| { |
| "best_global_step": 225, |
| "best_metric": 0.05929639935493469, |
| "best_model_checkpoint": "saves_stability/lntuning/llama-3-8b-instruct/train_copa_1757340255/checkpoint-225", |
| "epoch": 10.0, |
| "eval_steps": 45, |
| "global_step": 900, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05555555555555555, |
| "grad_norm": 12.413766860961914, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 0.5401, |
| "num_input_tokens_seen": 1632, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 10.926953315734863, |
| "learning_rate": 5e-06, |
| "loss": 0.589, |
| "num_input_tokens_seen": 3232, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 11.459014892578125, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.5822, |
| "num_input_tokens_seen": 4832, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 9.86060619354248, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.6143, |
| "num_input_tokens_seen": 6432, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 8.448033332824707, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.4866, |
| "num_input_tokens_seen": 7968, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 8.12926197052002, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.2775, |
| "num_input_tokens_seen": 9504, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 6.437356472015381, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.3527, |
| "num_input_tokens_seen": 11104, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 6.665041923522949, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.4136, |
| "num_input_tokens_seen": 12704, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 2.534313201904297, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.2081, |
| "num_input_tokens_seen": 14240, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.11573882400989532, |
| "eval_runtime": 0.5875, |
| "eval_samples_per_second": 68.083, |
| "eval_steps_per_second": 17.021, |
| "num_input_tokens_seen": 14240, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 5.011988639831543, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.1648, |
| "num_input_tokens_seen": 15808, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 0.8455270528793335, |
| "learning_rate": 3e-05, |
| "loss": 0.1657, |
| "num_input_tokens_seen": 17344, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 6.417602062225342, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.0886, |
| "num_input_tokens_seen": 18912, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 0.35575202107429504, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.1172, |
| "num_input_tokens_seen": 20448, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 0.35196706652641296, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.0435, |
| "num_input_tokens_seen": 21984, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 0.32802021503448486, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.1677, |
| "num_input_tokens_seen": 23552, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 0.16649259626865387, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.1363, |
| "num_input_tokens_seen": 25120, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 0.053206074982881546, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.0173, |
| "num_input_tokens_seen": 26656, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 11.185739517211914, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.2591, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.07012443244457245, |
| "eval_runtime": 0.5977, |
| "eval_samples_per_second": 66.923, |
| "eval_steps_per_second": 16.731, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 4.218883991241455, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.0344, |
| "num_input_tokens_seen": 29792, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.030968718230724335, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.0927, |
| "num_input_tokens_seen": 31328, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 4.61262321472168, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.0883, |
| "num_input_tokens_seen": 32832, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 5.351931571960449, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.0412, |
| "num_input_tokens_seen": 34304, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 0.05530461296439171, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.1526, |
| "num_input_tokens_seen": 35840, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 5.916304588317871, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.2116, |
| "num_input_tokens_seen": 37376, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 0.3439420759677887, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.0965, |
| "num_input_tokens_seen": 38944, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 0.2820066511631012, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.0462, |
| "num_input_tokens_seen": 40512, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.4881448745727539, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.0367, |
| "num_input_tokens_seen": 42080, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.06827564537525177, |
| "eval_runtime": 0.5886, |
| "eval_samples_per_second": 67.952, |
| "eval_steps_per_second": 16.988, |
| "num_input_tokens_seen": 42080, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 4.441239833831787, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.143, |
| "num_input_tokens_seen": 43680, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 0.04426378384232521, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.0385, |
| "num_input_tokens_seen": 45216, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 2.143786907196045, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.192, |
| "num_input_tokens_seen": 46720, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 0.15666843950748444, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.0042, |
| "num_input_tokens_seen": 48288, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 4.361499786376953, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.0937, |
| "num_input_tokens_seen": 49888, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 0.02156808227300644, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.0234, |
| "num_input_tokens_seen": 51424, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 6.929230690002441, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.1932, |
| "num_input_tokens_seen": 52992, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 0.6252282857894897, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.0798, |
| "num_input_tokens_seen": 54592, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.18464066088199615, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.1331, |
| "num_input_tokens_seen": 56192, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.05967045947909355, |
| "eval_runtime": 0.5919, |
| "eval_samples_per_second": 67.58, |
| "eval_steps_per_second": 16.895, |
| "num_input_tokens_seen": 56192, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 1.8507845401763916, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0186, |
| "num_input_tokens_seen": 57760, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 0.018928587436676025, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.0333, |
| "num_input_tokens_seen": 59264, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 0.03185804933309555, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.0572, |
| "num_input_tokens_seen": 60768, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 6.1525678634643555, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.1263, |
| "num_input_tokens_seen": 62272, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 0.24679124355316162, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.0479, |
| "num_input_tokens_seen": 63808, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 4.617246150970459, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.1704, |
| "num_input_tokens_seen": 65344, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 1.2960439920425415, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.0061, |
| "num_input_tokens_seen": 66912, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 6.159663677215576, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.1972, |
| "num_input_tokens_seen": 68480, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.7147485017776489, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.0047, |
| "num_input_tokens_seen": 70048, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.05929639935493469, |
| "eval_runtime": 0.5943, |
| "eval_samples_per_second": 67.308, |
| "eval_steps_per_second": 16.827, |
| "num_input_tokens_seen": 70048, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 0.47322940826416016, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0096, |
| "num_input_tokens_seen": 71680, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 5.344168663024902, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.1928, |
| "num_input_tokens_seen": 73280, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.012929716147482395, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 74752, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 0.05127738416194916, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.106, |
| "num_input_tokens_seen": 76288, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.03636276721954346, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0649, |
| "num_input_tokens_seen": 77888, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.31299880146980286, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 79456, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.5931472778320312, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.0646, |
| "num_input_tokens_seen": 81088, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 0.7903435230255127, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.0188, |
| "num_input_tokens_seen": 82592, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 4.708324909210205, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.0918, |
| "num_input_tokens_seen": 84192, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.059632014483213425, |
| "eval_runtime": 0.5811, |
| "eval_samples_per_second": 68.838, |
| "eval_steps_per_second": 17.21, |
| "num_input_tokens_seen": 84192, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 0.02156168594956398, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.0337, |
| "num_input_tokens_seen": 85696, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 1.075160264968872, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.0222, |
| "num_input_tokens_seen": 87296, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.09805523604154587, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.01, |
| "num_input_tokens_seen": 88832, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.028949294239282608, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0024, |
| "num_input_tokens_seen": 90368, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 0.4382733702659607, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.0078, |
| "num_input_tokens_seen": 91968, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 2.2116539478302, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0124, |
| "num_input_tokens_seen": 93568, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 0.011147049255669117, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.01, |
| "num_input_tokens_seen": 95168, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.057261109352111816, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.0729, |
| "num_input_tokens_seen": 96704, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.028863638639450073, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.1091, |
| "num_input_tokens_seen": 98304, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.06172233074903488, |
| "eval_runtime": 0.5816, |
| "eval_samples_per_second": 68.77, |
| "eval_steps_per_second": 17.193, |
| "num_input_tokens_seen": 98304, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.021796802058815956, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 99840, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 0.8977848291397095, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.032, |
| "num_input_tokens_seen": 101376, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 0.11903944611549377, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.0704, |
| "num_input_tokens_seen": 102976, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 3.8452537059783936, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.0592, |
| "num_input_tokens_seen": 104576, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 1.4291119575500488, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.1871, |
| "num_input_tokens_seen": 106112, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.04512714594602585, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.1058, |
| "num_input_tokens_seen": 107712, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 0.29252851009368896, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.1107, |
| "num_input_tokens_seen": 109344, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 6.142524719238281, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.1373, |
| "num_input_tokens_seen": 110944, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.19722937047481537, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0053, |
| "num_input_tokens_seen": 112544, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.062212228775024414, |
| "eval_runtime": 0.5812, |
| "eval_samples_per_second": 68.82, |
| "eval_steps_per_second": 17.205, |
| "num_input_tokens_seen": 112544, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.09612889587879181, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.0397, |
| "num_input_tokens_seen": 114144, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 2.852125406265259, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.0304, |
| "num_input_tokens_seen": 115712, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.6219462156295776, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.1283, |
| "num_input_tokens_seen": 117312, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.216511532664299, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.0795, |
| "num_input_tokens_seen": 118848, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 2.406106472015381, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0155, |
| "num_input_tokens_seen": 120416, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 0.8766459822654724, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0599, |
| "num_input_tokens_seen": 122016, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 6.439513683319092, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0956, |
| "num_input_tokens_seen": 123584, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.011177383363246918, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0031, |
| "num_input_tokens_seen": 125216, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.6205223798751831, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.0101, |
| "num_input_tokens_seen": 126784, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.06303347647190094, |
| "eval_runtime": 0.5821, |
| "eval_samples_per_second": 68.711, |
| "eval_steps_per_second": 17.178, |
| "num_input_tokens_seen": 126784, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 0.12966503202915192, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.0062, |
| "num_input_tokens_seen": 128352, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.27338626980781555, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0062, |
| "num_input_tokens_seen": 129920, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.04085496813058853, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 131488, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.0658736526966095, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.0676, |
| "num_input_tokens_seen": 133056, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.40652352571487427, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.038, |
| "num_input_tokens_seen": 134624, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.22422264516353607, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0227, |
| "num_input_tokens_seen": 136192, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.009336884133517742, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0115, |
| "num_input_tokens_seen": 137824, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 0.28929874300956726, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0154, |
| "num_input_tokens_seen": 139392, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.1447066068649292, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0808, |
| "num_input_tokens_seen": 140960, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.06195898726582527, |
| "eval_runtime": 0.5844, |
| "eval_samples_per_second": 68.443, |
| "eval_steps_per_second": 17.111, |
| "num_input_tokens_seen": 140960, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.04051095247268677, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.0635, |
| "num_input_tokens_seen": 142560, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.023053204640746117, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0316, |
| "num_input_tokens_seen": 144096, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.23852871358394623, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0049, |
| "num_input_tokens_seen": 145632, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 2.9483938217163086, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0206, |
| "num_input_tokens_seen": 147232, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 0.058457694947719574, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.0586, |
| "num_input_tokens_seen": 148864, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.023417210206389427, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.003, |
| "num_input_tokens_seen": 150432, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.29810523986816406, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0771, |
| "num_input_tokens_seen": 152032, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 0.020725004374980927, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.025, |
| "num_input_tokens_seen": 153600, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.4676690399646759, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0104, |
| "num_input_tokens_seen": 155200, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.0627414733171463, |
| "eval_runtime": 0.5812, |
| "eval_samples_per_second": 68.818, |
| "eval_steps_per_second": 17.204, |
| "num_input_tokens_seen": 155200, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.01173495128750801, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.0801, |
| "num_input_tokens_seen": 156800, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 0.007062096148729324, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 158336, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 0.04896526783704758, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 159904, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 6.96190071105957, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0464, |
| "num_input_tokens_seen": 161504, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.008594535291194916, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 163008, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.40661683678627014, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0174, |
| "num_input_tokens_seen": 164512, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 0.24199238419532776, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.1174, |
| "num_input_tokens_seen": 166080, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.025658387690782547, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0317, |
| "num_input_tokens_seen": 167680, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.006939691957086325, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 169216, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.06366995722055435, |
| "eval_runtime": 0.5877, |
| "eval_samples_per_second": 68.063, |
| "eval_steps_per_second": 17.016, |
| "num_input_tokens_seen": 169216, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.017029551789164543, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 170784, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 4.089417457580566, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0248, |
| "num_input_tokens_seen": 172352, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.15079845488071442, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 173920, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.006432878784835339, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 175488, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 2.5554280281066895, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0127, |
| "num_input_tokens_seen": 177024, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.41308820247650146, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.025, |
| "num_input_tokens_seen": 178624, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.43808722496032715, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.1628, |
| "num_input_tokens_seen": 180096, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 2.488274335861206, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0157, |
| "num_input_tokens_seen": 181600, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 1.1927860975265503, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0056, |
| "num_input_tokens_seen": 183232, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.06768044084310532, |
| "eval_runtime": 0.5879, |
| "eval_samples_per_second": 68.036, |
| "eval_steps_per_second": 17.009, |
| "num_input_tokens_seen": 183232, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.0071192653849720955, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0707, |
| "num_input_tokens_seen": 184768, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 0.004635022487491369, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0261, |
| "num_input_tokens_seen": 186208, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.007054431829601526, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.0091, |
| "num_input_tokens_seen": 187776, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 0.7310401797294617, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0049, |
| "num_input_tokens_seen": 189344, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.06609062105417252, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 190912, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.014985096640884876, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.0607, |
| "num_input_tokens_seen": 192512, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.010730482637882233, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 194112, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.30494803190231323, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0662, |
| "num_input_tokens_seen": 195712, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.06056293845176697, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 197248, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.07016871124505997, |
| "eval_runtime": 0.5849, |
| "eval_samples_per_second": 68.386, |
| "eval_steps_per_second": 17.096, |
| "num_input_tokens_seen": 197248, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.03782691806554794, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 198784, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 4.731870651245117, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0685, |
| "num_input_tokens_seen": 200352, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.1799234002828598, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 201856, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.010190362110733986, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0256, |
| "num_input_tokens_seen": 203488, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 0.007492389529943466, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 205056, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.06244223192334175, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0582, |
| "num_input_tokens_seen": 206656, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.04081568121910095, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0057, |
| "num_input_tokens_seen": 208288, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 7.480345726013184, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.096, |
| "num_input_tokens_seen": 209856, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 0.7831881046295166, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.0042, |
| "num_input_tokens_seen": 211424, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.06858132034540176, |
| "eval_runtime": 0.5834, |
| "eval_samples_per_second": 68.565, |
| "eval_steps_per_second": 17.141, |
| "num_input_tokens_seen": 211424, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.01823975145816803, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 212928, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.018310269340872765, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0025, |
| "num_input_tokens_seen": 214496, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.01911945454776287, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 216064, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 0.017379632219672203, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.0439, |
| "num_input_tokens_seen": 217600, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 0.5448654890060425, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0035, |
| "num_input_tokens_seen": 219136, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.15058286488056183, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0106, |
| "num_input_tokens_seen": 220704, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.2723049223423004, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0235, |
| "num_input_tokens_seen": 222272, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.11821077018976212, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 223872, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.420344352722168, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0692, |
| "num_input_tokens_seen": 225440, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.06697994470596313, |
| "eval_runtime": 0.5855, |
| "eval_samples_per_second": 68.323, |
| "eval_steps_per_second": 17.081, |
| "num_input_tokens_seen": 225440, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 1.6140986680984497, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0069, |
| "num_input_tokens_seen": 226912, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.10575804114341736, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 228512, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.4418221414089203, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.05, |
| "num_input_tokens_seen": 230080, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.006636911537498236, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 231616, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.005838257260620594, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0232, |
| "num_input_tokens_seen": 233184, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.16078728437423706, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 234688, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.15908759832382202, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.1145, |
| "num_input_tokens_seen": 236256, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.19336074590682983, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0028, |
| "num_input_tokens_seen": 237856, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 0.05375547334551811, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 239392, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.06791551411151886, |
| "eval_runtime": 0.5839, |
| "eval_samples_per_second": 68.503, |
| "eval_steps_per_second": 17.126, |
| "num_input_tokens_seen": 239392, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.12801700830459595, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0756, |
| "num_input_tokens_seen": 240960, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.013426029123365879, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.0063, |
| "num_input_tokens_seen": 242496, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 1.9325268268585205, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0073, |
| "num_input_tokens_seen": 244064, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.0063558281399309635, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 245664, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.04126088321208954, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 247264, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.07367562502622604, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.015, |
| "num_input_tokens_seen": 248832, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.00970037654042244, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0739, |
| "num_input_tokens_seen": 250432, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.00786888599395752, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0179, |
| "num_input_tokens_seen": 252064, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.005709424614906311, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 253632, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.0697808712720871, |
| "eval_runtime": 0.5791, |
| "eval_samples_per_second": 69.068, |
| "eval_steps_per_second": 17.267, |
| "num_input_tokens_seen": 253632, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.16491225361824036, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0478, |
| "num_input_tokens_seen": 255232, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.006427465938031673, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 256800, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.00451459689065814, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.0648, |
| "num_input_tokens_seen": 258368, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.03183142468333244, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 259872, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.0683126449584961, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 261376, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.4682014286518097, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0036, |
| "num_input_tokens_seen": 262944, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 0.07767174392938614, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0904, |
| "num_input_tokens_seen": 264544, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.2937130331993103, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.002, |
| "num_input_tokens_seen": 266080, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 0.15090857446193695, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0069, |
| "num_input_tokens_seen": 267680, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.06901142001152039, |
| "eval_runtime": 0.5858, |
| "eval_samples_per_second": 68.285, |
| "eval_steps_per_second": 17.071, |
| "num_input_tokens_seen": 267680, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 5.201120376586914, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.1035, |
| "num_input_tokens_seen": 269280, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.4644763171672821, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0035, |
| "num_input_tokens_seen": 270912, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.033365171402692795, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0094, |
| "num_input_tokens_seen": 272512, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.2439211755990982, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0165, |
| "num_input_tokens_seen": 274048, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 2.971935987472534, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0171, |
| "num_input_tokens_seen": 275648, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 0.022272281348705292, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0065, |
| "num_input_tokens_seen": 277248, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 0.04621336981654167, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.0028, |
| "num_input_tokens_seen": 278816, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.011941147036850452, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.001, |
| "num_input_tokens_seen": 280416, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.061786286532878876, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.003, |
| "num_input_tokens_seen": 281984, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.06747323274612427, |
| "eval_runtime": 0.5823, |
| "eval_samples_per_second": 68.688, |
| "eval_steps_per_second": 17.172, |
| "num_input_tokens_seen": 281984, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281984, |
| "step": 900, |
| "total_flos": 1.2698062524776448e+16, |
| "train_loss": 0.06748742392027958, |
| "train_runtime": 155.7099, |
| "train_samples_per_second": 23.12, |
| "train_steps_per_second": 5.78 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 900, |
| "num_input_tokens_seen": 281984, |
| "num_train_epochs": 10, |
| "save_steps": 45, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.2698062524776448e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
