| { | |
| "best_metric": 0.8136550784111023, | |
| "best_model_checkpoint": "outputs/checkpoint-92", | |
| "epoch": 1.0, | |
| "eval_steps": 23, | |
| "global_step": 92, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010869565217391304, | |
| "grad_norm": 0.6224715709686279, | |
| "learning_rate": 0.001, | |
| "loss": 3.1291, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021739130434782608, | |
| "grad_norm": 1.0227481126785278, | |
| "learning_rate": 0.001, | |
| "loss": 2.5435, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03260869565217391, | |
| "grad_norm": 0.5493069887161255, | |
| "learning_rate": 0.001, | |
| "loss": 2.2639, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.043478260869565216, | |
| "grad_norm": 1.1812320947647095, | |
| "learning_rate": 0.001, | |
| "loss": 2.0138, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05434782608695652, | |
| "grad_norm": 0.5967501401901245, | |
| "learning_rate": 0.001, | |
| "loss": 1.8423, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06521739130434782, | |
| "grad_norm": 0.4586184620857239, | |
| "learning_rate": 0.001, | |
| "loss": 1.6938, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07608695652173914, | |
| "grad_norm": 0.3564240038394928, | |
| "learning_rate": 0.001, | |
| "loss": 1.6358, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.08695652173913043, | |
| "grad_norm": 0.3485143482685089, | |
| "learning_rate": 0.001, | |
| "loss": 1.4822, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09782608695652174, | |
| "grad_norm": 0.26243680715560913, | |
| "learning_rate": 0.001, | |
| "loss": 1.4112, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "grad_norm": 0.24434934556484222, | |
| "learning_rate": 0.001, | |
| "loss": 1.3732, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11956521739130435, | |
| "grad_norm": 0.2579119801521301, | |
| "learning_rate": 0.001, | |
| "loss": 1.3184, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.13043478260869565, | |
| "grad_norm": 0.3096484839916229, | |
| "learning_rate": 0.001, | |
| "loss": 1.2864, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.14130434782608695, | |
| "grad_norm": 0.34087467193603516, | |
| "learning_rate": 0.001, | |
| "loss": 1.3093, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.15217391304347827, | |
| "grad_norm": 0.4372682571411133, | |
| "learning_rate": 0.001, | |
| "loss": 1.2093, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.16304347826086957, | |
| "grad_norm": 0.36609384417533875, | |
| "learning_rate": 0.001, | |
| "loss": 1.1981, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 0.23835352063179016, | |
| "learning_rate": 0.001, | |
| "loss": 1.2141, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.18478260869565216, | |
| "grad_norm": 0.23766446113586426, | |
| "learning_rate": 0.001, | |
| "loss": 1.1933, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1956521739130435, | |
| "grad_norm": 0.19143910706043243, | |
| "learning_rate": 0.001, | |
| "loss": 1.1518, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20652173913043478, | |
| "grad_norm": 0.17653922736644745, | |
| "learning_rate": 0.001, | |
| "loss": 1.1841, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 0.197570338845253, | |
| "learning_rate": 0.001, | |
| "loss": 1.1515, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22826086956521738, | |
| "grad_norm": 0.22776907682418823, | |
| "learning_rate": 0.001, | |
| "loss": 1.0765, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2391304347826087, | |
| "grad_norm": 0.22448290884494781, | |
| "learning_rate": 0.001, | |
| "loss": 1.0842, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.18711359798908234, | |
| "learning_rate": 0.001, | |
| "loss": 1.107, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.078507661819458, | |
| "eval_runtime": 59.3243, | |
| "eval_samples_per_second": 29.6, | |
| "eval_steps_per_second": 3.708, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 0.19963952898979187, | |
| "learning_rate": 0.001, | |
| "loss": 1.0325, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2717391304347826, | |
| "grad_norm": 0.18940940499305725, | |
| "learning_rate": 0.001, | |
| "loss": 1.0672, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2826086956521739, | |
| "grad_norm": 0.2163034975528717, | |
| "learning_rate": 0.001, | |
| "loss": 1.0506, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.29347826086956524, | |
| "grad_norm": 0.20538264513015747, | |
| "learning_rate": 0.001, | |
| "loss": 1.0806, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.30434782608695654, | |
| "grad_norm": 0.2238105982542038, | |
| "learning_rate": 0.001, | |
| "loss": 1.023, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.31521739130434784, | |
| "grad_norm": 0.25053274631500244, | |
| "learning_rate": 0.001, | |
| "loss": 1.0543, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.32608695652173914, | |
| "grad_norm": 0.2265777587890625, | |
| "learning_rate": 0.001, | |
| "loss": 0.9892, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.33695652173913043, | |
| "grad_norm": 0.2253606915473938, | |
| "learning_rate": 0.001, | |
| "loss": 1.0861, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.23203177750110626, | |
| "learning_rate": 0.001, | |
| "loss": 0.9886, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.358695652173913, | |
| "grad_norm": 0.2023775726556778, | |
| "learning_rate": 0.001, | |
| "loss": 0.9306, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3695652173913043, | |
| "grad_norm": 0.18947750329971313, | |
| "learning_rate": 0.001, | |
| "loss": 1.0285, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.3804347826086957, | |
| "grad_norm": 0.2040388137102127, | |
| "learning_rate": 0.001, | |
| "loss": 0.9631, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.391304347826087, | |
| "grad_norm": 0.194660022854805, | |
| "learning_rate": 0.001, | |
| "loss": 0.9291, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.40217391304347827, | |
| "grad_norm": 0.18031583726406097, | |
| "learning_rate": 0.001, | |
| "loss": 0.9478, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.41304347826086957, | |
| "grad_norm": 0.16710995137691498, | |
| "learning_rate": 0.001, | |
| "loss": 0.9605, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.42391304347826086, | |
| "grad_norm": 0.17387554049491882, | |
| "learning_rate": 0.001, | |
| "loss": 0.9359, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 0.1895458847284317, | |
| "learning_rate": 0.001, | |
| "loss": 0.9893, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.44565217391304346, | |
| "grad_norm": 0.1683230996131897, | |
| "learning_rate": 0.001, | |
| "loss": 0.9672, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.45652173913043476, | |
| "grad_norm": 0.1675473004579544, | |
| "learning_rate": 0.001, | |
| "loss": 0.9251, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4673913043478261, | |
| "grad_norm": 0.1792047619819641, | |
| "learning_rate": 0.001, | |
| "loss": 0.9478, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.4782608695652174, | |
| "grad_norm": 0.15780995786190033, | |
| "learning_rate": 0.001, | |
| "loss": 0.9098, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4891304347826087, | |
| "grad_norm": 0.18159078061580658, | |
| "learning_rate": 0.001, | |
| "loss": 0.9709, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.172611802816391, | |
| "learning_rate": 0.001, | |
| "loss": 0.951, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.9241428971290588, | |
| "eval_runtime": 59.13, | |
| "eval_samples_per_second": 29.697, | |
| "eval_steps_per_second": 3.721, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5108695652173914, | |
| "grad_norm": 0.1777033656835556, | |
| "learning_rate": 0.001, | |
| "loss": 0.8208, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 0.16995656490325928, | |
| "learning_rate": 0.001, | |
| "loss": 0.8468, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.532608695652174, | |
| "grad_norm": 0.19403420388698578, | |
| "learning_rate": 0.001, | |
| "loss": 0.9483, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 0.18332509696483612, | |
| "learning_rate": 0.001, | |
| "loss": 0.9222, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5543478260869565, | |
| "grad_norm": 0.16083504259586334, | |
| "learning_rate": 0.001, | |
| "loss": 0.9148, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.5652173913043478, | |
| "grad_norm": 0.17736805975437164, | |
| "learning_rate": 0.001, | |
| "loss": 0.9633, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.5760869565217391, | |
| "grad_norm": 0.17776110768318176, | |
| "learning_rate": 0.001, | |
| "loss": 0.8997, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.5869565217391305, | |
| "grad_norm": 0.16772133111953735, | |
| "learning_rate": 0.001, | |
| "loss": 0.9124, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5978260869565217, | |
| "grad_norm": 0.15817341208457947, | |
| "learning_rate": 0.001, | |
| "loss": 0.8638, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6086956521739131, | |
| "grad_norm": 0.16134516894817352, | |
| "learning_rate": 0.001, | |
| "loss": 0.903, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.6195652173913043, | |
| "grad_norm": 0.16825178265571594, | |
| "learning_rate": 0.001, | |
| "loss": 0.9136, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6304347826086957, | |
| "grad_norm": 0.16397567093372345, | |
| "learning_rate": 0.001, | |
| "loss": 0.8483, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.6413043478260869, | |
| "grad_norm": 0.1606256663799286, | |
| "learning_rate": 0.001, | |
| "loss": 0.9523, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 0.15459905564785004, | |
| "learning_rate": 0.001, | |
| "loss": 0.7925, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6630434782608695, | |
| "grad_norm": 0.16441817581653595, | |
| "learning_rate": 0.001, | |
| "loss": 0.861, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.6739130434782609, | |
| "grad_norm": 6.199015140533447, | |
| "learning_rate": 0.001, | |
| "loss": 0.9204, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.6847826086956522, | |
| "grad_norm": 0.17940112948417664, | |
| "learning_rate": 0.001, | |
| "loss": 0.9263, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.16865596175193787, | |
| "learning_rate": 0.001, | |
| "loss": 0.8912, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.7065217391304348, | |
| "grad_norm": 0.1812729686498642, | |
| "learning_rate": 0.001, | |
| "loss": 0.8756, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.717391304347826, | |
| "grad_norm": 0.17968200147151947, | |
| "learning_rate": 0.001, | |
| "loss": 0.9425, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7282608695652174, | |
| "grad_norm": 0.16858625411987305, | |
| "learning_rate": 0.001, | |
| "loss": 0.7713, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.7391304347826086, | |
| "grad_norm": 0.19516509771347046, | |
| "learning_rate": 0.001, | |
| "loss": 0.8782, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.20293837785720825, | |
| "learning_rate": 0.001, | |
| "loss": 0.9118, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.8631891012191772, | |
| "eval_runtime": 59.0622, | |
| "eval_samples_per_second": 29.731, | |
| "eval_steps_per_second": 3.725, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.7608695652173914, | |
| "grad_norm": 0.16690152883529663, | |
| "learning_rate": 0.001, | |
| "loss": 0.8347, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7717391304347826, | |
| "grad_norm": 0.1717895120382309, | |
| "learning_rate": 0.001, | |
| "loss": 0.805, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 0.1770227551460266, | |
| "learning_rate": 0.001, | |
| "loss": 0.9328, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7934782608695652, | |
| "grad_norm": 0.17513279616832733, | |
| "learning_rate": 0.001, | |
| "loss": 0.8617, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.8043478260869565, | |
| "grad_norm": 0.1699601709842682, | |
| "learning_rate": 0.001, | |
| "loss": 0.882, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.8152173913043478, | |
| "grad_norm": 0.18770118057727814, | |
| "learning_rate": 0.001, | |
| "loss": 0.9139, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8260869565217391, | |
| "grad_norm": 0.17997430264949799, | |
| "learning_rate": 0.001, | |
| "loss": 0.9387, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.8369565217391305, | |
| "grad_norm": 0.1689220368862152, | |
| "learning_rate": 0.001, | |
| "loss": 0.8536, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.8478260869565217, | |
| "grad_norm": 0.16486524045467377, | |
| "learning_rate": 0.001, | |
| "loss": 0.8825, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.8586956521739131, | |
| "grad_norm": 0.1535789668560028, | |
| "learning_rate": 0.001, | |
| "loss": 0.8315, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 0.15810497105121613, | |
| "learning_rate": 0.001, | |
| "loss": 0.8779, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8804347826086957, | |
| "grad_norm": 0.1733904778957367, | |
| "learning_rate": 0.001, | |
| "loss": 0.8488, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8913043478260869, | |
| "grad_norm": 0.15034274756908417, | |
| "learning_rate": 0.001, | |
| "loss": 0.8371, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.9021739130434783, | |
| "grad_norm": 0.1523410528898239, | |
| "learning_rate": 0.001, | |
| "loss": 0.8339, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.9130434782608695, | |
| "grad_norm": 0.1558963656425476, | |
| "learning_rate": 0.001, | |
| "loss": 0.833, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.9239130434782609, | |
| "grad_norm": 0.14974445104599, | |
| "learning_rate": 0.001, | |
| "loss": 0.7828, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.9347826086956522, | |
| "grad_norm": 0.17819815874099731, | |
| "learning_rate": 0.001, | |
| "loss": 0.8727, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.9456521739130435, | |
| "grad_norm": 0.15585996210575104, | |
| "learning_rate": 0.001, | |
| "loss": 0.8063, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.9565217391304348, | |
| "grad_norm": 0.14100810885429382, | |
| "learning_rate": 0.001, | |
| "loss": 0.8116, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.967391304347826, | |
| "grad_norm": 0.17345181107521057, | |
| "learning_rate": 0.001, | |
| "loss": 0.8708, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.9782608695652174, | |
| "grad_norm": 0.1612260788679123, | |
| "learning_rate": 0.001, | |
| "loss": 0.8239, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9891304347826086, | |
| "grad_norm": 0.16653840243816376, | |
| "learning_rate": 0.001, | |
| "loss": 0.897, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.22386449575424194, | |
| "learning_rate": 0.001, | |
| "loss": 0.847, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.8136550784111023, | |
| "eval_runtime": 59.2537, | |
| "eval_samples_per_second": 29.635, | |
| "eval_steps_per_second": 3.713, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 92, | |
| "total_flos": 2.3790723506425037e+17, | |
| "train_loss": 1.062613074546275, | |
| "train_runtime": 1607.4958, | |
| "train_samples_per_second": 7.282, | |
| "train_steps_per_second": 0.057 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 92, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 23, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "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": 2.3790723506425037e+17, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |