| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002, | |
| "grad_norm": 1.1219478845596313, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 2.2204, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0004, | |
| "grad_norm": 1.0918514728546143, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 2.7879, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0006, | |
| "grad_norm": 0.5491233468055725, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6259, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008, | |
| "grad_norm": 0.5795004963874817, | |
| "learning_rate": 0.00019793814432989693, | |
| "loss": 2.7188, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.001, | |
| "grad_norm": 0.4776715934276581, | |
| "learning_rate": 0.00019587628865979381, | |
| "loss": 2.3886, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0012, | |
| "grad_norm": 0.5436866283416748, | |
| "learning_rate": 0.00019381443298969073, | |
| "loss": 2.2441, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0014, | |
| "grad_norm": 0.5175201296806335, | |
| "learning_rate": 0.00019175257731958765, | |
| "loss": 2.5653, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0016, | |
| "grad_norm": 0.4766971170902252, | |
| "learning_rate": 0.00018969072164948454, | |
| "loss": 2.1695, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0018, | |
| "grad_norm": 0.534247875213623, | |
| "learning_rate": 0.00018762886597938145, | |
| "loss": 1.9714, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002, | |
| "grad_norm": 0.47341188788414, | |
| "learning_rate": 0.00018556701030927837, | |
| "loss": 2.01, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0022, | |
| "grad_norm": 0.5537518262863159, | |
| "learning_rate": 0.00018350515463917526, | |
| "loss": 2.3351, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0024, | |
| "grad_norm": 0.6112836003303528, | |
| "learning_rate": 0.00018144329896907217, | |
| "loss": 1.8101, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0026, | |
| "grad_norm": 0.4965449869632721, | |
| "learning_rate": 0.0001793814432989691, | |
| "loss": 1.7917, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0028, | |
| "grad_norm": 0.515200138092041, | |
| "learning_rate": 0.00017731958762886598, | |
| "loss": 1.9197, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.003, | |
| "grad_norm": 0.5032430291175842, | |
| "learning_rate": 0.0001752577319587629, | |
| "loss": 2.0435, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0032, | |
| "grad_norm": 0.4674118459224701, | |
| "learning_rate": 0.0001731958762886598, | |
| "loss": 1.7773, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0034, | |
| "grad_norm": 0.5223268866539001, | |
| "learning_rate": 0.0001711340206185567, | |
| "loss": 1.9218, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0036, | |
| "grad_norm": 1.04414963722229, | |
| "learning_rate": 0.00016907216494845361, | |
| "loss": 2.0631, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0038, | |
| "grad_norm": 0.5213132500648499, | |
| "learning_rate": 0.00016701030927835053, | |
| "loss": 2.1922, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.004, | |
| "grad_norm": 0.5390024781227112, | |
| "learning_rate": 0.00016494845360824742, | |
| "loss": 2.2059, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0042, | |
| "grad_norm": 0.4950883388519287, | |
| "learning_rate": 0.00016288659793814434, | |
| "loss": 1.9657, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0044, | |
| "grad_norm": 0.4840719699859619, | |
| "learning_rate": 0.00016082474226804125, | |
| "loss": 1.9423, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0046, | |
| "grad_norm": 0.48367372155189514, | |
| "learning_rate": 0.00015876288659793814, | |
| "loss": 1.8103, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0048, | |
| "grad_norm": 0.704979419708252, | |
| "learning_rate": 0.00015670103092783506, | |
| "loss": 2.0108, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 0.5910339951515198, | |
| "learning_rate": 0.00015463917525773197, | |
| "loss": 1.9767, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0052, | |
| "grad_norm": 0.7773292660713196, | |
| "learning_rate": 0.00015257731958762886, | |
| "loss": 2.1536, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0054, | |
| "grad_norm": 0.5155345797538757, | |
| "learning_rate": 0.00015051546391752578, | |
| "loss": 2.1462, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0056, | |
| "grad_norm": 0.510045051574707, | |
| "learning_rate": 0.0001484536082474227, | |
| "loss": 1.9734, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0058, | |
| "grad_norm": 0.527519166469574, | |
| "learning_rate": 0.00014639175257731958, | |
| "loss": 2.0963, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.006, | |
| "grad_norm": 0.4819324016571045, | |
| "learning_rate": 0.0001443298969072165, | |
| "loss": 1.969, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0062, | |
| "grad_norm": 0.44363436102867126, | |
| "learning_rate": 0.00014226804123711342, | |
| "loss": 1.8159, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0064, | |
| "grad_norm": 0.4479809105396271, | |
| "learning_rate": 0.0001402061855670103, | |
| "loss": 1.6981, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0066, | |
| "grad_norm": 0.42961710691452026, | |
| "learning_rate": 0.00013814432989690722, | |
| "loss": 1.6154, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0068, | |
| "grad_norm": 0.47089216113090515, | |
| "learning_rate": 0.00013608247422680414, | |
| "loss": 1.8148, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.007, | |
| "grad_norm": 0.5263468623161316, | |
| "learning_rate": 0.00013402061855670103, | |
| "loss": 2.1879, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0072, | |
| "grad_norm": 0.4599800109863281, | |
| "learning_rate": 0.00013195876288659794, | |
| "loss": 1.6743, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0074, | |
| "grad_norm": 0.505596399307251, | |
| "learning_rate": 0.00012989690721649486, | |
| "loss": 1.6814, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0076, | |
| "grad_norm": 0.5242177248001099, | |
| "learning_rate": 0.00012783505154639175, | |
| "loss": 1.9174, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0078, | |
| "grad_norm": 0.5053271651268005, | |
| "learning_rate": 0.00012577319587628866, | |
| "loss": 1.7547, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.008, | |
| "grad_norm": 0.5670127868652344, | |
| "learning_rate": 0.00012371134020618558, | |
| "loss": 2.0127, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0082, | |
| "grad_norm": 0.608784019947052, | |
| "learning_rate": 0.00012164948453608247, | |
| "loss": 1.401, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0084, | |
| "grad_norm": 0.48027142882347107, | |
| "learning_rate": 0.00011958762886597938, | |
| "loss": 1.5524, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0086, | |
| "grad_norm": 0.4793825149536133, | |
| "learning_rate": 0.0001175257731958763, | |
| "loss": 1.4128, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0088, | |
| "grad_norm": 0.5311229825019836, | |
| "learning_rate": 0.00011546391752577319, | |
| "loss": 1.5239, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.009, | |
| "grad_norm": 0.4911600351333618, | |
| "learning_rate": 0.0001134020618556701, | |
| "loss": 1.4944, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0092, | |
| "grad_norm": 0.5163685083389282, | |
| "learning_rate": 0.00011134020618556702, | |
| "loss": 1.5343, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0094, | |
| "grad_norm": 0.47748181223869324, | |
| "learning_rate": 0.00010927835051546391, | |
| "loss": 1.2547, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0096, | |
| "grad_norm": 0.6499949097633362, | |
| "learning_rate": 0.00010721649484536083, | |
| "loss": 1.8584, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0098, | |
| "grad_norm": 0.6305411458015442, | |
| "learning_rate": 0.00010515463917525774, | |
| "loss": 1.2266, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.48604005575180054, | |
| "learning_rate": 0.00010309278350515463, | |
| "loss": 1.0111, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0102, | |
| "grad_norm": 0.42038169503211975, | |
| "learning_rate": 0.00010103092783505155, | |
| "loss": 2.0069, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0104, | |
| "grad_norm": 0.41026800870895386, | |
| "learning_rate": 9.896907216494846e-05, | |
| "loss": 1.9782, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.0106, | |
| "grad_norm": 0.42623117566108704, | |
| "learning_rate": 9.690721649484537e-05, | |
| "loss": 2.1141, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.0108, | |
| "grad_norm": 0.4193218946456909, | |
| "learning_rate": 9.484536082474227e-05, | |
| "loss": 1.8971, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.011, | |
| "grad_norm": 0.43847334384918213, | |
| "learning_rate": 9.278350515463918e-05, | |
| "loss": 1.7533, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0112, | |
| "grad_norm": 0.46704810857772827, | |
| "learning_rate": 9.072164948453609e-05, | |
| "loss": 2.0783, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.0114, | |
| "grad_norm": 0.47083616256713867, | |
| "learning_rate": 8.865979381443299e-05, | |
| "loss": 2.2859, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0116, | |
| "grad_norm": 0.3852657079696655, | |
| "learning_rate": 8.65979381443299e-05, | |
| "loss": 1.6736, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.0118, | |
| "grad_norm": 0.4302668273448944, | |
| "learning_rate": 8.453608247422681e-05, | |
| "loss": 2.144, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.012, | |
| "grad_norm": 0.4399866759777069, | |
| "learning_rate": 8.247422680412371e-05, | |
| "loss": 2.2605, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0122, | |
| "grad_norm": 0.4025079905986786, | |
| "learning_rate": 8.041237113402063e-05, | |
| "loss": 1.9658, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.0124, | |
| "grad_norm": 0.4048742949962616, | |
| "learning_rate": 7.835051546391753e-05, | |
| "loss": 2.0939, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.0126, | |
| "grad_norm": 0.38197118043899536, | |
| "learning_rate": 7.628865979381443e-05, | |
| "loss": 1.974, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0128, | |
| "grad_norm": 0.423852801322937, | |
| "learning_rate": 7.422680412371135e-05, | |
| "loss": 1.8456, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.013, | |
| "grad_norm": 0.3875732123851776, | |
| "learning_rate": 7.216494845360825e-05, | |
| "loss": 1.8775, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.0132, | |
| "grad_norm": 0.400006502866745, | |
| "learning_rate": 7.010309278350515e-05, | |
| "loss": 1.9451, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0134, | |
| "grad_norm": 0.4780627191066742, | |
| "learning_rate": 6.804123711340207e-05, | |
| "loss": 2.1757, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.0136, | |
| "grad_norm": 0.4084922969341278, | |
| "learning_rate": 6.597938144329897e-05, | |
| "loss": 2.0411, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.0138, | |
| "grad_norm": 0.42807289958000183, | |
| "learning_rate": 6.391752577319587e-05, | |
| "loss": 2.3942, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.014, | |
| "grad_norm": 0.411408394575119, | |
| "learning_rate": 6.185567010309279e-05, | |
| "loss": 2.0569, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0142, | |
| "grad_norm": 0.3994385004043579, | |
| "learning_rate": 5.979381443298969e-05, | |
| "loss": 1.8811, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.0144, | |
| "grad_norm": 0.4302988648414612, | |
| "learning_rate": 5.7731958762886594e-05, | |
| "loss": 1.8287, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0146, | |
| "grad_norm": 0.43813201785087585, | |
| "learning_rate": 5.567010309278351e-05, | |
| "loss": 2.0936, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.0148, | |
| "grad_norm": 0.41323378682136536, | |
| "learning_rate": 5.360824742268041e-05, | |
| "loss": 1.8877, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 0.43512141704559326, | |
| "learning_rate": 5.1546391752577315e-05, | |
| "loss": 1.8419, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0152, | |
| "grad_norm": 0.46052488684654236, | |
| "learning_rate": 4.948453608247423e-05, | |
| "loss": 1.9927, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.0154, | |
| "grad_norm": 0.3917546272277832, | |
| "learning_rate": 4.7422680412371134e-05, | |
| "loss": 1.6892, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.0156, | |
| "grad_norm": 0.4031961262226105, | |
| "learning_rate": 4.536082474226804e-05, | |
| "loss": 1.7624, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0158, | |
| "grad_norm": 0.4103787839412689, | |
| "learning_rate": 4.329896907216495e-05, | |
| "loss": 1.8082, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.016, | |
| "grad_norm": 0.41262394189834595, | |
| "learning_rate": 4.1237113402061855e-05, | |
| "loss": 1.8105, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0162, | |
| "grad_norm": 0.41158345341682434, | |
| "learning_rate": 3.9175257731958764e-05, | |
| "loss": 1.8007, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0164, | |
| "grad_norm": 0.4416099786758423, | |
| "learning_rate": 3.7113402061855674e-05, | |
| "loss": 1.7926, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.0166, | |
| "grad_norm": 0.45428088307380676, | |
| "learning_rate": 3.5051546391752576e-05, | |
| "loss": 1.9448, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.0168, | |
| "grad_norm": 0.48291468620300293, | |
| "learning_rate": 3.2989690721649485e-05, | |
| "loss": 2.0836, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.017, | |
| "grad_norm": 0.46375176310539246, | |
| "learning_rate": 3.0927835051546395e-05, | |
| "loss": 1.9663, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.0172, | |
| "grad_norm": 0.4698142409324646, | |
| "learning_rate": 2.8865979381443297e-05, | |
| "loss": 1.8841, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.0174, | |
| "grad_norm": 0.48520392179489136, | |
| "learning_rate": 2.6804123711340206e-05, | |
| "loss": 1.9849, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0176, | |
| "grad_norm": 0.49598732590675354, | |
| "learning_rate": 2.4742268041237116e-05, | |
| "loss": 1.9452, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.0178, | |
| "grad_norm": 0.4195699691772461, | |
| "learning_rate": 2.268041237113402e-05, | |
| "loss": 1.3229, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.018, | |
| "grad_norm": 0.6012292504310608, | |
| "learning_rate": 2.0618556701030927e-05, | |
| "loss": 1.7777, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0182, | |
| "grad_norm": 0.44116055965423584, | |
| "learning_rate": 1.8556701030927837e-05, | |
| "loss": 1.5625, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.0184, | |
| "grad_norm": 0.5609015226364136, | |
| "learning_rate": 1.6494845360824743e-05, | |
| "loss": 1.9485, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.0186, | |
| "grad_norm": 0.5063101053237915, | |
| "learning_rate": 1.4432989690721649e-05, | |
| "loss": 1.5872, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.0188, | |
| "grad_norm": 0.45711398124694824, | |
| "learning_rate": 1.2371134020618558e-05, | |
| "loss": 1.4548, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.019, | |
| "grad_norm": 0.5572103261947632, | |
| "learning_rate": 1.0309278350515464e-05, | |
| "loss": 1.8041, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.0192, | |
| "grad_norm": 0.5026547908782959, | |
| "learning_rate": 8.247422680412371e-06, | |
| "loss": 1.4015, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.0194, | |
| "grad_norm": 0.5131281018257141, | |
| "learning_rate": 6.185567010309279e-06, | |
| "loss": 1.3779, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.0196, | |
| "grad_norm": 0.5810936689376831, | |
| "learning_rate": 4.123711340206186e-06, | |
| "loss": 1.4948, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.0198, | |
| "grad_norm": 0.5234789848327637, | |
| "learning_rate": 2.061855670103093e-06, | |
| "loss": 1.4101, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.5946288704872131, | |
| "learning_rate": 0.0, | |
| "loss": 1.1919, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 529353635839488.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |