| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "eval_steps": 100, | |
| "global_step": 28800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10416666666666667, | |
| "eval_loss": 0.07793249189853668, | |
| "eval_runtime": 3.1433, | |
| "eval_samples_per_second": 159.069, | |
| "eval_steps_per_second": 15.907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.20833333333333334, | |
| "eval_loss": 0.060976069420576096, | |
| "eval_runtime": 5.2088, | |
| "eval_samples_per_second": 95.992, | |
| "eval_steps_per_second": 9.599, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "eval_loss": 0.06986938416957855, | |
| "eval_runtime": 5.0795, | |
| "eval_samples_per_second": 98.436, | |
| "eval_steps_per_second": 9.844, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "eval_loss": 0.049215417355298996, | |
| "eval_runtime": 3.1571, | |
| "eval_samples_per_second": 158.372, | |
| "eval_steps_per_second": 15.837, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5208333333333334, | |
| "grad_norm": 2.454777240753174, | |
| "learning_rate": 8.680555555555556e-06, | |
| "loss": 0.1299, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5208333333333334, | |
| "eval_loss": 0.05049608275294304, | |
| "eval_runtime": 3.1336, | |
| "eval_samples_per_second": 159.56, | |
| "eval_steps_per_second": 15.956, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "eval_loss": 0.05414561927318573, | |
| "eval_runtime": 3.0995, | |
| "eval_samples_per_second": 161.315, | |
| "eval_steps_per_second": 16.131, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7291666666666666, | |
| "eval_loss": 0.058882277458906174, | |
| "eval_runtime": 3.2082, | |
| "eval_samples_per_second": 155.852, | |
| "eval_steps_per_second": 15.585, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "eval_loss": 0.05974648892879486, | |
| "eval_runtime": 4.9789, | |
| "eval_samples_per_second": 100.424, | |
| "eval_steps_per_second": 10.042, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "eval_loss": 0.06364227831363678, | |
| "eval_runtime": 3.1858, | |
| "eval_samples_per_second": 156.948, | |
| "eval_steps_per_second": 15.695, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.0416666666666667, | |
| "grad_norm": 0.6711832284927368, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.1028, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0416666666666667, | |
| "eval_loss": 0.07583945989608765, | |
| "eval_runtime": 3.1485, | |
| "eval_samples_per_second": 158.808, | |
| "eval_steps_per_second": 15.881, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.1458333333333333, | |
| "eval_loss": 0.06294066458940506, | |
| "eval_runtime": 3.7666, | |
| "eval_samples_per_second": 132.746, | |
| "eval_steps_per_second": 13.275, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 0.06619992107152939, | |
| "eval_runtime": 3.2391, | |
| "eval_samples_per_second": 154.363, | |
| "eval_steps_per_second": 15.436, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.3541666666666667, | |
| "eval_loss": 0.09093113243579865, | |
| "eval_runtime": 5.1419, | |
| "eval_samples_per_second": 97.241, | |
| "eval_steps_per_second": 9.724, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.4583333333333333, | |
| "eval_loss": 0.10987065732479095, | |
| "eval_runtime": 3.0882, | |
| "eval_samples_per_second": 161.906, | |
| "eval_steps_per_second": 16.191, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 23.796688079833984, | |
| "learning_rate": 2.604166666666667e-05, | |
| "loss": 0.091, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "eval_loss": 0.10730883479118347, | |
| "eval_runtime": 3.112, | |
| "eval_samples_per_second": 160.667, | |
| "eval_steps_per_second": 16.067, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "eval_loss": 0.11321321874856949, | |
| "eval_runtime": 3.1391, | |
| "eval_samples_per_second": 159.279, | |
| "eval_steps_per_second": 15.928, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.7708333333333335, | |
| "eval_loss": 0.09883977472782135, | |
| "eval_runtime": 5.0716, | |
| "eval_samples_per_second": 98.587, | |
| "eval_steps_per_second": 9.859, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "eval_loss": 0.11777649074792862, | |
| "eval_runtime": 5.1982, | |
| "eval_samples_per_second": 96.187, | |
| "eval_steps_per_second": 9.619, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.9791666666666665, | |
| "eval_loss": 0.09968377649784088, | |
| "eval_runtime": 3.1348, | |
| "eval_samples_per_second": 159.499, | |
| "eval_steps_per_second": 15.95, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 1.0350979566574097, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.1147, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "eval_loss": 0.0993076041340828, | |
| "eval_runtime": 3.1904, | |
| "eval_samples_per_second": 156.719, | |
| "eval_steps_per_second": 15.672, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "eval_loss": 0.09541837126016617, | |
| "eval_runtime": 3.0402, | |
| "eval_samples_per_second": 164.463, | |
| "eval_steps_per_second": 16.446, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.2916666666666665, | |
| "eval_loss": 0.1211095005273819, | |
| "eval_runtime": 3.0483, | |
| "eval_samples_per_second": 164.026, | |
| "eval_steps_per_second": 16.403, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.3958333333333335, | |
| "eval_loss": 0.10396522283554077, | |
| "eval_runtime": 3.0322, | |
| "eval_samples_per_second": 164.895, | |
| "eval_steps_per_second": 16.489, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.15123824775218964, | |
| "eval_runtime": 3.0437, | |
| "eval_samples_per_second": 164.275, | |
| "eval_steps_per_second": 16.427, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.6041666666666665, | |
| "grad_norm": 2.144970178604126, | |
| "learning_rate": 4.340277777777778e-05, | |
| "loss": 0.0832, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.6041666666666665, | |
| "eval_loss": 0.16908738017082214, | |
| "eval_runtime": 4.9423, | |
| "eval_samples_per_second": 101.167, | |
| "eval_steps_per_second": 10.117, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.7083333333333335, | |
| "eval_loss": 0.16467152535915375, | |
| "eval_runtime": 3.999, | |
| "eval_samples_per_second": 125.033, | |
| "eval_steps_per_second": 12.503, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "eval_loss": 0.19610275328159332, | |
| "eval_runtime": 4.7478, | |
| "eval_samples_per_second": 105.311, | |
| "eval_steps_per_second": 10.531, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.9166666666666665, | |
| "eval_loss": 0.2006768137216568, | |
| "eval_runtime": 4.7959, | |
| "eval_samples_per_second": 104.256, | |
| "eval_steps_per_second": 10.426, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.0208333333333335, | |
| "eval_loss": 0.1452997922897339, | |
| "eval_runtime": 4.8491, | |
| "eval_samples_per_second": 103.111, | |
| "eval_steps_per_second": 10.311, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 8.211258888244629, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 0.115, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "eval_loss": 0.17185473442077637, | |
| "eval_runtime": 4.968, | |
| "eval_samples_per_second": 100.644, | |
| "eval_steps_per_second": 10.064, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.2291666666666665, | |
| "eval_loss": 0.1926930844783783, | |
| "eval_runtime": 3.0476, | |
| "eval_samples_per_second": 164.064, | |
| "eval_steps_per_second": 16.406, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "eval_loss": 0.20631477236747742, | |
| "eval_runtime": 3.0178, | |
| "eval_samples_per_second": 165.682, | |
| "eval_steps_per_second": 16.568, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.4375, | |
| "eval_loss": 0.15469783544540405, | |
| "eval_runtime": 3.0052, | |
| "eval_samples_per_second": 166.38, | |
| "eval_steps_per_second": 16.638, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.5416666666666665, | |
| "eval_loss": 0.13334515690803528, | |
| "eval_runtime": 3.0679, | |
| "eval_samples_per_second": 162.979, | |
| "eval_steps_per_second": 16.298, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.6458333333333335, | |
| "grad_norm": 1.2755959033966064, | |
| "learning_rate": 4.880401234567901e-05, | |
| "loss": 0.0954, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.6458333333333335, | |
| "eval_loss": 0.16306829452514648, | |
| "eval_runtime": 3.1165, | |
| "eval_samples_per_second": 160.436, | |
| "eval_steps_per_second": 16.044, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_loss": 0.211606964468956, | |
| "eval_runtime": 2.9864, | |
| "eval_samples_per_second": 167.428, | |
| "eval_steps_per_second": 16.743, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.8541666666666665, | |
| "eval_loss": 0.18188658356666565, | |
| "eval_runtime": 2.984, | |
| "eval_samples_per_second": 167.563, | |
| "eval_steps_per_second": 16.756, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.9583333333333335, | |
| "eval_loss": 0.1797247678041458, | |
| "eval_runtime": 2.9814, | |
| "eval_samples_per_second": 167.708, | |
| "eval_steps_per_second": 16.771, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "eval_loss": 0.16481846570968628, | |
| "eval_runtime": 3.059, | |
| "eval_samples_per_second": 163.454, | |
| "eval_steps_per_second": 16.345, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 7.385697364807129, | |
| "learning_rate": 4.783950617283951e-05, | |
| "loss": 0.0823, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "eval_loss": 0.20793281495571136, | |
| "eval_runtime": 2.996, | |
| "eval_samples_per_second": 166.89, | |
| "eval_steps_per_second": 16.689, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.270833333333333, | |
| "eval_loss": 0.1974990963935852, | |
| "eval_runtime": 3.0328, | |
| "eval_samples_per_second": 164.862, | |
| "eval_steps_per_second": 16.486, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "eval_loss": 0.24861294031143188, | |
| "eval_runtime": 2.9906, | |
| "eval_samples_per_second": 167.189, | |
| "eval_steps_per_second": 16.719, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.479166666666667, | |
| "eval_loss": 0.1775345653295517, | |
| "eval_runtime": 2.9925, | |
| "eval_samples_per_second": 167.083, | |
| "eval_steps_per_second": 16.708, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 4.583333333333333, | |
| "eval_loss": 0.1704624593257904, | |
| "eval_runtime": 3.0684, | |
| "eval_samples_per_second": 162.95, | |
| "eval_steps_per_second": 16.295, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "grad_norm": 0.6013280749320984, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 0.0549, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "eval_loss": 0.22908172011375427, | |
| "eval_runtime": 3.0692, | |
| "eval_samples_per_second": 162.907, | |
| "eval_steps_per_second": 16.291, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.791666666666667, | |
| "eval_loss": 0.2069811075925827, | |
| "eval_runtime": 2.9792, | |
| "eval_samples_per_second": 167.83, | |
| "eval_steps_per_second": 16.783, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.895833333333333, | |
| "eval_loss": 0.20501156151294708, | |
| "eval_runtime": 2.9759, | |
| "eval_samples_per_second": 168.015, | |
| "eval_steps_per_second": 16.801, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.1834719032049179, | |
| "eval_runtime": 2.9965, | |
| "eval_samples_per_second": 166.864, | |
| "eval_steps_per_second": 16.686, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.104166666666667, | |
| "eval_loss": 0.1747453212738037, | |
| "eval_runtime": 3.0464, | |
| "eval_samples_per_second": 164.127, | |
| "eval_steps_per_second": 16.413, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 5.208333333333333, | |
| "grad_norm": 0.5279375910758972, | |
| "learning_rate": 4.591049382716049e-05, | |
| "loss": 0.0518, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.208333333333333, | |
| "eval_loss": 0.21121186017990112, | |
| "eval_runtime": 2.9779, | |
| "eval_samples_per_second": 167.902, | |
| "eval_steps_per_second": 16.79, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "eval_loss": 0.20287540555000305, | |
| "eval_runtime": 3.0068, | |
| "eval_samples_per_second": 166.292, | |
| "eval_steps_per_second": 16.629, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 5.416666666666667, | |
| "eval_loss": 0.20998916029930115, | |
| "eval_runtime": 3.0065, | |
| "eval_samples_per_second": 166.307, | |
| "eval_steps_per_second": 16.631, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.520833333333333, | |
| "eval_loss": 0.2178017646074295, | |
| "eval_runtime": 2.9901, | |
| "eval_samples_per_second": 167.22, | |
| "eval_steps_per_second": 16.722, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "eval_loss": 0.2072754055261612, | |
| "eval_runtime": 3.0001, | |
| "eval_samples_per_second": 166.662, | |
| "eval_steps_per_second": 16.666, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.729166666666667, | |
| "grad_norm": 1.3259690999984741, | |
| "learning_rate": 4.494598765432099e-05, | |
| "loss": 0.0363, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.729166666666667, | |
| "eval_loss": 0.16645069420337677, | |
| "eval_runtime": 3.0133, | |
| "eval_samples_per_second": 165.933, | |
| "eval_steps_per_second": 16.593, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "eval_loss": 0.18585015833377838, | |
| "eval_runtime": 3.0016, | |
| "eval_samples_per_second": 166.577, | |
| "eval_steps_per_second": 16.658, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 5.9375, | |
| "eval_loss": 0.24183067679405212, | |
| "eval_runtime": 3.0665, | |
| "eval_samples_per_second": 163.05, | |
| "eval_steps_per_second": 16.305, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.041666666666667, | |
| "eval_loss": 0.22054076194763184, | |
| "eval_runtime": 2.9689, | |
| "eval_samples_per_second": 168.413, | |
| "eval_steps_per_second": 16.841, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 6.145833333333333, | |
| "eval_loss": 0.17449523508548737, | |
| "eval_runtime": 3.0168, | |
| "eval_samples_per_second": 165.739, | |
| "eval_steps_per_second": 16.574, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.3047299087047577, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.0305, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_loss": 0.20409347116947174, | |
| "eval_runtime": 2.9665, | |
| "eval_samples_per_second": 168.549, | |
| "eval_steps_per_second": 16.855, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.354166666666667, | |
| "eval_loss": 0.23080451786518097, | |
| "eval_runtime": 3.0955, | |
| "eval_samples_per_second": 161.524, | |
| "eval_steps_per_second": 16.152, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 6.458333333333333, | |
| "eval_loss": 0.23451215028762817, | |
| "eval_runtime": 2.9667, | |
| "eval_samples_per_second": 168.535, | |
| "eval_steps_per_second": 16.854, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "eval_loss": 0.1807590276002884, | |
| "eval_runtime": 3.0657, | |
| "eval_samples_per_second": 163.093, | |
| "eval_steps_per_second": 16.309, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "eval_loss": 0.21056562662124634, | |
| "eval_runtime": 2.9809, | |
| "eval_samples_per_second": 167.732, | |
| "eval_steps_per_second": 16.773, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 6.770833333333333, | |
| "grad_norm": 0.7045068144798279, | |
| "learning_rate": 4.301697530864198e-05, | |
| "loss": 0.0221, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.770833333333333, | |
| "eval_loss": 0.2072342485189438, | |
| "eval_runtime": 2.97, | |
| "eval_samples_per_second": 168.353, | |
| "eval_steps_per_second": 16.835, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "eval_loss": 0.2312939465045929, | |
| "eval_runtime": 3.0018, | |
| "eval_samples_per_second": 166.566, | |
| "eval_steps_per_second": 16.657, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.979166666666667, | |
| "eval_loss": 0.2178051918745041, | |
| "eval_runtime": 3.0112, | |
| "eval_samples_per_second": 166.045, | |
| "eval_steps_per_second": 16.605, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 7.083333333333333, | |
| "eval_loss": 0.1962529867887497, | |
| "eval_runtime": 2.9373, | |
| "eval_samples_per_second": 170.222, | |
| "eval_steps_per_second": 17.022, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 7.1875, | |
| "eval_loss": 0.23245805501937866, | |
| "eval_runtime": 2.9287, | |
| "eval_samples_per_second": 170.727, | |
| "eval_steps_per_second": 17.073, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 7.291666666666667, | |
| "grad_norm": 0.3293949365615845, | |
| "learning_rate": 4.205246913580247e-05, | |
| "loss": 0.0191, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.291666666666667, | |
| "eval_loss": 0.20682944357395172, | |
| "eval_runtime": 2.922, | |
| "eval_samples_per_second": 171.118, | |
| "eval_steps_per_second": 17.112, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.395833333333333, | |
| "eval_loss": 0.2781427800655365, | |
| "eval_runtime": 3.0086, | |
| "eval_samples_per_second": 166.191, | |
| "eval_steps_per_second": 16.619, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_loss": 0.23378807306289673, | |
| "eval_runtime": 2.928, | |
| "eval_samples_per_second": 170.763, | |
| "eval_steps_per_second": 17.076, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 7.604166666666667, | |
| "eval_loss": 0.20770449936389923, | |
| "eval_runtime": 2.9745, | |
| "eval_samples_per_second": 168.097, | |
| "eval_steps_per_second": 16.81, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 7.708333333333333, | |
| "eval_loss": 0.19365566968917847, | |
| "eval_runtime": 2.9194, | |
| "eval_samples_per_second": 171.267, | |
| "eval_steps_per_second": 17.127, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "grad_norm": 0.15344583988189697, | |
| "learning_rate": 4.1087962962962965e-05, | |
| "loss": 0.0185, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "eval_loss": 0.22151848673820496, | |
| "eval_runtime": 2.901, | |
| "eval_samples_per_second": 172.356, | |
| "eval_steps_per_second": 17.236, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.916666666666667, | |
| "eval_loss": 0.18537622690200806, | |
| "eval_runtime": 2.9225, | |
| "eval_samples_per_second": 171.086, | |
| "eval_steps_per_second": 17.109, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 8.020833333333334, | |
| "eval_loss": 0.21262435615062714, | |
| "eval_runtime": 2.9556, | |
| "eval_samples_per_second": 169.168, | |
| "eval_steps_per_second": 16.917, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "eval_loss": 0.24334414303302765, | |
| "eval_runtime": 2.9324, | |
| "eval_samples_per_second": 170.51, | |
| "eval_steps_per_second": 17.051, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 8.229166666666666, | |
| "eval_loss": 0.2494126558303833, | |
| "eval_runtime": 2.9435, | |
| "eval_samples_per_second": 169.864, | |
| "eval_steps_per_second": 16.986, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 0.12840139865875244, | |
| "learning_rate": 4.012345679012346e-05, | |
| "loss": 0.0152, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "eval_loss": 0.18908418715000153, | |
| "eval_runtime": 2.9931, | |
| "eval_samples_per_second": 167.052, | |
| "eval_steps_per_second": 16.705, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.4375, | |
| "eval_loss": 0.21541763842105865, | |
| "eval_runtime": 2.9627, | |
| "eval_samples_per_second": 168.765, | |
| "eval_steps_per_second": 16.876, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 8.541666666666666, | |
| "eval_loss": 0.1759854555130005, | |
| "eval_runtime": 2.9612, | |
| "eval_samples_per_second": 168.849, | |
| "eval_steps_per_second": 16.885, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 8.645833333333334, | |
| "eval_loss": 0.19276337325572968, | |
| "eval_runtime": 3.0833, | |
| "eval_samples_per_second": 162.166, | |
| "eval_steps_per_second": 16.217, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "eval_loss": 0.2762078642845154, | |
| "eval_runtime": 2.9688, | |
| "eval_samples_per_second": 168.42, | |
| "eval_steps_per_second": 16.842, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 8.854166666666666, | |
| "grad_norm": 0.2024412602186203, | |
| "learning_rate": 3.915895061728395e-05, | |
| "loss": 0.0131, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 8.854166666666666, | |
| "eval_loss": 0.2193083018064499, | |
| "eval_runtime": 3.0051, | |
| "eval_samples_per_second": 166.385, | |
| "eval_steps_per_second": 16.639, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 8.958333333333334, | |
| "eval_loss": 0.2521556615829468, | |
| "eval_runtime": 2.9901, | |
| "eval_samples_per_second": 167.221, | |
| "eval_steps_per_second": 16.722, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 9.0625, | |
| "eval_loss": 0.25700661540031433, | |
| "eval_runtime": 2.9726, | |
| "eval_samples_per_second": 168.202, | |
| "eval_steps_per_second": 16.82, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "eval_loss": 0.2857784628868103, | |
| "eval_runtime": 3.0972, | |
| "eval_samples_per_second": 161.436, | |
| "eval_steps_per_second": 16.144, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 9.270833333333334, | |
| "eval_loss": 0.24449588358402252, | |
| "eval_runtime": 2.9884, | |
| "eval_samples_per_second": 167.312, | |
| "eval_steps_per_second": 16.731, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.406023770570755, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.0109, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "eval_loss": 0.2514427900314331, | |
| "eval_runtime": 2.9485, | |
| "eval_samples_per_second": 169.58, | |
| "eval_steps_per_second": 16.958, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.479166666666666, | |
| "eval_loss": 0.2540574371814728, | |
| "eval_runtime": 2.9427, | |
| "eval_samples_per_second": 169.914, | |
| "eval_steps_per_second": 16.991, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 9.583333333333334, | |
| "eval_loss": 0.24942024052143097, | |
| "eval_runtime": 2.8732, | |
| "eval_samples_per_second": 174.023, | |
| "eval_steps_per_second": 17.402, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 9.6875, | |
| "eval_loss": 0.25903967022895813, | |
| "eval_runtime": 2.8416, | |
| "eval_samples_per_second": 175.956, | |
| "eval_steps_per_second": 17.596, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 9.791666666666666, | |
| "eval_loss": 0.21632665395736694, | |
| "eval_runtime": 2.787, | |
| "eval_samples_per_second": 179.408, | |
| "eval_steps_per_second": 17.941, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 9.895833333333334, | |
| "grad_norm": 0.3254503309726715, | |
| "learning_rate": 3.722993827160494e-05, | |
| "loss": 0.0117, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 9.895833333333334, | |
| "eval_loss": 0.22778008878231049, | |
| "eval_runtime": 2.7626, | |
| "eval_samples_per_second": 180.991, | |
| "eval_steps_per_second": 18.099, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.25083282589912415, | |
| "eval_runtime": 2.6929, | |
| "eval_samples_per_second": 185.676, | |
| "eval_steps_per_second": 18.568, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 10.104166666666666, | |
| "eval_loss": 0.27844151854515076, | |
| "eval_runtime": 2.6854, | |
| "eval_samples_per_second": 186.195, | |
| "eval_steps_per_second": 18.62, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 10.208333333333334, | |
| "eval_loss": 0.2655443549156189, | |
| "eval_runtime": 2.6574, | |
| "eval_samples_per_second": 188.155, | |
| "eval_steps_per_second": 18.815, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 10.3125, | |
| "eval_loss": 0.27213552594184875, | |
| "eval_runtime": 2.7641, | |
| "eval_samples_per_second": 180.888, | |
| "eval_steps_per_second": 18.089, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 10.416666666666666, | |
| "grad_norm": 0.01915908418595791, | |
| "learning_rate": 3.626543209876543e-05, | |
| "loss": 0.01, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.416666666666666, | |
| "eval_loss": 0.2086724191904068, | |
| "eval_runtime": 2.6276, | |
| "eval_samples_per_second": 190.287, | |
| "eval_steps_per_second": 19.029, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.520833333333334, | |
| "eval_loss": 0.21828189492225647, | |
| "eval_runtime": 2.7043, | |
| "eval_samples_per_second": 184.893, | |
| "eval_steps_per_second": 18.489, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 10.625, | |
| "eval_loss": 0.17361362278461456, | |
| "eval_runtime": 2.6359, | |
| "eval_samples_per_second": 189.686, | |
| "eval_steps_per_second": 18.969, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 10.729166666666666, | |
| "eval_loss": 0.23900029063224792, | |
| "eval_runtime": 2.6622, | |
| "eval_samples_per_second": 187.815, | |
| "eval_steps_per_second": 18.781, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 10.833333333333334, | |
| "eval_loss": 0.21116891503334045, | |
| "eval_runtime": 2.7512, | |
| "eval_samples_per_second": 181.737, | |
| "eval_steps_per_second": 18.174, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 10.9375, | |
| "grad_norm": 0.15779471397399902, | |
| "learning_rate": 3.530092592592593e-05, | |
| "loss": 0.0079, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 10.9375, | |
| "eval_loss": 0.22312691807746887, | |
| "eval_runtime": 2.6146, | |
| "eval_samples_per_second": 191.236, | |
| "eval_steps_per_second": 19.124, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 11.041666666666666, | |
| "eval_loss": 0.19351090490818024, | |
| "eval_runtime": 2.6666, | |
| "eval_samples_per_second": 187.503, | |
| "eval_steps_per_second": 18.75, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 11.145833333333334, | |
| "eval_loss": 0.22429315745830536, | |
| "eval_runtime": 2.7037, | |
| "eval_samples_per_second": 184.933, | |
| "eval_steps_per_second": 18.493, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "eval_loss": 0.2761378884315491, | |
| "eval_runtime": 2.6376, | |
| "eval_samples_per_second": 189.563, | |
| "eval_steps_per_second": 18.956, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 11.354166666666666, | |
| "eval_loss": 0.21286267042160034, | |
| "eval_runtime": 2.7376, | |
| "eval_samples_per_second": 182.64, | |
| "eval_steps_per_second": 18.264, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 11.458333333333334, | |
| "grad_norm": 28.01507568359375, | |
| "learning_rate": 3.4336419753086427e-05, | |
| "loss": 0.0076, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.458333333333334, | |
| "eval_loss": 0.16870196163654327, | |
| "eval_runtime": 2.6961, | |
| "eval_samples_per_second": 185.451, | |
| "eval_steps_per_second": 18.545, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.5625, | |
| "eval_loss": 0.24428625404834747, | |
| "eval_runtime": 2.6808, | |
| "eval_samples_per_second": 186.514, | |
| "eval_steps_per_second": 18.651, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 11.666666666666666, | |
| "eval_loss": 0.24231122434139252, | |
| "eval_runtime": 2.6904, | |
| "eval_samples_per_second": 185.843, | |
| "eval_steps_per_second": 18.584, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 11.770833333333334, | |
| "eval_loss": 0.29215526580810547, | |
| "eval_runtime": 2.6528, | |
| "eval_samples_per_second": 188.481, | |
| "eval_steps_per_second": 18.848, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 11.875, | |
| "eval_loss": 0.22481079399585724, | |
| "eval_runtime": 2.6638, | |
| "eval_samples_per_second": 187.703, | |
| "eval_steps_per_second": 18.77, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 11.979166666666666, | |
| "grad_norm": 0.09294537454843521, | |
| "learning_rate": 3.337191358024692e-05, | |
| "loss": 0.0074, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 11.979166666666666, | |
| "eval_loss": 0.21293191611766815, | |
| "eval_runtime": 2.706, | |
| "eval_samples_per_second": 184.775, | |
| "eval_steps_per_second": 18.478, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 12.083333333333334, | |
| "eval_loss": 0.23273849487304688, | |
| "eval_runtime": 2.6588, | |
| "eval_samples_per_second": 188.053, | |
| "eval_steps_per_second": 18.805, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 12.1875, | |
| "eval_loss": 0.23750942945480347, | |
| "eval_runtime": 2.6738, | |
| "eval_samples_per_second": 186.999, | |
| "eval_steps_per_second": 18.7, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 12.291666666666666, | |
| "eval_loss": 0.26990386843681335, | |
| "eval_runtime": 2.6509, | |
| "eval_samples_per_second": 188.618, | |
| "eval_steps_per_second": 18.862, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 12.395833333333334, | |
| "eval_loss": 0.24741919338703156, | |
| "eval_runtime": 2.6554, | |
| "eval_samples_per_second": 188.299, | |
| "eval_steps_per_second": 18.83, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 1.0478854179382324, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.0056, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_loss": 0.22709093987941742, | |
| "eval_runtime": 2.7368, | |
| "eval_samples_per_second": 182.693, | |
| "eval_steps_per_second": 18.269, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.604166666666666, | |
| "eval_loss": 0.2405824214220047, | |
| "eval_runtime": 2.6757, | |
| "eval_samples_per_second": 186.87, | |
| "eval_steps_per_second": 18.687, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 12.708333333333334, | |
| "eval_loss": 0.28877463936805725, | |
| "eval_runtime": 2.671, | |
| "eval_samples_per_second": 187.194, | |
| "eval_steps_per_second": 18.719, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 12.8125, | |
| "eval_loss": 0.3177616596221924, | |
| "eval_runtime": 2.6749, | |
| "eval_samples_per_second": 186.922, | |
| "eval_steps_per_second": 18.692, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 12.916666666666666, | |
| "eval_loss": 0.2896740734577179, | |
| "eval_runtime": 2.6676, | |
| "eval_samples_per_second": 187.434, | |
| "eval_steps_per_second": 18.743, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 13.020833333333334, | |
| "grad_norm": 0.09383181482553482, | |
| "learning_rate": 3.14429012345679e-05, | |
| "loss": 0.0068, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 13.020833333333334, | |
| "eval_loss": 0.2690945267677307, | |
| "eval_runtime": 2.6854, | |
| "eval_samples_per_second": 186.195, | |
| "eval_steps_per_second": 18.619, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 13.125, | |
| "eval_loss": 0.28043872117996216, | |
| "eval_runtime": 2.7005, | |
| "eval_samples_per_second": 185.153, | |
| "eval_steps_per_second": 18.515, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 13.229166666666666, | |
| "eval_loss": 0.2700558006763458, | |
| "eval_runtime": 2.664, | |
| "eval_samples_per_second": 187.69, | |
| "eval_steps_per_second": 18.769, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "eval_loss": 0.2863883078098297, | |
| "eval_runtime": 2.6459, | |
| "eval_samples_per_second": 188.974, | |
| "eval_steps_per_second": 18.897, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 13.4375, | |
| "eval_loss": 0.31038790941238403, | |
| "eval_runtime": 2.6638, | |
| "eval_samples_per_second": 187.7, | |
| "eval_steps_per_second": 18.77, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 13.541666666666666, | |
| "grad_norm": 0.030933663249015808, | |
| "learning_rate": 3.04783950617284e-05, | |
| "loss": 0.0053, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 13.541666666666666, | |
| "eval_loss": 0.2956462502479553, | |
| "eval_runtime": 2.7335, | |
| "eval_samples_per_second": 182.914, | |
| "eval_steps_per_second": 18.291, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 13.645833333333334, | |
| "eval_loss": 0.24711303412914276, | |
| "eval_runtime": 2.6773, | |
| "eval_samples_per_second": 186.754, | |
| "eval_steps_per_second": 18.675, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "eval_loss": 0.2535928785800934, | |
| "eval_runtime": 2.6697, | |
| "eval_samples_per_second": 187.286, | |
| "eval_steps_per_second": 18.729, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 13.854166666666666, | |
| "eval_loss": 0.23705850541591644, | |
| "eval_runtime": 2.7698, | |
| "eval_samples_per_second": 180.518, | |
| "eval_steps_per_second": 18.052, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 13.958333333333334, | |
| "eval_loss": 0.23593567311763763, | |
| "eval_runtime": 2.6612, | |
| "eval_samples_per_second": 187.888, | |
| "eval_steps_per_second": 18.789, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 14.0625, | |
| "grad_norm": 0.031106941401958466, | |
| "learning_rate": 2.951388888888889e-05, | |
| "loss": 0.0049, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 14.0625, | |
| "eval_loss": 0.2466021180152893, | |
| "eval_runtime": 2.6433, | |
| "eval_samples_per_second": 189.159, | |
| "eval_steps_per_second": 18.916, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 14.166666666666666, | |
| "eval_loss": 0.27085715532302856, | |
| "eval_runtime": 2.6665, | |
| "eval_samples_per_second": 187.513, | |
| "eval_steps_per_second": 18.751, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 14.270833333333334, | |
| "eval_loss": 0.23776446282863617, | |
| "eval_runtime": 2.6623, | |
| "eval_samples_per_second": 187.806, | |
| "eval_steps_per_second": 18.781, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 14.375, | |
| "eval_loss": 0.22552721202373505, | |
| "eval_runtime": 2.696, | |
| "eval_samples_per_second": 185.463, | |
| "eval_steps_per_second": 18.546, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 14.479166666666666, | |
| "eval_loss": 0.25723254680633545, | |
| "eval_runtime": 2.759, | |
| "eval_samples_per_second": 181.228, | |
| "eval_steps_per_second": 18.123, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 14.583333333333334, | |
| "grad_norm": 0.014980652369558811, | |
| "learning_rate": 2.8549382716049384e-05, | |
| "loss": 0.0034, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 14.583333333333334, | |
| "eval_loss": 0.24281509220600128, | |
| "eval_runtime": 2.6514, | |
| "eval_samples_per_second": 188.58, | |
| "eval_steps_per_second": 18.858, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 14.6875, | |
| "eval_loss": 0.28589245676994324, | |
| "eval_runtime": 2.6758, | |
| "eval_samples_per_second": 186.862, | |
| "eval_steps_per_second": 18.686, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 14.791666666666666, | |
| "eval_loss": 0.23985494673252106, | |
| "eval_runtime": 2.6688, | |
| "eval_samples_per_second": 187.353, | |
| "eval_steps_per_second": 18.735, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 14.895833333333334, | |
| "eval_loss": 0.2654513716697693, | |
| "eval_runtime": 2.7853, | |
| "eval_samples_per_second": 179.514, | |
| "eval_steps_per_second": 17.951, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.24361200630664825, | |
| "eval_runtime": 2.6858, | |
| "eval_samples_per_second": 186.162, | |
| "eval_steps_per_second": 18.616, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 15.104166666666666, | |
| "grad_norm": 0.06275557726621628, | |
| "learning_rate": 2.7584876543209875e-05, | |
| "loss": 0.0025, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 15.104166666666666, | |
| "eval_loss": 0.22555358707904816, | |
| "eval_runtime": 2.6629, | |
| "eval_samples_per_second": 187.767, | |
| "eval_steps_per_second": 18.777, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 15.208333333333334, | |
| "eval_loss": 0.2367800772190094, | |
| "eval_runtime": 2.6382, | |
| "eval_samples_per_second": 189.52, | |
| "eval_steps_per_second": 18.952, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 15.3125, | |
| "eval_loss": 0.2452574521303177, | |
| "eval_runtime": 2.7269, | |
| "eval_samples_per_second": 183.358, | |
| "eval_steps_per_second": 18.336, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 15.416666666666666, | |
| "eval_loss": 0.23891642689704895, | |
| "eval_runtime": 2.7402, | |
| "eval_samples_per_second": 182.469, | |
| "eval_steps_per_second": 18.247, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 15.520833333333334, | |
| "eval_loss": 0.25828251242637634, | |
| "eval_runtime": 2.6596, | |
| "eval_samples_per_second": 187.998, | |
| "eval_steps_per_second": 18.8, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "grad_norm": 0.019014783203601837, | |
| "learning_rate": 2.6620370370370372e-05, | |
| "loss": 0.0028, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "eval_loss": 0.21553362905979156, | |
| "eval_runtime": 2.685, | |
| "eval_samples_per_second": 186.223, | |
| "eval_steps_per_second": 18.622, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 15.729166666666666, | |
| "eval_loss": 0.1873685121536255, | |
| "eval_runtime": 2.679, | |
| "eval_samples_per_second": 186.639, | |
| "eval_steps_per_second": 18.664, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 15.833333333333334, | |
| "eval_loss": 0.23436906933784485, | |
| "eval_runtime": 2.6734, | |
| "eval_samples_per_second": 187.027, | |
| "eval_steps_per_second": 18.703, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 15.9375, | |
| "eval_loss": 0.21372827887535095, | |
| "eval_runtime": 2.6738, | |
| "eval_samples_per_second": 187.0, | |
| "eval_steps_per_second": 18.7, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 16.041666666666668, | |
| "eval_loss": 0.28128960728645325, | |
| "eval_runtime": 2.721, | |
| "eval_samples_per_second": 183.754, | |
| "eval_steps_per_second": 18.375, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 16.145833333333332, | |
| "grad_norm": 0.001963632879778743, | |
| "learning_rate": 2.5655864197530866e-05, | |
| "loss": 0.0038, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 16.145833333333332, | |
| "eval_loss": 0.2411942183971405, | |
| "eval_runtime": 2.6679, | |
| "eval_samples_per_second": 187.412, | |
| "eval_steps_per_second": 18.741, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "eval_loss": 0.27309754490852356, | |
| "eval_runtime": 2.6866, | |
| "eval_samples_per_second": 186.109, | |
| "eval_steps_per_second": 18.611, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 16.354166666666668, | |
| "eval_loss": 0.26635468006134033, | |
| "eval_runtime": 2.6688, | |
| "eval_samples_per_second": 187.351, | |
| "eval_steps_per_second": 18.735, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 16.458333333333332, | |
| "eval_loss": 0.2760668694972992, | |
| "eval_runtime": 2.7331, | |
| "eval_samples_per_second": 182.941, | |
| "eval_steps_per_second": 18.294, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 16.5625, | |
| "eval_loss": 0.2560645341873169, | |
| "eval_runtime": 2.662, | |
| "eval_samples_per_second": 187.83, | |
| "eval_steps_per_second": 18.783, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "grad_norm": 0.04559716209769249, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.0024, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "eval_loss": 0.31702032685279846, | |
| "eval_runtime": 2.6834, | |
| "eval_samples_per_second": 186.33, | |
| "eval_steps_per_second": 18.633, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 16.770833333333332, | |
| "eval_loss": 0.23096273839473724, | |
| "eval_runtime": 2.643, | |
| "eval_samples_per_second": 189.182, | |
| "eval_steps_per_second": 18.918, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 16.875, | |
| "eval_loss": 0.2803973853588104, | |
| "eval_runtime": 2.6697, | |
| "eval_samples_per_second": 187.288, | |
| "eval_steps_per_second": 18.729, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 16.979166666666668, | |
| "eval_loss": 0.2678157091140747, | |
| "eval_runtime": 2.7817, | |
| "eval_samples_per_second": 179.745, | |
| "eval_steps_per_second": 17.975, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 17.083333333333332, | |
| "eval_loss": 0.27891919016838074, | |
| "eval_runtime": 2.8206, | |
| "eval_samples_per_second": 177.27, | |
| "eval_steps_per_second": 17.727, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 17.1875, | |
| "grad_norm": 0.025386014953255653, | |
| "learning_rate": 2.3726851851851854e-05, | |
| "loss": 0.0021, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 17.1875, | |
| "eval_loss": 0.2423981875181198, | |
| "eval_runtime": 2.6721, | |
| "eval_samples_per_second": 187.12, | |
| "eval_steps_per_second": 18.712, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 17.291666666666668, | |
| "eval_loss": 0.26049262285232544, | |
| "eval_runtime": 2.7078, | |
| "eval_samples_per_second": 184.654, | |
| "eval_steps_per_second": 18.465, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 17.395833333333332, | |
| "eval_loss": 0.30610647797584534, | |
| "eval_runtime": 2.6975, | |
| "eval_samples_per_second": 185.357, | |
| "eval_steps_per_second": 18.536, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "eval_loss": 0.26282158493995667, | |
| "eval_runtime": 2.7046, | |
| "eval_samples_per_second": 184.872, | |
| "eval_steps_per_second": 18.487, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 17.604166666666668, | |
| "eval_loss": 0.24968083202838898, | |
| "eval_runtime": 2.6698, | |
| "eval_samples_per_second": 187.28, | |
| "eval_steps_per_second": 18.728, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 17.708333333333332, | |
| "grad_norm": 0.038250915706157684, | |
| "learning_rate": 2.2762345679012348e-05, | |
| "loss": 0.0033, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 17.708333333333332, | |
| "eval_loss": 0.27164506912231445, | |
| "eval_runtime": 2.6512, | |
| "eval_samples_per_second": 188.591, | |
| "eval_steps_per_second": 18.859, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 17.8125, | |
| "eval_loss": 0.26982831954956055, | |
| "eval_runtime": 2.7234, | |
| "eval_samples_per_second": 183.593, | |
| "eval_steps_per_second": 18.359, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 17.916666666666668, | |
| "eval_loss": 0.2670990824699402, | |
| "eval_runtime": 2.7233, | |
| "eval_samples_per_second": 183.601, | |
| "eval_steps_per_second": 18.36, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 18.020833333333332, | |
| "eval_loss": 0.24990878999233246, | |
| "eval_runtime": 2.6806, | |
| "eval_samples_per_second": 186.525, | |
| "eval_steps_per_second": 18.653, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 18.125, | |
| "eval_loss": 0.2912808656692505, | |
| "eval_runtime": 2.6688, | |
| "eval_samples_per_second": 187.351, | |
| "eval_steps_per_second": 18.735, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 18.229166666666668, | |
| "grad_norm": 0.021504106000065804, | |
| "learning_rate": 2.179783950617284e-05, | |
| "loss": 0.003, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 18.229166666666668, | |
| "eval_loss": 0.26864093542099, | |
| "eval_runtime": 2.6597, | |
| "eval_samples_per_second": 187.99, | |
| "eval_steps_per_second": 18.799, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 18.333333333333332, | |
| "eval_loss": 0.2561093866825104, | |
| "eval_runtime": 2.7405, | |
| "eval_samples_per_second": 182.449, | |
| "eval_steps_per_second": 18.245, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 18.4375, | |
| "eval_loss": 0.2752689719200134, | |
| "eval_runtime": 2.8003, | |
| "eval_samples_per_second": 178.554, | |
| "eval_steps_per_second": 17.855, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 18.541666666666668, | |
| "eval_loss": 0.25926899909973145, | |
| "eval_runtime": 2.6884, | |
| "eval_samples_per_second": 185.987, | |
| "eval_steps_per_second": 18.599, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 18.645833333333332, | |
| "eval_loss": 0.26980525255203247, | |
| "eval_runtime": 2.7348, | |
| "eval_samples_per_second": 182.831, | |
| "eval_steps_per_second": 18.283, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 0.17659549415111542, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.0023, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "eval_loss": 0.22183029353618622, | |
| "eval_runtime": 2.7514, | |
| "eval_samples_per_second": 181.729, | |
| "eval_steps_per_second": 18.173, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 18.854166666666668, | |
| "eval_loss": 0.245712548494339, | |
| "eval_runtime": 2.6909, | |
| "eval_samples_per_second": 185.812, | |
| "eval_steps_per_second": 18.581, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 18.958333333333332, | |
| "eval_loss": 0.23490136861801147, | |
| "eval_runtime": 2.6739, | |
| "eval_samples_per_second": 186.991, | |
| "eval_steps_per_second": 18.699, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 19.0625, | |
| "eval_loss": 0.2799266278743744, | |
| "eval_runtime": 2.6965, | |
| "eval_samples_per_second": 185.424, | |
| "eval_steps_per_second": 18.542, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 19.166666666666668, | |
| "eval_loss": 0.2952316403388977, | |
| "eval_runtime": 2.7321, | |
| "eval_samples_per_second": 183.006, | |
| "eval_steps_per_second": 18.301, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 19.270833333333332, | |
| "grad_norm": 0.023679744452238083, | |
| "learning_rate": 1.9868827160493827e-05, | |
| "loss": 0.0018, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 19.270833333333332, | |
| "eval_loss": 0.23330457508563995, | |
| "eval_runtime": 2.6526, | |
| "eval_samples_per_second": 188.496, | |
| "eval_steps_per_second": 18.85, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 19.375, | |
| "eval_loss": 0.26152676343917847, | |
| "eval_runtime": 2.6547, | |
| "eval_samples_per_second": 188.345, | |
| "eval_steps_per_second": 18.834, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 19.479166666666668, | |
| "eval_loss": 0.2680070996284485, | |
| "eval_runtime": 2.6519, | |
| "eval_samples_per_second": 188.545, | |
| "eval_steps_per_second": 18.855, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 19.583333333333332, | |
| "eval_loss": 0.219146266579628, | |
| "eval_runtime": 2.6887, | |
| "eval_samples_per_second": 185.964, | |
| "eval_steps_per_second": 18.596, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 19.6875, | |
| "eval_loss": 0.2633296251296997, | |
| "eval_runtime": 2.716, | |
| "eval_samples_per_second": 184.095, | |
| "eval_steps_per_second": 18.41, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 19.791666666666668, | |
| "grad_norm": 0.1822015345096588, | |
| "learning_rate": 1.8904320987654324e-05, | |
| "loss": 0.0019, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 19.791666666666668, | |
| "eval_loss": 0.22770513594150543, | |
| "eval_runtime": 2.6476, | |
| "eval_samples_per_second": 188.849, | |
| "eval_steps_per_second": 18.885, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 19.895833333333332, | |
| "eval_loss": 0.2615948021411896, | |
| "eval_runtime": 2.6985, | |
| "eval_samples_per_second": 185.288, | |
| "eval_steps_per_second": 18.529, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.23876366019248962, | |
| "eval_runtime": 2.6951, | |
| "eval_samples_per_second": 185.519, | |
| "eval_steps_per_second": 18.552, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 20.104166666666668, | |
| "eval_loss": 0.22909638285636902, | |
| "eval_runtime": 2.6935, | |
| "eval_samples_per_second": 185.63, | |
| "eval_steps_per_second": 18.563, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 20.208333333333332, | |
| "eval_loss": 0.2323145866394043, | |
| "eval_runtime": 2.7334, | |
| "eval_samples_per_second": 182.92, | |
| "eval_steps_per_second": 18.292, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 20.3125, | |
| "grad_norm": 0.013318472541868687, | |
| "learning_rate": 1.7939814814814815e-05, | |
| "loss": 0.0013, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 20.3125, | |
| "eval_loss": 0.2226209044456482, | |
| "eval_runtime": 2.655, | |
| "eval_samples_per_second": 188.322, | |
| "eval_steps_per_second": 18.832, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 20.416666666666668, | |
| "eval_loss": 0.24030183255672455, | |
| "eval_runtime": 2.6564, | |
| "eval_samples_per_second": 188.222, | |
| "eval_steps_per_second": 18.822, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 20.520833333333332, | |
| "eval_loss": 0.2417810559272766, | |
| "eval_runtime": 2.7353, | |
| "eval_samples_per_second": 182.794, | |
| "eval_steps_per_second": 18.279, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 20.625, | |
| "eval_loss": 0.2691485583782196, | |
| "eval_runtime": 2.7924, | |
| "eval_samples_per_second": 179.056, | |
| "eval_steps_per_second": 17.906, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 20.729166666666668, | |
| "eval_loss": 0.21901768445968628, | |
| "eval_runtime": 2.6311, | |
| "eval_samples_per_second": 190.038, | |
| "eval_steps_per_second": 19.004, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 20.833333333333332, | |
| "grad_norm": 0.024880312383174896, | |
| "learning_rate": 1.697530864197531e-05, | |
| "loss": 0.0014, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 20.833333333333332, | |
| "eval_loss": 0.24805013835430145, | |
| "eval_runtime": 2.6373, | |
| "eval_samples_per_second": 189.588, | |
| "eval_steps_per_second": 18.959, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 20.9375, | |
| "eval_loss": 0.22968819737434387, | |
| "eval_runtime": 2.6422, | |
| "eval_samples_per_second": 189.237, | |
| "eval_steps_per_second": 18.924, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 21.041666666666668, | |
| "eval_loss": 0.2395009696483612, | |
| "eval_runtime": 2.6643, | |
| "eval_samples_per_second": 187.67, | |
| "eval_steps_per_second": 18.767, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 21.145833333333332, | |
| "eval_loss": 0.2686944603919983, | |
| "eval_runtime": 2.7421, | |
| "eval_samples_per_second": 182.345, | |
| "eval_steps_per_second": 18.235, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "eval_loss": 0.2748400866985321, | |
| "eval_runtime": 2.6544, | |
| "eval_samples_per_second": 188.369, | |
| "eval_steps_per_second": 18.837, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 21.354166666666668, | |
| "grad_norm": 0.020624302327632904, | |
| "learning_rate": 1.6010802469135803e-05, | |
| "loss": 0.0013, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 21.354166666666668, | |
| "eval_loss": 0.2482401579618454, | |
| "eval_runtime": 2.6547, | |
| "eval_samples_per_second": 188.344, | |
| "eval_steps_per_second": 18.834, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 21.458333333333332, | |
| "eval_loss": 0.2579880952835083, | |
| "eval_runtime": 2.6892, | |
| "eval_samples_per_second": 185.929, | |
| "eval_steps_per_second": 18.593, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 21.5625, | |
| "eval_loss": 0.2505187690258026, | |
| "eval_runtime": 2.7076, | |
| "eval_samples_per_second": 184.666, | |
| "eval_steps_per_second": 18.467, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 21.666666666666668, | |
| "eval_loss": 0.24572314321994781, | |
| "eval_runtime": 2.6772, | |
| "eval_samples_per_second": 186.761, | |
| "eval_steps_per_second": 18.676, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 21.770833333333332, | |
| "eval_loss": 0.2615715265274048, | |
| "eval_runtime": 2.6506, | |
| "eval_samples_per_second": 188.636, | |
| "eval_steps_per_second": 18.864, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 21.875, | |
| "grad_norm": 0.10809088498353958, | |
| "learning_rate": 1.5046296296296297e-05, | |
| "loss": 0.0012, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 21.875, | |
| "eval_loss": 0.24613332748413086, | |
| "eval_runtime": 2.664, | |
| "eval_samples_per_second": 187.687, | |
| "eval_steps_per_second": 18.769, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 21.979166666666668, | |
| "eval_loss": 0.2297743260860443, | |
| "eval_runtime": 2.8188, | |
| "eval_samples_per_second": 177.378, | |
| "eval_steps_per_second": 17.738, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 22.083333333333332, | |
| "eval_loss": 0.20882727205753326, | |
| "eval_runtime": 2.6539, | |
| "eval_samples_per_second": 188.4, | |
| "eval_steps_per_second": 18.84, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 22.1875, | |
| "eval_loss": 0.24317896366119385, | |
| "eval_runtime": 2.6727, | |
| "eval_samples_per_second": 187.08, | |
| "eval_steps_per_second": 18.708, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 22.291666666666668, | |
| "eval_loss": 0.24090611934661865, | |
| "eval_runtime": 2.6745, | |
| "eval_samples_per_second": 186.949, | |
| "eval_steps_per_second": 18.695, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 22.395833333333332, | |
| "grad_norm": 0.013452223502099514, | |
| "learning_rate": 1.4081790123456789e-05, | |
| "loss": 0.0006, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 22.395833333333332, | |
| "eval_loss": 0.23966069519519806, | |
| "eval_runtime": 2.6467, | |
| "eval_samples_per_second": 188.915, | |
| "eval_steps_per_second": 18.892, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "eval_loss": 0.2442026436328888, | |
| "eval_runtime": 2.752, | |
| "eval_samples_per_second": 181.687, | |
| "eval_steps_per_second": 18.169, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 22.604166666666668, | |
| "eval_loss": 0.2610822916030884, | |
| "eval_runtime": 2.7823, | |
| "eval_samples_per_second": 179.709, | |
| "eval_steps_per_second": 17.971, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 22.708333333333332, | |
| "eval_loss": 0.2411614954471588, | |
| "eval_runtime": 2.6894, | |
| "eval_samples_per_second": 185.916, | |
| "eval_steps_per_second": 18.592, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 22.8125, | |
| "eval_loss": 0.28125372529029846, | |
| "eval_runtime": 2.6995, | |
| "eval_samples_per_second": 185.22, | |
| "eval_steps_per_second": 18.522, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 22.916666666666668, | |
| "grad_norm": 0.01698753982782364, | |
| "learning_rate": 1.3117283950617285e-05, | |
| "loss": 0.0016, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 22.916666666666668, | |
| "eval_loss": 0.2897321879863739, | |
| "eval_runtime": 2.6638, | |
| "eval_samples_per_second": 187.699, | |
| "eval_steps_per_second": 18.77, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 23.020833333333332, | |
| "eval_loss": 0.2508152425289154, | |
| "eval_runtime": 2.6526, | |
| "eval_samples_per_second": 188.492, | |
| "eval_steps_per_second": 18.849, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 23.125, | |
| "eval_loss": 0.2747707664966583, | |
| "eval_runtime": 2.7474, | |
| "eval_samples_per_second": 181.988, | |
| "eval_steps_per_second": 18.199, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 23.229166666666668, | |
| "eval_loss": 0.24645893275737762, | |
| "eval_runtime": 2.7117, | |
| "eval_samples_per_second": 184.383, | |
| "eval_steps_per_second": 18.438, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 23.333333333333332, | |
| "eval_loss": 0.22453027963638306, | |
| "eval_runtime": 2.6879, | |
| "eval_samples_per_second": 186.022, | |
| "eval_steps_per_second": 18.602, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 23.4375, | |
| "grad_norm": 0.020492762327194214, | |
| "learning_rate": 1.2152777777777779e-05, | |
| "loss": 0.0012, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 23.4375, | |
| "eval_loss": 0.2575179934501648, | |
| "eval_runtime": 2.657, | |
| "eval_samples_per_second": 188.185, | |
| "eval_steps_per_second": 18.818, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 23.541666666666668, | |
| "eval_loss": 0.2540989816188812, | |
| "eval_runtime": 2.6727, | |
| "eval_samples_per_second": 187.077, | |
| "eval_steps_per_second": 18.708, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 23.645833333333332, | |
| "eval_loss": 0.26418963074684143, | |
| "eval_runtime": 2.6889, | |
| "eval_samples_per_second": 185.951, | |
| "eval_steps_per_second": 18.595, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "eval_loss": 0.27296414971351624, | |
| "eval_runtime": 2.6695, | |
| "eval_samples_per_second": 187.304, | |
| "eval_steps_per_second": 18.73, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 23.854166666666668, | |
| "eval_loss": 0.28103941679000854, | |
| "eval_runtime": 2.6878, | |
| "eval_samples_per_second": 186.024, | |
| "eval_steps_per_second": 18.602, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 23.958333333333332, | |
| "grad_norm": 0.019232362508773804, | |
| "learning_rate": 1.1188271604938271e-05, | |
| "loss": 0.0009, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 23.958333333333332, | |
| "eval_loss": 0.29488101601600647, | |
| "eval_runtime": 2.671, | |
| "eval_samples_per_second": 187.195, | |
| "eval_steps_per_second": 18.72, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 24.0625, | |
| "eval_loss": 0.2643510103225708, | |
| "eval_runtime": 2.6969, | |
| "eval_samples_per_second": 185.398, | |
| "eval_steps_per_second": 18.54, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 24.166666666666668, | |
| "eval_loss": 0.28387248516082764, | |
| "eval_runtime": 2.8223, | |
| "eval_samples_per_second": 177.159, | |
| "eval_steps_per_second": 17.716, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 24.270833333333332, | |
| "eval_loss": 0.28446924686431885, | |
| "eval_runtime": 2.79, | |
| "eval_samples_per_second": 179.211, | |
| "eval_steps_per_second": 17.921, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 24.375, | |
| "eval_loss": 0.24371571838855743, | |
| "eval_runtime": 2.6806, | |
| "eval_samples_per_second": 186.527, | |
| "eval_steps_per_second": 18.653, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 24.479166666666668, | |
| "grad_norm": 0.09378495067358017, | |
| "learning_rate": 1.0223765432098765e-05, | |
| "loss": 0.0012, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 24.479166666666668, | |
| "eval_loss": 0.2529699206352234, | |
| "eval_runtime": 2.7479, | |
| "eval_samples_per_second": 181.955, | |
| "eval_steps_per_second": 18.196, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 24.583333333333332, | |
| "eval_loss": 0.23549579083919525, | |
| "eval_runtime": 2.7468, | |
| "eval_samples_per_second": 182.03, | |
| "eval_steps_per_second": 18.203, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 24.6875, | |
| "eval_loss": 0.24692101776599884, | |
| "eval_runtime": 2.687, | |
| "eval_samples_per_second": 186.078, | |
| "eval_steps_per_second": 18.608, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 24.791666666666668, | |
| "eval_loss": 0.2594464123249054, | |
| "eval_runtime": 2.7055, | |
| "eval_samples_per_second": 184.81, | |
| "eval_steps_per_second": 18.481, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 24.895833333333332, | |
| "eval_loss": 0.2568516731262207, | |
| "eval_runtime": 2.7026, | |
| "eval_samples_per_second": 185.01, | |
| "eval_steps_per_second": 18.501, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.01909373700618744, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0006, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.23979052901268005, | |
| "eval_runtime": 2.6347, | |
| "eval_samples_per_second": 189.774, | |
| "eval_steps_per_second": 18.977, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 25.104166666666668, | |
| "eval_loss": 0.25141003727912903, | |
| "eval_runtime": 2.6995, | |
| "eval_samples_per_second": 185.219, | |
| "eval_steps_per_second": 18.522, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 25.208333333333332, | |
| "eval_loss": 0.2534993886947632, | |
| "eval_runtime": 2.6878, | |
| "eval_samples_per_second": 186.027, | |
| "eval_steps_per_second": 18.603, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 25.3125, | |
| "eval_loss": 0.2161099910736084, | |
| "eval_runtime": 2.6906, | |
| "eval_samples_per_second": 185.834, | |
| "eval_steps_per_second": 18.583, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 25.416666666666668, | |
| "eval_loss": 0.2284189760684967, | |
| "eval_runtime": 2.6783, | |
| "eval_samples_per_second": 186.688, | |
| "eval_steps_per_second": 18.669, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 25.520833333333332, | |
| "grad_norm": 0.01035739853978157, | |
| "learning_rate": 8.294753086419753e-06, | |
| "loss": 0.0005, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 25.520833333333332, | |
| "eval_loss": 0.23142491281032562, | |
| "eval_runtime": 2.7424, | |
| "eval_samples_per_second": 182.32, | |
| "eval_steps_per_second": 18.232, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 25.625, | |
| "eval_loss": 0.22284680604934692, | |
| "eval_runtime": 2.667, | |
| "eval_samples_per_second": 187.477, | |
| "eval_steps_per_second": 18.748, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 25.729166666666668, | |
| "eval_loss": 0.23332533240318298, | |
| "eval_runtime": 2.6695, | |
| "eval_samples_per_second": 187.298, | |
| "eval_steps_per_second": 18.73, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 25.833333333333332, | |
| "eval_loss": 0.2309611290693283, | |
| "eval_runtime": 2.7134, | |
| "eval_samples_per_second": 184.273, | |
| "eval_steps_per_second": 18.427, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 25.9375, | |
| "eval_loss": 0.2237754762172699, | |
| "eval_runtime": 2.6731, | |
| "eval_samples_per_second": 187.045, | |
| "eval_steps_per_second": 18.705, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 26.041666666666668, | |
| "grad_norm": 0.023038456216454506, | |
| "learning_rate": 7.330246913580248e-06, | |
| "loss": 0.0015, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 26.041666666666668, | |
| "eval_loss": 0.22988422214984894, | |
| "eval_runtime": 2.6773, | |
| "eval_samples_per_second": 186.754, | |
| "eval_steps_per_second": 18.675, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 26.145833333333332, | |
| "eval_loss": 0.2248789370059967, | |
| "eval_runtime": 2.6901, | |
| "eval_samples_per_second": 185.864, | |
| "eval_steps_per_second": 18.586, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "eval_loss": 0.21588632464408875, | |
| "eval_runtime": 2.7349, | |
| "eval_samples_per_second": 182.823, | |
| "eval_steps_per_second": 18.282, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 26.354166666666668, | |
| "eval_loss": 0.22912240028381348, | |
| "eval_runtime": 2.6991, | |
| "eval_samples_per_second": 185.249, | |
| "eval_steps_per_second": 18.525, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 26.458333333333332, | |
| "eval_loss": 0.24369725584983826, | |
| "eval_runtime": 2.6901, | |
| "eval_samples_per_second": 185.865, | |
| "eval_steps_per_second": 18.586, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 26.5625, | |
| "grad_norm": 0.004963716492056847, | |
| "learning_rate": 6.365740740740741e-06, | |
| "loss": 0.0009, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 26.5625, | |
| "eval_loss": 0.24551524221897125, | |
| "eval_runtime": 2.6597, | |
| "eval_samples_per_second": 187.991, | |
| "eval_steps_per_second": 18.799, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "eval_loss": 0.2321571707725525, | |
| "eval_runtime": 2.7245, | |
| "eval_samples_per_second": 183.522, | |
| "eval_steps_per_second": 18.352, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 26.770833333333332, | |
| "eval_loss": 0.2371620088815689, | |
| "eval_runtime": 2.6794, | |
| "eval_samples_per_second": 186.607, | |
| "eval_steps_per_second": 18.661, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 26.875, | |
| "eval_loss": 0.24450993537902832, | |
| "eval_runtime": 2.6866, | |
| "eval_samples_per_second": 186.112, | |
| "eval_steps_per_second": 18.611, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 26.979166666666668, | |
| "eval_loss": 0.24821239709854126, | |
| "eval_runtime": 2.6972, | |
| "eval_samples_per_second": 185.379, | |
| "eval_steps_per_second": 18.538, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 27.083333333333332, | |
| "grad_norm": 0.018454568460583687, | |
| "learning_rate": 5.401234567901234e-06, | |
| "loss": 0.0004, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 27.083333333333332, | |
| "eval_loss": 0.23901359736919403, | |
| "eval_runtime": 2.6658, | |
| "eval_samples_per_second": 187.559, | |
| "eval_steps_per_second": 18.756, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 27.1875, | |
| "eval_loss": 0.24414636194705963, | |
| "eval_runtime": 2.7605, | |
| "eval_samples_per_second": 181.124, | |
| "eval_steps_per_second": 18.112, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 27.291666666666668, | |
| "eval_loss": 0.22809472680091858, | |
| "eval_runtime": 2.6599, | |
| "eval_samples_per_second": 187.975, | |
| "eval_steps_per_second": 18.798, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 27.395833333333332, | |
| "eval_loss": 0.23070019483566284, | |
| "eval_runtime": 2.6735, | |
| "eval_samples_per_second": 187.017, | |
| "eval_steps_per_second": 18.702, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "eval_loss": 0.22740352153778076, | |
| "eval_runtime": 2.6801, | |
| "eval_samples_per_second": 186.562, | |
| "eval_steps_per_second": 18.656, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 27.604166666666668, | |
| "grad_norm": 0.024484924972057343, | |
| "learning_rate": 4.436728395061729e-06, | |
| "loss": 0.0003, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 27.604166666666668, | |
| "eval_loss": 0.23273907601833344, | |
| "eval_runtime": 2.6798, | |
| "eval_samples_per_second": 186.583, | |
| "eval_steps_per_second": 18.658, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 27.708333333333332, | |
| "eval_loss": 0.246334508061409, | |
| "eval_runtime": 2.7151, | |
| "eval_samples_per_second": 184.157, | |
| "eval_steps_per_second": 18.416, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 27.8125, | |
| "eval_loss": 0.2416531890630722, | |
| "eval_runtime": 2.7757, | |
| "eval_samples_per_second": 180.134, | |
| "eval_steps_per_second": 18.013, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 27.916666666666668, | |
| "eval_loss": 0.2483607828617096, | |
| "eval_runtime": 2.7131, | |
| "eval_samples_per_second": 184.294, | |
| "eval_steps_per_second": 18.429, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 28.020833333333332, | |
| "eval_loss": 0.252419114112854, | |
| "eval_runtime": 2.6895, | |
| "eval_samples_per_second": 185.906, | |
| "eval_steps_per_second": 18.591, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 28.125, | |
| "grad_norm": 0.0021363645792007446, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.0005, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 28.125, | |
| "eval_loss": 0.24915319681167603, | |
| "eval_runtime": 2.6788, | |
| "eval_samples_per_second": 186.648, | |
| "eval_steps_per_second": 18.665, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 28.229166666666668, | |
| "eval_loss": 0.2466410994529724, | |
| "eval_runtime": 2.7305, | |
| "eval_samples_per_second": 183.118, | |
| "eval_steps_per_second": 18.312, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 28.333333333333332, | |
| "eval_loss": 0.23784339427947998, | |
| "eval_runtime": 2.7758, | |
| "eval_samples_per_second": 180.131, | |
| "eval_steps_per_second": 18.013, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 28.4375, | |
| "eval_loss": 0.23728083074092865, | |
| "eval_runtime": 2.7003, | |
| "eval_samples_per_second": 185.168, | |
| "eval_steps_per_second": 18.517, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 28.541666666666668, | |
| "eval_loss": 0.23528096079826355, | |
| "eval_runtime": 2.6782, | |
| "eval_samples_per_second": 186.695, | |
| "eval_steps_per_second": 18.669, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 28.645833333333332, | |
| "grad_norm": 0.004232197534292936, | |
| "learning_rate": 2.5077160493827164e-06, | |
| "loss": 0.0007, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 28.645833333333332, | |
| "eval_loss": 0.22034627199172974, | |
| "eval_runtime": 2.6537, | |
| "eval_samples_per_second": 188.417, | |
| "eval_steps_per_second": 18.842, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "eval_loss": 0.22053499519824982, | |
| "eval_runtime": 2.7004, | |
| "eval_samples_per_second": 185.159, | |
| "eval_steps_per_second": 18.516, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 28.854166666666668, | |
| "eval_loss": 0.2195887714624405, | |
| "eval_runtime": 2.7033, | |
| "eval_samples_per_second": 184.958, | |
| "eval_steps_per_second": 18.496, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 28.958333333333332, | |
| "eval_loss": 0.2224828451871872, | |
| "eval_runtime": 2.7202, | |
| "eval_samples_per_second": 183.811, | |
| "eval_steps_per_second": 18.381, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 29.0625, | |
| "eval_loss": 0.22565923631191254, | |
| "eval_runtime": 2.7224, | |
| "eval_samples_per_second": 183.659, | |
| "eval_steps_per_second": 18.366, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 29.166666666666668, | |
| "grad_norm": 0.005333024077117443, | |
| "learning_rate": 1.5432098765432098e-06, | |
| "loss": 0.0001, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 29.166666666666668, | |
| "eval_loss": 0.2284005731344223, | |
| "eval_runtime": 2.7316, | |
| "eval_samples_per_second": 183.043, | |
| "eval_steps_per_second": 18.304, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 29.270833333333332, | |
| "eval_loss": 0.22724701464176178, | |
| "eval_runtime": 2.7988, | |
| "eval_samples_per_second": 178.649, | |
| "eval_steps_per_second": 17.865, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 29.375, | |
| "eval_loss": 0.23275510966777802, | |
| "eval_runtime": 2.8416, | |
| "eval_samples_per_second": 175.955, | |
| "eval_steps_per_second": 17.595, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 29.479166666666668, | |
| "eval_loss": 0.23228801786899567, | |
| "eval_runtime": 2.8546, | |
| "eval_samples_per_second": 175.155, | |
| "eval_steps_per_second": 17.516, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 29.583333333333332, | |
| "eval_loss": 0.23211157321929932, | |
| "eval_runtime": 2.8814, | |
| "eval_samples_per_second": 173.527, | |
| "eval_steps_per_second": 17.353, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 29.6875, | |
| "grad_norm": 0.001249138847924769, | |
| "learning_rate": 5.787037037037037e-07, | |
| "loss": 0.0007, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 29.6875, | |
| "eval_loss": 0.23103657364845276, | |
| "eval_runtime": 2.9118, | |
| "eval_samples_per_second": 171.717, | |
| "eval_steps_per_second": 17.172, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 29.791666666666668, | |
| "eval_loss": 0.23130209743976593, | |
| "eval_runtime": 3.0208, | |
| "eval_samples_per_second": 165.52, | |
| "eval_steps_per_second": 16.552, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 29.895833333333332, | |
| "eval_loss": 0.2313297837972641, | |
| "eval_runtime": 3.1209, | |
| "eval_samples_per_second": 160.208, | |
| "eval_steps_per_second": 16.021, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.23098842799663544, | |
| "eval_runtime": 3.0691, | |
| "eval_samples_per_second": 162.912, | |
| "eval_steps_per_second": 16.291, | |
| "step": 28800 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 28800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 1200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |