| { | |
| "best_metric": 0.155701145529747, | |
| "best_model_checkpoint": "saves/psy-course-info/Llama-3.1-8B-Instruct/train/fold4/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": 2.194518804550171, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 1.0283, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 2.1176815032958984, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0506, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 2.350534200668335, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 1.1433, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 2.3229384422302246, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 1.1949, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 2.4669547080993652, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.1299, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 2.586879014968872, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.8563, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 2.568798065185547, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 1.0405, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 2.2546226978302, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.8014, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 2.58187198638916, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.8424, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 1.5734577178955078, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.481, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.4470711648464203, | |
| "eval_runtime": 20.9596, | |
| "eval_samples_per_second": 2.147, | |
| "eval_steps_per_second": 2.147, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 1.2128584384918213, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.4775, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.8648531436920166, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.4034, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 1.3788645267486572, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5011, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.8829809427261353, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.276, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 1.1226744651794434, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.4457, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.7899404764175415, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.3465, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.6820979714393616, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.3289, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.7585280537605286, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.3761, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.5556255578994751, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.2452, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.642232358455658, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.227, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.26709282398223877, | |
| "eval_runtime": 20.7787, | |
| "eval_samples_per_second": 2.166, | |
| "eval_steps_per_second": 2.166, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.7405572533607483, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.2376, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.7726360559463501, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1854, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.6125112771987915, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.2192, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.5291657447814941, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.2319, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.5333792567253113, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.2028, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.8214161396026611, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.3072, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.5902807116508484, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.271, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.4466434121131897, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1567, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.4218069016933441, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.1632, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.6221491098403931, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.1733, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.19363102316856384, | |
| "eval_runtime": 20.8028, | |
| "eval_samples_per_second": 2.163, | |
| "eval_steps_per_second": 2.163, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.5914718508720398, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1592, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.5006762742996216, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.177, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.570541501045227, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1633, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.8022737503051758, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.206, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.8812099099159241, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.1768, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.6670054793357849, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.1688, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.5354920029640198, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.181, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.5447942614555359, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.1782, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.6054803133010864, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.187, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.7443354725837708, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.1622, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.17835888266563416, | |
| "eval_runtime": 20.8179, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.6073779463768005, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1902, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.6843507289886475, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.2264, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.49217379093170166, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.1347, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.47513994574546814, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.1763, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.368058443069458, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.0792, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.4543582499027252, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.1573, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.5805642008781433, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.1807, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.6300663948059082, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1542, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.5194098353385925, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.1581, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.5455979108810425, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.1527, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.1648695170879364, | |
| "eval_runtime": 20.8121, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 1.3302276134490967, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.3565, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.43565472960472107, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.1353, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.480875700712204, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.1243, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.48461875319480896, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.0971, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.5279831886291504, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1223, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.4635251462459564, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.1153, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.47293245792388916, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.1522, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.5724309682846069, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1439, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.5045034289360046, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.1376, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.5327277779579163, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.1002, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.1610100269317627, | |
| "eval_runtime": 20.8074, | |
| "eval_samples_per_second": 2.163, | |
| "eval_steps_per_second": 2.163, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.5841977000236511, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.1289, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.5795934796333313, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.1142, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.6151239275932312, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.1144, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.6791741251945496, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.1314, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.5318293571472168, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.1415, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.5658901333808899, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.1278, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.5097365975379944, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.0888, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.5901567339897156, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.1541, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.6255803108215332, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0974, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.44344496726989746, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0798, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.155701145529747, | |
| "eval_runtime": 20.8127, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.6580880284309387, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.1337, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.46482139825820923, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0847, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.5961633920669556, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.1053, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.5351518988609314, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.1274, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.5729773044586182, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.1015, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.8544556498527527, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1564, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.3593759536743164, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0591, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.4639565348625183, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0951, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.5681014060974121, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0883, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.6761187314987183, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0812, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.1632390022277832, | |
| "eval_runtime": 20.8096, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.47050565481185913, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0762, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.4817729592323303, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0641, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.5061681270599365, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0777, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.6684750318527222, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.1087, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.5278833508491516, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0912, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.6312245726585388, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0951, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.6985181570053101, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0889, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.6942794322967529, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.1148, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.7768173217773438, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.1133, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.820288896560669, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0872, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.16232681274414062, | |
| "eval_runtime": 20.816, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.6442907452583313, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0784, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.5543602705001831, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0957, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.7335277795791626, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.1067, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.6988636255264282, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.1022, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.7920371890068054, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0715, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.5601813793182373, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0722, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.6194542050361633, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0974, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.546447217464447, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0849, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.5601685643196106, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0751, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.5766252279281616, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.079, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.1592065542936325, | |
| "eval_runtime": 20.8082, | |
| "eval_samples_per_second": 2.163, | |
| "eval_steps_per_second": 2.163, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 1.0218473672866821, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.0927, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 1.091865062713623, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.1495, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.479623943567276, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0718, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.5074496865272522, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0626, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.4844929575920105, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0774, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.3479042649269104, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0411, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.5858511328697205, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0984, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.4459031820297241, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0601, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.6543456315994263, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0943, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.5373428463935852, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0801, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.1588951051235199, | |
| "eval_runtime": 20.813, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.5837863087654114, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0776, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.6311441659927368, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0686, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.5876055955886841, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.075, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.536090075969696, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0712, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.48177963495254517, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0607, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.6054865717887878, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.077, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.5703210830688477, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0955, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.5592256188392639, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0504, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.5410389304161072, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.084, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.6372770071029663, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0932, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.15954476594924927, | |
| "eval_runtime": 20.8113, | |
| "eval_samples_per_second": 2.162, | |
| "eval_steps_per_second": 2.162, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.42266181111335754, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0586, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.5649014115333557, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0747, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.4727138578891754, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0621, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.7124543190002441, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0657, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.3849957585334778, | |
| "learning_rate": 0.0, | |
| "loss": 0.0333, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.2628766887457587e+17, | |
| "train_loss": 0.20852369618415834, | |
| "train_runtime": 3506.5823, | |
| "train_samples_per_second": 0.577, | |
| "train_steps_per_second": 0.036 | |
| } | |
| ], | |
| "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": 1.2628766887457587e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |