| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 12900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7751937984496124, | |
| "grad_norm": 126.4291763305664, | |
| "learning_rate": 0.0003969, | |
| "loss": 57.794, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 47.50299072265625, | |
| "eval_runtime": 18.4297, | |
| "eval_samples_per_second": 2.604, | |
| "eval_steps_per_second": 0.651, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 1.550387596899225, | |
| "grad_norm": 110.41481018066406, | |
| "learning_rate": 0.00039380000000000003, | |
| "loss": 45.635, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 42.4621696472168, | |
| "eval_runtime": 17.0743, | |
| "eval_samples_per_second": 2.811, | |
| "eval_steps_per_second": 0.703, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 2.3255813953488373, | |
| "grad_norm": 119.6634750366211, | |
| "learning_rate": 0.0003907, | |
| "loss": 43.6742, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 40.13833999633789, | |
| "eval_runtime": 18.4155, | |
| "eval_samples_per_second": 2.606, | |
| "eval_steps_per_second": 0.652, | |
| "step": 387 | |
| }, | |
| { | |
| "epoch": 3.10077519379845, | |
| "grad_norm": 118.22651672363281, | |
| "learning_rate": 0.0003876, | |
| "loss": 40.4744, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.875968992248062, | |
| "grad_norm": 134.10031127929688, | |
| "learning_rate": 0.0003845, | |
| "loss": 37.5286, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 41.196353912353516, | |
| "eval_runtime": 17.0535, | |
| "eval_samples_per_second": 2.815, | |
| "eval_steps_per_second": 0.704, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.651162790697675, | |
| "grad_norm": 80.17464447021484, | |
| "learning_rate": 0.00038140000000000005, | |
| "loss": 33.7618, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.43735122680664, | |
| "eval_runtime": 19.3245, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 0.621, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 5.426356589147287, | |
| "grad_norm": 372.6417236328125, | |
| "learning_rate": 0.00037830000000000003, | |
| "loss": 31.5899, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.824180603027344, | |
| "eval_runtime": 17.2226, | |
| "eval_samples_per_second": 2.787, | |
| "eval_steps_per_second": 0.697, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 6.2015503875969, | |
| "grad_norm": 226.50611877441406, | |
| "learning_rate": 0.0003752, | |
| "loss": 31.2097, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.976744186046512, | |
| "grad_norm": 99.72407531738281, | |
| "learning_rate": 0.0003721, | |
| "loss": 29.0727, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.32234191894531, | |
| "eval_runtime": 18.5858, | |
| "eval_samples_per_second": 2.583, | |
| "eval_steps_per_second": 0.646, | |
| "step": 903 | |
| }, | |
| { | |
| "epoch": 7.751937984496124, | |
| "grad_norm": 107.07185363769531, | |
| "learning_rate": 0.000369, | |
| "loss": 27.8483, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.962478637695312, | |
| "eval_runtime": 17.3408, | |
| "eval_samples_per_second": 2.768, | |
| "eval_steps_per_second": 0.692, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 8.527131782945737, | |
| "grad_norm": 82.22681427001953, | |
| "learning_rate": 0.0003659, | |
| "loss": 26.0904, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.708364486694336, | |
| "eval_runtime": 18.5166, | |
| "eval_samples_per_second": 2.592, | |
| "eval_steps_per_second": 0.648, | |
| "step": 1161 | |
| }, | |
| { | |
| "epoch": 9.30232558139535, | |
| "grad_norm": 184.18087768554688, | |
| "learning_rate": 0.00036280000000000004, | |
| "loss": 26.1043, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.808786392211914, | |
| "eval_runtime": 17.2835, | |
| "eval_samples_per_second": 2.777, | |
| "eval_steps_per_second": 0.694, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 10.077519379844961, | |
| "grad_norm": 82.96527862548828, | |
| "learning_rate": 0.0003597, | |
| "loss": 26.1461, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.852713178294573, | |
| "grad_norm": 76.06727600097656, | |
| "learning_rate": 0.00035660000000000005, | |
| "loss": 24.3038, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.336084365844727, | |
| "eval_runtime": 18.0862, | |
| "eval_samples_per_second": 2.654, | |
| "eval_steps_per_second": 0.663, | |
| "step": 1419 | |
| }, | |
| { | |
| "epoch": 11.627906976744185, | |
| "grad_norm": 57.900630950927734, | |
| "learning_rate": 0.00035350000000000003, | |
| "loss": 23.6493, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.202993392944336, | |
| "eval_runtime": 16.9497, | |
| "eval_samples_per_second": 2.832, | |
| "eval_steps_per_second": 0.708, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 12.4031007751938, | |
| "grad_norm": 60.565650939941406, | |
| "learning_rate": 0.0003504, | |
| "loss": 23.9146, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.08062744140625, | |
| "eval_runtime": 17.9909, | |
| "eval_samples_per_second": 2.668, | |
| "eval_steps_per_second": 0.667, | |
| "step": 1677 | |
| }, | |
| { | |
| "epoch": 13.178294573643411, | |
| "grad_norm": 75.52674865722656, | |
| "learning_rate": 0.00034730000000000004, | |
| "loss": 23.033, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.953488372093023, | |
| "grad_norm": 45.41800308227539, | |
| "learning_rate": 0.0003442, | |
| "loss": 21.9133, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.397354125976562, | |
| "eval_runtime": 17.0643, | |
| "eval_samples_per_second": 2.813, | |
| "eval_steps_per_second": 0.703, | |
| "step": 1806 | |
| }, | |
| { | |
| "epoch": 14.728682170542635, | |
| "grad_norm": 46.21736145019531, | |
| "learning_rate": 0.0003411, | |
| "loss": 22.3071, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.09249496459961, | |
| "eval_runtime": 17.5136, | |
| "eval_samples_per_second": 2.741, | |
| "eval_steps_per_second": 0.685, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 15.503875968992247, | |
| "grad_norm": 29.049968719482422, | |
| "learning_rate": 0.000338, | |
| "loss": 21.0819, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.936742782592773, | |
| "eval_runtime": 17.2024, | |
| "eval_samples_per_second": 2.79, | |
| "eval_steps_per_second": 0.698, | |
| "step": 2064 | |
| }, | |
| { | |
| "epoch": 16.27906976744186, | |
| "grad_norm": 70.16988372802734, | |
| "learning_rate": 0.0003349, | |
| "loss": 21.0089, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.042001724243164, | |
| "eval_runtime": 17.9057, | |
| "eval_samples_per_second": 2.681, | |
| "eval_steps_per_second": 0.67, | |
| "step": 2193 | |
| }, | |
| { | |
| "epoch": 17.05426356589147, | |
| "grad_norm": 77.12342834472656, | |
| "learning_rate": 0.00033180000000000004, | |
| "loss": 21.1193, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.829457364341085, | |
| "grad_norm": 40.200958251953125, | |
| "learning_rate": 0.0003287, | |
| "loss": 20.9169, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.293771743774414, | |
| "eval_runtime": 17.1083, | |
| "eval_samples_per_second": 2.806, | |
| "eval_steps_per_second": 0.701, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 18.6046511627907, | |
| "grad_norm": 31.95384979248047, | |
| "learning_rate": 0.0003256, | |
| "loss": 19.7935, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.394454956054688, | |
| "eval_runtime": 18.6297, | |
| "eval_samples_per_second": 2.577, | |
| "eval_steps_per_second": 0.644, | |
| "step": 2451 | |
| }, | |
| { | |
| "epoch": 19.37984496124031, | |
| "grad_norm": 54.4798698425293, | |
| "learning_rate": 0.00032250000000000003, | |
| "loss": 19.8749, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.845718383789062, | |
| "eval_runtime": 17.4128, | |
| "eval_samples_per_second": 2.757, | |
| "eval_steps_per_second": 0.689, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 20.155038759689923, | |
| "grad_norm": 61.0432243347168, | |
| "learning_rate": 0.0003194, | |
| "loss": 19.6959, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.930232558139537, | |
| "grad_norm": 64.08226013183594, | |
| "learning_rate": 0.0003163, | |
| "loss": 19.2973, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.071313858032227, | |
| "eval_runtime": 18.6354, | |
| "eval_samples_per_second": 2.576, | |
| "eval_steps_per_second": 0.644, | |
| "step": 2709 | |
| }, | |
| { | |
| "epoch": 21.705426356589147, | |
| "grad_norm": 54.097721099853516, | |
| "learning_rate": 0.0003132, | |
| "loss": 18.5436, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.084577560424805, | |
| "eval_runtime": 17.3593, | |
| "eval_samples_per_second": 2.765, | |
| "eval_steps_per_second": 0.691, | |
| "step": 2838 | |
| }, | |
| { | |
| "epoch": 22.48062015503876, | |
| "grad_norm": 37.09468460083008, | |
| "learning_rate": 0.00031010000000000006, | |
| "loss": 18.5996, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.88102149963379, | |
| "eval_runtime": 17.5696, | |
| "eval_samples_per_second": 2.732, | |
| "eval_steps_per_second": 0.683, | |
| "step": 2967 | |
| }, | |
| { | |
| "epoch": 23.25581395348837, | |
| "grad_norm": 40.900291442871094, | |
| "learning_rate": 0.00030700000000000004, | |
| "loss": 19.1228, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.301599502563477, | |
| "eval_runtime": 17.3752, | |
| "eval_samples_per_second": 2.763, | |
| "eval_steps_per_second": 0.691, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 24.031007751937985, | |
| "grad_norm": 16.258365631103516, | |
| "learning_rate": 0.0003039, | |
| "loss": 18.2692, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 24.8062015503876, | |
| "grad_norm": 53.048091888427734, | |
| "learning_rate": 0.0003008, | |
| "loss": 18.0519, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.71547508239746, | |
| "eval_runtime": 18.0599, | |
| "eval_samples_per_second": 2.658, | |
| "eval_steps_per_second": 0.664, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 25.58139534883721, | |
| "grad_norm": 24.09917640686035, | |
| "learning_rate": 0.0002977, | |
| "loss": 17.7073, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.716806411743164, | |
| "eval_runtime": 17.32, | |
| "eval_samples_per_second": 2.771, | |
| "eval_steps_per_second": 0.693, | |
| "step": 3354 | |
| }, | |
| { | |
| "epoch": 26.356589147286822, | |
| "grad_norm": 86.21530151367188, | |
| "learning_rate": 0.0002946, | |
| "loss": 17.5055, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.989931106567383, | |
| "eval_runtime": 17.6249, | |
| "eval_samples_per_second": 2.723, | |
| "eval_steps_per_second": 0.681, | |
| "step": 3483 | |
| }, | |
| { | |
| "epoch": 27.131782945736433, | |
| "grad_norm": 20.129249572753906, | |
| "learning_rate": 0.0002915, | |
| "loss": 18.054, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 27.906976744186046, | |
| "grad_norm": 22.863677978515625, | |
| "learning_rate": 0.0002884, | |
| "loss": 17.4854, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.19437599182129, | |
| "eval_runtime": 17.6593, | |
| "eval_samples_per_second": 2.718, | |
| "eval_steps_per_second": 0.68, | |
| "step": 3612 | |
| }, | |
| { | |
| "epoch": 28.68217054263566, | |
| "grad_norm": 13.843172073364258, | |
| "learning_rate": 0.0002853, | |
| "loss": 17.0048, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.28289031982422, | |
| "eval_runtime": 18.1574, | |
| "eval_samples_per_second": 2.644, | |
| "eval_steps_per_second": 0.661, | |
| "step": 3741 | |
| }, | |
| { | |
| "epoch": 29.45736434108527, | |
| "grad_norm": 20.727048873901367, | |
| "learning_rate": 0.0002822, | |
| "loss": 16.8731, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.1208438873291, | |
| "eval_runtime": 16.9906, | |
| "eval_samples_per_second": 2.825, | |
| "eval_steps_per_second": 0.706, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 30.232558139534884, | |
| "grad_norm": 19.721155166625977, | |
| "learning_rate": 0.0002791, | |
| "loss": 16.683, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.758291244506836, | |
| "eval_runtime": 17.7849, | |
| "eval_samples_per_second": 2.699, | |
| "eval_steps_per_second": 0.675, | |
| "step": 3999 | |
| }, | |
| { | |
| "epoch": 31.007751937984494, | |
| "grad_norm": 25.496213912963867, | |
| "learning_rate": 0.00027600000000000004, | |
| "loss": 16.9178, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 31.782945736434108, | |
| "grad_norm": 26.640628814697266, | |
| "learning_rate": 0.0002729, | |
| "loss": 16.6109, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.623199462890625, | |
| "eval_runtime": 16.843, | |
| "eval_samples_per_second": 2.85, | |
| "eval_steps_per_second": 0.712, | |
| "step": 4128 | |
| }, | |
| { | |
| "epoch": 32.55813953488372, | |
| "grad_norm": 25.067975997924805, | |
| "learning_rate": 0.0002698, | |
| "loss": 15.8261, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.416189193725586, | |
| "eval_runtime": 17.6752, | |
| "eval_samples_per_second": 2.716, | |
| "eval_steps_per_second": 0.679, | |
| "step": 4257 | |
| }, | |
| { | |
| "epoch": 33.333333333333336, | |
| "grad_norm": 30.74283218383789, | |
| "learning_rate": 0.00026670000000000003, | |
| "loss": 16.9002, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.438751220703125, | |
| "eval_runtime": 16.8785, | |
| "eval_samples_per_second": 2.844, | |
| "eval_steps_per_second": 0.711, | |
| "step": 4386 | |
| }, | |
| { | |
| "epoch": 34.10852713178294, | |
| "grad_norm": 72.6050033569336, | |
| "learning_rate": 0.0002636, | |
| "loss": 15.7742, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 34.883720930232556, | |
| "grad_norm": 93.86290740966797, | |
| "learning_rate": 0.00026050000000000004, | |
| "loss": 16.3081, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.97564697265625, | |
| "eval_runtime": 17.4518, | |
| "eval_samples_per_second": 2.75, | |
| "eval_steps_per_second": 0.688, | |
| "step": 4515 | |
| }, | |
| { | |
| "epoch": 35.65891472868217, | |
| "grad_norm": 24.93766975402832, | |
| "learning_rate": 0.0002574, | |
| "loss": 15.4745, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.821380615234375, | |
| "eval_runtime": 16.9764, | |
| "eval_samples_per_second": 2.827, | |
| "eval_steps_per_second": 0.707, | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 36.434108527131784, | |
| "grad_norm": 33.70745849609375, | |
| "learning_rate": 0.0002543, | |
| "loss": 15.938, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.100107192993164, | |
| "eval_runtime": 17.5981, | |
| "eval_samples_per_second": 2.728, | |
| "eval_steps_per_second": 0.682, | |
| "step": 4773 | |
| }, | |
| { | |
| "epoch": 37.2093023255814, | |
| "grad_norm": 59.88523864746094, | |
| "learning_rate": 0.00025120000000000003, | |
| "loss": 14.9862, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 37.98449612403101, | |
| "grad_norm": 20.979228973388672, | |
| "learning_rate": 0.0002481, | |
| "loss": 15.9947, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.053319931030273, | |
| "eval_runtime": 17.0472, | |
| "eval_samples_per_second": 2.816, | |
| "eval_steps_per_second": 0.704, | |
| "step": 4902 | |
| }, | |
| { | |
| "epoch": 38.75968992248062, | |
| "grad_norm": 17.90158462524414, | |
| "learning_rate": 0.000245, | |
| "loss": 15.2328, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.62113380432129, | |
| "eval_runtime": 17.318, | |
| "eval_samples_per_second": 2.772, | |
| "eval_steps_per_second": 0.693, | |
| "step": 5031 | |
| }, | |
| { | |
| "epoch": 39.53488372093023, | |
| "grad_norm": 33.11941909790039, | |
| "learning_rate": 0.00024190000000000003, | |
| "loss": 15.202, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.138301849365234, | |
| "eval_runtime": 17.0128, | |
| "eval_samples_per_second": 2.821, | |
| "eval_steps_per_second": 0.705, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 40.310077519379846, | |
| "grad_norm": 15.685113906860352, | |
| "learning_rate": 0.0002388, | |
| "loss": 15.0583, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.408859252929688, | |
| "eval_runtime": 17.9066, | |
| "eval_samples_per_second": 2.681, | |
| "eval_steps_per_second": 0.67, | |
| "step": 5289 | |
| }, | |
| { | |
| "epoch": 41.08527131782946, | |
| "grad_norm": 20.353235244750977, | |
| "learning_rate": 0.00023569999999999998, | |
| "loss": 14.7257, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 41.86046511627907, | |
| "grad_norm": 22.713470458984375, | |
| "learning_rate": 0.00023259999999999996, | |
| "loss": 14.573, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.568130493164062, | |
| "eval_runtime": 17.7042, | |
| "eval_samples_per_second": 2.711, | |
| "eval_steps_per_second": 0.678, | |
| "step": 5418 | |
| }, | |
| { | |
| "epoch": 42.63565891472868, | |
| "grad_norm": 24.60871696472168, | |
| "learning_rate": 0.0002295, | |
| "loss": 14.7401, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.554765701293945, | |
| "eval_runtime": 18.0539, | |
| "eval_samples_per_second": 2.659, | |
| "eval_steps_per_second": 0.665, | |
| "step": 5547 | |
| }, | |
| { | |
| "epoch": 43.41085271317829, | |
| "grad_norm": 36.38352966308594, | |
| "learning_rate": 0.0002264, | |
| "loss": 14.6052, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.39527702331543, | |
| "eval_runtime": 17.3086, | |
| "eval_samples_per_second": 2.773, | |
| "eval_steps_per_second": 0.693, | |
| "step": 5676 | |
| }, | |
| { | |
| "epoch": 44.18604651162791, | |
| "grad_norm": 14.869057655334473, | |
| "learning_rate": 0.00022330000000000003, | |
| "loss": 13.9636, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 44.96124031007752, | |
| "grad_norm": 12.379744529724121, | |
| "learning_rate": 0.0002202, | |
| "loss": 14.1299, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.81528663635254, | |
| "eval_runtime": 17.5777, | |
| "eval_samples_per_second": 2.731, | |
| "eval_steps_per_second": 0.683, | |
| "step": 5805 | |
| }, | |
| { | |
| "epoch": 45.736434108527135, | |
| "grad_norm": 51.40928649902344, | |
| "learning_rate": 0.00021710000000000005, | |
| "loss": 13.6851, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.969324111938477, | |
| "eval_runtime": 17.5744, | |
| "eval_samples_per_second": 2.731, | |
| "eval_steps_per_second": 0.683, | |
| "step": 5934 | |
| }, | |
| { | |
| "epoch": 46.51162790697674, | |
| "grad_norm": 13.089680671691895, | |
| "learning_rate": 0.00021400000000000002, | |
| "loss": 14.6677, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.936065673828125, | |
| "eval_runtime": 18.038, | |
| "eval_samples_per_second": 2.661, | |
| "eval_steps_per_second": 0.665, | |
| "step": 6063 | |
| }, | |
| { | |
| "epoch": 47.286821705426355, | |
| "grad_norm": 15.198484420776367, | |
| "learning_rate": 0.0002109, | |
| "loss": 13.6493, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.3327751159668, | |
| "eval_runtime": 17.0231, | |
| "eval_samples_per_second": 2.82, | |
| "eval_steps_per_second": 0.705, | |
| "step": 6192 | |
| }, | |
| { | |
| "epoch": 48.06201550387597, | |
| "grad_norm": 17.197790145874023, | |
| "learning_rate": 0.00020780000000000004, | |
| "loss": 13.7191, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 48.83720930232558, | |
| "grad_norm": 22.198528289794922, | |
| "learning_rate": 0.00020470000000000002, | |
| "loss": 14.166, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.62310791015625, | |
| "eval_runtime": 17.8418, | |
| "eval_samples_per_second": 2.69, | |
| "eval_steps_per_second": 0.673, | |
| "step": 6321 | |
| }, | |
| { | |
| "epoch": 49.6124031007752, | |
| "grad_norm": 12.192609786987305, | |
| "learning_rate": 0.00020160000000000002, | |
| "loss": 13.7388, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.17361831665039, | |
| "eval_runtime": 16.9796, | |
| "eval_samples_per_second": 2.827, | |
| "eval_steps_per_second": 0.707, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 50.3875968992248, | |
| "grad_norm": 24.24190330505371, | |
| "learning_rate": 0.0001985, | |
| "loss": 13.0849, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.95216369628906, | |
| "eval_runtime": 17.6852, | |
| "eval_samples_per_second": 2.714, | |
| "eval_steps_per_second": 0.679, | |
| "step": 6579 | |
| }, | |
| { | |
| "epoch": 51.16279069767442, | |
| "grad_norm": 10.653921127319336, | |
| "learning_rate": 0.0001954, | |
| "loss": 13.7478, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 51.93798449612403, | |
| "grad_norm": 12.344590187072754, | |
| "learning_rate": 0.00019229999999999999, | |
| "loss": 13.2502, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.79899215698242, | |
| "eval_runtime": 17.0885, | |
| "eval_samples_per_second": 2.809, | |
| "eval_steps_per_second": 0.702, | |
| "step": 6708 | |
| }, | |
| { | |
| "epoch": 52.713178294573645, | |
| "grad_norm": 11.102241516113281, | |
| "learning_rate": 0.0001892, | |
| "loss": 13.5116, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.57374382019043, | |
| "eval_runtime": 17.6271, | |
| "eval_samples_per_second": 2.723, | |
| "eval_steps_per_second": 0.681, | |
| "step": 6837 | |
| }, | |
| { | |
| "epoch": 53.48837209302326, | |
| "grad_norm": 10.652983665466309, | |
| "learning_rate": 0.00018610000000000002, | |
| "loss": 12.6993, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.26504898071289, | |
| "eval_runtime": 17.4525, | |
| "eval_samples_per_second": 2.75, | |
| "eval_steps_per_second": 0.688, | |
| "step": 6966 | |
| }, | |
| { | |
| "epoch": 54.263565891472865, | |
| "grad_norm": 12.211697578430176, | |
| "learning_rate": 0.00018300000000000003, | |
| "loss": 13.3602, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.891380310058594, | |
| "eval_runtime": 18.9822, | |
| "eval_samples_per_second": 2.529, | |
| "eval_steps_per_second": 0.632, | |
| "step": 7095 | |
| }, | |
| { | |
| "epoch": 55.03875968992248, | |
| "grad_norm": 14.056374549865723, | |
| "learning_rate": 0.0001799, | |
| "loss": 12.9955, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 55.81395348837209, | |
| "grad_norm": 10.69999885559082, | |
| "learning_rate": 0.00017680000000000001, | |
| "loss": 12.9585, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.98616409301758, | |
| "eval_runtime": 17.2599, | |
| "eval_samples_per_second": 2.781, | |
| "eval_steps_per_second": 0.695, | |
| "step": 7224 | |
| }, | |
| { | |
| "epoch": 56.58914728682171, | |
| "grad_norm": 7.194685459136963, | |
| "learning_rate": 0.00017370000000000002, | |
| "loss": 12.7434, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.91057205200195, | |
| "eval_runtime": 18.4913, | |
| "eval_samples_per_second": 2.596, | |
| "eval_steps_per_second": 0.649, | |
| "step": 7353 | |
| }, | |
| { | |
| "epoch": 57.36434108527132, | |
| "grad_norm": 9.769908905029297, | |
| "learning_rate": 0.0001706, | |
| "loss": 12.7299, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.010562896728516, | |
| "eval_runtime": 17.0454, | |
| "eval_samples_per_second": 2.816, | |
| "eval_steps_per_second": 0.704, | |
| "step": 7482 | |
| }, | |
| { | |
| "epoch": 58.13953488372093, | |
| "grad_norm": 18.091665267944336, | |
| "learning_rate": 0.0001675, | |
| "loss": 12.3929, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 58.91472868217054, | |
| "grad_norm": 8.427603721618652, | |
| "learning_rate": 0.00016439999999999998, | |
| "loss": 12.717, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.35882568359375, | |
| "eval_runtime": 18.0781, | |
| "eval_samples_per_second": 2.655, | |
| "eval_steps_per_second": 0.664, | |
| "step": 7611 | |
| }, | |
| { | |
| "epoch": 59.689922480620154, | |
| "grad_norm": 15.918642044067383, | |
| "learning_rate": 0.00016130000000000002, | |
| "loss": 12.0563, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.09232711791992, | |
| "eval_runtime": 16.9066, | |
| "eval_samples_per_second": 2.839, | |
| "eval_steps_per_second": 0.71, | |
| "step": 7740 | |
| }, | |
| { | |
| "epoch": 60.46511627906977, | |
| "grad_norm": 13.870895385742188, | |
| "learning_rate": 0.00015820000000000002, | |
| "loss": 13.012, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.73225402832031, | |
| "eval_runtime": 17.9846, | |
| "eval_samples_per_second": 2.669, | |
| "eval_steps_per_second": 0.667, | |
| "step": 7869 | |
| }, | |
| { | |
| "epoch": 61.24031007751938, | |
| "grad_norm": 7.798965930938721, | |
| "learning_rate": 0.00015510000000000003, | |
| "loss": 12.2878, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.9967155456543, | |
| "eval_runtime": 17.0439, | |
| "eval_samples_per_second": 2.816, | |
| "eval_steps_per_second": 0.704, | |
| "step": 7998 | |
| }, | |
| { | |
| "epoch": 62.01550387596899, | |
| "grad_norm": 8.46688461303711, | |
| "learning_rate": 0.000152, | |
| "loss": 12.3515, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 62.7906976744186, | |
| "grad_norm": 9.745466232299805, | |
| "learning_rate": 0.00014890000000000001, | |
| "loss": 12.2794, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 37.55772399902344, | |
| "eval_runtime": 18.0451, | |
| "eval_samples_per_second": 2.66, | |
| "eval_steps_per_second": 0.665, | |
| "step": 8127 | |
| }, | |
| { | |
| "epoch": 63.565891472868216, | |
| "grad_norm": 7.328401565551758, | |
| "learning_rate": 0.0001458, | |
| "loss": 12.4147, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 37.27333068847656, | |
| "eval_runtime": 19.2621, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 0.623, | |
| "step": 8256 | |
| }, | |
| { | |
| "epoch": 64.34108527131782, | |
| "grad_norm": 12.89833927154541, | |
| "learning_rate": 0.0001427, | |
| "loss": 12.0032, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.3015022277832, | |
| "eval_runtime": 17.8838, | |
| "eval_samples_per_second": 2.684, | |
| "eval_steps_per_second": 0.671, | |
| "step": 8385 | |
| }, | |
| { | |
| "epoch": 65.11627906976744, | |
| "grad_norm": 15.308392524719238, | |
| "learning_rate": 0.00013959999999999998, | |
| "loss": 11.7392, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 65.89147286821705, | |
| "grad_norm": 12.101038932800293, | |
| "learning_rate": 0.00013650000000000004, | |
| "loss": 12.2793, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.280582427978516, | |
| "eval_runtime": 17.6628, | |
| "eval_samples_per_second": 2.718, | |
| "eval_steps_per_second": 0.679, | |
| "step": 8514 | |
| }, | |
| { | |
| "epoch": 66.66666666666667, | |
| "grad_norm": 12.754354476928711, | |
| "learning_rate": 0.00013340000000000002, | |
| "loss": 12.2309, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.24875259399414, | |
| "eval_runtime": 17.2522, | |
| "eval_samples_per_second": 2.782, | |
| "eval_steps_per_second": 0.696, | |
| "step": 8643 | |
| }, | |
| { | |
| "epoch": 67.44186046511628, | |
| "grad_norm": 9.756113052368164, | |
| "learning_rate": 0.00013030000000000002, | |
| "loss": 11.7082, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.66865158081055, | |
| "eval_runtime": 18.2695, | |
| "eval_samples_per_second": 2.627, | |
| "eval_steps_per_second": 0.657, | |
| "step": 8772 | |
| }, | |
| { | |
| "epoch": 68.21705426356588, | |
| "grad_norm": 9.372435569763184, | |
| "learning_rate": 0.0001272, | |
| "loss": 11.5136, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 68.9922480620155, | |
| "grad_norm": 15.044282913208008, | |
| "learning_rate": 0.0001241, | |
| "loss": 11.8694, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.04698944091797, | |
| "eval_runtime": 17.3888, | |
| "eval_samples_per_second": 2.76, | |
| "eval_steps_per_second": 0.69, | |
| "step": 8901 | |
| }, | |
| { | |
| "epoch": 69.76744186046511, | |
| "grad_norm": 9.250027656555176, | |
| "learning_rate": 0.000121, | |
| "loss": 11.782, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.40549087524414, | |
| "eval_runtime": 18.0194, | |
| "eval_samples_per_second": 2.664, | |
| "eval_steps_per_second": 0.666, | |
| "step": 9030 | |
| }, | |
| { | |
| "epoch": 70.54263565891473, | |
| "grad_norm": 16.64485740661621, | |
| "learning_rate": 0.00011789999999999999, | |
| "loss": 11.6254, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.70663070678711, | |
| "eval_runtime": 17.3613, | |
| "eval_samples_per_second": 2.765, | |
| "eval_steps_per_second": 0.691, | |
| "step": 9159 | |
| }, | |
| { | |
| "epoch": 71.31782945736434, | |
| "grad_norm": 15.693510055541992, | |
| "learning_rate": 0.00011479999999999997, | |
| "loss": 11.5873, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.10844421386719, | |
| "eval_runtime": 17.5839, | |
| "eval_samples_per_second": 2.73, | |
| "eval_steps_per_second": 0.682, | |
| "step": 9288 | |
| }, | |
| { | |
| "epoch": 72.09302325581395, | |
| "grad_norm": 7.485771179199219, | |
| "learning_rate": 0.00011170000000000003, | |
| "loss": 11.6159, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 72.86821705426357, | |
| "grad_norm": 15.41925048828125, | |
| "learning_rate": 0.00010860000000000004, | |
| "loss": 11.6251, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.29316329956055, | |
| "eval_runtime": 17.0634, | |
| "eval_samples_per_second": 2.813, | |
| "eval_steps_per_second": 0.703, | |
| "step": 9417 | |
| }, | |
| { | |
| "epoch": 73.64341085271317, | |
| "grad_norm": 16.74988555908203, | |
| "learning_rate": 0.00010550000000000002, | |
| "loss": 11.4589, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.55695724487305, | |
| "eval_runtime": 17.9041, | |
| "eval_samples_per_second": 2.681, | |
| "eval_steps_per_second": 0.67, | |
| "step": 9546 | |
| }, | |
| { | |
| "epoch": 74.4186046511628, | |
| "grad_norm": 146.4043426513672, | |
| "learning_rate": 0.00010240000000000001, | |
| "loss": 11.7378, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.988651275634766, | |
| "eval_runtime": 17.0167, | |
| "eval_samples_per_second": 2.821, | |
| "eval_steps_per_second": 0.705, | |
| "step": 9675 | |
| }, | |
| { | |
| "epoch": 75.1937984496124, | |
| "grad_norm": 10.800848960876465, | |
| "learning_rate": 9.93e-05, | |
| "loss": 11.4043, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 75.96899224806202, | |
| "grad_norm": 9.41010570526123, | |
| "learning_rate": 9.62e-05, | |
| "loss": 11.4933, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.47134017944336, | |
| "eval_runtime": 17.8569, | |
| "eval_samples_per_second": 2.688, | |
| "eval_steps_per_second": 0.672, | |
| "step": 9804 | |
| }, | |
| { | |
| "epoch": 76.74418604651163, | |
| "grad_norm": 19.61960220336914, | |
| "learning_rate": 9.31e-05, | |
| "loss": 11.2566, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.96221923828125, | |
| "eval_runtime": 17.0325, | |
| "eval_samples_per_second": 2.818, | |
| "eval_steps_per_second": 0.705, | |
| "step": 9933 | |
| }, | |
| { | |
| "epoch": 77.51937984496124, | |
| "grad_norm": 9.528326034545898, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 11.25, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 37.10159683227539, | |
| "eval_runtime": 18.225, | |
| "eval_samples_per_second": 2.634, | |
| "eval_steps_per_second": 0.658, | |
| "step": 10062 | |
| }, | |
| { | |
| "epoch": 78.29457364341086, | |
| "grad_norm": 8.064875602722168, | |
| "learning_rate": 8.690000000000003e-05, | |
| "loss": 11.2962, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 37.87105178833008, | |
| "eval_runtime": 16.9948, | |
| "eval_samples_per_second": 2.824, | |
| "eval_steps_per_second": 0.706, | |
| "step": 10191 | |
| }, | |
| { | |
| "epoch": 79.06976744186046, | |
| "grad_norm": 10.451505661010742, | |
| "learning_rate": 8.380000000000002e-05, | |
| "loss": 11.1642, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 79.84496124031008, | |
| "grad_norm": 7.318461894989014, | |
| "learning_rate": 8.070000000000001e-05, | |
| "loss": 11.0868, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.571414947509766, | |
| "eval_runtime": 18.0923, | |
| "eval_samples_per_second": 2.653, | |
| "eval_steps_per_second": 0.663, | |
| "step": 10320 | |
| }, | |
| { | |
| "epoch": 80.62015503875969, | |
| "grad_norm": 10.301888465881348, | |
| "learning_rate": 7.760000000000002e-05, | |
| "loss": 11.2786, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.1493034362793, | |
| "eval_runtime": 17.0167, | |
| "eval_samples_per_second": 2.821, | |
| "eval_steps_per_second": 0.705, | |
| "step": 10449 | |
| }, | |
| { | |
| "epoch": 81.3953488372093, | |
| "grad_norm": 8.667201042175293, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 11.1528, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.0099983215332, | |
| "eval_runtime": 17.9494, | |
| "eval_samples_per_second": 2.674, | |
| "eval_steps_per_second": 0.669, | |
| "step": 10578 | |
| }, | |
| { | |
| "epoch": 82.17054263565892, | |
| "grad_norm": 5.117663860321045, | |
| "learning_rate": 7.14e-05, | |
| "loss": 10.9299, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 82.94573643410853, | |
| "grad_norm": 5.9621806144714355, | |
| "learning_rate": 6.829999999999999e-05, | |
| "loss": 11.089, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.5473518371582, | |
| "eval_runtime": 17.0039, | |
| "eval_samples_per_second": 2.823, | |
| "eval_steps_per_second": 0.706, | |
| "step": 10707 | |
| }, | |
| { | |
| "epoch": 83.72093023255815, | |
| "grad_norm": 6.17501974105835, | |
| "learning_rate": 6.519999999999999e-05, | |
| "loss": 10.954, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.940486907958984, | |
| "eval_runtime": 17.721, | |
| "eval_samples_per_second": 2.709, | |
| "eval_steps_per_second": 0.677, | |
| "step": 10836 | |
| }, | |
| { | |
| "epoch": 84.49612403100775, | |
| "grad_norm": 31.69377326965332, | |
| "learning_rate": 6.210000000000003e-05, | |
| "loss": 11.0157, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.3872184753418, | |
| "eval_runtime": 16.9062, | |
| "eval_samples_per_second": 2.839, | |
| "eval_steps_per_second": 0.71, | |
| "step": 10965 | |
| }, | |
| { | |
| "epoch": 85.27131782945736, | |
| "grad_norm": 7.1002984046936035, | |
| "learning_rate": 5.9000000000000025e-05, | |
| "loss": 10.9849, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.4875373840332, | |
| "eval_runtime": 17.7347, | |
| "eval_samples_per_second": 2.707, | |
| "eval_steps_per_second": 0.677, | |
| "step": 11094 | |
| }, | |
| { | |
| "epoch": 86.04651162790698, | |
| "grad_norm": 13.370129585266113, | |
| "learning_rate": 5.590000000000002e-05, | |
| "loss": 11.0614, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 86.82170542635659, | |
| "grad_norm": 5.192051887512207, | |
| "learning_rate": 5.28e-05, | |
| "loss": 10.5423, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.11787796020508, | |
| "eval_runtime": 16.9675, | |
| "eval_samples_per_second": 2.829, | |
| "eval_steps_per_second": 0.707, | |
| "step": 11223 | |
| }, | |
| { | |
| "epoch": 87.59689922480621, | |
| "grad_norm": 5.747579097747803, | |
| "learning_rate": 4.97e-05, | |
| "loss": 11.1968, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.4084358215332, | |
| "eval_runtime": 17.9374, | |
| "eval_samples_per_second": 2.676, | |
| "eval_steps_per_second": 0.669, | |
| "step": 11352 | |
| }, | |
| { | |
| "epoch": 88.37209302325581, | |
| "grad_norm": 11.57238483428955, | |
| "learning_rate": 4.66e-05, | |
| "loss": 10.6376, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.82176971435547, | |
| "eval_runtime": 16.9422, | |
| "eval_samples_per_second": 2.833, | |
| "eval_steps_per_second": 0.708, | |
| "step": 11481 | |
| }, | |
| { | |
| "epoch": 89.14728682170542, | |
| "grad_norm": 8.44890308380127, | |
| "learning_rate": 4.3499999999999993e-05, | |
| "loss": 10.8035, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 89.92248062015504, | |
| "grad_norm": 7.732810974121094, | |
| "learning_rate": 4.0399999999999986e-05, | |
| "loss": 10.7131, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.25526428222656, | |
| "eval_runtime": 18.1455, | |
| "eval_samples_per_second": 2.645, | |
| "eval_steps_per_second": 0.661, | |
| "step": 11610 | |
| }, | |
| { | |
| "epoch": 90.69767441860465, | |
| "grad_norm": 6.144818305969238, | |
| "learning_rate": 3.7300000000000026e-05, | |
| "loss": 10.8252, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.136837005615234, | |
| "eval_runtime": 16.9741, | |
| "eval_samples_per_second": 2.828, | |
| "eval_steps_per_second": 0.707, | |
| "step": 11739 | |
| }, | |
| { | |
| "epoch": 91.47286821705427, | |
| "grad_norm": 4.7243475914001465, | |
| "learning_rate": 3.420000000000002e-05, | |
| "loss": 10.6456, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 38.91936111450195, | |
| "eval_runtime": 17.9915, | |
| "eval_samples_per_second": 2.668, | |
| "eval_steps_per_second": 0.667, | |
| "step": 11868 | |
| }, | |
| { | |
| "epoch": 92.24806201550388, | |
| "grad_norm": 13.374404907226562, | |
| "learning_rate": 3.110000000000002e-05, | |
| "loss": 10.8488, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.595462799072266, | |
| "eval_runtime": 16.9478, | |
| "eval_samples_per_second": 2.832, | |
| "eval_steps_per_second": 0.708, | |
| "step": 11997 | |
| }, | |
| { | |
| "epoch": 93.02325581395348, | |
| "grad_norm": 9.406952857971191, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 10.5219, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 93.7984496124031, | |
| "grad_norm": 5.360720634460449, | |
| "learning_rate": 2.49e-05, | |
| "loss": 10.8675, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.47597885131836, | |
| "eval_runtime": 17.7005, | |
| "eval_samples_per_second": 2.712, | |
| "eval_steps_per_second": 0.678, | |
| "step": 12126 | |
| }, | |
| { | |
| "epoch": 94.57364341085271, | |
| "grad_norm": 3.855013132095337, | |
| "learning_rate": 2.1799999999999995e-05, | |
| "loss": 10.4757, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 40.484397888183594, | |
| "eval_runtime": 16.9799, | |
| "eval_samples_per_second": 2.827, | |
| "eval_steps_per_second": 0.707, | |
| "step": 12255 | |
| }, | |
| { | |
| "epoch": 95.34883720930233, | |
| "grad_norm": 7.718498229980469, | |
| "learning_rate": 1.8699999999999987e-05, | |
| "loss": 10.3191, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.06733322143555, | |
| "eval_runtime": 17.7268, | |
| "eval_samples_per_second": 2.708, | |
| "eval_steps_per_second": 0.677, | |
| "step": 12384 | |
| }, | |
| { | |
| "epoch": 96.12403100775194, | |
| "grad_norm": 3.9156548976898193, | |
| "learning_rate": 1.5599999999999983e-05, | |
| "loss": 10.6169, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 96.89922480620154, | |
| "grad_norm": 3.464470148086548, | |
| "learning_rate": 1.2500000000000023e-05, | |
| "loss": 10.6073, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.37672805786133, | |
| "eval_runtime": 16.9277, | |
| "eval_samples_per_second": 2.836, | |
| "eval_steps_per_second": 0.709, | |
| "step": 12513 | |
| }, | |
| { | |
| "epoch": 97.67441860465117, | |
| "grad_norm": 5.994311332702637, | |
| "learning_rate": 9.400000000000018e-06, | |
| "loss": 10.3038, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.69685745239258, | |
| "eval_runtime": 18.0251, | |
| "eval_samples_per_second": 2.663, | |
| "eval_steps_per_second": 0.666, | |
| "step": 12642 | |
| }, | |
| { | |
| "epoch": 98.44961240310077, | |
| "grad_norm": 4.125715732574463, | |
| "learning_rate": 6.300000000000012e-06, | |
| "loss": 11.0709, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.93254470825195, | |
| "eval_runtime": 17.9269, | |
| "eval_samples_per_second": 2.678, | |
| "eval_steps_per_second": 0.669, | |
| "step": 12771 | |
| }, | |
| { | |
| "epoch": 99.2248062015504, | |
| "grad_norm": 4.381137371063232, | |
| "learning_rate": 3.200000000000005e-06, | |
| "loss": 10.2398, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 4.022866249084473, | |
| "learning_rate": 9.999999999999998e-08, | |
| "loss": 10.5951, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 39.87546157836914, | |
| "eval_runtime": 17.8005, | |
| "eval_samples_per_second": 2.697, | |
| "eval_steps_per_second": 0.674, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 12900, | |
| "total_flos": 1.4631500418239693e+19, | |
| "train_loss": 16.309644344832545, | |
| "train_runtime": 27199.2721, | |
| "train_samples_per_second": 1.897, | |
| "train_steps_per_second": 0.474 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4631500418239693e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |