| { | |
| "best_metric": 0.057771410793066025, | |
| "best_model_checkpoint": "saves/psy-course-doc/Llama-3.1-8B-Instruct/train/fold10/checkpoint-110", | |
| "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": 0.18021754920482635, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.2175, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 0.1665724813938141, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2108, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 0.18097135424613953, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.1637, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 0.16470956802368164, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.2571, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 0.17214342951774597, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.2045, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 0.17039048671722412, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.2308, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 0.18549735844135284, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.303, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 0.18758919835090637, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.2258, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.17439790070056915, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.1689, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 0.1339806616306305, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.1641, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.13725969195365906, | |
| "eval_runtime": 23.4094, | |
| "eval_samples_per_second": 1.922, | |
| "eval_steps_per_second": 1.922, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 0.1528996229171753, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.1572, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.1592380851507187, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.1704, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.16079406440258026, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1391, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.15372821688652039, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.103, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.14978763461112976, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.0821, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.1451999694108963, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.1356, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.140438973903656, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.1454, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.12988929450511932, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.1359, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.11539097875356674, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.1142, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.09883014857769012, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.0909, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.08548525720834732, | |
| "eval_runtime": 23.3429, | |
| "eval_samples_per_second": 1.928, | |
| "eval_steps_per_second": 1.928, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.11481934785842896, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.1143, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.10597186535596848, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1422, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.08763527125120163, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.1071, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.10206720978021622, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.07, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.10408683121204376, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.1037, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.1995173692703247, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.2778, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.09984336048364639, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0894, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.09152824431657791, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.0897, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.10456233471632004, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.129, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.1197863519191742, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.0908, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.07011803239583969, | |
| "eval_runtime": 23.335, | |
| "eval_samples_per_second": 1.928, | |
| "eval_steps_per_second": 1.928, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.10065857321023941, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.0848, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.11766823381185532, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.1377, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.09367826581001282, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1131, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.10166818648576736, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.0812, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.1027480959892273, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.1289, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.09376318007707596, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.1242, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.07300736010074615, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.0746, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.11094027757644653, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.0658, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.08517828583717346, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.0936, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.0935039073228836, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.118, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.0645856112241745, | |
| "eval_runtime": 23.3232, | |
| "eval_samples_per_second": 1.929, | |
| "eval_steps_per_second": 1.929, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.076003298163414, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0807, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.0890469178557396, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0599, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.09568687528371811, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.097, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.08569841086864471, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.0976, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.07479286938905716, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.046, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.11956445872783661, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.0621, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.0969395637512207, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.119, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.06882420927286148, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.056, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.09234333783388138, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.0638, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.0963311493396759, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.0867, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.061559222638607025, | |
| "eval_runtime": 23.3639, | |
| "eval_samples_per_second": 1.926, | |
| "eval_steps_per_second": 1.926, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.14070098102092743, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.1004, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.07476837933063507, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.0857, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.0877503752708435, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.0866, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.07462172210216522, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.0879, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.09036128968000412, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1084, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.07215496897697449, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.0665, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.10548751056194305, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.1744, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.07763306051492691, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.0593, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.08848672360181808, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0471, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.09191033989191055, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.0622, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.06012409180402756, | |
| "eval_runtime": 23.3184, | |
| "eval_samples_per_second": 1.93, | |
| "eval_steps_per_second": 1.93, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.08398331701755524, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.0668, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.069621242582798, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.043, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.06780290603637695, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.0253, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.08619555085897446, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0599, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.11270365118980408, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.1456, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.05919177457690239, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.0308, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.09078177809715271, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.1022, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.0790449008345604, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.0788, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.11816563457250595, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1171, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.08456794917583466, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0977, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.05951467156410217, | |
| "eval_runtime": 23.3282, | |
| "eval_samples_per_second": 1.929, | |
| "eval_steps_per_second": 1.929, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.0906873494386673, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.0724, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.10217013210058212, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.098, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.10844023525714874, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0678, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.09364091604948044, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.107, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.09139963984489441, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0784, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.14828111231327057, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.0437, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.10030606389045715, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0603, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.09666232019662857, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0927, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.08029602468013763, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0315, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.10753578692674637, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.1045, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.05842176452279091, | |
| "eval_runtime": 23.3221, | |
| "eval_samples_per_second": 1.93, | |
| "eval_steps_per_second": 1.93, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.0931696742773056, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0556, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.08699537068605423, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0459, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.09282850474119186, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0903, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.08753346651792526, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0976, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.08252975344657898, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0864, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.11231781542301178, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.1066, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.09453340619802475, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0855, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.0992935374379158, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.1148, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.09019789844751358, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0503, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.08707164973020554, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0473, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.058295316994190216, | |
| "eval_runtime": 23.3947, | |
| "eval_samples_per_second": 1.924, | |
| "eval_steps_per_second": 1.924, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.10828964412212372, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0753, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.11449547857046127, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.1041, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.06713505834341049, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0359, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.11314048618078232, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0785, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.11881639063358307, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0876, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.09872547537088394, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0536, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.06874336302280426, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0266, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.11092659085988998, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.1177, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.0744275152683258, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.052, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.10704729706048965, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0568, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.05799093842506409, | |
| "eval_runtime": 23.3343, | |
| "eval_samples_per_second": 1.928, | |
| "eval_steps_per_second": 1.928, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.14050760865211487, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.1122, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.1410040259361267, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.0723, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.09531880915164948, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.049, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.09136135131120682, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0522, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.08001089841127396, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0524, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.08216973394155502, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0439, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.11158059537410736, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.1087, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.06513264030218124, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0197, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.11665146052837372, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.1011, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.08367304503917694, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0772, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.057771410793066025, | |
| "eval_runtime": 23.3467, | |
| "eval_samples_per_second": 1.927, | |
| "eval_steps_per_second": 1.927, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.07535257935523987, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0459, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.1000489816069603, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.066, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.07393460720777512, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0326, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.09692024439573288, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0878, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.07969095557928085, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0452, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.09920325130224228, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0888, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.11619863659143448, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.1302, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.110574871301651, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.1038, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.13501973450183868, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.1353, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.10522457957267761, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0765, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.05783239006996155, | |
| "eval_runtime": 23.395, | |
| "eval_samples_per_second": 1.923, | |
| "eval_steps_per_second": 1.923, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.10014230012893677, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0771, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.06939004361629486, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0218, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.07230408489704132, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0423, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.08347835391759872, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0346, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.09913475066423416, | |
| "learning_rate": 0.0, | |
| "loss": 0.0806, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.4964997728396902e+17, | |
| "train_loss": 0.09641678650677205, | |
| "train_runtime": 3727.2923, | |
| "train_samples_per_second": 0.543, | |
| "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": 1.4964997728396902e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |