| { | |
| "best_metric": 0.7156062722206116, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_60-samples_config-3/checkpoint-161", | |
| "epoch": 34.95652173913044, | |
| "eval_steps": 500, | |
| "global_step": 201, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 0.7016680836677551, | |
| "learning_rate": 1.3333333333333336e-07, | |
| "loss": 1.1058, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.7510942816734314, | |
| "learning_rate": 2.666666666666667e-07, | |
| "loss": 1.1306, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.7180184125900269, | |
| "learning_rate": 5.333333333333335e-07, | |
| "loss": 1.0751, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.0960289239883423, | |
| "eval_runtime": 12.9777, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.696365475654602, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.0591, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.7511110901832581, | |
| "learning_rate": 1.066666666666667e-06, | |
| "loss": 1.0881, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.6216956973075867, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.1004, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.0923012495040894, | |
| "eval_runtime": 12.9789, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.6951440572738647, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 1.0571, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.6762353777885437, | |
| "learning_rate": 1.8666666666666669e-06, | |
| "loss": 1.0138, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.7078174352645874, | |
| "learning_rate": 2.133333333333334e-06, | |
| "loss": 1.1552, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 1.0845402479171753, | |
| "eval_runtime": 12.9749, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.7293493747711182, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.0645, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.7216983437538147, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.0369, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.6637709140777588, | |
| "learning_rate": 2.9333333333333338e-06, | |
| "loss": 1.0884, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.0730719566345215, | |
| "eval_runtime": 12.9771, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.7498079538345337, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 1.0667, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.741428792476654, | |
| "learning_rate": 3.4666666666666672e-06, | |
| "loss": 1.0362, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.7704049348831177, | |
| "learning_rate": 3.7333333333333337e-06, | |
| "loss": 1.0984, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 1.0593057870864868, | |
| "eval_runtime": 12.9818, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.6860598921775818, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.022, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.7669686079025269, | |
| "learning_rate": 4.266666666666668e-06, | |
| "loss": 1.022, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.6506966948509216, | |
| "learning_rate": 4.533333333333334e-06, | |
| "loss": 1.054, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.0363248586654663, | |
| "eval_runtime": 12.9751, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.7007432579994202, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.9417, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.7524662613868713, | |
| "learning_rate": 5.0666666666666676e-06, | |
| "loss": 1.0873, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.6942258477210999, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.9646, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.0060027837753296, | |
| "eval_runtime": 12.9773, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.6106516718864441, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.9713, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.6019290089607239, | |
| "learning_rate": 5.8666666666666675e-06, | |
| "loss": 0.9628, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.5895516872406006, | |
| "learning_rate": 6.133333333333334e-06, | |
| "loss": 0.9982, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9700033664703369, | |
| "eval_runtime": 12.976, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.6009705066680908, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.9219, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.5637978315353394, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.9649, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.9379908442497253, | |
| "eval_runtime": 12.9746, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.552122175693512, | |
| "learning_rate": 6.9333333333333344e-06, | |
| "loss": 0.9063, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.5362454652786255, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.8733, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.4836583137512207, | |
| "learning_rate": 7.4666666666666675e-06, | |
| "loss": 0.9161, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.9016616344451904, | |
| "eval_runtime": 12.9785, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.4063497483730316, | |
| "learning_rate": 7.733333333333334e-06, | |
| "loss": 0.9077, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.36825188994407654, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8492, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.42374539375305176, | |
| "learning_rate": 8.266666666666667e-06, | |
| "loss": 0.8966, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.8721893429756165, | |
| "eval_runtime": 12.9792, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.39867985248565674, | |
| "learning_rate": 8.533333333333335e-06, | |
| "loss": 0.8529, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.3662341833114624, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.8509, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.31930872797966003, | |
| "learning_rate": 9.066666666666667e-06, | |
| "loss": 0.8314, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8468105792999268, | |
| "eval_runtime": 12.9771, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 0.3196379244327545, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.7984, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.37208741903305054, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.8613, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 0.333916574716568, | |
| "learning_rate": 9.866666666666668e-06, | |
| "loss": 0.7747, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.8285906314849854, | |
| "eval_runtime": 12.9738, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.3695715367794037, | |
| "learning_rate": 9.999945845889795e-06, | |
| "loss": 0.8164, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 0.3073723316192627, | |
| "learning_rate": 9.999512620046523e-06, | |
| "loss": 0.786, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.3305300176143646, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.8162, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.8081052899360657, | |
| "eval_runtime": 12.9758, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.26086956521739, | |
| "grad_norm": 0.2998150587081909, | |
| "learning_rate": 9.99734667851357e-06, | |
| "loss": 0.697, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.304190069437027, | |
| "learning_rate": 9.995614150494293e-06, | |
| "loss": 0.7373, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "grad_norm": 0.3205544948577881, | |
| "learning_rate": 9.993448771956285e-06, | |
| "loss": 0.8422, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 0.7906327843666077, | |
| "eval_runtime": 12.981, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.29707759618759155, | |
| "learning_rate": 9.99085073052117e-06, | |
| "loss": 0.7246, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 15.652173913043478, | |
| "grad_norm": 0.36958566308021545, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.7637, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.2745814919471741, | |
| "learning_rate": 9.984357596869369e-06, | |
| "loss": 0.7802, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.7775909304618835, | |
| "eval_runtime": 12.9815, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.347826086956523, | |
| "grad_norm": 0.3392060399055481, | |
| "learning_rate": 9.980463067257437e-06, | |
| "loss": 0.7707, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.2813577651977539, | |
| "learning_rate": 9.976136999909156e-06, | |
| "loss": 0.7179, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 16.869565217391305, | |
| "eval_loss": 0.7692317962646484, | |
| "eval_runtime": 12.9784, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 17.043478260869566, | |
| "grad_norm": 0.29758134484291077, | |
| "learning_rate": 9.971379769661422e-06, | |
| "loss": 0.7172, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.3041021525859833, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.7507, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "grad_norm": 0.29003143310546875, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 0.7191, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 17.91304347826087, | |
| "eval_loss": 0.7605112195014954, | |
| "eval_runtime": 12.9797, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 18.08695652173913, | |
| "grad_norm": 0.29067984223365784, | |
| "learning_rate": 9.95452541005172e-06, | |
| "loss": 0.7032, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 18.434782608695652, | |
| "grad_norm": 0.26193752884864807, | |
| "learning_rate": 9.948048023191728e-06, | |
| "loss": 0.7523, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "grad_norm": 0.30279889702796936, | |
| "learning_rate": 9.941141907232766e-06, | |
| "loss": 0.6644, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 18.956521739130434, | |
| "eval_loss": 0.7523689270019531, | |
| "eval_runtime": 12.9798, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 19.130434782608695, | |
| "grad_norm": 0.29118812084198, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.7396, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 19.47826086956522, | |
| "grad_norm": 0.3351486623287201, | |
| "learning_rate": 9.926045918666045e-06, | |
| "loss": 0.7189, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 19.82608695652174, | |
| "grad_norm": 0.300157755613327, | |
| "learning_rate": 9.91785735406693e-06, | |
| "loss": 0.6898, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.7455799579620361, | |
| "eval_runtime": 12.9769, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 20.17391304347826, | |
| "grad_norm": 0.28359395265579224, | |
| "learning_rate": 9.909242676272797e-06, | |
| "loss": 0.6711, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 20.52173913043478, | |
| "grad_norm": 0.2949537932872772, | |
| "learning_rate": 9.90020263171194e-06, | |
| "loss": 0.7176, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "grad_norm": 0.3114525079727173, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.6776, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "eval_loss": 0.740369975566864, | |
| "eval_runtime": 12.9813, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 21.217391304347824, | |
| "grad_norm": 0.2974396049976349, | |
| "learning_rate": 9.880849612217238e-06, | |
| "loss": 0.6606, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 21.565217391304348, | |
| "grad_norm": 0.3048945963382721, | |
| "learning_rate": 9.870538314147194e-06, | |
| "loss": 0.7063, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "grad_norm": 0.32696473598480225, | |
| "learning_rate": 9.859805002892733e-06, | |
| "loss": 0.6571, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "eval_loss": 0.7337734699249268, | |
| "eval_runtime": 12.9777, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 22.26086956521739, | |
| "grad_norm": 0.31726372241973877, | |
| "learning_rate": 9.84865060845349e-06, | |
| "loss": 0.7094, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 22.608695652173914, | |
| "grad_norm": 0.31174156069755554, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.6608, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "grad_norm": 0.31973257660865784, | |
| "learning_rate": 9.825082472361558e-06, | |
| "loss": 0.6177, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "eval_loss": 0.7288563847541809, | |
| "eval_runtime": 12.9756, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 23.304347826086957, | |
| "grad_norm": 0.36018624901771545, | |
| "learning_rate": 9.812670772796113e-06, | |
| "loss": 0.6802, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 23.652173913043477, | |
| "grad_norm": 0.30724474787712097, | |
| "learning_rate": 9.799842074043438e-06, | |
| "loss": 0.6582, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.3061894178390503, | |
| "learning_rate": 9.786597487660336e-06, | |
| "loss": 0.6361, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.724605143070221, | |
| "eval_runtime": 12.9779, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "grad_norm": 0.3347689211368561, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.6304, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 24.695652173913043, | |
| "grad_norm": 0.3372015058994293, | |
| "learning_rate": 9.75886527830587e-06, | |
| "loss": 0.6357, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 24.869565217391305, | |
| "eval_loss": 0.7214323878288269, | |
| "eval_runtime": 12.9829, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 25.043478260869566, | |
| "grad_norm": 0.3627227544784546, | |
| "learning_rate": 9.744380058222483e-06, | |
| "loss": 0.6527, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 25.391304347826086, | |
| "grad_norm": 0.32768282294273376, | |
| "learning_rate": 9.729483756076436e-06, | |
| "loss": 0.6396, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "grad_norm": 0.3657958209514618, | |
| "learning_rate": 9.714177662574316e-06, | |
| "loss": 0.6767, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 25.91304347826087, | |
| "eval_loss": 0.7174214720726013, | |
| "eval_runtime": 12.9804, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 0.38097909092903137, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.5935, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 26.434782608695652, | |
| "grad_norm": 0.3344641327857971, | |
| "learning_rate": 9.682341441747446e-06, | |
| "loss": 0.6088, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 26.782608695652176, | |
| "grad_norm": 0.35751739144325256, | |
| "learning_rate": 9.665814072907293e-06, | |
| "loss": 0.5947, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 26.956521739130434, | |
| "eval_loss": 0.7170318961143494, | |
| "eval_runtime": 12.9784, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 27.130434782608695, | |
| "grad_norm": 0.3365950584411621, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 0.6198, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 27.47826086956522, | |
| "grad_norm": 0.419979989528656, | |
| "learning_rate": 9.63154797841033e-06, | |
| "loss": 0.5974, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "grad_norm": 0.3486815392971039, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.6182, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.7156062722206116, | |
| "eval_runtime": 12.977, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 28.17391304347826, | |
| "grad_norm": 0.37861475348472595, | |
| "learning_rate": 9.595676696276173e-06, | |
| "loss": 0.5702, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 28.52173913043478, | |
| "grad_norm": 0.3779279589653015, | |
| "learning_rate": 9.577142973279896e-06, | |
| "loss": 0.5992, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "grad_norm": 0.4336070418357849, | |
| "learning_rate": 9.55821265866333e-06, | |
| "loss": 0.5899, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "eval_loss": 0.7157405018806458, | |
| "eval_runtime": 12.9775, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 29.217391304347824, | |
| "grad_norm": 0.37057119607925415, | |
| "learning_rate": 9.538887392664544e-06, | |
| "loss": 0.588, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 29.565217391304348, | |
| "grad_norm": 0.4328691065311432, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.5622, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "grad_norm": 0.4100225567817688, | |
| "learning_rate": 9.499058738432492e-06, | |
| "loss": 0.5612, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "eval_loss": 0.7161919474601746, | |
| "eval_runtime": 12.9765, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 30.26086956521739, | |
| "grad_norm": 0.4240186810493469, | |
| "learning_rate": 9.478558801197065e-06, | |
| "loss": 0.5735, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 30.608695652173914, | |
| "grad_norm": 0.43012985587120056, | |
| "learning_rate": 9.457670814276083e-06, | |
| "loss": 0.5602, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "grad_norm": 0.4227677881717682, | |
| "learning_rate": 9.436396587532297e-06, | |
| "loss": 0.5545, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "eval_loss": 0.7185451984405518, | |
| "eval_runtime": 12.9819, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 31.304347826086957, | |
| "grad_norm": 0.45781397819519043, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.519, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 31.652173913043477, | |
| "grad_norm": 0.4655354619026184, | |
| "learning_rate": 9.392696821198488e-06, | |
| "loss": 0.5702, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.5229057669639587, | |
| "learning_rate": 9.370275068023097e-06, | |
| "loss": 0.5574, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.7232272624969482, | |
| "eval_runtime": 12.9824, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 32.34782608695652, | |
| "grad_norm": 0.4893576204776764, | |
| "learning_rate": 9.347474647526095e-06, | |
| "loss": 0.5103, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "grad_norm": 0.5043169260025024, | |
| "learning_rate": 9.324297535275156e-06, | |
| "loss": 0.5316, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 32.869565217391305, | |
| "eval_loss": 0.725393533706665, | |
| "eval_runtime": 12.9763, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 33.04347826086956, | |
| "grad_norm": 0.6029254198074341, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 0.5299, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 33.391304347826086, | |
| "grad_norm": 0.4804014265537262, | |
| "learning_rate": 9.276821300802535e-06, | |
| "loss": 0.5041, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 33.73913043478261, | |
| "grad_norm": 0.5214328169822693, | |
| "learning_rate": 9.25252629221175e-06, | |
| "loss": 0.5276, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 33.91304347826087, | |
| "eval_loss": 0.7338247299194336, | |
| "eval_runtime": 12.98, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 34.08695652173913, | |
| "grad_norm": 0.5305810570716858, | |
| "learning_rate": 9.227862818772392e-06, | |
| "loss": 0.4906, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 34.43478260869565, | |
| "grad_norm": 0.5043832659721375, | |
| "learning_rate": 9.202833017478421e-06, | |
| "loss": 0.5014, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 0.7232289910316467, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.4653, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 34.95652173913044, | |
| "eval_loss": 0.7406532764434814, | |
| "eval_runtime": 12.9783, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 34.95652173913044, | |
| "step": 201, | |
| "total_flos": 1.5560139400124826e+17, | |
| "train_loss": 0.769242295281804, | |
| "train_runtime": 5595.4258, | |
| "train_samples_per_second": 1.233, | |
| "train_steps_per_second": 0.134 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5560139400124826e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |