| { | |
| "best_metric": 0.13208191096782684, | |
| "best_model_checkpoint": "saves/psy-course-info/Ministral-8B-Instruct-2410/train/fold1/checkpoint-70", | |
| "epoch": 4.938271604938271, | |
| "eval_steps": 10, | |
| "global_step": 125, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03950617283950617, | |
| "grad_norm": 2.0392611026763916, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.8023, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 1.9608478546142578, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.8556, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 2.205570697784424, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.8928, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 2.0169217586517334, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.9065, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 2.3407657146453857, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.9812, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 2.2169551849365234, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.802, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 1.6904208660125732, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.6066, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 1.587057113647461, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.5914, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.3649190664291382, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.513, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 1.4411005973815918, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.4441, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.4035516381263733, | |
| "eval_runtime": 20.7099, | |
| "eval_samples_per_second": 2.173, | |
| "eval_steps_per_second": 2.173, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 1.3062949180603027, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.3498, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 1.3776774406433105, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.3552, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 1.1252286434173584, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3254, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.8249286413192749, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.2193, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.972386360168457, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.2636, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.9211995005607605, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.2853, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.6744509935379028, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.171, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.8162732720375061, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.1822, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.6481461524963379, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.2107, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.7848970293998718, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.1788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.18836185336112976, | |
| "eval_runtime": 20.5343, | |
| "eval_samples_per_second": 2.191, | |
| "eval_steps_per_second": 2.191, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.6704379916191101, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.1674, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.6506070494651794, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1724, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.798513650894165, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.1568, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.6190807223320007, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.1723, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.7190565466880798, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.2087, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.7314072847366333, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.1588, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.7282370924949646, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1423, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.7284507155418396, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.2039, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.5586243867874146, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.1179, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.6185597777366638, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.1678, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.1535085290670395, | |
| "eval_runtime": 20.5339, | |
| "eval_samples_per_second": 2.192, | |
| "eval_steps_per_second": 2.192, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.6984698176383972, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1476, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.6719391345977783, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.1528, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.4915294051170349, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1493, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.6415784955024719, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.1129, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.5630152821540833, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.1423, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.8457736372947693, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.13, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.5293275117874146, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.0947, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.5480828285217285, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.1014, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.6636469960212708, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.1176, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.6313031911849976, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.1048, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.13932707905769348, | |
| "eval_runtime": 20.5439, | |
| "eval_samples_per_second": 2.19, | |
| "eval_steps_per_second": 2.19, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.7543724179267883, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.124, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.5054760575294495, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0757, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.606337308883667, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.1321, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.8879940509796143, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.0993, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.8698432445526123, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.1611, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.526561439037323, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.1087, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.6857826709747314, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.0907, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.5263163447380066, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.093, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.8170616030693054, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.1662, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.7097113728523254, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.1062, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.13875317573547363, | |
| "eval_runtime": 20.534, | |
| "eval_samples_per_second": 2.191, | |
| "eval_steps_per_second": 2.191, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.6411317586898804, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.106, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.49972841143608093, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.0672, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.5446736216545105, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.085, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.6477044224739075, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.1257, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.5469913482666016, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0976, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.619965136051178, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.0836, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.4521268308162689, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.059, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.5939432382583618, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.0735, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.4468541443347931, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0558, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.3871822655200958, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.064, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.1354343146085739, | |
| "eval_runtime": 20.5259, | |
| "eval_samples_per_second": 2.192, | |
| "eval_steps_per_second": 2.192, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.6722210049629211, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.0788, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.7275375723838806, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.1106, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.5644099116325378, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.1084, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.6620832085609436, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.1214, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.600744366645813, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.0753, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.5558615922927856, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.0733, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.8500597476959229, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.0789, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.6937291622161865, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.0794, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.5254883766174316, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0657, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.6716175079345703, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0644, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.13208191096782684, | |
| "eval_runtime": 20.5205, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 2.193, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.4559856653213501, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.061, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.7319750189781189, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.1086, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.5549283623695374, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0892, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.8358809351921082, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0917, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.6367297768592834, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0839, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.6357365250587463, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.0892, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.32559823989868164, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0413, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.5467402338981628, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0565, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.4324474036693573, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.061, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.5900253057479858, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0551, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.13344335556030273, | |
| "eval_runtime": 20.5121, | |
| "eval_samples_per_second": 2.194, | |
| "eval_steps_per_second": 2.194, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.466666579246521, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0552, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.41671082377433777, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0515, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.44352683424949646, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0482, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.6872423887252808, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.076, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.4340912997722626, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0561, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.5467881560325623, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.067, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.5328499674797058, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0587, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.365951269865036, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0335, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.5092597603797913, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0641, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.5500108003616333, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0525, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.13489142060279846, | |
| "eval_runtime": 20.5035, | |
| "eval_samples_per_second": 2.195, | |
| "eval_steps_per_second": 2.195, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.7681226134300232, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0821, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.775193452835083, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0808, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.7246379852294922, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0672, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.5509033799171448, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0592, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.44705384969711304, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0342, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.6264105439186096, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0526, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.6991032361984253, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0693, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.7212735414505005, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0497, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.5432716608047485, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0463, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.5329452157020569, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0615, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.1327144056558609, | |
| "eval_runtime": 20.4873, | |
| "eval_samples_per_second": 2.196, | |
| "eval_steps_per_second": 2.196, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.5892925262451172, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.0544, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.7355815768241882, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.0641, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.45382779836654663, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0331, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.46838146448135376, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0454, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.5255027413368225, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0368, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.41906651854515076, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0558, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.48942694067955017, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.052, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.782687246799469, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.062, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.6439091563224792, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0538, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.5299074649810791, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0367, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.13360659778118134, | |
| "eval_runtime": 20.4657, | |
| "eval_samples_per_second": 2.199, | |
| "eval_steps_per_second": 2.199, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.550788938999176, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0558, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.4954306185245514, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0486, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.3741088807582855, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0299, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.43452662229537964, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0407, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.5362085700035095, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0311, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.3964925706386566, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0296, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.47591346502304077, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.04, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.3618928790092468, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0371, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.4418576657772064, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.0365, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.5662739276885986, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0396, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.13423092663288116, | |
| "eval_runtime": 20.4566, | |
| "eval_samples_per_second": 2.2, | |
| "eval_steps_per_second": 2.2, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.5295315980911255, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0374, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.47646480798721313, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0565, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.4440726339817047, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.034, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.5347505211830139, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0522, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.5972172021865845, | |
| "learning_rate": 0.0, | |
| "loss": 0.0512, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.3014876311027712e+17, | |
| "train_loss": 0.14965366646647454, | |
| "train_runtime": 3648.9705, | |
| "train_samples_per_second": 0.555, | |
| "train_steps_per_second": 0.034 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 125, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3014876311027712e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |