| { | |
| "best_metric": 0.15355931222438812, | |
| "best_model_checkpoint": "saves/psy-course-info/MentaLLaMA-chat-7B/train/fold5/checkpoint-90", | |
| "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": 0.9562548995018005, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 1.0442, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 0.8733327388763428, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.1092, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 1.0652928352355957, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 1.1194, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 0.9854899048805237, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 1.1127, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 1.0494587421417236, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.111, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 1.0473530292510986, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.9938, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 1.1149609088897705, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.9941, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 1.2149229049682617, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.9588, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.3304028511047363, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 1.0645, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 1.3816691637039185, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.7894, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.6198992729187012, | |
| "eval_runtime": 24.8341, | |
| "eval_samples_per_second": 1.812, | |
| "eval_steps_per_second": 1.812, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 1.1097314357757568, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.6047, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.8636482954025269, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.5891, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.7059842944145203, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4594, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.5071170926094055, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.4696, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.5149194002151489, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.3311, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.7308483719825745, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.4072, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.5209848284721375, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.2868, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.49168670177459717, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.3598, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.43376973271369934, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.3178, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.42548102140426636, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.3067, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.26275914907455444, | |
| "eval_runtime": 24.8286, | |
| "eval_samples_per_second": 1.812, | |
| "eval_steps_per_second": 1.812, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.46806931495666504, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.2804, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.5101025104522705, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1882, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.47250258922576904, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.2337, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.504750669002533, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.2085, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.40929704904556274, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.2143, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.6398062705993652, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.3534, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.3896842300891876, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1758, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.3708793818950653, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.2061, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.385396808385849, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.216, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.3281359076499939, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.1744, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.19748829305171967, | |
| "eval_runtime": 24.829, | |
| "eval_samples_per_second": 1.812, | |
| "eval_steps_per_second": 1.812, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.2807298004627228, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1131, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.46649572253227234, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.193, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.4385913908481598, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1889, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.4009336233139038, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.2312, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.38049182295799255, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.161, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.2896342873573303, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.138, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.3436133563518524, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.1347, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.41206976771354675, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.2242, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.4409872591495514, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.1486, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.3228354752063751, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.1489, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.17217481136322021, | |
| "eval_runtime": 24.861, | |
| "eval_samples_per_second": 1.81, | |
| "eval_steps_per_second": 1.81, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.37731462717056274, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1571, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.4434247612953186, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.1996, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.30294740200042725, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.1422, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.3518632650375366, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.1618, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.2825823724269867, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.151, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.34738558530807495, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.16, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.37332749366760254, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.149, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.37660354375839233, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1955, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.36386626958847046, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.1553, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.37852463126182556, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.1622, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.16764000058174133, | |
| "eval_runtime": 24.8815, | |
| "eval_samples_per_second": 1.809, | |
| "eval_steps_per_second": 1.809, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.831751823425293, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.2592, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.32533928751945496, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.1291, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.258903831243515, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.0939, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.32664772868156433, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.1463, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.39440107345581055, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1058, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.46934986114501953, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.1663, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.3761884272098541, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.126, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.37222376465797424, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1716, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.37241289019584656, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.1465, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.3771013021469116, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.1611, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.1615438312292099, | |
| "eval_runtime": 24.887, | |
| "eval_samples_per_second": 1.808, | |
| "eval_steps_per_second": 1.808, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.2557624578475952, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.1018, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.35138729214668274, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.1299, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.3328498303890228, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.12, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.32999247312545776, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0977, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.387717604637146, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.1286, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.27463188767433167, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.1151, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.30502501130104065, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.1191, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.2334824651479721, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.0841, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.3771997094154358, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1659, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.38378649950027466, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.1067, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.1636304259300232, | |
| "eval_runtime": 24.8745, | |
| "eval_samples_per_second": 1.809, | |
| "eval_steps_per_second": 1.809, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.39505815505981445, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.1067, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.39354029297828674, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.1398, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.3442412316799164, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.1148, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.31601643562316895, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.1247, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.45984143018722534, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.156, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.625750720500946, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1434, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.3247933089733124, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0838, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.30317744612693787, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.1249, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.44227707386016846, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.1026, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.310411661863327, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0909, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.15589037537574768, | |
| "eval_runtime": 24.8714, | |
| "eval_samples_per_second": 1.809, | |
| "eval_steps_per_second": 1.809, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.42356765270233154, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.1104, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.3408052325248718, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.1093, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.3002070486545563, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0927, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.2932652533054352, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0963, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.3612593412399292, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.1065, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.3108547329902649, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.1075, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.2927524745464325, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.1093, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.27479830384254456, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0872, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.3041783571243286, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.1009, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.29193615913391113, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0873, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.15355931222438812, | |
| "eval_runtime": 24.8991, | |
| "eval_samples_per_second": 1.807, | |
| "eval_steps_per_second": 1.807, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.2939348816871643, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0877, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.26824069023132324, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0749, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.2405591458082199, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0745, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.40852129459381104, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.1235, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.40543147921562195, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.1317, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.2982388436794281, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0964, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.3460067808628082, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1353, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.43082395195961, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.1526, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.329779714345932, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0879, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.32258978486061096, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0842, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.15459902584552765, | |
| "eval_runtime": 24.8843, | |
| "eval_samples_per_second": 1.808, | |
| "eval_steps_per_second": 1.808, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.3776220679283142, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.1017, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.9673794507980347, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.1614, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.3300073742866516, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0824, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.2507697343826294, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0638, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.26608824729919434, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0792, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.29626861214637756, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0932, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.3560323119163513, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0853, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.29537203907966614, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0867, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.32487908005714417, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.1084, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.31251293420791626, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0893, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.15510736405849457, | |
| "eval_runtime": 24.9161, | |
| "eval_samples_per_second": 1.806, | |
| "eval_steps_per_second": 1.806, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.28879591822624207, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0718, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.37039247155189514, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.1311, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.30211618542671204, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0756, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.3424753248691559, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.087, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.3468068838119507, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0901, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.3149239718914032, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0835, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.3526304364204407, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0946, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.36930376291275024, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.1277, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.37969374656677246, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.1015, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.37840989232063293, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.1005, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.15540145337581635, | |
| "eval_runtime": 24.9415, | |
| "eval_samples_per_second": 1.804, | |
| "eval_steps_per_second": 1.804, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.35852643847465515, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.1087, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.3703908324241638, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0832, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.4437415897846222, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0838, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.35072624683380127, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.1182, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.3396915793418884, | |
| "learning_rate": 0.0, | |
| "loss": 0.078, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.3713455836476211e+17, | |
| "train_loss": 0.22640113997459413, | |
| "train_runtime": 4002.4779, | |
| "train_samples_per_second": 0.506, | |
| "train_steps_per_second": 0.031 | |
| } | |
| ], | |
| "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": 1.3713455836476211e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |