| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1788282447247436, | |
| "eval_steps": 200, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007858854964831625, | |
| "eval_loss": 3.1892831325531006, | |
| "eval_runtime": 159.5257, | |
| "eval_samples_per_second": 35.455, | |
| "eval_steps_per_second": 4.432, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.01571770992966325, | |
| "eval_loss": 2.780208110809326, | |
| "eval_runtime": 157.5706, | |
| "eval_samples_per_second": 35.895, | |
| "eval_steps_per_second": 4.487, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.01964713741207906, | |
| "grad_norm": 4.9997968673706055, | |
| "learning_rate": 0.0002982, | |
| "loss": 4.719, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.023576564894494872, | |
| "eval_loss": 1.4220576286315918, | |
| "eval_runtime": 158.6041, | |
| "eval_samples_per_second": 35.661, | |
| "eval_steps_per_second": 4.458, | |
| "eval_wer": 0.8876923817624497, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.0314354198593265, | |
| "eval_loss": 1.227359414100647, | |
| "eval_runtime": 158.5238, | |
| "eval_samples_per_second": 35.679, | |
| "eval_steps_per_second": 4.46, | |
| "eval_wer": 0.8224390557044503, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.03929427482415812, | |
| "grad_norm": 2.6001393795013428, | |
| "learning_rate": 0.0002949457627118644, | |
| "loss": 1.0441, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03929427482415812, | |
| "eval_loss": 1.1094719171524048, | |
| "eval_runtime": 161.2687, | |
| "eval_samples_per_second": 35.072, | |
| "eval_steps_per_second": 4.384, | |
| "eval_wer": 0.7886729469917029, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.047153129788989744, | |
| "eval_loss": 1.091428279876709, | |
| "eval_runtime": 158.6043, | |
| "eval_samples_per_second": 35.661, | |
| "eval_steps_per_second": 4.458, | |
| "eval_wer": 0.7549228868097125, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.055011984753821366, | |
| "eval_loss": 1.0177329778671265, | |
| "eval_runtime": 159.8661, | |
| "eval_samples_per_second": 35.38, | |
| "eval_steps_per_second": 4.422, | |
| "eval_wer": 0.7354881160629745, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.05894141223623718, | |
| "grad_norm": 2.7494542598724365, | |
| "learning_rate": 0.0002898610169491525, | |
| "loss": 0.8033, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.062870839718653, | |
| "eval_loss": 0.9907466769218445, | |
| "eval_runtime": 159.6438, | |
| "eval_samples_per_second": 35.429, | |
| "eval_steps_per_second": 4.429, | |
| "eval_wer": 0.7232912326876475, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.07072969468348461, | |
| "eval_loss": 0.9761303067207336, | |
| "eval_runtime": 159.1571, | |
| "eval_samples_per_second": 35.537, | |
| "eval_steps_per_second": 4.442, | |
| "eval_wer": 0.7145287348943204, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.07858854964831624, | |
| "grad_norm": 2.753251314163208, | |
| "learning_rate": 0.00028477627118644064, | |
| "loss": 0.7227, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07858854964831624, | |
| "eval_loss": 0.9555273056030273, | |
| "eval_runtime": 159.3414, | |
| "eval_samples_per_second": 35.496, | |
| "eval_steps_per_second": 4.437, | |
| "eval_wer": 0.6902794049204796, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08644740461314787, | |
| "eval_loss": 0.8994919061660767, | |
| "eval_runtime": 159.0426, | |
| "eval_samples_per_second": 35.563, | |
| "eval_steps_per_second": 4.445, | |
| "eval_wer": 0.6747765242092086, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.09430625957797949, | |
| "eval_loss": 0.8897404670715332, | |
| "eval_runtime": 158.5492, | |
| "eval_samples_per_second": 35.673, | |
| "eval_steps_per_second": 4.459, | |
| "eval_wer": 0.66655967646162, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.0982356870603953, | |
| "grad_norm": 2.404499053955078, | |
| "learning_rate": 0.00027969152542372877, | |
| "loss": 0.6794, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.10216511454281112, | |
| "eval_loss": 0.8826168775558472, | |
| "eval_runtime": 159.8456, | |
| "eval_samples_per_second": 35.384, | |
| "eval_steps_per_second": 4.423, | |
| "eval_wer": 0.6559676461619939, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.11002396950764273, | |
| "eval_loss": 0.8744593858718872, | |
| "eval_runtime": 159.6838, | |
| "eval_samples_per_second": 35.42, | |
| "eval_steps_per_second": 4.427, | |
| "eval_wer": 0.6445571407937604, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.11788282447247436, | |
| "grad_norm": 2.406255006790161, | |
| "learning_rate": 0.00027460677966101695, | |
| "loss": 0.6513, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.11788282447247436, | |
| "eval_loss": 0.8450209498405457, | |
| "eval_runtime": 159.2776, | |
| "eval_samples_per_second": 35.51, | |
| "eval_steps_per_second": 4.439, | |
| "eval_wer": 0.6436905201328819, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.125741679437306, | |
| "eval_loss": 0.8596389293670654, | |
| "eval_runtime": 159.3028, | |
| "eval_samples_per_second": 35.505, | |
| "eval_steps_per_second": 4.438, | |
| "eval_wer": 0.6510888928118631, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.13360053440213762, | |
| "eval_loss": 0.8597909212112427, | |
| "eval_runtime": 159.9064, | |
| "eval_samples_per_second": 35.371, | |
| "eval_steps_per_second": 4.421, | |
| "eval_wer": 0.6376402240374893, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.1375299618845534, | |
| "grad_norm": 2.2046961784362793, | |
| "learning_rate": 0.000269522033898305, | |
| "loss": 0.6147, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.14145938936696922, | |
| "eval_loss": 0.8516111969947815, | |
| "eval_runtime": 160.4338, | |
| "eval_samples_per_second": 35.254, | |
| "eval_steps_per_second": 4.407, | |
| "eval_wer": 0.6375439328529473, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.14931824433180085, | |
| "eval_loss": 0.8251617550849915, | |
| "eval_runtime": 160.6004, | |
| "eval_samples_per_second": 35.218, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.6100367511354335, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.15717709929663248, | |
| "grad_norm": 1.520897388458252, | |
| "learning_rate": 0.0002644372881355932, | |
| "loss": 0.6092, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.15717709929663248, | |
| "eval_loss": 0.8579581379890442, | |
| "eval_runtime": 159.0993, | |
| "eval_samples_per_second": 35.55, | |
| "eval_steps_per_second": 4.444, | |
| "eval_wer": 0.6822551395419749, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1650359542614641, | |
| "eval_loss": 0.8204948306083679, | |
| "eval_runtime": 159.818, | |
| "eval_samples_per_second": 35.39, | |
| "eval_steps_per_second": 4.424, | |
| "eval_wer": 0.6135674279019756, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.17289480922629574, | |
| "eval_loss": 0.8033376336097717, | |
| "eval_runtime": 159.2754, | |
| "eval_samples_per_second": 35.511, | |
| "eval_steps_per_second": 4.439, | |
| "eval_wer": 0.6385068446983678, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.17682423670871153, | |
| "grad_norm": 2.3011837005615234, | |
| "learning_rate": 0.00025936271186440674, | |
| "loss": 0.5928, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.18075366419112734, | |
| "eval_loss": 0.7927771210670471, | |
| "eval_runtime": 160.1079, | |
| "eval_samples_per_second": 35.326, | |
| "eval_steps_per_second": 4.416, | |
| "eval_wer": 0.6005039238657701, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.18861251915595897, | |
| "eval_loss": 0.7911030054092407, | |
| "eval_runtime": 160.2559, | |
| "eval_samples_per_second": 35.294, | |
| "eval_steps_per_second": 4.412, | |
| "eval_wer": 0.5923512702412094, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.1964713741207906, | |
| "grad_norm": 6.133739948272705, | |
| "learning_rate": 0.0002542779661016949, | |
| "loss": 0.5681, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1964713741207906, | |
| "eval_loss": 0.7968648076057434, | |
| "eval_runtime": 160.0012, | |
| "eval_samples_per_second": 35.35, | |
| "eval_steps_per_second": 4.419, | |
| "eval_wer": 0.5944375792396206, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.20433022908562223, | |
| "eval_loss": 0.7932958602905273, | |
| "eval_runtime": 159.7818, | |
| "eval_samples_per_second": 35.398, | |
| "eval_steps_per_second": 4.425, | |
| "eval_wer": 0.5898958450353871, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.21218908405045384, | |
| "eval_loss": 0.7830468416213989, | |
| "eval_runtime": 160.2841, | |
| "eval_samples_per_second": 35.287, | |
| "eval_steps_per_second": 4.411, | |
| "eval_wer": 0.6012742533421065, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.21611851153286965, | |
| "grad_norm": 2.9641568660736084, | |
| "learning_rate": 0.00024919322033898305, | |
| "loss": 0.5806, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.22004793901528547, | |
| "eval_loss": 0.7702626585960388, | |
| "eval_runtime": 160.806, | |
| "eval_samples_per_second": 35.173, | |
| "eval_steps_per_second": 4.397, | |
| "eval_wer": 0.5789026014668357, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.2279067939801171, | |
| "eval_loss": 0.7665734887123108, | |
| "eval_runtime": 160.6796, | |
| "eval_samples_per_second": 35.2, | |
| "eval_steps_per_second": 4.4, | |
| "eval_wer": 0.589831650912359, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.23576564894494872, | |
| "grad_norm": 2.6571083068847656, | |
| "learning_rate": 0.00024410847457627117, | |
| "loss": 0.5608, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.23576564894494872, | |
| "eval_loss": 0.7580233216285706, | |
| "eval_runtime": 160.371, | |
| "eval_samples_per_second": 35.268, | |
| "eval_steps_per_second": 4.409, | |
| "eval_wer": 0.5694500168509573, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24362450390978035, | |
| "eval_loss": 0.7478851675987244, | |
| "eval_runtime": 162.164, | |
| "eval_samples_per_second": 34.878, | |
| "eval_steps_per_second": 4.36, | |
| "eval_wer": 0.5650848164850508, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.251483358874612, | |
| "eval_loss": 0.7638738751411438, | |
| "eval_runtime": 160.257, | |
| "eval_samples_per_second": 35.293, | |
| "eval_steps_per_second": 4.412, | |
| "eval_wer": 0.5846640240086021, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.2554127863570278, | |
| "grad_norm": 1.5677289962768555, | |
| "learning_rate": 0.0002390237288135593, | |
| "loss": 0.5333, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.2593422138394436, | |
| "eval_loss": 0.7297228574752808, | |
| "eval_runtime": 160.7223, | |
| "eval_samples_per_second": 35.191, | |
| "eval_steps_per_second": 4.399, | |
| "eval_wer": 0.5676044358139012, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.26720106880427524, | |
| "eval_loss": 0.7441245913505554, | |
| "eval_runtime": 160.37, | |
| "eval_samples_per_second": 35.268, | |
| "eval_steps_per_second": 4.409, | |
| "eval_wer": 0.5590345203896583, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.2750599237691068, | |
| "grad_norm": 3.644160032272339, | |
| "learning_rate": 0.00023393898305084743, | |
| "loss": 0.5406, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.2750599237691068, | |
| "eval_loss": 0.7404661774635315, | |
| "eval_runtime": 160.5995, | |
| "eval_samples_per_second": 35.218, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.5491165283818267, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.28291877873393845, | |
| "eval_loss": 0.7237815856933594, | |
| "eval_runtime": 160.4373, | |
| "eval_samples_per_second": 35.254, | |
| "eval_steps_per_second": 4.407, | |
| "eval_wer": 0.5529039816404808, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.2907776336987701, | |
| "eval_loss": 0.7328305840492249, | |
| "eval_runtime": 161.925, | |
| "eval_samples_per_second": 34.93, | |
| "eval_steps_per_second": 4.366, | |
| "eval_wer": 0.5543964950008826, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.2947070611811859, | |
| "grad_norm": 3.6030795574188232, | |
| "learning_rate": 0.00022885423728813558, | |
| "loss": 0.535, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2986364886636017, | |
| "eval_loss": 0.7263395190238953, | |
| "eval_runtime": 160.6865, | |
| "eval_samples_per_second": 35.199, | |
| "eval_steps_per_second": 4.4, | |
| "eval_wer": 0.5598690439890228, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.30649534362843334, | |
| "eval_loss": 0.7421374320983887, | |
| "eval_runtime": 160.2249, | |
| "eval_samples_per_second": 35.3, | |
| "eval_steps_per_second": 4.413, | |
| "eval_wer": 0.5594357336585836, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.31435419859326497, | |
| "grad_norm": 3.376089096069336, | |
| "learning_rate": 0.0002237694915254237, | |
| "loss": 0.5195, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.31435419859326497, | |
| "eval_loss": 0.7434934377670288, | |
| "eval_runtime": 161.1972, | |
| "eval_samples_per_second": 35.087, | |
| "eval_steps_per_second": 4.386, | |
| "eval_wer": 0.5543804464701256, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.3222130535580966, | |
| "eval_loss": 0.7186952233314514, | |
| "eval_runtime": 162.677, | |
| "eval_samples_per_second": 34.768, | |
| "eval_steps_per_second": 4.346, | |
| "eval_wer": 0.5423921939946398, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.3300719085229282, | |
| "eval_loss": 0.6976691484451294, | |
| "eval_runtime": 159.5716, | |
| "eval_samples_per_second": 35.445, | |
| "eval_steps_per_second": 4.431, | |
| "eval_wer": 0.5353308404615558, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.33400133600534404, | |
| "grad_norm": 1.9758217334747314, | |
| "learning_rate": 0.00021868474576271186, | |
| "loss": 0.5023, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.33793076348775986, | |
| "eval_loss": 0.6949788928031921, | |
| "eval_runtime": 160.6972, | |
| "eval_samples_per_second": 35.197, | |
| "eval_steps_per_second": 4.4, | |
| "eval_wer": 0.5385565951437146, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.3457896184525915, | |
| "eval_loss": 0.7155033946037292, | |
| "eval_runtime": 159.9521, | |
| "eval_samples_per_second": 35.361, | |
| "eval_steps_per_second": 4.42, | |
| "eval_wer": 0.5450883471618173, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.35364847341742306, | |
| "grad_norm": 3.3146464824676514, | |
| "learning_rate": 0.00021361016949152543, | |
| "loss": 0.5106, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.35364847341742306, | |
| "eval_loss": 0.6857195496559143, | |
| "eval_runtime": 160.5474, | |
| "eval_samples_per_second": 35.229, | |
| "eval_steps_per_second": 4.404, | |
| "eval_wer": 0.5379467509749483, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.3615073283822547, | |
| "eval_loss": 0.68482905626297, | |
| "eval_runtime": 161.0662, | |
| "eval_samples_per_second": 35.116, | |
| "eval_steps_per_second": 4.389, | |
| "eval_wer": 0.5329075123172473, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.3693661833470863, | |
| "eval_loss": 0.6732301712036133, | |
| "eval_runtime": 160.6243, | |
| "eval_samples_per_second": 35.213, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.5202291730192101, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.37329561082950213, | |
| "grad_norm": 4.61689567565918, | |
| "learning_rate": 0.00020852542372881352, | |
| "loss": 0.4968, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.37722503831191795, | |
| "eval_loss": 0.6839133501052856, | |
| "eval_runtime": 161.2367, | |
| "eval_samples_per_second": 35.079, | |
| "eval_steps_per_second": 4.385, | |
| "eval_wer": 0.5274510118598642, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.3850838932767496, | |
| "eval_loss": 0.6766842603683472, | |
| "eval_runtime": 160.827, | |
| "eval_samples_per_second": 35.168, | |
| "eval_steps_per_second": 4.396, | |
| "eval_wer": 0.5198279597502848, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.3929427482415812, | |
| "grad_norm": 3.5624563694000244, | |
| "learning_rate": 0.0002034508474576271, | |
| "loss": 0.4824, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.3929427482415812, | |
| "eval_loss": 0.6718243956565857, | |
| "eval_runtime": 161.1794, | |
| "eval_samples_per_second": 35.091, | |
| "eval_steps_per_second": 4.386, | |
| "eval_wer": 0.5334531623629857, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.40080160320641284, | |
| "eval_loss": 0.6593254804611206, | |
| "eval_runtime": 160.9535, | |
| "eval_samples_per_second": 35.141, | |
| "eval_steps_per_second": 4.393, | |
| "eval_wer": 0.5175169713212755, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.40866045817124447, | |
| "eval_loss": 0.6799437403678894, | |
| "eval_runtime": 159.6664, | |
| "eval_samples_per_second": 35.424, | |
| "eval_steps_per_second": 4.428, | |
| "eval_wer": 0.5173885830752195, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.4125898856536603, | |
| "grad_norm": 2.189781427383423, | |
| "learning_rate": 0.00019836610169491524, | |
| "loss": 0.48, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.4165193131360761, | |
| "eval_loss": 0.6662308573722839, | |
| "eval_runtime": 160.8779, | |
| "eval_samples_per_second": 35.157, | |
| "eval_steps_per_second": 4.395, | |
| "eval_wer": 0.5128949944632569, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.42437816810090767, | |
| "eval_loss": 0.6619213223457336, | |
| "eval_runtime": 160.6185, | |
| "eval_samples_per_second": 35.214, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.5005536743111169, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.4322370230657393, | |
| "grad_norm": 10.41739559173584, | |
| "learning_rate": 0.00019328135593220337, | |
| "loss": 0.4693, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.4322370230657393, | |
| "eval_loss": 0.6576216220855713, | |
| "eval_runtime": 160.9844, | |
| "eval_samples_per_second": 35.134, | |
| "eval_steps_per_second": 4.392, | |
| "eval_wer": 0.519940299465584, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.44009587803057093, | |
| "eval_loss": 0.6406122446060181, | |
| "eval_runtime": 160.4456, | |
| "eval_samples_per_second": 35.252, | |
| "eval_steps_per_second": 4.406, | |
| "eval_wer": 0.5018696538331916, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.44795473299540256, | |
| "eval_loss": 0.6408420205116272, | |
| "eval_runtime": 161.6075, | |
| "eval_samples_per_second": 34.998, | |
| "eval_steps_per_second": 4.375, | |
| "eval_wer": 0.5066039704065093, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.4518841604778184, | |
| "grad_norm": 3.5733156204223633, | |
| "learning_rate": 0.00018819661016949152, | |
| "loss": 0.4691, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.4558135879602342, | |
| "eval_loss": 0.6476473212242126, | |
| "eval_runtime": 161.2518, | |
| "eval_samples_per_second": 35.076, | |
| "eval_steps_per_second": 4.384, | |
| "eval_wer": 0.5019498964869766, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.4636724429250658, | |
| "eval_loss": 0.6423429846763611, | |
| "eval_runtime": 161.3676, | |
| "eval_samples_per_second": 35.05, | |
| "eval_steps_per_second": 4.381, | |
| "eval_wer": 0.4945996694002664, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.47153129788989745, | |
| "grad_norm": 2.3962831497192383, | |
| "learning_rate": 0.00018311186440677962, | |
| "loss": 0.4444, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.47153129788989745, | |
| "eval_loss": 0.6374172568321228, | |
| "eval_runtime": 162.3359, | |
| "eval_samples_per_second": 34.841, | |
| "eval_steps_per_second": 4.355, | |
| "eval_wer": 0.4975846961210701, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.4793901528547291, | |
| "eval_loss": 0.6312358379364014, | |
| "eval_runtime": 162.5747, | |
| "eval_samples_per_second": 34.79, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.4961403283529393, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 0.4872490078195607, | |
| "eval_loss": 0.6170411109924316, | |
| "eval_runtime": 161.58, | |
| "eval_samples_per_second": 35.004, | |
| "eval_steps_per_second": 4.376, | |
| "eval_wer": 0.4818571359792011, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 0.4911784353019765, | |
| "grad_norm": 2.623764753341675, | |
| "learning_rate": 0.0001780372881355932, | |
| "loss": 0.4474, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.49510786278439234, | |
| "eval_loss": 0.6300910115242004, | |
| "eval_runtime": 164.417, | |
| "eval_samples_per_second": 34.4, | |
| "eval_steps_per_second": 4.3, | |
| "eval_wer": 0.49325159281667763, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 0.502966717749224, | |
| "eval_loss": 0.6253496408462524, | |
| "eval_runtime": 161.3418, | |
| "eval_samples_per_second": 35.056, | |
| "eval_steps_per_second": 4.382, | |
| "eval_wer": 0.4862383848758646, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 0.5108255727140556, | |
| "grad_norm": 2.9566869735717773, | |
| "learning_rate": 0.00017295254237288134, | |
| "loss": 0.4471, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.5108255727140556, | |
| "eval_loss": 0.622020959854126, | |
| "eval_runtime": 161.5861, | |
| "eval_samples_per_second": 35.003, | |
| "eval_steps_per_second": 4.375, | |
| "eval_wer": 0.4849224053537899, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.5186844276788872, | |
| "eval_loss": 0.6201028823852539, | |
| "eval_runtime": 160.9515, | |
| "eval_samples_per_second": 35.141, | |
| "eval_steps_per_second": 4.393, | |
| "eval_wer": 0.48527547303044405, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 0.5265432826437189, | |
| "eval_loss": 0.6168439984321594, | |
| "eval_runtime": 162.0987, | |
| "eval_samples_per_second": 34.892, | |
| "eval_steps_per_second": 4.362, | |
| "eval_wer": 0.4848261141692478, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 0.5304727101261346, | |
| "grad_norm": 1.5596935749053955, | |
| "learning_rate": 0.0001678677966101695, | |
| "loss": 0.4323, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.5344021376085505, | |
| "eval_loss": 0.6172667741775513, | |
| "eval_runtime": 162.3681, | |
| "eval_samples_per_second": 34.834, | |
| "eval_steps_per_second": 4.354, | |
| "eval_wer": 0.47707467381361235, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 0.542260992573382, | |
| "eval_loss": 0.603190004825592, | |
| "eval_runtime": 161.2926, | |
| "eval_samples_per_second": 35.067, | |
| "eval_steps_per_second": 4.383, | |
| "eval_wer": 0.4656160228531078, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 0.5501198475382136, | |
| "grad_norm": 2.978868246078491, | |
| "learning_rate": 0.0001627830508474576, | |
| "loss": 0.4575, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.5501198475382136, | |
| "eval_loss": 0.6097469925880432, | |
| "eval_runtime": 161.1042, | |
| "eval_samples_per_second": 35.108, | |
| "eval_steps_per_second": 4.388, | |
| "eval_wer": 0.4678307200975751, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.5579787025030453, | |
| "eval_loss": 0.5970696806907654, | |
| "eval_runtime": 161.5846, | |
| "eval_samples_per_second": 35.003, | |
| "eval_steps_per_second": 4.375, | |
| "eval_wer": 0.4673653127056218, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 0.5658375574678769, | |
| "eval_loss": 0.5976916551589966, | |
| "eval_runtime": 161.7136, | |
| "eval_samples_per_second": 34.975, | |
| "eval_steps_per_second": 4.372, | |
| "eval_wer": 0.4697565437884162, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 0.5697669849502928, | |
| "grad_norm": 3.0501327514648438, | |
| "learning_rate": 0.00015769830508474575, | |
| "loss": 0.4395, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.5736964124327085, | |
| "eval_loss": 0.6056780815124512, | |
| "eval_runtime": 162.5963, | |
| "eval_samples_per_second": 34.786, | |
| "eval_steps_per_second": 4.348, | |
| "eval_wer": 0.4734316573317713, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 0.5815552673975402, | |
| "eval_loss": 0.582733690738678, | |
| "eval_runtime": 162.9467, | |
| "eval_samples_per_second": 34.711, | |
| "eval_steps_per_second": 4.339, | |
| "eval_wer": 0.4574152236362761, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 0.5894141223623718, | |
| "grad_norm": 4.3484697341918945, | |
| "learning_rate": 0.00015261355932203388, | |
| "loss": 0.4119, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5894141223623718, | |
| "eval_loss": 0.5946210622787476, | |
| "eval_runtime": 162.2892, | |
| "eval_samples_per_second": 34.851, | |
| "eval_steps_per_second": 4.356, | |
| "eval_wer": 0.4640432668389209, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5972729773272034, | |
| "eval_loss": 0.602292001247406, | |
| "eval_runtime": 161.4334, | |
| "eval_samples_per_second": 35.036, | |
| "eval_steps_per_second": 4.38, | |
| "eval_wer": 0.47707467381361235, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 0.605131832292035, | |
| "eval_loss": 0.6129310727119446, | |
| "eval_runtime": 161.8649, | |
| "eval_samples_per_second": 34.943, | |
| "eval_steps_per_second": 4.368, | |
| "eval_wer": 0.47266132785543485, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 0.6090612597744509, | |
| "grad_norm": 4.229031085968018, | |
| "learning_rate": 0.00014752881355932203, | |
| "loss": 0.4125, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.6129906872568667, | |
| "eval_loss": 0.590186595916748, | |
| "eval_runtime": 162.4898, | |
| "eval_samples_per_second": 34.808, | |
| "eval_steps_per_second": 4.351, | |
| "eval_wer": 0.45837813548169665, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 0.6208495422216983, | |
| "eval_loss": 0.5955421328544617, | |
| "eval_runtime": 161.8228, | |
| "eval_samples_per_second": 34.952, | |
| "eval_steps_per_second": 4.369, | |
| "eval_wer": 0.46537529489175267, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 0.6287083971865299, | |
| "grad_norm": 1.4181621074676514, | |
| "learning_rate": 0.00014244406779661016, | |
| "loss": 0.4039, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6287083971865299, | |
| "eval_loss": 0.5955237150192261, | |
| "eval_runtime": 161.3699, | |
| "eval_samples_per_second": 35.05, | |
| "eval_steps_per_second": 4.381, | |
| "eval_wer": 0.45946943557317327, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6365672521513616, | |
| "eval_loss": 0.578912079334259, | |
| "eval_runtime": 163.2091, | |
| "eval_samples_per_second": 34.655, | |
| "eval_steps_per_second": 4.332, | |
| "eval_wer": 0.4497279774036687, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 0.6444261071161932, | |
| "eval_loss": 0.5779294371604919, | |
| "eval_runtime": 164.0491, | |
| "eval_samples_per_second": 34.477, | |
| "eval_steps_per_second": 4.31, | |
| "eval_wer": 0.4630322094012293, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 0.648355534598609, | |
| "grad_norm": 2.0229876041412354, | |
| "learning_rate": 0.00013736949152542372, | |
| "loss": 0.3969, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.6522849620810248, | |
| "eval_loss": 0.5677434802055359, | |
| "eval_runtime": 161.201, | |
| "eval_samples_per_second": 35.087, | |
| "eval_steps_per_second": 4.386, | |
| "eval_wer": 0.45507213814575276, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 0.6601438170458565, | |
| "eval_loss": 0.586939811706543, | |
| "eval_runtime": 161.4539, | |
| "eval_samples_per_second": 35.032, | |
| "eval_steps_per_second": 4.379, | |
| "eval_wer": 0.46062492978767794, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 0.6680026720106881, | |
| "grad_norm": 4.166793346405029, | |
| "learning_rate": 0.00013229491525423729, | |
| "loss": 0.3923, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.6680026720106881, | |
| "eval_loss": 0.5710186958312988, | |
| "eval_runtime": 160.5637, | |
| "eval_samples_per_second": 35.226, | |
| "eval_steps_per_second": 4.403, | |
| "eval_wer": 0.45017733626486495, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.6758615269755197, | |
| "eval_loss": 0.5639811158180237, | |
| "eval_runtime": 161.7944, | |
| "eval_samples_per_second": 34.958, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.44741698897465937, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 0.6837203819403513, | |
| "eval_loss": 0.5841760039329529, | |
| "eval_runtime": 161.0184, | |
| "eval_samples_per_second": 35.126, | |
| "eval_steps_per_second": 4.391, | |
| "eval_wer": 0.4497921715266967, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 0.6876498094227671, | |
| "grad_norm": 3.127680778503418, | |
| "learning_rate": 0.0001272101694915254, | |
| "loss": 0.386, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.691579236905183, | |
| "eval_loss": 0.5596618056297302, | |
| "eval_runtime": 160.919, | |
| "eval_samples_per_second": 35.148, | |
| "eval_steps_per_second": 4.394, | |
| "eval_wer": 0.44403074898493045, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 0.6994380918700145, | |
| "eval_loss": 0.5620830059051514, | |
| "eval_runtime": 160.6614, | |
| "eval_samples_per_second": 35.204, | |
| "eval_steps_per_second": 4.401, | |
| "eval_wer": 0.43812488966635105, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 0.7072969468348461, | |
| "grad_norm": 17.387800216674805, | |
| "learning_rate": 0.00012213559322033898, | |
| "loss": 0.3851, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.7072969468348461, | |
| "eval_loss": 0.566453218460083, | |
| "eval_runtime": 161.6574, | |
| "eval_samples_per_second": 34.988, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.434562115838295, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.7151558017996777, | |
| "eval_loss": 0.5572646260261536, | |
| "eval_runtime": 162.4898, | |
| "eval_samples_per_second": 34.808, | |
| "eval_steps_per_second": 4.351, | |
| "eval_wer": 0.4356213188682576, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 0.7230146567645094, | |
| "eval_loss": 0.5548349022865295, | |
| "eval_runtime": 161.0153, | |
| "eval_samples_per_second": 35.127, | |
| "eval_steps_per_second": 4.391, | |
| "eval_wer": 0.4344337275922389, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 0.7269440842469252, | |
| "grad_norm": 9.4507417678833, | |
| "learning_rate": 0.00011705084745762712, | |
| "loss": 0.369, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.730873511729341, | |
| "eval_loss": 0.5616690516471863, | |
| "eval_runtime": 161.4318, | |
| "eval_samples_per_second": 35.036, | |
| "eval_steps_per_second": 4.38, | |
| "eval_wer": 0.43637559981383706, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 0.7387323666941726, | |
| "eval_loss": 0.5595532655715942, | |
| "eval_runtime": 160.8301, | |
| "eval_samples_per_second": 35.168, | |
| "eval_steps_per_second": 4.396, | |
| "eval_wer": 0.4393927235961548, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 0.7465912216590043, | |
| "grad_norm": 1.8793506622314453, | |
| "learning_rate": 0.00011196610169491524, | |
| "loss": 0.3738, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.7465912216590043, | |
| "eval_loss": 0.549248218536377, | |
| "eval_runtime": 161.3194, | |
| "eval_samples_per_second": 35.061, | |
| "eval_steps_per_second": 4.383, | |
| "eval_wer": 0.42923400362696795, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.7544500766238359, | |
| "eval_loss": 0.5478147268295288, | |
| "eval_runtime": 162.2231, | |
| "eval_samples_per_second": 34.866, | |
| "eval_steps_per_second": 4.358, | |
| "eval_wer": 0.4372261719439585, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 0.7623089315886675, | |
| "eval_loss": 0.5375632047653198, | |
| "eval_runtime": 161.0297, | |
| "eval_samples_per_second": 35.124, | |
| "eval_steps_per_second": 4.39, | |
| "eval_wer": 0.42873649917350065, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 0.7662383590710834, | |
| "grad_norm": 2.159616708755493, | |
| "learning_rate": 0.00010688135593220338, | |
| "loss": 0.368, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.7701677865534992, | |
| "eval_loss": 0.5282244086265564, | |
| "eval_runtime": 163.0357, | |
| "eval_samples_per_second": 34.692, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.4193481086806503, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 0.7780266415183308, | |
| "eval_loss": 0.5348193049430847, | |
| "eval_runtime": 162.5531, | |
| "eval_samples_per_second": 34.795, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.42507743416090255, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 0.7858854964831624, | |
| "grad_norm": 2.2020351886749268, | |
| "learning_rate": 0.00010179661016949151, | |
| "loss": 0.3629, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.7858854964831624, | |
| "eval_loss": 0.5367931723594666, | |
| "eval_runtime": 162.0053, | |
| "eval_samples_per_second": 34.912, | |
| "eval_steps_per_second": 4.364, | |
| "eval_wer": 0.43130426409462214, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.793744351447994, | |
| "eval_loss": 0.5550614595413208, | |
| "eval_runtime": 161.9948, | |
| "eval_samples_per_second": 34.915, | |
| "eval_steps_per_second": 4.364, | |
| "eval_wer": 0.44123830463321084, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 0.8016032064128257, | |
| "eval_loss": 0.5251778364181519, | |
| "eval_runtime": 162.6214, | |
| "eval_samples_per_second": 34.78, | |
| "eval_steps_per_second": 4.348, | |
| "eval_wer": 0.4105214167642952, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 0.8055326338952414, | |
| "grad_norm": 2.7725887298583984, | |
| "learning_rate": 9.671186440677966e-05, | |
| "loss": 0.3638, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.8094620613776573, | |
| "eval_loss": 0.5242481827735901, | |
| "eval_runtime": 162.5731, | |
| "eval_samples_per_second": 34.791, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.41174110510182793, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 0.8173209163424889, | |
| "eval_loss": 0.5233432054519653, | |
| "eval_runtime": 161.9438, | |
| "eval_samples_per_second": 34.926, | |
| "eval_steps_per_second": 4.366, | |
| "eval_wer": 0.4165877613904447, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 0.8251797713073206, | |
| "grad_norm": 2.733196496963501, | |
| "learning_rate": 9.162711864406779e-05, | |
| "loss": 0.3512, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.8251797713073206, | |
| "eval_loss": 0.524342954158783, | |
| "eval_runtime": 161.947, | |
| "eval_samples_per_second": 34.925, | |
| "eval_steps_per_second": 4.366, | |
| "eval_wer": 0.4160581598754634, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.8330386262721522, | |
| "eval_loss": 0.5150259733200073, | |
| "eval_runtime": 162.0793, | |
| "eval_samples_per_second": 34.896, | |
| "eval_steps_per_second": 4.362, | |
| "eval_wer": 0.4123028036783232, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 0.8408974812369838, | |
| "eval_loss": 0.5088914632797241, | |
| "eval_runtime": 161.2392, | |
| "eval_samples_per_second": 35.078, | |
| "eval_steps_per_second": 4.385, | |
| "eval_wer": 0.4079536518431738, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 0.8448269087193996, | |
| "grad_norm": 4.562708377838135, | |
| "learning_rate": 8.654237288135593e-05, | |
| "loss": 0.3536, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.8487563362018153, | |
| "eval_loss": 0.515373170375824, | |
| "eval_runtime": 162.8063, | |
| "eval_samples_per_second": 34.741, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.40899680634237934, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 0.856615191166647, | |
| "eval_loss": 0.5161571502685547, | |
| "eval_runtime": 162.7678, | |
| "eval_samples_per_second": 34.749, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.4091893887114635, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 0.8644740461314786, | |
| "grad_norm": 2.272256374359131, | |
| "learning_rate": 8.146779661016948e-05, | |
| "loss": 0.3464, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.8644740461314786, | |
| "eval_loss": 0.5097736716270447, | |
| "eval_runtime": 162.1935, | |
| "eval_samples_per_second": 34.872, | |
| "eval_steps_per_second": 4.359, | |
| "eval_wer": 0.40527354720675324, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.8723329010963102, | |
| "eval_loss": 0.5069981813430786, | |
| "eval_runtime": 162.5966, | |
| "eval_samples_per_second": 34.785, | |
| "eval_steps_per_second": 4.348, | |
| "eval_wer": 0.4022724719551925, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 0.8801917560611419, | |
| "eval_loss": 0.5070444345474243, | |
| "eval_runtime": 162.5617, | |
| "eval_samples_per_second": 34.793, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.40707098265153824, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 0.8841211835435577, | |
| "grad_norm": 2.9740068912506104, | |
| "learning_rate": 7.638305084745762e-05, | |
| "loss": 0.3377, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.8880506110259735, | |
| "eval_loss": 0.5028176307678223, | |
| "eval_runtime": 162.4451, | |
| "eval_samples_per_second": 34.818, | |
| "eval_steps_per_second": 4.352, | |
| "eval_wer": 0.39670363178251034, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 0.8959094659908051, | |
| "eval_loss": 0.5036062002182007, | |
| "eval_runtime": 162.5763, | |
| "eval_samples_per_second": 34.79, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.39784307746625797, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 0.9037683209556368, | |
| "grad_norm": 1.9388916492462158, | |
| "learning_rate": 7.129830508474575e-05, | |
| "loss": 0.3272, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.9037683209556368, | |
| "eval_loss": 0.5020586848258972, | |
| "eval_runtime": 161.6894, | |
| "eval_samples_per_second": 34.981, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.39538765226043554, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.9116271759204684, | |
| "eval_loss": 0.5032612085342407, | |
| "eval_runtime": 163.6786, | |
| "eval_samples_per_second": 34.556, | |
| "eval_steps_per_second": 4.319, | |
| "eval_wer": 0.3984529216350243, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 0.9194860308853, | |
| "eval_loss": 0.49842530488967896, | |
| "eval_runtime": 162.0701, | |
| "eval_samples_per_second": 34.898, | |
| "eval_steps_per_second": 4.362, | |
| "eval_wer": 0.3971850877052206, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 0.9234154583677158, | |
| "grad_norm": 3.9436373710632324, | |
| "learning_rate": 6.621355932203389e-05, | |
| "loss": 0.319, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.9273448858501316, | |
| "eval_loss": 0.4928737282752991, | |
| "eval_runtime": 163.9597, | |
| "eval_samples_per_second": 34.496, | |
| "eval_steps_per_second": 4.312, | |
| "eval_wer": 0.39243472260114587, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 0.9352037408149633, | |
| "eval_loss": 0.49405232071876526, | |
| "eval_runtime": 161.8803, | |
| "eval_samples_per_second": 34.939, | |
| "eval_steps_per_second": 4.367, | |
| "eval_wer": 0.4013095601097719, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 0.9430625957797949, | |
| "grad_norm": 3.4186201095581055, | |
| "learning_rate": 6.112881355932203e-05, | |
| "loss": 0.3184, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.9430625957797949, | |
| "eval_loss": 0.4856198728084564, | |
| "eval_runtime": 163.6122, | |
| "eval_samples_per_second": 34.57, | |
| "eval_steps_per_second": 4.321, | |
| "eval_wer": 0.387411532474202, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.9509214507446265, | |
| "eval_loss": 0.48915818333625793, | |
| "eval_runtime": 162.8317, | |
| "eval_samples_per_second": 34.735, | |
| "eval_steps_per_second": 4.342, | |
| "eval_wer": 0.3913755195711833, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 0.9587803057094582, | |
| "eval_loss": 0.48598504066467285, | |
| "eval_runtime": 160.6269, | |
| "eval_samples_per_second": 35.212, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.3813772849095665, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 0.9627097331918739, | |
| "grad_norm": 2.70164155960083, | |
| "learning_rate": 5.6044067796610164e-05, | |
| "loss": 0.3091, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.9666391606742898, | |
| "eval_loss": 0.4825168251991272, | |
| "eval_runtime": 162.6242, | |
| "eval_samples_per_second": 34.78, | |
| "eval_steps_per_second": 4.347, | |
| "eval_wer": 0.38336730272343567, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 0.9744980156391214, | |
| "eval_loss": 0.4784228205680847, | |
| "eval_runtime": 162.0189, | |
| "eval_samples_per_second": 34.91, | |
| "eval_steps_per_second": 4.364, | |
| "eval_wer": 0.3866893485901366, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 0.982356870603953, | |
| "grad_norm": 9.408166885375977, | |
| "learning_rate": 5.096949152542373e-05, | |
| "loss": 0.3154, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.982356870603953, | |
| "eval_loss": 0.47507792711257935, | |
| "eval_runtime": 161.9422, | |
| "eval_samples_per_second": 34.926, | |
| "eval_steps_per_second": 4.366, | |
| "eval_wer": 0.3807834892715572, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.9902157255687847, | |
| "eval_loss": 0.4778765141963959, | |
| "eval_runtime": 162.3405, | |
| "eval_samples_per_second": 34.84, | |
| "eval_steps_per_second": 4.355, | |
| "eval_wer": 0.38492401020686556, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 0.9980745805336163, | |
| "eval_loss": 0.477267324924469, | |
| "eval_runtime": 161.2107, | |
| "eval_samples_per_second": 35.085, | |
| "eval_steps_per_second": 4.386, | |
| "eval_wer": 0.38084768339458525, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 1.002004008016032, | |
| "grad_norm": 0.7003775835037231, | |
| "learning_rate": 4.589491525423728e-05, | |
| "loss": 0.312, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.005933435498448, | |
| "eval_loss": 0.47774726152420044, | |
| "eval_runtime": 160.8535, | |
| "eval_samples_per_second": 35.162, | |
| "eval_steps_per_second": 4.395, | |
| "eval_wer": 0.3757923962061273, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 1.0137922904632795, | |
| "eval_loss": 0.4752050042152405, | |
| "eval_runtime": 159.7765, | |
| "eval_samples_per_second": 35.399, | |
| "eval_steps_per_second": 4.425, | |
| "eval_wer": 0.3820513232013609, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 1.0216511454281112, | |
| "grad_norm": 0.702942430973053, | |
| "learning_rate": 4.081016949152542e-05, | |
| "loss": 0.2651, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.0216511454281112, | |
| "eval_loss": 0.4700838327407837, | |
| "eval_runtime": 163.2858, | |
| "eval_samples_per_second": 34.639, | |
| "eval_steps_per_second": 4.33, | |
| "eval_wer": 0.37750958899712733, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.0295100003929427, | |
| "eval_loss": 0.47011885046958923, | |
| "eval_runtime": 160.7741, | |
| "eval_samples_per_second": 35.18, | |
| "eval_steps_per_second": 4.397, | |
| "eval_wer": 0.3760652212289965, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 1.0373688553577745, | |
| "eval_loss": 0.471804678440094, | |
| "eval_runtime": 160.2455, | |
| "eval_samples_per_second": 35.296, | |
| "eval_steps_per_second": 4.412, | |
| "eval_wer": 0.37755773458939834, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 1.0412982828401902, | |
| "grad_norm": 0.98069828748703, | |
| "learning_rate": 3.572542372881355e-05, | |
| "loss": 0.2627, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.045227710322606, | |
| "eval_loss": 0.4638473391532898, | |
| "eval_runtime": 160.1121, | |
| "eval_samples_per_second": 35.325, | |
| "eval_steps_per_second": 4.416, | |
| "eval_wer": 0.37296785479289374, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 1.0530865652874377, | |
| "eval_loss": 0.4677112400531769, | |
| "eval_runtime": 159.9389, | |
| "eval_samples_per_second": 35.364, | |
| "eval_steps_per_second": 4.42, | |
| "eval_wer": 0.3720370400089872, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 1.0609454202522692, | |
| "grad_norm": 0.8780287504196167, | |
| "learning_rate": 3.0640677966101693e-05, | |
| "loss": 0.2427, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.0609454202522692, | |
| "eval_loss": 0.4642546474933624, | |
| "eval_runtime": 160.0541, | |
| "eval_samples_per_second": 35.338, | |
| "eval_steps_per_second": 4.417, | |
| "eval_wer": 0.36985443982603394, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.0688042752171008, | |
| "eval_loss": 0.46017909049987793, | |
| "eval_runtime": 159.9066, | |
| "eval_samples_per_second": 35.371, | |
| "eval_steps_per_second": 4.421, | |
| "eval_wer": 0.3713469531864358, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 1.0766631301819325, | |
| "eval_loss": 0.46644654870033264, | |
| "eval_runtime": 160.7516, | |
| "eval_samples_per_second": 35.185, | |
| "eval_steps_per_second": 4.398, | |
| "eval_wer": 0.3703037986872302, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 1.0805925576643483, | |
| "grad_norm": 0.8659859895706177, | |
| "learning_rate": 2.556610169491525e-05, | |
| "loss": 0.2464, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.0845219851467642, | |
| "eval_loss": 0.4609028100967407, | |
| "eval_runtime": 161.4502, | |
| "eval_samples_per_second": 35.032, | |
| "eval_steps_per_second": 4.379, | |
| "eval_wer": 0.36770393670459467, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 1.0923808401115958, | |
| "eval_loss": 0.4613707363605499, | |
| "eval_runtime": 160.5963, | |
| "eval_samples_per_second": 35.219, | |
| "eval_steps_per_second": 4.402, | |
| "eval_wer": 0.3687310426730433, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 1.1002396950764273, | |
| "grad_norm": 1.6944918632507324, | |
| "learning_rate": 2.0481355932203388e-05, | |
| "loss": 0.2537, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.1002396950764273, | |
| "eval_loss": 0.45553678274154663, | |
| "eval_runtime": 160.1154, | |
| "eval_samples_per_second": 35.325, | |
| "eval_steps_per_second": 4.416, | |
| "eval_wer": 0.36545714239861343, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.108098550041259, | |
| "eval_loss": 0.456032931804657, | |
| "eval_runtime": 160.97, | |
| "eval_samples_per_second": 35.137, | |
| "eval_steps_per_second": 4.392, | |
| "eval_wer": 0.36447818202243587, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 1.1159574050060905, | |
| "eval_loss": 0.45427000522613525, | |
| "eval_runtime": 160.1348, | |
| "eval_samples_per_second": 35.32, | |
| "eval_steps_per_second": 4.415, | |
| "eval_wer": 0.36261655245462276, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 1.1198868324885065, | |
| "grad_norm": 0.8318812251091003, | |
| "learning_rate": 1.5396610169491525e-05, | |
| "loss": 0.2313, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.1238162599709223, | |
| "eval_loss": 0.45402956008911133, | |
| "eval_runtime": 160.7545, | |
| "eval_samples_per_second": 35.184, | |
| "eval_steps_per_second": 4.398, | |
| "eval_wer": 0.3631461539696041, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 1.1316751149357538, | |
| "eval_loss": 0.4536111354827881, | |
| "eval_runtime": 165.4654, | |
| "eval_samples_per_second": 34.182, | |
| "eval_steps_per_second": 4.273, | |
| "eval_wer": 0.3626326009853798, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 1.1395339699005855, | |
| "grad_norm": 0.7866860032081604, | |
| "learning_rate": 1.031186440677966e-05, | |
| "loss": 0.2451, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.1395339699005855, | |
| "eval_loss": 0.45293620228767395, | |
| "eval_runtime": 160.3649, | |
| "eval_samples_per_second": 35.27, | |
| "eval_steps_per_second": 4.409, | |
| "eval_wer": 0.3617338832629873, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.147392824865417, | |
| "eval_loss": 0.4530145823955536, | |
| "eval_runtime": 160.576, | |
| "eval_samples_per_second": 35.223, | |
| "eval_steps_per_second": 4.403, | |
| "eval_wer": 0.3598401566336602, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 1.1552516798302488, | |
| "eval_loss": 0.4515323042869568, | |
| "eval_runtime": 160.1136, | |
| "eval_samples_per_second": 35.325, | |
| "eval_steps_per_second": 4.416, | |
| "eval_wer": 0.3591500698111088, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 1.1591811073126645, | |
| "grad_norm": 3.2193210124969482, | |
| "learning_rate": 5.227118644067796e-06, | |
| "loss": 0.2445, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 1.1631105347950803, | |
| "eval_loss": 0.451358437538147, | |
| "eval_runtime": 160.6595, | |
| "eval_samples_per_second": 35.205, | |
| "eval_steps_per_second": 4.401, | |
| "eval_wer": 0.3590056330342957, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 1.170969389759912, | |
| "eval_loss": 0.4514302611351013, | |
| "eval_runtime": 160.1434, | |
| "eval_samples_per_second": 35.318, | |
| "eval_steps_per_second": 4.415, | |
| "eval_wer": 0.3588772447882396, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 1.1788282447247436, | |
| "grad_norm": 0.5669330358505249, | |
| "learning_rate": 1.423728813559322e-07, | |
| "loss": 0.2364, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.1788282447247436, | |
| "eval_loss": 0.4510672390460968, | |
| "eval_runtime": 160.6855, | |
| "eval_samples_per_second": 35.199, | |
| "eval_steps_per_second": 4.4, | |
| "eval_wer": 0.3591179727495948, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.1788282447247436, | |
| "step": 30000, | |
| "total_flos": 3.731985674211105e+19, | |
| "train_loss": 0.5082863594055176, | |
| "train_runtime": 37313.8627, | |
| "train_samples_per_second": 6.432, | |
| "train_steps_per_second": 0.804 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 30000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.731985674211105e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |