| { | |
| "best_metric": 39.72011741415796, | |
| "best_model_checkpoint": "./checkpoint-4500", | |
| "epoch": 15.01, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.6e-07, | |
| "loss": 1.447, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.6e-07, | |
| "loss": 1.2062, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.953535353535352e-07, | |
| "loss": 0.9199, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.903030303030303e-07, | |
| "loss": 0.8626, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.852525252525253e-07, | |
| "loss": 0.6832, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.802020202020202e-07, | |
| "loss": 0.6653, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.751515151515151e-07, | |
| "loss": 0.5903, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.7010101010101e-07, | |
| "loss": 0.6102, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.65050505050505e-07, | |
| "loss": 0.5404, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.6e-07, | |
| "loss": 0.5417, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.549494949494948e-07, | |
| "loss": 0.4839, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.498989898989899e-07, | |
| "loss": 0.6314, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.448484848484848e-07, | |
| "loss": 0.4998, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.397979797979797e-07, | |
| "loss": 0.5843, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.347474747474747e-07, | |
| "loss": 0.5397, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.296969696969696e-07, | |
| "loss": 0.483, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.246464646464645e-07, | |
| "loss": 0.5599, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.195959595959596e-07, | |
| "loss": 0.4669, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.145454545454545e-07, | |
| "loss": 0.4666, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.094949494949494e-07, | |
| "loss": 0.4531, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.6448386907577515, | |
| "eval_runtime": 2680.2862, | |
| "eval_samples_per_second": 3.935, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 50.73929961089494, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.044444444444445e-07, | |
| "loss": 0.4615, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 8.993939393939394e-07, | |
| "loss": 0.4126, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 8.943434343434342e-07, | |
| "loss": 0.4149, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 8.892929292929292e-07, | |
| "loss": 0.3856, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 8.842424242424242e-07, | |
| "loss": 0.503, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 8.791919191919191e-07, | |
| "loss": 0.3849, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.741414141414141e-07, | |
| "loss": 0.4925, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.690909090909091e-07, | |
| "loss": 0.4559, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.64040404040404e-07, | |
| "loss": 0.4114, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.58989898989899e-07, | |
| "loss": 0.4575, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.539393939393939e-07, | |
| "loss": 0.396, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.488888888888888e-07, | |
| "loss": 0.4015, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.438383838383838e-07, | |
| "loss": 0.3955, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.387878787878787e-07, | |
| "loss": 0.375, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.337373737373737e-07, | |
| "loss": 0.369, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 8.286868686868687e-07, | |
| "loss": 0.3435, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 8.236363636363636e-07, | |
| "loss": 0.3368, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.185858585858586e-07, | |
| "loss": 0.4251, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.135353535353536e-07, | |
| "loss": 0.3489, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.084848484848484e-07, | |
| "loss": 0.4031, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5755370259284973, | |
| "eval_runtime": 2679.7676, | |
| "eval_samples_per_second": 3.936, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 46.50010239606799, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.034343434343433e-07, | |
| "loss": 0.404, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 7.983838383838384e-07, | |
| "loss": 0.3739, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 7.933333333333333e-07, | |
| "loss": 0.3828, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 7.882828282828282e-07, | |
| "loss": 0.3757, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 7.832323232323233e-07, | |
| "loss": 0.3461, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 7.781818181818182e-07, | |
| "loss": 0.3482, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 7.731313131313131e-07, | |
| "loss": 0.327, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 7.68080808080808e-07, | |
| "loss": 0.3374, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.63030303030303e-07, | |
| "loss": 0.294, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.579797979797979e-07, | |
| "loss": 0.3151, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.529292929292928e-07, | |
| "loss": 0.3525, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.478787878787879e-07, | |
| "loss": 0.3425, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.428282828282828e-07, | |
| "loss": 0.3327, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.377777777777777e-07, | |
| "loss": 0.3716, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.327272727272728e-07, | |
| "loss": 0.3497, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.276767676767677e-07, | |
| "loss": 0.3224, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.226262626262625e-07, | |
| "loss": 0.3713, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.175757575757575e-07, | |
| "loss": 0.3016, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.125252525252525e-07, | |
| "loss": 0.3327, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.074747474747474e-07, | |
| "loss": 0.2745, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_loss": 0.5389129519462585, | |
| "eval_runtime": 2674.3453, | |
| "eval_samples_per_second": 3.944, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 43.71902518943273, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.024242424242424e-07, | |
| "loss": 0.3222, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 6.973737373737374e-07, | |
| "loss": 0.2568, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 6.923232323232323e-07, | |
| "loss": 0.2884, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 6.872727272727273e-07, | |
| "loss": 0.2843, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 6.822222222222221e-07, | |
| "loss": 0.3463, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 6.771717171717171e-07, | |
| "loss": 0.2729, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 6.721212121212121e-07, | |
| "loss": 0.3626, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.67070707070707e-07, | |
| "loss": 0.3229, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.62020202020202e-07, | |
| "loss": 0.2853, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 6.56969696969697e-07, | |
| "loss": 0.3527, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.519191919191919e-07, | |
| "loss": 0.2876, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.468686868686868e-07, | |
| "loss": 0.2958, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.418181818181818e-07, | |
| "loss": 0.2592, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.367676767676767e-07, | |
| "loss": 0.2922, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.317171717171716e-07, | |
| "loss": 0.2516, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.266666666666667e-07, | |
| "loss": 0.2588, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.216161616161616e-07, | |
| "loss": 0.2404, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.165656565656565e-07, | |
| "loss": 0.3323, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.115151515151516e-07, | |
| "loss": 0.251, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.064646464646465e-07, | |
| "loss": 0.336, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.5165684819221497, | |
| "eval_runtime": 2687.4563, | |
| "eval_samples_per_second": 3.925, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 42.40562495733497, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 6.014141414141413e-07, | |
| "loss": 0.3095, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 5.963636363636363e-07, | |
| "loss": 0.263, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 5.913131313131313e-07, | |
| "loss": 0.3225, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 5.862626262626262e-07, | |
| "loss": 0.2666, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 5.812121212121212e-07, | |
| "loss": 0.2649, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 5.761616161616162e-07, | |
| "loss": 0.2585, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.711111111111111e-07, | |
| "loss": 0.2639, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.660606060606061e-07, | |
| "loss": 0.2388, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5.610101010101009e-07, | |
| "loss": 0.2425, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5.559595959595959e-07, | |
| "loss": 0.2217, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.509090909090909e-07, | |
| "loss": 0.3125, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.458585858585858e-07, | |
| "loss": 0.2252, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 5.408080808080808e-07, | |
| "loss": 0.3136, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.357575757575758e-07, | |
| "loss": 0.2938, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.307070707070707e-07, | |
| "loss": 0.2586, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.256565656565657e-07, | |
| "loss": 0.2889, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.206060606060607e-07, | |
| "loss": 0.2535, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.155555555555555e-07, | |
| "loss": 0.2527, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.105050505050504e-07, | |
| "loss": 0.2468, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 5.054545454545454e-07, | |
| "loss": 0.2429, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "eval_loss": 0.50449538230896, | |
| "eval_runtime": 2691.9546, | |
| "eval_samples_per_second": 3.918, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 41.18096798416274, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 5.004040404040404e-07, | |
| "loss": 0.2291, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 4.953535353535353e-07, | |
| "loss": 0.2151, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 4.903030303030302e-07, | |
| "loss": 0.2196, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 4.852525252525253e-07, | |
| "loss": 0.2827, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 4.802020202020202e-07, | |
| "loss": 0.2262, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.7515151515151514e-07, | |
| "loss": 0.2713, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.701010101010101e-07, | |
| "loss": 0.2776, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.65050505050505e-07, | |
| "loss": 0.2581, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.6e-07, | |
| "loss": 0.2559, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.549494949494949e-07, | |
| "loss": 0.2657, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.498989898989899e-07, | |
| "loss": 0.233, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.448484848484848e-07, | |
| "loss": 0.236, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.3979797979797977e-07, | |
| "loss": 0.2141, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.3474747474747475e-07, | |
| "loss": 0.233, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.296969696969697e-07, | |
| "loss": 0.1993, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.246464646464646e-07, | |
| "loss": 0.2134, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.1959595959595955e-07, | |
| "loss": 0.2408, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.1454545454545453e-07, | |
| "loss": 0.2425, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.0949494949494946e-07, | |
| "loss": 0.2227, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 4.044444444444444e-07, | |
| "loss": 0.2852, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "eval_loss": 0.4940944314002991, | |
| "eval_runtime": 2678.7609, | |
| "eval_samples_per_second": 3.937, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 40.64441258788996, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 3.993939393939394e-07, | |
| "loss": 0.2513, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.943434343434343e-07, | |
| "loss": 0.2307, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.892929292929293e-07, | |
| "loss": 0.2637, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.8424242424242423e-07, | |
| "loss": 0.2197, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.7919191919191916e-07, | |
| "loss": 0.2289, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.7414141414141414e-07, | |
| "loss": 0.1967, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.690909090909091e-07, | |
| "loss": 0.2304, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.6404040404040406e-07, | |
| "loss": 0.1926, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.5898989898989894e-07, | |
| "loss": 0.2027, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.539393939393939e-07, | |
| "loss": 0.205, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.488888888888889e-07, | |
| "loss": 0.2529, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 3.4383838383838384e-07, | |
| "loss": 0.1958, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 3.3878787878787877e-07, | |
| "loss": 0.2827, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 3.337373737373737e-07, | |
| "loss": 0.2434, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3.286868686868687e-07, | |
| "loss": 0.216, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3.236363636363636e-07, | |
| "loss": 0.2628, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 3.1858585858585855e-07, | |
| "loss": 0.2141, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 3.1353535353535353e-07, | |
| "loss": 0.2097, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 3.0848484848484847e-07, | |
| "loss": 0.1984, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 3.0343434343434345e-07, | |
| "loss": 0.2217, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "eval_loss": 0.4887745678424835, | |
| "eval_runtime": 2675.8042, | |
| "eval_samples_per_second": 3.942, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 40.110587753430266, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.9838383838383833e-07, | |
| "loss": 0.1831, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.933333333333333e-07, | |
| "loss": 0.2014, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.882828282828283e-07, | |
| "loss": 0.1742, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.8323232323232323e-07, | |
| "loss": 0.2636, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.7818181818181816e-07, | |
| "loss": 0.1826, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.731313131313131e-07, | |
| "loss": 0.261, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.680808080808081e-07, | |
| "loss": 0.2434, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.63030303030303e-07, | |
| "loss": 0.2108, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.5797979797979794e-07, | |
| "loss": 0.2477, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.529292929292929e-07, | |
| "loss": 0.2099, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.4787878787878786e-07, | |
| "loss": 0.2025, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.4282828282828284e-07, | |
| "loss": 0.1996, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 2.3777777777777777e-07, | |
| "loss": 0.2068, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 2.3272727272727273e-07, | |
| "loss": 0.1837, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.2767676767676766e-07, | |
| "loss": 0.1894, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.2262626262626262e-07, | |
| "loss": 0.1692, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 2.1757575757575758e-07, | |
| "loss": 0.2537, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 2.1252525252525254e-07, | |
| "loss": 0.1746, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.0747474747474747e-07, | |
| "loss": 0.2555, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 2.0242424242424242e-07, | |
| "loss": 0.2384, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "eval_loss": 0.48731526732444763, | |
| "eval_runtime": 2698.6, | |
| "eval_samples_per_second": 3.908, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 39.9208137074203, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.9737373737373736e-07, | |
| "loss": 0.212, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.9232323232323231e-07, | |
| "loss": 0.2351, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.8727272727272727e-07, | |
| "loss": 0.2033, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.8222222222222223e-07, | |
| "loss": 0.2018, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.7737373737373736e-07, | |
| "loss": 0.1997, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.7232323232323232e-07, | |
| "loss": 0.1918, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.6727272727272725e-07, | |
| "loss": 0.1905, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.622222222222222e-07, | |
| "loss": 0.1716, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.5717171717171717e-07, | |
| "loss": 0.178, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.5212121212121213e-07, | |
| "loss": 0.2347, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.4707070707070706e-07, | |
| "loss": 0.1808, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.4202020202020202e-07, | |
| "loss": 0.224, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.3696969696969695e-07, | |
| "loss": 0.2412, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.319191919191919e-07, | |
| "loss": 0.2154, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.2686868686868686e-07, | |
| "loss": 0.2228, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.218181818181818e-07, | |
| "loss": 0.2142, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 1.1676767676767677e-07, | |
| "loss": 0.1944, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 1.1171717171717171e-07, | |
| "loss": 0.2003, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 1.0666666666666667e-07, | |
| "loss": 0.1805, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 1.0161616161616161e-07, | |
| "loss": 0.1889, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "eval_loss": 0.48579299449920654, | |
| "eval_runtime": 2674.4789, | |
| "eval_samples_per_second": 3.944, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 39.72011741415796, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 9.656565656565656e-08, | |
| "loss": 0.1683, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 9.151515151515152e-08, | |
| "loss": 0.1798, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 8.646464646464646e-08, | |
| "loss": 0.2081, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 8.14141414141414e-08, | |
| "loss": 0.2018, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 7.636363636363636e-08, | |
| "loss": 0.2006, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 7.131313131313131e-08, | |
| "loss": 0.2452, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 6.626262626262625e-08, | |
| "loss": 0.2166, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 6.121212121212121e-08, | |
| "loss": 0.201, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 5.616161616161616e-08, | |
| "loss": 0.2311, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 5.1111111111111114e-08, | |
| "loss": 0.1939, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 4.606060606060606e-08, | |
| "loss": 0.2, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 4.101010101010101e-08, | |
| "loss": 0.1681, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 3.595959595959596e-08, | |
| "loss": 0.1962, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 3.090909090909091e-08, | |
| "loss": 0.1646, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 2.5858585858585858e-08, | |
| "loss": 0.1801, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 2.0808080808080806e-08, | |
| "loss": 0.1839, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 1.5757575757575758e-08, | |
| "loss": 0.2263, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 1.0707070707070706e-08, | |
| "loss": 0.1765, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 5.656565656565656e-09, | |
| "loss": 0.2496, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 6.06060606060606e-10, | |
| "loss": 0.2202, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "eval_loss": 0.48882269859313965, | |
| "eval_runtime": 2675.553, | |
| "eval_samples_per_second": 3.942, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 39.72284797597106, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "step": 5000, | |
| "total_flos": 9.22347797962752e+19, | |
| "train_loss": 0.30594726686477663, | |
| "train_runtime": 69031.1556, | |
| "train_samples_per_second": 4.636, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 9.22347797962752e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |