{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.5, "eval_steps": 500, "global_step": 5000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "grad_norm": 0.13945120573043823, "learning_rate": 1.8e-06, "loss": 1.1792, "step": 10 }, { "grad_norm": 0.14891666173934937, "learning_rate": 3.8e-06, "loss": 1.1842, "step": 20 }, { "grad_norm": 0.15832659602165222, "learning_rate": 5.8e-06, "loss": 1.1695, "step": 30 }, { "grad_norm": 0.16203539073467255, "learning_rate": 7.8e-06, "loss": 1.1497, "step": 40 }, { "grad_norm": 0.2627524137496948, "learning_rate": 9.800000000000001e-06, "loss": 1.153, "step": 50 }, { "grad_norm": 0.2639051675796509, "learning_rate": 1.18e-05, "loss": 1.1328, "step": 60 }, { "grad_norm": 0.2185407280921936, "learning_rate": 1.3800000000000002e-05, "loss": 1.1178, "step": 70 }, { "grad_norm": 0.34473440051078796, "learning_rate": 1.58e-05, "loss": 1.1078, "step": 80 }, { "grad_norm": 0.18536631762981415, "learning_rate": 1.78e-05, "loss": 1.1055, "step": 90 }, { "grad_norm": 0.5128417611122131, "learning_rate": 1.9800000000000004e-05, "loss": 1.1067, "step": 100 }, { "grad_norm": 0.28990018367767334, "learning_rate": 2.18e-05, "loss": 1.1033, "step": 110 }, { "grad_norm": 0.21424825489521027, "learning_rate": 2.38e-05, "loss": 1.101, "step": 120 }, { "grad_norm": 0.2597063481807709, "learning_rate": 2.58e-05, "loss": 1.0932, "step": 130 }, { "grad_norm": 0.3148020803928375, "learning_rate": 2.7800000000000005e-05, "loss": 1.094, "step": 140 }, { "grad_norm": 0.3488495349884033, "learning_rate": 2.98e-05, "loss": 1.0821, "step": 150 }, { "grad_norm": 0.3580583333969116, "learning_rate": 3.18e-05, "loss": 1.0609, "step": 160 }, { "grad_norm": 0.4472530484199524, "learning_rate": 3.38e-05, "loss": 1.0338, "step": 170 }, { "grad_norm": 0.534553050994873, "learning_rate": 3.58e-05, "loss": 0.9983, "step": 180 }, { "grad_norm": 0.6643158197402954, "learning_rate": 3.7800000000000004e-05, "loss": 0.9619, "step": 190 }, { "grad_norm": 0.7085841298103333, "learning_rate": 3.9800000000000005e-05, "loss": 0.9224, "step": 200 }, { "grad_norm": 0.7006478905677795, "learning_rate": 4.18e-05, "loss": 0.8822, "step": 210 }, { "grad_norm": 1.0977967977523804, "learning_rate": 4.38e-05, "loss": 0.8636, "step": 220 }, { "grad_norm": 0.7735133767127991, "learning_rate": 4.58e-05, "loss": 0.8314, "step": 230 }, { "grad_norm": 1.304010033607483, "learning_rate": 4.78e-05, "loss": 0.8028, "step": 240 }, { "grad_norm": 0.858792245388031, "learning_rate": 4.9800000000000004e-05, "loss": 0.7667, "step": 250 }, { "grad_norm": 0.9611242413520813, "learning_rate": 5.1800000000000005e-05, "loss": 0.7465, "step": 260 }, { "grad_norm": 0.9728292226791382, "learning_rate": 5.380000000000001e-05, "loss": 0.7271, "step": 270 }, { "grad_norm": 0.9787989854812622, "learning_rate": 5.580000000000001e-05, "loss": 0.7047, "step": 280 }, { "grad_norm": 0.9946438074111938, "learning_rate": 5.7799999999999995e-05, "loss": 0.6934, "step": 290 }, { "grad_norm": 0.9978992342948914, "learning_rate": 5.9800000000000003e-05, "loss": 0.6607, "step": 300 }, { "grad_norm": 0.9351487755775452, "learning_rate": 6.18e-05, "loss": 0.6528, "step": 310 }, { "grad_norm": 1.142807960510254, "learning_rate": 6.38e-05, "loss": 0.6212, "step": 320 }, { "grad_norm": 0.9334923028945923, "learning_rate": 6.58e-05, "loss": 0.6125, "step": 330 }, { "grad_norm": 0.8837148547172546, "learning_rate": 6.780000000000001e-05, "loss": 0.5951, "step": 340 }, { "grad_norm": 0.9366855025291443, "learning_rate": 6.98e-05, "loss": 0.5671, "step": 350 }, { "grad_norm": 0.9530870318412781, "learning_rate": 7.18e-05, "loss": 0.5615, "step": 360 }, { "grad_norm": 1.0772392749786377, "learning_rate": 7.38e-05, "loss": 0.5644, "step": 370 }, { "grad_norm": 0.9228338003158569, "learning_rate": 7.58e-05, "loss": 0.55, "step": 380 }, { "grad_norm": 1.1477776765823364, "learning_rate": 7.780000000000001e-05, "loss": 0.5417, "step": 390 }, { "grad_norm": 0.9487908482551575, "learning_rate": 7.98e-05, "loss": 0.5299, "step": 400 }, { "grad_norm": 0.8648481965065002, "learning_rate": 8.18e-05, "loss": 0.5245, "step": 410 }, { "grad_norm": 1.0358718633651733, "learning_rate": 8.38e-05, "loss": 0.5066, "step": 420 }, { "grad_norm": 0.9835023283958435, "learning_rate": 8.58e-05, "loss": 0.5033, "step": 430 }, { "grad_norm": 0.9648961424827576, "learning_rate": 8.78e-05, "loss": 0.4882, "step": 440 }, { "grad_norm": 1.1135525703430176, "learning_rate": 8.98e-05, "loss": 0.4808, "step": 450 }, { "grad_norm": 0.9057552218437195, "learning_rate": 9.180000000000001e-05, "loss": 0.4701, "step": 460 }, { "grad_norm": 1.0914181470870972, "learning_rate": 9.38e-05, "loss": 0.4617, "step": 470 }, { "grad_norm": 1.1427769660949707, "learning_rate": 9.58e-05, "loss": 0.4483, "step": 480 }, { "grad_norm": 0.9296786785125732, "learning_rate": 9.78e-05, "loss": 0.4222, "step": 490 }, { "eval/loss": 0.4169562757015228, "step": 500 }, { "grad_norm": 1.1534019708633423, "learning_rate": 9.98e-05, "loss": 0.4134, "step": 500 }, { "grad_norm": 1.0778305530548096, "learning_rate": 9.9999778549206e-05, "loss": 0.4081, "step": 510 }, { "grad_norm": 1.007461428642273, "learning_rate": 9.999901304280685e-05, "loss": 0.3938, "step": 520 }, { "grad_norm": 0.8525221943855286, "learning_rate": 9.999770075521164e-05, "loss": 0.3874, "step": 530 }, { "grad_norm": 1.109859585762024, "learning_rate": 9.99958417007713e-05, "loss": 0.387, "step": 540 }, { "grad_norm": 1.0042812824249268, "learning_rate": 9.999343589981615e-05, "loss": 0.373, "step": 550 }, { "grad_norm": 0.9587410092353821, "learning_rate": 9.999048337865568e-05, "loss": 0.3665, "step": 560 }, { "grad_norm": 1.1748636960983276, "learning_rate": 9.998698416957815e-05, "loss": 0.3461, "step": 570 }, { "grad_norm": 1.0844266414642334, "learning_rate": 9.998293831085037e-05, "loss": 0.3271, "step": 580 }, { "grad_norm": 0.927900493144989, "learning_rate": 9.997834584671719e-05, "loss": 0.3336, "step": 590 }, { "grad_norm": 1.1865487098693848, "learning_rate": 9.997320682740107e-05, "loss": 0.3229, "step": 600 }, { "grad_norm": 1.2733619213104248, "learning_rate": 9.996752130910149e-05, "loss": 0.317, "step": 610 }, { "grad_norm": 1.2515984773635864, "learning_rate": 9.99612893539944e-05, "loss": 0.3263, "step": 620 }, { "grad_norm": 1.2944176197052002, "learning_rate": 9.995451103023144e-05, "loss": 0.3099, "step": 630 }, { "grad_norm": 1.0592591762542725, "learning_rate": 9.994718641193928e-05, "loss": 0.2869, "step": 640 }, { "grad_norm": 1.1084046363830566, "learning_rate": 9.993931557921874e-05, "loss": 0.2858, "step": 650 }, { "grad_norm": 0.9912689924240112, "learning_rate": 9.993089861814402e-05, "loss": 0.281, "step": 660 }, { "grad_norm": 1.1200611591339111, "learning_rate": 9.992193562076166e-05, "loss": 0.2603, "step": 670 }, { "grad_norm": 1.3195865154266357, "learning_rate": 9.991242668508954e-05, "loss": 0.2551, "step": 680 }, { "grad_norm": 1.0591305494308472, "learning_rate": 9.990237191511587e-05, "loss": 0.2296, "step": 690 }, { "grad_norm": 1.5685760974884033, "learning_rate": 9.989177142079802e-05, "loss": 0.2412, "step": 700 }, { "grad_norm": 0.9255881309509277, "learning_rate": 9.988062531806126e-05, "loss": 0.2232, "step": 710 }, { "grad_norm": 1.0331252813339233, "learning_rate": 9.986893372879762e-05, "loss": 0.2208, "step": 720 }, { "grad_norm": 1.0819493532180786, "learning_rate": 9.985669678086443e-05, "loss": 0.2124, "step": 730 }, { "grad_norm": 0.947820782661438, "learning_rate": 9.984391460808298e-05, "loss": 0.2011, "step": 740 }, { "grad_norm": 1.2813969850540161, "learning_rate": 9.983058735023709e-05, "loss": 0.204, "step": 750 }, { "grad_norm": 0.9526956081390381, "learning_rate": 9.98167151530715e-05, "loss": 0.2, "step": 760 }, { "grad_norm": 1.2552719116210938, "learning_rate": 9.980229816829034e-05, "loss": 0.1923, "step": 770 }, { "grad_norm": 1.0731266736984253, "learning_rate": 9.978733655355544e-05, "loss": 0.1874, "step": 780 }, { "grad_norm": 1.1821141242980957, "learning_rate": 9.977183047248464e-05, "loss": 0.1694, "step": 790 }, { "grad_norm": 1.3457095623016357, "learning_rate": 9.975578009464992e-05, "loss": 0.1903, "step": 800 }, { "grad_norm": 0.9491983652114868, "learning_rate": 9.97391855955757e-05, "loss": 0.1668, "step": 810 }, { "grad_norm": 1.257034182548523, "learning_rate": 9.972204715673669e-05, "loss": 0.1644, "step": 820 }, { "grad_norm": 1.0270410776138306, "learning_rate": 9.970436496555617e-05, "loss": 0.1504, "step": 830 }, { "grad_norm": 0.8814347386360168, "learning_rate": 9.968613921540373e-05, "loss": 0.1473, "step": 840 }, { "grad_norm": 1.0740413665771484, "learning_rate": 9.966737010559326e-05, "loss": 0.1602, "step": 850 }, { "grad_norm": 1.1755180358886719, "learning_rate": 9.964805784138072e-05, "loss": 0.1368, "step": 860 }, { "grad_norm": 1.1243144273757935, "learning_rate": 9.962820263396195e-05, "loss": 0.1473, "step": 870 }, { "grad_norm": 1.0385600328445435, "learning_rate": 9.960780470047033e-05, "loss": 0.1404, "step": 880 }, { "grad_norm": 0.8931643962860107, "learning_rate": 9.958686426397437e-05, "loss": 0.1312, "step": 890 }, { "grad_norm": 1.0461937189102173, "learning_rate": 9.956538155347534e-05, "loss": 0.1442, "step": 900 }, { "grad_norm": 1.0877867937088013, "learning_rate": 9.95433568039047e-05, "loss": 0.1284, "step": 910 }, { "grad_norm": 1.0202831029891968, "learning_rate": 9.952079025612162e-05, "loss": 0.1243, "step": 920 }, { "grad_norm": 1.262707233428955, "learning_rate": 9.949768215691022e-05, "loss": 0.1352, "step": 930 }, { "grad_norm": 1.084681749343872, "learning_rate": 9.9474032758977e-05, "loss": 0.1437, "step": 940 }, { "grad_norm": 0.9750522375106812, "learning_rate": 9.944984232094794e-05, "loss": 0.1335, "step": 950 }, { "grad_norm": 0.7702094912528992, "learning_rate": 9.942511110736584e-05, "loss": 0.127, "step": 960 }, { "grad_norm": 1.0232125520706177, "learning_rate": 9.939983938868726e-05, "loss": 0.1325, "step": 970 }, { "grad_norm": 0.9044125080108643, "learning_rate": 9.93740274412797e-05, "loss": 0.1276, "step": 980 }, { "grad_norm": 0.8179964423179626, "learning_rate": 9.934767554741846e-05, "loss": 0.1345, "step": 990 }, { "eval/loss": 0.1256265753507614, "step": 1000 }, { "grad_norm": 1.036224603652954, "learning_rate": 9.932078399528361e-05, "loss": 0.1203, "step": 1000 }, { "grad_norm": 1.0834065675735474, "learning_rate": 9.929335307895689e-05, "loss": 0.1415, "step": 1010 }, { "grad_norm": 0.9903861880302429, "learning_rate": 9.926538309841839e-05, "loss": 0.1327, "step": 1020 }, { "grad_norm": 0.9907166361808777, "learning_rate": 9.923687435954334e-05, "loss": 0.1397, "step": 1030 }, { "grad_norm": 0.8730444312095642, "learning_rate": 9.920782717409873e-05, "loss": 0.1352, "step": 1040 }, { "grad_norm": 0.9731714725494385, "learning_rate": 9.917824185973994e-05, "loss": 0.1304, "step": 1050 }, { "grad_norm": 0.9861385226249695, "learning_rate": 9.914811874000723e-05, "loss": 0.12, "step": 1060 }, { "grad_norm": 1.0249139070510864, "learning_rate": 9.911745814432218e-05, "loss": 0.1179, "step": 1070 }, { "grad_norm": 1.009698748588562, "learning_rate": 9.90862604079842e-05, "loss": 0.1244, "step": 1080 }, { "grad_norm": 0.8057038187980652, "learning_rate": 9.90545258721667e-05, "loss": 0.1287, "step": 1090 }, { "grad_norm": 0.8731530904769897, "learning_rate": 9.90222548839135e-05, "loss": 0.1301, "step": 1100 }, { "grad_norm": 0.8571906089782715, "learning_rate": 9.898944779613495e-05, "loss": 0.121, "step": 1110 }, { "grad_norm": 0.8374456167221069, "learning_rate": 9.89561049676041e-05, "loss": 0.1155, "step": 1120 }, { "grad_norm": 0.9036489129066467, "learning_rate": 9.89222267629528e-05, "loss": 0.1264, "step": 1130 }, { "grad_norm": 0.7708646655082703, "learning_rate": 9.888781355266763e-05, "loss": 0.1218, "step": 1140 }, { "grad_norm": 0.8521364331245422, "learning_rate": 9.885286571308598e-05, "loss": 0.1231, "step": 1150 }, { "grad_norm": 1.0965105295181274, "learning_rate": 9.881738362639182e-05, "loss": 0.1211, "step": 1160 }, { "grad_norm": 0.9882417917251587, "learning_rate": 9.878136768061154e-05, "loss": 0.1314, "step": 1170 }, { "grad_norm": 0.8928432464599609, "learning_rate": 9.874481826960979e-05, "loss": 0.1318, "step": 1180 }, { "grad_norm": 1.0171260833740234, "learning_rate": 9.870773579308503e-05, "loss": 0.1298, "step": 1190 }, { "grad_norm": 0.8835505843162537, "learning_rate": 9.867012065656533e-05, "loss": 0.1226, "step": 1200 }, { "grad_norm": 1.0140732526779175, "learning_rate": 9.863197327140376e-05, "loss": 0.1249, "step": 1210 }, { "grad_norm": 0.9845730662345886, "learning_rate": 9.859329405477403e-05, "loss": 0.1303, "step": 1220 }, { "grad_norm": 0.8532151579856873, "learning_rate": 9.855408342966585e-05, "loss": 0.1168, "step": 1230 }, { "grad_norm": 0.8756526112556458, "learning_rate": 9.851434182488033e-05, "loss": 0.1205, "step": 1240 }, { "grad_norm": 0.869296669960022, "learning_rate": 9.84740696750253e-05, "loss": 0.1105, "step": 1250 }, { "grad_norm": 0.9367215037345886, "learning_rate": 9.843326742051055e-05, "loss": 0.1219, "step": 1260 }, { "grad_norm": 0.8888024687767029, "learning_rate": 9.839193550754297e-05, "loss": 0.1048, "step": 1270 }, { "grad_norm": 0.9300942420959473, "learning_rate": 9.835007438812177e-05, "loss": 0.1188, "step": 1280 }, { "grad_norm": 0.7654188871383667, "learning_rate": 9.830768452003341e-05, "loss": 0.1077, "step": 1290 }, { "grad_norm": 0.872153103351593, "learning_rate": 9.826476636684671e-05, "loss": 0.1212, "step": 1300 }, { "grad_norm": 1.01703679561615, "learning_rate": 9.822132039790773e-05, "loss": 0.1032, "step": 1310 }, { "grad_norm": 0.8411621451377869, "learning_rate": 9.817734708833461e-05, "loss": 0.1247, "step": 1320 }, { "grad_norm": 0.9694423675537109, "learning_rate": 9.813284691901243e-05, "loss": 0.1177, "step": 1330 }, { "grad_norm": 0.9464964866638184, "learning_rate": 9.808782037658792e-05, "loss": 0.1174, "step": 1340 }, { "grad_norm": 0.7668417096138, "learning_rate": 9.804226795346411e-05, "loss": 0.1188, "step": 1350 }, { "grad_norm": 0.8842877745628357, "learning_rate": 9.799619014779503e-05, "loss": 0.1204, "step": 1360 }, { "grad_norm": 0.6962409019470215, "learning_rate": 9.794958746348013e-05, "loss": 0.1211, "step": 1370 }, { "grad_norm": 1.0325098037719727, "learning_rate": 9.790246041015896e-05, "loss": 0.1185, "step": 1380 }, { "grad_norm": 0.7428703904151917, "learning_rate": 9.785480950320538e-05, "loss": 0.1083, "step": 1390 }, { "grad_norm": 0.973091721534729, "learning_rate": 9.78066352637221e-05, "loss": 0.1255, "step": 1400 }, { "grad_norm": 0.8553988933563232, "learning_rate": 9.775793821853488e-05, "loss": 0.1143, "step": 1410 }, { "grad_norm": 0.8368469476699829, "learning_rate": 9.77087189001868e-05, "loss": 0.1137, "step": 1420 }, { "grad_norm": 0.8512156009674072, "learning_rate": 9.765897784693243e-05, "loss": 0.1153, "step": 1430 }, { "grad_norm": 0.7669680118560791, "learning_rate": 9.760871560273197e-05, "loss": 0.119, "step": 1440 }, { "grad_norm": 0.8029971718788147, "learning_rate": 9.755793271724526e-05, "loss": 0.1129, "step": 1450 }, { "grad_norm": 0.8904449343681335, "learning_rate": 9.750662974582584e-05, "loss": 0.1119, "step": 1460 }, { "grad_norm": 0.791136622428894, "learning_rate": 9.745480724951473e-05, "loss": 0.1013, "step": 1470 }, { "grad_norm": 0.910875141620636, "learning_rate": 9.740246579503447e-05, "loss": 0.1008, "step": 1480 }, { "grad_norm": 1.0056816339492798, "learning_rate": 9.734960595478284e-05, "loss": 0.1064, "step": 1490 }, { "eval/loss": 0.11019096165895462, "step": 1500 }, { "grad_norm": 0.7263178825378418, "learning_rate": 9.729622830682657e-05, "loss": 0.1081, "step": 1500 }, { "grad_norm": 0.8493738770484924, "learning_rate": 9.724233343489504e-05, "loss": 0.0943, "step": 1510 }, { "grad_norm": 0.9054257273674011, "learning_rate": 9.718792192837396e-05, "loss": 0.1086, "step": 1520 }, { "grad_norm": 0.8628945350646973, "learning_rate": 9.713299438229886e-05, "loss": 0.1176, "step": 1530 }, { "grad_norm": 0.9462695717811584, "learning_rate": 9.707755139734855e-05, "loss": 0.1104, "step": 1540 }, { "grad_norm": 0.8205919861793518, "learning_rate": 9.702159357983866e-05, "loss": 0.1055, "step": 1550 }, { "grad_norm": 0.87127685546875, "learning_rate": 9.696512154171492e-05, "loss": 0.1058, "step": 1560 }, { "grad_norm": 0.8638626337051392, "learning_rate": 9.690813590054645e-05, "loss": 0.1122, "step": 1570 }, { "grad_norm": 0.817553699016571, "learning_rate": 9.685063727951914e-05, "loss": 0.111, "step": 1580 }, { "grad_norm": 0.9839280247688293, "learning_rate": 9.679262630742865e-05, "loss": 0.1011, "step": 1590 }, { "grad_norm": 0.8560328483581543, "learning_rate": 9.673410361867373e-05, "loss": 0.1046, "step": 1600 }, { "grad_norm": 0.7241759300231934, "learning_rate": 9.667506985324909e-05, "loss": 0.1094, "step": 1610 }, { "grad_norm": 0.9413039684295654, "learning_rate": 9.661552565673855e-05, "loss": 0.1043, "step": 1620 }, { "grad_norm": 0.7728748917579651, "learning_rate": 9.655547168030789e-05, "loss": 0.099, "step": 1630 }, { "grad_norm": 0.7551400065422058, "learning_rate": 9.649490858069777e-05, "loss": 0.1027, "step": 1640 }, { "grad_norm": 1.0061802864074707, "learning_rate": 9.643383702021658e-05, "loss": 0.1038, "step": 1650 }, { "grad_norm": 0.7926415801048279, "learning_rate": 9.637225766673307e-05, "loss": 0.1029, "step": 1660 }, { "grad_norm": 0.8423016667366028, "learning_rate": 9.631017119366922e-05, "loss": 0.0949, "step": 1670 }, { "grad_norm": 0.8185058236122131, "learning_rate": 9.624757827999273e-05, "loss": 0.0972, "step": 1680 }, { "grad_norm": 0.773152768611908, "learning_rate": 9.618447961020971e-05, "loss": 0.0961, "step": 1690 }, { "grad_norm": 0.7074319124221802, "learning_rate": 9.612087587435707e-05, "loss": 0.0986, "step": 1700 }, { "grad_norm": 0.7399337291717529, "learning_rate": 9.605676776799508e-05, "loss": 0.1067, "step": 1710 }, { "grad_norm": 0.8363661170005798, "learning_rate": 9.599215599219973e-05, "loss": 0.1078, "step": 1720 }, { "grad_norm": 0.8980876207351685, "learning_rate": 9.592704125355505e-05, "loss": 0.0942, "step": 1730 }, { "grad_norm": 0.6812839508056641, "learning_rate": 9.586142426414538e-05, "loss": 0.0944, "step": 1740 }, { "grad_norm": 0.8147590756416321, "learning_rate": 9.57953057415476e-05, "loss": 0.1143, "step": 1750 }, { "grad_norm": 0.7535157203674316, "learning_rate": 9.572868640882328e-05, "loss": 0.1043, "step": 1760 }, { "grad_norm": 0.7984067797660828, "learning_rate": 9.56615669945108e-05, "loss": 0.1069, "step": 1770 }, { "grad_norm": 0.8090737462043762, "learning_rate": 9.55939482326173e-05, "loss": 0.1036, "step": 1780 }, { "grad_norm": 0.7418506741523743, "learning_rate": 9.552583086261069e-05, "loss": 0.1003, "step": 1790 }, { "grad_norm": 0.7592194080352783, "learning_rate": 9.545721562941168e-05, "loss": 0.1012, "step": 1800 }, { "grad_norm": 0.8451941609382629, "learning_rate": 9.538810328338543e-05, "loss": 0.0948, "step": 1810 }, { "grad_norm": 0.8155949115753174, "learning_rate": 9.531849458033349e-05, "loss": 0.1133, "step": 1820 }, { "grad_norm": 0.8501562476158142, "learning_rate": 9.524839028148547e-05, "loss": 0.1124, "step": 1830 }, { "grad_norm": 0.768547534942627, "learning_rate": 9.517779115349077e-05, "loss": 0.0973, "step": 1840 }, { "grad_norm": 0.8181756138801575, "learning_rate": 9.510669796841014e-05, "loss": 0.0955, "step": 1850 }, { "grad_norm": 0.8555414080619812, "learning_rate": 9.503511150370727e-05, "loss": 0.0968, "step": 1860 }, { "grad_norm": 0.664882242679596, "learning_rate": 9.496303254224024e-05, "loss": 0.0988, "step": 1870 }, { "grad_norm": 0.9401973485946655, "learning_rate": 9.489046187225306e-05, "loss": 0.1097, "step": 1880 }, { "grad_norm": 0.7436525821685791, "learning_rate": 9.481740028736692e-05, "loss": 0.1073, "step": 1890 }, { "grad_norm": 0.7996155619621277, "learning_rate": 9.474384858657164e-05, "loss": 0.099, "step": 1900 }, { "grad_norm": 0.836756706237793, "learning_rate": 9.466980757421679e-05, "loss": 0.0952, "step": 1910 }, { "grad_norm": 0.7757591605186462, "learning_rate": 9.459527806000305e-05, "loss": 0.0881, "step": 1920 }, { "grad_norm": 0.7440276741981506, "learning_rate": 9.452026085897325e-05, "loss": 0.0914, "step": 1930 }, { "grad_norm": 0.9869599342346191, "learning_rate": 9.444475679150348e-05, "loss": 0.0932, "step": 1940 }, { "grad_norm": 0.639369547367096, "learning_rate": 9.436876668329411e-05, "loss": 0.0941, "step": 1950 }, { "grad_norm": 0.9225316643714905, "learning_rate": 9.429229136536079e-05, "loss": 0.0954, "step": 1960 }, { "grad_norm": 0.8412165641784668, "learning_rate": 9.421533167402534e-05, "loss": 0.0918, "step": 1970 }, { "grad_norm": 0.8764362335205078, "learning_rate": 9.413788845090666e-05, "loss": 0.0958, "step": 1980 }, { "grad_norm": 0.8007136583328247, "learning_rate": 9.405996254291136e-05, "loss": 0.0971, "step": 1990 }, { "eval/loss": 0.1067818658053875, "step": 2000 }, { "grad_norm": 0.7538358569145203, "learning_rate": 9.398155480222474e-05, "loss": 0.1051, "step": 2000 }, { "grad_norm": 0.7017872333526611, "learning_rate": 9.390266608630128e-05, "loss": 0.0866, "step": 2010 }, { "grad_norm": 0.8721462488174438, "learning_rate": 9.38232972578553e-05, "loss": 0.0908, "step": 2020 }, { "grad_norm": 0.7407742738723755, "learning_rate": 9.374344918485164e-05, "loss": 0.0936, "step": 2030 }, { "grad_norm": 0.7346519231796265, "learning_rate": 9.366312274049602e-05, "loss": 0.0914, "step": 2040 }, { "grad_norm": 0.8163416981697083, "learning_rate": 9.358231880322554e-05, "loss": 0.1, "step": 2050 }, { "grad_norm": 0.7716719508171082, "learning_rate": 9.350103825669916e-05, "loss": 0.0972, "step": 2060 }, { "grad_norm": 0.8662499785423279, "learning_rate": 9.341928198978787e-05, "loss": 0.0881, "step": 2070 }, { "grad_norm": 0.8245924115180969, "learning_rate": 9.333705089656512e-05, "loss": 0.0968, "step": 2080 }, { "grad_norm": 0.6623910665512085, "learning_rate": 9.325434587629698e-05, "loss": 0.0928, "step": 2090 }, { "grad_norm": 0.6419522762298584, "learning_rate": 9.31711678334323e-05, "loss": 0.0865, "step": 2100 }, { "grad_norm": 0.7035567760467529, "learning_rate": 9.308751767759282e-05, "loss": 0.098, "step": 2110 }, { "grad_norm": 1.060964822769165, "learning_rate": 9.300339632356325e-05, "loss": 0.0875, "step": 2120 }, { "grad_norm": 0.8512191772460938, "learning_rate": 9.291880469128124e-05, "loss": 0.102, "step": 2130 }, { "grad_norm": 0.6871913075447083, "learning_rate": 9.283374370582732e-05, "loss": 0.0903, "step": 2140 }, { "grad_norm": 0.7258641123771667, "learning_rate": 9.274821429741482e-05, "loss": 0.0971, "step": 2150 }, { "grad_norm": 0.769487202167511, "learning_rate": 9.266221740137961e-05, "loss": 0.0921, "step": 2160 }, { "grad_norm": 0.7882146239280701, "learning_rate": 9.257575395817001e-05, "loss": 0.0895, "step": 2170 }, { "grad_norm": 0.7510284185409546, "learning_rate": 9.248882491333637e-05, "loss": 0.0856, "step": 2180 }, { "grad_norm": 0.7088679671287537, "learning_rate": 9.240143121752076e-05, "loss": 0.0949, "step": 2190 }, { "grad_norm": 0.7769065499305725, "learning_rate": 9.23135738264467e-05, "loss": 0.0948, "step": 2200 }, { "grad_norm": 0.7561931014060974, "learning_rate": 9.222525370090849e-05, "loss": 0.0853, "step": 2210 }, { "grad_norm": 0.5801048874855042, "learning_rate": 9.213647180676088e-05, "loss": 0.0903, "step": 2220 }, { "grad_norm": 0.797909140586853, "learning_rate": 9.204722911490846e-05, "loss": 0.086, "step": 2230 }, { "grad_norm": 0.8954686522483826, "learning_rate": 9.1957526601295e-05, "loss": 0.0966, "step": 2240 }, { "grad_norm": 0.7500966787338257, "learning_rate": 9.186736524689281e-05, "loss": 0.0891, "step": 2250 }, { "grad_norm": 0.8346981406211853, "learning_rate": 9.177674603769204e-05, "loss": 0.1022, "step": 2260 }, { "grad_norm": 0.6813527941703796, "learning_rate": 9.168566996468983e-05, "loss": 0.0892, "step": 2270 }, { "grad_norm": 0.6238582730293274, "learning_rate": 9.159413802387951e-05, "loss": 0.0906, "step": 2280 }, { "grad_norm": 0.7254133224487305, "learning_rate": 9.150215121623974e-05, "loss": 0.0815, "step": 2290 }, { "grad_norm": 0.7142599821090698, "learning_rate": 9.140971054772349e-05, "loss": 0.09, "step": 2300 }, { "grad_norm": 0.7069528102874756, "learning_rate": 9.131681702924713e-05, "loss": 0.0922, "step": 2310 }, { "grad_norm": 0.6693764925003052, "learning_rate": 9.122347167667926e-05, "loss": 0.0844, "step": 2320 }, { "grad_norm": 0.947096049785614, "learning_rate": 9.112967551082973e-05, "loss": 0.0852, "step": 2330 }, { "grad_norm": 0.7382696866989136, "learning_rate": 9.103542955743835e-05, "loss": 0.0995, "step": 2340 }, { "grad_norm": 0.7165326476097107, "learning_rate": 9.094073484716381e-05, "loss": 0.1076, "step": 2350 }, { "grad_norm": 0.6296545267105103, "learning_rate": 9.084559241557226e-05, "loss": 0.0919, "step": 2360 }, { "grad_norm": 0.7040671706199646, "learning_rate": 9.075000330312608e-05, "loss": 0.0869, "step": 2370 }, { "grad_norm": 0.626624584197998, "learning_rate": 9.065396855517253e-05, "loss": 0.0872, "step": 2380 }, { "grad_norm": 0.8399067521095276, "learning_rate": 9.055748922193219e-05, "loss": 0.0942, "step": 2390 }, { "grad_norm": 0.7377070188522339, "learning_rate": 9.046056635848761e-05, "loss": 0.0915, "step": 2400 }, { "grad_norm": 0.6742762327194214, "learning_rate": 9.036320102477169e-05, "loss": 0.0825, "step": 2410 }, { "grad_norm": 0.6328057646751404, "learning_rate": 9.02653942855561e-05, "loss": 0.0874, "step": 2420 }, { "grad_norm": 0.8291254043579102, "learning_rate": 9.016714721043971e-05, "loss": 0.0815, "step": 2430 }, { "grad_norm": 0.647943377494812, "learning_rate": 9.006846087383675e-05, "loss": 0.0801, "step": 2440 }, { "grad_norm": 0.6599271893501282, "learning_rate": 8.996933635496523e-05, "loss": 0.0889, "step": 2450 }, { "grad_norm": 0.6580495238304138, "learning_rate": 8.986977473783498e-05, "loss": 0.1002, "step": 2460 }, { "grad_norm": 0.7060482501983643, "learning_rate": 8.97697771112359e-05, "loss": 0.0866, "step": 2470 }, { "grad_norm": 0.7797880172729492, "learning_rate": 8.966934456872602e-05, "loss": 0.0891, "step": 2480 }, { "grad_norm": 0.7028229832649231, "learning_rate": 8.95684782086195e-05, "loss": 0.0843, "step": 2490 }, { "eval/loss": 0.0992034749686718, "step": 2500 }, { "grad_norm": 0.7305085062980652, "learning_rate": 8.946717913397476e-05, "loss": 0.0925, "step": 2500 }, { "grad_norm": 0.59566330909729, "learning_rate": 8.93654484525822e-05, "loss": 0.084, "step": 2510 }, { "grad_norm": 0.7499849796295166, "learning_rate": 8.926328727695226e-05, "loss": 0.0851, "step": 2520 }, { "grad_norm": 0.6647255420684814, "learning_rate": 8.916069672430319e-05, "loss": 0.0868, "step": 2530 }, { "grad_norm": 0.7804701328277588, "learning_rate": 8.905767791654884e-05, "loss": 0.0776, "step": 2540 }, { "grad_norm": 0.7224317193031311, "learning_rate": 8.895423198028638e-05, "loss": 0.0867, "step": 2550 }, { "grad_norm": 0.6948305368423462, "learning_rate": 8.885036004678402e-05, "loss": 0.0847, "step": 2560 }, { "grad_norm": 0.7347127199172974, "learning_rate": 8.874606325196857e-05, "loss": 0.0793, "step": 2570 }, { "grad_norm": 0.6937545537948608, "learning_rate": 8.864134273641304e-05, "loss": 0.0866, "step": 2580 }, { "grad_norm": 0.6622210144996643, "learning_rate": 8.853619964532427e-05, "loss": 0.084, "step": 2590 }, { "grad_norm": 0.7432724833488464, "learning_rate": 8.843063512853019e-05, "loss": 0.0833, "step": 2600 }, { "grad_norm": 0.6376204490661621, "learning_rate": 8.832465034046749e-05, "loss": 0.0818, "step": 2610 }, { "grad_norm": 0.7311689853668213, "learning_rate": 8.821824644016882e-05, "loss": 0.0777, "step": 2620 }, { "grad_norm": 0.6861463189125061, "learning_rate": 8.811142459125019e-05, "loss": 0.0824, "step": 2630 }, { "grad_norm": 0.57241290807724, "learning_rate": 8.800418596189822e-05, "loss": 0.0804, "step": 2640 }, { "grad_norm": 0.6616230010986328, "learning_rate": 8.789653172485737e-05, "loss": 0.0835, "step": 2650 }, { "grad_norm": 0.7123761177062988, "learning_rate": 8.778846305741715e-05, "loss": 0.1028, "step": 2660 }, { "grad_norm": 0.6900753378868103, "learning_rate": 8.767998114139918e-05, "loss": 0.0784, "step": 2670 }, { "grad_norm": 0.7233784794807434, "learning_rate": 8.757108716314429e-05, "loss": 0.0813, "step": 2680 }, { "grad_norm": 0.7010366320610046, "learning_rate": 8.746178231349962e-05, "loss": 0.0809, "step": 2690 }, { "grad_norm": 0.670228123664856, "learning_rate": 8.735206778780549e-05, "loss": 0.0839, "step": 2700 }, { "grad_norm": 0.7604944705963135, "learning_rate": 8.724194478588234e-05, "loss": 0.096, "step": 2710 }, { "grad_norm": 0.6636898517608643, "learning_rate": 8.713141451201772e-05, "loss": 0.0862, "step": 2720 }, { "grad_norm": 0.5826030373573303, "learning_rate": 8.702047817495295e-05, "loss": 0.084, "step": 2730 }, { "grad_norm": 0.7790019512176514, "learning_rate": 8.69091369878701e-05, "loss": 0.0749, "step": 2740 }, { "grad_norm": 0.6743026375770569, "learning_rate": 8.679739216837849e-05, "loss": 0.0867, "step": 2750 }, { "grad_norm": 0.7239930629730225, "learning_rate": 8.66852449385016e-05, "loss": 0.0824, "step": 2760 }, { "grad_norm": 0.8334649801254272, "learning_rate": 8.657269652466356e-05, "loss": 0.0905, "step": 2770 }, { "grad_norm": 0.6980594396591187, "learning_rate": 8.645974815767577e-05, "loss": 0.081, "step": 2780 }, { "grad_norm": 0.5909601449966431, "learning_rate": 8.634640107272351e-05, "loss": 0.0752, "step": 2790 }, { "grad_norm": 0.7336029410362244, "learning_rate": 8.623265650935234e-05, "loss": 0.0845, "step": 2800 }, { "grad_norm": 0.746120035648346, "learning_rate": 8.611851571145456e-05, "loss": 0.0857, "step": 2810 }, { "grad_norm": 0.6810508370399475, "learning_rate": 8.600397992725566e-05, "loss": 0.0898, "step": 2820 }, { "grad_norm": 0.6810851097106934, "learning_rate": 8.588905040930061e-05, "loss": 0.0808, "step": 2830 }, { "grad_norm": 0.6361064910888672, "learning_rate": 8.577372841444022e-05, "loss": 0.0853, "step": 2840 }, { "grad_norm": 0.640917956829071, "learning_rate": 8.565801520381736e-05, "loss": 0.0879, "step": 2850 }, { "grad_norm": 0.7023554444313049, "learning_rate": 8.554191204285313e-05, "loss": 0.0883, "step": 2860 }, { "grad_norm": 0.6683006286621094, "learning_rate": 8.542542020123315e-05, "loss": 0.0763, "step": 2870 }, { "grad_norm": 0.5859060287475586, "learning_rate": 8.530854095289347e-05, "loss": 0.0816, "step": 2880 }, { "grad_norm": 0.7188131213188171, "learning_rate": 8.519127557600688e-05, "loss": 0.0814, "step": 2890 }, { "grad_norm": 0.5783581137657166, "learning_rate": 8.507362535296871e-05, "loss": 0.0821, "step": 2900 }, { "grad_norm": 0.5532498359680176, "learning_rate": 8.495559157038299e-05, "loss": 0.0746, "step": 2910 }, { "grad_norm": 0.7175596356391907, "learning_rate": 8.483717551904823e-05, "loss": 0.0859, "step": 2920 }, { "grad_norm": 0.6711920499801636, "learning_rate": 8.47183784939434e-05, "loss": 0.0778, "step": 2930 }, { "grad_norm": 0.756864070892334, "learning_rate": 8.459920179421374e-05, "loss": 0.0847, "step": 2940 }, { "grad_norm": 0.7185421586036682, "learning_rate": 8.447964672315656e-05, "loss": 0.1006, "step": 2950 }, { "grad_norm": 0.7799793481826782, "learning_rate": 8.435971458820692e-05, "loss": 0.0878, "step": 2960 }, { "grad_norm": 0.6690899729728699, "learning_rate": 8.423940670092345e-05, "loss": 0.0809, "step": 2970 }, { "grad_norm": 0.7314770817756653, "learning_rate": 8.411872437697394e-05, "loss": 0.08, "step": 2980 }, { "grad_norm": 0.7342603206634521, "learning_rate": 8.399766893612096e-05, "loss": 0.0732, "step": 2990 }, { "eval/loss": 0.09841185182332993, "step": 3000 }, { "grad_norm": 0.829379141330719, "learning_rate": 8.38762417022074e-05, "loss": 0.0746, "step": 3000 }, { "grad_norm": 0.6351096630096436, "learning_rate": 8.375444400314204e-05, "loss": 0.0736, "step": 3010 }, { "grad_norm": 0.6351186037063599, "learning_rate": 8.3632277170885e-05, "loss": 0.0809, "step": 3020 }, { "grad_norm": 0.8107452392578125, "learning_rate": 8.350974254143318e-05, "loss": 0.0807, "step": 3030 }, { "grad_norm": 0.7463685274124146, "learning_rate": 8.338684145480566e-05, "loss": 0.0801, "step": 3040 }, { "grad_norm": 0.6013656854629517, "learning_rate": 8.326357525502904e-05, "loss": 0.085, "step": 3050 }, { "grad_norm": 0.5978612303733826, "learning_rate": 8.313994529012273e-05, "loss": 0.0767, "step": 3060 }, { "grad_norm": 0.6979216933250427, "learning_rate": 8.301595291208422e-05, "loss": 0.0948, "step": 3070 }, { "grad_norm": 0.564122200012207, "learning_rate": 8.289159947687427e-05, "loss": 0.0863, "step": 3080 }, { "grad_norm": 0.6440123319625854, "learning_rate": 8.276688634440216e-05, "loss": 0.0861, "step": 3090 }, { "grad_norm": 0.6634021997451782, "learning_rate": 8.26418148785107e-05, "loss": 0.0912, "step": 3100 }, { "grad_norm": 0.6206750273704529, "learning_rate": 8.251638644696141e-05, "loss": 0.08, "step": 3110 }, { "grad_norm": 0.6597092151641846, "learning_rate": 8.23906024214195e-05, "loss": 0.0823, "step": 3120 }, { "grad_norm": 0.7210692763328552, "learning_rate": 8.226446417743897e-05, "loss": 0.077, "step": 3130 }, { "grad_norm": 0.613390326499939, "learning_rate": 8.213797309444742e-05, "loss": 0.0795, "step": 3140 }, { "grad_norm": 0.5780166387557983, "learning_rate": 8.201113055573105e-05, "loss": 0.0742, "step": 3150 }, { "grad_norm": 0.6366831064224243, "learning_rate": 8.188393794841958e-05, "loss": 0.0873, "step": 3160 }, { "grad_norm": 0.809515118598938, "learning_rate": 8.175639666347094e-05, "loss": 0.0753, "step": 3170 }, { "grad_norm": 0.7801691889762878, "learning_rate": 8.162850809565623e-05, "loss": 0.0835, "step": 3180 }, { "grad_norm": 0.6820812821388245, "learning_rate": 8.150027364354431e-05, "loss": 0.0763, "step": 3190 }, { "grad_norm": 0.7042929530143738, "learning_rate": 8.137169470948662e-05, "loss": 0.0891, "step": 3200 }, { "grad_norm": 0.599989116191864, "learning_rate": 8.124277269960179e-05, "loss": 0.0804, "step": 3210 }, { "grad_norm": 0.7096439003944397, "learning_rate": 8.111350902376023e-05, "loss": 0.0784, "step": 3220 }, { "grad_norm": 0.6960294842720032, "learning_rate": 8.098390509556883e-05, "loss": 0.077, "step": 3230 }, { "grad_norm": 0.7003394365310669, "learning_rate": 8.085396233235536e-05, "loss": 0.0843, "step": 3240 }, { "grad_norm": 0.8249390125274658, "learning_rate": 8.072368215515306e-05, "loss": 0.0893, "step": 3250 }, { "grad_norm": 0.6759915947914124, "learning_rate": 8.059306598868506e-05, "loss": 0.083, "step": 3260 }, { "grad_norm": 0.6599640250205994, "learning_rate": 8.046211526134888e-05, "loss": 0.0909, "step": 3270 }, { "grad_norm": 0.5901545286178589, "learning_rate": 8.033083140520065e-05, "loss": 0.0781, "step": 3280 }, { "grad_norm": 0.6479388475418091, "learning_rate": 8.019921585593962e-05, "loss": 0.0792, "step": 3290 }, { "grad_norm": 0.6577224135398865, "learning_rate": 8.006727005289232e-05, "loss": 0.0874, "step": 3300 }, { "grad_norm": 0.6495193243026733, "learning_rate": 7.993499543899692e-05, "loss": 0.0776, "step": 3310 }, { "grad_norm": 0.60964035987854, "learning_rate": 7.980239346078742e-05, "loss": 0.0763, "step": 3320 }, { "grad_norm": 0.6103464365005493, "learning_rate": 7.966946556837778e-05, "loss": 0.0813, "step": 3330 }, { "grad_norm": 0.6395727396011353, "learning_rate": 7.953621321544616e-05, "loss": 0.0826, "step": 3340 }, { "grad_norm": 0.763139009475708, "learning_rate": 7.940263785921896e-05, "loss": 0.0833, "step": 3350 }, { "grad_norm": 0.6148580312728882, "learning_rate": 7.926874096045482e-05, "loss": 0.073, "step": 3360 }, { "grad_norm": 0.6035718321800232, "learning_rate": 7.913452398342881e-05, "loss": 0.08, "step": 3370 }, { "grad_norm": 0.5609293580055237, "learning_rate": 7.89999883959163e-05, "loss": 0.0842, "step": 3380 }, { "grad_norm": 0.6097584366798401, "learning_rate": 7.886513566917687e-05, "loss": 0.0874, "step": 3390 }, { "grad_norm": 0.6336648464202881, "learning_rate": 7.872996727793838e-05, "loss": 0.0787, "step": 3400 }, { "grad_norm": 0.6050973534584045, "learning_rate": 7.859448470038069e-05, "loss": 0.0848, "step": 3410 }, { "grad_norm": 0.6018955707550049, "learning_rate": 7.845868941811956e-05, "loss": 0.0768, "step": 3420 }, { "grad_norm": 0.7884665131568909, "learning_rate": 7.832258291619043e-05, "loss": 0.0732, "step": 3430 }, { "grad_norm": 0.552079975605011, "learning_rate": 7.81861666830322e-05, "loss": 0.0776, "step": 3440 }, { "grad_norm": 0.5439578890800476, "learning_rate": 7.804944221047097e-05, "loss": 0.0796, "step": 3450 }, { "grad_norm": 0.7273028492927551, "learning_rate": 7.791241099370364e-05, "loss": 0.0784, "step": 3460 }, { "grad_norm": 0.6264462471008301, "learning_rate": 7.777507453128163e-05, "loss": 0.0889, "step": 3470 }, { "grad_norm": 0.5623062252998352, "learning_rate": 7.763743432509451e-05, "loss": 0.078, "step": 3480 }, { "grad_norm": 0.622681200504303, "learning_rate": 7.749949188035353e-05, "loss": 0.0755, "step": 3490 }, { "eval/loss": 0.09640956759452819, "step": 3500 }, { "grad_norm": 0.5653059482574463, "learning_rate": 7.736124870557516e-05, "loss": 0.0754, "step": 3500 }, { "grad_norm": 0.5897047519683838, "learning_rate": 7.722270631256459e-05, "loss": 0.0805, "step": 3510 }, { "grad_norm": 0.8080556988716125, "learning_rate": 7.708386621639925e-05, "loss": 0.0858, "step": 3520 }, { "grad_norm": 0.808615505695343, "learning_rate": 7.694472993541219e-05, "loss": 0.0786, "step": 3530 }, { "grad_norm": 0.6473509073257446, "learning_rate": 7.680529899117547e-05, "loss": 0.0739, "step": 3540 }, { "grad_norm": 0.6235729455947876, "learning_rate": 7.666557490848358e-05, "loss": 0.0884, "step": 3550 }, { "grad_norm": 0.587633490562439, "learning_rate": 7.65255592153367e-05, "loss": 0.0782, "step": 3560 }, { "grad_norm": 0.6852300763130188, "learning_rate": 7.638525344292402e-05, "loss": 0.0813, "step": 3570 }, { "grad_norm": 0.6202707886695862, "learning_rate": 7.624465912560697e-05, "loss": 0.0802, "step": 3580 }, { "grad_norm": 0.5827509164810181, "learning_rate": 7.610377780090249e-05, "loss": 0.0804, "step": 3590 }, { "grad_norm": 0.6174249649047852, "learning_rate": 7.596261100946618e-05, "loss": 0.0821, "step": 3600 }, { "grad_norm": 0.6508203148841858, "learning_rate": 7.582116029507542e-05, "loss": 0.075, "step": 3610 }, { "grad_norm": 0.7538310885429382, "learning_rate": 7.56794272046126e-05, "loss": 0.0881, "step": 3620 }, { "grad_norm": 0.6635108590126038, "learning_rate": 7.55374132880481e-05, "loss": 0.0804, "step": 3630 }, { "grad_norm": 0.5851735472679138, "learning_rate": 7.539512009842333e-05, "loss": 0.0767, "step": 3640 }, { "grad_norm": 0.5552839636802673, "learning_rate": 7.525254919183382e-05, "loss": 0.0801, "step": 3650 }, { "grad_norm": 0.5505324602127075, "learning_rate": 7.510970212741215e-05, "loss": 0.0762, "step": 3660 }, { "grad_norm": 0.6016770005226135, "learning_rate": 7.496658046731096e-05, "loss": 0.0734, "step": 3670 }, { "grad_norm": 0.6466760635375977, "learning_rate": 7.482318577668578e-05, "loss": 0.0795, "step": 3680 }, { "grad_norm": 0.5529597401618958, "learning_rate": 7.467951962367796e-05, "loss": 0.0827, "step": 3690 }, { "grad_norm": 0.4894736707210541, "learning_rate": 7.453558357939755e-05, "loss": 0.0715, "step": 3700 }, { "grad_norm": 0.5741005539894104, "learning_rate": 7.439137921790606e-05, "loss": 0.0841, "step": 3710 }, { "grad_norm": 0.5835148096084595, "learning_rate": 7.42469081161993e-05, "loss": 0.0777, "step": 3720 }, { "grad_norm": 0.641464114189148, "learning_rate": 7.410217185419006e-05, "loss": 0.0725, "step": 3730 }, { "grad_norm": 0.6527230143547058, "learning_rate": 7.395717201469095e-05, "loss": 0.0758, "step": 3740 }, { "grad_norm": 0.633739709854126, "learning_rate": 7.381191018339696e-05, "loss": 0.083, "step": 3750 }, { "grad_norm": 0.5952082872390747, "learning_rate": 7.36663879488682e-05, "loss": 0.079, "step": 3760 }, { "grad_norm": 0.6131454706192017, "learning_rate": 7.352060690251254e-05, "loss": 0.075, "step": 3770 }, { "grad_norm": 0.6006690263748169, "learning_rate": 7.337456863856811e-05, "loss": 0.0758, "step": 3780 }, { "grad_norm": 0.5432766079902649, "learning_rate": 7.3228274754086e-05, "loss": 0.0787, "step": 3790 }, { "grad_norm": 0.694949746131897, "learning_rate": 7.308172684891267e-05, "loss": 0.0739, "step": 3800 }, { "grad_norm": 0.7675940990447998, "learning_rate": 7.293492652567255e-05, "loss": 0.0677, "step": 3810 }, { "grad_norm": 0.69501793384552, "learning_rate": 7.278787538975043e-05, "loss": 0.0789, "step": 3820 }, { "grad_norm": 0.6174379587173462, "learning_rate": 7.2640575049274e-05, "loss": 0.0828, "step": 3830 }, { "grad_norm": 0.5873093605041504, "learning_rate": 7.249302711509616e-05, "loss": 0.0811, "step": 3840 }, { "grad_norm": 0.5893351435661316, "learning_rate": 7.23452332007775e-05, "loss": 0.0715, "step": 3850 }, { "grad_norm": 0.7077029347419739, "learning_rate": 7.219719492256858e-05, "loss": 0.0706, "step": 3860 }, { "grad_norm": 0.5139503479003906, "learning_rate": 7.20489138993923e-05, "loss": 0.0706, "step": 3870 }, { "grad_norm": 0.5573222637176514, "learning_rate": 7.190039175282614e-05, "loss": 0.0683, "step": 3880 }, { "grad_norm": 0.569512128829956, "learning_rate": 7.175163010708455e-05, "loss": 0.0822, "step": 3890 }, { "grad_norm": 0.5464267134666443, "learning_rate": 7.1602630589001e-05, "loss": 0.0764, "step": 3900 }, { "grad_norm": 0.5804369449615479, "learning_rate": 7.14533948280104e-05, "loss": 0.0789, "step": 3910 }, { "grad_norm": 0.5735273957252502, "learning_rate": 7.130392445613109e-05, "loss": 0.0683, "step": 3920 }, { "grad_norm": 0.6597340106964111, "learning_rate": 7.115422110794711e-05, "loss": 0.0695, "step": 3930 }, { "grad_norm": 0.5542530417442322, "learning_rate": 7.100428642059033e-05, "loss": 0.0703, "step": 3940 }, { "grad_norm": 0.6993482708930969, "learning_rate": 7.08541220337224e-05, "loss": 0.0767, "step": 3950 }, { "grad_norm": 0.5256457924842834, "learning_rate": 7.070372958951706e-05, "loss": 0.0688, "step": 3960 }, { "grad_norm": 0.547267496585846, "learning_rate": 7.055311073264194e-05, "loss": 0.0696, "step": 3970 }, { "grad_norm": 0.5011000037193298, "learning_rate": 7.040226711024077e-05, "loss": 0.0676, "step": 3980 }, { "grad_norm": 0.6491769552230835, "learning_rate": 7.02512003719152e-05, "loss": 0.0712, "step": 3990 }, { "eval/loss": 0.09708894073963165, "step": 4000 }, { "grad_norm": 0.5656026601791382, "learning_rate": 7.00999121697069e-05, "loss": 0.0716, "step": 4000 }, { "grad_norm": 0.5710963010787964, "learning_rate": 6.99484041580794e-05, "loss": 0.0744, "step": 4010 }, { "grad_norm": 0.5667974352836609, "learning_rate": 6.979667799390004e-05, "loss": 0.0767, "step": 4020 }, { "grad_norm": 0.4471123516559601, "learning_rate": 6.964473533642185e-05, "loss": 0.0735, "step": 4030 }, { "grad_norm": 0.5450411438941956, "learning_rate": 6.949257784726539e-05, "loss": 0.0813, "step": 4040 }, { "grad_norm": 0.5534598231315613, "learning_rate": 6.934020719040056e-05, "loss": 0.0657, "step": 4050 }, { "grad_norm": 0.5396161675453186, "learning_rate": 6.918762503212848e-05, "loss": 0.0755, "step": 4060 }, { "grad_norm": 0.6875944137573242, "learning_rate": 6.903483304106319e-05, "loss": 0.0751, "step": 4070 }, { "grad_norm": 0.661061704158783, "learning_rate": 6.888183288811341e-05, "loss": 0.0754, "step": 4080 }, { "grad_norm": 0.5431751608848572, "learning_rate": 6.87286262464643e-05, "loss": 0.0763, "step": 4090 }, { "grad_norm": 0.5911586284637451, "learning_rate": 6.857521479155915e-05, "loss": 0.0713, "step": 4100 }, { "grad_norm": 0.5542717576026917, "learning_rate": 6.842160020108104e-05, "loss": 0.0858, "step": 4110 }, { "grad_norm": 0.5695271492004395, "learning_rate": 6.826778415493455e-05, "loss": 0.0719, "step": 4120 }, { "grad_norm": 0.6780376434326172, "learning_rate": 6.811376833522729e-05, "loss": 0.0729, "step": 4130 }, { "grad_norm": 0.5736984610557556, "learning_rate": 6.795955442625159e-05, "loss": 0.0729, "step": 4140 }, { "grad_norm": 0.5208945870399475, "learning_rate": 6.780514411446608e-05, "loss": 0.0688, "step": 4150 }, { "grad_norm": 0.548612654209137, "learning_rate": 6.765053908847716e-05, "loss": 0.0664, "step": 4160 }, { "grad_norm": 0.5552271604537964, "learning_rate": 6.749574103902064e-05, "loss": 0.0693, "step": 4170 }, { "grad_norm": 0.4939730167388916, "learning_rate": 6.734075165894317e-05, "loss": 0.07, "step": 4180 }, { "grad_norm": 0.5729985237121582, "learning_rate": 6.71855726431838e-05, "loss": 0.0679, "step": 4190 }, { "grad_norm": 0.5953519940376282, "learning_rate": 6.703020568875538e-05, "loss": 0.0727, "step": 4200 }, { "grad_norm": 0.6199027299880981, "learning_rate": 6.687465249472603e-05, "loss": 0.0716, "step": 4210 }, { "grad_norm": 0.5546400547027588, "learning_rate": 6.671891476220055e-05, "loss": 0.0709, "step": 4220 }, { "grad_norm": 0.46127140522003174, "learning_rate": 6.656299419430183e-05, "loss": 0.0658, "step": 4230 }, { "grad_norm": 0.577629029750824, "learning_rate": 6.640689249615223e-05, "loss": 0.0721, "step": 4240 }, { "grad_norm": 0.6214181780815125, "learning_rate": 6.625061137485491e-05, "loss": 0.0704, "step": 4250 }, { "grad_norm": 0.4957064390182495, "learning_rate": 6.609415253947517e-05, "loss": 0.0801, "step": 4260 }, { "grad_norm": 0.5450161099433899, "learning_rate": 6.593751770102178e-05, "loss": 0.0733, "step": 4270 }, { "grad_norm": 0.5261049270629883, "learning_rate": 6.578070857242823e-05, "loss": 0.0589, "step": 4280 }, { "grad_norm": 0.5709089040756226, "learning_rate": 6.562372686853402e-05, "loss": 0.0755, "step": 4290 }, { "grad_norm": 0.59120774269104, "learning_rate": 6.546657430606593e-05, "loss": 0.0759, "step": 4300 }, { "grad_norm": 0.6718469858169556, "learning_rate": 6.530925260361918e-05, "loss": 0.0714, "step": 4310 }, { "grad_norm": 0.5931757092475891, "learning_rate": 6.515176348163871e-05, "loss": 0.0743, "step": 4320 }, { "grad_norm": 0.5021288394927979, "learning_rate": 6.499410866240032e-05, "loss": 0.0689, "step": 4330 }, { "grad_norm": 0.49451255798339844, "learning_rate": 6.48362898699919e-05, "loss": 0.0633, "step": 4340 }, { "grad_norm": 0.6468533277511597, "learning_rate": 6.467830883029443e-05, "loss": 0.0697, "step": 4350 }, { "grad_norm": 0.6061948537826538, "learning_rate": 6.452016727096326e-05, "loss": 0.069, "step": 4360 }, { "grad_norm": 0.4761110842227936, "learning_rate": 6.436186692140916e-05, "loss": 0.0714, "step": 4370 }, { "grad_norm": 0.5411221385002136, "learning_rate": 6.420340951277938e-05, "loss": 0.0735, "step": 4380 }, { "grad_norm": 0.6015405654907227, "learning_rate": 6.404479677793874e-05, "loss": 0.0667, "step": 4390 }, { "grad_norm": 0.5228006839752197, "learning_rate": 6.388603045145075e-05, "loss": 0.0699, "step": 4400 }, { "grad_norm": 0.6190991997718811, "learning_rate": 6.372711226955843e-05, "loss": 0.0687, "step": 4410 }, { "grad_norm": 0.6963168382644653, "learning_rate": 6.356804397016564e-05, "loss": 0.0644, "step": 4420 }, { "grad_norm": 0.5099686980247498, "learning_rate": 6.340882729281779e-05, "loss": 0.0766, "step": 4430 }, { "grad_norm": 0.5546655058860779, "learning_rate": 6.324946397868294e-05, "loss": 0.0807, "step": 4440 }, { "grad_norm": 0.5260878205299377, "learning_rate": 6.308995577053276e-05, "loss": 0.0703, "step": 4450 }, { "grad_norm": 0.520291805267334, "learning_rate": 6.293030441272347e-05, "loss": 0.0679, "step": 4460 }, { "grad_norm": 0.46433889865875244, "learning_rate": 6.277051165117677e-05, "loss": 0.0675, "step": 4470 }, { "grad_norm": 0.5065008997917175, "learning_rate": 6.261057923336064e-05, "loss": 0.0712, "step": 4480 }, { "grad_norm": 0.5910764336585999, "learning_rate": 6.245050890827042e-05, "loss": 0.0711, "step": 4490 }, { "eval/loss": 0.0965606552362442, "step": 4500 }, { "grad_norm": 0.6224591135978699, "learning_rate": 6.229030242640952e-05, "loss": 0.0674, "step": 4500 }, { "grad_norm": 0.6442241072654724, "learning_rate": 6.212996153977037e-05, "loss": 0.0643, "step": 4510 }, { "grad_norm": 0.5292598009109497, "learning_rate": 6.196948800181523e-05, "loss": 0.0738, "step": 4520 }, { "grad_norm": 0.49640458822250366, "learning_rate": 6.180888356745695e-05, "loss": 0.0742, "step": 4530 }, { "grad_norm": 0.5422980189323425, "learning_rate": 6.164814999303995e-05, "loss": 0.0707, "step": 4540 }, { "grad_norm": 0.5385618209838867, "learning_rate": 6.148728903632081e-05, "loss": 0.0715, "step": 4550 }, { "grad_norm": 0.5169300436973572, "learning_rate": 6.132630245644921e-05, "loss": 0.0696, "step": 4560 }, { "grad_norm": 0.6023463010787964, "learning_rate": 6.116519201394857e-05, "loss": 0.0726, "step": 4570 }, { "grad_norm": 0.5851990580558777, "learning_rate": 6.10039594706969e-05, "loss": 0.0719, "step": 4580 }, { "grad_norm": 0.5784035921096802, "learning_rate": 6.084260658990744e-05, "loss": 0.0704, "step": 4590 }, { "grad_norm": 0.6393957734107971, "learning_rate": 6.068113513610943e-05, "loss": 0.0685, "step": 4600 }, { "grad_norm": 0.5436510443687439, "learning_rate": 6.0519546875128876e-05, "loss": 0.0724, "step": 4610 }, { "grad_norm": 0.517709493637085, "learning_rate": 6.035784357406906e-05, "loss": 0.0639, "step": 4620 }, { "grad_norm": 0.5001219511032104, "learning_rate": 6.01960270012914e-05, "loss": 0.0761, "step": 4630 }, { "grad_norm": 0.5643563866615295, "learning_rate": 6.003409892639599e-05, "loss": 0.0764, "step": 4640 }, { "grad_norm": 0.6038289666175842, "learning_rate": 5.9872061120202336e-05, "loss": 0.0756, "step": 4650 }, { "grad_norm": 0.6031238436698914, "learning_rate": 5.9709915354729914e-05, "loss": 0.0719, "step": 4660 }, { "grad_norm": 0.6391696929931641, "learning_rate": 5.9547663403178824e-05, "loss": 0.0839, "step": 4670 }, { "grad_norm": 0.5629892349243164, "learning_rate": 5.9385307039910445e-05, "loss": 0.0651, "step": 4680 }, { "grad_norm": 0.6046562790870667, "learning_rate": 5.922284804042792e-05, "loss": 0.0666, "step": 4690 }, { "grad_norm": 0.6428664922714233, "learning_rate": 5.906028818135687e-05, "loss": 0.0672, "step": 4700 }, { "grad_norm": 0.49685654044151306, "learning_rate": 5.889762924042585e-05, "loss": 0.0736, "step": 4710 }, { "grad_norm": 0.542539119720459, "learning_rate": 5.873487299644699e-05, "loss": 0.0604, "step": 4720 }, { "grad_norm": 0.5018340945243835, "learning_rate": 5.857202122929649e-05, "loss": 0.0708, "step": 4730 }, { "grad_norm": 0.5392184257507324, "learning_rate": 5.840907571989518e-05, "loss": 0.0767, "step": 4740 }, { "grad_norm": 0.49123615026474, "learning_rate": 5.824603825018904e-05, "loss": 0.0821, "step": 4750 }, { "grad_norm": 0.5722000002861023, "learning_rate": 5.808291060312975e-05, "loss": 0.0753, "step": 4760 }, { "grad_norm": 0.49736565351486206, "learning_rate": 5.7919694562655083e-05, "loss": 0.0734, "step": 4770 }, { "grad_norm": 0.597605288028717, "learning_rate": 5.775639191366954e-05, "loss": 0.0702, "step": 4780 }, { "grad_norm": 0.5945085287094116, "learning_rate": 5.75930044420247e-05, "loss": 0.0697, "step": 4790 }, { "grad_norm": 0.5786648988723755, "learning_rate": 5.74295339344998e-05, "loss": 0.063, "step": 4800 }, { "grad_norm": 0.5226376056671143, "learning_rate": 5.726598217878211e-05, "loss": 0.0725, "step": 4810 }, { "grad_norm": 0.6626110076904297, "learning_rate": 5.71023509634474e-05, "loss": 0.0783, "step": 4820 }, { "grad_norm": 0.6417555212974548, "learning_rate": 5.693864207794049e-05, "loss": 0.0737, "step": 4830 }, { "grad_norm": 0.4717249572277069, "learning_rate": 5.677485731255545e-05, "loss": 0.0751, "step": 4840 }, { "grad_norm": 0.5369865894317627, "learning_rate": 5.6610998458416296e-05, "loss": 0.0766, "step": 4850 }, { "grad_norm": 0.6259351968765259, "learning_rate": 5.644706730745716e-05, "loss": 0.069, "step": 4860 }, { "grad_norm": 0.5614010095596313, "learning_rate": 5.628306565240287e-05, "loss": 0.0746, "step": 4870 }, { "grad_norm": 0.6325775980949402, "learning_rate": 5.611899528674923e-05, "loss": 0.0643, "step": 4880 }, { "grad_norm": 0.5036805868148804, "learning_rate": 5.595485800474349e-05, "loss": 0.0754, "step": 4890 }, { "grad_norm": 0.5675097107887268, "learning_rate": 5.579065560136467e-05, "loss": 0.0684, "step": 4900 }, { "grad_norm": 0.6225830912590027, "learning_rate": 5.562638987230392e-05, "loss": 0.077, "step": 4910 }, { "grad_norm": 0.5745845437049866, "learning_rate": 5.546206261394498e-05, "loss": 0.079, "step": 4920 }, { "grad_norm": 0.6023924350738525, "learning_rate": 5.529767562334437e-05, "loss": 0.0683, "step": 4930 }, { "grad_norm": 0.6788532137870789, "learning_rate": 5.5133230698211926e-05, "loss": 0.0695, "step": 4940 }, { "grad_norm": 0.617906391620636, "learning_rate": 5.496872963689096e-05, "loss": 0.0684, "step": 4950 }, { "grad_norm": 0.5210251808166504, "learning_rate": 5.4804174238338756e-05, "loss": 0.0697, "step": 4960 }, { "grad_norm": 0.6266798377037048, "learning_rate": 5.463956630210678e-05, "loss": 0.0628, "step": 4970 }, { "grad_norm": 0.5564236044883728, "learning_rate": 5.4474907628321046e-05, "loss": 0.0742, "step": 4980 }, { "grad_norm": 0.5566116571426392, "learning_rate": 5.431020001766244e-05, "loss": 0.0655, "step": 4990 }, { "eval/loss": 0.09668893575668334, "step": 5000 }, { "grad_norm": 0.5017706751823425, "learning_rate": 5.4145445271346986e-05, "loss": 0.0798, "step": 5000 } ], "logging_steps": 10, "max_steps": 10000, "num_input_tokens_seen": 0, "num_train_epochs": 9223372036854775807, "save_steps": 1000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 32, "trial_name": null, "trial_params": null }