| { | |
| "best_metric": 0.08806045353412628, | |
| "best_model_checkpoint": "saves/psy-course-info-chain/Llama3-OpenBioLLM-8B/train/fold6/checkpoint-80", | |
| "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": 1.2098701000213623, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.3537, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 1.3167098760604858, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.3394, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 1.1081513166427612, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.3491, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 1.3030786514282227, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.3357, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 1.3628971576690674, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.3759, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 1.2690974473953247, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.335, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 1.0661746263504028, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.254, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 1.0394080877304077, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.2655, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.032701849937439, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.1916, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 0.9631877541542053, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.2109, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.17117230594158173, | |
| "eval_runtime": 20.2136, | |
| "eval_samples_per_second": 2.226, | |
| "eval_steps_per_second": 2.226, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 0.9776226878166199, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.2453, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.9095515012741089, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.1919, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.7622648477554321, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1735, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 1.0270005464553833, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.2146, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.8578253388404846, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.1702, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.7143718004226685, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.1581, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.5231890678405762, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.1064, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.6164197325706482, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.1413, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.4978383481502533, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.1006, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.5567417740821838, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.1534, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.11690770089626312, | |
| "eval_runtime": 20.1725, | |
| "eval_samples_per_second": 2.231, | |
| "eval_steps_per_second": 2.231, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.5125220417976379, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.1322, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.5450965762138367, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1737, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.6283867955207825, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.0818, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.5701476335525513, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.1366, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.5162869691848755, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.1236, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.9549514651298523, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.1744, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.5940822958946228, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0781, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.5215325951576233, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.094, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.6191585063934326, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.1205, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.6245893239974976, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.128, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.10279463976621628, | |
| "eval_runtime": 20.1937, | |
| "eval_samples_per_second": 2.228, | |
| "eval_steps_per_second": 2.228, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.6656838059425354, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1184, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.41577374935150146, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.0845, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.42081260681152344, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.076, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.5547986626625061, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.0776, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.5910443067550659, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.1184, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.5948926210403442, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.1289, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.4887751638889313, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.0823, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.5978832244873047, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.0753, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.6041213870048523, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.1244, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.4581482410430908, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.0865, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.09400840103626251, | |
| "eval_runtime": 20.2025, | |
| "eval_samples_per_second": 2.227, | |
| "eval_steps_per_second": 2.227, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.49227094650268555, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0546, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.4849826991558075, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0701, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.6303190588951111, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.0881, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.48153242468833923, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.039, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.5066944360733032, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.0732, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.5285928249359131, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.0785, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.6026076674461365, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.091, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.5043290853500366, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.0674, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.8468377590179443, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.1303, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.6623268127441406, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.1088, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.0900571271777153, | |
| "eval_runtime": 20.1992, | |
| "eval_samples_per_second": 2.228, | |
| "eval_steps_per_second": 2.228, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.9407477974891663, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.1363, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.3677526116371155, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.0586, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.3780328929424286, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.0593, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.38486793637275696, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.0723, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.34154024720191956, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0651, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.5469534993171692, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.0825, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.37920835614204407, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.0531, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.4694402515888214, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.0764, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.34996798634529114, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0517, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.4056988060474396, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.076, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.08984256535768509, | |
| "eval_runtime": 20.2019, | |
| "eval_samples_per_second": 2.228, | |
| "eval_steps_per_second": 2.228, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.42485472559928894, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.0499, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.5835949182510376, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0465, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.5139871835708618, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.074, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.6129758954048157, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0406, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.499253511428833, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.0672, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.5918835997581482, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.109, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.5507199764251709, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.0634, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.3790760934352875, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.039, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.9641798138618469, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0633, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.48686787486076355, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0628, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.09097646176815033, | |
| "eval_runtime": 20.1965, | |
| "eval_samples_per_second": 2.228, | |
| "eval_steps_per_second": 2.228, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.6990075707435608, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.052, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.6660770177841187, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0734, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.5313035845756531, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.052, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.7404530644416809, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0808, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.7202139496803284, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0532, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 1.460814118385315, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1364, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.7142069935798645, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0488, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.3973197638988495, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0282, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.4896116554737091, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0508, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.4304870069026947, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0255, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.08806045353412628, | |
| "eval_runtime": 20.2325, | |
| "eval_samples_per_second": 2.224, | |
| "eval_steps_per_second": 2.224, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.7037865519523621, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0469, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.5505902171134949, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0606, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.47218337655067444, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0557, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.4788568615913391, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0394, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.5414381623268127, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0435, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.4069317579269409, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0331, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.6086376309394836, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0589, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.4970571994781494, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.069, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.46633851528167725, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.059, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.6050962209701538, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0582, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.09193442761898041, | |
| "eval_runtime": 20.2356, | |
| "eval_samples_per_second": 2.224, | |
| "eval_steps_per_second": 2.224, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.4856192171573639, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0587, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.6240473389625549, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0475, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.3966387212276459, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0344, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.3751712143421173, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0344, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.4704873561859131, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0455, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.30209898948669434, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0268, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.4116058051586151, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0397, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.34586301445961, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0489, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.4923393428325653, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0538, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.604167103767395, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0389, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.08921382576227188, | |
| "eval_runtime": 20.2385, | |
| "eval_samples_per_second": 2.223, | |
| "eval_steps_per_second": 2.223, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.9553114175796509, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.0802, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.9266524314880371, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.0618, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.3909324109554291, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0415, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.2993374764919281, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0211, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.37430423498153687, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0396, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.48327693343162537, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0471, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.38522160053253174, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0301, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.4199317991733551, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0377, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.40575993061065674, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0261, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.3058774769306183, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0396, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.09071807563304901, | |
| "eval_runtime": 20.232, | |
| "eval_samples_per_second": 2.224, | |
| "eval_steps_per_second": 2.224, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.4861699938774109, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0334, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.48257943987846375, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0511, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.45548227429389954, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0498, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.30992263555526733, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0292, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.4637013077735901, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0415, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.5105428695678711, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0444, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.42670539021492004, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0463, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.4397276043891907, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0348, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.3670833110809326, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.0238, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.3886328339576721, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0371, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.09216072410345078, | |
| "eval_runtime": 20.2084, | |
| "eval_samples_per_second": 2.227, | |
| "eval_steps_per_second": 2.227, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.4576190412044525, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0484, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.37236690521240234, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0347, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.4739607572555542, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0533, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.3876991868019104, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0313, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.34178659319877625, | |
| "learning_rate": 0.0, | |
| "loss": 0.0341, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.3555716469633843e+17, | |
| "train_loss": 0.09280947175621987, | |
| "train_runtime": 3762.3288, | |
| "train_samples_per_second": 0.538, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "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": 1.3555716469633843e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |