| { |
| "best_global_step": 200, |
| "best_metric": 0.0016466762172058225, |
| "best_model_checkpoint": "unified_sft_output/checkpoint-200", |
| "epoch": 50.0, |
| "eval_steps": 10, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.27586206896551724, |
| "grad_norm": 11.375, |
| "learning_rate": 0.0, |
| "loss": 1.2371, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.5517241379310345, |
| "grad_norm": 10.75, |
| "learning_rate": 5.000000000000001e-07, |
| "loss": 1.249, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.8275862068965517, |
| "grad_norm": 12.375, |
| "learning_rate": 1.0000000000000002e-06, |
| "loss": 1.4344, |
| "step": 3 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 12.25, |
| "learning_rate": 1.5e-06, |
| "loss": 1.2445, |
| "step": 4 |
| }, |
| { |
| "epoch": 1.2758620689655173, |
| "grad_norm": 10.1875, |
| "learning_rate": 2.0000000000000003e-06, |
| "loss": 1.1814, |
| "step": 5 |
| }, |
| { |
| "epoch": 1.5517241379310345, |
| "grad_norm": 13.25, |
| "learning_rate": 2.5e-06, |
| "loss": 1.2665, |
| "step": 6 |
| }, |
| { |
| "epoch": 1.8275862068965516, |
| "grad_norm": 21.375, |
| "learning_rate": 3e-06, |
| "loss": 1.3003, |
| "step": 7 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 12.0625, |
| "learning_rate": 3.5e-06, |
| "loss": 1.4292, |
| "step": 8 |
| }, |
| { |
| "epoch": 2.2758620689655173, |
| "grad_norm": 10.8125, |
| "learning_rate": 4.000000000000001e-06, |
| "loss": 1.1558, |
| "step": 9 |
| }, |
| { |
| "epoch": 2.5517241379310347, |
| "grad_norm": 8.375, |
| "learning_rate": 4.5e-06, |
| "loss": 1.2145, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.5517241379310347, |
| "eval_loss": 1.21233069896698, |
| "eval_runtime": 1.7228, |
| "eval_samples_per_second": 1.741, |
| "eval_steps_per_second": 1.741, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.8275862068965516, |
| "grad_norm": 8.8125, |
| "learning_rate": 5e-06, |
| "loss": 1.2393, |
| "step": 11 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 6.875, |
| "learning_rate": 5.500000000000001e-06, |
| "loss": 1.1221, |
| "step": 12 |
| }, |
| { |
| "epoch": 3.2758620689655173, |
| "grad_norm": 5.5625, |
| "learning_rate": 6e-06, |
| "loss": 1.0858, |
| "step": 13 |
| }, |
| { |
| "epoch": 3.5517241379310347, |
| "grad_norm": 5.59375, |
| "learning_rate": 6.5000000000000004e-06, |
| "loss": 1.1747, |
| "step": 14 |
| }, |
| { |
| "epoch": 3.8275862068965516, |
| "grad_norm": 3.625, |
| "learning_rate": 7e-06, |
| "loss": 0.9076, |
| "step": 15 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 4.53125, |
| "learning_rate": 7.500000000000001e-06, |
| "loss": 1.1572, |
| "step": 16 |
| }, |
| { |
| "epoch": 4.275862068965517, |
| "grad_norm": 3.453125, |
| "learning_rate": 8.000000000000001e-06, |
| "loss": 1.0446, |
| "step": 17 |
| }, |
| { |
| "epoch": 4.551724137931035, |
| "grad_norm": 3.078125, |
| "learning_rate": 8.5e-06, |
| "loss": 1.011, |
| "step": 18 |
| }, |
| { |
| "epoch": 4.827586206896552, |
| "grad_norm": 2.375, |
| "learning_rate": 9e-06, |
| "loss": 0.9294, |
| "step": 19 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.9453125, |
| "learning_rate": 9.5e-06, |
| "loss": 0.7185, |
| "step": 20 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.8018922209739685, |
| "eval_runtime": 1.7241, |
| "eval_samples_per_second": 1.74, |
| "eval_steps_per_second": 1.74, |
| "step": 20 |
| }, |
| { |
| "epoch": 5.275862068965517, |
| "grad_norm": 2.453125, |
| "learning_rate": 1e-05, |
| "loss": 0.8795, |
| "step": 21 |
| }, |
| { |
| "epoch": 5.551724137931035, |
| "grad_norm": 1.90625, |
| "learning_rate": 1.0500000000000001e-05, |
| "loss": 0.8248, |
| "step": 22 |
| }, |
| { |
| "epoch": 5.827586206896552, |
| "grad_norm": 1.7421875, |
| "learning_rate": 1.1000000000000001e-05, |
| "loss": 0.6881, |
| "step": 23 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 1.671875, |
| "learning_rate": 1.15e-05, |
| "loss": 0.7153, |
| "step": 24 |
| }, |
| { |
| "epoch": 6.275862068965517, |
| "grad_norm": 3.125, |
| "learning_rate": 1.2e-05, |
| "loss": 0.6096, |
| "step": 25 |
| }, |
| { |
| "epoch": 6.551724137931035, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.25e-05, |
| "loss": 0.6779, |
| "step": 26 |
| }, |
| { |
| "epoch": 6.827586206896552, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.3000000000000001e-05, |
| "loss": 0.6229, |
| "step": 27 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.3500000000000001e-05, |
| "loss": 0.5901, |
| "step": 28 |
| }, |
| { |
| "epoch": 7.275862068965517, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.4e-05, |
| "loss": 0.593, |
| "step": 29 |
| }, |
| { |
| "epoch": 7.551724137931035, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.45e-05, |
| "loss": 0.5158, |
| "step": 30 |
| }, |
| { |
| "epoch": 7.551724137931035, |
| "eval_loss": 0.4224577248096466, |
| "eval_runtime": 1.7416, |
| "eval_samples_per_second": 1.723, |
| "eval_steps_per_second": 1.723, |
| "step": 30 |
| }, |
| { |
| "epoch": 7.827586206896552, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.5000000000000002e-05, |
| "loss": 0.4175, |
| "step": 31 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.96484375, |
| "learning_rate": 1.55e-05, |
| "loss": 0.478, |
| "step": 32 |
| }, |
| { |
| "epoch": 8.275862068965518, |
| "grad_norm": 0.79296875, |
| "learning_rate": 1.6000000000000003e-05, |
| "loss": 0.4238, |
| "step": 33 |
| }, |
| { |
| "epoch": 8.551724137931034, |
| "grad_norm": 1.0703125, |
| "learning_rate": 1.65e-05, |
| "loss": 0.4365, |
| "step": 34 |
| }, |
| { |
| "epoch": 8.827586206896552, |
| "grad_norm": 0.9453125, |
| "learning_rate": 1.7e-05, |
| "loss": 0.3686, |
| "step": 35 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 2.328125, |
| "learning_rate": 1.7500000000000002e-05, |
| "loss": 0.4172, |
| "step": 36 |
| }, |
| { |
| "epoch": 9.275862068965518, |
| "grad_norm": 0.73046875, |
| "learning_rate": 1.8e-05, |
| "loss": 0.3731, |
| "step": 37 |
| }, |
| { |
| "epoch": 9.551724137931034, |
| "grad_norm": 0.953125, |
| "learning_rate": 1.8500000000000002e-05, |
| "loss": 0.3885, |
| "step": 38 |
| }, |
| { |
| "epoch": 9.827586206896552, |
| "grad_norm": 0.79296875, |
| "learning_rate": 1.9e-05, |
| "loss": 0.3008, |
| "step": 39 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.95e-05, |
| "loss": 0.2395, |
| "step": 40 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.21218304336071014, |
| "eval_runtime": 1.7107, |
| "eval_samples_per_second": 1.754, |
| "eval_steps_per_second": 1.754, |
| "step": 40 |
| }, |
| { |
| "epoch": 10.275862068965518, |
| "grad_norm": 0.91796875, |
| "learning_rate": 2e-05, |
| "loss": 0.2379, |
| "step": 41 |
| }, |
| { |
| "epoch": 10.551724137931034, |
| "grad_norm": 0.859375, |
| "learning_rate": 1.9999619230641714e-05, |
| "loss": 0.267, |
| "step": 42 |
| }, |
| { |
| "epoch": 10.827586206896552, |
| "grad_norm": 0.953125, |
| "learning_rate": 1.9998476951563914e-05, |
| "loss": 0.2913, |
| "step": 43 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.9996573249755573e-05, |
| "loss": 0.2742, |
| "step": 44 |
| }, |
| { |
| "epoch": 11.275862068965518, |
| "grad_norm": 0.875, |
| "learning_rate": 1.999390827019096e-05, |
| "loss": 0.1832, |
| "step": 45 |
| }, |
| { |
| "epoch": 11.551724137931034, |
| "grad_norm": 0.87890625, |
| "learning_rate": 1.999048221581858e-05, |
| "loss": 0.2744, |
| "step": 46 |
| }, |
| { |
| "epoch": 11.827586206896552, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.9986295347545738e-05, |
| "loss": 0.1503, |
| "step": 47 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.998134798421867e-05, |
| "loss": 0.1235, |
| "step": 48 |
| }, |
| { |
| "epoch": 12.275862068965518, |
| "grad_norm": 0.82421875, |
| "learning_rate": 1.9975640502598243e-05, |
| "loss": 0.1688, |
| "step": 49 |
| }, |
| { |
| "epoch": 12.551724137931034, |
| "grad_norm": 0.8671875, |
| "learning_rate": 1.9969173337331283e-05, |
| "loss": 0.1676, |
| "step": 50 |
| }, |
| { |
| "epoch": 12.551724137931034, |
| "eval_loss": 0.09635955095291138, |
| "eval_runtime": 1.7755, |
| "eval_samples_per_second": 1.69, |
| "eval_steps_per_second": 1.69, |
| "step": 50 |
| }, |
| { |
| "epoch": 12.827586206896552, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.9961946980917457e-05, |
| "loss": 0.0832, |
| "step": 51 |
| }, |
| { |
| "epoch": 13.0, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.9953961983671792e-05, |
| "loss": 0.1309, |
| "step": 52 |
| }, |
| { |
| "epoch": 13.275862068965518, |
| "grad_norm": 0.78125, |
| "learning_rate": 1.9945218953682736e-05, |
| "loss": 0.0907, |
| "step": 53 |
| }, |
| { |
| "epoch": 13.551724137931034, |
| "grad_norm": 0.796875, |
| "learning_rate": 1.9935718556765878e-05, |
| "loss": 0.094, |
| "step": 54 |
| }, |
| { |
| "epoch": 13.827586206896552, |
| "grad_norm": 1.0703125, |
| "learning_rate": 1.9925461516413224e-05, |
| "loss": 0.0985, |
| "step": 55 |
| }, |
| { |
| "epoch": 14.0, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.9914448613738107e-05, |
| "loss": 0.1094, |
| "step": 56 |
| }, |
| { |
| "epoch": 14.275862068965518, |
| "grad_norm": 0.66015625, |
| "learning_rate": 1.9902680687415704e-05, |
| "loss": 0.0448, |
| "step": 57 |
| }, |
| { |
| "epoch": 14.551724137931034, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.989015863361917e-05, |
| "loss": 0.0836, |
| "step": 58 |
| }, |
| { |
| "epoch": 14.827586206896552, |
| "grad_norm": 0.8125, |
| "learning_rate": 1.9876883405951378e-05, |
| "loss": 0.0687, |
| "step": 59 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 0.7890625, |
| "learning_rate": 1.9862856015372315e-05, |
| "loss": 0.0276, |
| "step": 60 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.03497372567653656, |
| "eval_runtime": 1.7662, |
| "eval_samples_per_second": 1.699, |
| "eval_steps_per_second": 1.699, |
| "step": 60 |
| }, |
| { |
| "epoch": 15.275862068965518, |
| "grad_norm": 0.7265625, |
| "learning_rate": 1.9848077530122083e-05, |
| "loss": 0.0395, |
| "step": 61 |
| }, |
| { |
| "epoch": 15.551724137931034, |
| "grad_norm": 0.51953125, |
| "learning_rate": 1.983254907563955e-05, |
| "loss": 0.0248, |
| "step": 62 |
| }, |
| { |
| "epoch": 15.827586206896552, |
| "grad_norm": 1.1640625, |
| "learning_rate": 1.9816271834476642e-05, |
| "loss": 0.0355, |
| "step": 63 |
| }, |
| { |
| "epoch": 16.0, |
| "grad_norm": 0.80859375, |
| "learning_rate": 1.9799247046208297e-05, |
| "loss": 0.035, |
| "step": 64 |
| }, |
| { |
| "epoch": 16.275862068965516, |
| "grad_norm": 0.494140625, |
| "learning_rate": 1.9781476007338058e-05, |
| "loss": 0.0172, |
| "step": 65 |
| }, |
| { |
| "epoch": 16.551724137931036, |
| "grad_norm": 0.59765625, |
| "learning_rate": 1.9762960071199334e-05, |
| "loss": 0.0216, |
| "step": 66 |
| }, |
| { |
| "epoch": 16.82758620689655, |
| "grad_norm": 0.984375, |
| "learning_rate": 1.9743700647852356e-05, |
| "loss": 0.0215, |
| "step": 67 |
| }, |
| { |
| "epoch": 17.0, |
| "grad_norm": 0.875, |
| "learning_rate": 1.9723699203976768e-05, |
| "loss": 0.0206, |
| "step": 68 |
| }, |
| { |
| "epoch": 17.275862068965516, |
| "grad_norm": 0.49609375, |
| "learning_rate": 1.9702957262759964e-05, |
| "loss": 0.0142, |
| "step": 69 |
| }, |
| { |
| "epoch": 17.551724137931036, |
| "grad_norm": 0.546875, |
| "learning_rate": 1.968147640378108e-05, |
| "loss": 0.0107, |
| "step": 70 |
| }, |
| { |
| "epoch": 17.551724137931036, |
| "eval_loss": 0.010336539708077908, |
| "eval_runtime": 1.7352, |
| "eval_samples_per_second": 1.729, |
| "eval_steps_per_second": 1.729, |
| "step": 70 |
| }, |
| { |
| "epoch": 17.82758620689655, |
| "grad_norm": 0.50390625, |
| "learning_rate": 1.9659258262890683e-05, |
| "loss": 0.0105, |
| "step": 71 |
| }, |
| { |
| "epoch": 18.0, |
| "grad_norm": 0.58203125, |
| "learning_rate": 1.963630453208623e-05, |
| "loss": 0.012, |
| "step": 72 |
| }, |
| { |
| "epoch": 18.275862068965516, |
| "grad_norm": 0.361328125, |
| "learning_rate": 1.961261695938319e-05, |
| "loss": 0.0079, |
| "step": 73 |
| }, |
| { |
| "epoch": 18.551724137931036, |
| "grad_norm": 0.376953125, |
| "learning_rate": 1.958819734868193e-05, |
| "loss": 0.0083, |
| "step": 74 |
| }, |
| { |
| "epoch": 18.82758620689655, |
| "grad_norm": 0.2734375, |
| "learning_rate": 1.9563047559630356e-05, |
| "loss": 0.0072, |
| "step": 75 |
| }, |
| { |
| "epoch": 19.0, |
| "grad_norm": 0.310546875, |
| "learning_rate": 1.953716950748227e-05, |
| "loss": 0.0072, |
| "step": 76 |
| }, |
| { |
| "epoch": 19.275862068965516, |
| "grad_norm": 0.2392578125, |
| "learning_rate": 1.9510565162951538e-05, |
| "loss": 0.006, |
| "step": 77 |
| }, |
| { |
| "epoch": 19.551724137931036, |
| "grad_norm": 0.259765625, |
| "learning_rate": 1.9483236552061996e-05, |
| "loss": 0.0056, |
| "step": 78 |
| }, |
| { |
| "epoch": 19.82758620689655, |
| "grad_norm": 0.28125, |
| "learning_rate": 1.945518575599317e-05, |
| "loss": 0.005, |
| "step": 79 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 0.439453125, |
| "learning_rate": 1.9426414910921785e-05, |
| "loss": 0.0061, |
| "step": 80 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.005222354084253311, |
| "eval_runtime": 1.7373, |
| "eval_samples_per_second": 1.727, |
| "eval_steps_per_second": 1.727, |
| "step": 80 |
| }, |
| { |
| "epoch": 20.275862068965516, |
| "grad_norm": 0.2353515625, |
| "learning_rate": 1.9396926207859085e-05, |
| "loss": 0.0047, |
| "step": 81 |
| }, |
| { |
| "epoch": 20.551724137931036, |
| "grad_norm": 0.16796875, |
| "learning_rate": 1.9366721892483976e-05, |
| "loss": 0.0034, |
| "step": 82 |
| }, |
| { |
| "epoch": 20.82758620689655, |
| "grad_norm": 0.609375, |
| "learning_rate": 1.9335804264972018e-05, |
| "loss": 0.0056, |
| "step": 83 |
| }, |
| { |
| "epoch": 21.0, |
| "grad_norm": 0.1884765625, |
| "learning_rate": 1.9304175679820247e-05, |
| "loss": 0.0043, |
| "step": 84 |
| }, |
| { |
| "epoch": 21.275862068965516, |
| "grad_norm": 0.1630859375, |
| "learning_rate": 1.9271838545667876e-05, |
| "loss": 0.0031, |
| "step": 85 |
| }, |
| { |
| "epoch": 21.551724137931036, |
| "grad_norm": 0.1552734375, |
| "learning_rate": 1.9238795325112867e-05, |
| "loss": 0.0041, |
| "step": 86 |
| }, |
| { |
| "epoch": 21.82758620689655, |
| "grad_norm": 0.189453125, |
| "learning_rate": 1.9205048534524405e-05, |
| "loss": 0.0034, |
| "step": 87 |
| }, |
| { |
| "epoch": 22.0, |
| "grad_norm": 0.36328125, |
| "learning_rate": 1.917060074385124e-05, |
| "loss": 0.0043, |
| "step": 88 |
| }, |
| { |
| "epoch": 22.275862068965516, |
| "grad_norm": 0.1279296875, |
| "learning_rate": 1.913545457642601e-05, |
| "loss": 0.003, |
| "step": 89 |
| }, |
| { |
| "epoch": 22.551724137931036, |
| "grad_norm": 0.1513671875, |
| "learning_rate": 1.9099612708765432e-05, |
| "loss": 0.0033, |
| "step": 90 |
| }, |
| { |
| "epoch": 22.551724137931036, |
| "eval_loss": 0.0038691603112965822, |
| "eval_runtime": 1.7266, |
| "eval_samples_per_second": 1.738, |
| "eval_steps_per_second": 1.738, |
| "step": 90 |
| }, |
| { |
| "epoch": 22.82758620689655, |
| "grad_norm": 0.1689453125, |
| "learning_rate": 1.9063077870366504e-05, |
| "loss": 0.0029, |
| "step": 91 |
| }, |
| { |
| "epoch": 23.0, |
| "grad_norm": 0.1435546875, |
| "learning_rate": 1.902585284349861e-05, |
| "loss": 0.0032, |
| "step": 92 |
| }, |
| { |
| "epoch": 23.275862068965516, |
| "grad_norm": 0.1376953125, |
| "learning_rate": 1.8987940462991673e-05, |
| "loss": 0.003, |
| "step": 93 |
| }, |
| { |
| "epoch": 23.551724137931036, |
| "grad_norm": 0.1083984375, |
| "learning_rate": 1.894934361602025e-05, |
| "loss": 0.0024, |
| "step": 94 |
| }, |
| { |
| "epoch": 23.82758620689655, |
| "grad_norm": 0.08203125, |
| "learning_rate": 1.891006524188368e-05, |
| "loss": 0.0026, |
| "step": 95 |
| }, |
| { |
| "epoch": 24.0, |
| "grad_norm": 0.1748046875, |
| "learning_rate": 1.887010833178222e-05, |
| "loss": 0.0031, |
| "step": 96 |
| }, |
| { |
| "epoch": 24.275862068965516, |
| "grad_norm": 0.11474609375, |
| "learning_rate": 1.8829475928589272e-05, |
| "loss": 0.0026, |
| "step": 97 |
| }, |
| { |
| "epoch": 24.551724137931036, |
| "grad_norm": 0.109375, |
| "learning_rate": 1.8788171126619653e-05, |
| "loss": 0.002, |
| "step": 98 |
| }, |
| { |
| "epoch": 24.82758620689655, |
| "grad_norm": 0.1123046875, |
| "learning_rate": 1.874619707139396e-05, |
| "loss": 0.0026, |
| "step": 99 |
| }, |
| { |
| "epoch": 25.0, |
| "grad_norm": 0.09912109375, |
| "learning_rate": 1.8703556959398998e-05, |
| "loss": 0.0025, |
| "step": 100 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_loss": 0.0027635388541966677, |
| "eval_runtime": 1.7315, |
| "eval_samples_per_second": 1.733, |
| "eval_steps_per_second": 1.733, |
| "step": 100 |
| }, |
| { |
| "epoch": 25.275862068965516, |
| "grad_norm": 0.0986328125, |
| "learning_rate": 1.866025403784439e-05, |
| "loss": 0.0022, |
| "step": 101 |
| }, |
| { |
| "epoch": 25.551724137931036, |
| "grad_norm": 0.08837890625, |
| "learning_rate": 1.861629160441526e-05, |
| "loss": 0.0024, |
| "step": 102 |
| }, |
| { |
| "epoch": 25.82758620689655, |
| "grad_norm": 0.09814453125, |
| "learning_rate": 1.8571673007021124e-05, |
| "loss": 0.002, |
| "step": 103 |
| }, |
| { |
| "epoch": 26.0, |
| "grad_norm": 0.103515625, |
| "learning_rate": 1.8526401643540924e-05, |
| "loss": 0.0023, |
| "step": 104 |
| }, |
| { |
| "epoch": 26.275862068965516, |
| "grad_norm": 0.09716796875, |
| "learning_rate": 1.848048096156426e-05, |
| "loss": 0.0021, |
| "step": 105 |
| }, |
| { |
| "epoch": 26.551724137931036, |
| "grad_norm": 0.0830078125, |
| "learning_rate": 1.843391445812886e-05, |
| "loss": 0.0021, |
| "step": 106 |
| }, |
| { |
| "epoch": 26.82758620689655, |
| "grad_norm": 0.06982421875, |
| "learning_rate": 1.8386705679454243e-05, |
| "loss": 0.0021, |
| "step": 107 |
| }, |
| { |
| "epoch": 27.0, |
| "grad_norm": 0.1494140625, |
| "learning_rate": 1.8338858220671683e-05, |
| "loss": 0.0022, |
| "step": 108 |
| }, |
| { |
| "epoch": 27.275862068965516, |
| "grad_norm": 0.09423828125, |
| "learning_rate": 1.8290375725550417e-05, |
| "loss": 0.0025, |
| "step": 109 |
| }, |
| { |
| "epoch": 27.551724137931036, |
| "grad_norm": 0.06982421875, |
| "learning_rate": 1.8241261886220155e-05, |
| "loss": 0.0019, |
| "step": 110 |
| }, |
| { |
| "epoch": 27.551724137931036, |
| "eval_loss": 0.0022178783547133207, |
| "eval_runtime": 1.7843, |
| "eval_samples_per_second": 1.681, |
| "eval_steps_per_second": 1.681, |
| "step": 110 |
| }, |
| { |
| "epoch": 27.82758620689655, |
| "grad_norm": 0.14453125, |
| "learning_rate": 1.819152044288992e-05, |
| "loss": 0.0022, |
| "step": 111 |
| }, |
| { |
| "epoch": 28.0, |
| "grad_norm": 0.06689453125, |
| "learning_rate": 1.8141155183563195e-05, |
| "loss": 0.0016, |
| "step": 112 |
| }, |
| { |
| "epoch": 28.275862068965516, |
| "grad_norm": 0.0771484375, |
| "learning_rate": 1.8090169943749477e-05, |
| "loss": 0.002, |
| "step": 113 |
| }, |
| { |
| "epoch": 28.551724137931036, |
| "grad_norm": 0.06298828125, |
| "learning_rate": 1.8038568606172172e-05, |
| "loss": 0.0018, |
| "step": 114 |
| }, |
| { |
| "epoch": 28.82758620689655, |
| "grad_norm": 0.07958984375, |
| "learning_rate": 1.798635510047293e-05, |
| "loss": 0.0017, |
| "step": 115 |
| }, |
| { |
| "epoch": 29.0, |
| "grad_norm": 0.16015625, |
| "learning_rate": 1.7933533402912354e-05, |
| "loss": 0.0022, |
| "step": 116 |
| }, |
| { |
| "epoch": 29.275862068965516, |
| "grad_norm": 0.07373046875, |
| "learning_rate": 1.788010753606722e-05, |
| "loss": 0.0018, |
| "step": 117 |
| }, |
| { |
| "epoch": 29.551724137931036, |
| "grad_norm": 0.07861328125, |
| "learning_rate": 1.782608156852414e-05, |
| "loss": 0.002, |
| "step": 118 |
| }, |
| { |
| "epoch": 29.82758620689655, |
| "grad_norm": 0.0634765625, |
| "learning_rate": 1.777145961456971e-05, |
| "loss": 0.0017, |
| "step": 119 |
| }, |
| { |
| "epoch": 30.0, |
| "grad_norm": 0.14453125, |
| "learning_rate": 1.7716245833877202e-05, |
| "loss": 0.0019, |
| "step": 120 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_loss": 0.001993604702875018, |
| "eval_runtime": 1.7303, |
| "eval_samples_per_second": 1.734, |
| "eval_steps_per_second": 1.734, |
| "step": 120 |
| }, |
| { |
| "epoch": 30.275862068965516, |
| "grad_norm": 0.06884765625, |
| "learning_rate": 1.766044443118978e-05, |
| "loss": 0.0018, |
| "step": 121 |
| }, |
| { |
| "epoch": 30.551724137931036, |
| "grad_norm": 0.064453125, |
| "learning_rate": 1.7604059656000313e-05, |
| "loss": 0.0015, |
| "step": 122 |
| }, |
| { |
| "epoch": 30.82758620689655, |
| "grad_norm": 0.119140625, |
| "learning_rate": 1.7547095802227723e-05, |
| "loss": 0.002, |
| "step": 123 |
| }, |
| { |
| "epoch": 31.0, |
| "grad_norm": 0.09814453125, |
| "learning_rate": 1.7489557207890025e-05, |
| "loss": 0.0021, |
| "step": 124 |
| }, |
| { |
| "epoch": 31.275862068965516, |
| "grad_norm": 0.0703125, |
| "learning_rate": 1.7431448254773943e-05, |
| "loss": 0.0015, |
| "step": 125 |
| }, |
| { |
| "epoch": 31.551724137931036, |
| "grad_norm": 0.0810546875, |
| "learning_rate": 1.737277336810124e-05, |
| "loss": 0.0017, |
| "step": 126 |
| }, |
| { |
| "epoch": 31.82758620689655, |
| "grad_norm": 0.0810546875, |
| "learning_rate": 1.7313537016191706e-05, |
| "loss": 0.0018, |
| "step": 127 |
| }, |
| { |
| "epoch": 32.0, |
| "grad_norm": 0.10302734375, |
| "learning_rate": 1.7253743710122877e-05, |
| "loss": 0.0021, |
| "step": 128 |
| }, |
| { |
| "epoch": 32.275862068965516, |
| "grad_norm": 0.07080078125, |
| "learning_rate": 1.7193398003386514e-05, |
| "loss": 0.0016, |
| "step": 129 |
| }, |
| { |
| "epoch": 32.55172413793103, |
| "grad_norm": 0.08349609375, |
| "learning_rate": 1.713250449154182e-05, |
| "loss": 0.0019, |
| "step": 130 |
| }, |
| { |
| "epoch": 32.55172413793103, |
| "eval_loss": 0.0018896261462941766, |
| "eval_runtime": 1.787, |
| "eval_samples_per_second": 1.679, |
| "eval_steps_per_second": 1.679, |
| "step": 130 |
| }, |
| { |
| "epoch": 32.827586206896555, |
| "grad_norm": 0.0625, |
| "learning_rate": 1.7071067811865477e-05, |
| "loss": 0.0016, |
| "step": 131 |
| }, |
| { |
| "epoch": 33.0, |
| "grad_norm": 0.11865234375, |
| "learning_rate": 1.700909264299851e-05, |
| "loss": 0.0015, |
| "step": 132 |
| }, |
| { |
| "epoch": 33.275862068965516, |
| "grad_norm": 0.09326171875, |
| "learning_rate": 1.6946583704589973e-05, |
| "loss": 0.0016, |
| "step": 133 |
| }, |
| { |
| "epoch": 33.55172413793103, |
| "grad_norm": 0.056396484375, |
| "learning_rate": 1.688354575693754e-05, |
| "loss": 0.0015, |
| "step": 134 |
| }, |
| { |
| "epoch": 33.827586206896555, |
| "grad_norm": 0.0615234375, |
| "learning_rate": 1.6819983600624986e-05, |
| "loss": 0.0018, |
| "step": 135 |
| }, |
| { |
| "epoch": 34.0, |
| "grad_norm": 0.1123046875, |
| "learning_rate": 1.6755902076156606e-05, |
| "loss": 0.0015, |
| "step": 136 |
| }, |
| { |
| "epoch": 34.275862068965516, |
| "grad_norm": 0.0810546875, |
| "learning_rate": 1.6691306063588583e-05, |
| "loss": 0.0015, |
| "step": 137 |
| }, |
| { |
| "epoch": 34.55172413793103, |
| "grad_norm": 0.06591796875, |
| "learning_rate": 1.6626200482157378e-05, |
| "loss": 0.0017, |
| "step": 138 |
| }, |
| { |
| "epoch": 34.827586206896555, |
| "grad_norm": 0.0703125, |
| "learning_rate": 1.6560590289905074e-05, |
| "loss": 0.0016, |
| "step": 139 |
| }, |
| { |
| "epoch": 35.0, |
| "grad_norm": 0.08544921875, |
| "learning_rate": 1.6494480483301836e-05, |
| "loss": 0.0018, |
| "step": 140 |
| }, |
| { |
| "epoch": 35.0, |
| "eval_loss": 0.0019512426806613803, |
| "eval_runtime": 1.7323, |
| "eval_samples_per_second": 1.732, |
| "eval_steps_per_second": 1.732, |
| "step": 140 |
| }, |
| { |
| "epoch": 35.275862068965516, |
| "grad_norm": 0.0498046875, |
| "learning_rate": 1.6427876096865394e-05, |
| "loss": 0.0015, |
| "step": 141 |
| }, |
| { |
| "epoch": 35.55172413793103, |
| "grad_norm": 0.1044921875, |
| "learning_rate": 1.636078220277764e-05, |
| "loss": 0.0017, |
| "step": 142 |
| }, |
| { |
| "epoch": 35.827586206896555, |
| "grad_norm": 0.0927734375, |
| "learning_rate": 1.6293203910498375e-05, |
| "loss": 0.0019, |
| "step": 143 |
| }, |
| { |
| "epoch": 36.0, |
| "grad_norm": 0.10595703125, |
| "learning_rate": 1.6225146366376198e-05, |
| "loss": 0.0014, |
| "step": 144 |
| }, |
| { |
| "epoch": 36.275862068965516, |
| "grad_norm": 0.057373046875, |
| "learning_rate": 1.6156614753256583e-05, |
| "loss": 0.0017, |
| "step": 145 |
| }, |
| { |
| "epoch": 36.55172413793103, |
| "grad_norm": 0.07470703125, |
| "learning_rate": 1.608761429008721e-05, |
| "loss": 0.0014, |
| "step": 146 |
| }, |
| { |
| "epoch": 36.827586206896555, |
| "grad_norm": 0.0712890625, |
| "learning_rate": 1.6018150231520486e-05, |
| "loss": 0.0014, |
| "step": 147 |
| }, |
| { |
| "epoch": 37.0, |
| "grad_norm": 0.11328125, |
| "learning_rate": 1.5948227867513416e-05, |
| "loss": 0.0017, |
| "step": 148 |
| }, |
| { |
| "epoch": 37.275862068965516, |
| "grad_norm": 0.06103515625, |
| "learning_rate": 1.5877852522924733e-05, |
| "loss": 0.0016, |
| "step": 149 |
| }, |
| { |
| "epoch": 37.55172413793103, |
| "grad_norm": 0.08056640625, |
| "learning_rate": 1.5807029557109398e-05, |
| "loss": 0.0016, |
| "step": 150 |
| }, |
| { |
| "epoch": 37.55172413793103, |
| "eval_loss": 0.001894477754831314, |
| "eval_runtime": 1.7336, |
| "eval_samples_per_second": 1.731, |
| "eval_steps_per_second": 1.731, |
| "step": 150 |
| }, |
| { |
| "epoch": 37.827586206896555, |
| "grad_norm": 0.06982421875, |
| "learning_rate": 1.573576436351046e-05, |
| "loss": 0.0016, |
| "step": 151 |
| }, |
| { |
| "epoch": 38.0, |
| "grad_norm": 0.091796875, |
| "learning_rate": 1.566406236924833e-05, |
| "loss": 0.0014, |
| "step": 152 |
| }, |
| { |
| "epoch": 38.275862068965516, |
| "grad_norm": 0.05517578125, |
| "learning_rate": 1.5591929034707468e-05, |
| "loss": 0.0014, |
| "step": 153 |
| }, |
| { |
| "epoch": 38.55172413793103, |
| "grad_norm": 0.0732421875, |
| "learning_rate": 1.5519369853120584e-05, |
| "loss": 0.0018, |
| "step": 154 |
| }, |
| { |
| "epoch": 38.827586206896555, |
| "grad_norm": 0.08154296875, |
| "learning_rate": 1.5446390350150272e-05, |
| "loss": 0.0015, |
| "step": 155 |
| }, |
| { |
| "epoch": 39.0, |
| "grad_norm": 0.1083984375, |
| "learning_rate": 1.5372996083468242e-05, |
| "loss": 0.0015, |
| "step": 156 |
| }, |
| { |
| "epoch": 39.275862068965516, |
| "grad_norm": 0.06884765625, |
| "learning_rate": 1.529919264233205e-05, |
| "loss": 0.0017, |
| "step": 157 |
| }, |
| { |
| "epoch": 39.55172413793103, |
| "grad_norm": 0.052978515625, |
| "learning_rate": 1.5224985647159489e-05, |
| "loss": 0.0014, |
| "step": 158 |
| }, |
| { |
| "epoch": 39.827586206896555, |
| "grad_norm": 0.0712890625, |
| "learning_rate": 1.5150380749100545e-05, |
| "loss": 0.0015, |
| "step": 159 |
| }, |
| { |
| "epoch": 40.0, |
| "grad_norm": 0.10888671875, |
| "learning_rate": 1.5075383629607043e-05, |
| "loss": 0.0016, |
| "step": 160 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_loss": 0.0018837498500943184, |
| "eval_runtime": 1.7505, |
| "eval_samples_per_second": 1.714, |
| "eval_steps_per_second": 1.714, |
| "step": 160 |
| }, |
| { |
| "epoch": 40.275862068965516, |
| "grad_norm": 0.07373046875, |
| "learning_rate": 1.5000000000000002e-05, |
| "loss": 0.0015, |
| "step": 161 |
| }, |
| { |
| "epoch": 40.55172413793103, |
| "grad_norm": 0.078125, |
| "learning_rate": 1.4924235601034673e-05, |
| "loss": 0.0018, |
| "step": 162 |
| }, |
| { |
| "epoch": 40.827586206896555, |
| "grad_norm": 0.06884765625, |
| "learning_rate": 1.4848096202463373e-05, |
| "loss": 0.0015, |
| "step": 163 |
| }, |
| { |
| "epoch": 41.0, |
| "grad_norm": 0.058837890625, |
| "learning_rate": 1.4771587602596085e-05, |
| "loss": 0.0013, |
| "step": 164 |
| }, |
| { |
| "epoch": 41.275862068965516, |
| "grad_norm": 0.05859375, |
| "learning_rate": 1.469471562785891e-05, |
| "loss": 0.0017, |
| "step": 165 |
| }, |
| { |
| "epoch": 41.55172413793103, |
| "grad_norm": 0.06494140625, |
| "learning_rate": 1.4617486132350343e-05, |
| "loss": 0.0014, |
| "step": 166 |
| }, |
| { |
| "epoch": 41.827586206896555, |
| "grad_norm": 0.05908203125, |
| "learning_rate": 1.4539904997395468e-05, |
| "loss": 0.0013, |
| "step": 167 |
| }, |
| { |
| "epoch": 42.0, |
| "grad_norm": 0.201171875, |
| "learning_rate": 1.4461978131098089e-05, |
| "loss": 0.0021, |
| "step": 168 |
| }, |
| { |
| "epoch": 42.275862068965516, |
| "grad_norm": 0.055908203125, |
| "learning_rate": 1.4383711467890776e-05, |
| "loss": 0.0014, |
| "step": 169 |
| }, |
| { |
| "epoch": 42.55172413793103, |
| "grad_norm": 0.07470703125, |
| "learning_rate": 1.4305110968082953e-05, |
| "loss": 0.0014, |
| "step": 170 |
| }, |
| { |
| "epoch": 42.55172413793103, |
| "eval_loss": 0.0018982401816174388, |
| "eval_runtime": 1.7258, |
| "eval_samples_per_second": 1.738, |
| "eval_steps_per_second": 1.738, |
| "step": 170 |
| }, |
| { |
| "epoch": 42.827586206896555, |
| "grad_norm": 0.05322265625, |
| "learning_rate": 1.4226182617406996e-05, |
| "loss": 0.0014, |
| "step": 171 |
| }, |
| { |
| "epoch": 43.0, |
| "grad_norm": 0.16015625, |
| "learning_rate": 1.4146932426562391e-05, |
| "loss": 0.0021, |
| "step": 172 |
| }, |
| { |
| "epoch": 43.275862068965516, |
| "grad_norm": 0.068359375, |
| "learning_rate": 1.4067366430758004e-05, |
| "loss": 0.0016, |
| "step": 173 |
| }, |
| { |
| "epoch": 43.55172413793103, |
| "grad_norm": 0.0693359375, |
| "learning_rate": 1.3987490689252463e-05, |
| "loss": 0.0014, |
| "step": 174 |
| }, |
| { |
| "epoch": 43.827586206896555, |
| "grad_norm": 0.07177734375, |
| "learning_rate": 1.3907311284892737e-05, |
| "loss": 0.0016, |
| "step": 175 |
| }, |
| { |
| "epoch": 44.0, |
| "grad_norm": 0.07568359375, |
| "learning_rate": 1.3826834323650899e-05, |
| "loss": 0.0013, |
| "step": 176 |
| }, |
| { |
| "epoch": 44.275862068965516, |
| "grad_norm": 0.07177734375, |
| "learning_rate": 1.3746065934159123e-05, |
| "loss": 0.0013, |
| "step": 177 |
| }, |
| { |
| "epoch": 44.55172413793103, |
| "grad_norm": 0.0595703125, |
| "learning_rate": 1.3665012267242974e-05, |
| "loss": 0.0015, |
| "step": 178 |
| }, |
| { |
| "epoch": 44.827586206896555, |
| "grad_norm": 0.0771484375, |
| "learning_rate": 1.3583679495453e-05, |
| "loss": 0.0016, |
| "step": 179 |
| }, |
| { |
| "epoch": 45.0, |
| "grad_norm": 0.11083984375, |
| "learning_rate": 1.3502073812594677e-05, |
| "loss": 0.0016, |
| "step": 180 |
| }, |
| { |
| "epoch": 45.0, |
| "eval_loss": 0.0018364647403359413, |
| "eval_runtime": 1.7371, |
| "eval_samples_per_second": 1.727, |
| "eval_steps_per_second": 1.727, |
| "step": 180 |
| }, |
| { |
| "epoch": 45.275862068965516, |
| "grad_norm": 0.046142578125, |
| "learning_rate": 1.342020143325669e-05, |
| "loss": 0.0014, |
| "step": 181 |
| }, |
| { |
| "epoch": 45.55172413793103, |
| "grad_norm": 0.08251953125, |
| "learning_rate": 1.333806859233771e-05, |
| "loss": 0.0015, |
| "step": 182 |
| }, |
| { |
| "epoch": 45.827586206896555, |
| "grad_norm": 0.09912109375, |
| "learning_rate": 1.3255681544571568e-05, |
| "loss": 0.0015, |
| "step": 183 |
| }, |
| { |
| "epoch": 46.0, |
| "grad_norm": 0.0986328125, |
| "learning_rate": 1.3173046564050923e-05, |
| "loss": 0.0013, |
| "step": 184 |
| }, |
| { |
| "epoch": 46.275862068965516, |
| "grad_norm": 0.07275390625, |
| "learning_rate": 1.3090169943749475e-05, |
| "loss": 0.0013, |
| "step": 185 |
| }, |
| { |
| "epoch": 46.55172413793103, |
| "grad_norm": 0.05859375, |
| "learning_rate": 1.300705799504273e-05, |
| "loss": 0.0015, |
| "step": 186 |
| }, |
| { |
| "epoch": 46.827586206896555, |
| "grad_norm": 0.06396484375, |
| "learning_rate": 1.2923717047227368e-05, |
| "loss": 0.0014, |
| "step": 187 |
| }, |
| { |
| "epoch": 47.0, |
| "grad_norm": 0.09033203125, |
| "learning_rate": 1.284015344703923e-05, |
| "loss": 0.0016, |
| "step": 188 |
| }, |
| { |
| "epoch": 47.275862068965516, |
| "grad_norm": 0.09423828125, |
| "learning_rate": 1.2756373558169992e-05, |
| "loss": 0.0015, |
| "step": 189 |
| }, |
| { |
| "epoch": 47.55172413793103, |
| "grad_norm": 0.044189453125, |
| "learning_rate": 1.267238376078257e-05, |
| "loss": 0.0013, |
| "step": 190 |
| }, |
| { |
| "epoch": 47.55172413793103, |
| "eval_loss": 0.0017142044380307198, |
| "eval_runtime": 1.7594, |
| "eval_samples_per_second": 1.705, |
| "eval_steps_per_second": 1.705, |
| "step": 190 |
| }, |
| { |
| "epoch": 47.827586206896555, |
| "grad_norm": 0.263671875, |
| "learning_rate": 1.2588190451025209e-05, |
| "loss": 0.0024, |
| "step": 191 |
| }, |
| { |
| "epoch": 48.0, |
| "grad_norm": 0.0712890625, |
| "learning_rate": 1.2503800040544417e-05, |
| "loss": 0.0014, |
| "step": 192 |
| }, |
| { |
| "epoch": 48.275862068965516, |
| "grad_norm": 0.08154296875, |
| "learning_rate": 1.2419218955996677e-05, |
| "loss": 0.0017, |
| "step": 193 |
| }, |
| { |
| "epoch": 48.55172413793103, |
| "grad_norm": 0.0595703125, |
| "learning_rate": 1.2334453638559057e-05, |
| "loss": 0.0013, |
| "step": 194 |
| }, |
| { |
| "epoch": 48.827586206896555, |
| "grad_norm": 0.055419921875, |
| "learning_rate": 1.2249510543438652e-05, |
| "loss": 0.0013, |
| "step": 195 |
| }, |
| { |
| "epoch": 49.0, |
| "grad_norm": 0.142578125, |
| "learning_rate": 1.2164396139381029e-05, |
| "loss": 0.0016, |
| "step": 196 |
| }, |
| { |
| "epoch": 49.275862068965516, |
| "grad_norm": 0.07177734375, |
| "learning_rate": 1.2079116908177592e-05, |
| "loss": 0.0015, |
| "step": 197 |
| }, |
| { |
| "epoch": 49.55172413793103, |
| "grad_norm": 0.059814453125, |
| "learning_rate": 1.1993679344171973e-05, |
| "loss": 0.0014, |
| "step": 198 |
| }, |
| { |
| "epoch": 49.827586206896555, |
| "grad_norm": 0.1376953125, |
| "learning_rate": 1.190808995376545e-05, |
| "loss": 0.0018, |
| "step": 199 |
| }, |
| { |
| "epoch": 50.0, |
| "grad_norm": 0.0654296875, |
| "learning_rate": 1.1822355254921478e-05, |
| "loss": 0.0012, |
| "step": 200 |
| }, |
| { |
| "epoch": 50.0, |
| "eval_loss": 0.0016466762172058225, |
| "eval_runtime": 1.723, |
| "eval_samples_per_second": 1.741, |
| "eval_steps_per_second": 1.741, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 400, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 100, |
| "save_steps": 40, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.072322663530496e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
