| { | |
| "best_metric": 0.06968548893928528, | |
| "best_model_checkpoint": "saves/psy-course-doc/Llama-3.1-8B-Instruct/train/fold4/checkpoint-100", | |
| "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": 0.1788543462753296, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.2003, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 0.15522800385951996, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2058, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 0.16844213008880615, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.2201, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 0.17402788996696472, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.1343, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 0.17550528049468994, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.2522, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 0.18190859258174896, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.1913, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 0.19788743555545807, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.2414, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 0.17828775942325592, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.1717, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.17290732264518738, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.1929, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 0.14847496151924133, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.1379, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.15000608563423157, | |
| "eval_runtime": 21.538, | |
| "eval_samples_per_second": 2.089, | |
| "eval_steps_per_second": 2.089, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 0.1618112325668335, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.1544, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.15634895861148834, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.1459, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.17232589423656464, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1712, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.17381250858306885, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.1528, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.1521753966808319, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.143, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.16353924572467804, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.1105, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.13930287957191467, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.1906, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.12930214405059814, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.1686, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.11505807191133499, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.0906, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.1082131415605545, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.1556, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.09602971374988556, | |
| "eval_runtime": 21.6977, | |
| "eval_samples_per_second": 2.074, | |
| "eval_steps_per_second": 2.074, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.13616810739040375, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.1162, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.09093806892633438, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.063, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.09628450870513916, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.1232, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.11707480251789093, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.1204, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.08480218052864075, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.0971, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.17484459280967712, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.1681, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.11269538104534149, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1341, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.09481585770845413, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1004, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.0998338907957077, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.0935, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.11120180785655975, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.1246, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.08139553666114807, | |
| "eval_runtime": 21.6913, | |
| "eval_samples_per_second": 2.075, | |
| "eval_steps_per_second": 2.075, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.11669044196605682, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1211, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.09925449639558792, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.0678, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.07805059105157852, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.0445, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.10079403221607208, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.0958, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.080256387591362, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.0373, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.09927237033843994, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.0946, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.10970957577228546, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.1434, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.107501320540905, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.092, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.08530978113412857, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.0973, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.09136773645877838, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.0919, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.07570390403270721, | |
| "eval_runtime": 21.7202, | |
| "eval_samples_per_second": 2.072, | |
| "eval_steps_per_second": 2.072, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.10633984953165054, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1092, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.07765901833772659, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0614, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.08163449913263321, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.0597, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.0758323222398758, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.071, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.093803770840168, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.0772, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.0847662165760994, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.121, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.08161197602748871, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.1075, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.0909591019153595, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1148, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.07414304465055466, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.057, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.07943868637084961, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.0345, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.07385429739952087, | |
| "eval_runtime": 21.7488, | |
| "eval_samples_per_second": 2.069, | |
| "eval_steps_per_second": 2.069, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.15582451224327087, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.1573, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.08236896991729736, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.1171, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.08111395686864853, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.094, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.0707816332578659, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.0507, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.06753059476613998, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0631, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.07619708776473999, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.0878, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.09822479635477066, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.1414, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.09035561978816986, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1415, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.07979875057935715, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0519, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.08112189918756485, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.0412, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.0714641660451889, | |
| "eval_runtime": 21.7698, | |
| "eval_samples_per_second": 2.067, | |
| "eval_steps_per_second": 2.067, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.08404616266489029, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.064, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.07832326740026474, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0501, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.08300691843032837, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.0601, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.09299349784851074, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0827, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.09701690077781677, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.1216, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.09380342811346054, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.0541, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.08833695203065872, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.109, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.0866917073726654, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.059, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.09058818221092224, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0615, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.08380809426307678, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0616, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.07049601525068283, | |
| "eval_runtime": 21.7497, | |
| "eval_samples_per_second": 2.069, | |
| "eval_steps_per_second": 2.069, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.08423994481563568, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.0343, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.09435715526342392, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0665, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.09836889803409576, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0583, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.10150245577096939, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0799, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.11149924248456955, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.1349, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.1692681908607483, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1208, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.08621936291456223, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0689, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.0897722989320755, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0939, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.09457165002822876, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0973, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.11529839783906937, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.1267, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.07009167969226837, | |
| "eval_runtime": 21.7772, | |
| "eval_samples_per_second": 2.066, | |
| "eval_steps_per_second": 2.066, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.09495740383863449, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0922, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.08246947079896927, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0353, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.09393060952425003, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.084, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.0827101618051529, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0517, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.08497904986143112, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0573, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.10192100703716278, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0945, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.09473208338022232, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0732, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.10408113151788712, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0865, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.1016639694571495, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0642, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.11952732503414154, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0488, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.07015260308980942, | |
| "eval_runtime": 21.8065, | |
| "eval_samples_per_second": 2.064, | |
| "eval_steps_per_second": 2.064, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.12438849359750748, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.1102, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.09344843029975891, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0846, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.09065958112478256, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0617, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.1886902004480362, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0522, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.09225163608789444, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.048, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.07945277541875839, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0122, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.11571206152439117, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0893, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.1022576168179512, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0564, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.11757322400808334, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.1029, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.10314515978097916, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0548, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.06968548893928528, | |
| "eval_runtime": 21.8041, | |
| "eval_samples_per_second": 2.064, | |
| "eval_steps_per_second": 2.064, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.0980178639292717, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.0534, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.16884902119636536, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.1559, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.09049151837825775, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.047, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.07266122847795486, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0305, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.1037364974617958, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0883, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.06758195906877518, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0217, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.12144804745912552, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.1045, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.07703778892755508, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0295, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.10785632580518723, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0831, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.10297554731369019, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0773, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.06995988637208939, | |
| "eval_runtime": 21.8259, | |
| "eval_samples_per_second": 2.062, | |
| "eval_steps_per_second": 2.062, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.09017482399940491, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0839, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.09495143592357635, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0545, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.09739414602518082, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.07, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.09476127475500107, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0704, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.09785457700490952, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0694, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.11750278621912003, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.1076, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.10429148375988007, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0823, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.07405687868595123, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0269, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.11975529789924622, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.1154, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.09624066203832626, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0525, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.06996127963066101, | |
| "eval_runtime": 21.7895, | |
| "eval_samples_per_second": 2.065, | |
| "eval_steps_per_second": 2.065, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.08022435754537582, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0431, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.10740126669406891, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0665, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.09020163118839264, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0601, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.0656752958893776, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0108, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.09660234302282333, | |
| "learning_rate": 0.0, | |
| "loss": 0.0561, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.5152446730030285e+17, | |
| "train_loss": 0.0947436870187521, | |
| "train_runtime": 3639.791, | |
| "train_samples_per_second": 0.556, | |
| "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": 1.5152446730030285e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |