fpadovani/rus-cyrl-10mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
19449022671cb5399182193585c681ecd65046fa
rus-cyrl-10mb-100mb_seed3407/checkpoint-6000/trainer_state.json
ModelHub XC 1944902267 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/rus-cyrl-10mb-100mb_seed3407
Source: Original Platform
2026-07-12 10:01:17 +08:00

12057 lines
330 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5243380232456524,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742575645446777,
"epoch": 0.0004369483527047103,
"grad_norm": 4.875,
"learning_rate": 2e-06,
"loss": 10.7635,
"mean_token_accuracy": 0.00012376237427815794,
"num_tokens": 10478.0,
"step": 5
},
{
"entropy": 10.742580890655518,
"epoch": 0.0008738967054094206,
"grad_norm": 5.25,
"learning_rate": 4.5e-06,
"loss": 10.7564,
"mean_token_accuracy": 5.527915782295167e-05,
"num_tokens": 21152.0,
"step": 10
},
{
"entropy": 10.74257001876831,
"epoch": 0.0013108450581141308,
"grad_norm": 5.15625,
"learning_rate": 7e-06,
"loss": 10.7163,
"mean_token_accuracy": 6.680026999674737e-05,
"num_tokens": 31246.0,
"step": 15
},
{
"entropy": 10.742575645446777,
"epoch": 0.0017477934108188413,
"grad_norm": 4.75,
"learning_rate": 9.5e-06,
"loss": 10.6955,
"mean_token_accuracy": 0.0014378628693521023,
"num_tokens": 41934.0,
"step": 20
},
{
"entropy": 10.742451190948486,
"epoch": 0.0021847417635235517,
"grad_norm": 4.09375,
"learning_rate": 1.2e-05,
"loss": 10.5979,
"mean_token_accuracy": 0.01751632974483073,
"num_tokens": 52830.0,
"step": 25
},
{
"entropy": 10.741966247558594,
"epoch": 0.0026216901162282617,
"grad_norm": 3.875,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.4324,
"mean_token_accuracy": 0.046310391277074814,
"num_tokens": 62706.0,
"step": 30
},
{
"entropy": 10.739739894866943,
"epoch": 0.003058638468932972,
"grad_norm": 3.15625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.3067,
"mean_token_accuracy": 0.046505602076649666,
"num_tokens": 72619.0,
"step": 35
},
{
"entropy": 10.732788467407227,
"epoch": 0.0034955868216376825,
"grad_norm": 2.90625,
"learning_rate": 1.95e-05,
"loss": 10.198,
"mean_token_accuracy": 0.04933134503662586,
"num_tokens": 82903.0,
"step": 40
},
{
"entropy": 10.720102882385254,
"epoch": 0.003932535174342393,
"grad_norm": 2.21875,
"learning_rate": 2.2e-05,
"loss": 10.0464,
"mean_token_accuracy": 0.06326089426875114,
"num_tokens": 94243.0,
"step": 45
},
{
"entropy": 10.703330898284912,
"epoch": 0.004369483527047103,
"grad_norm": 2.125,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.0094,
"mean_token_accuracy": 0.05352335274219513,
"num_tokens": 103618.0,
"step": 50
},
{
"entropy": 10.69249677658081,
"epoch": 0.004806431879751813,
"grad_norm": 1.8671875,
"learning_rate": 2.7e-05,
"loss": 9.9487,
"mean_token_accuracy": 0.048276785761117935,
"num_tokens": 112180.0,
"step": 55
},
{
"entropy": 10.693581295013427,
"epoch": 0.005243380232456523,
"grad_norm": 2.140625,
"learning_rate": 2.95e-05,
"loss": 9.9131,
"mean_token_accuracy": 0.05066986009478569,
"num_tokens": 122462.0,
"step": 60
},
{
"entropy": 10.694738101959228,
"epoch": 0.005680328585161234,
"grad_norm": 1.8671875,
"learning_rate": 3.2e-05,
"loss": 9.8716,
"mean_token_accuracy": 0.05316077768802643,
"num_tokens": 132741.0,
"step": 65
},
{
"entropy": 10.687106227874756,
"epoch": 0.006117276937865944,
"grad_norm": 2.015625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.7556,
"mean_token_accuracy": 0.055888672918081285,
"num_tokens": 142973.0,
"step": 70
},
{
"entropy": 10.674335384368897,
"epoch": 0.006554225290570655,
"grad_norm": 1.8046875,
"learning_rate": 3.7e-05,
"loss": 9.7947,
"mean_token_accuracy": 0.04946387633681297,
"num_tokens": 153199.0,
"step": 75
},
{
"entropy": 10.660592651367187,
"epoch": 0.006991173643275365,
"grad_norm": 1.8203125,
"learning_rate": 3.95e-05,
"loss": 9.6937,
"mean_token_accuracy": 0.05424195118248463,
"num_tokens": 163079.0,
"step": 80
},
{
"entropy": 10.647120380401612,
"epoch": 0.0074281219959800755,
"grad_norm": 1.7578125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.6379,
"mean_token_accuracy": 0.05559157766401768,
"num_tokens": 173308.0,
"step": 85
},
{
"entropy": 10.629187202453613,
"epoch": 0.007865070348684786,
"grad_norm": 1.65625,
"learning_rate": 4.45e-05,
"loss": 9.5576,
"mean_token_accuracy": 0.0650589868426323,
"num_tokens": 183856.0,
"step": 90
},
{
"entropy": 10.605023765563965,
"epoch": 0.008302018701389496,
"grad_norm": 1.796875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.4922,
"mean_token_accuracy": 0.055145786702632905,
"num_tokens": 193368.0,
"step": 95
},
{
"entropy": 10.602028179168702,
"epoch": 0.008738967054094207,
"grad_norm": 1.78125,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.4365,
"mean_token_accuracy": 0.05791991353034973,
"num_tokens": 202686.0,
"step": 100
},
{
"entropy": 10.589914703369141,
"epoch": 0.009175915406798917,
"grad_norm": 1.7578125,
"learning_rate": 5.2e-05,
"loss": 9.3332,
"mean_token_accuracy": 0.06775561273097992,
"num_tokens": 213237.0,
"step": 105
},
{
"entropy": 10.540931034088135,
"epoch": 0.009612863759503626,
"grad_norm": 1.59375,
"learning_rate": 5.45e-05,
"loss": 9.2707,
"mean_token_accuracy": 0.0691582515835762,
"num_tokens": 222406.0,
"step": 110
},
{
"entropy": 10.49779634475708,
"epoch": 0.010049812112208336,
"grad_norm": 1.671875,
"learning_rate": 5.7e-05,
"loss": 9.2158,
"mean_token_accuracy": 0.06427372209727764,
"num_tokens": 232139.0,
"step": 115
},
{
"entropy": 10.439636135101319,
"epoch": 0.010486760464913047,
"grad_norm": 1.40625,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.1331,
"mean_token_accuracy": 0.06220344454050064,
"num_tokens": 242438.0,
"step": 120
},
{
"entropy": 10.37880210876465,
"epoch": 0.010923708817617757,
"grad_norm": 1.4375,
"learning_rate": 6.2e-05,
"loss": 9.0192,
"mean_token_accuracy": 0.06697547994554043,
"num_tokens": 253082.0,
"step": 125
},
{
"entropy": 10.290444278717041,
"epoch": 0.011360657170322468,
"grad_norm": 1.4140625,
"learning_rate": 6.450000000000001e-05,
"loss": 8.9381,
"mean_token_accuracy": 0.07001210227608681,
"num_tokens": 264067.0,
"step": 130
},
{
"entropy": 10.230575561523438,
"epoch": 0.011797605523027178,
"grad_norm": 1.34375,
"learning_rate": 6.7e-05,
"loss": 8.8679,
"mean_token_accuracy": 0.06120241619646549,
"num_tokens": 274571.0,
"step": 135
},
{
"entropy": 10.212278461456298,
"epoch": 0.012234553875731888,
"grad_norm": 1.40625,
"learning_rate": 6.950000000000001e-05,
"loss": 8.7477,
"mean_token_accuracy": 0.06530353389680385,
"num_tokens": 284561.0,
"step": 140
},
{
"entropy": 10.070102214813232,
"epoch": 0.012671502228436599,
"grad_norm": 1.1796875,
"learning_rate": 7.2e-05,
"loss": 8.6078,
"mean_token_accuracy": 0.06655799858272075,
"num_tokens": 295366.0,
"step": 145
},
{
"entropy": 9.980097579956055,
"epoch": 0.01310845058114131,
"grad_norm": 1.0546875,
"learning_rate": 7.45e-05,
"loss": 8.6389,
"mean_token_accuracy": 0.06603652313351631,
"num_tokens": 305423.0,
"step": 150
},
{
"entropy": 9.907401084899902,
"epoch": 0.01354539893384602,
"grad_norm": 1.046875,
"learning_rate": 7.7e-05,
"loss": 8.4744,
"mean_token_accuracy": 0.06612432450056076,
"num_tokens": 315182.0,
"step": 155
},
{
"entropy": 9.672210597991944,
"epoch": 0.01398234728655073,
"grad_norm": 1.0,
"learning_rate": 7.950000000000001e-05,
"loss": 8.4101,
"mean_token_accuracy": 0.07041444033384323,
"num_tokens": 324780.0,
"step": 160
},
{
"entropy": 9.59818925857544,
"epoch": 0.01441929563925544,
"grad_norm": 0.8515625,
"learning_rate": 8.2e-05,
"loss": 8.3853,
"mean_token_accuracy": 0.06835086569190026,
"num_tokens": 335104.0,
"step": 165
},
{
"entropy": 9.408317279815673,
"epoch": 0.014856243991960151,
"grad_norm": 0.99609375,
"learning_rate": 8.450000000000001e-05,
"loss": 8.3064,
"mean_token_accuracy": 0.06708988398313523,
"num_tokens": 345073.0,
"step": 170
},
{
"entropy": 9.27857847213745,
"epoch": 0.015293192344664861,
"grad_norm": 0.87109375,
"learning_rate": 8.7e-05,
"loss": 8.2728,
"mean_token_accuracy": 0.06384357884526252,
"num_tokens": 354342.0,
"step": 175
},
{
"entropy": 9.09966974258423,
"epoch": 0.015730140697369572,
"grad_norm": 0.8203125,
"learning_rate": 8.95e-05,
"loss": 8.1513,
"mean_token_accuracy": 0.07314991652965545,
"num_tokens": 364441.0,
"step": 180
},
{
"entropy": 8.942261791229248,
"epoch": 0.01616708905007428,
"grad_norm": 0.7421875,
"learning_rate": 9.2e-05,
"loss": 8.1527,
"mean_token_accuracy": 0.06924629844725132,
"num_tokens": 373918.0,
"step": 185
},
{
"entropy": 8.88406400680542,
"epoch": 0.016604037402778993,
"grad_norm": 0.75,
"learning_rate": 9.45e-05,
"loss": 8.1632,
"mean_token_accuracy": 0.0712675966322422,
"num_tokens": 383273.0,
"step": 190
},
{
"entropy": 8.708986186981202,
"epoch": 0.0170409857554837,
"grad_norm": 0.796875,
"learning_rate": 9.7e-05,
"loss": 8.0308,
"mean_token_accuracy": 0.08221853896975517,
"num_tokens": 393262.0,
"step": 195
},
{
"entropy": 8.639312553405762,
"epoch": 0.017477934108188414,
"grad_norm": 0.73046875,
"learning_rate": 9.95e-05,
"loss": 8.1392,
"mean_token_accuracy": 0.06822756379842758,
"num_tokens": 402389.0,
"step": 200
},
{
"entropy": 8.639290618896485,
"epoch": 0.017914882460893122,
"grad_norm": 0.77734375,
"learning_rate": 0.000102,
"loss": 8.0721,
"mean_token_accuracy": 0.06872168891131877,
"num_tokens": 412997.0,
"step": 205
},
{
"entropy": 8.622207546234131,
"epoch": 0.018351830813597834,
"grad_norm": 0.8203125,
"learning_rate": 0.00010449999999999999,
"loss": 8.0805,
"mean_token_accuracy": 0.07161624394357205,
"num_tokens": 422222.0,
"step": 210
},
{
"entropy": 8.481709098815918,
"epoch": 0.018788779166302543,
"grad_norm": 0.71875,
"learning_rate": 0.000107,
"loss": 8.0159,
"mean_token_accuracy": 0.07780317105352878,
"num_tokens": 432539.0,
"step": 215
},
{
"entropy": 8.539759254455566,
"epoch": 0.019225727519007252,
"grad_norm": 0.84765625,
"learning_rate": 0.0001095,
"loss": 8.018,
"mean_token_accuracy": 0.07488755434751511,
"num_tokens": 442311.0,
"step": 220
},
{
"entropy": 8.473109245300293,
"epoch": 0.019662675871711964,
"grad_norm": 0.7421875,
"learning_rate": 0.000112,
"loss": 8.0033,
"mean_token_accuracy": 0.07176320515573024,
"num_tokens": 452727.0,
"step": 225
},
{
"entropy": 8.49320011138916,
"epoch": 0.020099624224416673,
"grad_norm": 0.84375,
"learning_rate": 0.0001145,
"loss": 8.0269,
"mean_token_accuracy": 0.0723928865045309,
"num_tokens": 462461.0,
"step": 230
},
{
"entropy": 8.517394256591796,
"epoch": 0.020536572577121385,
"grad_norm": 1.0078125,
"learning_rate": 0.00011700000000000001,
"loss": 7.9988,
"mean_token_accuracy": 0.07492853850126266,
"num_tokens": 472362.0,
"step": 235
},
{
"entropy": 8.411276721954346,
"epoch": 0.020973520929826094,
"grad_norm": 0.79296875,
"learning_rate": 0.00011949999999999999,
"loss": 7.9901,
"mean_token_accuracy": 0.07653542906045914,
"num_tokens": 482718.0,
"step": 240
},
{
"entropy": 8.451101970672607,
"epoch": 0.021410469282530806,
"grad_norm": 0.76953125,
"learning_rate": 0.000122,
"loss": 7.9973,
"mean_token_accuracy": 0.07303892523050308,
"num_tokens": 492947.0,
"step": 245
},
{
"entropy": 8.451656246185303,
"epoch": 0.021847417635235514,
"grad_norm": 0.875,
"learning_rate": 0.0001245,
"loss": 8.0293,
"mean_token_accuracy": 0.07184793651103974,
"num_tokens": 502957.0,
"step": 250
},
{
"entropy": 8.451603507995605,
"epoch": 0.022284365987940227,
"grad_norm": 0.90625,
"learning_rate": 0.000127,
"loss": 7.9786,
"mean_token_accuracy": 0.07364074736833573,
"num_tokens": 513220.0,
"step": 255
},
{
"entropy": 8.423027420043946,
"epoch": 0.022721314340644935,
"grad_norm": 0.8671875,
"learning_rate": 0.0001295,
"loss": 7.927,
"mean_token_accuracy": 0.07086114585399628,
"num_tokens": 523217.0,
"step": 260
},
{
"entropy": 8.423391246795655,
"epoch": 0.023158262693349647,
"grad_norm": 0.85546875,
"learning_rate": 0.000132,
"loss": 7.9515,
"mean_token_accuracy": 0.0709399688988924,
"num_tokens": 533789.0,
"step": 265
},
{
"entropy": 8.354116344451905,
"epoch": 0.023595211046054356,
"grad_norm": 0.80078125,
"learning_rate": 0.00013450000000000002,
"loss": 7.9324,
"mean_token_accuracy": 0.07723044976592064,
"num_tokens": 544146.0,
"step": 270
},
{
"entropy": 8.401555442810059,
"epoch": 0.02403215939875907,
"grad_norm": 0.84375,
"learning_rate": 0.00013700000000000002,
"loss": 7.9685,
"mean_token_accuracy": 0.07366384789347649,
"num_tokens": 554215.0,
"step": 275
},
{
"entropy": 8.307286357879638,
"epoch": 0.024469107751463777,
"grad_norm": 0.77734375,
"learning_rate": 0.0001395,
"loss": 7.9334,
"mean_token_accuracy": 0.0711975559592247,
"num_tokens": 564587.0,
"step": 280
},
{
"entropy": 8.396804809570312,
"epoch": 0.024906056104168486,
"grad_norm": 0.73046875,
"learning_rate": 0.00014199999999999998,
"loss": 7.9794,
"mean_token_accuracy": 0.07162599861621857,
"num_tokens": 574819.0,
"step": 285
},
{
"entropy": 8.358118438720703,
"epoch": 0.025343004456873198,
"grad_norm": 0.796875,
"learning_rate": 0.0001445,
"loss": 7.8989,
"mean_token_accuracy": 0.07322670742869378,
"num_tokens": 585178.0,
"step": 290
},
{
"entropy": 8.294723510742188,
"epoch": 0.025779952809577907,
"grad_norm": 0.90625,
"learning_rate": 0.000147,
"loss": 7.8845,
"mean_token_accuracy": 0.07678839713335037,
"num_tokens": 595086.0,
"step": 295
},
{
"entropy": 8.285739517211914,
"epoch": 0.02621690116228262,
"grad_norm": 0.8125,
"learning_rate": 0.0001495,
"loss": 7.89,
"mean_token_accuracy": 0.08111602552235127,
"num_tokens": 604942.0,
"step": 300
},
{
"entropy": 8.310109424591065,
"epoch": 0.026653849514987327,
"grad_norm": 0.765625,
"learning_rate": 0.000152,
"loss": 7.874,
"mean_token_accuracy": 0.07880265414714813,
"num_tokens": 616082.0,
"step": 305
},
{
"entropy": 8.292809677124023,
"epoch": 0.02709079786769204,
"grad_norm": 1.0234375,
"learning_rate": 0.00015450000000000001,
"loss": 7.9016,
"mean_token_accuracy": 0.07674356997013092,
"num_tokens": 626375.0,
"step": 310
},
{
"entropy": 8.272511959075928,
"epoch": 0.027527746220396748,
"grad_norm": 0.82421875,
"learning_rate": 0.000157,
"loss": 7.8627,
"mean_token_accuracy": 0.07318086177110672,
"num_tokens": 637563.0,
"step": 315
},
{
"entropy": 8.222749423980712,
"epoch": 0.02796469457310146,
"grad_norm": 0.88671875,
"learning_rate": 0.0001595,
"loss": 7.8692,
"mean_token_accuracy": 0.07608433291316033,
"num_tokens": 647256.0,
"step": 320
},
{
"entropy": 8.308216381072999,
"epoch": 0.02840164292580617,
"grad_norm": 0.75,
"learning_rate": 0.000162,
"loss": 7.9245,
"mean_token_accuracy": 0.0743330918252468,
"num_tokens": 657247.0,
"step": 325
},
{
"entropy": 8.2763596534729,
"epoch": 0.02883859127851088,
"grad_norm": 1.0546875,
"learning_rate": 0.00016450000000000001,
"loss": 7.8367,
"mean_token_accuracy": 0.07250913083553315,
"num_tokens": 667613.0,
"step": 330
},
{
"entropy": 8.169933319091797,
"epoch": 0.02927553963121559,
"grad_norm": 1.0625,
"learning_rate": 0.00016700000000000002,
"loss": 7.8732,
"mean_token_accuracy": 0.07495946660637856,
"num_tokens": 678048.0,
"step": 335
},
{
"entropy": 8.294293117523193,
"epoch": 0.029712487983920302,
"grad_norm": 0.98046875,
"learning_rate": 0.00016950000000000003,
"loss": 7.865,
"mean_token_accuracy": 0.07813518717885018,
"num_tokens": 688468.0,
"step": 340
},
{
"entropy": 8.193762683868409,
"epoch": 0.03014943633662501,
"grad_norm": 0.84765625,
"learning_rate": 0.00017199999999999998,
"loss": 7.84,
"mean_token_accuracy": 0.07896131426095962,
"num_tokens": 699809.0,
"step": 345
},
{
"entropy": 8.181721496582032,
"epoch": 0.030586384689329723,
"grad_norm": 0.92578125,
"learning_rate": 0.00017449999999999999,
"loss": 7.7969,
"mean_token_accuracy": 0.08212044797837734,
"num_tokens": 709227.0,
"step": 350
},
{
"entropy": 8.286689472198486,
"epoch": 0.03102333304203443,
"grad_norm": 0.88671875,
"learning_rate": 0.000177,
"loss": 7.904,
"mean_token_accuracy": 0.0743263080716133,
"num_tokens": 720178.0,
"step": 355
},
{
"entropy": 8.163193130493164,
"epoch": 0.031460281394739144,
"grad_norm": 0.78125,
"learning_rate": 0.0001795,
"loss": 7.8202,
"mean_token_accuracy": 0.0798851165920496,
"num_tokens": 730514.0,
"step": 360
},
{
"entropy": 8.259191036224365,
"epoch": 0.03189722974744385,
"grad_norm": 0.8203125,
"learning_rate": 0.000182,
"loss": 7.8692,
"mean_token_accuracy": 0.07668052241206169,
"num_tokens": 741345.0,
"step": 365
},
{
"entropy": 8.17900333404541,
"epoch": 0.03233417810014856,
"grad_norm": 0.97265625,
"learning_rate": 0.0001845,
"loss": 7.7991,
"mean_token_accuracy": 0.08136411309242249,
"num_tokens": 751163.0,
"step": 370
},
{
"entropy": 8.175147342681885,
"epoch": 0.03277112645285327,
"grad_norm": 0.81640625,
"learning_rate": 0.000187,
"loss": 7.7993,
"mean_token_accuracy": 0.0767936147749424,
"num_tokens": 762090.0,
"step": 375
},
{
"entropy": 8.196157550811767,
"epoch": 0.033208074805557986,
"grad_norm": 1.125,
"learning_rate": 0.0001895,
"loss": 7.7674,
"mean_token_accuracy": 0.08601637631654739,
"num_tokens": 772049.0,
"step": 380
},
{
"entropy": 8.117916297912597,
"epoch": 0.03364502315826269,
"grad_norm": 1.109375,
"learning_rate": 0.000192,
"loss": 7.7365,
"mean_token_accuracy": 0.08381113037467003,
"num_tokens": 781280.0,
"step": 385
},
{
"entropy": 8.152852725982665,
"epoch": 0.0340819715109674,
"grad_norm": 0.89453125,
"learning_rate": 0.0001945,
"loss": 7.8156,
"mean_token_accuracy": 0.07903518378734589,
"num_tokens": 791642.0,
"step": 390
},
{
"entropy": 8.175984954833984,
"epoch": 0.034518919863672115,
"grad_norm": 0.9609375,
"learning_rate": 0.00019700000000000002,
"loss": 7.7973,
"mean_token_accuracy": 0.07584034129977227,
"num_tokens": 801224.0,
"step": 395
},
{
"entropy": 8.157032155990601,
"epoch": 0.03495586821637683,
"grad_norm": 0.94921875,
"learning_rate": 0.00019950000000000002,
"loss": 7.7379,
"mean_token_accuracy": 0.08061860464513301,
"num_tokens": 810727.0,
"step": 400
},
{
"entropy": 8.220977401733398,
"epoch": 0.03539281656908153,
"grad_norm": 1.0078125,
"learning_rate": 0.000202,
"loss": 7.8159,
"mean_token_accuracy": 0.07617722004652024,
"num_tokens": 819943.0,
"step": 405
},
{
"entropy": 8.027306842803956,
"epoch": 0.035829764921786245,
"grad_norm": 0.91796875,
"learning_rate": 0.00020449999999999998,
"loss": 7.744,
"mean_token_accuracy": 0.07786332815885544,
"num_tokens": 829474.0,
"step": 410
},
{
"entropy": 8.135616159439087,
"epoch": 0.03626671327449096,
"grad_norm": 1.0390625,
"learning_rate": 0.000207,
"loss": 7.7154,
"mean_token_accuracy": 0.07687773257493973,
"num_tokens": 839288.0,
"step": 415
},
{
"entropy": 8.123063278198241,
"epoch": 0.03670366162719567,
"grad_norm": 0.8984375,
"learning_rate": 0.0002095,
"loss": 7.7375,
"mean_token_accuracy": 0.08089107796549797,
"num_tokens": 850064.0,
"step": 420
},
{
"entropy": 8.010475349426269,
"epoch": 0.037140609979900374,
"grad_norm": 0.84765625,
"learning_rate": 0.000212,
"loss": 7.7137,
"mean_token_accuracy": 0.08113521486520767,
"num_tokens": 860732.0,
"step": 425
},
{
"entropy": 8.06641664505005,
"epoch": 0.037577558332605086,
"grad_norm": 0.9375,
"learning_rate": 0.0002145,
"loss": 7.7097,
"mean_token_accuracy": 0.08033490926027298,
"num_tokens": 870076.0,
"step": 430
},
{
"entropy": 7.975166702270508,
"epoch": 0.0380145066853098,
"grad_norm": 0.93359375,
"learning_rate": 0.00021700000000000002,
"loss": 7.6087,
"mean_token_accuracy": 0.08692155629396439,
"num_tokens": 880201.0,
"step": 435
},
{
"entropy": 8.02914743423462,
"epoch": 0.038451455038014504,
"grad_norm": 0.78125,
"learning_rate": 0.0002195,
"loss": 7.6725,
"mean_token_accuracy": 0.08405178114771843,
"num_tokens": 890344.0,
"step": 440
},
{
"entropy": 8.042040014266968,
"epoch": 0.038888403390719216,
"grad_norm": 0.921875,
"learning_rate": 0.000222,
"loss": 7.6622,
"mean_token_accuracy": 0.0840606488287449,
"num_tokens": 900443.0,
"step": 445
},
{
"entropy": 8.088546133041381,
"epoch": 0.03932535174342393,
"grad_norm": 0.89453125,
"learning_rate": 0.0002245,
"loss": 7.6868,
"mean_token_accuracy": 0.07948172986507415,
"num_tokens": 910284.0,
"step": 450
},
{
"entropy": 8.075697088241578,
"epoch": 0.03976230009612864,
"grad_norm": 0.984375,
"learning_rate": 0.00022700000000000002,
"loss": 7.7751,
"mean_token_accuracy": 0.07789854928851128,
"num_tokens": 920986.0,
"step": 455
},
{
"entropy": 7.980392408370972,
"epoch": 0.040199248448833345,
"grad_norm": 0.92578125,
"learning_rate": 0.00022950000000000002,
"loss": 7.6705,
"mean_token_accuracy": 0.08652443066239357,
"num_tokens": 930481.0,
"step": 460
},
{
"entropy": 8.108790636062622,
"epoch": 0.04063619680153806,
"grad_norm": 0.96484375,
"learning_rate": 0.00023200000000000003,
"loss": 7.6989,
"mean_token_accuracy": 0.08597837761044502,
"num_tokens": 940554.0,
"step": 465
},
{
"entropy": 7.986668729782105,
"epoch": 0.04107314515424277,
"grad_norm": 1.0078125,
"learning_rate": 0.00023449999999999998,
"loss": 7.6399,
"mean_token_accuracy": 0.08571613058447838,
"num_tokens": 950005.0,
"step": 470
},
{
"entropy": 7.950281715393066,
"epoch": 0.04151009350694748,
"grad_norm": 0.8125,
"learning_rate": 0.000237,
"loss": 7.6361,
"mean_token_accuracy": 0.0789151020348072,
"num_tokens": 959414.0,
"step": 475
},
{
"entropy": 8.029509019851684,
"epoch": 0.04194704185965219,
"grad_norm": 0.9296875,
"learning_rate": 0.0002395,
"loss": 7.6712,
"mean_token_accuracy": 0.07935962677001954,
"num_tokens": 969352.0,
"step": 480
},
{
"entropy": 8.051045751571655,
"epoch": 0.0423839902123569,
"grad_norm": 0.85546875,
"learning_rate": 0.000242,
"loss": 7.6912,
"mean_token_accuracy": 0.07945066690444946,
"num_tokens": 980294.0,
"step": 485
},
{
"entropy": 7.899024248123169,
"epoch": 0.04282093856506161,
"grad_norm": 0.921875,
"learning_rate": 0.0002445,
"loss": 7.5774,
"mean_token_accuracy": 0.08797174990177155,
"num_tokens": 990453.0,
"step": 490
},
{
"entropy": 8.07900128364563,
"epoch": 0.04325788691776632,
"grad_norm": 0.91796875,
"learning_rate": 0.000247,
"loss": 7.5947,
"mean_token_accuracy": 0.08882696852087975,
"num_tokens": 1000509.0,
"step": 495
},
{
"entropy": 7.993640470504761,
"epoch": 0.04369483527047103,
"grad_norm": 0.98046875,
"learning_rate": 0.0002495,
"loss": 7.7357,
"mean_token_accuracy": 0.07507672794163227,
"num_tokens": 1011747.0,
"step": 500
},
{
"entropy": 7.9472064018249515,
"epoch": 0.04413178362317574,
"grad_norm": 0.9765625,
"learning_rate": 0.000252,
"loss": 7.6713,
"mean_token_accuracy": 0.08197397291660309,
"num_tokens": 1021511.0,
"step": 505
},
{
"entropy": 8.02770299911499,
"epoch": 0.04456873197588045,
"grad_norm": 0.90625,
"learning_rate": 0.0002545,
"loss": 7.6761,
"mean_token_accuracy": 0.0799028217792511,
"num_tokens": 1032116.0,
"step": 510
},
{
"entropy": 7.903038883209229,
"epoch": 0.04500568032858516,
"grad_norm": 1.015625,
"learning_rate": 0.000257,
"loss": 7.5751,
"mean_token_accuracy": 0.09080960974097252,
"num_tokens": 1041356.0,
"step": 515
},
{
"entropy": 7.906881475448609,
"epoch": 0.04544262868128987,
"grad_norm": 1.015625,
"learning_rate": 0.0002595,
"loss": 7.6368,
"mean_token_accuracy": 0.08153252303600311,
"num_tokens": 1051314.0,
"step": 520
},
{
"entropy": 8.004989957809448,
"epoch": 0.04587957703399458,
"grad_norm": 0.859375,
"learning_rate": 0.000262,
"loss": 7.6159,
"mean_token_accuracy": 0.08653398081660271,
"num_tokens": 1060857.0,
"step": 525
},
{
"entropy": 7.971346616744995,
"epoch": 0.046316525386699295,
"grad_norm": 0.78515625,
"learning_rate": 0.00026450000000000003,
"loss": 7.624,
"mean_token_accuracy": 0.08459379523992538,
"num_tokens": 1071744.0,
"step": 530
},
{
"entropy": 7.969670820236206,
"epoch": 0.046753473739404,
"grad_norm": 1.046875,
"learning_rate": 0.00026700000000000004,
"loss": 7.5816,
"mean_token_accuracy": 0.08811679482460022,
"num_tokens": 1080777.0,
"step": 535
},
{
"entropy": 7.867853498458862,
"epoch": 0.04719042209210871,
"grad_norm": 1.09375,
"learning_rate": 0.00026950000000000005,
"loss": 7.5781,
"mean_token_accuracy": 0.0892080545425415,
"num_tokens": 1090951.0,
"step": 540
},
{
"entropy": 7.928713369369507,
"epoch": 0.047627370444813424,
"grad_norm": 1.09375,
"learning_rate": 0.00027200000000000005,
"loss": 7.5405,
"mean_token_accuracy": 0.08424938097596169,
"num_tokens": 1101667.0,
"step": 545
},
{
"entropy": 7.9725940227508545,
"epoch": 0.04806431879751814,
"grad_norm": 0.9296875,
"learning_rate": 0.0002745,
"loss": 7.6535,
"mean_token_accuracy": 0.0810074783861637,
"num_tokens": 1112652.0,
"step": 550
},
{
"entropy": 7.903535270690918,
"epoch": 0.04850126715022284,
"grad_norm": 0.9609375,
"learning_rate": 0.000277,
"loss": 7.6086,
"mean_token_accuracy": 0.08384414836764335,
"num_tokens": 1122518.0,
"step": 555
},
{
"entropy": 7.999388360977173,
"epoch": 0.048938215502927554,
"grad_norm": 0.98046875,
"learning_rate": 0.0002795,
"loss": 7.6081,
"mean_token_accuracy": 0.08308766409754753,
"num_tokens": 1132867.0,
"step": 560
},
{
"entropy": 7.961488819122314,
"epoch": 0.049375163855632266,
"grad_norm": 0.93359375,
"learning_rate": 0.00028199999999999997,
"loss": 7.6766,
"mean_token_accuracy": 0.07999438382685184,
"num_tokens": 1142524.0,
"step": 565
},
{
"entropy": 8.025870513916015,
"epoch": 0.04981211220833697,
"grad_norm": 1.0703125,
"learning_rate": 0.0002845,
"loss": 7.5489,
"mean_token_accuracy": 0.08902617059648037,
"num_tokens": 1152983.0,
"step": 570
},
{
"entropy": 7.944514799118042,
"epoch": 0.050249060561041684,
"grad_norm": 0.91015625,
"learning_rate": 0.000287,
"loss": 7.6243,
"mean_token_accuracy": 0.08356612995266914,
"num_tokens": 1164271.0,
"step": 575
},
{
"entropy": 7.8507874488830565,
"epoch": 0.050686008913746396,
"grad_norm": 1.046875,
"learning_rate": 0.0002895,
"loss": 7.4777,
"mean_token_accuracy": 0.08888518437743187,
"num_tokens": 1173567.0,
"step": 580
},
{
"entropy": 7.87242693901062,
"epoch": 0.05112295726645111,
"grad_norm": 1.0078125,
"learning_rate": 0.000292,
"loss": 7.5272,
"mean_token_accuracy": 0.08922602906823159,
"num_tokens": 1183619.0,
"step": 585
},
{
"entropy": 7.838789176940918,
"epoch": 0.05155990561915581,
"grad_norm": 1.0625,
"learning_rate": 0.0002945,
"loss": 7.4864,
"mean_token_accuracy": 0.08544198125600815,
"num_tokens": 1193922.0,
"step": 590
},
{
"entropy": 7.948909044265747,
"epoch": 0.051996853971860525,
"grad_norm": 0.95703125,
"learning_rate": 0.000297,
"loss": 7.5839,
"mean_token_accuracy": 0.08625762760639191,
"num_tokens": 1203665.0,
"step": 595
},
{
"entropy": 7.866623973846435,
"epoch": 0.05243380232456524,
"grad_norm": 0.92578125,
"learning_rate": 0.0002995,
"loss": 7.5713,
"mean_token_accuracy": 0.0907437652349472,
"num_tokens": 1214119.0,
"step": 600
},
{
"entropy": 7.910204887390137,
"epoch": 0.05287075067726995,
"grad_norm": 1.15625,
"learning_rate": 0.000302,
"loss": 7.5954,
"mean_token_accuracy": 0.08006279319524764,
"num_tokens": 1225275.0,
"step": 605
},
{
"entropy": 7.817413377761841,
"epoch": 0.053307699029974655,
"grad_norm": 0.9375,
"learning_rate": 0.0003045,
"loss": 7.4647,
"mean_token_accuracy": 0.08570721000432968,
"num_tokens": 1235199.0,
"step": 610
},
{
"entropy": 7.833220481872559,
"epoch": 0.05374464738267937,
"grad_norm": 0.91015625,
"learning_rate": 0.000307,
"loss": 7.4878,
"mean_token_accuracy": 0.08858786523342133,
"num_tokens": 1245648.0,
"step": 615
},
{
"entropy": 7.811017656326294,
"epoch": 0.05418159573538408,
"grad_norm": 0.9609375,
"learning_rate": 0.0003095,
"loss": 7.4341,
"mean_token_accuracy": 0.09172307997941971,
"num_tokens": 1254793.0,
"step": 620
},
{
"entropy": 7.904883337020874,
"epoch": 0.05461854408808879,
"grad_norm": 0.90625,
"learning_rate": 0.000312,
"loss": 7.4908,
"mean_token_accuracy": 0.09243529438972473,
"num_tokens": 1264429.0,
"step": 625
},
{
"entropy": 7.71052098274231,
"epoch": 0.055055492440793496,
"grad_norm": 1.1015625,
"learning_rate": 0.0003145,
"loss": 7.4554,
"mean_token_accuracy": 0.0870548702776432,
"num_tokens": 1274449.0,
"step": 630
},
{
"entropy": 7.8729815006256105,
"epoch": 0.05549244079349821,
"grad_norm": 0.8984375,
"learning_rate": 0.000317,
"loss": 7.5543,
"mean_token_accuracy": 0.08215527608990669,
"num_tokens": 1284987.0,
"step": 635
},
{
"entropy": 7.923571491241455,
"epoch": 0.05592938914620292,
"grad_norm": 0.98046875,
"learning_rate": 0.0003195,
"loss": 7.5696,
"mean_token_accuracy": 0.08023575767874717,
"num_tokens": 1295984.0,
"step": 640
},
{
"entropy": 7.778188037872314,
"epoch": 0.056366337498907626,
"grad_norm": 1.0,
"learning_rate": 0.000322,
"loss": 7.4517,
"mean_token_accuracy": 0.08863993138074874,
"num_tokens": 1305738.0,
"step": 645
},
{
"entropy": 7.7744215965271,
"epoch": 0.05680328585161234,
"grad_norm": 1.015625,
"learning_rate": 0.00032450000000000003,
"loss": 7.4031,
"mean_token_accuracy": 0.08852944150567055,
"num_tokens": 1315691.0,
"step": 650
},
{
"entropy": 7.780942630767822,
"epoch": 0.05724023420431705,
"grad_norm": 0.96484375,
"learning_rate": 0.00032700000000000003,
"loss": 7.4193,
"mean_token_accuracy": 0.0951805904507637,
"num_tokens": 1326195.0,
"step": 655
},
{
"entropy": 7.719406795501709,
"epoch": 0.05767718255702176,
"grad_norm": 1.0625,
"learning_rate": 0.00032950000000000004,
"loss": 7.4457,
"mean_token_accuracy": 0.08878713175654411,
"num_tokens": 1336916.0,
"step": 660
},
{
"entropy": 7.77421612739563,
"epoch": 0.05811413090972647,
"grad_norm": 1.0625,
"learning_rate": 0.00033200000000000005,
"loss": 7.4653,
"mean_token_accuracy": 0.08794940188527108,
"num_tokens": 1347386.0,
"step": 665
},
{
"entropy": 7.852012014389038,
"epoch": 0.05855107926243118,
"grad_norm": 1.171875,
"learning_rate": 0.00033450000000000005,
"loss": 7.4342,
"mean_token_accuracy": 0.08515654727816582,
"num_tokens": 1357959.0,
"step": 670
},
{
"entropy": 7.772040128707886,
"epoch": 0.05898802761513589,
"grad_norm": 1.1015625,
"learning_rate": 0.000337,
"loss": 7.4385,
"mean_token_accuracy": 0.08945433050394058,
"num_tokens": 1368135.0,
"step": 675
},
{
"entropy": 7.773286199569702,
"epoch": 0.059424975967840604,
"grad_norm": 0.953125,
"learning_rate": 0.0003395,
"loss": 7.4691,
"mean_token_accuracy": 0.08570723086595536,
"num_tokens": 1378702.0,
"step": 680
},
{
"entropy": 7.67702841758728,
"epoch": 0.05986192432054531,
"grad_norm": 1.0625,
"learning_rate": 0.000342,
"loss": 7.3237,
"mean_token_accuracy": 0.09294968172907829,
"num_tokens": 1387902.0,
"step": 685
},
{
"entropy": 7.789173412322998,
"epoch": 0.06029887267325002,
"grad_norm": 0.98828125,
"learning_rate": 0.00034449999999999997,
"loss": 7.3411,
"mean_token_accuracy": 0.09882994741201401,
"num_tokens": 1397288.0,
"step": 690
},
{
"entropy": 7.668393468856811,
"epoch": 0.060735821025954734,
"grad_norm": 0.9609375,
"learning_rate": 0.000347,
"loss": 7.3877,
"mean_token_accuracy": 0.09437731429934501,
"num_tokens": 1407416.0,
"step": 695
},
{
"entropy": 7.834095811843872,
"epoch": 0.061172769378659446,
"grad_norm": 0.91015625,
"learning_rate": 0.0003495,
"loss": 7.443,
"mean_token_accuracy": 0.08851219043135643,
"num_tokens": 1417733.0,
"step": 700
},
{
"entropy": 7.720336055755615,
"epoch": 0.06160971773136415,
"grad_norm": 1.2578125,
"learning_rate": 0.000352,
"loss": 7.3365,
"mean_token_accuracy": 0.09252023622393608,
"num_tokens": 1426723.0,
"step": 705
},
{
"entropy": 7.701070976257324,
"epoch": 0.06204666608406886,
"grad_norm": 1.109375,
"learning_rate": 0.0003545,
"loss": 7.3991,
"mean_token_accuracy": 0.09218236356973648,
"num_tokens": 1437290.0,
"step": 710
},
{
"entropy": 7.631638383865356,
"epoch": 0.062483614436773575,
"grad_norm": 0.9921875,
"learning_rate": 0.000357,
"loss": 7.3274,
"mean_token_accuracy": 0.09394551664590836,
"num_tokens": 1446756.0,
"step": 715
},
{
"entropy": 7.714466619491577,
"epoch": 0.06292056278947829,
"grad_norm": 1.0390625,
"learning_rate": 0.0003595,
"loss": 7.4124,
"mean_token_accuracy": 0.08511779010295868,
"num_tokens": 1457074.0,
"step": 720
},
{
"entropy": 7.746994829177856,
"epoch": 0.06335751114218299,
"grad_norm": 1.1171875,
"learning_rate": 0.000362,
"loss": 7.4538,
"mean_token_accuracy": 0.08377371467649937,
"num_tokens": 1467785.0,
"step": 725
},
{
"entropy": 7.678491926193237,
"epoch": 0.0637944594948877,
"grad_norm": 0.88671875,
"learning_rate": 0.0003645,
"loss": 7.3443,
"mean_token_accuracy": 0.09074902832508087,
"num_tokens": 1477532.0,
"step": 730
},
{
"entropy": 7.6045910835266115,
"epoch": 0.06423140784759242,
"grad_norm": 1.0,
"learning_rate": 0.000367,
"loss": 7.2766,
"mean_token_accuracy": 0.09526442065834999,
"num_tokens": 1487102.0,
"step": 735
},
{
"entropy": 7.650178909301758,
"epoch": 0.06466835620029712,
"grad_norm": 0.99609375,
"learning_rate": 0.0003695,
"loss": 7.3517,
"mean_token_accuracy": 0.08858921527862548,
"num_tokens": 1497459.0,
"step": 740
},
{
"entropy": 7.654647779464722,
"epoch": 0.06510530455300184,
"grad_norm": 0.86328125,
"learning_rate": 0.000372,
"loss": 7.3427,
"mean_token_accuracy": 0.08997647315263749,
"num_tokens": 1508756.0,
"step": 745
},
{
"entropy": 7.6317675590515135,
"epoch": 0.06554225290570655,
"grad_norm": 0.94921875,
"learning_rate": 0.0003745,
"loss": 7.2637,
"mean_token_accuracy": 0.09751128628849984,
"num_tokens": 1518735.0,
"step": 750
},
{
"entropy": 7.698947954177856,
"epoch": 0.06597920125841125,
"grad_norm": 1.0078125,
"learning_rate": 0.000377,
"loss": 7.2964,
"mean_token_accuracy": 0.09505418241024018,
"num_tokens": 1528632.0,
"step": 755
},
{
"entropy": 7.500899124145508,
"epoch": 0.06641614961111597,
"grad_norm": 0.91796875,
"learning_rate": 0.0003795,
"loss": 7.295,
"mean_token_accuracy": 0.09173346012830734,
"num_tokens": 1538180.0,
"step": 760
},
{
"entropy": 7.681208324432373,
"epoch": 0.06685309796382068,
"grad_norm": 1.0234375,
"learning_rate": 0.000382,
"loss": 7.3137,
"mean_token_accuracy": 0.09170436710119248,
"num_tokens": 1547788.0,
"step": 765
},
{
"entropy": 7.601843214035034,
"epoch": 0.06729004631652538,
"grad_norm": 1.046875,
"learning_rate": 0.0003845,
"loss": 7.2644,
"mean_token_accuracy": 0.09326299503445626,
"num_tokens": 1558008.0,
"step": 770
},
{
"entropy": 7.5049737930297855,
"epoch": 0.0677269946692301,
"grad_norm": 1.0859375,
"learning_rate": 0.00038700000000000003,
"loss": 7.3246,
"mean_token_accuracy": 0.08936878740787506,
"num_tokens": 1567728.0,
"step": 775
},
{
"entropy": 7.653160285949707,
"epoch": 0.0681639430219348,
"grad_norm": 0.9609375,
"learning_rate": 0.00038950000000000003,
"loss": 7.2736,
"mean_token_accuracy": 0.09446480795741082,
"num_tokens": 1578488.0,
"step": 780
},
{
"entropy": 7.6056342124938965,
"epoch": 0.06860089137463951,
"grad_norm": 0.9453125,
"learning_rate": 0.00039200000000000004,
"loss": 7.2499,
"mean_token_accuracy": 0.09750332310795784,
"num_tokens": 1588070.0,
"step": 785
},
{
"entropy": 7.5712121486663815,
"epoch": 0.06903783972734423,
"grad_norm": 0.89453125,
"learning_rate": 0.00039450000000000005,
"loss": 7.3194,
"mean_token_accuracy": 0.09099319577217102,
"num_tokens": 1598572.0,
"step": 790
},
{
"entropy": 7.714870977401733,
"epoch": 0.06947478808004894,
"grad_norm": 1.109375,
"learning_rate": 0.00039700000000000005,
"loss": 7.3286,
"mean_token_accuracy": 0.09093371108174324,
"num_tokens": 1607982.0,
"step": 795
},
{
"entropy": 7.541324853897095,
"epoch": 0.06991173643275365,
"grad_norm": 1.078125,
"learning_rate": 0.0003995,
"loss": 7.3322,
"mean_token_accuracy": 0.0973043292760849,
"num_tokens": 1618785.0,
"step": 800
},
{
"entropy": 7.621754837036133,
"epoch": 0.07034868478545836,
"grad_norm": 1.1015625,
"learning_rate": 0.000402,
"loss": 7.2548,
"mean_token_accuracy": 0.09845281392335892,
"num_tokens": 1627988.0,
"step": 805
},
{
"entropy": 7.560144567489624,
"epoch": 0.07078563313816306,
"grad_norm": 1.1640625,
"learning_rate": 0.0004045,
"loss": 7.3355,
"mean_token_accuracy": 0.08645970746874809,
"num_tokens": 1638024.0,
"step": 810
},
{
"entropy": 7.686397504806519,
"epoch": 0.07122258149086778,
"grad_norm": 1.078125,
"learning_rate": 0.00040699999999999997,
"loss": 7.3101,
"mean_token_accuracy": 0.0890956200659275,
"num_tokens": 1649215.0,
"step": 815
},
{
"entropy": 7.442864656448364,
"epoch": 0.07165952984357249,
"grad_norm": 0.89453125,
"learning_rate": 0.0004095,
"loss": 7.1978,
"mean_token_accuracy": 0.09686766788363457,
"num_tokens": 1659198.0,
"step": 820
},
{
"entropy": 7.594072866439819,
"epoch": 0.0720964781962772,
"grad_norm": 1.0625,
"learning_rate": 0.000412,
"loss": 7.2452,
"mean_token_accuracy": 0.0892924964427948,
"num_tokens": 1668972.0,
"step": 825
},
{
"entropy": 7.439077568054199,
"epoch": 0.07253342654898191,
"grad_norm": 1.15625,
"learning_rate": 0.0004145,
"loss": 7.2232,
"mean_token_accuracy": 0.09631062857806683,
"num_tokens": 1679807.0,
"step": 830
},
{
"entropy": 7.600879383087158,
"epoch": 0.07297037490168662,
"grad_norm": 1.0078125,
"learning_rate": 0.000417,
"loss": 7.2853,
"mean_token_accuracy": 0.09247459471225739,
"num_tokens": 1690567.0,
"step": 835
},
{
"entropy": 7.469448566436768,
"epoch": 0.07340732325439134,
"grad_norm": 1.0859375,
"learning_rate": 0.0004195,
"loss": 7.1689,
"mean_token_accuracy": 0.09535667225718499,
"num_tokens": 1700939.0,
"step": 840
},
{
"entropy": 7.388382434844971,
"epoch": 0.07384427160709604,
"grad_norm": 0.9609375,
"learning_rate": 0.000422,
"loss": 7.2393,
"mean_token_accuracy": 0.099522565305233,
"num_tokens": 1711181.0,
"step": 845
},
{
"entropy": 7.600938224792481,
"epoch": 0.07428121995980075,
"grad_norm": 1.1015625,
"learning_rate": 0.0004245,
"loss": 7.2539,
"mean_token_accuracy": 0.09072699621319771,
"num_tokens": 1721392.0,
"step": 850
},
{
"entropy": 7.495168161392212,
"epoch": 0.07471816831250547,
"grad_norm": 0.99609375,
"learning_rate": 0.000427,
"loss": 7.1767,
"mean_token_accuracy": 0.09786228016018868,
"num_tokens": 1731942.0,
"step": 855
},
{
"entropy": 7.469488048553467,
"epoch": 0.07515511666521017,
"grad_norm": 0.9140625,
"learning_rate": 0.0004295,
"loss": 7.2758,
"mean_token_accuracy": 0.09157036766409873,
"num_tokens": 1743409.0,
"step": 860
},
{
"entropy": 7.502536010742188,
"epoch": 0.07559206501791488,
"grad_norm": 0.85546875,
"learning_rate": 0.000432,
"loss": 7.1633,
"mean_token_accuracy": 0.09798691645264626,
"num_tokens": 1753425.0,
"step": 865
},
{
"entropy": 7.4638996601104735,
"epoch": 0.0760290133706196,
"grad_norm": 0.984375,
"learning_rate": 0.0004345,
"loss": 7.1934,
"mean_token_accuracy": 0.09164758920669555,
"num_tokens": 1762722.0,
"step": 870
},
{
"entropy": 7.642696189880371,
"epoch": 0.0764659617233243,
"grad_norm": 1.0390625,
"learning_rate": 0.000437,
"loss": 7.2749,
"mean_token_accuracy": 0.0860434666275978,
"num_tokens": 1771895.0,
"step": 875
},
{
"entropy": 7.42297739982605,
"epoch": 0.07690291007602901,
"grad_norm": 0.9765625,
"learning_rate": 0.0004395,
"loss": 7.1698,
"mean_token_accuracy": 0.09690217077732086,
"num_tokens": 1782175.0,
"step": 880
},
{
"entropy": 7.478855276107788,
"epoch": 0.07733985842873373,
"grad_norm": 1.015625,
"learning_rate": 0.000442,
"loss": 7.2251,
"mean_token_accuracy": 0.09283168092370034,
"num_tokens": 1792409.0,
"step": 885
},
{
"entropy": 7.5595293045043945,
"epoch": 0.07777680678143843,
"grad_norm": 1.1171875,
"learning_rate": 0.0004445,
"loss": 7.2736,
"mean_token_accuracy": 0.08964755833148956,
"num_tokens": 1802721.0,
"step": 890
},
{
"entropy": 7.382868146896362,
"epoch": 0.07821375513414315,
"grad_norm": 1.03125,
"learning_rate": 0.000447,
"loss": 7.1126,
"mean_token_accuracy": 0.09378557428717613,
"num_tokens": 1812064.0,
"step": 895
},
{
"entropy": 7.530679559707641,
"epoch": 0.07865070348684786,
"grad_norm": 1.0390625,
"learning_rate": 0.00044950000000000003,
"loss": 7.215,
"mean_token_accuracy": 0.08957840651273727,
"num_tokens": 1822284.0,
"step": 900
},
{
"entropy": 7.388239574432373,
"epoch": 0.07908765183955256,
"grad_norm": 0.9375,
"learning_rate": 0.00045200000000000004,
"loss": 7.1062,
"mean_token_accuracy": 0.10064681470394135,
"num_tokens": 1832150.0,
"step": 905
},
{
"entropy": 7.545283079147339,
"epoch": 0.07952460019225728,
"grad_norm": 1.0859375,
"learning_rate": 0.00045450000000000004,
"loss": 7.2334,
"mean_token_accuracy": 0.0913564309477806,
"num_tokens": 1842870.0,
"step": 910
},
{
"entropy": 7.420826816558838,
"epoch": 0.07996154854496199,
"grad_norm": 1.09375,
"learning_rate": 0.00045700000000000005,
"loss": 7.0721,
"mean_token_accuracy": 0.10421185195446014,
"num_tokens": 1852771.0,
"step": 915
},
{
"entropy": 7.382372570037842,
"epoch": 0.08039849689766669,
"grad_norm": 1.03125,
"learning_rate": 0.00045950000000000006,
"loss": 7.1698,
"mean_token_accuracy": 0.09922162219882011,
"num_tokens": 1863095.0,
"step": 920
},
{
"entropy": 7.418209648132324,
"epoch": 0.08083544525037141,
"grad_norm": 1.09375,
"learning_rate": 0.000462,
"loss": 7.171,
"mean_token_accuracy": 0.09901941046118737,
"num_tokens": 1873060.0,
"step": 925
},
{
"entropy": 7.4146593570709225,
"epoch": 0.08127239360307612,
"grad_norm": 0.9609375,
"learning_rate": 0.0004645,
"loss": 7.1089,
"mean_token_accuracy": 0.10169215872883797,
"num_tokens": 1882574.0,
"step": 930
},
{
"entropy": 7.437223148345947,
"epoch": 0.08170934195578082,
"grad_norm": 0.94921875,
"learning_rate": 0.000467,
"loss": 7.1725,
"mean_token_accuracy": 0.09495756924152374,
"num_tokens": 1892227.0,
"step": 935
},
{
"entropy": 7.439269733428955,
"epoch": 0.08214629030848554,
"grad_norm": 1.046875,
"learning_rate": 0.0004695,
"loss": 7.1934,
"mean_token_accuracy": 0.09354586526751518,
"num_tokens": 1902868.0,
"step": 940
},
{
"entropy": 7.387948656082154,
"epoch": 0.08258323866119024,
"grad_norm": 0.94921875,
"learning_rate": 0.000472,
"loss": 7.1448,
"mean_token_accuracy": 0.09095190614461898,
"num_tokens": 1912781.0,
"step": 945
},
{
"entropy": 7.391385173797607,
"epoch": 0.08302018701389496,
"grad_norm": 0.91796875,
"learning_rate": 0.0004745,
"loss": 7.1389,
"mean_token_accuracy": 0.0988696813583374,
"num_tokens": 1923638.0,
"step": 950
},
{
"entropy": 7.388245868682861,
"epoch": 0.08345713536659967,
"grad_norm": 1.0546875,
"learning_rate": 0.000477,
"loss": 7.1097,
"mean_token_accuracy": 0.10090045481920243,
"num_tokens": 1933498.0,
"step": 955
},
{
"entropy": 7.285632038116455,
"epoch": 0.08389408371930437,
"grad_norm": 0.98828125,
"learning_rate": 0.0004795,
"loss": 7.1227,
"mean_token_accuracy": 0.09693519324064255,
"num_tokens": 1943798.0,
"step": 960
},
{
"entropy": 7.4096606254577635,
"epoch": 0.0843310320720091,
"grad_norm": 1.1015625,
"learning_rate": 0.000482,
"loss": 7.1564,
"mean_token_accuracy": 0.09475521519780158,
"num_tokens": 1954235.0,
"step": 965
},
{
"entropy": 7.368670797348022,
"epoch": 0.0847679804247138,
"grad_norm": 0.95703125,
"learning_rate": 0.0004845,
"loss": 7.152,
"mean_token_accuracy": 0.09138476997613906,
"num_tokens": 1964421.0,
"step": 970
},
{
"entropy": 7.36945948600769,
"epoch": 0.0852049287774185,
"grad_norm": 0.86328125,
"learning_rate": 0.000487,
"loss": 7.1236,
"mean_token_accuracy": 0.09725102558732032,
"num_tokens": 1975862.0,
"step": 975
},
{
"entropy": 7.358737993240356,
"epoch": 0.08564187713012322,
"grad_norm": 0.98828125,
"learning_rate": 0.0004895,
"loss": 7.1205,
"mean_token_accuracy": 0.09733417257666588,
"num_tokens": 1985758.0,
"step": 980
},
{
"entropy": 7.362492322921753,
"epoch": 0.08607882548282793,
"grad_norm": 0.98046875,
"learning_rate": 0.000492,
"loss": 7.0829,
"mean_token_accuracy": 0.09430268853902816,
"num_tokens": 1996904.0,
"step": 985
},
{
"entropy": 7.448752498626709,
"epoch": 0.08651577383553263,
"grad_norm": 0.94140625,
"learning_rate": 0.0004945,
"loss": 7.1749,
"mean_token_accuracy": 0.09713461846113206,
"num_tokens": 2006114.0,
"step": 990
},
{
"entropy": 7.365045785903931,
"epoch": 0.08695272218823735,
"grad_norm": 0.9375,
"learning_rate": 0.000497,
"loss": 7.1111,
"mean_token_accuracy": 0.09818708300590515,
"num_tokens": 2016349.0,
"step": 995
},
{
"entropy": 7.291970348358154,
"epoch": 0.08738967054094206,
"grad_norm": 1.015625,
"learning_rate": 0.0004995,
"loss": 7.061,
"mean_token_accuracy": 0.09953764006495476,
"num_tokens": 2026701.0,
"step": 1000
},
{
"entropy": 7.302675056457519,
"epoch": 0.08782661889364678,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999999986190023,
"loss": 7.0537,
"mean_token_accuracy": 0.09898199439048767,
"num_tokens": 2036950.0,
"step": 1005
},
{
"entropy": 7.409235525131225,
"epoch": 0.08826356724635148,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999999930086991,
"loss": 7.0489,
"mean_token_accuracy": 0.10240035951137542,
"num_tokens": 2047655.0,
"step": 1010
},
{
"entropy": 7.260290336608887,
"epoch": 0.08870051559905619,
"grad_norm": 1.0390625,
"learning_rate": 0.000499999983082778,
"loss": 7.0274,
"mean_token_accuracy": 0.10356488525867462,
"num_tokens": 2056800.0,
"step": 1015
},
{
"entropy": 7.276587057113647,
"epoch": 0.0891374639517609,
"grad_norm": 1.0625,
"learning_rate": 0.0004999999688412394,
"loss": 7.07,
"mean_token_accuracy": 0.09646826907992363,
"num_tokens": 2066862.0,
"step": 1020
},
{
"entropy": 7.327617454528808,
"epoch": 0.08957441230446561,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999502840836,
"loss": 7.0896,
"mean_token_accuracy": 0.10041970610618592,
"num_tokens": 2076585.0,
"step": 1025
},
{
"entropy": 7.3417980670928955,
"epoch": 0.09001136065717032,
"grad_norm": 0.921875,
"learning_rate": 0.0004999999274113107,
"loss": 7.1548,
"mean_token_accuracy": 0.09449673742055893,
"num_tokens": 2087768.0,
"step": 1030
},
{
"entropy": 7.343121814727783,
"epoch": 0.09044830900987504,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999999002229214,
"loss": 7.0622,
"mean_token_accuracy": 0.09806089103221893,
"num_tokens": 2097027.0,
"step": 1035
},
{
"entropy": 7.191218137741089,
"epoch": 0.09088525736257974,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999998687189161,
"loss": 6.9901,
"mean_token_accuracy": 0.10912425816059113,
"num_tokens": 2107055.0,
"step": 1040
},
{
"entropy": 7.271266222000122,
"epoch": 0.09132220571528446,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999998328992954,
"loss": 6.9198,
"mean_token_accuracy": 0.10159473046660424,
"num_tokens": 2116771.0,
"step": 1045
},
{
"entropy": 7.177909517288208,
"epoch": 0.09175915406798917,
"grad_norm": 1.0546875,
"learning_rate": 0.00049999979276406,
"loss": 7.0603,
"mean_token_accuracy": 0.10223758444190026,
"num_tokens": 2126881.0,
"step": 1050
},
{
"entropy": 7.254304838180542,
"epoch": 0.09219610242069387,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999997483132108,
"loss": 7.0248,
"mean_token_accuracy": 0.101267521828413,
"num_tokens": 2136964.0,
"step": 1055
},
{
"entropy": 7.332895040512085,
"epoch": 0.09263305077339859,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999996995467483,
"loss": 7.0764,
"mean_token_accuracy": 0.09840327501296997,
"num_tokens": 2147263.0,
"step": 1060
},
{
"entropy": 7.248973989486695,
"epoch": 0.0930699991261033,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999996464646738,
"loss": 7.089,
"mean_token_accuracy": 0.09868241250514984,
"num_tokens": 2157474.0,
"step": 1065
},
{
"entropy": 7.242979240417481,
"epoch": 0.093506947478808,
"grad_norm": 0.96875,
"learning_rate": 0.0004999995890669882,
"loss": 7.0217,
"mean_token_accuracy": 0.10121526122093201,
"num_tokens": 2167701.0,
"step": 1070
},
{
"entropy": 7.1951741695404055,
"epoch": 0.09394389583151272,
"grad_norm": 1.0,
"learning_rate": 0.0004999995273536925,
"loss": 6.9862,
"mean_token_accuracy": 0.10409378558397293,
"num_tokens": 2178581.0,
"step": 1075
},
{
"entropy": 7.250149822235107,
"epoch": 0.09438084418421742,
"grad_norm": 0.93359375,
"learning_rate": 0.000499999461324788,
"loss": 6.9949,
"mean_token_accuracy": 0.10199672728776932,
"num_tokens": 2189529.0,
"step": 1080
},
{
"entropy": 7.254613780975342,
"epoch": 0.09481779253692213,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999993909802759,
"loss": 7.0541,
"mean_token_accuracy": 0.09610985442996026,
"num_tokens": 2199805.0,
"step": 1085
},
{
"entropy": 7.232923269271851,
"epoch": 0.09525474088962685,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999993163201577,
"loss": 7.0206,
"mean_token_accuracy": 0.09527453482151031,
"num_tokens": 2210108.0,
"step": 1090
},
{
"entropy": 7.358099699020386,
"epoch": 0.09569168924233155,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999992373444347,
"loss": 7.01,
"mean_token_accuracy": 0.10141385495662689,
"num_tokens": 2220325.0,
"step": 1095
},
{
"entropy": 7.133218908309937,
"epoch": 0.09612863759503627,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999991540531084,
"loss": 7.0098,
"mean_token_accuracy": 0.0968212477862835,
"num_tokens": 2230521.0,
"step": 1100
},
{
"entropy": 7.322919178009033,
"epoch": 0.09656558594774098,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999990664461804,
"loss": 7.0377,
"mean_token_accuracy": 0.09720605388283729,
"num_tokens": 2240372.0,
"step": 1105
},
{
"entropy": 7.171909475326538,
"epoch": 0.09700253430044568,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999989745236524,
"loss": 6.992,
"mean_token_accuracy": 0.10702305659651756,
"num_tokens": 2250522.0,
"step": 1110
},
{
"entropy": 7.190585279464722,
"epoch": 0.0974394826531504,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999988782855261,
"loss": 6.9584,
"mean_token_accuracy": 0.09914659708738327,
"num_tokens": 2260142.0,
"step": 1115
},
{
"entropy": 7.227755498886109,
"epoch": 0.09787643100585511,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999987777318035,
"loss": 6.9664,
"mean_token_accuracy": 0.10190813839435578,
"num_tokens": 2270372.0,
"step": 1120
},
{
"entropy": 7.071720218658447,
"epoch": 0.09831337935855981,
"grad_norm": 1.03125,
"learning_rate": 0.0004999986728624865,
"loss": 6.9076,
"mean_token_accuracy": 0.11096167787909508,
"num_tokens": 2280517.0,
"step": 1125
},
{
"entropy": 7.284544038772583,
"epoch": 0.09875032771126453,
"grad_norm": 1.0234375,
"learning_rate": 0.000499998563677577,
"loss": 6.9999,
"mean_token_accuracy": 0.09647511318325996,
"num_tokens": 2290778.0,
"step": 1130
},
{
"entropy": 7.183425855636597,
"epoch": 0.09918727606396924,
"grad_norm": 1.015625,
"learning_rate": 0.0004999984501770771,
"loss": 7.0455,
"mean_token_accuracy": 0.09870154932141303,
"num_tokens": 2301692.0,
"step": 1135
},
{
"entropy": 7.21610689163208,
"epoch": 0.09962422441667394,
"grad_norm": 1.15625,
"learning_rate": 0.0004999983323609891,
"loss": 6.9337,
"mean_token_accuracy": 0.10188546925783157,
"num_tokens": 2312759.0,
"step": 1140
},
{
"entropy": 7.203960323333741,
"epoch": 0.10006117276937866,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999982102293153,
"loss": 6.9185,
"mean_token_accuracy": 0.10292895063757897,
"num_tokens": 2322016.0,
"step": 1145
},
{
"entropy": 7.224305295944214,
"epoch": 0.10049812112208337,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999980837820578,
"loss": 6.9399,
"mean_token_accuracy": 0.10199728310108185,
"num_tokens": 2332476.0,
"step": 1150
},
{
"entropy": 7.190159654617309,
"epoch": 0.10093506947478809,
"grad_norm": 1.015625,
"learning_rate": 0.0004999979530192192,
"loss": 6.9518,
"mean_token_accuracy": 0.10138825252652169,
"num_tokens": 2342914.0,
"step": 1155
},
{
"entropy": 7.146100616455078,
"epoch": 0.10137201782749279,
"grad_norm": 0.96875,
"learning_rate": 0.0004999978179408019,
"loss": 6.9927,
"mean_token_accuracy": 0.0981225997209549,
"num_tokens": 2353381.0,
"step": 1160
},
{
"entropy": 7.201415205001831,
"epoch": 0.1018089661801975,
"grad_norm": 0.84375,
"learning_rate": 0.0004999976785468085,
"loss": 6.982,
"mean_token_accuracy": 0.10512297451496125,
"num_tokens": 2364207.0,
"step": 1165
},
{
"entropy": 7.082000207901001,
"epoch": 0.10224591453290222,
"grad_norm": 1.046875,
"learning_rate": 0.0004999975348372418,
"loss": 6.8996,
"mean_token_accuracy": 0.1049137569963932,
"num_tokens": 2374985.0,
"step": 1170
},
{
"entropy": 7.163199710845947,
"epoch": 0.10268286288560692,
"grad_norm": 0.96875,
"learning_rate": 0.0004999973868121044,
"loss": 6.9749,
"mean_token_accuracy": 0.1049682468175888,
"num_tokens": 2385684.0,
"step": 1175
},
{
"entropy": 7.2960728168487545,
"epoch": 0.10311981123831163,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999972344713994,
"loss": 6.8993,
"mean_token_accuracy": 0.09997786954045296,
"num_tokens": 2395177.0,
"step": 1180
},
{
"entropy": 7.023489665985108,
"epoch": 0.10355675959101635,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999970778151295,
"loss": 6.9248,
"mean_token_accuracy": 0.109105683863163,
"num_tokens": 2405154.0,
"step": 1185
},
{
"entropy": 7.2286418914794925,
"epoch": 0.10399370794372105,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999969168432976,
"loss": 6.9182,
"mean_token_accuracy": 0.10396889075636864,
"num_tokens": 2415818.0,
"step": 1190
},
{
"entropy": 7.086079740524292,
"epoch": 0.10443065629642576,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999967515559071,
"loss": 6.8955,
"mean_token_accuracy": 0.1030263639986515,
"num_tokens": 2425072.0,
"step": 1195
},
{
"entropy": 7.1413719177246096,
"epoch": 0.10486760464913047,
"grad_norm": 1.0703125,
"learning_rate": 0.000499996581952961,
"loss": 6.9562,
"mean_token_accuracy": 0.10766926109790802,
"num_tokens": 2435055.0,
"step": 1200
},
{
"entropy": 7.168883657455444,
"epoch": 0.10530455300183518,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999964080344625,
"loss": 6.8422,
"mean_token_accuracy": 0.11093364953994751,
"num_tokens": 2444434.0,
"step": 1205
},
{
"entropy": 7.101231050491333,
"epoch": 0.1057415013545399,
"grad_norm": 1.015625,
"learning_rate": 0.0004999962298004151,
"loss": 6.8734,
"mean_token_accuracy": 0.10424663722515107,
"num_tokens": 2453860.0,
"step": 1210
},
{
"entropy": 7.190755844116211,
"epoch": 0.1061784497072446,
"grad_norm": 1.09375,
"learning_rate": 0.0004999960472508219,
"loss": 6.9733,
"mean_token_accuracy": 0.09948998987674713,
"num_tokens": 2464235.0,
"step": 1215
},
{
"entropy": 7.079598331451416,
"epoch": 0.10661539805994931,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999958603856869,
"loss": 6.8763,
"mean_token_accuracy": 0.10838079452514648,
"num_tokens": 2474561.0,
"step": 1220
},
{
"entropy": 7.070672941207886,
"epoch": 0.10705234641265403,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999956692050132,
"loss": 6.8808,
"mean_token_accuracy": 0.10815191268920898,
"num_tokens": 2484574.0,
"step": 1225
},
{
"entropy": 7.15245943069458,
"epoch": 0.10748929476535873,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999954737088048,
"loss": 6.8478,
"mean_token_accuracy": 0.11184538155794144,
"num_tokens": 2494146.0,
"step": 1230
},
{
"entropy": 6.9708349227905275,
"epoch": 0.10792624311806344,
"grad_norm": 0.984375,
"learning_rate": 0.0004999952738970652,
"loss": 6.9382,
"mean_token_accuracy": 0.10538864210247993,
"num_tokens": 2504157.0,
"step": 1235
},
{
"entropy": 7.157891273498535,
"epoch": 0.10836319147076816,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999950697697983,
"loss": 6.8751,
"mean_token_accuracy": 0.10475126504898072,
"num_tokens": 2512979.0,
"step": 1240
},
{
"entropy": 7.113284254074097,
"epoch": 0.10880013982347286,
"grad_norm": 1.046875,
"learning_rate": 0.0004999948613270081,
"loss": 6.9312,
"mean_token_accuracy": 0.10144072771072388,
"num_tokens": 2523202.0,
"step": 1245
},
{
"entropy": 7.073151969909668,
"epoch": 0.10923708817617758,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999946485686987,
"loss": 6.8758,
"mean_token_accuracy": 0.10646733790636062,
"num_tokens": 2532790.0,
"step": 1250
},
{
"entropy": 7.101035165786743,
"epoch": 0.10967403652888229,
"grad_norm": 1.0859375,
"learning_rate": 0.000499994431494874,
"loss": 6.8643,
"mean_token_accuracy": 0.10860786586999893,
"num_tokens": 2542337.0,
"step": 1255
},
{
"entropy": 7.028286504745483,
"epoch": 0.11011098488158699,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999942101055381,
"loss": 6.8175,
"mean_token_accuracy": 0.10841023772954941,
"num_tokens": 2552762.0,
"step": 1260
},
{
"entropy": 7.066199254989624,
"epoch": 0.11054793323429171,
"grad_norm": 1.046875,
"learning_rate": 0.0004999939844006955,
"loss": 6.858,
"mean_token_accuracy": 0.10801806300878525,
"num_tokens": 2563261.0,
"step": 1265
},
{
"entropy": 7.134337902069092,
"epoch": 0.11098488158699642,
"grad_norm": 1.03125,
"learning_rate": 0.0004999937543803503,
"loss": 6.8252,
"mean_token_accuracy": 0.1146186038851738,
"num_tokens": 2573089.0,
"step": 1270
},
{
"entropy": 7.0005029201507565,
"epoch": 0.11142182993970112,
"grad_norm": 0.96875,
"learning_rate": 0.000499993520044507,
"loss": 6.9056,
"mean_token_accuracy": 0.10176412537693977,
"num_tokens": 2584358.0,
"step": 1275
},
{
"entropy": 7.139933633804321,
"epoch": 0.11185877829240584,
"grad_norm": 1.03125,
"learning_rate": 0.0004999932813931702,
"loss": 6.8709,
"mean_token_accuracy": 0.1116573728621006,
"num_tokens": 2594981.0,
"step": 1280
},
{
"entropy": 7.084459447860718,
"epoch": 0.11229572664511055,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999930384263441,
"loss": 6.8717,
"mean_token_accuracy": 0.1069402813911438,
"num_tokens": 2603508.0,
"step": 1285
},
{
"entropy": 7.084566974639893,
"epoch": 0.11273267499781525,
"grad_norm": 1.0390625,
"learning_rate": 0.000499992791144034,
"loss": 6.8535,
"mean_token_accuracy": 0.10854770466685296,
"num_tokens": 2613666.0,
"step": 1290
},
{
"entropy": 7.203876972198486,
"epoch": 0.11316962335051997,
"grad_norm": 1.140625,
"learning_rate": 0.0004999925395462439,
"loss": 6.9042,
"mean_token_accuracy": 0.10472902953624726,
"num_tokens": 2624441.0,
"step": 1295
},
{
"entropy": 7.066425085067749,
"epoch": 0.11360657170322468,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999922836329793,
"loss": 6.8576,
"mean_token_accuracy": 0.10966517627239228,
"num_tokens": 2635261.0,
"step": 1300
},
{
"entropy": 6.97562608718872,
"epoch": 0.1140435200559294,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999920234042446,
"loss": 6.8321,
"mean_token_accuracy": 0.11323759481310844,
"num_tokens": 2645493.0,
"step": 1305
},
{
"entropy": 7.0524732112884525,
"epoch": 0.1144804684086341,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999917588600451,
"loss": 6.7956,
"mean_token_accuracy": 0.11606079787015915,
"num_tokens": 2655430.0,
"step": 1310
},
{
"entropy": 7.048898983001709,
"epoch": 0.1149174167613388,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999914900003857,
"loss": 6.8529,
"mean_token_accuracy": 0.10828904882073402,
"num_tokens": 2665675.0,
"step": 1315
},
{
"entropy": 7.087590456008911,
"epoch": 0.11535436511404352,
"grad_norm": 1.015625,
"learning_rate": 0.0004999912168252717,
"loss": 6.8204,
"mean_token_accuracy": 0.10795803815126419,
"num_tokens": 2675962.0,
"step": 1320
},
{
"entropy": 7.021550798416138,
"epoch": 0.11579131346674823,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999909393347083,
"loss": 6.8477,
"mean_token_accuracy": 0.10889611020684242,
"num_tokens": 2687015.0,
"step": 1325
},
{
"entropy": 7.038705635070801,
"epoch": 0.11622826181945294,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999906575287006,
"loss": 6.8956,
"mean_token_accuracy": 0.10390607044100761,
"num_tokens": 2698065.0,
"step": 1330
},
{
"entropy": 7.041507291793823,
"epoch": 0.11666521017215765,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999903714072544,
"loss": 6.7831,
"mean_token_accuracy": 0.11800387650728225,
"num_tokens": 2709046.0,
"step": 1335
},
{
"entropy": 7.065563631057739,
"epoch": 0.11710215852486236,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999900809703749,
"loss": 6.8565,
"mean_token_accuracy": 0.11037417724728585,
"num_tokens": 2719705.0,
"step": 1340
},
{
"entropy": 7.038755893707275,
"epoch": 0.11753910687756706,
"grad_norm": 1.484375,
"learning_rate": 0.0004999897862180678,
"loss": 6.7328,
"mean_token_accuracy": 0.11823359355330468,
"num_tokens": 2729556.0,
"step": 1345
},
{
"entropy": 6.916531419754028,
"epoch": 0.11797605523027178,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999894871503386,
"loss": 6.8563,
"mean_token_accuracy": 0.10672288462519645,
"num_tokens": 2741144.0,
"step": 1350
},
{
"entropy": 7.106829738616943,
"epoch": 0.11841300358297649,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999891837671932,
"loss": 6.7757,
"mean_token_accuracy": 0.11289330124855042,
"num_tokens": 2751384.0,
"step": 1355
},
{
"entropy": 6.979836463928223,
"epoch": 0.11884995193568121,
"grad_norm": 1.109375,
"learning_rate": 0.0004999888760686375,
"loss": 6.7905,
"mean_token_accuracy": 0.10638567060232162,
"num_tokens": 2761044.0,
"step": 1360
},
{
"entropy": 6.934152507781983,
"epoch": 0.11928690028838591,
"grad_norm": 0.9140625,
"learning_rate": 0.000499988564054677,
"loss": 6.7447,
"mean_token_accuracy": 0.11002607867121697,
"num_tokens": 2770946.0,
"step": 1365
},
{
"entropy": 7.068941926956176,
"epoch": 0.11972384864109062,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999882477253182,
"loss": 6.8113,
"mean_token_accuracy": 0.1063067264854908,
"num_tokens": 2781247.0,
"step": 1370
},
{
"entropy": 6.94280161857605,
"epoch": 0.12016079699379534,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999879270805667,
"loss": 6.6546,
"mean_token_accuracy": 0.12014054954051971,
"num_tokens": 2792078.0,
"step": 1375
},
{
"entropy": 6.9622045993804935,
"epoch": 0.12059774534650004,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999876021204291,
"loss": 6.7449,
"mean_token_accuracy": 0.11271467059850693,
"num_tokens": 2802458.0,
"step": 1380
},
{
"entropy": 6.967449569702149,
"epoch": 0.12103469369920475,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999872728449112,
"loss": 6.886,
"mean_token_accuracy": 0.10394820868968964,
"num_tokens": 2813352.0,
"step": 1385
},
{
"entropy": 7.045512008666992,
"epoch": 0.12147164205190947,
"grad_norm": 1.03125,
"learning_rate": 0.0004999869392540195,
"loss": 6.7739,
"mean_token_accuracy": 0.10725075080990791,
"num_tokens": 2822954.0,
"step": 1390
},
{
"entropy": 7.02110652923584,
"epoch": 0.12190859040461417,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999866013477605,
"loss": 6.8006,
"mean_token_accuracy": 0.10808226242661476,
"num_tokens": 2833375.0,
"step": 1395
},
{
"entropy": 6.927406930923462,
"epoch": 0.12234553875731889,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999862591261407,
"loss": 6.7472,
"mean_token_accuracy": 0.10812352672219276,
"num_tokens": 2843103.0,
"step": 1400
},
{
"entropy": 7.004934597015381,
"epoch": 0.1227824871100236,
"grad_norm": 0.953125,
"learning_rate": 0.0004999859125891664,
"loss": 6.8174,
"mean_token_accuracy": 0.10998742878437043,
"num_tokens": 2853801.0,
"step": 1405
},
{
"entropy": 7.010592699050903,
"epoch": 0.1232194354627283,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999855617368445,
"loss": 6.7962,
"mean_token_accuracy": 0.10651249438524246,
"num_tokens": 2863851.0,
"step": 1410
},
{
"entropy": 6.922228670120239,
"epoch": 0.12365638381543302,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999852065691816,
"loss": 6.7389,
"mean_token_accuracy": 0.11268523782491684,
"num_tokens": 2875073.0,
"step": 1415
},
{
"entropy": 7.0324512958526615,
"epoch": 0.12409333216813773,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999848470861846,
"loss": 6.7683,
"mean_token_accuracy": 0.11101210638880729,
"num_tokens": 2885676.0,
"step": 1420
},
{
"entropy": 6.929794263839722,
"epoch": 0.12453028052084243,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999844832878602,
"loss": 6.7645,
"mean_token_accuracy": 0.11524223163723946,
"num_tokens": 2896663.0,
"step": 1425
},
{
"entropy": 7.003040027618408,
"epoch": 0.12496722887354715,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999841151742156,
"loss": 6.7606,
"mean_token_accuracy": 0.11545668840408325,
"num_tokens": 2907555.0,
"step": 1430
},
{
"entropy": 6.966159200668335,
"epoch": 0.12540417722625186,
"grad_norm": 1.0625,
"learning_rate": 0.0004999837427452578,
"loss": 6.7548,
"mean_token_accuracy": 0.1095295563340187,
"num_tokens": 2917173.0,
"step": 1435
},
{
"entropy": 6.940775394439697,
"epoch": 0.12584112557895658,
"grad_norm": 1.015625,
"learning_rate": 0.000499983366000994,
"loss": 6.7623,
"mean_token_accuracy": 0.1120285913348198,
"num_tokens": 2928268.0,
"step": 1440
},
{
"entropy": 7.006380558013916,
"epoch": 0.12627807393166127,
"grad_norm": 1.046875,
"learning_rate": 0.0004999829849414312,
"loss": 6.7183,
"mean_token_accuracy": 0.11839316189289092,
"num_tokens": 2938203.0,
"step": 1445
},
{
"entropy": 6.905020713806152,
"epoch": 0.12671502228436599,
"grad_norm": 1.0078125,
"learning_rate": 0.000499982599566577,
"loss": 6.743,
"mean_token_accuracy": 0.11780567914247513,
"num_tokens": 2948094.0,
"step": 1450
},
{
"entropy": 6.976879024505616,
"epoch": 0.1271519706370707,
"grad_norm": 1.0625,
"learning_rate": 0.0004999822098764385,
"loss": 6.7402,
"mean_token_accuracy": 0.10916817337274551,
"num_tokens": 2957865.0,
"step": 1455
},
{
"entropy": 6.912422466278076,
"epoch": 0.1275889189897754,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999818158710234,
"loss": 6.6597,
"mean_token_accuracy": 0.11891222149133682,
"num_tokens": 2967880.0,
"step": 1460
},
{
"entropy": 6.932916593551636,
"epoch": 0.12802586734248012,
"grad_norm": 1.046875,
"learning_rate": 0.0004999814175503392,
"loss": 6.7246,
"mean_token_accuracy": 0.11173751205205917,
"num_tokens": 2979537.0,
"step": 1465
},
{
"entropy": 6.902735710144043,
"epoch": 0.12846281569518483,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999810149143935,
"loss": 6.7097,
"mean_token_accuracy": 0.11276535987854004,
"num_tokens": 2989700.0,
"step": 1470
},
{
"entropy": 6.83418345451355,
"epoch": 0.12889976404788953,
"grad_norm": 1.0,
"learning_rate": 0.000499980607963194,
"loss": 6.7178,
"mean_token_accuracy": 0.11464745998382568,
"num_tokens": 2999141.0,
"step": 1475
},
{
"entropy": 6.9692826747894285,
"epoch": 0.12933671240059424,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999801966967486,
"loss": 6.6815,
"mean_token_accuracy": 0.11767361015081405,
"num_tokens": 3008259.0,
"step": 1480
},
{
"entropy": 6.939387321472168,
"epoch": 0.12977366075329896,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999797811150652,
"loss": 6.7097,
"mean_token_accuracy": 0.11343906447291374,
"num_tokens": 3018281.0,
"step": 1485
},
{
"entropy": 6.963279438018799,
"epoch": 0.13021060910600368,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999793612181516,
"loss": 6.7186,
"mean_token_accuracy": 0.11935974359512329,
"num_tokens": 3029062.0,
"step": 1490
},
{
"entropy": 6.882570552825928,
"epoch": 0.13064755745870837,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999789370060161,
"loss": 6.6863,
"mean_token_accuracy": 0.12017739489674568,
"num_tokens": 3038142.0,
"step": 1495
},
{
"entropy": 6.990508508682251,
"epoch": 0.1310845058114131,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999785084786667,
"loss": 6.7944,
"mean_token_accuracy": 0.1040409967303276,
"num_tokens": 3049123.0,
"step": 1500
},
{
"entropy": 6.859305667877197,
"epoch": 0.1315214541641178,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999780756361114,
"loss": 6.6964,
"mean_token_accuracy": 0.11377365663647651,
"num_tokens": 3059031.0,
"step": 1505
},
{
"entropy": 6.957183074951172,
"epoch": 0.1319584025168225,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999776384783589,
"loss": 6.7387,
"mean_token_accuracy": 0.11297168284654617,
"num_tokens": 3068830.0,
"step": 1510
},
{
"entropy": 6.836016607284546,
"epoch": 0.13239535086952722,
"grad_norm": 0.8046875,
"learning_rate": 0.0004999771970054174,
"loss": 6.7098,
"mean_token_accuracy": 0.11699092388153076,
"num_tokens": 3080124.0,
"step": 1515
},
{
"entropy": 6.9651100635528564,
"epoch": 0.13283229922223194,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999767512172953,
"loss": 6.72,
"mean_token_accuracy": 0.10890761166810989,
"num_tokens": 3089607.0,
"step": 1520
},
{
"entropy": 6.871801853179932,
"epoch": 0.13326924757493663,
"grad_norm": 0.90625,
"learning_rate": 0.0004999763011140013,
"loss": 6.7972,
"mean_token_accuracy": 0.10377966463565827,
"num_tokens": 3099363.0,
"step": 1525
},
{
"entropy": 7.077904081344604,
"epoch": 0.13370619592764135,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999758466955439,
"loss": 6.7009,
"mean_token_accuracy": 0.11966917067766189,
"num_tokens": 3109552.0,
"step": 1530
},
{
"entropy": 6.822536611557007,
"epoch": 0.13414314428034607,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999753879619319,
"loss": 6.6813,
"mean_token_accuracy": 0.11428188979625702,
"num_tokens": 3119536.0,
"step": 1535
},
{
"entropy": 6.9331159591674805,
"epoch": 0.13458009263305076,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999749249131741,
"loss": 6.7238,
"mean_token_accuracy": 0.11595344543457031,
"num_tokens": 3130552.0,
"step": 1540
},
{
"entropy": 6.9094908237457275,
"epoch": 0.13501704098575548,
"grad_norm": 0.9375,
"learning_rate": 0.0004999744575492793,
"loss": 6.6219,
"mean_token_accuracy": 0.11641833782196045,
"num_tokens": 3140723.0,
"step": 1545
},
{
"entropy": 6.898475456237793,
"epoch": 0.1354539893384602,
"grad_norm": 0.90625,
"learning_rate": 0.0004999739858702567,
"loss": 6.6648,
"mean_token_accuracy": 0.12024381458759308,
"num_tokens": 3151031.0,
"step": 1550
},
{
"entropy": 6.850230026245117,
"epoch": 0.1358909376911649,
"grad_norm": 0.9140625,
"learning_rate": 0.000499973509876115,
"loss": 6.6687,
"mean_token_accuracy": 0.11914398446679116,
"num_tokens": 3161373.0,
"step": 1555
},
{
"entropy": 6.880463647842407,
"epoch": 0.1363278860438696,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999730295668634,
"loss": 6.6831,
"mean_token_accuracy": 0.11570274606347083,
"num_tokens": 3171094.0,
"step": 1560
},
{
"entropy": 6.895293521881103,
"epoch": 0.13676483439657433,
"grad_norm": 1.046875,
"learning_rate": 0.0004999725449425114,
"loss": 6.6283,
"mean_token_accuracy": 0.1216744564473629,
"num_tokens": 3181646.0,
"step": 1565
},
{
"entropy": 6.8003284454345705,
"epoch": 0.13720178274927902,
"grad_norm": 1.0703125,
"learning_rate": 0.000499972056003068,
"loss": 6.603,
"mean_token_accuracy": 0.12167230620980263,
"num_tokens": 3190727.0,
"step": 1570
},
{
"entropy": 6.9626048564910885,
"epoch": 0.13763873110198374,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999715627485426,
"loss": 6.7085,
"mean_token_accuracy": 0.10727820545434952,
"num_tokens": 3200316.0,
"step": 1575
},
{
"entropy": 6.841419315338134,
"epoch": 0.13807567945468846,
"grad_norm": 0.8984375,
"learning_rate": 0.000499971065178945,
"loss": 6.6663,
"mean_token_accuracy": 0.11115863621234894,
"num_tokens": 3211830.0,
"step": 1580
},
{
"entropy": 6.864039182662964,
"epoch": 0.13851262780739318,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999705632942843,
"loss": 6.6778,
"mean_token_accuracy": 0.11532776281237603,
"num_tokens": 3222711.0,
"step": 1585
},
{
"entropy": 6.850924634933472,
"epoch": 0.13894957616009787,
"grad_norm": 0.921875,
"learning_rate": 0.0004999700570945704,
"loss": 6.5875,
"mean_token_accuracy": 0.1223737932741642,
"num_tokens": 3232810.0,
"step": 1590
},
{
"entropy": 6.806768321990967,
"epoch": 0.1393865245128026,
"grad_norm": 1.0625,
"learning_rate": 0.0004999695465798128,
"loss": 6.6716,
"mean_token_accuracy": 0.11834604367613792,
"num_tokens": 3243366.0,
"step": 1595
},
{
"entropy": 6.887040901184082,
"epoch": 0.1398234728655073,
"grad_norm": 0.9375,
"learning_rate": 0.0004999690317500215,
"loss": 6.6644,
"mean_token_accuracy": 0.1145846463739872,
"num_tokens": 3254000.0,
"step": 1600
},
{
"entropy": 6.924470329284668,
"epoch": 0.140260421218212,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999685126052063,
"loss": 6.6944,
"mean_token_accuracy": 0.11372155621647835,
"num_tokens": 3263601.0,
"step": 1605
},
{
"entropy": 6.830214357376098,
"epoch": 0.14069736957091672,
"grad_norm": 0.953125,
"learning_rate": 0.0004999679891453772,
"loss": 6.7183,
"mean_token_accuracy": 0.11202776506543159,
"num_tokens": 3273243.0,
"step": 1610
},
{
"entropy": 6.92011456489563,
"epoch": 0.14113431792362144,
"grad_norm": 0.86328125,
"learning_rate": 0.000499967461370544,
"loss": 6.6652,
"mean_token_accuracy": 0.11686113476753235,
"num_tokens": 3284208.0,
"step": 1615
},
{
"entropy": 6.883434915542603,
"epoch": 0.14157126627632613,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999669292807172,
"loss": 6.646,
"mean_token_accuracy": 0.11703974828124046,
"num_tokens": 3294216.0,
"step": 1620
},
{
"entropy": 6.908010053634643,
"epoch": 0.14200821462903085,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999663928759066,
"loss": 6.6971,
"mean_token_accuracy": 0.11337215676903725,
"num_tokens": 3304576.0,
"step": 1625
},
{
"entropy": 6.85626482963562,
"epoch": 0.14244516298173557,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999658521561229,
"loss": 6.6682,
"mean_token_accuracy": 0.10966120883822442,
"num_tokens": 3314214.0,
"step": 1630
},
{
"entropy": 6.887621068954468,
"epoch": 0.14288211133444026,
"grad_norm": 0.953125,
"learning_rate": 0.0004999653071213763,
"loss": 6.6773,
"mean_token_accuracy": 0.11030459403991699,
"num_tokens": 3325527.0,
"step": 1635
},
{
"entropy": 6.751001405715942,
"epoch": 0.14331905968714498,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999647577716771,
"loss": 6.5835,
"mean_token_accuracy": 0.11718848198652268,
"num_tokens": 3334511.0,
"step": 1640
},
{
"entropy": 6.8687464714050295,
"epoch": 0.1437560080398497,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999642041070361,
"loss": 6.5728,
"mean_token_accuracy": 0.12193055376410485,
"num_tokens": 3343922.0,
"step": 1645
},
{
"entropy": 6.801642322540284,
"epoch": 0.1441929563925544,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999636461274637,
"loss": 6.6072,
"mean_token_accuracy": 0.11523821577429771,
"num_tokens": 3353895.0,
"step": 1650
},
{
"entropy": 6.818548059463501,
"epoch": 0.1446299047452591,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999630838329706,
"loss": 6.6567,
"mean_token_accuracy": 0.11228160336613655,
"num_tokens": 3364637.0,
"step": 1655
},
{
"entropy": 6.882724046707153,
"epoch": 0.14506685309796383,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999625172235679,
"loss": 6.6184,
"mean_token_accuracy": 0.11534506604075431,
"num_tokens": 3374308.0,
"step": 1660
},
{
"entropy": 6.775670385360717,
"epoch": 0.14550380145066852,
"grad_norm": 0.921875,
"learning_rate": 0.000499961946299266,
"loss": 6.6,
"mean_token_accuracy": 0.11504803895950318,
"num_tokens": 3383637.0,
"step": 1665
},
{
"entropy": 6.865798473358154,
"epoch": 0.14594074980337324,
"grad_norm": 1.03125,
"learning_rate": 0.0004999613710600763,
"loss": 6.6789,
"mean_token_accuracy": 0.11600840091705322,
"num_tokens": 3394150.0,
"step": 1670
},
{
"entropy": 6.889613914489746,
"epoch": 0.14637769815607796,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999607915060094,
"loss": 6.6368,
"mean_token_accuracy": 0.11584951207041741,
"num_tokens": 3404445.0,
"step": 1675
},
{
"entropy": 6.7219658374786375,
"epoch": 0.14681464650878268,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999602076370768,
"loss": 6.541,
"mean_token_accuracy": 0.12608898878097535,
"num_tokens": 3415183.0,
"step": 1680
},
{
"entropy": 6.776702976226806,
"epoch": 0.14725159486148737,
"grad_norm": 1.046875,
"learning_rate": 0.0004999596194532895,
"loss": 6.5253,
"mean_token_accuracy": 0.12415588200092316,
"num_tokens": 3424824.0,
"step": 1685
},
{
"entropy": 6.871752786636352,
"epoch": 0.1476885432141921,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999590269546588,
"loss": 6.6382,
"mean_token_accuracy": 0.11570465192198753,
"num_tokens": 3434789.0,
"step": 1690
},
{
"entropy": 6.707127237319947,
"epoch": 0.1481254915668968,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999584301411961,
"loss": 6.5214,
"mean_token_accuracy": 0.1381271928548813,
"num_tokens": 3445810.0,
"step": 1695
},
{
"entropy": 6.895666456222534,
"epoch": 0.1485624399196015,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999578290129129,
"loss": 6.661,
"mean_token_accuracy": 0.11268147677183152,
"num_tokens": 3455510.0,
"step": 1700
},
{
"entropy": 6.774993848800659,
"epoch": 0.14899938827230622,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999572235698206,
"loss": 6.6274,
"mean_token_accuracy": 0.11752628609538078,
"num_tokens": 3465785.0,
"step": 1705
},
{
"entropy": 6.814649772644043,
"epoch": 0.14943633662501093,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999566138119308,
"loss": 6.6843,
"mean_token_accuracy": 0.1124463550746441,
"num_tokens": 3477568.0,
"step": 1710
},
{
"entropy": 6.754849433898926,
"epoch": 0.14987328497771563,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999559997392554,
"loss": 6.5138,
"mean_token_accuracy": 0.13679737225174904,
"num_tokens": 3487703.0,
"step": 1715
},
{
"entropy": 6.805766534805298,
"epoch": 0.15031023333042035,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999553813518061,
"loss": 6.5906,
"mean_token_accuracy": 0.1175839364528656,
"num_tokens": 3497757.0,
"step": 1720
},
{
"entropy": 6.736756134033203,
"epoch": 0.15074718168312506,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999547586495946,
"loss": 6.5553,
"mean_token_accuracy": 0.1203412413597107,
"num_tokens": 3507594.0,
"step": 1725
},
{
"entropy": 6.828785610198975,
"epoch": 0.15118413003582976,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999541316326329,
"loss": 6.6505,
"mean_token_accuracy": 0.11356838271021844,
"num_tokens": 3517889.0,
"step": 1730
},
{
"entropy": 6.788142585754395,
"epoch": 0.15162107838853447,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999535003009332,
"loss": 6.621,
"mean_token_accuracy": 0.11979384943842888,
"num_tokens": 3528158.0,
"step": 1735
},
{
"entropy": 6.820278310775757,
"epoch": 0.1520580267412392,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999528646545074,
"loss": 6.6304,
"mean_token_accuracy": 0.1174927718937397,
"num_tokens": 3539099.0,
"step": 1740
},
{
"entropy": 6.8182226657867435,
"epoch": 0.15249497509394389,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999522246933678,
"loss": 6.631,
"mean_token_accuracy": 0.11622772663831711,
"num_tokens": 3549819.0,
"step": 1745
},
{
"entropy": 6.893373823165893,
"epoch": 0.1529319234466486,
"grad_norm": 1.09375,
"learning_rate": 0.0004999515804175268,
"loss": 6.632,
"mean_token_accuracy": 0.11604494601488113,
"num_tokens": 3560022.0,
"step": 1750
},
{
"entropy": 6.761956310272216,
"epoch": 0.15336887179935332,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999509318269964,
"loss": 6.5647,
"mean_token_accuracy": 0.1173597402870655,
"num_tokens": 3569681.0,
"step": 1755
},
{
"entropy": 6.857851600646972,
"epoch": 0.15380582015205801,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999502789217894,
"loss": 6.6579,
"mean_token_accuracy": 0.11668316721916198,
"num_tokens": 3580731.0,
"step": 1760
},
{
"entropy": 6.75460410118103,
"epoch": 0.15424276850476273,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999496217019181,
"loss": 6.6264,
"mean_token_accuracy": 0.12062862291932105,
"num_tokens": 3590482.0,
"step": 1765
},
{
"entropy": 6.796440553665161,
"epoch": 0.15467971685746745,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999489601673952,
"loss": 6.6159,
"mean_token_accuracy": 0.11088166609406472,
"num_tokens": 3600988.0,
"step": 1770
},
{
"entropy": 6.9524415016174315,
"epoch": 0.15511666521017214,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999482943182334,
"loss": 6.6848,
"mean_token_accuracy": 0.1131463699042797,
"num_tokens": 3611226.0,
"step": 1775
},
{
"entropy": 6.716507911682129,
"epoch": 0.15555361356287686,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999476241544454,
"loss": 6.5993,
"mean_token_accuracy": 0.11544200181961059,
"num_tokens": 3621324.0,
"step": 1780
},
{
"entropy": 6.776705932617188,
"epoch": 0.15599056191558158,
"grad_norm": 0.96875,
"learning_rate": 0.0004999469496760441,
"loss": 6.5311,
"mean_token_accuracy": 0.11900553330779076,
"num_tokens": 3631650.0,
"step": 1785
},
{
"entropy": 6.790343856811523,
"epoch": 0.1564275102682863,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999462708830425,
"loss": 6.5413,
"mean_token_accuracy": 0.13263114020228386,
"num_tokens": 3641173.0,
"step": 1790
},
{
"entropy": 6.7471825122833256,
"epoch": 0.156864458620991,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999455877754535,
"loss": 6.5587,
"mean_token_accuracy": 0.1188740074634552,
"num_tokens": 3651846.0,
"step": 1795
},
{
"entropy": 6.735989761352539,
"epoch": 0.1573014069736957,
"grad_norm": 0.96875,
"learning_rate": 0.0004999449003532903,
"loss": 6.5573,
"mean_token_accuracy": 0.116403878480196,
"num_tokens": 3661174.0,
"step": 1800
},
{
"entropy": 6.798489284515381,
"epoch": 0.15773835532640043,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999442086165661,
"loss": 6.5437,
"mean_token_accuracy": 0.11836761459708214,
"num_tokens": 3671042.0,
"step": 1805
},
{
"entropy": 6.753465700149536,
"epoch": 0.15817530367910512,
"grad_norm": 0.94140625,
"learning_rate": 0.000499943512565294,
"loss": 6.5084,
"mean_token_accuracy": 0.11789297163486481,
"num_tokens": 3681206.0,
"step": 1810
},
{
"entropy": 6.780641365051269,
"epoch": 0.15861225203180984,
"grad_norm": 0.9375,
"learning_rate": 0.0004999428121994874,
"loss": 6.6077,
"mean_token_accuracy": 0.12206789627671241,
"num_tokens": 3691748.0,
"step": 1815
},
{
"entropy": 6.706904029846191,
"epoch": 0.15904920038451456,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999421075191599,
"loss": 6.505,
"mean_token_accuracy": 0.12005864605307578,
"num_tokens": 3701730.0,
"step": 1820
},
{
"entropy": 6.79358229637146,
"epoch": 0.15948614873721925,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999413985243249,
"loss": 6.6198,
"mean_token_accuracy": 0.11459683477878571,
"num_tokens": 3712525.0,
"step": 1825
},
{
"entropy": 6.8429272174835205,
"epoch": 0.15992309708992397,
"grad_norm": 0.95703125,
"learning_rate": 0.000499940685214996,
"loss": 6.5678,
"mean_token_accuracy": 0.11963910087943078,
"num_tokens": 3722016.0,
"step": 1830
},
{
"entropy": 6.786110496520996,
"epoch": 0.1603600454426287,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999399675911869,
"loss": 6.476,
"mean_token_accuracy": 0.12395038083195686,
"num_tokens": 3731299.0,
"step": 1835
},
{
"entropy": 6.662138032913208,
"epoch": 0.16079699379533338,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999392456529113,
"loss": 6.5372,
"mean_token_accuracy": 0.1260882593691349,
"num_tokens": 3740922.0,
"step": 1840
},
{
"entropy": 6.7078369617462155,
"epoch": 0.1612339421480381,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999385194001831,
"loss": 6.5398,
"mean_token_accuracy": 0.11944171339273453,
"num_tokens": 3750339.0,
"step": 1845
},
{
"entropy": 6.79173173904419,
"epoch": 0.16167089050074282,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999377888330164,
"loss": 6.5166,
"mean_token_accuracy": 0.12448024600744248,
"num_tokens": 3761379.0,
"step": 1850
},
{
"entropy": 6.675159645080567,
"epoch": 0.1621078388534475,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999370539514249,
"loss": 6.5731,
"mean_token_accuracy": 0.11835632175207138,
"num_tokens": 3771649.0,
"step": 1855
},
{
"entropy": 6.754816055297852,
"epoch": 0.16254478720615223,
"grad_norm": 0.953125,
"learning_rate": 0.0004999363147554229,
"loss": 6.4983,
"mean_token_accuracy": 0.1242656834423542,
"num_tokens": 3782393.0,
"step": 1860
},
{
"entropy": 6.758715486526489,
"epoch": 0.16298173555885695,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999355712450244,
"loss": 6.5619,
"mean_token_accuracy": 0.12336914837360383,
"num_tokens": 3792849.0,
"step": 1865
},
{
"entropy": 6.747960758209229,
"epoch": 0.16341868391156164,
"grad_norm": 1.015625,
"learning_rate": 0.0004999348234202438,
"loss": 6.5009,
"mean_token_accuracy": 0.1251760870218277,
"num_tokens": 3803848.0,
"step": 1870
},
{
"entropy": 6.647700023651123,
"epoch": 0.16385563226426636,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999340712810956,
"loss": 6.4194,
"mean_token_accuracy": 0.13567735776305198,
"num_tokens": 3814704.0,
"step": 1875
},
{
"entropy": 6.708374786376953,
"epoch": 0.16429258061697108,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999333148275939,
"loss": 6.5592,
"mean_token_accuracy": 0.11908423453569413,
"num_tokens": 3824955.0,
"step": 1880
},
{
"entropy": 6.752207708358765,
"epoch": 0.1647295289696758,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999325540597535,
"loss": 6.5372,
"mean_token_accuracy": 0.12644217535853386,
"num_tokens": 3836540.0,
"step": 1885
},
{
"entropy": 6.604652452468872,
"epoch": 0.1651664773223805,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999317889775887,
"loss": 6.431,
"mean_token_accuracy": 0.13235808461904525,
"num_tokens": 3846472.0,
"step": 1890
},
{
"entropy": 6.771966743469238,
"epoch": 0.1656034256750852,
"grad_norm": 0.875,
"learning_rate": 0.0004999310195811143,
"loss": 6.5333,
"mean_token_accuracy": 0.11936069577932358,
"num_tokens": 3857187.0,
"step": 1895
},
{
"entropy": 6.751552248001099,
"epoch": 0.16604037402778993,
"grad_norm": 0.984375,
"learning_rate": 0.0004999302458703451,
"loss": 6.5008,
"mean_token_accuracy": 0.12447258234024047,
"num_tokens": 3866259.0,
"step": 1900
},
{
"entropy": 6.657085752487182,
"epoch": 0.16647732238049462,
"grad_norm": 1.0390625,
"learning_rate": 0.000499929467845296,
"loss": 6.5587,
"mean_token_accuracy": 0.11343137621879577,
"num_tokens": 3876869.0,
"step": 1905
},
{
"entropy": 6.7602463722229,
"epoch": 0.16691427073319934,
"grad_norm": 0.875,
"learning_rate": 0.0004999286855059818,
"loss": 6.5404,
"mean_token_accuracy": 0.12126922458410264,
"num_tokens": 3888278.0,
"step": 1910
},
{
"entropy": 6.7652349948883055,
"epoch": 0.16735121908590406,
"grad_norm": 1.125,
"learning_rate": 0.0004999278988524177,
"loss": 6.6004,
"mean_token_accuracy": 0.12169176936149598,
"num_tokens": 3898648.0,
"step": 1915
},
{
"entropy": 6.759922075271606,
"epoch": 0.16778816743860875,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999271078846184,
"loss": 6.5229,
"mean_token_accuracy": 0.11797695755958557,
"num_tokens": 3908282.0,
"step": 1920
},
{
"entropy": 6.548764705657959,
"epoch": 0.16822511579131347,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999263126025994,
"loss": 6.5006,
"mean_token_accuracy": 0.12412167713046074,
"num_tokens": 3918333.0,
"step": 1925
},
{
"entropy": 6.82745590209961,
"epoch": 0.1686620641440182,
"grad_norm": 1.0,
"learning_rate": 0.0004999255130063759,
"loss": 6.51,
"mean_token_accuracy": 0.1215559110045433,
"num_tokens": 3927403.0,
"step": 1930
},
{
"entropy": 6.691656970977784,
"epoch": 0.16909901249672288,
"grad_norm": 0.96875,
"learning_rate": 0.0004999247090959633,
"loss": 6.5286,
"mean_token_accuracy": 0.12117488756775856,
"num_tokens": 3937352.0,
"step": 1935
},
{
"entropy": 6.6749176502227785,
"epoch": 0.1695359608494276,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999239008713767,
"loss": 6.3616,
"mean_token_accuracy": 0.12652190998196602,
"num_tokens": 3947164.0,
"step": 1940
},
{
"entropy": 6.7179159164428714,
"epoch": 0.16997290920213232,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999230883326319,
"loss": 6.5359,
"mean_token_accuracy": 0.1260188527405262,
"num_tokens": 3956887.0,
"step": 1945
},
{
"entropy": 6.743377208709717,
"epoch": 0.170409857554837,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999222714797445,
"loss": 6.5084,
"mean_token_accuracy": 0.1210441879928112,
"num_tokens": 3966542.0,
"step": 1950
},
{
"entropy": 6.723532009124756,
"epoch": 0.17084680590754173,
"grad_norm": 1.0078125,
"learning_rate": 0.00049992145031273,
"loss": 6.4706,
"mean_token_accuracy": 0.13221301287412643,
"num_tokens": 3976127.0,
"step": 1955
},
{
"entropy": 6.577977323532105,
"epoch": 0.17128375426024645,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999206248316043,
"loss": 6.4402,
"mean_token_accuracy": 0.12495635226368904,
"num_tokens": 3987371.0,
"step": 1960
},
{
"entropy": 6.643271446228027,
"epoch": 0.17172070261295114,
"grad_norm": 0.98046875,
"learning_rate": 0.000499919795036383,
"loss": 6.441,
"mean_token_accuracy": 0.12783539444208145,
"num_tokens": 3997924.0,
"step": 1965
},
{
"entropy": 6.687552833557129,
"epoch": 0.17215765096565586,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999189609270823,
"loss": 6.5346,
"mean_token_accuracy": 0.12447880506515503,
"num_tokens": 4009150.0,
"step": 1970
},
{
"entropy": 6.702529954910278,
"epoch": 0.17259459931836058,
"grad_norm": 0.93359375,
"learning_rate": 0.000499918122503718,
"loss": 6.5461,
"mean_token_accuracy": 0.12213218584656715,
"num_tokens": 4018899.0,
"step": 1975
},
{
"entropy": 6.684345149993897,
"epoch": 0.17303154767106527,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999172797663064,
"loss": 6.489,
"mean_token_accuracy": 0.12457576245069504,
"num_tokens": 4028652.0,
"step": 1980
},
{
"entropy": 6.704040813446045,
"epoch": 0.17346849602376999,
"grad_norm": 0.90625,
"learning_rate": 0.0004999164327148634,
"loss": 6.53,
"mean_token_accuracy": 0.11930816993117332,
"num_tokens": 4039819.0,
"step": 1985
},
{
"entropy": 6.703826856613159,
"epoch": 0.1739054443764747,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999155813494053,
"loss": 6.4394,
"mean_token_accuracy": 0.12949882224202156,
"num_tokens": 4050029.0,
"step": 1990
},
{
"entropy": 6.5715407371521,
"epoch": 0.17434239272917942,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999147256699487,
"loss": 6.4687,
"mean_token_accuracy": 0.1293756529688835,
"num_tokens": 4060497.0,
"step": 1995
},
{
"entropy": 6.748078489303589,
"epoch": 0.17477934108188412,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999138656765096,
"loss": 6.45,
"mean_token_accuracy": 0.1262899488210678,
"num_tokens": 4071477.0,
"step": 2000
},
{
"entropy": 6.660907793045044,
"epoch": 0.17521628943458883,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999130013691049,
"loss": 6.5565,
"mean_token_accuracy": 0.12222278639674186,
"num_tokens": 4081869.0,
"step": 2005
},
{
"entropy": 6.7544700622558596,
"epoch": 0.17565323778729355,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999121327477509,
"loss": 6.528,
"mean_token_accuracy": 0.12365535125136376,
"num_tokens": 4092797.0,
"step": 2010
},
{
"entropy": 6.605335807800293,
"epoch": 0.17609018613999824,
"grad_norm": 0.921875,
"learning_rate": 0.0004999112598124642,
"loss": 6.4776,
"mean_token_accuracy": 0.12264428436756133,
"num_tokens": 4103688.0,
"step": 2015
},
{
"entropy": 6.75774884223938,
"epoch": 0.17652713449270296,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999103825632618,
"loss": 6.4325,
"mean_token_accuracy": 0.12662287577986717,
"num_tokens": 4114037.0,
"step": 2020
},
{
"entropy": 6.603244209289551,
"epoch": 0.17696408284540768,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999095010001605,
"loss": 6.4215,
"mean_token_accuracy": 0.12197433784604073,
"num_tokens": 4124164.0,
"step": 2025
},
{
"entropy": 6.765624523162842,
"epoch": 0.17740103119811237,
"grad_norm": 0.9375,
"learning_rate": 0.000499908615123177,
"loss": 6.5396,
"mean_token_accuracy": 0.11871138587594032,
"num_tokens": 4134225.0,
"step": 2030
},
{
"entropy": 6.675165271759033,
"epoch": 0.1778379795508171,
"grad_norm": 0.890625,
"learning_rate": 0.0004999077249323284,
"loss": 6.5901,
"mean_token_accuracy": 0.11778633445501327,
"num_tokens": 4145962.0,
"step": 2035
},
{
"entropy": 6.702001953125,
"epoch": 0.1782749279035218,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999068304276319,
"loss": 6.4709,
"mean_token_accuracy": 0.12179390788078308,
"num_tokens": 4156572.0,
"step": 2040
},
{
"entropy": 6.660705947875977,
"epoch": 0.1787118762562265,
"grad_norm": 0.9375,
"learning_rate": 0.0004999059316091045,
"loss": 6.5431,
"mean_token_accuracy": 0.12079031616449357,
"num_tokens": 4166960.0,
"step": 2045
},
{
"entropy": 6.70681414604187,
"epoch": 0.17914882460893122,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999050284767634,
"loss": 6.4874,
"mean_token_accuracy": 0.12624370977282523,
"num_tokens": 4177833.0,
"step": 2050
},
{
"entropy": 6.578025674819946,
"epoch": 0.17958577296163594,
"grad_norm": 1.015625,
"learning_rate": 0.0004999041210306261,
"loss": 6.4109,
"mean_token_accuracy": 0.12834745794534683,
"num_tokens": 4187858.0,
"step": 2055
},
{
"entropy": 6.776542139053345,
"epoch": 0.18002272131434063,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999032092707098,
"loss": 6.545,
"mean_token_accuracy": 0.12043375000357628,
"num_tokens": 4198207.0,
"step": 2060
},
{
"entropy": 6.6667626857757565,
"epoch": 0.18045966966704535,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999022931970323,
"loss": 6.425,
"mean_token_accuracy": 0.12915648147463799,
"num_tokens": 4207594.0,
"step": 2065
},
{
"entropy": 6.629980421066284,
"epoch": 0.18089661801975007,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999013728096107,
"loss": 6.431,
"mean_token_accuracy": 0.12719237804412842,
"num_tokens": 4217334.0,
"step": 2070
},
{
"entropy": 6.644939041137695,
"epoch": 0.18133356637245476,
"grad_norm": 0.9765625,
"learning_rate": 0.000499900448108463,
"loss": 6.4226,
"mean_token_accuracy": 0.13288431987166405,
"num_tokens": 4226407.0,
"step": 2075
},
{
"entropy": 6.604313325881958,
"epoch": 0.18177051472515948,
"grad_norm": 0.9765625,
"learning_rate": 0.000499899519093607,
"loss": 6.376,
"mean_token_accuracy": 0.13293503150343894,
"num_tokens": 4237818.0,
"step": 2080
},
{
"entropy": 6.664392614364624,
"epoch": 0.1822074630778642,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998985857650603,
"loss": 6.4837,
"mean_token_accuracy": 0.12130877822637558,
"num_tokens": 4247783.0,
"step": 2085
},
{
"entropy": 6.606699132919312,
"epoch": 0.18264441143056892,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998976481228409,
"loss": 6.4436,
"mean_token_accuracy": 0.12809643447399138,
"num_tokens": 4258049.0,
"step": 2090
},
{
"entropy": 6.618226480484009,
"epoch": 0.1830813597832736,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998967061669668,
"loss": 6.4403,
"mean_token_accuracy": 0.11963641420006751,
"num_tokens": 4269357.0,
"step": 2095
},
{
"entropy": 6.6365663528442385,
"epoch": 0.18351830813597833,
"grad_norm": 0.984375,
"learning_rate": 0.0004998957598974559,
"loss": 6.3581,
"mean_token_accuracy": 0.13459884151816368,
"num_tokens": 4279576.0,
"step": 2100
},
{
"entropy": 6.598727512359619,
"epoch": 0.18395525648868305,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998948093143266,
"loss": 6.4112,
"mean_token_accuracy": 0.1286292240023613,
"num_tokens": 4289761.0,
"step": 2105
},
{
"entropy": 6.582360744476318,
"epoch": 0.18439220484138774,
"grad_norm": 0.98828125,
"learning_rate": 0.000499893854417597,
"loss": 6.355,
"mean_token_accuracy": 0.12943675965070725,
"num_tokens": 4299231.0,
"step": 2110
},
{
"entropy": 6.613821411132813,
"epoch": 0.18482915319409246,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998928952072854,
"loss": 6.5108,
"mean_token_accuracy": 0.12351166307926179,
"num_tokens": 4308687.0,
"step": 2115
},
{
"entropy": 6.6865287780761715,
"epoch": 0.18526610154679718,
"grad_norm": 1.03125,
"learning_rate": 0.0004998919316834102,
"loss": 6.4583,
"mean_token_accuracy": 0.12498997375369073,
"num_tokens": 4318183.0,
"step": 2120
},
{
"entropy": 6.668766021728516,
"epoch": 0.18570304989950187,
"grad_norm": 0.98046875,
"learning_rate": 0.00049989096384599,
"loss": 6.5085,
"mean_token_accuracy": 0.12343138679862023,
"num_tokens": 4328482.0,
"step": 2125
},
{
"entropy": 6.633816528320312,
"epoch": 0.1861399982522066,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998899916950431,
"loss": 6.3824,
"mean_token_accuracy": 0.1267977774143219,
"num_tokens": 4339313.0,
"step": 2130
},
{
"entropy": 6.633962535858155,
"epoch": 0.1865769466049113,
"grad_norm": 1.015625,
"learning_rate": 0.0004998890152305885,
"loss": 6.4954,
"mean_token_accuracy": 0.12780801132321357,
"num_tokens": 4350299.0,
"step": 2135
},
{
"entropy": 6.715720367431641,
"epoch": 0.187013894957616,
"grad_norm": 0.984375,
"learning_rate": 0.0004998880344526447,
"loss": 6.4273,
"mean_token_accuracy": 0.13316971361637114,
"num_tokens": 4361424.0,
"step": 2140
},
{
"entropy": 6.532921981811524,
"epoch": 0.18745084331032072,
"grad_norm": 0.9375,
"learning_rate": 0.0004998870493612306,
"loss": 6.3943,
"mean_token_accuracy": 0.1273239456117153,
"num_tokens": 4371480.0,
"step": 2145
},
{
"entropy": 6.555461359024048,
"epoch": 0.18788779166302544,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998860599563649,
"loss": 6.4256,
"mean_token_accuracy": 0.12563225403428077,
"num_tokens": 4381951.0,
"step": 2150
},
{
"entropy": 6.6292718887329105,
"epoch": 0.18832474001573013,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998850662380668,
"loss": 6.4814,
"mean_token_accuracy": 0.12124204561114311,
"num_tokens": 4392290.0,
"step": 2155
},
{
"entropy": 6.614552593231201,
"epoch": 0.18876168836843485,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998840682063553,
"loss": 6.2488,
"mean_token_accuracy": 0.14245351031422615,
"num_tokens": 4403350.0,
"step": 2160
},
{
"entropy": 6.56493353843689,
"epoch": 0.18919863672113957,
"grad_norm": 0.875,
"learning_rate": 0.0004998830658612496,
"loss": 6.4447,
"mean_token_accuracy": 0.12840750589966773,
"num_tokens": 4413171.0,
"step": 2165
},
{
"entropy": 6.563513231277466,
"epoch": 0.18963558507384426,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998820592027688,
"loss": 6.3957,
"mean_token_accuracy": 0.1307518906891346,
"num_tokens": 4424371.0,
"step": 2170
},
{
"entropy": 6.649606466293335,
"epoch": 0.19007253342654898,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998810482309322,
"loss": 6.3667,
"mean_token_accuracy": 0.12607548013329506,
"num_tokens": 4434040.0,
"step": 2175
},
{
"entropy": 6.6798676490783695,
"epoch": 0.1905094817792537,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998800329457595,
"loss": 6.4636,
"mean_token_accuracy": 0.12016530409455299,
"num_tokens": 4444663.0,
"step": 2180
},
{
"entropy": 6.602048730850219,
"epoch": 0.1909464301319584,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998790133472697,
"loss": 6.4146,
"mean_token_accuracy": 0.1262466661632061,
"num_tokens": 4454733.0,
"step": 2185
},
{
"entropy": 6.578805065155029,
"epoch": 0.1913833784846631,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998779894354827,
"loss": 6.3965,
"mean_token_accuracy": 0.13079310059547425,
"num_tokens": 4464286.0,
"step": 2190
},
{
"entropy": 6.633804512023926,
"epoch": 0.19182032683736783,
"grad_norm": 0.88671875,
"learning_rate": 0.000499876961210418,
"loss": 6.389,
"mean_token_accuracy": 0.13454255908727647,
"num_tokens": 4475005.0,
"step": 2195
},
{
"entropy": 6.588158702850341,
"epoch": 0.19225727519007255,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998759286720953,
"loss": 6.336,
"mean_token_accuracy": 0.13137971311807634,
"num_tokens": 4484606.0,
"step": 2200
},
{
"entropy": 6.507455158233642,
"epoch": 0.19269422354277724,
"grad_norm": 0.96875,
"learning_rate": 0.0004998748918205345,
"loss": 6.4482,
"mean_token_accuracy": 0.12594832628965377,
"num_tokens": 4494930.0,
"step": 2205
},
{
"entropy": 6.659773206710815,
"epoch": 0.19313117189548196,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998738506557554,
"loss": 6.4699,
"mean_token_accuracy": 0.12103351801633835,
"num_tokens": 4505304.0,
"step": 2210
},
{
"entropy": 6.693742942810059,
"epoch": 0.19356812024818668,
"grad_norm": 0.921875,
"learning_rate": 0.000499872805177778,
"loss": 6.392,
"mean_token_accuracy": 0.1304948538541794,
"num_tokens": 4514333.0,
"step": 2215
},
{
"entropy": 6.483904218673706,
"epoch": 0.19400506860089137,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998717553866225,
"loss": 6.4429,
"mean_token_accuracy": 0.13353654071688653,
"num_tokens": 4525175.0,
"step": 2220
},
{
"entropy": 6.648404788970947,
"epoch": 0.1944420169535961,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998707012823087,
"loss": 6.3151,
"mean_token_accuracy": 0.13554888740181922,
"num_tokens": 4534212.0,
"step": 2225
},
{
"entropy": 6.492750024795532,
"epoch": 0.1948789653063008,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998696428648573,
"loss": 6.4957,
"mean_token_accuracy": 0.12358262240886689,
"num_tokens": 4545141.0,
"step": 2230
},
{
"entropy": 6.6611635208129885,
"epoch": 0.1953159136590055,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998685801342881,
"loss": 6.3623,
"mean_token_accuracy": 0.13657205328345298,
"num_tokens": 4554410.0,
"step": 2235
},
{
"entropy": 6.515309810638428,
"epoch": 0.19575286201171022,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998675130906217,
"loss": 6.3629,
"mean_token_accuracy": 0.1314949244260788,
"num_tokens": 4564397.0,
"step": 2240
},
{
"entropy": 6.506466865539551,
"epoch": 0.19618981036441493,
"grad_norm": 0.921875,
"learning_rate": 0.0004998664417338787,
"loss": 6.3565,
"mean_token_accuracy": 0.13614993840456008,
"num_tokens": 4573986.0,
"step": 2245
},
{
"entropy": 6.616925239562988,
"epoch": 0.19662675871711963,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998653660640794,
"loss": 6.3557,
"mean_token_accuracy": 0.1368450790643692,
"num_tokens": 4583449.0,
"step": 2250
},
{
"entropy": 6.583423757553101,
"epoch": 0.19706370706982435,
"grad_norm": 1.0625,
"learning_rate": 0.0004998642860812447,
"loss": 6.393,
"mean_token_accuracy": 0.12896343395113946,
"num_tokens": 4593637.0,
"step": 2255
},
{
"entropy": 6.550680303573609,
"epoch": 0.19750065542252906,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998632017853951,
"loss": 6.3727,
"mean_token_accuracy": 0.13199751377105712,
"num_tokens": 4603865.0,
"step": 2260
},
{
"entropy": 6.628088903427124,
"epoch": 0.19793760377523376,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998621131765514,
"loss": 6.4037,
"mean_token_accuracy": 0.1369698889553547,
"num_tokens": 4613502.0,
"step": 2265
},
{
"entropy": 6.536065626144409,
"epoch": 0.19837455212793847,
"grad_norm": 0.96875,
"learning_rate": 0.0004998610202547345,
"loss": 6.3718,
"mean_token_accuracy": 0.1352036103606224,
"num_tokens": 4623163.0,
"step": 2270
},
{
"entropy": 6.6453217506408695,
"epoch": 0.1988115004806432,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998599230199654,
"loss": 6.3806,
"mean_token_accuracy": 0.13569082021713258,
"num_tokens": 4632850.0,
"step": 2275
},
{
"entropy": 6.540207004547119,
"epoch": 0.19924844883334789,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998588214722653,
"loss": 6.39,
"mean_token_accuracy": 0.13032061085104943,
"num_tokens": 4642858.0,
"step": 2280
},
{
"entropy": 6.648379802703857,
"epoch": 0.1996853971860526,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998577156116551,
"loss": 6.4224,
"mean_token_accuracy": 0.12049075067043305,
"num_tokens": 4653083.0,
"step": 2285
},
{
"entropy": 6.599198913574218,
"epoch": 0.20012234553875732,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998566054381561,
"loss": 6.4329,
"mean_token_accuracy": 0.12729515209794046,
"num_tokens": 4663353.0,
"step": 2290
},
{
"entropy": 6.545134258270264,
"epoch": 0.20055929389146204,
"grad_norm": 1.2890625,
"learning_rate": 0.0004998554909517895,
"loss": 6.2739,
"mean_token_accuracy": 0.13422200456261635,
"num_tokens": 4673878.0,
"step": 2295
},
{
"entropy": 6.544426107406617,
"epoch": 0.20099624224416673,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998543721525769,
"loss": 6.3686,
"mean_token_accuracy": 0.13058247044682503,
"num_tokens": 4683661.0,
"step": 2300
},
{
"entropy": 6.487976789474487,
"epoch": 0.20143319059687145,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998532490405397,
"loss": 6.4136,
"mean_token_accuracy": 0.1305684819817543,
"num_tokens": 4693817.0,
"step": 2305
},
{
"entropy": 6.587650442123413,
"epoch": 0.20187013894957617,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998521216156991,
"loss": 6.3358,
"mean_token_accuracy": 0.12866818010807038,
"num_tokens": 4703550.0,
"step": 2310
},
{
"entropy": 6.52202959060669,
"epoch": 0.20230708730228086,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998509898780771,
"loss": 6.4504,
"mean_token_accuracy": 0.12608920335769652,
"num_tokens": 4714872.0,
"step": 2315
},
{
"entropy": 6.569240665435791,
"epoch": 0.20274403565498558,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998498538276954,
"loss": 6.3384,
"mean_token_accuracy": 0.13502655774354935,
"num_tokens": 4725745.0,
"step": 2320
},
{
"entropy": 6.534421777725219,
"epoch": 0.2031809840076903,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998487134645755,
"loss": 6.4199,
"mean_token_accuracy": 0.1276063710451126,
"num_tokens": 4735560.0,
"step": 2325
},
{
"entropy": 6.615751838684082,
"epoch": 0.203617932360395,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998475687887396,
"loss": 6.4198,
"mean_token_accuracy": 0.12353309392929077,
"num_tokens": 4746304.0,
"step": 2330
},
{
"entropy": 6.603571605682373,
"epoch": 0.2040548807130997,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998464198002095,
"loss": 6.3777,
"mean_token_accuracy": 0.1305947408080101,
"num_tokens": 4756200.0,
"step": 2335
},
{
"entropy": 6.568876934051514,
"epoch": 0.20449182906580443,
"grad_norm": 0.953125,
"learning_rate": 0.0004998452664990072,
"loss": 6.3913,
"mean_token_accuracy": 0.12563365921378136,
"num_tokens": 4766249.0,
"step": 2340
},
{
"entropy": 6.6453080654144285,
"epoch": 0.20492877741850912,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998441088851548,
"loss": 6.4154,
"mean_token_accuracy": 0.12701233252882957,
"num_tokens": 4776510.0,
"step": 2345
},
{
"entropy": 6.529261922836303,
"epoch": 0.20536572577121384,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998429469586748,
"loss": 6.383,
"mean_token_accuracy": 0.1331693261861801,
"num_tokens": 4786798.0,
"step": 2350
},
{
"entropy": 6.568332815170288,
"epoch": 0.20580267412391856,
"grad_norm": 0.91015625,
"learning_rate": 0.000499841780719589,
"loss": 6.3123,
"mean_token_accuracy": 0.12657533809542656,
"num_tokens": 4796001.0,
"step": 2355
},
{
"entropy": 6.49063138961792,
"epoch": 0.20623962247662325,
"grad_norm": 1.171875,
"learning_rate": 0.0004998406101679202,
"loss": 6.2915,
"mean_token_accuracy": 0.13698873668909073,
"num_tokens": 4807634.0,
"step": 2360
},
{
"entropy": 6.5027268409729,
"epoch": 0.20667657082932797,
"grad_norm": 0.953125,
"learning_rate": 0.0004998394353036906,
"loss": 6.3005,
"mean_token_accuracy": 0.1387121118605137,
"num_tokens": 4817840.0,
"step": 2365
},
{
"entropy": 6.520350217819214,
"epoch": 0.2071135191820327,
"grad_norm": 1.015625,
"learning_rate": 0.0004998382561269229,
"loss": 6.2813,
"mean_token_accuracy": 0.13542777374386789,
"num_tokens": 4827186.0,
"step": 2370
},
{
"entropy": 6.492212581634521,
"epoch": 0.20755046753473738,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998370726376395,
"loss": 6.3373,
"mean_token_accuracy": 0.12861326411366464,
"num_tokens": 4837140.0,
"step": 2375
},
{
"entropy": 6.579593086242676,
"epoch": 0.2079874158874421,
"grad_norm": 1.0,
"learning_rate": 0.0004998358848358634,
"loss": 6.359,
"mean_token_accuracy": 0.13045982643961906,
"num_tokens": 4846907.0,
"step": 2380
},
{
"entropy": 6.4922003746032715,
"epoch": 0.20842436424014682,
"grad_norm": 0.90234375,
"learning_rate": 0.000499834692721617,
"loss": 6.3765,
"mean_token_accuracy": 0.1315027058124542,
"num_tokens": 4856165.0,
"step": 2385
},
{
"entropy": 6.563181781768799,
"epoch": 0.2088613125928515,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998334962949237,
"loss": 6.2397,
"mean_token_accuracy": 0.14191331565380097,
"num_tokens": 4865308.0,
"step": 2390
},
{
"entropy": 6.4551702499389645,
"epoch": 0.20929826094555623,
"grad_norm": 1.140625,
"learning_rate": 0.0004998322955558059,
"loss": 6.311,
"mean_token_accuracy": 0.14102340936660768,
"num_tokens": 4876286.0,
"step": 2395
},
{
"entropy": 6.538469457626343,
"epoch": 0.20973520929826095,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998310905042869,
"loss": 6.3287,
"mean_token_accuracy": 0.13503796905279158,
"num_tokens": 4887227.0,
"step": 2400
},
{
"entropy": 6.518013572692871,
"epoch": 0.21017215765096567,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998298811403898,
"loss": 6.3721,
"mean_token_accuracy": 0.13244589641690255,
"num_tokens": 4896994.0,
"step": 2405
},
{
"entropy": 6.554981422424317,
"epoch": 0.21060910600367036,
"grad_norm": 1.0,
"learning_rate": 0.0004998286674641378,
"loss": 6.3652,
"mean_token_accuracy": 0.1293633349239826,
"num_tokens": 4907623.0,
"step": 2410
},
{
"entropy": 6.4484740734100345,
"epoch": 0.21104605435637508,
"grad_norm": 0.94921875,
"learning_rate": 0.000499827449475554,
"loss": 6.2625,
"mean_token_accuracy": 0.133323784917593,
"num_tokens": 4917040.0,
"step": 2415
},
{
"entropy": 6.612685298919677,
"epoch": 0.2114830027090798,
"grad_norm": 0.9375,
"learning_rate": 0.000499826227174662,
"loss": 6.406,
"mean_token_accuracy": 0.12849230915308,
"num_tokens": 4928390.0,
"step": 2420
},
{
"entropy": 6.526340484619141,
"epoch": 0.2119199510617845,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998250005614852,
"loss": 6.3792,
"mean_token_accuracy": 0.130333261936903,
"num_tokens": 4939040.0,
"step": 2425
},
{
"entropy": 6.447541618347168,
"epoch": 0.2123568994144892,
"grad_norm": 0.8828125,
"learning_rate": 0.000499823769636047,
"loss": 6.3954,
"mean_token_accuracy": 0.1286868393421173,
"num_tokens": 4949915.0,
"step": 2430
},
{
"entropy": 6.531896543502808,
"epoch": 0.21279384776719393,
"grad_norm": 0.890625,
"learning_rate": 0.0004998225343983712,
"loss": 6.2937,
"mean_token_accuracy": 0.13460685759782792,
"num_tokens": 4960127.0,
"step": 2435
},
{
"entropy": 6.602890491485596,
"epoch": 0.21323079611989862,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998212948484814,
"loss": 6.3519,
"mean_token_accuracy": 0.12345878928899764,
"num_tokens": 4970152.0,
"step": 2440
},
{
"entropy": 6.47441291809082,
"epoch": 0.21366774447260334,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998200509864012,
"loss": 6.3047,
"mean_token_accuracy": 0.13682056069374085,
"num_tokens": 4979745.0,
"step": 2445
},
{
"entropy": 6.590406131744385,
"epoch": 0.21410469282530806,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998188028121547,
"loss": 6.3249,
"mean_token_accuracy": 0.14112883135676385,
"num_tokens": 4989924.0,
"step": 2450
},
{
"entropy": 6.459373140335083,
"epoch": 0.21454164117801275,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998175503257658,
"loss": 6.2795,
"mean_token_accuracy": 0.1387402407824993,
"num_tokens": 4999698.0,
"step": 2455
},
{
"entropy": 6.507392549514771,
"epoch": 0.21497858953071747,
"grad_norm": 0.921875,
"learning_rate": 0.0004998162935272585,
"loss": 6.374,
"mean_token_accuracy": 0.13244377076625824,
"num_tokens": 5010625.0,
"step": 2460
},
{
"entropy": 6.44816575050354,
"epoch": 0.2154155378834222,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998150324166568,
"loss": 6.2678,
"mean_token_accuracy": 0.14076413363218307,
"num_tokens": 5019740.0,
"step": 2465
},
{
"entropy": 6.608205986022949,
"epoch": 0.21585248623612688,
"grad_norm": 0.78515625,
"learning_rate": 0.000499813766993985,
"loss": 6.3888,
"mean_token_accuracy": 0.12639941275119781,
"num_tokens": 5030601.0,
"step": 2470
},
{
"entropy": 6.60090184211731,
"epoch": 0.2162894345888316,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998124972592674,
"loss": 6.316,
"mean_token_accuracy": 0.13324218690395356,
"num_tokens": 5040567.0,
"step": 2475
},
{
"entropy": 6.433722639083863,
"epoch": 0.21672638294153632,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998112232125283,
"loss": 6.2152,
"mean_token_accuracy": 0.13335614651441574,
"num_tokens": 5049512.0,
"step": 2480
},
{
"entropy": 6.4912238121032715,
"epoch": 0.217163331294241,
"grad_norm": 0.87109375,
"learning_rate": 0.000499809944853792,
"loss": 6.4084,
"mean_token_accuracy": 0.12704317271709442,
"num_tokens": 5060986.0,
"step": 2485
},
{
"entropy": 6.486314105987549,
"epoch": 0.21760027964694573,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998086621830833,
"loss": 6.3174,
"mean_token_accuracy": 0.13398604840040207,
"num_tokens": 5071492.0,
"step": 2490
},
{
"entropy": 6.531875467300415,
"epoch": 0.21803722799965045,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998073752004267,
"loss": 6.3477,
"mean_token_accuracy": 0.1316291257739067,
"num_tokens": 5081722.0,
"step": 2495
},
{
"entropy": 6.555519723892212,
"epoch": 0.21847417635235516,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998060839058467,
"loss": 6.2966,
"mean_token_accuracy": 0.13120478838682176,
"num_tokens": 5091906.0,
"step": 2500
},
{
"entropy": 6.486591482162476,
"epoch": 0.21891112470505986,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998047882993683,
"loss": 6.2947,
"mean_token_accuracy": 0.13860670775175093,
"num_tokens": 5102025.0,
"step": 2505
},
{
"entropy": 6.444411563873291,
"epoch": 0.21934807305776458,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998034883810162,
"loss": 6.2987,
"mean_token_accuracy": 0.13405173942446708,
"num_tokens": 5112354.0,
"step": 2510
},
{
"entropy": 6.513107395172119,
"epoch": 0.2197850214104693,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998021841508155,
"loss": 6.2366,
"mean_token_accuracy": 0.14100577235221862,
"num_tokens": 5121695.0,
"step": 2515
},
{
"entropy": 6.501870727539062,
"epoch": 0.22022196976317399,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998008756087911,
"loss": 6.3494,
"mean_token_accuracy": 0.13254786729812623,
"num_tokens": 5132561.0,
"step": 2520
},
{
"entropy": 6.552183055877686,
"epoch": 0.2206589181158787,
"grad_norm": 0.92578125,
"learning_rate": 0.000499799562754968,
"loss": 6.3747,
"mean_token_accuracy": 0.12984378933906554,
"num_tokens": 5141981.0,
"step": 2525
},
{
"entropy": 6.45419282913208,
"epoch": 0.22109586646858342,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997982455893716,
"loss": 6.2503,
"mean_token_accuracy": 0.13417460918426513,
"num_tokens": 5151550.0,
"step": 2530
},
{
"entropy": 6.523354196548462,
"epoch": 0.22153281482128812,
"grad_norm": 1.046875,
"learning_rate": 0.0004997969241120271,
"loss": 6.3171,
"mean_token_accuracy": 0.12860540449619293,
"num_tokens": 5161679.0,
"step": 2535
},
{
"entropy": 6.444218873977661,
"epoch": 0.22196976317399283,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997955983229597,
"loss": 6.3836,
"mean_token_accuracy": 0.13248728290200235,
"num_tokens": 5171575.0,
"step": 2540
},
{
"entropy": 6.49551568031311,
"epoch": 0.22240671152669755,
"grad_norm": 1.0234375,
"learning_rate": 0.000499794268222195,
"loss": 6.2317,
"mean_token_accuracy": 0.13764370903372763,
"num_tokens": 5180152.0,
"step": 2545
},
{
"entropy": 6.458237552642823,
"epoch": 0.22284365987940224,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997929338097583,
"loss": 6.3363,
"mean_token_accuracy": 0.13028131797909737,
"num_tokens": 5190518.0,
"step": 2550
},
{
"entropy": 6.557815790176392,
"epoch": 0.22328060823210696,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997915950856755,
"loss": 6.3577,
"mean_token_accuracy": 0.12511950880289077,
"num_tokens": 5201729.0,
"step": 2555
},
{
"entropy": 6.497034549713135,
"epoch": 0.22371755658481168,
"grad_norm": 0.953125,
"learning_rate": 0.0004997902520499721,
"loss": 6.3337,
"mean_token_accuracy": 0.1315643459558487,
"num_tokens": 5211940.0,
"step": 2560
},
{
"entropy": 6.42897891998291,
"epoch": 0.22415450493751637,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997889047026738,
"loss": 6.2651,
"mean_token_accuracy": 0.13462026566267013,
"num_tokens": 5221028.0,
"step": 2565
},
{
"entropy": 6.471619081497193,
"epoch": 0.2245914532902211,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997875530438066,
"loss": 6.2416,
"mean_token_accuracy": 0.13898374885320663,
"num_tokens": 5230498.0,
"step": 2570
},
{
"entropy": 6.477230548858643,
"epoch": 0.2250284016429258,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997861970733963,
"loss": 6.2929,
"mean_token_accuracy": 0.13128770291805267,
"num_tokens": 5240112.0,
"step": 2575
},
{
"entropy": 6.502867269515991,
"epoch": 0.2254653499956305,
"grad_norm": 0.9140625,
"learning_rate": 0.000499784836791469,
"loss": 6.2492,
"mean_token_accuracy": 0.13940685987472534,
"num_tokens": 5249201.0,
"step": 2580
},
{
"entropy": 6.488351058959961,
"epoch": 0.22590229834833522,
"grad_norm": 0.90625,
"learning_rate": 0.0004997834721980507,
"loss": 6.2683,
"mean_token_accuracy": 0.1345277301967144,
"num_tokens": 5258541.0,
"step": 2585
},
{
"entropy": 6.57556848526001,
"epoch": 0.22633924670103994,
"grad_norm": 1.390625,
"learning_rate": 0.0004997821032931676,
"loss": 6.3723,
"mean_token_accuracy": 0.1347046300768852,
"num_tokens": 5268524.0,
"step": 2590
},
{
"entropy": 6.418022775650025,
"epoch": 0.22677619505374463,
"grad_norm": 0.83203125,
"learning_rate": 0.000499780730076846,
"loss": 6.2883,
"mean_token_accuracy": 0.13537805154919624,
"num_tokens": 5280702.0,
"step": 2595
},
{
"entropy": 6.522402095794678,
"epoch": 0.22721314340644935,
"grad_norm": 0.90625,
"learning_rate": 0.0004997793525491122,
"loss": 6.2628,
"mean_token_accuracy": 0.1282507784664631,
"num_tokens": 5291214.0,
"step": 2600
},
{
"entropy": 6.40453028678894,
"epoch": 0.22765009175915407,
"grad_norm": 1.0,
"learning_rate": 0.0004997779707099926,
"loss": 6.2218,
"mean_token_accuracy": 0.14659756794571877,
"num_tokens": 5300729.0,
"step": 2605
},
{
"entropy": 6.47467451095581,
"epoch": 0.2280870401118588,
"grad_norm": 0.875,
"learning_rate": 0.0004997765845595138,
"loss": 6.3621,
"mean_token_accuracy": 0.13119488582015038,
"num_tokens": 5310551.0,
"step": 2610
},
{
"entropy": 6.483618068695068,
"epoch": 0.22852398846456348,
"grad_norm": 0.96875,
"learning_rate": 0.0004997751940977024,
"loss": 6.2198,
"mean_token_accuracy": 0.13728965371847152,
"num_tokens": 5320038.0,
"step": 2615
},
{
"entropy": 6.453027200698853,
"epoch": 0.2289609368172682,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997737993245848,
"loss": 6.2905,
"mean_token_accuracy": 0.1325733795762062,
"num_tokens": 5329753.0,
"step": 2620
},
{
"entropy": 6.422470378875732,
"epoch": 0.22939788516997292,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997724002401881,
"loss": 6.2727,
"mean_token_accuracy": 0.13691200613975524,
"num_tokens": 5340546.0,
"step": 2625
},
{
"entropy": 6.40334906578064,
"epoch": 0.2298348335226776,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997709968445391,
"loss": 6.3444,
"mean_token_accuracy": 0.13128848671913146,
"num_tokens": 5352097.0,
"step": 2630
},
{
"entropy": 6.561565923690796,
"epoch": 0.23027178187538233,
"grad_norm": 0.9375,
"learning_rate": 0.0004997695891376644,
"loss": 6.2735,
"mean_token_accuracy": 0.13817958086729049,
"num_tokens": 5361646.0,
"step": 2635
},
{
"entropy": 6.418984413146973,
"epoch": 0.23070873022808705,
"grad_norm": 1.0625,
"learning_rate": 0.0004997681771195912,
"loss": 6.2515,
"mean_token_accuracy": 0.14123460948467254,
"num_tokens": 5371695.0,
"step": 2640
},
{
"entropy": 6.445744562149048,
"epoch": 0.23114567858079174,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997667607903466,
"loss": 6.3225,
"mean_token_accuracy": 0.13134846240282058,
"num_tokens": 5381978.0,
"step": 2645
},
{
"entropy": 6.554618072509766,
"epoch": 0.23158262693349646,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997653401499578,
"loss": 6.3257,
"mean_token_accuracy": 0.1353774681687355,
"num_tokens": 5391891.0,
"step": 2650
},
{
"entropy": 6.483434438705444,
"epoch": 0.23201957528620118,
"grad_norm": 0.94140625,
"learning_rate": 0.000499763915198452,
"loss": 6.2978,
"mean_token_accuracy": 0.13159872964024544,
"num_tokens": 5403172.0,
"step": 2655
},
{
"entropy": 6.40730848312378,
"epoch": 0.23245652363890587,
"grad_norm": 0.90625,
"learning_rate": 0.0004997624859358565,
"loss": 6.2971,
"mean_token_accuracy": 0.13805141896009446,
"num_tokens": 5413130.0,
"step": 2660
},
{
"entropy": 6.454382991790771,
"epoch": 0.2328934719916106,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997610523621988,
"loss": 6.332,
"mean_token_accuracy": 0.13280672580003738,
"num_tokens": 5423510.0,
"step": 2665
},
{
"entropy": 6.487873840332031,
"epoch": 0.2333304203443153,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997596144775063,
"loss": 6.1581,
"mean_token_accuracy": 0.14178016260266305,
"num_tokens": 5433707.0,
"step": 2670
},
{
"entropy": 6.299017810821534,
"epoch": 0.23376736869702,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997581722818067,
"loss": 6.2586,
"mean_token_accuracy": 0.137728713452816,
"num_tokens": 5444381.0,
"step": 2675
},
{
"entropy": 6.607342195510864,
"epoch": 0.23420431704972472,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997567257751275,
"loss": 6.2167,
"mean_token_accuracy": 0.13631812259554862,
"num_tokens": 5453905.0,
"step": 2680
},
{
"entropy": 6.388966417312622,
"epoch": 0.23464126540242944,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997552749574965,
"loss": 6.2393,
"mean_token_accuracy": 0.13831414729356767,
"num_tokens": 5464839.0,
"step": 2685
},
{
"entropy": 6.426655292510986,
"epoch": 0.23507821375513413,
"grad_norm": 0.96875,
"learning_rate": 0.0004997538198289416,
"loss": 6.2768,
"mean_token_accuracy": 0.12901907563209533,
"num_tokens": 5475459.0,
"step": 2690
},
{
"entropy": 6.431175565719604,
"epoch": 0.23551516210783885,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997523603894907,
"loss": 6.285,
"mean_token_accuracy": 0.1379141092300415,
"num_tokens": 5484886.0,
"step": 2695
},
{
"entropy": 6.4651857852935795,
"epoch": 0.23595211046054357,
"grad_norm": 0.9375,
"learning_rate": 0.0004997508966391717,
"loss": 6.271,
"mean_token_accuracy": 0.13728889897465707,
"num_tokens": 5494803.0,
"step": 2700
},
{
"entropy": 6.40422306060791,
"epoch": 0.2363890588132483,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997494285780129,
"loss": 6.2203,
"mean_token_accuracy": 0.13441247269511222,
"num_tokens": 5504997.0,
"step": 2705
},
{
"entropy": 6.3673491954803465,
"epoch": 0.23682600716595298,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997479562060421,
"loss": 6.2201,
"mean_token_accuracy": 0.14114237129688262,
"num_tokens": 5514492.0,
"step": 2710
},
{
"entropy": 6.491820907592773,
"epoch": 0.2372629555186577,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997464795232877,
"loss": 6.2568,
"mean_token_accuracy": 0.13033155351877213,
"num_tokens": 5524419.0,
"step": 2715
},
{
"entropy": 6.461743450164795,
"epoch": 0.23769990387136242,
"grad_norm": 0.90625,
"learning_rate": 0.0004997449985297782,
"loss": 6.282,
"mean_token_accuracy": 0.13783859759569167,
"num_tokens": 5534289.0,
"step": 2720
},
{
"entropy": 6.40246434211731,
"epoch": 0.2381368522240671,
"grad_norm": 1.078125,
"learning_rate": 0.0004997435132255418,
"loss": 6.2304,
"mean_token_accuracy": 0.13647418320178986,
"num_tokens": 5543034.0,
"step": 2725
},
{
"entropy": 6.480632209777832,
"epoch": 0.23857380057677183,
"grad_norm": 0.921875,
"learning_rate": 0.0004997420236106071,
"loss": 6.2934,
"mean_token_accuracy": 0.1379154346883297,
"num_tokens": 5553291.0,
"step": 2730
},
{
"entropy": 6.460249185562134,
"epoch": 0.23901074892947655,
"grad_norm": 0.890625,
"learning_rate": 0.0004997405296850026,
"loss": 6.2744,
"mean_token_accuracy": 0.13468145504593848,
"num_tokens": 5563772.0,
"step": 2735
},
{
"entropy": 6.447478199005127,
"epoch": 0.23944769728218124,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997390314487569,
"loss": 6.3068,
"mean_token_accuracy": 0.13267375007271767,
"num_tokens": 5574348.0,
"step": 2740
},
{
"entropy": 6.4441228866577145,
"epoch": 0.23988464563488596,
"grad_norm": 0.90625,
"learning_rate": 0.0004997375289018989,
"loss": 6.1637,
"mean_token_accuracy": 0.13476863205432893,
"num_tokens": 5583491.0,
"step": 2745
},
{
"entropy": 6.34928069114685,
"epoch": 0.24032159398759068,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997360220444574,
"loss": 6.324,
"mean_token_accuracy": 0.12789161056280135,
"num_tokens": 5594273.0,
"step": 2750
},
{
"entropy": 6.478325939178466,
"epoch": 0.24075854234029537,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997345108764611,
"loss": 6.2209,
"mean_token_accuracy": 0.13871316835284234,
"num_tokens": 5604529.0,
"step": 2755
},
{
"entropy": 6.397528219223022,
"epoch": 0.2411954906930001,
"grad_norm": 0.921875,
"learning_rate": 0.0004997329953979391,
"loss": 6.2102,
"mean_token_accuracy": 0.14513446986675263,
"num_tokens": 5614036.0,
"step": 2760
},
{
"entropy": 6.458429145812988,
"epoch": 0.2416324390457048,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997314756089206,
"loss": 6.2591,
"mean_token_accuracy": 0.13734069988131523,
"num_tokens": 5624036.0,
"step": 2765
},
{
"entropy": 6.424458265304565,
"epoch": 0.2420693873984095,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997299515094347,
"loss": 6.2456,
"mean_token_accuracy": 0.1391109459102154,
"num_tokens": 5633987.0,
"step": 2770
},
{
"entropy": 6.383187675476075,
"epoch": 0.24250633575111422,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997284230995105,
"loss": 6.1199,
"mean_token_accuracy": 0.14982096254825591,
"num_tokens": 5643905.0,
"step": 2775
},
{
"entropy": 6.436825084686279,
"epoch": 0.24294328410381894,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997268903791773,
"loss": 6.3289,
"mean_token_accuracy": 0.1315071016550064,
"num_tokens": 5654922.0,
"step": 2780
},
{
"entropy": 6.39974889755249,
"epoch": 0.24338023245652363,
"grad_norm": 0.8515625,
"learning_rate": 0.0004997253533484647,
"loss": 6.1586,
"mean_token_accuracy": 0.1425160974264145,
"num_tokens": 5664407.0,
"step": 2785
},
{
"entropy": 6.417563819885254,
"epoch": 0.24381718080922835,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997238120074021,
"loss": 6.2272,
"mean_token_accuracy": 0.13456432819366454,
"num_tokens": 5674524.0,
"step": 2790
},
{
"entropy": 6.452868366241455,
"epoch": 0.24425412916193306,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997222663560188,
"loss": 6.2615,
"mean_token_accuracy": 0.1321194626390934,
"num_tokens": 5684691.0,
"step": 2795
},
{
"entropy": 6.3742493152618405,
"epoch": 0.24469107751463778,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997207163943449,
"loss": 6.1785,
"mean_token_accuracy": 0.13999853581190108,
"num_tokens": 5693683.0,
"step": 2800
},
{
"entropy": 6.406622314453125,
"epoch": 0.24512802586734248,
"grad_norm": 0.890625,
"learning_rate": 0.0004997191621224098,
"loss": 6.2671,
"mean_token_accuracy": 0.13562537357211113,
"num_tokens": 5704394.0,
"step": 2805
},
{
"entropy": 6.394683933258056,
"epoch": 0.2455649742200472,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997176035402435,
"loss": 6.1658,
"mean_token_accuracy": 0.1439206138253212,
"num_tokens": 5715480.0,
"step": 2810
},
{
"entropy": 6.465519857406616,
"epoch": 0.2460019225727519,
"grad_norm": 0.890625,
"learning_rate": 0.0004997160406478758,
"loss": 6.1749,
"mean_token_accuracy": 0.14369347989559172,
"num_tokens": 5725973.0,
"step": 2815
},
{
"entropy": 6.26147494316101,
"epoch": 0.2464388709254566,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997144734453367,
"loss": 6.2044,
"mean_token_accuracy": 0.1463428631424904,
"num_tokens": 5735798.0,
"step": 2820
},
{
"entropy": 6.491905450820923,
"epoch": 0.24687581927816132,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997129019326562,
"loss": 6.2764,
"mean_token_accuracy": 0.12938766777515412,
"num_tokens": 5746623.0,
"step": 2825
},
{
"entropy": 6.469847059249878,
"epoch": 0.24731276763086604,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997113261098645,
"loss": 6.2206,
"mean_token_accuracy": 0.14192579537630082,
"num_tokens": 5756526.0,
"step": 2830
},
{
"entropy": 6.384304094314575,
"epoch": 0.24774971598357073,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997097459769919,
"loss": 6.2083,
"mean_token_accuracy": 0.13960244581103326,
"num_tokens": 5766670.0,
"step": 2835
},
{
"entropy": 6.459981393814087,
"epoch": 0.24818666433627545,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997081615340685,
"loss": 6.2636,
"mean_token_accuracy": 0.13670481294393538,
"num_tokens": 5776473.0,
"step": 2840
},
{
"entropy": 6.503982400894165,
"epoch": 0.24862361268898017,
"grad_norm": 0.953125,
"learning_rate": 0.0004997065727811249,
"loss": 6.2949,
"mean_token_accuracy": 0.13849538564682007,
"num_tokens": 5787059.0,
"step": 2845
},
{
"entropy": 6.414534044265747,
"epoch": 0.24906056104168486,
"grad_norm": 0.86328125,
"learning_rate": 0.0004997049797181915,
"loss": 6.2884,
"mean_token_accuracy": 0.1313557967543602,
"num_tokens": 5797367.0,
"step": 2850
},
{
"entropy": 6.3486549854278564,
"epoch": 0.24949750939438958,
"grad_norm": 0.984375,
"learning_rate": 0.0004997033823452988,
"loss": 6.1751,
"mean_token_accuracy": 0.13914304301142694,
"num_tokens": 5806704.0,
"step": 2855
},
{
"entropy": 6.40527114868164,
"epoch": 0.2499344577470943,
"grad_norm": 0.828125,
"learning_rate": 0.0004997017806624774,
"loss": 6.2468,
"mean_token_accuracy": 0.1393037997186184,
"num_tokens": 5817797.0,
"step": 2860
},
{
"entropy": 6.466370344161987,
"epoch": 0.250371406099799,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997001746697582,
"loss": 6.2764,
"mean_token_accuracy": 0.14098547995090485,
"num_tokens": 5829270.0,
"step": 2865
},
{
"entropy": 6.405104541778565,
"epoch": 0.2508083544525037,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996985643671719,
"loss": 6.2277,
"mean_token_accuracy": 0.1410501480102539,
"num_tokens": 5839214.0,
"step": 2870
},
{
"entropy": 6.361758279800415,
"epoch": 0.2512453028052084,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996969497547495,
"loss": 6.2165,
"mean_token_accuracy": 0.13901832550764084,
"num_tokens": 5848171.0,
"step": 2875
},
{
"entropy": 6.408371257781982,
"epoch": 0.25168225115791315,
"grad_norm": 0.875,
"learning_rate": 0.0004996953308325217,
"loss": 6.1341,
"mean_token_accuracy": 0.14583304151892662,
"num_tokens": 5858399.0,
"step": 2880
},
{
"entropy": 6.343305540084839,
"epoch": 0.25211919951061784,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996937076005198,
"loss": 6.1379,
"mean_token_accuracy": 0.14553209990262986,
"num_tokens": 5868212.0,
"step": 2885
},
{
"entropy": 6.380623054504395,
"epoch": 0.25255614786332253,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996920800587749,
"loss": 6.207,
"mean_token_accuracy": 0.13883035629987717,
"num_tokens": 5877883.0,
"step": 2890
},
{
"entropy": 6.433513402938843,
"epoch": 0.2529930962160273,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996904482073181,
"loss": 6.2114,
"mean_token_accuracy": 0.14031296521425246,
"num_tokens": 5887090.0,
"step": 2895
},
{
"entropy": 6.32197265625,
"epoch": 0.25343004456873197,
"grad_norm": 0.9375,
"learning_rate": 0.0004996888120461808,
"loss": 6.2558,
"mean_token_accuracy": 0.1309097707271576,
"num_tokens": 5897491.0,
"step": 2900
},
{
"entropy": 6.420869779586792,
"epoch": 0.25386699292143666,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996871715753943,
"loss": 6.2715,
"mean_token_accuracy": 0.13663778081536293,
"num_tokens": 5908494.0,
"step": 2905
},
{
"entropy": 6.50685076713562,
"epoch": 0.2543039412741414,
"grad_norm": 0.875,
"learning_rate": 0.00049968552679499,
"loss": 6.3246,
"mean_token_accuracy": 0.12492352053523063,
"num_tokens": 5919708.0,
"step": 2910
},
{
"entropy": 6.412569522857666,
"epoch": 0.2547408896268461,
"grad_norm": 0.984375,
"learning_rate": 0.0004996838777049998,
"loss": 6.1817,
"mean_token_accuracy": 0.1387775108218193,
"num_tokens": 5930122.0,
"step": 2915
},
{
"entropy": 6.335930967330933,
"epoch": 0.2551778379795508,
"grad_norm": 0.93359375,
"learning_rate": 0.000499682224305455,
"loss": 6.1921,
"mean_token_accuracy": 0.1372269608080387,
"num_tokens": 5940033.0,
"step": 2920
},
{
"entropy": 6.3563535690307615,
"epoch": 0.25561478633225554,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996805665963874,
"loss": 6.1798,
"mean_token_accuracy": 0.13515137359499932,
"num_tokens": 5949335.0,
"step": 2925
},
{
"entropy": 6.468244028091431,
"epoch": 0.25605173468496023,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996789045778287,
"loss": 6.2149,
"mean_token_accuracy": 0.1392026238143444,
"num_tokens": 5959299.0,
"step": 2930
},
{
"entropy": 6.408330249786377,
"epoch": 0.2564886830376649,
"grad_norm": 0.83984375,
"learning_rate": 0.0004996772382498111,
"loss": 6.1833,
"mean_token_accuracy": 0.13642103523015975,
"num_tokens": 5970114.0,
"step": 2935
},
{
"entropy": 6.343743133544922,
"epoch": 0.25692563139036967,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996755676123662,
"loss": 6.281,
"mean_token_accuracy": 0.13536796644330024,
"num_tokens": 5979630.0,
"step": 2940
},
{
"entropy": 6.489995050430298,
"epoch": 0.25736257974307436,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996738926655262,
"loss": 6.2612,
"mean_token_accuracy": 0.13175804391503335,
"num_tokens": 5990190.0,
"step": 2945
},
{
"entropy": 6.450542163848877,
"epoch": 0.25779952809577905,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996722134093231,
"loss": 6.2849,
"mean_token_accuracy": 0.1331914894282818,
"num_tokens": 6000706.0,
"step": 2950
},
{
"entropy": 6.346250867843628,
"epoch": 0.2582364764484838,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996705298437894,
"loss": 6.2405,
"mean_token_accuracy": 0.13917525857686996,
"num_tokens": 6011274.0,
"step": 2955
},
{
"entropy": 6.465360450744629,
"epoch": 0.2586734248011885,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996688419689572,
"loss": 6.1539,
"mean_token_accuracy": 0.148239216953516,
"num_tokens": 6021481.0,
"step": 2960
},
{
"entropy": 6.345787000656128,
"epoch": 0.25911037315389324,
"grad_norm": 0.953125,
"learning_rate": 0.0004996671497848589,
"loss": 6.2279,
"mean_token_accuracy": 0.13937210515141488,
"num_tokens": 6032091.0,
"step": 2965
},
{
"entropy": 6.386220216751099,
"epoch": 0.25954732150659793,
"grad_norm": 0.890625,
"learning_rate": 0.0004996654532915269,
"loss": 6.2324,
"mean_token_accuracy": 0.13179687783122063,
"num_tokens": 6043371.0,
"step": 2970
},
{
"entropy": 6.471800041198731,
"epoch": 0.2599842698593026,
"grad_norm": 0.953125,
"learning_rate": 0.0004996637524889937,
"loss": 6.1863,
"mean_token_accuracy": 0.13913432955741883,
"num_tokens": 6052812.0,
"step": 2975
},
{
"entropy": 6.313665294647217,
"epoch": 0.26042121821200737,
"grad_norm": 0.828125,
"learning_rate": 0.0004996620473772921,
"loss": 6.1687,
"mean_token_accuracy": 0.14615661054849624,
"num_tokens": 6064033.0,
"step": 2980
},
{
"entropy": 6.398783588409424,
"epoch": 0.26085816656471206,
"grad_norm": 0.84375,
"learning_rate": 0.0004996603379564546,
"loss": 6.3042,
"mean_token_accuracy": 0.13552941903471946,
"num_tokens": 6075837.0,
"step": 2985
},
{
"entropy": 6.529281282424927,
"epoch": 0.26129511491741675,
"grad_norm": 0.859375,
"learning_rate": 0.0004996586242265142,
"loss": 6.1871,
"mean_token_accuracy": 0.14371934831142424,
"num_tokens": 6085514.0,
"step": 2990
},
{
"entropy": 6.355500984191894,
"epoch": 0.2617320632701215,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996569061875036,
"loss": 6.1057,
"mean_token_accuracy": 0.14477267414331435,
"num_tokens": 6094992.0,
"step": 2995
},
{
"entropy": 6.338720464706421,
"epoch": 0.2621690116228262,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996551838394559,
"loss": 6.1787,
"mean_token_accuracy": 0.13958787322044372,
"num_tokens": 6104841.0,
"step": 3000
},
{
"epoch": 0.2621690116228262,
"eval_entropy": 6.140728340206636,
"eval_loss": 6.207319259643555,
"eval_mean_token_accuracy": 0.1469711780999768,
"eval_num_tokens": 6104841.0,
"eval_runtime": 21.6443,
"eval_samples_per_second": 1710.102,
"eval_steps_per_second": 213.774,
"step": 3000
},
{
"entropy": 6.2954404830932615,
"epoch": 0.2626059599755309,
"grad_norm": 0.953125,
"learning_rate": 0.000499653457182404,
"loss": 6.0721,
"mean_token_accuracy": 0.14866723716259003,
"num_tokens": 6114830.0,
"step": 3005
},
{
"entropy": 6.330233383178711,
"epoch": 0.2630429083282356,
"grad_norm": 0.890625,
"learning_rate": 0.0004996517262163811,
"loss": 6.2002,
"mean_token_accuracy": 0.1417041540145874,
"num_tokens": 6125165.0,
"step": 3010
},
{
"entropy": 6.360240745544433,
"epoch": 0.2634798566809403,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996499909414204,
"loss": 6.1454,
"mean_token_accuracy": 0.14034123420715333,
"num_tokens": 6135670.0,
"step": 3015
},
{
"entropy": 6.297566843032837,
"epoch": 0.263916805033645,
"grad_norm": 0.87109375,
"learning_rate": 0.0004996482513575551,
"loss": 6.1294,
"mean_token_accuracy": 0.14982950687408447,
"num_tokens": 6146055.0,
"step": 3020
},
{
"entropy": 6.299193429946899,
"epoch": 0.26435375338634975,
"grad_norm": 1.0,
"learning_rate": 0.0004996465074648186,
"loss": 6.1479,
"mean_token_accuracy": 0.14521702080965043,
"num_tokens": 6155941.0,
"step": 3025
},
{
"entropy": 6.452114486694336,
"epoch": 0.26479070173905445,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996447592632444,
"loss": 6.2433,
"mean_token_accuracy": 0.1405558854341507,
"num_tokens": 6166291.0,
"step": 3030
},
{
"entropy": 6.352116060256958,
"epoch": 0.26522765009175914,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996430067528661,
"loss": 6.1996,
"mean_token_accuracy": 0.13908485621213912,
"num_tokens": 6175294.0,
"step": 3035
},
{
"entropy": 6.441894578933716,
"epoch": 0.2656645984444639,
"grad_norm": 0.875,
"learning_rate": 0.0004996412499337172,
"loss": 6.1799,
"mean_token_accuracy": 0.14599951803684236,
"num_tokens": 6185273.0,
"step": 3040
},
{
"entropy": 6.258168268203735,
"epoch": 0.2661015467971686,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996394888058314,
"loss": 6.1415,
"mean_token_accuracy": 0.1480218470096588,
"num_tokens": 6195075.0,
"step": 3045
},
{
"entropy": 6.221267890930176,
"epoch": 0.26653849514987327,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996377233692425,
"loss": 6.0302,
"mean_token_accuracy": 0.15916952788829802,
"num_tokens": 6206018.0,
"step": 3050
},
{
"entropy": 6.406425666809082,
"epoch": 0.266975443502578,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996359536239843,
"loss": 6.1703,
"mean_token_accuracy": 0.13967156410217285,
"num_tokens": 6216577.0,
"step": 3055
},
{
"entropy": 6.416021108627319,
"epoch": 0.2674123918552827,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996341795700908,
"loss": 6.1967,
"mean_token_accuracy": 0.14080278426408768,
"num_tokens": 6226521.0,
"step": 3060
},
{
"entropy": 6.317079877853393,
"epoch": 0.2678493402079874,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996324012075961,
"loss": 6.1758,
"mean_token_accuracy": 0.14496348276734353,
"num_tokens": 6236509.0,
"step": 3065
},
{
"entropy": 6.361893129348755,
"epoch": 0.26828628856069214,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996306185365341,
"loss": 6.161,
"mean_token_accuracy": 0.14295196756720543,
"num_tokens": 6246584.0,
"step": 3070
},
{
"entropy": 6.38335657119751,
"epoch": 0.26872323691339683,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996288315569393,
"loss": 6.184,
"mean_token_accuracy": 0.1432151563465595,
"num_tokens": 6256881.0,
"step": 3075
},
{
"entropy": 6.398205709457398,
"epoch": 0.2691601852661015,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996270402688457,
"loss": 6.146,
"mean_token_accuracy": 0.14504260420799256,
"num_tokens": 6266380.0,
"step": 3080
},
{
"entropy": 6.373401737213134,
"epoch": 0.2695971336188063,
"grad_norm": 1.0,
"learning_rate": 0.0004996252446722877,
"loss": 6.1964,
"mean_token_accuracy": 0.1460059978067875,
"num_tokens": 6274872.0,
"step": 3085
},
{
"entropy": 6.350259590148926,
"epoch": 0.27003408197151096,
"grad_norm": 0.890625,
"learning_rate": 0.0004996234447673,
"loss": 6.2075,
"mean_token_accuracy": 0.14173548221588134,
"num_tokens": 6286167.0,
"step": 3090
},
{
"entropy": 6.369588041305542,
"epoch": 0.27047103032421566,
"grad_norm": 0.890625,
"learning_rate": 0.0004996216405539169,
"loss": 6.156,
"mean_token_accuracy": 0.1371704325079918,
"num_tokens": 6295898.0,
"step": 3095
},
{
"entropy": 6.381024503707886,
"epoch": 0.2709079786769204,
"grad_norm": 0.890625,
"learning_rate": 0.0004996198320321728,
"loss": 6.1818,
"mean_token_accuracy": 0.14304329603910446,
"num_tokens": 6306543.0,
"step": 3100
},
{
"entropy": 6.351856899261475,
"epoch": 0.2713449270296251,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996180192021029,
"loss": 6.1697,
"mean_token_accuracy": 0.13672457933425902,
"num_tokens": 6317709.0,
"step": 3105
},
{
"entropy": 6.351310873031617,
"epoch": 0.2717818753823298,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996162020637415,
"loss": 6.1198,
"mean_token_accuracy": 0.14543213993310927,
"num_tokens": 6326874.0,
"step": 3110
},
{
"entropy": 6.359648990631103,
"epoch": 0.27221882373503453,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996143806171238,
"loss": 6.2203,
"mean_token_accuracy": 0.13802286610007286,
"num_tokens": 6337943.0,
"step": 3115
},
{
"entropy": 6.3505360126495365,
"epoch": 0.2726557720877392,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996125548622845,
"loss": 6.2814,
"mean_token_accuracy": 0.1384304866194725,
"num_tokens": 6348442.0,
"step": 3120
},
{
"entropy": 6.348107242584229,
"epoch": 0.2730927204404439,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996107247992587,
"loss": 6.1768,
"mean_token_accuracy": 0.14265852496027948,
"num_tokens": 6358675.0,
"step": 3125
},
{
"entropy": 6.382030820846557,
"epoch": 0.27352966879314866,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996088904280815,
"loss": 6.2168,
"mean_token_accuracy": 0.13871825635433196,
"num_tokens": 6368631.0,
"step": 3130
},
{
"entropy": 6.371310091018676,
"epoch": 0.27396661714585335,
"grad_norm": 0.890625,
"learning_rate": 0.0004996070517487881,
"loss": 6.0773,
"mean_token_accuracy": 0.14500803351402283,
"num_tokens": 6378709.0,
"step": 3135
},
{
"entropy": 6.318156099319458,
"epoch": 0.27440356549855804,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996052087614138,
"loss": 6.2322,
"mean_token_accuracy": 0.13356006070971488,
"num_tokens": 6389686.0,
"step": 3140
},
{
"entropy": 6.3309718608856205,
"epoch": 0.2748405138512628,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996033614659938,
"loss": 6.1964,
"mean_token_accuracy": 0.13752286732196808,
"num_tokens": 6399447.0,
"step": 3145
},
{
"entropy": 6.345420885086059,
"epoch": 0.2752774622039675,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996015098625638,
"loss": 6.1715,
"mean_token_accuracy": 0.14288139641284942,
"num_tokens": 6408891.0,
"step": 3150
},
{
"entropy": 6.360024499893188,
"epoch": 0.2757144105566722,
"grad_norm": 0.90625,
"learning_rate": 0.000499599653951159,
"loss": 6.144,
"mean_token_accuracy": 0.1456197828054428,
"num_tokens": 6419471.0,
"step": 3155
},
{
"entropy": 6.319950771331787,
"epoch": 0.2761513589093769,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995977937318153,
"loss": 6.1237,
"mean_token_accuracy": 0.14010287821292877,
"num_tokens": 6430445.0,
"step": 3160
},
{
"entropy": 6.3929389953613285,
"epoch": 0.2765883072620816,
"grad_norm": 0.91015625,
"learning_rate": 0.000499595929204568,
"loss": 6.1615,
"mean_token_accuracy": 0.14165643975138664,
"num_tokens": 6440821.0,
"step": 3165
},
{
"entropy": 6.299115943908691,
"epoch": 0.27702525561478636,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995940603694533,
"loss": 6.0938,
"mean_token_accuracy": 0.14386325478553771,
"num_tokens": 6450958.0,
"step": 3170
},
{
"entropy": 6.301024341583252,
"epoch": 0.27746220396749105,
"grad_norm": 0.875,
"learning_rate": 0.0004995921872265067,
"loss": 6.1834,
"mean_token_accuracy": 0.14242049753665925,
"num_tokens": 6462106.0,
"step": 3175
},
{
"entropy": 6.328290796279907,
"epoch": 0.27789915232019574,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995903097757643,
"loss": 6.0674,
"mean_token_accuracy": 0.15150198489427566,
"num_tokens": 6471890.0,
"step": 3180
},
{
"entropy": 6.29628005027771,
"epoch": 0.2783361006729005,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995884280172622,
"loss": 6.0887,
"mean_token_accuracy": 0.1439910836517811,
"num_tokens": 6481959.0,
"step": 3185
},
{
"entropy": 6.357741498947144,
"epoch": 0.2787730490256052,
"grad_norm": 1.0,
"learning_rate": 0.0004995865419510362,
"loss": 6.1264,
"mean_token_accuracy": 0.14735689684748648,
"num_tokens": 6492086.0,
"step": 3190
},
{
"entropy": 6.274436140060425,
"epoch": 0.27920999737830987,
"grad_norm": 0.984375,
"learning_rate": 0.0004995846515771227,
"loss": 6.16,
"mean_token_accuracy": 0.14143534004688263,
"num_tokens": 6502530.0,
"step": 3195
},
{
"entropy": 6.367669486999512,
"epoch": 0.2796469457310146,
"grad_norm": 0.93359375,
"learning_rate": 0.000499582756895558,
"loss": 6.161,
"mean_token_accuracy": 0.14963102713227272,
"num_tokens": 6511600.0,
"step": 3200
},
{
"entropy": 6.2996594429016115,
"epoch": 0.2800838940837193,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995808579063782,
"loss": 6.1345,
"mean_token_accuracy": 0.13933146744966507,
"num_tokens": 6522712.0,
"step": 3205
},
{
"entropy": 6.15234899520874,
"epoch": 0.280520842436424,
"grad_norm": 0.94140625,
"learning_rate": 0.00049957895460962,
"loss": 6.0024,
"mean_token_accuracy": 0.1613503262400627,
"num_tokens": 6532472.0,
"step": 3210
},
{
"entropy": 6.329534006118775,
"epoch": 0.28095779078912875,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995770470053197,
"loss": 6.0545,
"mean_token_accuracy": 0.15718746110796927,
"num_tokens": 6541778.0,
"step": 3215
},
{
"entropy": 6.323871088027954,
"epoch": 0.28139473914183344,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995751350935139,
"loss": 6.0655,
"mean_token_accuracy": 0.14491002410650253,
"num_tokens": 6552368.0,
"step": 3220
},
{
"entropy": 6.281246185302734,
"epoch": 0.28183168749453813,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995732188742395,
"loss": 6.0917,
"mean_token_accuracy": 0.14650973826646804,
"num_tokens": 6562593.0,
"step": 3225
},
{
"entropy": 6.327645111083984,
"epoch": 0.2822686358472429,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995712983475329,
"loss": 6.1427,
"mean_token_accuracy": 0.14498995393514633,
"num_tokens": 6572695.0,
"step": 3230
},
{
"entropy": 6.225468492507934,
"epoch": 0.28270558419994757,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995693735134313,
"loss": 6.1144,
"mean_token_accuracy": 0.1426160916686058,
"num_tokens": 6581671.0,
"step": 3235
},
{
"entropy": 6.335493373870849,
"epoch": 0.28314253255265226,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995674443719713,
"loss": 6.1229,
"mean_token_accuracy": 0.14055858105421065,
"num_tokens": 6590460.0,
"step": 3240
},
{
"entropy": 6.303600740432739,
"epoch": 0.283579480905357,
"grad_norm": 0.96484375,
"learning_rate": 0.00049956551092319,
"loss": 6.1244,
"mean_token_accuracy": 0.1433852382004261,
"num_tokens": 6599892.0,
"step": 3245
},
{
"entropy": 6.354545307159424,
"epoch": 0.2840164292580617,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995635731671247,
"loss": 6.2012,
"mean_token_accuracy": 0.13369778245687486,
"num_tokens": 6610530.0,
"step": 3250
},
{
"entropy": 6.2983636379241945,
"epoch": 0.2844533776107664,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995616311038124,
"loss": 6.1237,
"mean_token_accuracy": 0.14690847247838973,
"num_tokens": 6621117.0,
"step": 3255
},
{
"entropy": 6.379088354110718,
"epoch": 0.28489032596347114,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995596847332901,
"loss": 6.1595,
"mean_token_accuracy": 0.14022180885076524,
"num_tokens": 6631521.0,
"step": 3260
},
{
"entropy": 6.423089265823364,
"epoch": 0.2853272743161758,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995577340555956,
"loss": 6.247,
"mean_token_accuracy": 0.13523154854774475,
"num_tokens": 6642180.0,
"step": 3265
},
{
"entropy": 6.334853219985962,
"epoch": 0.2857642226688805,
"grad_norm": 0.85546875,
"learning_rate": 0.0004995557790707661,
"loss": 6.1713,
"mean_token_accuracy": 0.1466689735651016,
"num_tokens": 6652735.0,
"step": 3270
},
{
"entropy": 6.3520190715789795,
"epoch": 0.28620117102158527,
"grad_norm": 0.9140625,
"learning_rate": 0.000499553819778839,
"loss": 6.16,
"mean_token_accuracy": 0.14624010920524597,
"num_tokens": 6663135.0,
"step": 3275
},
{
"entropy": 6.2731767177581785,
"epoch": 0.28663811937428996,
"grad_norm": 0.9765625,
"learning_rate": 0.000499551856179852,
"loss": 6.1412,
"mean_token_accuracy": 0.1431369364261627,
"num_tokens": 6673375.0,
"step": 3280
},
{
"entropy": 6.350570297241211,
"epoch": 0.28707506772699465,
"grad_norm": 0.875,
"learning_rate": 0.0004995498882738428,
"loss": 6.1158,
"mean_token_accuracy": 0.14772647768259048,
"num_tokens": 6683101.0,
"step": 3285
},
{
"entropy": 6.329085493087769,
"epoch": 0.2875120160796994,
"grad_norm": 0.96875,
"learning_rate": 0.0004995479160608491,
"loss": 6.173,
"mean_token_accuracy": 0.14756054282188416,
"num_tokens": 6692763.0,
"step": 3290
},
{
"entropy": 6.351175737380982,
"epoch": 0.2879489644324041,
"grad_norm": 1.0,
"learning_rate": 0.0004995459395409085,
"loss": 6.155,
"mean_token_accuracy": 0.1399838872253895,
"num_tokens": 6702833.0,
"step": 3295
},
{
"entropy": 6.3768007278442385,
"epoch": 0.2883859127851088,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995439587140593,
"loss": 6.1495,
"mean_token_accuracy": 0.13922087401151656,
"num_tokens": 6713083.0,
"step": 3300
},
{
"entropy": 6.256941509246826,
"epoch": 0.2888228611378135,
"grad_norm": 0.953125,
"learning_rate": 0.0004995419735803392,
"loss": 6.0994,
"mean_token_accuracy": 0.14441604316234588,
"num_tokens": 6722501.0,
"step": 3305
},
{
"entropy": 6.345928049087524,
"epoch": 0.2892598094905182,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995399841397865,
"loss": 6.1313,
"mean_token_accuracy": 0.1453465297818184,
"num_tokens": 6731961.0,
"step": 3310
},
{
"entropy": 6.358537864685059,
"epoch": 0.2896967578432229,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995379903924392,
"loss": 6.1214,
"mean_token_accuracy": 0.1412414386868477,
"num_tokens": 6742576.0,
"step": 3315
},
{
"entropy": 6.313660049438477,
"epoch": 0.29013370619592765,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995359923383356,
"loss": 6.1247,
"mean_token_accuracy": 0.15020241886377333,
"num_tokens": 6753066.0,
"step": 3320
},
{
"entropy": 6.3176545143127445,
"epoch": 0.29057065454863235,
"grad_norm": 0.94140625,
"learning_rate": 0.000499533989977514,
"loss": 6.2618,
"mean_token_accuracy": 0.1356256164610386,
"num_tokens": 6764455.0,
"step": 3325
},
{
"entropy": 6.375495719909668,
"epoch": 0.29100760290133704,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995319833100128,
"loss": 6.1293,
"mean_token_accuracy": 0.1450370132923126,
"num_tokens": 6774655.0,
"step": 3330
},
{
"entropy": 6.2986647605896,
"epoch": 0.2914445512540418,
"grad_norm": 0.87890625,
"learning_rate": 0.0004995299723358705,
"loss": 6.1021,
"mean_token_accuracy": 0.15024321302771568,
"num_tokens": 6785942.0,
"step": 3335
},
{
"entropy": 6.265730381011963,
"epoch": 0.2918814996067465,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995279570551257,
"loss": 6.0624,
"mean_token_accuracy": 0.1497375935316086,
"num_tokens": 6795355.0,
"step": 3340
},
{
"entropy": 6.296811914443969,
"epoch": 0.29231844795945117,
"grad_norm": 0.9453125,
"learning_rate": 0.000499525937467817,
"loss": 6.1472,
"mean_token_accuracy": 0.14276159554719925,
"num_tokens": 6805579.0,
"step": 3345
},
{
"entropy": 6.325125026702881,
"epoch": 0.2927553963121559,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995239135739833,
"loss": 6.1636,
"mean_token_accuracy": 0.14122274965047837,
"num_tokens": 6814878.0,
"step": 3350
},
{
"entropy": 6.344474601745605,
"epoch": 0.2931923446648606,
"grad_norm": 0.96875,
"learning_rate": 0.0004995218853736631,
"loss": 6.055,
"mean_token_accuracy": 0.15251176059246063,
"num_tokens": 6824533.0,
"step": 3355
},
{
"entropy": 6.125179576873779,
"epoch": 0.29362929301756535,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995198528668956,
"loss": 5.9683,
"mean_token_accuracy": 0.159256049990654,
"num_tokens": 6835529.0,
"step": 3360
},
{
"entropy": 6.3220854759216305,
"epoch": 0.29406624137027004,
"grad_norm": 0.90625,
"learning_rate": 0.0004995178160537197,
"loss": 6.1517,
"mean_token_accuracy": 0.14569269567728044,
"num_tokens": 6845679.0,
"step": 3365
},
{
"entropy": 6.354899740219116,
"epoch": 0.29450318972297473,
"grad_norm": 0.921875,
"learning_rate": 0.0004995157749341744,
"loss": 6.1334,
"mean_token_accuracy": 0.13885997533798217,
"num_tokens": 6855876.0,
"step": 3370
},
{
"entropy": 6.297861623764038,
"epoch": 0.2949401380756795,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995137295082989,
"loss": 6.2052,
"mean_token_accuracy": 0.1401164412498474,
"num_tokens": 6866200.0,
"step": 3375
},
{
"entropy": 6.359179210662842,
"epoch": 0.2953770864283842,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995116797761324,
"loss": 6.0778,
"mean_token_accuracy": 0.14605300948023797,
"num_tokens": 6875191.0,
"step": 3380
},
{
"entropy": 6.273428821563721,
"epoch": 0.29581403478108886,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995096257377143,
"loss": 6.0599,
"mean_token_accuracy": 0.14512585699558259,
"num_tokens": 6885352.0,
"step": 3385
},
{
"entropy": 6.30303840637207,
"epoch": 0.2962509831337936,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995075673930839,
"loss": 6.1311,
"mean_token_accuracy": 0.14580143094062806,
"num_tokens": 6895585.0,
"step": 3390
},
{
"entropy": 6.382786750793457,
"epoch": 0.2966879314864983,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995055047422808,
"loss": 6.1014,
"mean_token_accuracy": 0.14060337394475936,
"num_tokens": 6905532.0,
"step": 3395
},
{
"entropy": 6.158539295196533,
"epoch": 0.297124879839203,
"grad_norm": 0.84375,
"learning_rate": 0.0004995034377853443,
"loss": 6.1029,
"mean_token_accuracy": 0.14624154269695283,
"num_tokens": 6915835.0,
"step": 3400
},
{
"entropy": 6.280689668655396,
"epoch": 0.29756182819190774,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995013665223143,
"loss": 6.1525,
"mean_token_accuracy": 0.14352616518735886,
"num_tokens": 6925964.0,
"step": 3405
},
{
"entropy": 6.314380741119384,
"epoch": 0.29799877654461243,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994992909532305,
"loss": 6.091,
"mean_token_accuracy": 0.14589581415057182,
"num_tokens": 6936145.0,
"step": 3410
},
{
"entropy": 6.29896092414856,
"epoch": 0.2984357248973171,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994972110781326,
"loss": 6.0656,
"mean_token_accuracy": 0.14432404488325118,
"num_tokens": 6945557.0,
"step": 3415
},
{
"entropy": 6.251769351959228,
"epoch": 0.29887267325002187,
"grad_norm": 0.984375,
"learning_rate": 0.0004994951268970606,
"loss": 6.091,
"mean_token_accuracy": 0.1475279785692692,
"num_tokens": 6955147.0,
"step": 3420
},
{
"entropy": 6.284211874008179,
"epoch": 0.29930962160272656,
"grad_norm": 0.984375,
"learning_rate": 0.0004994930384100545,
"loss": 6.059,
"mean_token_accuracy": 0.14195767417550087,
"num_tokens": 6964024.0,
"step": 3425
},
{
"entropy": 6.276056480407715,
"epoch": 0.29974656995543125,
"grad_norm": 0.90234375,
"learning_rate": 0.0004994909456171541,
"loss": 6.1159,
"mean_token_accuracy": 0.1480628401041031,
"num_tokens": 6974399.0,
"step": 3430
},
{
"entropy": 6.257720851898194,
"epoch": 0.300183518308136,
"grad_norm": 0.88671875,
"learning_rate": 0.0004994888485183998,
"loss": 6.0676,
"mean_token_accuracy": 0.14755637869238852,
"num_tokens": 6984324.0,
"step": 3435
},
{
"entropy": 6.278302526473999,
"epoch": 0.3006204666608407,
"grad_norm": 0.984375,
"learning_rate": 0.000499486747113832,
"loss": 6.0759,
"mean_token_accuracy": 0.15024591833353043,
"num_tokens": 6995659.0,
"step": 3440
},
{
"entropy": 6.34593620300293,
"epoch": 0.3010574150135454,
"grad_norm": 0.9375,
"learning_rate": 0.0004994846414034905,
"loss": 6.1757,
"mean_token_accuracy": 0.13790618106722832,
"num_tokens": 7006463.0,
"step": 3445
},
{
"entropy": 6.315510845184326,
"epoch": 0.30149436336625013,
"grad_norm": 0.97265625,
"learning_rate": 0.000499482531387416,
"loss": 6.0778,
"mean_token_accuracy": 0.1473211646080017,
"num_tokens": 7016188.0,
"step": 3450
},
{
"entropy": 6.284571266174316,
"epoch": 0.3019313117189548,
"grad_norm": 0.890625,
"learning_rate": 0.0004994804170656489,
"loss": 6.1437,
"mean_token_accuracy": 0.14328745901584625,
"num_tokens": 7028435.0,
"step": 3455
},
{
"entropy": 6.376664781570435,
"epoch": 0.3023682600716595,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994782984382299,
"loss": 6.1132,
"mean_token_accuracy": 0.14584433510899544,
"num_tokens": 7037992.0,
"step": 3460
},
{
"entropy": 6.252527952194214,
"epoch": 0.30280520842436426,
"grad_norm": 0.8671875,
"learning_rate": 0.0004994761755051995,
"loss": 6.1248,
"mean_token_accuracy": 0.14423885941505432,
"num_tokens": 7048743.0,
"step": 3465
},
{
"entropy": 6.2551322937011715,
"epoch": 0.30324215677706895,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994740482665982,
"loss": 6.0851,
"mean_token_accuracy": 0.14877668246626854,
"num_tokens": 7059524.0,
"step": 3470
},
{
"entropy": 6.315831565856934,
"epoch": 0.30367910512977364,
"grad_norm": 0.90625,
"learning_rate": 0.0004994719167224672,
"loss": 6.0897,
"mean_token_accuracy": 0.1395334392786026,
"num_tokens": 7069742.0,
"step": 3475
},
{
"entropy": 6.26900954246521,
"epoch": 0.3041160534824784,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994697808728472,
"loss": 6.1423,
"mean_token_accuracy": 0.1453932225704193,
"num_tokens": 7079909.0,
"step": 3480
},
{
"entropy": 6.348114776611328,
"epoch": 0.3045530018351831,
"grad_norm": 0.984375,
"learning_rate": 0.0004994676407177791,
"loss": 6.1526,
"mean_token_accuracy": 0.14140928015112877,
"num_tokens": 7089136.0,
"step": 3485
},
{
"entropy": 6.288559532165527,
"epoch": 0.30498995018788777,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994654962573043,
"loss": 6.1544,
"mean_token_accuracy": 0.14411381259560585,
"num_tokens": 7099621.0,
"step": 3490
},
{
"entropy": 6.291727304458618,
"epoch": 0.3054268985405925,
"grad_norm": 0.890625,
"learning_rate": 0.0004994633474914634,
"loss": 6.0123,
"mean_token_accuracy": 0.15279096812009813,
"num_tokens": 7109599.0,
"step": 3495
},
{
"entropy": 6.285858392715454,
"epoch": 0.3058638468932972,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994611944202979,
"loss": 6.0568,
"mean_token_accuracy": 0.15079958140850067,
"num_tokens": 7120733.0,
"step": 3500
},
{
"entropy": 6.276545238494873,
"epoch": 0.3063007952460019,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994590370438492,
"loss": 6.136,
"mean_token_accuracy": 0.1393118381500244,
"num_tokens": 7129871.0,
"step": 3505
},
{
"entropy": 6.256414079666138,
"epoch": 0.30673774359870665,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994568753621585,
"loss": 6.0527,
"mean_token_accuracy": 0.14701383113861083,
"num_tokens": 7139167.0,
"step": 3510
},
{
"entropy": 6.227274465560913,
"epoch": 0.30717469195141134,
"grad_norm": 0.859375,
"learning_rate": 0.0004994547093752673,
"loss": 6.1219,
"mean_token_accuracy": 0.13770676925778388,
"num_tokens": 7150534.0,
"step": 3515
},
{
"entropy": 6.320562839508057,
"epoch": 0.30761164030411603,
"grad_norm": 0.921875,
"learning_rate": 0.0004994525390832171,
"loss": 6.0774,
"mean_token_accuracy": 0.14651177376508712,
"num_tokens": 7160472.0,
"step": 3520
},
{
"entropy": 6.295433425903321,
"epoch": 0.3080485886568208,
"grad_norm": 0.90234375,
"learning_rate": 0.0004994503644860497,
"loss": 6.103,
"mean_token_accuracy": 0.1437145009636879,
"num_tokens": 7171636.0,
"step": 3525
},
{
"entropy": 6.27422308921814,
"epoch": 0.30848553700952547,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994481855838067,
"loss": 6.0513,
"mean_token_accuracy": 0.14762744009494783,
"num_tokens": 7182163.0,
"step": 3530
},
{
"entropy": 6.206560754776001,
"epoch": 0.30892248536223016,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994460023765298,
"loss": 6.0918,
"mean_token_accuracy": 0.14782901704311371,
"num_tokens": 7191694.0,
"step": 3535
},
{
"entropy": 6.269797325134277,
"epoch": 0.3093594337149349,
"grad_norm": 0.75390625,
"learning_rate": 0.000499443814864261,
"loss": 6.1028,
"mean_token_accuracy": 0.14703719019889833,
"num_tokens": 7202813.0,
"step": 3540
},
{
"entropy": 6.328323841094971,
"epoch": 0.3097963820676396,
"grad_norm": 0.890625,
"learning_rate": 0.0004994416230470422,
"loss": 6.0943,
"mean_token_accuracy": 0.1497291296720505,
"num_tokens": 7213226.0,
"step": 3545
},
{
"entropy": 6.286290502548217,
"epoch": 0.3102333304203443,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994394269249155,
"loss": 6.1446,
"mean_token_accuracy": 0.13995824977755547,
"num_tokens": 7223520.0,
"step": 3550
},
{
"entropy": 6.266398096084595,
"epoch": 0.31067027877304904,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994372264979231,
"loss": 6.0356,
"mean_token_accuracy": 0.15077984631061553,
"num_tokens": 7232651.0,
"step": 3555
},
{
"entropy": 6.315041875839233,
"epoch": 0.3111072271257537,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994350217661069,
"loss": 6.1171,
"mean_token_accuracy": 0.14419228360056877,
"num_tokens": 7242064.0,
"step": 3560
},
{
"entropy": 6.287273693084717,
"epoch": 0.3115441754784585,
"grad_norm": 0.875,
"learning_rate": 0.0004994328127295095,
"loss": 6.136,
"mean_token_accuracy": 0.1410224989056587,
"num_tokens": 7253048.0,
"step": 3565
},
{
"entropy": 6.305558681488037,
"epoch": 0.31198112383116317,
"grad_norm": 1.03125,
"learning_rate": 0.0004994305993881731,
"loss": 6.079,
"mean_token_accuracy": 0.14993986338377,
"num_tokens": 7262516.0,
"step": 3570
},
{
"entropy": 6.22655839920044,
"epoch": 0.31241807218386786,
"grad_norm": 0.859375,
"learning_rate": 0.0004994283817421402,
"loss": 6.0066,
"mean_token_accuracy": 0.15258399695158004,
"num_tokens": 7272079.0,
"step": 3575
},
{
"entropy": 6.220291090011597,
"epoch": 0.3128550205365726,
"grad_norm": 0.84765625,
"learning_rate": 0.0004994261597914534,
"loss": 5.9878,
"mean_token_accuracy": 0.15598128288984298,
"num_tokens": 7282150.0,
"step": 3580
},
{
"entropy": 6.26607666015625,
"epoch": 0.3132919688892773,
"grad_norm": 0.890625,
"learning_rate": 0.0004994239335361552,
"loss": 6.0999,
"mean_token_accuracy": 0.14056497439742088,
"num_tokens": 7291782.0,
"step": 3585
},
{
"entropy": 6.293278360366822,
"epoch": 0.313728917241982,
"grad_norm": 0.90625,
"learning_rate": 0.0004994217029762884,
"loss": 6.0599,
"mean_token_accuracy": 0.14566777795553207,
"num_tokens": 7301472.0,
"step": 3590
},
{
"entropy": 6.261723375320434,
"epoch": 0.31416586559468673,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994194681118957,
"loss": 6.0321,
"mean_token_accuracy": 0.1512453258037567,
"num_tokens": 7311456.0,
"step": 3595
},
{
"entropy": 6.295667886734009,
"epoch": 0.3146028139473914,
"grad_norm": 0.9765625,
"learning_rate": 0.00049941722894302,
"loss": 6.1232,
"mean_token_accuracy": 0.14058786481618882,
"num_tokens": 7321106.0,
"step": 3600
},
{
"entropy": 6.266748237609863,
"epoch": 0.3150397623000961,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994149854697042,
"loss": 6.1111,
"mean_token_accuracy": 0.14695683643221855,
"num_tokens": 7331232.0,
"step": 3605
},
{
"entropy": 6.314934539794922,
"epoch": 0.31547671065280086,
"grad_norm": 0.82421875,
"learning_rate": 0.0004994127376919916,
"loss": 6.0965,
"mean_token_accuracy": 0.14339785426855087,
"num_tokens": 7341560.0,
"step": 3610
},
{
"entropy": 6.326571130752564,
"epoch": 0.31591365900550555,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994104856099249,
"loss": 6.0868,
"mean_token_accuracy": 0.14662182778120042,
"num_tokens": 7350923.0,
"step": 3615
},
{
"entropy": 6.197239208221435,
"epoch": 0.31635060735821025,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994082292235476,
"loss": 5.9859,
"mean_token_accuracy": 0.15532282441854478,
"num_tokens": 7360121.0,
"step": 3620
},
{
"entropy": 6.209159564971924,
"epoch": 0.316787555710915,
"grad_norm": 0.890625,
"learning_rate": 0.0004994059685329027,
"loss": 6.0856,
"mean_token_accuracy": 0.14521137177944182,
"num_tokens": 7370823.0,
"step": 3625
},
{
"entropy": 6.269903326034546,
"epoch": 0.3172245040636197,
"grad_norm": 0.8671875,
"learning_rate": 0.000499403703538034,
"loss": 6.0541,
"mean_token_accuracy": 0.1409251481294632,
"num_tokens": 7380102.0,
"step": 3630
},
{
"entropy": 6.256821346282959,
"epoch": 0.3176614524163244,
"grad_norm": 0.921875,
"learning_rate": 0.0004994014342389845,
"loss": 6.0329,
"mean_token_accuracy": 0.14819468557834625,
"num_tokens": 7389909.0,
"step": 3635
},
{
"entropy": 6.277596092224121,
"epoch": 0.3180984007690291,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993991606357979,
"loss": 6.1144,
"mean_token_accuracy": 0.1460643008351326,
"num_tokens": 7401092.0,
"step": 3640
},
{
"entropy": 6.189544677734375,
"epoch": 0.3185353491217338,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993968827285179,
"loss": 6.026,
"mean_token_accuracy": 0.14508207887411118,
"num_tokens": 7412155.0,
"step": 3645
},
{
"entropy": 6.2149989128112795,
"epoch": 0.3189722974744385,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993946005171881,
"loss": 5.9921,
"mean_token_accuracy": 0.153763547539711,
"num_tokens": 7421981.0,
"step": 3650
},
{
"entropy": 6.267820215225219,
"epoch": 0.31940924582714325,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993923140018523,
"loss": 6.0067,
"mean_token_accuracy": 0.15336575210094452,
"num_tokens": 7432298.0,
"step": 3655
},
{
"entropy": 6.185194301605224,
"epoch": 0.31984619417984794,
"grad_norm": 0.87890625,
"learning_rate": 0.0004993900231825542,
"loss": 6.0497,
"mean_token_accuracy": 0.15190931037068367,
"num_tokens": 7442575.0,
"step": 3660
},
{
"entropy": 6.248135757446289,
"epoch": 0.32028314253255263,
"grad_norm": 0.84375,
"learning_rate": 0.0004993877280593381,
"loss": 6.0386,
"mean_token_accuracy": 0.15646653175354003,
"num_tokens": 7453072.0,
"step": 3665
},
{
"entropy": 6.271560096740723,
"epoch": 0.3207200908852574,
"grad_norm": 0.90625,
"learning_rate": 0.0004993854286322477,
"loss": 5.9975,
"mean_token_accuracy": 0.1494731992483139,
"num_tokens": 7463173.0,
"step": 3670
},
{
"entropy": 6.215144538879395,
"epoch": 0.32115703923796207,
"grad_norm": 1.078125,
"learning_rate": 0.0004993831249013272,
"loss": 6.1129,
"mean_token_accuracy": 0.1444126322865486,
"num_tokens": 7473751.0,
"step": 3675
},
{
"entropy": 6.231976127624511,
"epoch": 0.32159398759066676,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993808168666207,
"loss": 5.9103,
"mean_token_accuracy": 0.15974051058292388,
"num_tokens": 7483233.0,
"step": 3680
},
{
"entropy": 6.229567003250122,
"epoch": 0.3220309359433715,
"grad_norm": 0.84765625,
"learning_rate": 0.0004993785045281726,
"loss": 6.0487,
"mean_token_accuracy": 0.14229854047298432,
"num_tokens": 7493152.0,
"step": 3685
},
{
"entropy": 6.293207740783691,
"epoch": 0.3224678842960762,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993761878860272,
"loss": 6.008,
"mean_token_accuracy": 0.15029292106628417,
"num_tokens": 7502463.0,
"step": 3690
},
{
"entropy": 6.283467102050781,
"epoch": 0.3229048326487809,
"grad_norm": 0.91796875,
"learning_rate": 0.000499373866940229,
"loss": 6.0655,
"mean_token_accuracy": 0.15074712336063384,
"num_tokens": 7512301.0,
"step": 3695
},
{
"entropy": 6.14256911277771,
"epoch": 0.32334178100148564,
"grad_norm": 0.9609375,
"learning_rate": 0.0004993715416908225,
"loss": 6.0948,
"mean_token_accuracy": 0.14658709317445756,
"num_tokens": 7522591.0,
"step": 3700
},
{
"entropy": 6.322413969039917,
"epoch": 0.32377872935419033,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993692121378521,
"loss": 6.0879,
"mean_token_accuracy": 0.1442805364727974,
"num_tokens": 7532400.0,
"step": 3705
},
{
"entropy": 6.263008069992066,
"epoch": 0.324215677706895,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993668782813629,
"loss": 6.0059,
"mean_token_accuracy": 0.14632104933261872,
"num_tokens": 7542163.0,
"step": 3710
},
{
"entropy": 6.202136135101318,
"epoch": 0.32465262605959977,
"grad_norm": 0.93359375,
"learning_rate": 0.0004993645401213992,
"loss": 5.9957,
"mean_token_accuracy": 0.1582685887813568,
"num_tokens": 7551638.0,
"step": 3715
},
{
"entropy": 6.236076211929321,
"epoch": 0.32508957441230446,
"grad_norm": 0.93359375,
"learning_rate": 0.0004993621976580061,
"loss": 6.0695,
"mean_token_accuracy": 0.1474739357829094,
"num_tokens": 7561230.0,
"step": 3720
},
{
"entropy": 6.2406653881073,
"epoch": 0.32552652276500915,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993598508912285,
"loss": 5.9482,
"mean_token_accuracy": 0.15776676088571548,
"num_tokens": 7570962.0,
"step": 3725
},
{
"entropy": 6.198595762252808,
"epoch": 0.3259634711177139,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993574998211114,
"loss": 6.13,
"mean_token_accuracy": 0.1489112690091133,
"num_tokens": 7581454.0,
"step": 3730
},
{
"entropy": 6.233174753189087,
"epoch": 0.3264004194704186,
"grad_norm": 0.90625,
"learning_rate": 0.0004993551444476998,
"loss": 6.001,
"mean_token_accuracy": 0.15145405381917953,
"num_tokens": 7591497.0,
"step": 3735
},
{
"entropy": 6.271373987197876,
"epoch": 0.3268373678231233,
"grad_norm": 0.953125,
"learning_rate": 0.0004993527847710392,
"loss": 6.008,
"mean_token_accuracy": 0.15311285704374314,
"num_tokens": 7600719.0,
"step": 3740
},
{
"entropy": 6.20383505821228,
"epoch": 0.32727431617582803,
"grad_norm": 0.87109375,
"learning_rate": 0.0004993504207911744,
"loss": 6.04,
"mean_token_accuracy": 0.15261640399694443,
"num_tokens": 7610550.0,
"step": 3745
},
{
"entropy": 6.275401020050049,
"epoch": 0.3277112645285327,
"grad_norm": 0.8515625,
"learning_rate": 0.0004993480525081511,
"loss": 6.0498,
"mean_token_accuracy": 0.15051608979701997,
"num_tokens": 7620472.0,
"step": 3750
},
{
"entropy": 6.2765693187713625,
"epoch": 0.3281482128812374,
"grad_norm": 0.8984375,
"learning_rate": 0.0004993456799220145,
"loss": 5.9904,
"mean_token_accuracy": 0.146634604036808,
"num_tokens": 7630354.0,
"step": 3755
},
{
"entropy": 6.198559522628784,
"epoch": 0.32858516123394216,
"grad_norm": 0.921875,
"learning_rate": 0.0004993433030328103,
"loss": 6.0309,
"mean_token_accuracy": 0.14934386909008027,
"num_tokens": 7639689.0,
"step": 3760
},
{
"entropy": 6.244138765335083,
"epoch": 0.32902210958664685,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993409218405839,
"loss": 6.0687,
"mean_token_accuracy": 0.1417417049407959,
"num_tokens": 7649319.0,
"step": 3765
},
{
"entropy": 6.326660203933716,
"epoch": 0.3294590579393516,
"grad_norm": 0.87890625,
"learning_rate": 0.0004993385363453812,
"loss": 6.0627,
"mean_token_accuracy": 0.1466643713414669,
"num_tokens": 7659008.0,
"step": 3770
},
{
"entropy": 6.22724666595459,
"epoch": 0.3298960062920563,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993361465472477,
"loss": 6.0084,
"mean_token_accuracy": 0.15965658724308013,
"num_tokens": 7669064.0,
"step": 3775
},
{
"entropy": 6.2213051319122314,
"epoch": 0.330332954644761,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993337524462295,
"loss": 6.0794,
"mean_token_accuracy": 0.1521337777376175,
"num_tokens": 7679122.0,
"step": 3780
},
{
"entropy": 6.1721875190734865,
"epoch": 0.3307699029974657,
"grad_norm": 0.9765625,
"learning_rate": 0.0004993313540423722,
"loss": 6.0814,
"mean_token_accuracy": 0.14944267868995667,
"num_tokens": 7688373.0,
"step": 3785
},
{
"entropy": 6.261203241348267,
"epoch": 0.3312068513501704,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993289513357221,
"loss": 6.0095,
"mean_token_accuracy": 0.15269952714443208,
"num_tokens": 7697879.0,
"step": 3790
},
{
"entropy": 6.238248348236084,
"epoch": 0.3316437997028751,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993265443263251,
"loss": 6.0476,
"mean_token_accuracy": 0.14502775445580482,
"num_tokens": 7708104.0,
"step": 3795
},
{
"entropy": 6.258405303955078,
"epoch": 0.33208074805557986,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993241330142274,
"loss": 6.0196,
"mean_token_accuracy": 0.15770744234323503,
"num_tokens": 7717626.0,
"step": 3800
},
{
"entropy": 6.172434234619141,
"epoch": 0.33251769640828455,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993217173994754,
"loss": 6.0529,
"mean_token_accuracy": 0.15278787910938263,
"num_tokens": 7728221.0,
"step": 3805
},
{
"entropy": 6.2589140892028805,
"epoch": 0.33295464476098924,
"grad_norm": 0.828125,
"learning_rate": 0.0004993192974821154,
"loss": 6.0592,
"mean_token_accuracy": 0.15035932809114455,
"num_tokens": 7739083.0,
"step": 3810
},
{
"entropy": 6.2666344165802,
"epoch": 0.333391593113694,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993168732621935,
"loss": 6.0896,
"mean_token_accuracy": 0.144187543541193,
"num_tokens": 7748573.0,
"step": 3815
},
{
"entropy": 6.239906644821167,
"epoch": 0.3338285414663987,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993144447397565,
"loss": 6.036,
"mean_token_accuracy": 0.15224629193544387,
"num_tokens": 7758741.0,
"step": 3820
},
{
"entropy": 6.211666822433472,
"epoch": 0.33426548981910337,
"grad_norm": 1.0234375,
"learning_rate": 0.000499312011914851,
"loss": 6.0719,
"mean_token_accuracy": 0.14373058900237085,
"num_tokens": 7768504.0,
"step": 3825
},
{
"entropy": 6.250473356246948,
"epoch": 0.3347024381718081,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993095747875234,
"loss": 6.1943,
"mean_token_accuracy": 0.14370191246271133,
"num_tokens": 7779017.0,
"step": 3830
},
{
"entropy": 6.270117139816284,
"epoch": 0.3351393865245128,
"grad_norm": 0.9375,
"learning_rate": 0.0004993071333578207,
"loss": 5.9947,
"mean_token_accuracy": 0.15620245635509492,
"num_tokens": 7787630.0,
"step": 3835
},
{
"entropy": 6.185167360305786,
"epoch": 0.3355763348772175,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993046876257898,
"loss": 6.0063,
"mean_token_accuracy": 0.1563827708363533,
"num_tokens": 7797105.0,
"step": 3840
},
{
"entropy": 6.144869089126587,
"epoch": 0.33601328322992224,
"grad_norm": 0.85546875,
"learning_rate": 0.0004993022375914773,
"loss": 6.0018,
"mean_token_accuracy": 0.15615751594305038,
"num_tokens": 7807070.0,
"step": 3845
},
{
"entropy": 6.301814270019531,
"epoch": 0.33645023158262694,
"grad_norm": 0.921875,
"learning_rate": 0.0004992997832549303,
"loss": 6.0356,
"mean_token_accuracy": 0.1477530762553215,
"num_tokens": 7818090.0,
"step": 3850
},
{
"entropy": 6.200486707687378,
"epoch": 0.3368871799353316,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992973246161961,
"loss": 5.9792,
"mean_token_accuracy": 0.1545109361410141,
"num_tokens": 7827562.0,
"step": 3855
},
{
"entropy": 6.198225593566894,
"epoch": 0.3373241282880364,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992948616753215,
"loss": 6.038,
"mean_token_accuracy": 0.14645904898643494,
"num_tokens": 7837581.0,
"step": 3860
},
{
"entropy": 6.237083196640015,
"epoch": 0.33776107664074106,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992923944323541,
"loss": 6.0914,
"mean_token_accuracy": 0.14635520726442336,
"num_tokens": 7848729.0,
"step": 3865
},
{
"entropy": 6.258805942535401,
"epoch": 0.33819802499344576,
"grad_norm": 1.046875,
"learning_rate": 0.000499289922887341,
"loss": 6.0141,
"mean_token_accuracy": 0.15104980170726776,
"num_tokens": 7857379.0,
"step": 3870
},
{
"entropy": 6.239930486679077,
"epoch": 0.3386349733461505,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992874470403296,
"loss": 5.9866,
"mean_token_accuracy": 0.15481197834014893,
"num_tokens": 7867200.0,
"step": 3875
},
{
"entropy": 6.176450443267822,
"epoch": 0.3390719216988552,
"grad_norm": 0.96875,
"learning_rate": 0.0004992849668913674,
"loss": 5.9995,
"mean_token_accuracy": 0.1490779235959053,
"num_tokens": 7876709.0,
"step": 3880
},
{
"entropy": 6.228991889953614,
"epoch": 0.3395088700515599,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992824824405021,
"loss": 6.0927,
"mean_token_accuracy": 0.14840275645256043,
"num_tokens": 7886256.0,
"step": 3885
},
{
"entropy": 6.205230760574341,
"epoch": 0.33994581840426463,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992799936877813,
"loss": 6.029,
"mean_token_accuracy": 0.1480804704129696,
"num_tokens": 7896497.0,
"step": 3890
},
{
"entropy": 6.236300754547119,
"epoch": 0.3403827667569693,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992775006332526,
"loss": 6.0244,
"mean_token_accuracy": 0.14987484216690064,
"num_tokens": 7906608.0,
"step": 3895
},
{
"entropy": 6.279505777359009,
"epoch": 0.340819715109674,
"grad_norm": 0.96875,
"learning_rate": 0.000499275003276964,
"loss": 6.1049,
"mean_token_accuracy": 0.1473986364901066,
"num_tokens": 7916359.0,
"step": 3900
},
{
"entropy": 6.21343584060669,
"epoch": 0.34125666346237876,
"grad_norm": 0.9765625,
"learning_rate": 0.0004992725016189632,
"loss": 5.9918,
"mean_token_accuracy": 0.15183644890785217,
"num_tokens": 7925800.0,
"step": 3905
},
{
"entropy": 6.26988959312439,
"epoch": 0.34169361181508345,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992699956592984,
"loss": 6.0665,
"mean_token_accuracy": 0.14467344135046006,
"num_tokens": 7935424.0,
"step": 3910
},
{
"entropy": 6.236717939376831,
"epoch": 0.34213056016778814,
"grad_norm": 0.8046875,
"learning_rate": 0.0004992674853980175,
"loss": 6.0533,
"mean_token_accuracy": 0.14863845109939575,
"num_tokens": 7945766.0,
"step": 3915
},
{
"entropy": 6.269533777236939,
"epoch": 0.3425675085204929,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992649708351689,
"loss": 6.0645,
"mean_token_accuracy": 0.14135592579841613,
"num_tokens": 7955622.0,
"step": 3920
},
{
"entropy": 6.248694849014282,
"epoch": 0.3430044568731976,
"grad_norm": 0.8671875,
"learning_rate": 0.0004992624519708004,
"loss": 6.127,
"mean_token_accuracy": 0.14197314158082008,
"num_tokens": 7965750.0,
"step": 3925
},
{
"entropy": 6.266123914718628,
"epoch": 0.3434414052259023,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992599288049607,
"loss": 6.0537,
"mean_token_accuracy": 0.15157223641872405,
"num_tokens": 7977568.0,
"step": 3930
},
{
"entropy": 6.255645704269409,
"epoch": 0.343878353578607,
"grad_norm": 0.96875,
"learning_rate": 0.000499257401337698,
"loss": 6.1047,
"mean_token_accuracy": 0.14600171893835068,
"num_tokens": 7988411.0,
"step": 3935
},
{
"entropy": 6.17456579208374,
"epoch": 0.3443153019313117,
"grad_norm": 0.93359375,
"learning_rate": 0.0004992548695690608,
"loss": 5.9695,
"mean_token_accuracy": 0.1559397757053375,
"num_tokens": 7998315.0,
"step": 3940
},
{
"entropy": 6.207716941833496,
"epoch": 0.3447522502840164,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992523334990977,
"loss": 6.0146,
"mean_token_accuracy": 0.14814926981925963,
"num_tokens": 8007133.0,
"step": 3945
},
{
"entropy": 6.179518938064575,
"epoch": 0.34518919863672115,
"grad_norm": 0.8828125,
"learning_rate": 0.0004992497931278574,
"loss": 5.9608,
"mean_token_accuracy": 0.1595747470855713,
"num_tokens": 8016882.0,
"step": 3950
},
{
"entropy": 6.179951095581055,
"epoch": 0.34562614698942584,
"grad_norm": 0.85546875,
"learning_rate": 0.0004992472484553886,
"loss": 5.9728,
"mean_token_accuracy": 0.15136277675628662,
"num_tokens": 8026893.0,
"step": 3955
},
{
"entropy": 6.2377852439880375,
"epoch": 0.34606309534213053,
"grad_norm": 0.95703125,
"learning_rate": 0.00049924469948174,
"loss": 6.0043,
"mean_token_accuracy": 0.15436646193265915,
"num_tokens": 8036240.0,
"step": 3960
},
{
"entropy": 6.265040302276612,
"epoch": 0.3465000436948353,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992421462069606,
"loss": 5.9976,
"mean_token_accuracy": 0.14397231116890907,
"num_tokens": 8046086.0,
"step": 3965
},
{
"entropy": 6.084886646270752,
"epoch": 0.34693699204753997,
"grad_norm": 0.87890625,
"learning_rate": 0.0004992395886310993,
"loss": 5.9995,
"mean_token_accuracy": 0.1564622104167938,
"num_tokens": 8056294.0,
"step": 3970
},
{
"entropy": 6.250084590911865,
"epoch": 0.3473739404002447,
"grad_norm": 0.90625,
"learning_rate": 0.0004992370267542052,
"loss": 6.0348,
"mean_token_accuracy": 0.1465321108698845,
"num_tokens": 8067329.0,
"step": 3975
},
{
"entropy": 6.234533977508545,
"epoch": 0.3478108887529494,
"grad_norm": 0.9375,
"learning_rate": 0.0004992344605763275,
"loss": 5.99,
"mean_token_accuracy": 0.15098554342985154,
"num_tokens": 8077106.0,
"step": 3980
},
{
"entropy": 6.200424528121948,
"epoch": 0.3482478371056541,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992318900975153,
"loss": 5.9058,
"mean_token_accuracy": 0.15573831498622895,
"num_tokens": 8087766.0,
"step": 3985
},
{
"entropy": 6.1102227687835695,
"epoch": 0.34868478545835885,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992293153178179,
"loss": 6.012,
"mean_token_accuracy": 0.1529572382569313,
"num_tokens": 8098649.0,
"step": 3990
},
{
"entropy": 6.219414758682251,
"epoch": 0.34912173381106354,
"grad_norm": 0.859375,
"learning_rate": 0.0004992267362372848,
"loss": 6.0228,
"mean_token_accuracy": 0.14785912334918977,
"num_tokens": 8109237.0,
"step": 3995
},
{
"entropy": 6.186442136764526,
"epoch": 0.34955868216376823,
"grad_norm": 0.8828125,
"learning_rate": 0.0004992241528559653,
"loss": 6.0131,
"mean_token_accuracy": 0.15060167163610458,
"num_tokens": 8119731.0,
"step": 4000
},
{
"entropy": 6.268817949295044,
"epoch": 0.349995630516473,
"grad_norm": 0.91796875,
"learning_rate": 0.0004992215651739092,
"loss": 6.0347,
"mean_token_accuracy": 0.14275750368833542,
"num_tokens": 8129448.0,
"step": 4005
},
{
"entropy": 6.165897274017334,
"epoch": 0.35043257886917767,
"grad_norm": 0.92578125,
"learning_rate": 0.000499218973191166,
"loss": 5.9221,
"mean_token_accuracy": 0.15644895881414414,
"num_tokens": 8138345.0,
"step": 4010
},
{
"entropy": 6.224948310852051,
"epoch": 0.35086952722188236,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992163769077853,
"loss": 5.9829,
"mean_token_accuracy": 0.15417408645153047,
"num_tokens": 8148380.0,
"step": 4015
},
{
"entropy": 6.180328798294068,
"epoch": 0.3513064755745871,
"grad_norm": 1.0078125,
"learning_rate": 0.000499213776323817,
"loss": 6.0055,
"mean_token_accuracy": 0.15456762313842773,
"num_tokens": 8159359.0,
"step": 4020
},
{
"entropy": 6.2428892135620115,
"epoch": 0.3517434239272918,
"grad_norm": 0.94921875,
"learning_rate": 0.000499211171439311,
"loss": 6.0872,
"mean_token_accuracy": 0.14517719000577928,
"num_tokens": 8168586.0,
"step": 4025
},
{
"entropy": 6.204936695098877,
"epoch": 0.3521803722799965,
"grad_norm": 0.8125,
"learning_rate": 0.0004992085622543172,
"loss": 5.9854,
"mean_token_accuracy": 0.1576502501964569,
"num_tokens": 8178836.0,
"step": 4030
},
{
"entropy": 6.168943452835083,
"epoch": 0.35261732063270124,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992059487688859,
"loss": 6.008,
"mean_token_accuracy": 0.15231774598360062,
"num_tokens": 8188933.0,
"step": 4035
},
{
"entropy": 6.144512033462524,
"epoch": 0.35305426898540593,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992033309830668,
"loss": 5.9847,
"mean_token_accuracy": 0.16302024722099304,
"num_tokens": 8199366.0,
"step": 4040
},
{
"entropy": 6.314675426483154,
"epoch": 0.3534912173381106,
"grad_norm": 1.453125,
"learning_rate": 0.0004992007088969105,
"loss": 5.9634,
"mean_token_accuracy": 0.1587460979819298,
"num_tokens": 8209806.0,
"step": 4045
},
{
"entropy": 6.2659971714019775,
"epoch": 0.35392816569081537,
"grad_norm": 1.0078125,
"learning_rate": 0.000499198082510467,
"loss": 6.0697,
"mean_token_accuracy": 0.14272622764110565,
"num_tokens": 8219367.0,
"step": 4050
},
{
"entropy": 6.154402875900269,
"epoch": 0.35436511404352006,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991954518237868,
"loss": 5.9533,
"mean_token_accuracy": 0.151335009932518,
"num_tokens": 8228918.0,
"step": 4055
},
{
"entropy": 6.1810705184936525,
"epoch": 0.35480206239622475,
"grad_norm": 0.87890625,
"learning_rate": 0.0004991928168369205,
"loss": 6.039,
"mean_token_accuracy": 0.1519352450966835,
"num_tokens": 8239380.0,
"step": 4060
},
{
"entropy": 6.269686651229859,
"epoch": 0.3552390107489295,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991901775499183,
"loss": 6.1006,
"mean_token_accuracy": 0.14787229001522065,
"num_tokens": 8249131.0,
"step": 4065
},
{
"entropy": 6.278018474578857,
"epoch": 0.3556759591016342,
"grad_norm": 0.95703125,
"learning_rate": 0.0004991875339628311,
"loss": 5.984,
"mean_token_accuracy": 0.15121260285377502,
"num_tokens": 8258550.0,
"step": 4070
},
{
"entropy": 6.158737659454346,
"epoch": 0.3561129074543389,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991848860757095,
"loss": 6.0121,
"mean_token_accuracy": 0.15460924804210663,
"num_tokens": 8269089.0,
"step": 4075
},
{
"entropy": 6.170330953598023,
"epoch": 0.3565498558070436,
"grad_norm": 0.85546875,
"learning_rate": 0.0004991822338886044,
"loss": 5.9981,
"mean_token_accuracy": 0.1478424534201622,
"num_tokens": 8279735.0,
"step": 4080
},
{
"entropy": 6.221038961410523,
"epoch": 0.3569868041597483,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991795774015665,
"loss": 5.9616,
"mean_token_accuracy": 0.15689391344785691,
"num_tokens": 8289693.0,
"step": 4085
},
{
"entropy": 6.19352879524231,
"epoch": 0.357423752512453,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991769166146469,
"loss": 6.0344,
"mean_token_accuracy": 0.1523745611310005,
"num_tokens": 8300909.0,
"step": 4090
},
{
"entropy": 6.205101537704468,
"epoch": 0.35786070086515775,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991742515278965,
"loss": 6.0008,
"mean_token_accuracy": 0.15361023545265198,
"num_tokens": 8311479.0,
"step": 4095
},
{
"entropy": 6.265884780883789,
"epoch": 0.35829764921786245,
"grad_norm": 0.8828125,
"learning_rate": 0.0004991715821413665,
"loss": 5.9999,
"mean_token_accuracy": 0.15454096496105194,
"num_tokens": 8321267.0,
"step": 4100
},
{
"entropy": 6.146849822998047,
"epoch": 0.35873459757056714,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991689084551082,
"loss": 5.9561,
"mean_token_accuracy": 0.15801454037427903,
"num_tokens": 8331445.0,
"step": 4105
},
{
"entropy": 6.163304042816162,
"epoch": 0.3591715459232719,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991662304691727,
"loss": 6.0427,
"mean_token_accuracy": 0.14610980674624444,
"num_tokens": 8340519.0,
"step": 4110
},
{
"entropy": 6.259560871124267,
"epoch": 0.3596084942759766,
"grad_norm": 0.9375,
"learning_rate": 0.0004991635481836114,
"loss": 6.0199,
"mean_token_accuracy": 0.14872232228517532,
"num_tokens": 8350467.0,
"step": 4115
},
{
"entropy": 6.172140598297119,
"epoch": 0.36004544262868127,
"grad_norm": 0.8828125,
"learning_rate": 0.0004991608615984757,
"loss": 5.9613,
"mean_token_accuracy": 0.1463010773062706,
"num_tokens": 8360067.0,
"step": 4120
},
{
"entropy": 6.2325794219970705,
"epoch": 0.360482390981386,
"grad_norm": 0.96484375,
"learning_rate": 0.0004991581707138175,
"loss": 6.0362,
"mean_token_accuracy": 0.15214309245347976,
"num_tokens": 8370509.0,
"step": 4125
},
{
"entropy": 6.223120260238647,
"epoch": 0.3609193393340907,
"grad_norm": 0.7890625,
"learning_rate": 0.0004991554755296878,
"loss": 5.982,
"mean_token_accuracy": 0.15292935967445373,
"num_tokens": 8381489.0,
"step": 4130
},
{
"entropy": 6.122712326049805,
"epoch": 0.3613562876867954,
"grad_norm": 0.87890625,
"learning_rate": 0.0004991527760461389,
"loss": 6.0008,
"mean_token_accuracy": 0.15350275188684465,
"num_tokens": 8392287.0,
"step": 4135
},
{
"entropy": 6.170266437530517,
"epoch": 0.36179323603950014,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991500722632223,
"loss": 5.9812,
"mean_token_accuracy": 0.15097640305757523,
"num_tokens": 8402433.0,
"step": 4140
},
{
"entropy": 6.1707926273345945,
"epoch": 0.36223018439220483,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991473641809896,
"loss": 6.0225,
"mean_token_accuracy": 0.14983665943145752,
"num_tokens": 8412944.0,
"step": 4145
},
{
"entropy": 6.2134088516235355,
"epoch": 0.3626671327449095,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991446517994932,
"loss": 5.9384,
"mean_token_accuracy": 0.15240545570850372,
"num_tokens": 8422511.0,
"step": 4150
},
{
"entropy": 6.168175792694091,
"epoch": 0.3631040810976143,
"grad_norm": 1.0546875,
"learning_rate": 0.0004991419351187849,
"loss": 5.9653,
"mean_token_accuracy": 0.15691232234239577,
"num_tokens": 8433094.0,
"step": 4155
},
{
"entropy": 6.125294160842896,
"epoch": 0.36354102945031896,
"grad_norm": 0.79296875,
"learning_rate": 0.0004991392141389169,
"loss": 6.0623,
"mean_token_accuracy": 0.15332620590925217,
"num_tokens": 8444430.0,
"step": 4160
},
{
"entropy": 6.231081438064575,
"epoch": 0.36397797780302366,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991364888599412,
"loss": 6.1176,
"mean_token_accuracy": 0.1470529094338417,
"num_tokens": 8455171.0,
"step": 4165
},
{
"entropy": 6.221288394927979,
"epoch": 0.3644149261557284,
"grad_norm": 0.984375,
"learning_rate": 0.0004991337592819103,
"loss": 5.921,
"mean_token_accuracy": 0.1584060475230217,
"num_tokens": 8464724.0,
"step": 4170
},
{
"entropy": 6.2204231262207035,
"epoch": 0.3648518745084331,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991310254048763,
"loss": 6.0201,
"mean_token_accuracy": 0.15029047429561615,
"num_tokens": 8474732.0,
"step": 4175
},
{
"entropy": 6.133815145492553,
"epoch": 0.36528882286113784,
"grad_norm": 0.80078125,
"learning_rate": 0.0004991282872288919,
"loss": 5.9576,
"mean_token_accuracy": 0.15694391280412673,
"num_tokens": 8484731.0,
"step": 4180
},
{
"entropy": 6.150207233428955,
"epoch": 0.36572577121384253,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991255447540095,
"loss": 5.9559,
"mean_token_accuracy": 0.15796650052070618,
"num_tokens": 8494509.0,
"step": 4185
},
{
"entropy": 6.219148397445679,
"epoch": 0.3661627195665472,
"grad_norm": 0.99609375,
"learning_rate": 0.0004991227979802817,
"loss": 6.0406,
"mean_token_accuracy": 0.15501723140478135,
"num_tokens": 8504260.0,
"step": 4190
},
{
"entropy": 6.221963882446289,
"epoch": 0.36659966791925197,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991200469077612,
"loss": 6.0644,
"mean_token_accuracy": 0.14710740000009537,
"num_tokens": 8514222.0,
"step": 4195
},
{
"entropy": 6.22112398147583,
"epoch": 0.36703661627195666,
"grad_norm": 0.87890625,
"learning_rate": 0.0004991172915365008,
"loss": 6.0062,
"mean_token_accuracy": 0.15218688696622848,
"num_tokens": 8523979.0,
"step": 4200
},
{
"entropy": 6.199363327026367,
"epoch": 0.36747356462466135,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991145318665532,
"loss": 5.9521,
"mean_token_accuracy": 0.1610599249601364,
"num_tokens": 8533516.0,
"step": 4205
},
{
"entropy": 6.191833066940307,
"epoch": 0.3679105129773661,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991117678979716,
"loss": 6.0551,
"mean_token_accuracy": 0.1441046938300133,
"num_tokens": 8544444.0,
"step": 4210
},
{
"entropy": 6.20907039642334,
"epoch": 0.3683474613300708,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991089996308089,
"loss": 5.9794,
"mean_token_accuracy": 0.15260274708271027,
"num_tokens": 8555571.0,
"step": 4215
},
{
"entropy": 6.278723573684692,
"epoch": 0.3687844096827755,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991062270651179,
"loss": 6.0342,
"mean_token_accuracy": 0.14726278856396674,
"num_tokens": 8566486.0,
"step": 4220
},
{
"entropy": 6.21203727722168,
"epoch": 0.36922135803548023,
"grad_norm": 1.0,
"learning_rate": 0.0004991034502009522,
"loss": 5.9621,
"mean_token_accuracy": 0.15388412922620773,
"num_tokens": 8575324.0,
"step": 4225
},
{
"entropy": 6.1372088432312015,
"epoch": 0.3696583063881849,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991006690383649,
"loss": 5.9502,
"mean_token_accuracy": 0.15232342332601548,
"num_tokens": 8586015.0,
"step": 4230
},
{
"entropy": 6.132140016555786,
"epoch": 0.3700952547408896,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990978835774093,
"loss": 5.9306,
"mean_token_accuracy": 0.155072882771492,
"num_tokens": 8594748.0,
"step": 4235
},
{
"entropy": 6.13446011543274,
"epoch": 0.37053220309359436,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990950938181389,
"loss": 6.0013,
"mean_token_accuracy": 0.14853542819619178,
"num_tokens": 8605805.0,
"step": 4240
},
{
"entropy": 6.186568546295166,
"epoch": 0.37096915144629905,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990922997606071,
"loss": 5.9595,
"mean_token_accuracy": 0.15177000910043717,
"num_tokens": 8616591.0,
"step": 4245
},
{
"entropy": 6.214903497695923,
"epoch": 0.37140609979900374,
"grad_norm": 0.921875,
"learning_rate": 0.0004990895014048677,
"loss": 5.987,
"mean_token_accuracy": 0.1489131972193718,
"num_tokens": 8627063.0,
"step": 4250
},
{
"entropy": 6.300046157836914,
"epoch": 0.3718430481517085,
"grad_norm": 0.98828125,
"learning_rate": 0.0004990866987509741,
"loss": 6.0661,
"mean_token_accuracy": 0.14470021352171897,
"num_tokens": 8637041.0,
"step": 4255
},
{
"entropy": 6.190048170089722,
"epoch": 0.3722799965044132,
"grad_norm": 0.87109375,
"learning_rate": 0.0004990838917989802,
"loss": 5.9773,
"mean_token_accuracy": 0.15778657048940659,
"num_tokens": 8647485.0,
"step": 4260
},
{
"entropy": 6.16148829460144,
"epoch": 0.37271694485711787,
"grad_norm": 0.921875,
"learning_rate": 0.00049908108054894,
"loss": 5.9516,
"mean_token_accuracy": 0.15104584991931916,
"num_tokens": 8657185.0,
"step": 4265
},
{
"entropy": 6.20255479812622,
"epoch": 0.3731538932098226,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990782650009071,
"loss": 5.9613,
"mean_token_accuracy": 0.15804735124111174,
"num_tokens": 8667639.0,
"step": 4270
},
{
"entropy": 6.188499164581299,
"epoch": 0.3735908415625273,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990754451549358,
"loss": 5.9794,
"mean_token_accuracy": 0.15283505171537398,
"num_tokens": 8678003.0,
"step": 4275
},
{
"entropy": 6.210866785049438,
"epoch": 0.374027789915232,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990726210110798,
"loss": 6.0443,
"mean_token_accuracy": 0.15008555352687836,
"num_tokens": 8688837.0,
"step": 4280
},
{
"entropy": 6.155147314071655,
"epoch": 0.37446473826793675,
"grad_norm": 0.953125,
"learning_rate": 0.0004990697925693938,
"loss": 5.8289,
"mean_token_accuracy": 0.16587514132261277,
"num_tokens": 8698395.0,
"step": 4285
},
{
"entropy": 6.157758569717407,
"epoch": 0.37490168662064144,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990669598299316,
"loss": 5.9545,
"mean_token_accuracy": 0.15433542281389237,
"num_tokens": 8708031.0,
"step": 4290
},
{
"entropy": 6.12908787727356,
"epoch": 0.37533863497334613,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990641227927478,
"loss": 5.9331,
"mean_token_accuracy": 0.15757068544626235,
"num_tokens": 8717825.0,
"step": 4295
},
{
"entropy": 6.257285165786743,
"epoch": 0.3757755833260509,
"grad_norm": 0.90234375,
"learning_rate": 0.0004990612814578967,
"loss": 6.0194,
"mean_token_accuracy": 0.14891362860798835,
"num_tokens": 8728873.0,
"step": 4300
},
{
"entropy": 6.150735092163086,
"epoch": 0.37621253167875557,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990584358254327,
"loss": 5.9536,
"mean_token_accuracy": 0.15229746103286743,
"num_tokens": 8737917.0,
"step": 4305
},
{
"entropy": 6.098759412765503,
"epoch": 0.37664948003146026,
"grad_norm": 0.8984375,
"learning_rate": 0.0004990555858954107,
"loss": 5.9038,
"mean_token_accuracy": 0.1548462837934494,
"num_tokens": 8747466.0,
"step": 4310
},
{
"entropy": 6.17538890838623,
"epoch": 0.377086428384165,
"grad_norm": 0.85546875,
"learning_rate": 0.000499052731667885,
"loss": 5.9409,
"mean_token_accuracy": 0.1568036288022995,
"num_tokens": 8757037.0,
"step": 4315
},
{
"entropy": 6.196196222305298,
"epoch": 0.3775233767368697,
"grad_norm": 1.015625,
"learning_rate": 0.0004990498731429105,
"loss": 5.9771,
"mean_token_accuracy": 0.1564203307032585,
"num_tokens": 8766780.0,
"step": 4320
},
{
"entropy": 6.169270992279053,
"epoch": 0.3779603250895744,
"grad_norm": 0.86328125,
"learning_rate": 0.0004990470103205421,
"loss": 5.9469,
"mean_token_accuracy": 0.1535429298877716,
"num_tokens": 8777063.0,
"step": 4325
},
{
"entropy": 6.119955730438233,
"epoch": 0.37839727344227914,
"grad_norm": 0.94140625,
"learning_rate": 0.0004990441432008347,
"loss": 5.9802,
"mean_token_accuracy": 0.15425399094820022,
"num_tokens": 8786479.0,
"step": 4330
},
{
"entropy": 6.275782108306885,
"epoch": 0.3788342217949838,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990412717838431,
"loss": 5.9646,
"mean_token_accuracy": 0.15385251939296724,
"num_tokens": 8795663.0,
"step": 4335
},
{
"entropy": 6.1232274055480955,
"epoch": 0.3792711701476885,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990383960696226,
"loss": 5.9272,
"mean_token_accuracy": 0.15729418247938157,
"num_tokens": 8805854.0,
"step": 4340
},
{
"entropy": 6.137463521957398,
"epoch": 0.37970811850039327,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990355160582283,
"loss": 5.8988,
"mean_token_accuracy": 0.15592976957559584,
"num_tokens": 8816068.0,
"step": 4345
},
{
"entropy": 6.1173069953918455,
"epoch": 0.38014506685309796,
"grad_norm": 0.88671875,
"learning_rate": 0.0004990326317497153,
"loss": 5.9099,
"mean_token_accuracy": 0.15910188257694244,
"num_tokens": 8826301.0,
"step": 4350
},
{
"entropy": 6.201882314682007,
"epoch": 0.38058201520580265,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990297431441391,
"loss": 5.9579,
"mean_token_accuracy": 0.16176767647266388,
"num_tokens": 8837482.0,
"step": 4355
},
{
"entropy": 6.178161764144898,
"epoch": 0.3810189635585074,
"grad_norm": 0.9296875,
"learning_rate": 0.000499026850241555,
"loss": 5.9815,
"mean_token_accuracy": 0.15647393688559533,
"num_tokens": 8847041.0,
"step": 4360
},
{
"entropy": 6.138125228881836,
"epoch": 0.3814559119112121,
"grad_norm": 0.90234375,
"learning_rate": 0.0004990239530420185,
"loss": 5.971,
"mean_token_accuracy": 0.1511300563812256,
"num_tokens": 8856963.0,
"step": 4365
},
{
"entropy": 6.18984546661377,
"epoch": 0.3818928602639168,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990210515455853,
"loss": 5.9517,
"mean_token_accuracy": 0.1531401291489601,
"num_tokens": 8867318.0,
"step": 4370
},
{
"entropy": 6.149747371673584,
"epoch": 0.3823298086166215,
"grad_norm": 0.85546875,
"learning_rate": 0.0004990181457523109,
"loss": 5.9576,
"mean_token_accuracy": 0.15716585367918015,
"num_tokens": 8877634.0,
"step": 4375
},
{
"entropy": 6.152999687194824,
"epoch": 0.3827667569693262,
"grad_norm": 0.9296875,
"learning_rate": 0.000499015235662251,
"loss": 5.9975,
"mean_token_accuracy": 0.15242273956537247,
"num_tokens": 8887708.0,
"step": 4380
},
{
"entropy": 6.169225645065308,
"epoch": 0.38320370532203096,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990123212754616,
"loss": 5.925,
"mean_token_accuracy": 0.1560312107205391,
"num_tokens": 8897510.0,
"step": 4385
},
{
"entropy": 6.145167112350464,
"epoch": 0.38364065367473565,
"grad_norm": 0.90625,
"learning_rate": 0.0004990094025919985,
"loss": 5.9577,
"mean_token_accuracy": 0.1475707024335861,
"num_tokens": 8908282.0,
"step": 4390
},
{
"entropy": 6.280637264251709,
"epoch": 0.38407760202744035,
"grad_norm": 0.8515625,
"learning_rate": 0.0004990064796119175,
"loss": 5.9905,
"mean_token_accuracy": 0.14701349660754204,
"num_tokens": 8918332.0,
"step": 4395
},
{
"entropy": 6.1294549942016605,
"epoch": 0.3845145503801451,
"grad_norm": 0.90625,
"learning_rate": 0.0004990035523352751,
"loss": 5.957,
"mean_token_accuracy": 0.15417761653661727,
"num_tokens": 8929138.0,
"step": 4400
},
{
"entropy": 6.135020399093628,
"epoch": 0.3849514987328498,
"grad_norm": 0.86328125,
"learning_rate": 0.000499000620762127,
"loss": 5.9958,
"mean_token_accuracy": 0.14873793125152587,
"num_tokens": 8940871.0,
"step": 4405
},
{
"entropy": 6.231491804122925,
"epoch": 0.3853884470855545,
"grad_norm": 0.9765625,
"learning_rate": 0.0004989976848925297,
"loss": 5.9361,
"mean_token_accuracy": 0.15769388526678085,
"num_tokens": 8951199.0,
"step": 4410
},
{
"entropy": 6.145180130004883,
"epoch": 0.3858253954382592,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989947447265394,
"loss": 6.0565,
"mean_token_accuracy": 0.14883989095687866,
"num_tokens": 8961482.0,
"step": 4415
},
{
"entropy": 6.272399139404297,
"epoch": 0.3862623437909639,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989918002642126,
"loss": 5.966,
"mean_token_accuracy": 0.1529775321483612,
"num_tokens": 8971884.0,
"step": 4420
},
{
"entropy": 6.130031728744507,
"epoch": 0.3866992921436686,
"grad_norm": 0.86328125,
"learning_rate": 0.0004989888515056057,
"loss": 5.9206,
"mean_token_accuracy": 0.1541355162858963,
"num_tokens": 8982086.0,
"step": 4425
},
{
"entropy": 6.135330867767334,
"epoch": 0.38713624049637335,
"grad_norm": 0.953125,
"learning_rate": 0.0004989858984507752,
"loss": 5.9874,
"mean_token_accuracy": 0.152043230086565,
"num_tokens": 8991555.0,
"step": 4430
},
{
"entropy": 6.160719919204712,
"epoch": 0.38757318884907804,
"grad_norm": 0.9140625,
"learning_rate": 0.0004989829410997778,
"loss": 5.9709,
"mean_token_accuracy": 0.1504402533173561,
"num_tokens": 9002459.0,
"step": 4435
},
{
"entropy": 6.1761571884155275,
"epoch": 0.38801013720178273,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989799794526702,
"loss": 6.0036,
"mean_token_accuracy": 0.14963382706046105,
"num_tokens": 9012514.0,
"step": 4440
},
{
"entropy": 6.172939491271973,
"epoch": 0.3884470855544875,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989770135095093,
"loss": 5.9261,
"mean_token_accuracy": 0.1512054905295372,
"num_tokens": 9022636.0,
"step": 4445
},
{
"entropy": 6.146451759338379,
"epoch": 0.3888840339071922,
"grad_norm": 0.90234375,
"learning_rate": 0.0004989740432703519,
"loss": 5.9381,
"mean_token_accuracy": 0.15658134073019028,
"num_tokens": 9032868.0,
"step": 4450
},
{
"entropy": 6.137963438034058,
"epoch": 0.38932098225989686,
"grad_norm": 0.9140625,
"learning_rate": 0.000498971068735255,
"loss": 5.9352,
"mean_token_accuracy": 0.14985790252685546,
"num_tokens": 9043057.0,
"step": 4455
},
{
"entropy": 6.127811861038208,
"epoch": 0.3897579306126016,
"grad_norm": 0.86328125,
"learning_rate": 0.0004989680899042756,
"loss": 5.9517,
"mean_token_accuracy": 0.15457415729761123,
"num_tokens": 9053179.0,
"step": 4460
},
{
"entropy": 6.138548040390015,
"epoch": 0.3901948789653063,
"grad_norm": 0.94140625,
"learning_rate": 0.0004989651067774708,
"loss": 5.96,
"mean_token_accuracy": 0.15666047632694244,
"num_tokens": 9063112.0,
"step": 4465
},
{
"entropy": 6.1771155834198,
"epoch": 0.390631827318011,
"grad_norm": 0.90234375,
"learning_rate": 0.0004989621193548981,
"loss": 5.9731,
"mean_token_accuracy": 0.15245330333709717,
"num_tokens": 9073512.0,
"step": 4470
},
{
"entropy": 6.193724632263184,
"epoch": 0.39106877567071574,
"grad_norm": 0.97265625,
"learning_rate": 0.0004989591276366145,
"loss": 5.9841,
"mean_token_accuracy": 0.14708113372325898,
"num_tokens": 9083369.0,
"step": 4475
},
{
"entropy": 6.141264724731445,
"epoch": 0.39150572402342043,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989561316226774,
"loss": 5.9619,
"mean_token_accuracy": 0.15295498222112655,
"num_tokens": 9092814.0,
"step": 4480
},
{
"entropy": 6.042321491241455,
"epoch": 0.3919426723761251,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989531313131444,
"loss": 5.8825,
"mean_token_accuracy": 0.16421083658933638,
"num_tokens": 9102816.0,
"step": 4485
},
{
"entropy": 6.154022264480591,
"epoch": 0.39237962072882987,
"grad_norm": 0.90625,
"learning_rate": 0.000498950126708073,
"loss": 6.0144,
"mean_token_accuracy": 0.14581276923418046,
"num_tokens": 9114470.0,
"step": 4490
},
{
"entropy": 6.221161270141602,
"epoch": 0.39281656908153456,
"grad_norm": 0.87109375,
"learning_rate": 0.0004989471178075208,
"loss": 5.989,
"mean_token_accuracy": 0.14692097306251525,
"num_tokens": 9125604.0,
"step": 4495
},
{
"entropy": 6.133387279510498,
"epoch": 0.39325351743423925,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989441046115456,
"loss": 5.9294,
"mean_token_accuracy": 0.15933895260095596,
"num_tokens": 9135088.0,
"step": 4500
},
{
"entropy": 6.188969993591309,
"epoch": 0.393690465786944,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989410871202051,
"loss": 5.9832,
"mean_token_accuracy": 0.15392538905143738,
"num_tokens": 9145302.0,
"step": 4505
},
{
"entropy": 6.105658292770386,
"epoch": 0.3941274141396487,
"grad_norm": 0.91015625,
"learning_rate": 0.0004989380653335571,
"loss": 5.9199,
"mean_token_accuracy": 0.15435281097888948,
"num_tokens": 9155104.0,
"step": 4510
},
{
"entropy": 6.124040269851685,
"epoch": 0.3945643624923534,
"grad_norm": 0.890625,
"learning_rate": 0.0004989350392516598,
"loss": 5.8527,
"mean_token_accuracy": 0.15398550778627396,
"num_tokens": 9165055.0,
"step": 4515
},
{
"entropy": 6.115254163742065,
"epoch": 0.39500131084505813,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989320088745711,
"loss": 5.8808,
"mean_token_accuracy": 0.16048204451799392,
"num_tokens": 9175305.0,
"step": 4520
},
{
"entropy": 6.1552516460418705,
"epoch": 0.3954382591977628,
"grad_norm": 0.9609375,
"learning_rate": 0.000498928974202349,
"loss": 5.9706,
"mean_token_accuracy": 0.1471177026629448,
"num_tokens": 9185153.0,
"step": 4525
},
{
"entropy": 6.160642766952515,
"epoch": 0.3958752075504675,
"grad_norm": 0.97265625,
"learning_rate": 0.000498925935235052,
"loss": 5.9264,
"mean_token_accuracy": 0.1507927320897579,
"num_tokens": 9195385.0,
"step": 4530
},
{
"entropy": 6.099987506866455,
"epoch": 0.39631215590317226,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989228919727382,
"loss": 5.8825,
"mean_token_accuracy": 0.16080036759376526,
"num_tokens": 9205009.0,
"step": 4535
},
{
"entropy": 6.138633680343628,
"epoch": 0.39674910425587695,
"grad_norm": 0.859375,
"learning_rate": 0.000498919844415466,
"loss": 5.9448,
"mean_token_accuracy": 0.15411535501480103,
"num_tokens": 9214684.0,
"step": 4540
},
{
"entropy": 6.087051582336426,
"epoch": 0.39718605260858164,
"grad_norm": 0.890625,
"learning_rate": 0.0004989167925632937,
"loss": 5.8855,
"mean_token_accuracy": 0.1588042065501213,
"num_tokens": 9225721.0,
"step": 4545
},
{
"entropy": 6.1690984725952145,
"epoch": 0.3976230009612864,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989137364162801,
"loss": 5.967,
"mean_token_accuracy": 0.15482041388750076,
"num_tokens": 9236167.0,
"step": 4550
},
{
"entropy": 6.09454870223999,
"epoch": 0.3980599493139911,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989106759744837,
"loss": 5.9052,
"mean_token_accuracy": 0.16375220119953154,
"num_tokens": 9245716.0,
"step": 4555
},
{
"entropy": 6.142717027664185,
"epoch": 0.39849689766669577,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989076112379633,
"loss": 5.9952,
"mean_token_accuracy": 0.1506937101483345,
"num_tokens": 9256897.0,
"step": 4560
},
{
"entropy": 6.255190420150757,
"epoch": 0.3989338460194005,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989045422067775,
"loss": 6.0027,
"mean_token_accuracy": 0.14253595024347304,
"num_tokens": 9266080.0,
"step": 4565
},
{
"entropy": 6.113092374801636,
"epoch": 0.3993707943721052,
"grad_norm": 0.85546875,
"learning_rate": 0.0004989014688809853,
"loss": 5.9129,
"mean_token_accuracy": 0.1528676062822342,
"num_tokens": 9276815.0,
"step": 4570
},
{
"entropy": 6.166882181167603,
"epoch": 0.3998077427248099,
"grad_norm": 0.890625,
"learning_rate": 0.0004988983912606454,
"loss": 5.9845,
"mean_token_accuracy": 0.15317472666502,
"num_tokens": 9287339.0,
"step": 4575
},
{
"entropy": 6.104081726074218,
"epoch": 0.40024469107751465,
"grad_norm": 0.83203125,
"learning_rate": 0.0004988953093458173,
"loss": 5.8525,
"mean_token_accuracy": 0.160149547457695,
"num_tokens": 9298002.0,
"step": 4580
},
{
"entropy": 6.127166795730591,
"epoch": 0.40068163943021934,
"grad_norm": 0.91796875,
"learning_rate": 0.0004988922231365597,
"loss": 5.9923,
"mean_token_accuracy": 0.14690545797348023,
"num_tokens": 9309008.0,
"step": 4585
},
{
"entropy": 6.125999641418457,
"epoch": 0.4011185877829241,
"grad_norm": 0.84375,
"learning_rate": 0.000498889132632932,
"loss": 5.9939,
"mean_token_accuracy": 0.1510233238339424,
"num_tokens": 9319391.0,
"step": 4590
},
{
"entropy": 6.04268856048584,
"epoch": 0.4015555361356288,
"grad_norm": 0.84375,
"learning_rate": 0.0004988860378349934,
"loss": 5.8255,
"mean_token_accuracy": 0.1640730954706669,
"num_tokens": 9329808.0,
"step": 4595
},
{
"entropy": 6.1170600891113285,
"epoch": 0.40199248448833347,
"grad_norm": 0.90625,
"learning_rate": 0.0004988829387428033,
"loss": 5.8954,
"mean_token_accuracy": 0.158653524518013,
"num_tokens": 9339580.0,
"step": 4600
},
{
"entropy": 6.180007410049439,
"epoch": 0.4024294328410382,
"grad_norm": 0.96875,
"learning_rate": 0.0004988798353564211,
"loss": 5.914,
"mean_token_accuracy": 0.16047825962305068,
"num_tokens": 9350008.0,
"step": 4605
},
{
"entropy": 6.181416606903076,
"epoch": 0.4028663811937429,
"grad_norm": 1.0,
"learning_rate": 0.0004988767276759064,
"loss": 5.9099,
"mean_token_accuracy": 0.15556182861328124,
"num_tokens": 9359026.0,
"step": 4610
},
{
"entropy": 6.122279930114746,
"epoch": 0.4033033295464476,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988736157013186,
"loss": 5.8758,
"mean_token_accuracy": 0.15592683404684066,
"num_tokens": 9369084.0,
"step": 4615
},
{
"entropy": 6.068732404708863,
"epoch": 0.40374027789915234,
"grad_norm": 1.0,
"learning_rate": 0.0004988704994327178,
"loss": 5.8703,
"mean_token_accuracy": 0.16378343552351,
"num_tokens": 9378273.0,
"step": 4620
},
{
"entropy": 6.097204971313476,
"epoch": 0.40417722625185704,
"grad_norm": 0.88671875,
"learning_rate": 0.0004988673788701633,
"loss": 5.9002,
"mean_token_accuracy": 0.1529839888215065,
"num_tokens": 9388434.0,
"step": 4625
},
{
"entropy": 6.173472833633423,
"epoch": 0.4046141746045617,
"grad_norm": 0.88671875,
"learning_rate": 0.0004988642540137152,
"loss": 5.9829,
"mean_token_accuracy": 0.14962709844112396,
"num_tokens": 9398484.0,
"step": 4630
},
{
"entropy": 6.070959091186523,
"epoch": 0.4050511229572665,
"grad_norm": 0.83203125,
"learning_rate": 0.0004988611248634335,
"loss": 5.8537,
"mean_token_accuracy": 0.1586995616555214,
"num_tokens": 9409544.0,
"step": 4635
},
{
"entropy": 6.145827198028565,
"epoch": 0.40548807130997117,
"grad_norm": 0.9453125,
"learning_rate": 0.000498857991419378,
"loss": 5.9176,
"mean_token_accuracy": 0.15430696606636046,
"num_tokens": 9419391.0,
"step": 4640
},
{
"entropy": 6.167720651626587,
"epoch": 0.40592501966267586,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988548536816091,
"loss": 5.8823,
"mean_token_accuracy": 0.16215257942676545,
"num_tokens": 9429342.0,
"step": 4645
},
{
"entropy": 6.117645025253296,
"epoch": 0.4063619680153806,
"grad_norm": 0.84765625,
"learning_rate": 0.0004988517116501867,
"loss": 5.9971,
"mean_token_accuracy": 0.14209671840071678,
"num_tokens": 9440643.0,
"step": 4650
},
{
"entropy": 6.208621311187744,
"epoch": 0.4067989163680853,
"grad_norm": 0.96875,
"learning_rate": 0.0004988485653251711,
"loss": 5.9451,
"mean_token_accuracy": 0.15437053889036179,
"num_tokens": 9450674.0,
"step": 4655
},
{
"entropy": 6.1741351127624515,
"epoch": 0.40723586472079,
"grad_norm": 0.9375,
"learning_rate": 0.0004988454147066228,
"loss": 5.9197,
"mean_token_accuracy": 0.15716563463211058,
"num_tokens": 9460075.0,
"step": 4660
},
{
"entropy": 6.099549198150635,
"epoch": 0.40767281307349473,
"grad_norm": 0.9453125,
"learning_rate": 0.0004988422597946022,
"loss": 5.896,
"mean_token_accuracy": 0.16132495701313018,
"num_tokens": 9469327.0,
"step": 4665
},
{
"entropy": 6.170559120178223,
"epoch": 0.4081097614261994,
"grad_norm": 0.953125,
"learning_rate": 0.0004988391005891698,
"loss": 5.9437,
"mean_token_accuracy": 0.1529410094022751,
"num_tokens": 9478304.0,
"step": 4670
},
{
"entropy": 6.089248180389404,
"epoch": 0.4085467097789041,
"grad_norm": 0.984375,
"learning_rate": 0.000498835937090386,
"loss": 5.8651,
"mean_token_accuracy": 0.15965885519981385,
"num_tokens": 9487453.0,
"step": 4675
},
{
"entropy": 6.091420078277588,
"epoch": 0.40898365813160886,
"grad_norm": 0.96484375,
"learning_rate": 0.0004988327692983116,
"loss": 5.8263,
"mean_token_accuracy": 0.16226141899824142,
"num_tokens": 9496623.0,
"step": 4680
},
{
"entropy": 6.112924242019654,
"epoch": 0.40942060648431355,
"grad_norm": 0.89453125,
"learning_rate": 0.0004988295972130076,
"loss": 5.9733,
"mean_token_accuracy": 0.15255378037691117,
"num_tokens": 9506293.0,
"step": 4685
},
{
"entropy": 6.144012498855591,
"epoch": 0.40985755483701825,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988264208345346,
"loss": 5.8609,
"mean_token_accuracy": 0.1574437811970711,
"num_tokens": 9516127.0,
"step": 4690
},
{
"entropy": 6.063244152069092,
"epoch": 0.410294503189723,
"grad_norm": 0.921875,
"learning_rate": 0.0004988232401629535,
"loss": 5.8666,
"mean_token_accuracy": 0.15879199355840684,
"num_tokens": 9525597.0,
"step": 4695
},
{
"entropy": 6.0836179733276365,
"epoch": 0.4107314515424277,
"grad_norm": 0.96484375,
"learning_rate": 0.0004988200551983254,
"loss": 5.8454,
"mean_token_accuracy": 0.1594383656978607,
"num_tokens": 9535208.0,
"step": 4700
},
{
"entropy": 6.075837135314941,
"epoch": 0.4111683998951324,
"grad_norm": 0.92578125,
"learning_rate": 0.0004988168659407113,
"loss": 5.8929,
"mean_token_accuracy": 0.16378208845853806,
"num_tokens": 9545078.0,
"step": 4705
},
{
"entropy": 6.160520839691162,
"epoch": 0.4116053482478371,
"grad_norm": 0.921875,
"learning_rate": 0.0004988136723901725,
"loss": 5.9045,
"mean_token_accuracy": 0.1531546324491501,
"num_tokens": 9555074.0,
"step": 4710
},
{
"entropy": 6.109802627563477,
"epoch": 0.4120422966005418,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988104745467702,
"loss": 5.9027,
"mean_token_accuracy": 0.15896735340356827,
"num_tokens": 9564777.0,
"step": 4715
},
{
"entropy": 6.107122421264648,
"epoch": 0.4124792449532465,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988072724105657,
"loss": 5.8435,
"mean_token_accuracy": 0.1565806970000267,
"num_tokens": 9574551.0,
"step": 4720
},
{
"entropy": 6.101800775527954,
"epoch": 0.41291619330595125,
"grad_norm": 0.890625,
"learning_rate": 0.0004988040659816203,
"loss": 5.9663,
"mean_token_accuracy": 0.14592478796839714,
"num_tokens": 9585162.0,
"step": 4725
},
{
"entropy": 6.126856660842895,
"epoch": 0.41335314165865594,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988008552599958,
"loss": 5.8922,
"mean_token_accuracy": 0.16171422302722932,
"num_tokens": 9595422.0,
"step": 4730
},
{
"entropy": 6.096254062652588,
"epoch": 0.41379009001136063,
"grad_norm": 0.90625,
"learning_rate": 0.0004987976402457536,
"loss": 5.8825,
"mean_token_accuracy": 0.165338733792305,
"num_tokens": 9604315.0,
"step": 4735
},
{
"entropy": 6.0762707710266115,
"epoch": 0.4142270383640654,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987944209389555,
"loss": 5.8958,
"mean_token_accuracy": 0.15499772429466246,
"num_tokens": 9614369.0,
"step": 4740
},
{
"entropy": 6.158682441711425,
"epoch": 0.41466398671677007,
"grad_norm": 0.96875,
"learning_rate": 0.000498791197339663,
"loss": 5.9443,
"mean_token_accuracy": 0.1528699055314064,
"num_tokens": 9624154.0,
"step": 4745
},
{
"entropy": 6.150540733337403,
"epoch": 0.41510093506947476,
"grad_norm": 1.0625,
"learning_rate": 0.000498787969447938,
"loss": 5.9644,
"mean_token_accuracy": 0.15262885689735411,
"num_tokens": 9634903.0,
"step": 4750
},
{
"entropy": 6.073925113677978,
"epoch": 0.4155378834221795,
"grad_norm": 0.8984375,
"learning_rate": 0.0004987847372638427,
"loss": 5.9637,
"mean_token_accuracy": 0.1611965700984001,
"num_tokens": 9644873.0,
"step": 4755
},
{
"entropy": 6.153850030899048,
"epoch": 0.4159748317748842,
"grad_norm": 0.9296875,
"learning_rate": 0.0004987815007874388,
"loss": 5.8692,
"mean_token_accuracy": 0.17039722949266434,
"num_tokens": 9654661.0,
"step": 4760
},
{
"entropy": 6.119657421112061,
"epoch": 0.4164117801275889,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987782600187884,
"loss": 5.884,
"mean_token_accuracy": 0.16364409178495407,
"num_tokens": 9664980.0,
"step": 4765
},
{
"entropy": 6.09555344581604,
"epoch": 0.41684872848029364,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987750149579537,
"loss": 5.8792,
"mean_token_accuracy": 0.16119770854711532,
"num_tokens": 9675117.0,
"step": 4770
},
{
"entropy": 6.113473224639892,
"epoch": 0.41728567683299833,
"grad_norm": 0.9375,
"learning_rate": 0.000498771765604997,
"loss": 5.923,
"mean_token_accuracy": 0.15419435352087021,
"num_tokens": 9684900.0,
"step": 4775
},
{
"entropy": 6.145103168487549,
"epoch": 0.417722625185703,
"grad_norm": 0.85546875,
"learning_rate": 0.0004987685119599806,
"loss": 5.9188,
"mean_token_accuracy": 0.1530965194106102,
"num_tokens": 9694586.0,
"step": 4780
},
{
"entropy": 6.1405035018920895,
"epoch": 0.41815957353840777,
"grad_norm": 0.83984375,
"learning_rate": 0.0004987652540229668,
"loss": 5.9196,
"mean_token_accuracy": 0.15793851613998414,
"num_tokens": 9705676.0,
"step": 4785
},
{
"entropy": 6.1464728832244875,
"epoch": 0.41859652189111246,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987619917940183,
"loss": 5.9458,
"mean_token_accuracy": 0.15899438858032228,
"num_tokens": 9715442.0,
"step": 4790
},
{
"entropy": 6.219377660751343,
"epoch": 0.4190334702438172,
"grad_norm": 1.1484375,
"learning_rate": 0.0004987587252731975,
"loss": 6.0173,
"mean_token_accuracy": 0.15520615503191948,
"num_tokens": 9726621.0,
"step": 4795
},
{
"entropy": 6.143201494216919,
"epoch": 0.4194704185965219,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987554544605671,
"loss": 6.0238,
"mean_token_accuracy": 0.14427199214696884,
"num_tokens": 9736688.0,
"step": 4800
},
{
"entropy": 6.126335477828979,
"epoch": 0.4199073669492266,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987521793561898,
"loss": 5.8957,
"mean_token_accuracy": 0.16006623208522797,
"num_tokens": 9746478.0,
"step": 4805
},
{
"entropy": 6.120117235183716,
"epoch": 0.42034431530193134,
"grad_norm": 0.9375,
"learning_rate": 0.0004987488999601285,
"loss": 5.9316,
"mean_token_accuracy": 0.156678207218647,
"num_tokens": 9756370.0,
"step": 4810
},
{
"entropy": 6.161874628067016,
"epoch": 0.42078126365463603,
"grad_norm": 0.9375,
"learning_rate": 0.000498745616272446,
"loss": 5.9347,
"mean_token_accuracy": 0.14946951419115068,
"num_tokens": 9766293.0,
"step": 4815
},
{
"entropy": 6.136138820648194,
"epoch": 0.4212182120073407,
"grad_norm": 0.82421875,
"learning_rate": 0.0004987423282932054,
"loss": 5.9138,
"mean_token_accuracy": 0.15771846622228622,
"num_tokens": 9777172.0,
"step": 4820
},
{
"entropy": 6.083975219726563,
"epoch": 0.42165516036004547,
"grad_norm": 0.859375,
"learning_rate": 0.0004987390360224697,
"loss": 5.8839,
"mean_token_accuracy": 0.15716514885425567,
"num_tokens": 9787194.0,
"step": 4825
},
{
"entropy": 6.096252679824829,
"epoch": 0.42209210871275016,
"grad_norm": 0.875,
"learning_rate": 0.000498735739460302,
"loss": 5.9231,
"mean_token_accuracy": 0.15954737961292267,
"num_tokens": 9797528.0,
"step": 4830
},
{
"entropy": 6.119057941436767,
"epoch": 0.42252905706545485,
"grad_norm": 1.125,
"learning_rate": 0.0004987324386067656,
"loss": 5.9458,
"mean_token_accuracy": 0.15423630625009538,
"num_tokens": 9808412.0,
"step": 4835
},
{
"entropy": 6.164724588394165,
"epoch": 0.4229660054181596,
"grad_norm": 0.8359375,
"learning_rate": 0.0004987291334619239,
"loss": 6.0183,
"mean_token_accuracy": 0.14648769348859786,
"num_tokens": 9820047.0,
"step": 4840
},
{
"entropy": 6.1444111347198485,
"epoch": 0.4234029537708643,
"grad_norm": 1.0546875,
"learning_rate": 0.00049872582402584,
"loss": 5.8427,
"mean_token_accuracy": 0.16165557652711868,
"num_tokens": 9828812.0,
"step": 4845
},
{
"entropy": 6.07479567527771,
"epoch": 0.423839902123569,
"grad_norm": 1.0390625,
"learning_rate": 0.0004987225102985778,
"loss": 5.8709,
"mean_token_accuracy": 0.1626689463853836,
"num_tokens": 9838288.0,
"step": 4850
},
{
"entropy": 6.17829852104187,
"epoch": 0.4242768504762737,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987191922802003,
"loss": 5.8979,
"mean_token_accuracy": 0.15759438425302505,
"num_tokens": 9848536.0,
"step": 4855
},
{
"entropy": 6.081185388565063,
"epoch": 0.4247137988289784,
"grad_norm": 0.87109375,
"learning_rate": 0.0004987158699707717,
"loss": 5.9403,
"mean_token_accuracy": 0.15936280265450478,
"num_tokens": 9859741.0,
"step": 4860
},
{
"entropy": 6.154061698913575,
"epoch": 0.4251507471816831,
"grad_norm": 1.015625,
"learning_rate": 0.0004987125433703554,
"loss": 5.8932,
"mean_token_accuracy": 0.15802878588438035,
"num_tokens": 9869173.0,
"step": 4865
},
{
"entropy": 6.140933799743652,
"epoch": 0.42558769553438786,
"grad_norm": 0.8828125,
"learning_rate": 0.0004987092124790155,
"loss": 5.9152,
"mean_token_accuracy": 0.15708720907568932,
"num_tokens": 9879619.0,
"step": 4870
},
{
"entropy": 6.073929786682129,
"epoch": 0.42602464388709255,
"grad_norm": 0.8984375,
"learning_rate": 0.0004987058772968154,
"loss": 5.8475,
"mean_token_accuracy": 0.16600827127695084,
"num_tokens": 9890214.0,
"step": 4875
},
{
"entropy": 6.079315423965454,
"epoch": 0.42646159223979724,
"grad_norm": 0.8359375,
"learning_rate": 0.0004987025378238195,
"loss": 5.8829,
"mean_token_accuracy": 0.15276064574718476,
"num_tokens": 9900266.0,
"step": 4880
},
{
"entropy": 6.166507816314697,
"epoch": 0.426898540592502,
"grad_norm": 0.96875,
"learning_rate": 0.0004986991940600917,
"loss": 5.9015,
"mean_token_accuracy": 0.15680881440639496,
"num_tokens": 9910790.0,
"step": 4885
},
{
"entropy": 6.145452642440796,
"epoch": 0.4273354889452067,
"grad_norm": 0.92578125,
"learning_rate": 0.0004986958460056961,
"loss": 5.9196,
"mean_token_accuracy": 0.1490568682551384,
"num_tokens": 9921167.0,
"step": 4890
},
{
"entropy": 6.087120628356933,
"epoch": 0.42777243729791137,
"grad_norm": 0.90625,
"learning_rate": 0.000498692493660697,
"loss": 5.8626,
"mean_token_accuracy": 0.15905486345291137,
"num_tokens": 9931674.0,
"step": 4895
},
{
"entropy": 6.098935270309449,
"epoch": 0.4282093856506161,
"grad_norm": 0.9609375,
"learning_rate": 0.0004986891370251587,
"loss": 5.8165,
"mean_token_accuracy": 0.15766057521104812,
"num_tokens": 9941484.0,
"step": 4900
},
{
"entropy": 6.127752304077148,
"epoch": 0.4286463340033208,
"grad_norm": 0.90625,
"learning_rate": 0.0004986857760991454,
"loss": 5.9676,
"mean_token_accuracy": 0.15095995366573334,
"num_tokens": 9951885.0,
"step": 4905
},
{
"entropy": 6.1411848068237305,
"epoch": 0.4290832823560255,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986824108827218,
"loss": 5.7956,
"mean_token_accuracy": 0.16215234696865083,
"num_tokens": 9961658.0,
"step": 4910
},
{
"entropy": 5.999189138412476,
"epoch": 0.42952023070873024,
"grad_norm": 0.890625,
"learning_rate": 0.0004986790413759523,
"loss": 5.9074,
"mean_token_accuracy": 0.15364072769880294,
"num_tokens": 9971971.0,
"step": 4915
},
{
"entropy": 6.094682407379151,
"epoch": 0.42995717906143494,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986756675789016,
"loss": 5.8129,
"mean_token_accuracy": 0.15694656521081923,
"num_tokens": 9982211.0,
"step": 4920
},
{
"entropy": 6.112336492538452,
"epoch": 0.4303941274141396,
"grad_norm": 1.0,
"learning_rate": 0.0004986722894916344,
"loss": 5.8617,
"mean_token_accuracy": 0.16319485306739806,
"num_tokens": 9991935.0,
"step": 4925
},
{
"entropy": 6.052273893356324,
"epoch": 0.4308310757668444,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986689071142155,
"loss": 5.8917,
"mean_token_accuracy": 0.15127992033958435,
"num_tokens": 10002014.0,
"step": 4930
},
{
"entropy": 6.119452953338623,
"epoch": 0.43126802411954906,
"grad_norm": 1.0625,
"learning_rate": 0.0004986655204467097,
"loss": 5.9239,
"mean_token_accuracy": 0.15334465056657792,
"num_tokens": 10012030.0,
"step": 4935
},
{
"entropy": 6.091619825363159,
"epoch": 0.43170497247225376,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986621294891822,
"loss": 5.8779,
"mean_token_accuracy": 0.15746373534202576,
"num_tokens": 10022188.0,
"step": 4940
},
{
"entropy": 6.132827997207642,
"epoch": 0.4321419208249585,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986587342416976,
"loss": 5.7939,
"mean_token_accuracy": 0.16190674751996995,
"num_tokens": 10032131.0,
"step": 4945
},
{
"entropy": 6.1291422843933105,
"epoch": 0.4325788691776632,
"grad_norm": 0.8671875,
"learning_rate": 0.0004986553347043214,
"loss": 5.841,
"mean_token_accuracy": 0.15926993489265442,
"num_tokens": 10043204.0,
"step": 4950
},
{
"entropy": 6.077534294128418,
"epoch": 0.4330158175303679,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986519308771188,
"loss": 5.9102,
"mean_token_accuracy": 0.15396561622619628,
"num_tokens": 10052863.0,
"step": 4955
},
{
"entropy": 6.091691589355468,
"epoch": 0.43345276588307263,
"grad_norm": 0.83984375,
"learning_rate": 0.0004986485227601548,
"loss": 5.8971,
"mean_token_accuracy": 0.16426312774419785,
"num_tokens": 10063212.0,
"step": 4960
},
{
"entropy": 6.135089302062989,
"epoch": 0.4338897142357773,
"grad_norm": 0.8359375,
"learning_rate": 0.000498645110353495,
"loss": 5.9125,
"mean_token_accuracy": 0.1550646737217903,
"num_tokens": 10074180.0,
"step": 4965
},
{
"entropy": 6.034312391281128,
"epoch": 0.434326662588482,
"grad_norm": 0.87109375,
"learning_rate": 0.0004986416936572049,
"loss": 5.8586,
"mean_token_accuracy": 0.16182521730661392,
"num_tokens": 10084513.0,
"step": 4970
},
{
"entropy": 6.12773118019104,
"epoch": 0.43476361094118676,
"grad_norm": 0.91015625,
"learning_rate": 0.0004986382726713497,
"loss": 5.8815,
"mean_token_accuracy": 0.15795285105705262,
"num_tokens": 10094352.0,
"step": 4975
},
{
"entropy": 6.01217246055603,
"epoch": 0.43520055929389145,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986348473959955,
"loss": 5.7399,
"mean_token_accuracy": 0.17221405506134033,
"num_tokens": 10103677.0,
"step": 4980
},
{
"entropy": 6.117711782455444,
"epoch": 0.43563750764659614,
"grad_norm": 0.953125,
"learning_rate": 0.0004986314178312076,
"loss": 5.9636,
"mean_token_accuracy": 0.14685683250427245,
"num_tokens": 10113479.0,
"step": 4985
},
{
"entropy": 6.107839775085449,
"epoch": 0.4360744559993009,
"grad_norm": 0.92578125,
"learning_rate": 0.0004986279839770519,
"loss": 5.8576,
"mean_token_accuracy": 0.15548283010721206,
"num_tokens": 10123526.0,
"step": 4990
},
{
"entropy": 6.107575511932373,
"epoch": 0.4365114043520056,
"grad_norm": 0.9453125,
"learning_rate": 0.0004986245458335944,
"loss": 5.8764,
"mean_token_accuracy": 0.15630691051483153,
"num_tokens": 10134176.0,
"step": 4995
},
{
"entropy": 6.0929351329803465,
"epoch": 0.43694835270471033,
"grad_norm": 0.90625,
"learning_rate": 0.0004986211034009008,
"loss": 5.8435,
"mean_token_accuracy": 0.1651988849043846,
"num_tokens": 10144763.0,
"step": 5000
},
{
"entropy": 6.047188711166382,
"epoch": 0.437385301057415,
"grad_norm": 0.94921875,
"learning_rate": 0.0004986176566790375,
"loss": 5.8092,
"mean_token_accuracy": 0.16798071563243866,
"num_tokens": 10155011.0,
"step": 5005
},
{
"entropy": 6.10062165260315,
"epoch": 0.4378222494101197,
"grad_norm": 0.875,
"learning_rate": 0.0004986142056680701,
"loss": 5.9057,
"mean_token_accuracy": 0.15978446751832961,
"num_tokens": 10165474.0,
"step": 5010
},
{
"entropy": 6.080620527267456,
"epoch": 0.43825919776282446,
"grad_norm": 0.91796875,
"learning_rate": 0.0004986107503680652,
"loss": 5.9249,
"mean_token_accuracy": 0.15882788896560668,
"num_tokens": 10175137.0,
"step": 5015
},
{
"entropy": 6.098481130599976,
"epoch": 0.43869614611552915,
"grad_norm": 1.0,
"learning_rate": 0.0004986072907790889,
"loss": 5.9203,
"mean_token_accuracy": 0.1513497620820999,
"num_tokens": 10185623.0,
"step": 5020
},
{
"entropy": 6.172309303283692,
"epoch": 0.43913309446823384,
"grad_norm": 0.84765625,
"learning_rate": 0.0004986038269012076,
"loss": 5.8909,
"mean_token_accuracy": 0.1573339194059372,
"num_tokens": 10195201.0,
"step": 5025
},
{
"entropy": 6.099797010421753,
"epoch": 0.4395700428209386,
"grad_norm": 0.87890625,
"learning_rate": 0.0004986003587344877,
"loss": 5.9272,
"mean_token_accuracy": 0.15723302960395813,
"num_tokens": 10205951.0,
"step": 5030
},
{
"entropy": 6.0694938659667965,
"epoch": 0.4400069911736433,
"grad_norm": 0.97265625,
"learning_rate": 0.0004985968862789957,
"loss": 5.7995,
"mean_token_accuracy": 0.1665978655219078,
"num_tokens": 10215665.0,
"step": 5035
},
{
"entropy": 6.0943746089935305,
"epoch": 0.44044393952634797,
"grad_norm": 0.88671875,
"learning_rate": 0.0004985934095347985,
"loss": 5.8876,
"mean_token_accuracy": 0.1551556296646595,
"num_tokens": 10225037.0,
"step": 5040
},
{
"entropy": 6.0550844192504885,
"epoch": 0.4408808878790527,
"grad_norm": 0.91796875,
"learning_rate": 0.0004985899285019624,
"loss": 5.8357,
"mean_token_accuracy": 0.15570665746927262,
"num_tokens": 10234581.0,
"step": 5045
},
{
"entropy": 6.072161102294922,
"epoch": 0.4413178362317574,
"grad_norm": 0.9609375,
"learning_rate": 0.0004985864431805543,
"loss": 5.7679,
"mean_token_accuracy": 0.16553371846675874,
"num_tokens": 10243716.0,
"step": 5050
},
{
"entropy": 6.05171570777893,
"epoch": 0.4417547845844621,
"grad_norm": 0.96484375,
"learning_rate": 0.0004985829535706411,
"loss": 5.8731,
"mean_token_accuracy": 0.15933090895414354,
"num_tokens": 10253198.0,
"step": 5055
},
{
"entropy": 6.05954155921936,
"epoch": 0.44219173293716685,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985794596722897,
"loss": 5.9501,
"mean_token_accuracy": 0.15209509134292604,
"num_tokens": 10262912.0,
"step": 5060
},
{
"entropy": 6.088939428329468,
"epoch": 0.44262868128987154,
"grad_norm": 0.90234375,
"learning_rate": 0.000498575961485567,
"loss": 5.8864,
"mean_token_accuracy": 0.15810550898313522,
"num_tokens": 10272960.0,
"step": 5065
},
{
"entropy": 6.16075382232666,
"epoch": 0.44306562964257623,
"grad_norm": 0.9453125,
"learning_rate": 0.0004985724590105403,
"loss": 5.9316,
"mean_token_accuracy": 0.15536672621965408,
"num_tokens": 10283034.0,
"step": 5070
},
{
"entropy": 6.0558013916015625,
"epoch": 0.443502577995281,
"grad_norm": 0.953125,
"learning_rate": 0.0004985689522472767,
"loss": 5.8871,
"mean_token_accuracy": 0.15896483659744262,
"num_tokens": 10292553.0,
"step": 5075
},
{
"entropy": 6.028521251678467,
"epoch": 0.44393952634798567,
"grad_norm": 0.86328125,
"learning_rate": 0.0004985654411958434,
"loss": 5.853,
"mean_token_accuracy": 0.16116765141487122,
"num_tokens": 10302475.0,
"step": 5080
},
{
"entropy": 6.089235115051269,
"epoch": 0.44437647470069036,
"grad_norm": 0.91015625,
"learning_rate": 0.0004985619258563078,
"loss": 5.8595,
"mean_token_accuracy": 0.1636977016925812,
"num_tokens": 10311693.0,
"step": 5085
},
{
"entropy": 6.120714521408081,
"epoch": 0.4448134230533951,
"grad_norm": 0.90234375,
"learning_rate": 0.0004985584062287372,
"loss": 5.8539,
"mean_token_accuracy": 0.16536978930234908,
"num_tokens": 10323052.0,
"step": 5090
},
{
"entropy": 6.062410688400268,
"epoch": 0.4452503714060998,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985548823131994,
"loss": 5.8291,
"mean_token_accuracy": 0.16429725140333176,
"num_tokens": 10332997.0,
"step": 5095
},
{
"entropy": 6.1553514957427975,
"epoch": 0.4456873197588045,
"grad_norm": 1.0,
"learning_rate": 0.0004985513541097618,
"loss": 5.9593,
"mean_token_accuracy": 0.15591451823711394,
"num_tokens": 10343298.0,
"step": 5100
},
{
"entropy": 6.171523571014404,
"epoch": 0.44612426811150924,
"grad_norm": 0.94921875,
"learning_rate": 0.000498547821618492,
"loss": 5.8499,
"mean_token_accuracy": 0.15821425765752792,
"num_tokens": 10352485.0,
"step": 5105
},
{
"entropy": 5.98296799659729,
"epoch": 0.44656121646421393,
"grad_norm": 0.8984375,
"learning_rate": 0.0004985442848394579,
"loss": 5.9016,
"mean_token_accuracy": 0.15563226342201233,
"num_tokens": 10362810.0,
"step": 5110
},
{
"entropy": 6.0878349304199215,
"epoch": 0.4469981648169186,
"grad_norm": 0.91796875,
"learning_rate": 0.0004985407437727273,
"loss": 5.8386,
"mean_token_accuracy": 0.16296190321445464,
"num_tokens": 10372817.0,
"step": 5115
},
{
"entropy": 6.139986419677735,
"epoch": 0.44743511316962337,
"grad_norm": 1.0,
"learning_rate": 0.0004985371984183681,
"loss": 5.861,
"mean_token_accuracy": 0.16705801486968994,
"num_tokens": 10383046.0,
"step": 5120
},
{
"entropy": 6.054336261749268,
"epoch": 0.44787206152232806,
"grad_norm": 0.90625,
"learning_rate": 0.0004985336487764483,
"loss": 5.9406,
"mean_token_accuracy": 0.15005630254745483,
"num_tokens": 10393498.0,
"step": 5125
},
{
"entropy": 6.110240030288696,
"epoch": 0.44830900987503275,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985300948470359,
"loss": 5.9147,
"mean_token_accuracy": 0.15699672549962998,
"num_tokens": 10403595.0,
"step": 5130
},
{
"entropy": 6.111717271804809,
"epoch": 0.4487459582277375,
"grad_norm": 0.8515625,
"learning_rate": 0.0004985265366301992,
"loss": 5.8574,
"mean_token_accuracy": 0.15583672225475312,
"num_tokens": 10413475.0,
"step": 5135
},
{
"entropy": 6.064886283874512,
"epoch": 0.4491829065804422,
"grad_norm": 0.90234375,
"learning_rate": 0.0004985229741260065,
"loss": 5.8758,
"mean_token_accuracy": 0.1638584241271019,
"num_tokens": 10423512.0,
"step": 5140
},
{
"entropy": 6.083583545684815,
"epoch": 0.4496198549331469,
"grad_norm": 1.0546875,
"learning_rate": 0.0004985194073345259,
"loss": 5.8933,
"mean_token_accuracy": 0.15554327368736268,
"num_tokens": 10434198.0,
"step": 5145
},
{
"entropy": 6.109025287628174,
"epoch": 0.4500568032858516,
"grad_norm": 0.9453125,
"learning_rate": 0.0004985158362558261,
"loss": 5.8197,
"mean_token_accuracy": 0.1605295419692993,
"num_tokens": 10444077.0,
"step": 5150
},
{
"entropy": 6.064793586730957,
"epoch": 0.4504937516385563,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985122608899752,
"loss": 5.8867,
"mean_token_accuracy": 0.1586834102869034,
"num_tokens": 10454481.0,
"step": 5155
},
{
"entropy": 6.120577573776245,
"epoch": 0.450930699991261,
"grad_norm": 0.875,
"learning_rate": 0.0004985086812370422,
"loss": 5.8943,
"mean_token_accuracy": 0.16226042658090592,
"num_tokens": 10464453.0,
"step": 5160
},
{
"entropy": 6.134780406951904,
"epoch": 0.45136764834396575,
"grad_norm": 0.9375,
"learning_rate": 0.0004985050972970956,
"loss": 5.9678,
"mean_token_accuracy": 0.15403684377670288,
"num_tokens": 10474932.0,
"step": 5165
},
{
"entropy": 6.126642560958862,
"epoch": 0.45180459669667045,
"grad_norm": 0.91015625,
"learning_rate": 0.000498501509070204,
"loss": 5.9591,
"mean_token_accuracy": 0.1517454206943512,
"num_tokens": 10484719.0,
"step": 5170
},
{
"entropy": 6.129985237121582,
"epoch": 0.45224154504937514,
"grad_norm": 0.90625,
"learning_rate": 0.0004984979165564364,
"loss": 5.9909,
"mean_token_accuracy": 0.14962753057479858,
"num_tokens": 10495808.0,
"step": 5175
},
{
"entropy": 6.180895233154297,
"epoch": 0.4526784934020799,
"grad_norm": 0.91015625,
"learning_rate": 0.0004984943197558616,
"loss": 5.894,
"mean_token_accuracy": 0.15759684890508652,
"num_tokens": 10506528.0,
"step": 5180
},
{
"entropy": 6.100573635101318,
"epoch": 0.4531154417547846,
"grad_norm": 0.94140625,
"learning_rate": 0.0004984907186685487,
"loss": 5.8813,
"mean_token_accuracy": 0.16605969071388244,
"num_tokens": 10517117.0,
"step": 5185
},
{
"entropy": 6.166533660888672,
"epoch": 0.45355239010748927,
"grad_norm": 0.90625,
"learning_rate": 0.0004984871132945667,
"loss": 5.8792,
"mean_token_accuracy": 0.1546791136264801,
"num_tokens": 10527394.0,
"step": 5190
},
{
"entropy": 6.110329866409302,
"epoch": 0.453989338460194,
"grad_norm": 1.0546875,
"learning_rate": 0.0004984835036339848,
"loss": 5.8734,
"mean_token_accuracy": 0.15998434275388718,
"num_tokens": 10536779.0,
"step": 5195
},
{
"entropy": 6.09484977722168,
"epoch": 0.4544262868128987,
"grad_norm": 0.88671875,
"learning_rate": 0.0004984798896868722,
"loss": 5.95,
"mean_token_accuracy": 0.14918453395366668,
"num_tokens": 10547116.0,
"step": 5200
},
{
"entropy": 6.090693998336792,
"epoch": 0.45486323516560345,
"grad_norm": 0.95703125,
"learning_rate": 0.0004984762714532981,
"loss": 5.8305,
"mean_token_accuracy": 0.16339221000671386,
"num_tokens": 10556396.0,
"step": 5205
},
{
"entropy": 6.060329151153565,
"epoch": 0.45530018351830814,
"grad_norm": 0.9140625,
"learning_rate": 0.000498472648933332,
"loss": 6.017,
"mean_token_accuracy": 0.15394461452960967,
"num_tokens": 10568127.0,
"step": 5210
},
{
"entropy": 6.157448148727417,
"epoch": 0.45573713187101283,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984690221270436,
"loss": 5.8351,
"mean_token_accuracy": 0.163652141392231,
"num_tokens": 10578565.0,
"step": 5215
},
{
"entropy": 6.087811136245728,
"epoch": 0.4561740802237176,
"grad_norm": 0.91796875,
"learning_rate": 0.0004984653910345021,
"loss": 5.8763,
"mean_token_accuracy": 0.14887666031718255,
"num_tokens": 10588633.0,
"step": 5220
},
{
"entropy": 6.125527667999267,
"epoch": 0.4566110285764223,
"grad_norm": 0.9140625,
"learning_rate": 0.0004984617556557774,
"loss": 5.8284,
"mean_token_accuracy": 0.16143705323338509,
"num_tokens": 10597917.0,
"step": 5225
},
{
"entropy": 6.145944738388062,
"epoch": 0.45704797692912696,
"grad_norm": 0.90234375,
"learning_rate": 0.000498458115990939,
"loss": 5.8797,
"mean_token_accuracy": 0.1518850490450859,
"num_tokens": 10608396.0,
"step": 5230
},
{
"entropy": 6.045504951477051,
"epoch": 0.4574849252818317,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984544720400568,
"loss": 5.9012,
"mean_token_accuracy": 0.1609129011631012,
"num_tokens": 10618191.0,
"step": 5235
},
{
"entropy": 6.115693616867065,
"epoch": 0.4579218736345364,
"grad_norm": 0.9609375,
"learning_rate": 0.000498450823803201,
"loss": 5.8819,
"mean_token_accuracy": 0.15265067145228386,
"num_tokens": 10627594.0,
"step": 5240
},
{
"entropy": 6.037624931335449,
"epoch": 0.4583588219872411,
"grad_norm": 0.85546875,
"learning_rate": 0.0004984471712804412,
"loss": 5.8556,
"mean_token_accuracy": 0.16447641402482988,
"num_tokens": 10637260.0,
"step": 5245
},
{
"entropy": 6.126225137710572,
"epoch": 0.45879577033994584,
"grad_norm": 0.9609375,
"learning_rate": 0.0004984435144718475,
"loss": 5.8806,
"mean_token_accuracy": 0.16061059087514878,
"num_tokens": 10647968.0,
"step": 5250
},
{
"entropy": 6.047690582275391,
"epoch": 0.45923271869265053,
"grad_norm": 0.98828125,
"learning_rate": 0.0004984398533774901,
"loss": 5.8283,
"mean_token_accuracy": 0.1624503865838051,
"num_tokens": 10657760.0,
"step": 5255
},
{
"entropy": 6.077875328063965,
"epoch": 0.4596696670453552,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984361879974393,
"loss": 5.8548,
"mean_token_accuracy": 0.15239312797784804,
"num_tokens": 10667577.0,
"step": 5260
},
{
"entropy": 6.071667623519898,
"epoch": 0.46010661539805997,
"grad_norm": 0.8359375,
"learning_rate": 0.0004984325183317654,
"loss": 5.8095,
"mean_token_accuracy": 0.15807173699140548,
"num_tokens": 10677984.0,
"step": 5265
},
{
"entropy": 6.0686726570129395,
"epoch": 0.46054356375076466,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984288443805386,
"loss": 5.8646,
"mean_token_accuracy": 0.16376810520887375,
"num_tokens": 10687700.0,
"step": 5270
},
{
"entropy": 6.089095544815064,
"epoch": 0.46098051210346935,
"grad_norm": 0.92578125,
"learning_rate": 0.0004984251661438295,
"loss": 5.8463,
"mean_token_accuracy": 0.16278012990951538,
"num_tokens": 10697351.0,
"step": 5275
},
{
"entropy": 6.053745889663697,
"epoch": 0.4614174604561741,
"grad_norm": 1.125,
"learning_rate": 0.0004984214836217087,
"loss": 5.876,
"mean_token_accuracy": 0.1601707324385643,
"num_tokens": 10707498.0,
"step": 5280
},
{
"entropy": 6.080528354644775,
"epoch": 0.4618544088088788,
"grad_norm": 0.98828125,
"learning_rate": 0.0004984177968142467,
"loss": 5.9022,
"mean_token_accuracy": 0.1559564083814621,
"num_tokens": 10717724.0,
"step": 5285
},
{
"entropy": 6.118378019332885,
"epoch": 0.4622913571615835,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984141057215143,
"loss": 5.864,
"mean_token_accuracy": 0.15625895708799362,
"num_tokens": 10727030.0,
"step": 5290
},
{
"entropy": 6.041504955291748,
"epoch": 0.46272830551428823,
"grad_norm": 0.98046875,
"learning_rate": 0.0004984104103435824,
"loss": 5.7468,
"mean_token_accuracy": 0.16861266046762466,
"num_tokens": 10736861.0,
"step": 5295
},
{
"entropy": 6.121583938598633,
"epoch": 0.4631652538669929,
"grad_norm": 0.87109375,
"learning_rate": 0.0004984067106805217,
"loss": 5.9697,
"mean_token_accuracy": 0.15583069622516632,
"num_tokens": 10748310.0,
"step": 5300
},
{
"entropy": 6.073477172851563,
"epoch": 0.4636022022196976,
"grad_norm": 1.0,
"learning_rate": 0.0004984030067324032,
"loss": 5.8063,
"mean_token_accuracy": 0.1636292114853859,
"num_tokens": 10758110.0,
"step": 5305
},
{
"entropy": 6.064459705352784,
"epoch": 0.46403915057240236,
"grad_norm": 0.9765625,
"learning_rate": 0.0004983992984992979,
"loss": 5.8569,
"mean_token_accuracy": 0.16498022079467772,
"num_tokens": 10767919.0,
"step": 5310
},
{
"entropy": 6.065229988098144,
"epoch": 0.46447609892510705,
"grad_norm": 0.95703125,
"learning_rate": 0.0004983955859812771,
"loss": 5.7786,
"mean_token_accuracy": 0.1648388147354126,
"num_tokens": 10776888.0,
"step": 5315
},
{
"entropy": 6.072467899322509,
"epoch": 0.46491304727781174,
"grad_norm": 0.9765625,
"learning_rate": 0.0004983918691784119,
"loss": 5.8742,
"mean_token_accuracy": 0.1590124249458313,
"num_tokens": 10786806.0,
"step": 5320
},
{
"entropy": 6.10413556098938,
"epoch": 0.4653499956305165,
"grad_norm": 0.9140625,
"learning_rate": 0.0004983881480907736,
"loss": 5.8477,
"mean_token_accuracy": 0.1579391196370125,
"num_tokens": 10796452.0,
"step": 5325
},
{
"entropy": 6.104733037948608,
"epoch": 0.4657869439832212,
"grad_norm": 0.97265625,
"learning_rate": 0.0004983844227184334,
"loss": 5.8239,
"mean_token_accuracy": 0.15865322947502136,
"num_tokens": 10806718.0,
"step": 5330
},
{
"entropy": 6.097746181488037,
"epoch": 0.46622389233592587,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983806930614631,
"loss": 5.9065,
"mean_token_accuracy": 0.1553451269865036,
"num_tokens": 10817978.0,
"step": 5335
},
{
"entropy": 6.0263652324676515,
"epoch": 0.4666608406886306,
"grad_norm": 0.8359375,
"learning_rate": 0.000498376959119934,
"loss": 5.7833,
"mean_token_accuracy": 0.1663276508450508,
"num_tokens": 10827818.0,
"step": 5340
},
{
"entropy": 6.1345278263092045,
"epoch": 0.4670977890413353,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983732208939178,
"loss": 5.9218,
"mean_token_accuracy": 0.1574622869491577,
"num_tokens": 10838492.0,
"step": 5345
},
{
"entropy": 6.077729845046997,
"epoch": 0.46753473739404,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983694783834863,
"loss": 5.825,
"mean_token_accuracy": 0.15848338454961777,
"num_tokens": 10847807.0,
"step": 5350
},
{
"entropy": 6.045288324356079,
"epoch": 0.46797168574674475,
"grad_norm": 0.859375,
"learning_rate": 0.0004983657315887109,
"loss": 5.9006,
"mean_token_accuracy": 0.15541983991861344,
"num_tokens": 10858637.0,
"step": 5355
},
{
"entropy": 6.122717761993409,
"epoch": 0.46840863409944944,
"grad_norm": 1.03125,
"learning_rate": 0.0004983619805096639,
"loss": 5.8334,
"mean_token_accuracy": 0.1633836179971695,
"num_tokens": 10868042.0,
"step": 5360
},
{
"entropy": 6.083197355270386,
"epoch": 0.46884558245215413,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983582251464171,
"loss": 5.8429,
"mean_token_accuracy": 0.1656041830778122,
"num_tokens": 10877171.0,
"step": 5365
},
{
"entropy": 6.062127876281738,
"epoch": 0.4692825308048589,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983544654990423,
"loss": 5.8458,
"mean_token_accuracy": 0.15868037939071655,
"num_tokens": 10888434.0,
"step": 5370
},
{
"entropy": 6.090228271484375,
"epoch": 0.46971947915756357,
"grad_norm": 0.97265625,
"learning_rate": 0.000498350701567612,
"loss": 5.8758,
"mean_token_accuracy": 0.15781670212745666,
"num_tokens": 10898467.0,
"step": 5375
},
{
"entropy": 5.993190670013428,
"epoch": 0.47015642751026826,
"grad_norm": 0.98046875,
"learning_rate": 0.0004983469333521983,
"loss": 5.8475,
"mean_token_accuracy": 0.16673893183469773,
"num_tokens": 10908178.0,
"step": 5380
},
{
"entropy": 6.0736956119537355,
"epoch": 0.470593375862973,
"grad_norm": 0.8125,
"learning_rate": 0.0004983431608528732,
"loss": 5.9307,
"mean_token_accuracy": 0.1510281190276146,
"num_tokens": 10919957.0,
"step": 5385
},
{
"entropy": 6.17218861579895,
"epoch": 0.4710303242156777,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983393840697092,
"loss": 5.8789,
"mean_token_accuracy": 0.1570976510643959,
"num_tokens": 10929728.0,
"step": 5390
},
{
"entropy": 6.0762652397155765,
"epoch": 0.4714672725683824,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983356030027789,
"loss": 5.9169,
"mean_token_accuracy": 0.1537831172347069,
"num_tokens": 10940427.0,
"step": 5395
},
{
"entropy": 5.99635124206543,
"epoch": 0.47190422092108714,
"grad_norm": 0.890625,
"learning_rate": 0.0004983318176521546,
"loss": 5.8623,
"mean_token_accuracy": 0.16047628670930864,
"num_tokens": 10950641.0,
"step": 5400
},
{
"entropy": 6.131054782867432,
"epoch": 0.4723411692737918,
"grad_norm": 1.0625,
"learning_rate": 0.0004983280280179092,
"loss": 5.828,
"mean_token_accuracy": 0.16604136526584626,
"num_tokens": 10959341.0,
"step": 5405
},
{
"entropy": 6.076440000534058,
"epoch": 0.4727781176264966,
"grad_norm": 0.95703125,
"learning_rate": 0.000498324234100115,
"loss": 5.8583,
"mean_token_accuracy": 0.16322903633117675,
"num_tokens": 10967745.0,
"step": 5410
},
{
"entropy": 6.018764305114746,
"epoch": 0.47321506597920127,
"grad_norm": 0.921875,
"learning_rate": 0.0004983204358988451,
"loss": 5.8205,
"mean_token_accuracy": 0.16208201944828032,
"num_tokens": 10978345.0,
"step": 5415
},
{
"entropy": 6.096686792373657,
"epoch": 0.47365201433190596,
"grad_norm": 0.87109375,
"learning_rate": 0.0004983166334141722,
"loss": 5.8818,
"mean_token_accuracy": 0.15515933856368064,
"num_tokens": 10988030.0,
"step": 5420
},
{
"entropy": 6.145278215408325,
"epoch": 0.4740889626846107,
"grad_norm": 0.8203125,
"learning_rate": 0.0004983128266461693,
"loss": 5.8527,
"mean_token_accuracy": 0.16152588576078414,
"num_tokens": 10999169.0,
"step": 5425
},
{
"entropy": 6.05873441696167,
"epoch": 0.4745259110373154,
"grad_norm": 0.9609375,
"learning_rate": 0.0004983090155949093,
"loss": 5.9396,
"mean_token_accuracy": 0.1501481741666794,
"num_tokens": 11009608.0,
"step": 5430
},
{
"entropy": 6.13284592628479,
"epoch": 0.4749628593900201,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983052002604653,
"loss": 5.8727,
"mean_token_accuracy": 0.15617259219288826,
"num_tokens": 11020187.0,
"step": 5435
},
{
"entropy": 6.095590591430664,
"epoch": 0.47539980774272483,
"grad_norm": 0.890625,
"learning_rate": 0.0004983013806429107,
"loss": 5.8371,
"mean_token_accuracy": 0.16051960587501526,
"num_tokens": 11030459.0,
"step": 5440
},
{
"entropy": 6.00195426940918,
"epoch": 0.4758367560954295,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982975567423185,
"loss": 5.7936,
"mean_token_accuracy": 0.1667706102132797,
"num_tokens": 11040880.0,
"step": 5445
},
{
"entropy": 6.054232740402222,
"epoch": 0.4762737044481342,
"grad_norm": 0.984375,
"learning_rate": 0.0004982937285587623,
"loss": 5.8042,
"mean_token_accuracy": 0.16394554674625397,
"num_tokens": 11050901.0,
"step": 5450
},
{
"entropy": 6.088509798049927,
"epoch": 0.47671065280083896,
"grad_norm": 0.89453125,
"learning_rate": 0.0004982898960923152,
"loss": 5.9762,
"mean_token_accuracy": 0.14890825003385544,
"num_tokens": 11061274.0,
"step": 5455
},
{
"entropy": 6.127956914901733,
"epoch": 0.47714760115354365,
"grad_norm": 1.0,
"learning_rate": 0.0004982860593430509,
"loss": 5.794,
"mean_token_accuracy": 0.15740275830030442,
"num_tokens": 11070644.0,
"step": 5460
},
{
"entropy": 6.042313289642334,
"epoch": 0.47758454950624835,
"grad_norm": 0.875,
"learning_rate": 0.0004982822183110431,
"loss": 5.8447,
"mean_token_accuracy": 0.1646100953221321,
"num_tokens": 11081154.0,
"step": 5465
},
{
"entropy": 6.081279993057251,
"epoch": 0.4780214978589531,
"grad_norm": 0.9921875,
"learning_rate": 0.0004982783729963654,
"loss": 5.9438,
"mean_token_accuracy": 0.15628594309091567,
"num_tokens": 11091929.0,
"step": 5470
},
{
"entropy": 6.000416278839111,
"epoch": 0.4784584462116578,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982745233990914,
"loss": 5.7724,
"mean_token_accuracy": 0.17148518860340117,
"num_tokens": 11102075.0,
"step": 5475
},
{
"entropy": 6.0447105884552,
"epoch": 0.4788953945643625,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982706695192949,
"loss": 5.8729,
"mean_token_accuracy": 0.15767095237970352,
"num_tokens": 11111886.0,
"step": 5480
},
{
"entropy": 6.1248678207397464,
"epoch": 0.4793323429170672,
"grad_norm": 0.98828125,
"learning_rate": 0.0004982668113570503,
"loss": 5.8597,
"mean_token_accuracy": 0.15638172328472139,
"num_tokens": 11122329.0,
"step": 5485
},
{
"entropy": 6.062728977203369,
"epoch": 0.4797692912697719,
"grad_norm": 0.98828125,
"learning_rate": 0.000498262948912431,
"loss": 5.7911,
"mean_token_accuracy": 0.1594435304403305,
"num_tokens": 11132348.0,
"step": 5490
},
{
"entropy": 6.077106475830078,
"epoch": 0.4802062396224766,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982590821855115,
"loss": 5.8575,
"mean_token_accuracy": 0.16328706592321396,
"num_tokens": 11141975.0,
"step": 5495
},
{
"entropy": 6.098625373840332,
"epoch": 0.48064318797518135,
"grad_norm": 0.953125,
"learning_rate": 0.0004982552111763657,
"loss": 5.8586,
"mean_token_accuracy": 0.15738974511623383,
"num_tokens": 11152174.0,
"step": 5500
},
{
"entropy": 6.052024221420288,
"epoch": 0.48108013632788604,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982513358850681,
"loss": 5.8605,
"mean_token_accuracy": 0.15670732408761978,
"num_tokens": 11162563.0,
"step": 5505
},
{
"entropy": 6.084334468841552,
"epoch": 0.48151708468059073,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982474563116928,
"loss": 5.7676,
"mean_token_accuracy": 0.16274628937244415,
"num_tokens": 11172572.0,
"step": 5510
},
{
"entropy": 6.056139326095581,
"epoch": 0.4819540330332955,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982435724563143,
"loss": 5.8529,
"mean_token_accuracy": 0.1674552872776985,
"num_tokens": 11183404.0,
"step": 5515
},
{
"entropy": 6.025302410125732,
"epoch": 0.4823909813860002,
"grad_norm": 0.87890625,
"learning_rate": 0.0004982396843190072,
"loss": 5.8477,
"mean_token_accuracy": 0.16038714200258256,
"num_tokens": 11193905.0,
"step": 5520
},
{
"entropy": 6.099229335784912,
"epoch": 0.48282792973870486,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982357918998459,
"loss": 5.7562,
"mean_token_accuracy": 0.17048548460006713,
"num_tokens": 11203127.0,
"step": 5525
},
{
"entropy": 6.02909369468689,
"epoch": 0.4832648780914096,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982318951989053,
"loss": 5.7721,
"mean_token_accuracy": 0.16809311956167222,
"num_tokens": 11213468.0,
"step": 5530
},
{
"entropy": 6.023222398757935,
"epoch": 0.4837018264441143,
"grad_norm": 0.8984375,
"learning_rate": 0.0004982279942162597,
"loss": 5.8596,
"mean_token_accuracy": 0.16149110794067384,
"num_tokens": 11224722.0,
"step": 5535
},
{
"entropy": 6.100701427459716,
"epoch": 0.484138774796819,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982240889519844,
"loss": 5.8437,
"mean_token_accuracy": 0.16304720044136048,
"num_tokens": 11234565.0,
"step": 5540
},
{
"entropy": 6.084958171844482,
"epoch": 0.48457572314952374,
"grad_norm": 0.91015625,
"learning_rate": 0.000498220179406154,
"loss": 5.8372,
"mean_token_accuracy": 0.16213309168815612,
"num_tokens": 11244496.0,
"step": 5545
},
{
"entropy": 6.096549654006958,
"epoch": 0.48501267150222843,
"grad_norm": 0.87890625,
"learning_rate": 0.0004982162655788436,
"loss": 5.8194,
"mean_token_accuracy": 0.16341753602027892,
"num_tokens": 11254187.0,
"step": 5550
},
{
"entropy": 6.047856283187866,
"epoch": 0.4854496198549331,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982123474701283,
"loss": 5.7992,
"mean_token_accuracy": 0.16489096730947495,
"num_tokens": 11264580.0,
"step": 5555
},
{
"entropy": 6.067271184921265,
"epoch": 0.48588656820763787,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982084250800832,
"loss": 5.9469,
"mean_token_accuracy": 0.15531962588429452,
"num_tokens": 11275104.0,
"step": 5560
},
{
"entropy": 6.1391197681427006,
"epoch": 0.48632351656034256,
"grad_norm": 0.9921875,
"learning_rate": 0.0004982044984087836,
"loss": 5.9635,
"mean_token_accuracy": 0.1547545611858368,
"num_tokens": 11285983.0,
"step": 5565
},
{
"entropy": 6.03480339050293,
"epoch": 0.48676046491304725,
"grad_norm": 0.94140625,
"learning_rate": 0.0004982005674563048,
"loss": 5.8322,
"mean_token_accuracy": 0.16253410577774047,
"num_tokens": 11295378.0,
"step": 5570
},
{
"entropy": 6.071707391738892,
"epoch": 0.487197413265752,
"grad_norm": 0.95703125,
"learning_rate": 0.000498196632222722,
"loss": 5.9392,
"mean_token_accuracy": 0.15828753113746644,
"num_tokens": 11305500.0,
"step": 5575
},
{
"entropy": 6.089603424072266,
"epoch": 0.4876343616184567,
"grad_norm": 0.875,
"learning_rate": 0.0004981926927081109,
"loss": 5.8472,
"mean_token_accuracy": 0.1600751742720604,
"num_tokens": 11316455.0,
"step": 5580
},
{
"entropy": 6.074129629135132,
"epoch": 0.4880713099711614,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981887489125471,
"loss": 5.8112,
"mean_token_accuracy": 0.1575797788798809,
"num_tokens": 11325865.0,
"step": 5585
},
{
"entropy": 5.981272888183594,
"epoch": 0.48850825832386613,
"grad_norm": 0.86328125,
"learning_rate": 0.000498184800836106,
"loss": 5.7768,
"mean_token_accuracy": 0.17487731128931044,
"num_tokens": 11337368.0,
"step": 5590
},
{
"entropy": 6.0062541484832765,
"epoch": 0.4889452066765708,
"grad_norm": 0.8828125,
"learning_rate": 0.0004981808484788635,
"loss": 5.7523,
"mean_token_accuracy": 0.17290766686201095,
"num_tokens": 11347129.0,
"step": 5595
},
{
"entropy": 6.070178651809693,
"epoch": 0.48938215502927557,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981768918408955,
"loss": 5.831,
"mean_token_accuracy": 0.15914562493562698,
"num_tokens": 11356641.0,
"step": 5600
},
{
"entropy": 6.019480848312378,
"epoch": 0.48981910338198026,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981729309222777,
"loss": 5.7193,
"mean_token_accuracy": 0.17006878405809403,
"num_tokens": 11366485.0,
"step": 5605
},
{
"entropy": 6.009759902954102,
"epoch": 0.49025605173468495,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981689657230861,
"loss": 5.8558,
"mean_token_accuracy": 0.16212911680340766,
"num_tokens": 11376305.0,
"step": 5610
},
{
"entropy": 6.054521274566651,
"epoch": 0.4906930000873897,
"grad_norm": 0.8828125,
"learning_rate": 0.000498164996243397,
"loss": 5.8259,
"mean_token_accuracy": 0.16426292508840562,
"num_tokens": 11385837.0,
"step": 5615
},
{
"entropy": 6.0769446849822994,
"epoch": 0.4911299484400944,
"grad_norm": 0.921875,
"learning_rate": 0.0004981610224832861,
"loss": 5.8674,
"mean_token_accuracy": 0.16243572831153869,
"num_tokens": 11396719.0,
"step": 5620
},
{
"entropy": 6.109344911575318,
"epoch": 0.4915668967927991,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981570444428299,
"loss": 5.8794,
"mean_token_accuracy": 0.15813525915145873,
"num_tokens": 11408107.0,
"step": 5625
},
{
"entropy": 6.026615953445434,
"epoch": 0.4920038451455038,
"grad_norm": 0.89453125,
"learning_rate": 0.0004981530621221048,
"loss": 5.7906,
"mean_token_accuracy": 0.17031390815973282,
"num_tokens": 11417547.0,
"step": 5630
},
{
"entropy": 6.092897939682007,
"epoch": 0.4924407934982085,
"grad_norm": 0.953125,
"learning_rate": 0.0004981490755211869,
"loss": 5.8722,
"mean_token_accuracy": 0.16181053072214127,
"num_tokens": 11427660.0,
"step": 5635
},
{
"entropy": 5.994324016571045,
"epoch": 0.4928777418509132,
"grad_norm": 0.8671875,
"learning_rate": 0.0004981450846401528,
"loss": 5.7465,
"mean_token_accuracy": 0.16703419536352157,
"num_tokens": 11438032.0,
"step": 5640
},
{
"entropy": 6.065396118164062,
"epoch": 0.49331469020361796,
"grad_norm": 0.94921875,
"learning_rate": 0.000498141089479079,
"loss": 5.8522,
"mean_token_accuracy": 0.15965607166290283,
"num_tokens": 11448987.0,
"step": 5645
},
{
"entropy": 6.04461407661438,
"epoch": 0.49375163855632265,
"grad_norm": 0.9453125,
"learning_rate": 0.0004981370900380423,
"loss": 5.8294,
"mean_token_accuracy": 0.16260455697774887,
"num_tokens": 11458659.0,
"step": 5650
},
{
"entropy": 6.053560400009156,
"epoch": 0.49418858690902734,
"grad_norm": 0.96875,
"learning_rate": 0.0004981330863171194,
"loss": 5.8066,
"mean_token_accuracy": 0.16022174954414367,
"num_tokens": 11468104.0,
"step": 5655
},
{
"entropy": 6.050134658813477,
"epoch": 0.4946255352617321,
"grad_norm": 1.0625,
"learning_rate": 0.0004981290783163867,
"loss": 5.7662,
"mean_token_accuracy": 0.16689103543758393,
"num_tokens": 11479579.0,
"step": 5660
},
{
"entropy": 6.049282073974609,
"epoch": 0.4950624836144368,
"grad_norm": 1.0234375,
"learning_rate": 0.0004981250660359216,
"loss": 5.8378,
"mean_token_accuracy": 0.1668689325451851,
"num_tokens": 11489202.0,
"step": 5665
},
{
"entropy": 6.028315734863281,
"epoch": 0.49549943196714147,
"grad_norm": 0.99609375,
"learning_rate": 0.0004981210494758007,
"loss": 5.8735,
"mean_token_accuracy": 0.15290959030389786,
"num_tokens": 11499425.0,
"step": 5670
},
{
"entropy": 6.09428915977478,
"epoch": 0.4959363803198462,
"grad_norm": 1.0,
"learning_rate": 0.000498117028636101,
"loss": 5.8559,
"mean_token_accuracy": 0.1578807756304741,
"num_tokens": 11508983.0,
"step": 5675
},
{
"entropy": 6.036450290679932,
"epoch": 0.4963733286725509,
"grad_norm": 0.87890625,
"learning_rate": 0.0004981130035169,
"loss": 5.8957,
"mean_token_accuracy": 0.1656028926372528,
"num_tokens": 11519791.0,
"step": 5680
},
{
"entropy": 6.058498096466065,
"epoch": 0.4968102770252556,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981089741182747,
"loss": 5.8084,
"mean_token_accuracy": 0.1662641003727913,
"num_tokens": 11530537.0,
"step": 5685
},
{
"entropy": 6.059791564941406,
"epoch": 0.49724722537796034,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981049404403023,
"loss": 5.8354,
"mean_token_accuracy": 0.16185742169618605,
"num_tokens": 11540591.0,
"step": 5690
},
{
"entropy": 5.9863921165466305,
"epoch": 0.49768417373066504,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981009024830602,
"loss": 5.7767,
"mean_token_accuracy": 0.16423231661319732,
"num_tokens": 11550719.0,
"step": 5695
},
{
"entropy": 6.011070585250854,
"epoch": 0.4981211220833697,
"grad_norm": 0.89453125,
"learning_rate": 0.000498096860246626,
"loss": 5.8403,
"mean_token_accuracy": 0.16285658478736878,
"num_tokens": 11561805.0,
"step": 5700
},
{
"entropy": 6.165259504318238,
"epoch": 0.4985580704360745,
"grad_norm": 0.91796875,
"learning_rate": 0.000498092813731077,
"loss": 5.8735,
"mean_token_accuracy": 0.1558795154094696,
"num_tokens": 11571380.0,
"step": 5705
},
{
"entropy": 6.012100601196289,
"epoch": 0.49899501878877917,
"grad_norm": 0.87109375,
"learning_rate": 0.000498088762936491,
"loss": 5.8465,
"mean_token_accuracy": 0.15960172265768052,
"num_tokens": 11580853.0,
"step": 5710
},
{
"entropy": 5.926440191268921,
"epoch": 0.49943196714148386,
"grad_norm": 1.1015625,
"learning_rate": 0.0004980847078629457,
"loss": 5.7472,
"mean_token_accuracy": 0.17658197581768037,
"num_tokens": 11591966.0,
"step": 5715
},
{
"entropy": 6.049602222442627,
"epoch": 0.4998689154941886,
"grad_norm": 0.9921875,
"learning_rate": 0.0004980806485105186,
"loss": 5.8151,
"mean_token_accuracy": 0.16555469185113908,
"num_tokens": 11601239.0,
"step": 5720
},
{
"entropy": 6.057469320297241,
"epoch": 0.5003058638468934,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980765848792879,
"loss": 5.737,
"mean_token_accuracy": 0.1763659179210663,
"num_tokens": 11609857.0,
"step": 5725
},
{
"entropy": 6.077443647384643,
"epoch": 0.500742812199598,
"grad_norm": 0.9140625,
"learning_rate": 0.0004980725169693314,
"loss": 5.793,
"mean_token_accuracy": 0.16501347720623016,
"num_tokens": 11618952.0,
"step": 5730
},
{
"entropy": 6.033372402191162,
"epoch": 0.5011797605523027,
"grad_norm": 0.921875,
"learning_rate": 0.0004980684447807272,
"loss": 5.7241,
"mean_token_accuracy": 0.17032039165496826,
"num_tokens": 11627638.0,
"step": 5735
},
{
"entropy": 6.073156690597534,
"epoch": 0.5016167089050074,
"grad_norm": 0.90234375,
"learning_rate": 0.0004980643683135533,
"loss": 5.8382,
"mean_token_accuracy": 0.165236958861351,
"num_tokens": 11637842.0,
"step": 5740
},
{
"entropy": 6.023643064498901,
"epoch": 0.5020536572577121,
"grad_norm": 1.078125,
"learning_rate": 0.0004980602875678878,
"loss": 5.7428,
"mean_token_accuracy": 0.1751474380493164,
"num_tokens": 11647217.0,
"step": 5745
},
{
"entropy": 5.952885913848877,
"epoch": 0.5024906056104168,
"grad_norm": 0.90625,
"learning_rate": 0.0004980562025438093,
"loss": 5.7367,
"mean_token_accuracy": 0.16680333018302917,
"num_tokens": 11657558.0,
"step": 5750
},
{
"entropy": 6.0496491432189945,
"epoch": 0.5029275539631216,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980521132413958,
"loss": 5.749,
"mean_token_accuracy": 0.1631416290998459,
"num_tokens": 11668054.0,
"step": 5755
},
{
"entropy": 6.065122604370117,
"epoch": 0.5033645023158263,
"grad_norm": 0.9609375,
"learning_rate": 0.000498048019660726,
"loss": 5.7608,
"mean_token_accuracy": 0.1645866945385933,
"num_tokens": 11677779.0,
"step": 5760
},
{
"entropy": 6.0388609886169435,
"epoch": 0.503801450668531,
"grad_norm": 0.86328125,
"learning_rate": 0.0004980439218018783,
"loss": 5.8191,
"mean_token_accuracy": 0.16016570776700972,
"num_tokens": 11687577.0,
"step": 5765
},
{
"entropy": 6.143608713150025,
"epoch": 0.5042383990212357,
"grad_norm": 0.8125,
"learning_rate": 0.0004980398196649312,
"loss": 5.9225,
"mean_token_accuracy": 0.1562805950641632,
"num_tokens": 11699630.0,
"step": 5770
},
{
"entropy": 6.046047735214233,
"epoch": 0.5046753473739404,
"grad_norm": 0.89453125,
"learning_rate": 0.0004980357132499636,
"loss": 5.8515,
"mean_token_accuracy": 0.15963000506162645,
"num_tokens": 11710139.0,
"step": 5775
},
{
"entropy": 6.0711297512054445,
"epoch": 0.5051122957266451,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980316025570542,
"loss": 5.8034,
"mean_token_accuracy": 0.16327530294656753,
"num_tokens": 11719070.0,
"step": 5780
},
{
"entropy": 6.022385835647583,
"epoch": 0.5055492440793499,
"grad_norm": 0.8984375,
"learning_rate": 0.0004980274875862816,
"loss": 5.8419,
"mean_token_accuracy": 0.1634025901556015,
"num_tokens": 11728770.0,
"step": 5785
},
{
"entropy": 6.030874395370484,
"epoch": 0.5059861924320546,
"grad_norm": 0.98828125,
"learning_rate": 0.000498023368337725,
"loss": 5.7771,
"mean_token_accuracy": 0.16805996745824814,
"num_tokens": 11737753.0,
"step": 5790
},
{
"entropy": 5.990839719772339,
"epoch": 0.5064231407847593,
"grad_norm": 0.99609375,
"learning_rate": 0.0004980192448114635,
"loss": 5.7601,
"mean_token_accuracy": 0.1641977235674858,
"num_tokens": 11747837.0,
"step": 5795
},
{
"entropy": 6.0531830310821535,
"epoch": 0.5068600891374639,
"grad_norm": 0.91015625,
"learning_rate": 0.0004980151170075758,
"loss": 5.8853,
"mean_token_accuracy": 0.1629597380757332,
"num_tokens": 11757911.0,
"step": 5800
},
{
"entropy": 6.1099793910980225,
"epoch": 0.5072970374901686,
"grad_norm": 0.953125,
"learning_rate": 0.0004980109849261414,
"loss": 5.7496,
"mean_token_accuracy": 0.16389151364564897,
"num_tokens": 11766213.0,
"step": 5805
},
{
"entropy": 5.956259536743164,
"epoch": 0.5077339858428733,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980068485672394,
"loss": 5.7053,
"mean_token_accuracy": 0.17541376501321793,
"num_tokens": 11775640.0,
"step": 5810
},
{
"entropy": 6.052332258224487,
"epoch": 0.5081709341955781,
"grad_norm": 0.984375,
"learning_rate": 0.0004980027079309493,
"loss": 5.7603,
"mean_token_accuracy": 0.16962587982416152,
"num_tokens": 11785664.0,
"step": 5815
},
{
"entropy": 6.101211261749268,
"epoch": 0.5086078825482828,
"grad_norm": 1.0234375,
"learning_rate": 0.0004979985630173503,
"loss": 5.8708,
"mean_token_accuracy": 0.1568834140896797,
"num_tokens": 11795402.0,
"step": 5820
},
{
"entropy": 6.002501583099365,
"epoch": 0.5090448309009875,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979944138265221,
"loss": 5.7352,
"mean_token_accuracy": 0.16570198237895967,
"num_tokens": 11805923.0,
"step": 5825
},
{
"entropy": 6.021250009536743,
"epoch": 0.5094817792536922,
"grad_norm": 0.89453125,
"learning_rate": 0.0004979902603585442,
"loss": 5.8499,
"mean_token_accuracy": 0.159707410633564,
"num_tokens": 11816097.0,
"step": 5830
},
{
"entropy": 6.048093175888061,
"epoch": 0.5099187276063969,
"grad_norm": 0.87109375,
"learning_rate": 0.0004979861026134961,
"loss": 5.7511,
"mean_token_accuracy": 0.16532458364963531,
"num_tokens": 11826889.0,
"step": 5835
},
{
"entropy": 6.058287286758423,
"epoch": 0.5103556759591016,
"grad_norm": 0.86328125,
"learning_rate": 0.0004979819405914579,
"loss": 5.8224,
"mean_token_accuracy": 0.16387106627225875,
"num_tokens": 11837935.0,
"step": 5840
},
{
"entropy": 6.050278091430664,
"epoch": 0.5107926243118064,
"grad_norm": 0.9765625,
"learning_rate": 0.0004979777742925092,
"loss": 5.8071,
"mean_token_accuracy": 0.16187942028045654,
"num_tokens": 11847715.0,
"step": 5845
},
{
"entropy": 6.108582592010498,
"epoch": 0.5112295726645111,
"grad_norm": 0.92578125,
"learning_rate": 0.0004979736037167298,
"loss": 5.8401,
"mean_token_accuracy": 0.16293361485004426,
"num_tokens": 11857841.0,
"step": 5850
},
{
"entropy": 6.041381406784057,
"epoch": 0.5116665210172158,
"grad_norm": 0.9296875,
"learning_rate": 0.0004979694288642,
"loss": 5.828,
"mean_token_accuracy": 0.16747580021619796,
"num_tokens": 11868135.0,
"step": 5855
},
{
"entropy": 6.044627094268799,
"epoch": 0.5121034693699205,
"grad_norm": 0.9453125,
"learning_rate": 0.0004979652497349997,
"loss": 5.7595,
"mean_token_accuracy": 0.16045006960630417,
"num_tokens": 11877637.0,
"step": 5860
},
{
"entropy": 5.958197832107544,
"epoch": 0.5125404177226252,
"grad_norm": 0.93359375,
"learning_rate": 0.000497961066329209,
"loss": 5.8228,
"mean_token_accuracy": 0.1567137211561203,
"num_tokens": 11887128.0,
"step": 5865
},
{
"entropy": 6.115878343582153,
"epoch": 0.5129773660753298,
"grad_norm": 0.98046875,
"learning_rate": 0.0004979568786469083,
"loss": 5.8504,
"mean_token_accuracy": 0.1540878966450691,
"num_tokens": 11897885.0,
"step": 5870
},
{
"entropy": 6.017870473861694,
"epoch": 0.5134143144280346,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979526866881777,
"loss": 5.7764,
"mean_token_accuracy": 0.16695483028888702,
"num_tokens": 11907919.0,
"step": 5875
},
{
"entropy": 5.91297402381897,
"epoch": 0.5138512627807393,
"grad_norm": 0.90625,
"learning_rate": 0.000497948490453098,
"loss": 5.749,
"mean_token_accuracy": 0.16705156862735748,
"num_tokens": 11918612.0,
"step": 5880
},
{
"entropy": 5.986278533935547,
"epoch": 0.514288211133444,
"grad_norm": 0.85546875,
"learning_rate": 0.0004979442899417493,
"loss": 5.8105,
"mean_token_accuracy": 0.1623510330915451,
"num_tokens": 11927759.0,
"step": 5885
},
{
"entropy": 6.073221349716187,
"epoch": 0.5147251594861487,
"grad_norm": 0.90625,
"learning_rate": 0.0004979400851542122,
"loss": 5.7793,
"mean_token_accuracy": 0.16810414046049119,
"num_tokens": 11937522.0,
"step": 5890
},
{
"entropy": 5.950752019882202,
"epoch": 0.5151621078388534,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979358760905675,
"loss": 5.7659,
"mean_token_accuracy": 0.16226734668016435,
"num_tokens": 11947787.0,
"step": 5895
},
{
"entropy": 5.993432235717774,
"epoch": 0.5155990561915581,
"grad_norm": 0.890625,
"learning_rate": 0.000497931662750896,
"loss": 5.8018,
"mean_token_accuracy": 0.16491454988718032,
"num_tokens": 11958929.0,
"step": 5900
},
{
"entropy": 6.028296089172363,
"epoch": 0.5160360045442629,
"grad_norm": 0.98828125,
"learning_rate": 0.0004979274451352783,
"loss": 5.7525,
"mean_token_accuracy": 0.16111301630735397,
"num_tokens": 11969242.0,
"step": 5905
},
{
"entropy": 6.05470871925354,
"epoch": 0.5164729528969676,
"grad_norm": 0.90625,
"learning_rate": 0.0004979232232437954,
"loss": 5.8135,
"mean_token_accuracy": 0.1630721300840378,
"num_tokens": 11980149.0,
"step": 5910
},
{
"entropy": 6.071461725234985,
"epoch": 0.5169099012496723,
"grad_norm": 0.90625,
"learning_rate": 0.0004979189970765282,
"loss": 5.8085,
"mean_token_accuracy": 0.1617502823472023,
"num_tokens": 11990636.0,
"step": 5915
},
{
"entropy": 6.038999891281128,
"epoch": 0.517346849602377,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979147666335579,
"loss": 5.8137,
"mean_token_accuracy": 0.1656518757343292,
"num_tokens": 12000615.0,
"step": 5920
},
{
"entropy": 6.024601888656616,
"epoch": 0.5177837979550817,
"grad_norm": 0.93359375,
"learning_rate": 0.0004979105319149656,
"loss": 5.8525,
"mean_token_accuracy": 0.16385390758514404,
"num_tokens": 12011072.0,
"step": 5925
},
{
"entropy": 6.008143901824951,
"epoch": 0.5182207463077865,
"grad_norm": 0.9140625,
"learning_rate": 0.0004979062929208324,
"loss": 5.7506,
"mean_token_accuracy": 0.16876157373189926,
"num_tokens": 12021726.0,
"step": 5930
},
{
"entropy": 6.075357961654663,
"epoch": 0.5186576946604912,
"grad_norm": 0.94921875,
"learning_rate": 0.0004979020496512398,
"loss": 5.8845,
"mean_token_accuracy": 0.1558486267924309,
"num_tokens": 12031353.0,
"step": 5935
},
{
"entropy": 6.0586957931518555,
"epoch": 0.5190946430131959,
"grad_norm": 0.91796875,
"learning_rate": 0.000497897802106269,
"loss": 5.8251,
"mean_token_accuracy": 0.1579397052526474,
"num_tokens": 12041234.0,
"step": 5940
},
{
"entropy": 5.9722131252288815,
"epoch": 0.5195315913659005,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978935502860017,
"loss": 5.7596,
"mean_token_accuracy": 0.16703967601060868,
"num_tokens": 12051100.0,
"step": 5945
},
{
"entropy": 6.037038087844849,
"epoch": 0.5199685397186052,
"grad_norm": 0.875,
"learning_rate": 0.0004978892941905191,
"loss": 5.806,
"mean_token_accuracy": 0.16735392659902573,
"num_tokens": 12062012.0,
"step": 5950
},
{
"entropy": 6.086524915695191,
"epoch": 0.5204054880713099,
"grad_norm": 0.95703125,
"learning_rate": 0.0004978850338199033,
"loss": 5.8402,
"mean_token_accuracy": 0.1612513154745102,
"num_tokens": 12072763.0,
"step": 5955
},
{
"entropy": 5.9862555980682375,
"epoch": 0.5208424364240147,
"grad_norm": 0.9375,
"learning_rate": 0.0004978807691742356,
"loss": 5.6893,
"mean_token_accuracy": 0.1724808156490326,
"num_tokens": 12081892.0,
"step": 5960
},
{
"entropy": 6.019426059722901,
"epoch": 0.5212793847767194,
"grad_norm": 0.98046875,
"learning_rate": 0.000497876500253598,
"loss": 5.7704,
"mean_token_accuracy": 0.16536674797534942,
"num_tokens": 12091428.0,
"step": 5965
},
{
"entropy": 5.9819176197052,
"epoch": 0.5217163331294241,
"grad_norm": 0.92578125,
"learning_rate": 0.0004978722270580722,
"loss": 5.7747,
"mean_token_accuracy": 0.16727808862924576,
"num_tokens": 12101687.0,
"step": 5970
},
{
"entropy": 5.963487434387207,
"epoch": 0.5221532814821288,
"grad_norm": 0.84765625,
"learning_rate": 0.0004978679495877405,
"loss": 5.7282,
"mean_token_accuracy": 0.1675598442554474,
"num_tokens": 12111726.0,
"step": 5975
},
{
"entropy": 6.071728181838989,
"epoch": 0.5225902298348335,
"grad_norm": 0.90234375,
"learning_rate": 0.0004978636678426847,
"loss": 5.755,
"mean_token_accuracy": 0.1716315284371376,
"num_tokens": 12122171.0,
"step": 5980
},
{
"entropy": 6.087211227416992,
"epoch": 0.5230271781875382,
"grad_norm": 0.93359375,
"learning_rate": 0.0004978593818229869,
"loss": 5.8079,
"mean_token_accuracy": 0.16242510229349136,
"num_tokens": 12131414.0,
"step": 5985
},
{
"entropy": 5.968759298324585,
"epoch": 0.523464126540243,
"grad_norm": 1.0625,
"learning_rate": 0.0004978550915287294,
"loss": 5.7627,
"mean_token_accuracy": 0.15576607882976531,
"num_tokens": 12141665.0,
"step": 5990
},
{
"entropy": 6.049363803863526,
"epoch": 0.5239010748929477,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978507969599945,
"loss": 5.8318,
"mean_token_accuracy": 0.15878561586141587,
"num_tokens": 12151263.0,
"step": 5995
},
{
"entropy": 6.077195549011231,
"epoch": 0.5243380232456524,
"grad_norm": 0.83984375,
"learning_rate": 0.0004978464981168645,
"loss": 5.8145,
"mean_token_accuracy": 0.16041911840438844,
"num_tokens": 12161070.0,
"step": 6000
},
{
"epoch": 0.5243380232456524,
"eval_entropy": 5.812668307543265,
"eval_loss": 5.803792476654053,
"eval_mean_token_accuracy": 0.17073484680949289,
"eval_num_tokens": 12161070.0,
"eval_runtime": 21.6682,
"eval_samples_per_second": 1708.221,
"eval_steps_per_second": 213.539,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 114420,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2700955567226880.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink