval_steps_per_second": 6.895, | |
| "eval_wer": 0.500875656742557, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "eval_loss": 0.5727171301841736, | |
| "eval_runtime": 1.9071, | |
| "eval_samples_per_second": 52.437, | |
| "eval_steps_per_second": 6.817, | |
| "eval_wer": 0.5183887915936952, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "eval_loss": 0.5795398354530334, | |
| "eval_runtime": 1.8851, | |
| "eval_samples_per_second": 53.048, | |
| "eval_steps_per_second": 6.896, | |
| "eval_wer": 0.5043782837127846, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 6480, | |
| "total_flos": 1.5048474592820595e+19, | |
| "train_loss": 0.5518865008413055, | |
| "train_runtime": 5863.5283, | |
| "train_samples_per_second": 35.331, | |
| "train_steps_per_second": 1.105 | |
| } | |
| ], | |
| "max_steps": 6480, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.5048474592820595e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |