fpadovani/dan-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
dan-latn-100mb-after-ppt-sh…/checkpoint-6000/trainer_state.json
ModelHub XC c6d06f7aaa 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/dan-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
Source: Original Platform
2026-07-20 03:48:14 +08:00

12057 lines
329 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5011693952555963,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.839805936813354,
"epoch": 0.000417641162712997,
"grad_norm": 14.0,
"learning_rate": 2e-06,
"loss": 14.3117,
"mean_token_accuracy": 0.0,
"num_tokens": 9371.0,
"step": 5
},
{
"entropy": 4.85613489151001,
"epoch": 0.000835282325425994,
"grad_norm": 14.9375,
"learning_rate": 4.5e-06,
"loss": 14.2404,
"mean_token_accuracy": 0.0,
"num_tokens": 18155.0,
"step": 10
},
{
"entropy": 4.9069380283355715,
"epoch": 0.001252923488138991,
"grad_norm": 17.125,
"learning_rate": 7e-06,
"loss": 14.0553,
"mean_token_accuracy": 9.416195680387319e-05,
"num_tokens": 27776.0,
"step": 15
},
{
"entropy": 5.117074966430664,
"epoch": 0.001670564650851988,
"grad_norm": 26.25,
"learning_rate": 9.5e-06,
"loss": 13.5033,
"mean_token_accuracy": 0.0,
"num_tokens": 37166.0,
"step": 20
},
{
"entropy": 6.88233642578125,
"epoch": 0.002088205813564985,
"grad_norm": 27.25,
"learning_rate": 1.2e-05,
"loss": 12.0856,
"mean_token_accuracy": 0.0,
"num_tokens": 46667.0,
"step": 25
},
{
"entropy": 10.158202266693115,
"epoch": 0.002505846976277982,
"grad_norm": 4.21875,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.8984,
"mean_token_accuracy": 0.0,
"num_tokens": 55738.0,
"step": 30
},
{
"entropy": 10.693987464904785,
"epoch": 0.002923488138990979,
"grad_norm": 3.078125,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.6313,
"mean_token_accuracy": 0.010924705216893926,
"num_tokens": 65303.0,
"step": 35
},
{
"entropy": 10.707240295410156,
"epoch": 0.003341129301703976,
"grad_norm": 2.8125,
"learning_rate": 1.95e-05,
"loss": 10.333,
"mean_token_accuracy": 0.0401553388684988,
"num_tokens": 74876.0,
"step": 40
},
{
"entropy": 10.677993488311767,
"epoch": 0.003758770464416973,
"grad_norm": 2.453125,
"learning_rate": 2.2e-05,
"loss": 9.9854,
"mean_token_accuracy": 0.04825415499508381,
"num_tokens": 84683.0,
"step": 45
},
{
"entropy": 10.602502727508545,
"epoch": 0.00417641162712997,
"grad_norm": 2.46875,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.751,
"mean_token_accuracy": 0.04532666318118572,
"num_tokens": 94511.0,
"step": 50
},
{
"entropy": 10.550715732574464,
"epoch": 0.004594052789842967,
"grad_norm": 2.0625,
"learning_rate": 2.7e-05,
"loss": 9.5808,
"mean_token_accuracy": 0.04490456208586693,
"num_tokens": 103820.0,
"step": 55
},
{
"entropy": 10.557010459899903,
"epoch": 0.005011693952555964,
"grad_norm": 2.09375,
"learning_rate": 2.95e-05,
"loss": 9.487,
"mean_token_accuracy": 0.04334092810750008,
"num_tokens": 112398.0,
"step": 60
},
{
"entropy": 10.562658882141113,
"epoch": 0.005429335115268961,
"grad_norm": 2.140625,
"learning_rate": 3.2e-05,
"loss": 9.4118,
"mean_token_accuracy": 0.04636790603399277,
"num_tokens": 121167.0,
"step": 65
},
{
"entropy": 10.556397819519043,
"epoch": 0.005846976277981958,
"grad_norm": 1.9921875,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.3179,
"mean_token_accuracy": 0.04509260468184948,
"num_tokens": 129399.0,
"step": 70
},
{
"entropy": 10.528542232513427,
"epoch": 0.006264617440694955,
"grad_norm": 2.078125,
"learning_rate": 3.7e-05,
"loss": 9.2505,
"mean_token_accuracy": 0.04285807758569717,
"num_tokens": 138512.0,
"step": 75
},
{
"entropy": 10.51480073928833,
"epoch": 0.006682258603407952,
"grad_norm": 1.8515625,
"learning_rate": 3.95e-05,
"loss": 9.2265,
"mean_token_accuracy": 0.04041963405907154,
"num_tokens": 147712.0,
"step": 80
},
{
"entropy": 10.507884788513184,
"epoch": 0.007099899766120949,
"grad_norm": 1.921875,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.0905,
"mean_token_accuracy": 0.051801984757184984,
"num_tokens": 156759.0,
"step": 85
},
{
"entropy": 10.462992191314697,
"epoch": 0.007517540928833946,
"grad_norm": 2.0,
"learning_rate": 4.45e-05,
"loss": 8.9786,
"mean_token_accuracy": 0.05203719846904278,
"num_tokens": 166395.0,
"step": 90
},
{
"entropy": 10.411985778808594,
"epoch": 0.007935182091546943,
"grad_norm": 1.890625,
"learning_rate": 4.7000000000000004e-05,
"loss": 8.8518,
"mean_token_accuracy": 0.0536493469029665,
"num_tokens": 174739.0,
"step": 95
},
{
"entropy": 10.324466896057128,
"epoch": 0.00835282325425994,
"grad_norm": 1.90625,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.6885,
"mean_token_accuracy": 0.05701705515384674,
"num_tokens": 183010.0,
"step": 100
},
{
"entropy": 10.227258014678956,
"epoch": 0.008770464416972937,
"grad_norm": 1.7265625,
"learning_rate": 5.2e-05,
"loss": 8.6423,
"mean_token_accuracy": 0.053074367344379425,
"num_tokens": 192589.0,
"step": 105
},
{
"entropy": 10.162396335601807,
"epoch": 0.009188105579685933,
"grad_norm": 1.6796875,
"learning_rate": 5.45e-05,
"loss": 8.5017,
"mean_token_accuracy": 0.05809981971979141,
"num_tokens": 201638.0,
"step": 110
},
{
"entropy": 10.037503719329834,
"epoch": 0.00960574674239893,
"grad_norm": 1.6640625,
"learning_rate": 5.7e-05,
"loss": 8.3491,
"mean_token_accuracy": 0.0572802733629942,
"num_tokens": 209862.0,
"step": 115
},
{
"entropy": 9.966203022003175,
"epoch": 0.010023387905111928,
"grad_norm": 1.453125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.3354,
"mean_token_accuracy": 0.04967698715627193,
"num_tokens": 218897.0,
"step": 120
},
{
"entropy": 9.839297485351562,
"epoch": 0.010441029067824925,
"grad_norm": 1.359375,
"learning_rate": 6.2e-05,
"loss": 8.1976,
"mean_token_accuracy": 0.05662783086299896,
"num_tokens": 227682.0,
"step": 125
},
{
"entropy": 9.541645622253418,
"epoch": 0.010858670230537921,
"grad_norm": 1.546875,
"learning_rate": 6.450000000000001e-05,
"loss": 7.9999,
"mean_token_accuracy": 0.057890772446990015,
"num_tokens": 236170.0,
"step": 130
},
{
"entropy": 9.362167835235596,
"epoch": 0.011276311393250919,
"grad_norm": 1.4296875,
"learning_rate": 6.7e-05,
"loss": 7.9447,
"mean_token_accuracy": 0.053887016326189044,
"num_tokens": 245710.0,
"step": 135
},
{
"entropy": 9.22001256942749,
"epoch": 0.011693952555963916,
"grad_norm": 1.609375,
"learning_rate": 6.950000000000001e-05,
"loss": 7.7881,
"mean_token_accuracy": 0.0592480544000864,
"num_tokens": 254751.0,
"step": 140
},
{
"entropy": 8.881026935577392,
"epoch": 0.012111593718676913,
"grad_norm": 1.2109375,
"learning_rate": 7.2e-05,
"loss": 7.6882,
"mean_token_accuracy": 0.05897580310702324,
"num_tokens": 264210.0,
"step": 145
},
{
"entropy": 8.729740524291993,
"epoch": 0.01252923488138991,
"grad_norm": 1.1953125,
"learning_rate": 7.45e-05,
"loss": 7.6572,
"mean_token_accuracy": 0.061323515325784686,
"num_tokens": 272944.0,
"step": 150
},
{
"entropy": 8.590540027618408,
"epoch": 0.012946876044102907,
"grad_norm": 1.2578125,
"learning_rate": 7.7e-05,
"loss": 7.546,
"mean_token_accuracy": 0.05964183136820793,
"num_tokens": 281311.0,
"step": 155
},
{
"entropy": 8.28624792098999,
"epoch": 0.013364517206815904,
"grad_norm": 1.3515625,
"learning_rate": 7.950000000000001e-05,
"loss": 7.4735,
"mean_token_accuracy": 0.0655230738222599,
"num_tokens": 289701.0,
"step": 160
},
{
"entropy": 8.275508499145507,
"epoch": 0.013782158369528901,
"grad_norm": 0.93359375,
"learning_rate": 8.2e-05,
"loss": 7.5326,
"mean_token_accuracy": 0.06389093734323978,
"num_tokens": 299459.0,
"step": 165
},
{
"entropy": 8.162372493743897,
"epoch": 0.014199799532241897,
"grad_norm": 1.0234375,
"learning_rate": 8.450000000000001e-05,
"loss": 7.5028,
"mean_token_accuracy": 0.06458220668137074,
"num_tokens": 309343.0,
"step": 170
},
{
"entropy": 8.083532857894898,
"epoch": 0.014617440694954895,
"grad_norm": 1.1484375,
"learning_rate": 8.7e-05,
"loss": 7.4762,
"mean_token_accuracy": 0.06845348998904228,
"num_tokens": 318840.0,
"step": 175
},
{
"entropy": 8.051537418365479,
"epoch": 0.015035081857667892,
"grad_norm": 1.1953125,
"learning_rate": 8.95e-05,
"loss": 7.2904,
"mean_token_accuracy": 0.07184495069086552,
"num_tokens": 328689.0,
"step": 180
},
{
"entropy": 7.988371896743774,
"epoch": 0.01545272302038089,
"grad_norm": 1.0234375,
"learning_rate": 9.2e-05,
"loss": 7.4079,
"mean_token_accuracy": 0.06707764603197575,
"num_tokens": 338209.0,
"step": 185
},
{
"entropy": 7.929311227798462,
"epoch": 0.015870364183093885,
"grad_norm": 0.984375,
"learning_rate": 9.45e-05,
"loss": 7.2823,
"mean_token_accuracy": 0.0719263195991516,
"num_tokens": 347270.0,
"step": 190
},
{
"entropy": 7.857247591018677,
"epoch": 0.016288005345806884,
"grad_norm": 0.98828125,
"learning_rate": 9.7e-05,
"loss": 7.3071,
"mean_token_accuracy": 0.07316098734736443,
"num_tokens": 356366.0,
"step": 195
},
{
"entropy": 7.875389623641968,
"epoch": 0.01670564650851988,
"grad_norm": 1.46875,
"learning_rate": 9.95e-05,
"loss": 7.2708,
"mean_token_accuracy": 0.07387731000781059,
"num_tokens": 365980.0,
"step": 200
},
{
"entropy": 7.728267478942871,
"epoch": 0.017123287671232876,
"grad_norm": 0.97265625,
"learning_rate": 0.000102,
"loss": 7.2272,
"mean_token_accuracy": 0.07522129304707051,
"num_tokens": 374591.0,
"step": 205
},
{
"entropy": 7.748341989517212,
"epoch": 0.017540928833945875,
"grad_norm": 1.0234375,
"learning_rate": 0.00010449999999999999,
"loss": 7.2198,
"mean_token_accuracy": 0.07714342400431633,
"num_tokens": 383250.0,
"step": 210
},
{
"entropy": 7.837753868103027,
"epoch": 0.01795856999665887,
"grad_norm": 1.0859375,
"learning_rate": 0.000107,
"loss": 7.1392,
"mean_token_accuracy": 0.07376849949359894,
"num_tokens": 391502.0,
"step": 215
},
{
"entropy": 7.608754634857178,
"epoch": 0.018376211159371866,
"grad_norm": 1.0546875,
"learning_rate": 0.0001095,
"loss": 7.0676,
"mean_token_accuracy": 0.0789151631295681,
"num_tokens": 400225.0,
"step": 220
},
{
"entropy": 7.752291250228882,
"epoch": 0.018793852322084866,
"grad_norm": 0.91015625,
"learning_rate": 0.000112,
"loss": 7.274,
"mean_token_accuracy": 0.0727268923074007,
"num_tokens": 410008.0,
"step": 225
},
{
"entropy": 7.60542573928833,
"epoch": 0.01921149348479786,
"grad_norm": 1.0234375,
"learning_rate": 0.0001145,
"loss": 7.183,
"mean_token_accuracy": 0.07938049286603928,
"num_tokens": 418979.0,
"step": 230
},
{
"entropy": 7.717845916748047,
"epoch": 0.01962913464751086,
"grad_norm": 0.94140625,
"learning_rate": 0.00011700000000000001,
"loss": 7.1495,
"mean_token_accuracy": 0.08237392455339432,
"num_tokens": 428139.0,
"step": 235
},
{
"entropy": 7.551066637039185,
"epoch": 0.020046775810223856,
"grad_norm": 1.140625,
"learning_rate": 0.00011949999999999999,
"loss": 7.0674,
"mean_token_accuracy": 0.08105618506669998,
"num_tokens": 437113.0,
"step": 240
},
{
"entropy": 7.683083581924438,
"epoch": 0.020464416972936852,
"grad_norm": 1.40625,
"learning_rate": 0.000122,
"loss": 7.2807,
"mean_token_accuracy": 0.07711479477584363,
"num_tokens": 446879.0,
"step": 245
},
{
"entropy": 7.552040719985962,
"epoch": 0.02088205813564985,
"grad_norm": 1.109375,
"learning_rate": 0.0001245,
"loss": 7.1648,
"mean_token_accuracy": 0.08340629488229752,
"num_tokens": 456362.0,
"step": 250
},
{
"entropy": 7.564500951766968,
"epoch": 0.021299699298362847,
"grad_norm": 1.171875,
"learning_rate": 0.000127,
"loss": 6.9652,
"mean_token_accuracy": 0.08746102228760719,
"num_tokens": 464930.0,
"step": 255
},
{
"entropy": 7.649157238006592,
"epoch": 0.021717340461075842,
"grad_norm": 1.5,
"learning_rate": 0.0001295,
"loss": 7.1434,
"mean_token_accuracy": 0.07869515046477318,
"num_tokens": 474303.0,
"step": 260
},
{
"entropy": 7.556221199035645,
"epoch": 0.02213498162378884,
"grad_norm": 1.3203125,
"learning_rate": 0.000132,
"loss": 7.1414,
"mean_token_accuracy": 0.08254432752728462,
"num_tokens": 483750.0,
"step": 265
},
{
"entropy": 7.497516727447509,
"epoch": 0.022552622786501837,
"grad_norm": 1.0078125,
"learning_rate": 0.00013450000000000002,
"loss": 7.0328,
"mean_token_accuracy": 0.0869042843580246,
"num_tokens": 493325.0,
"step": 270
},
{
"entropy": 7.550202894210815,
"epoch": 0.022970263949214833,
"grad_norm": 1.1015625,
"learning_rate": 0.00013700000000000002,
"loss": 7.0252,
"mean_token_accuracy": 0.08019787892699241,
"num_tokens": 501907.0,
"step": 275
},
{
"entropy": 7.360218524932861,
"epoch": 0.023387905111927832,
"grad_norm": 1.1875,
"learning_rate": 0.0001395,
"loss": 7.079,
"mean_token_accuracy": 0.08745673149824143,
"num_tokens": 511522.0,
"step": 280
},
{
"entropy": 7.559610795974732,
"epoch": 0.023805546274640828,
"grad_norm": 1.2265625,
"learning_rate": 0.00014199999999999998,
"loss": 7.0269,
"mean_token_accuracy": 0.08385204374790192,
"num_tokens": 521287.0,
"step": 285
},
{
"entropy": 7.444974565505982,
"epoch": 0.024223187437353827,
"grad_norm": 1.0078125,
"learning_rate": 0.0001445,
"loss": 7.0961,
"mean_token_accuracy": 0.07927819266915322,
"num_tokens": 530631.0,
"step": 290
},
{
"entropy": 7.401523399353027,
"epoch": 0.024640828600066823,
"grad_norm": 1.1328125,
"learning_rate": 0.000147,
"loss": 6.9731,
"mean_token_accuracy": 0.09002419337630271,
"num_tokens": 540386.0,
"step": 295
},
{
"entropy": 7.445022916793823,
"epoch": 0.02505846976277982,
"grad_norm": 1.640625,
"learning_rate": 0.0001495,
"loss": 7.053,
"mean_token_accuracy": 0.08409906476736069,
"num_tokens": 550121.0,
"step": 300
},
{
"entropy": 7.3913647651672365,
"epoch": 0.025476110925492818,
"grad_norm": 0.984375,
"learning_rate": 0.000152,
"loss": 6.9113,
"mean_token_accuracy": 0.09003557711839676,
"num_tokens": 559460.0,
"step": 305
},
{
"entropy": 7.339849758148193,
"epoch": 0.025893752088205813,
"grad_norm": 1.171875,
"learning_rate": 0.00015450000000000001,
"loss": 6.8715,
"mean_token_accuracy": 0.09122911468148232,
"num_tokens": 568392.0,
"step": 310
},
{
"entropy": 7.381598663330078,
"epoch": 0.02631139325091881,
"grad_norm": 0.99609375,
"learning_rate": 0.000157,
"loss": 6.9572,
"mean_token_accuracy": 0.0872497834265232,
"num_tokens": 578248.0,
"step": 315
},
{
"entropy": 7.4210817337036135,
"epoch": 0.026729034413631808,
"grad_norm": 1.1328125,
"learning_rate": 0.0001595,
"loss": 7.0283,
"mean_token_accuracy": 0.08656517416238785,
"num_tokens": 588164.0,
"step": 320
},
{
"entropy": 7.34563045501709,
"epoch": 0.027146675576344804,
"grad_norm": 0.90625,
"learning_rate": 0.000162,
"loss": 6.9552,
"mean_token_accuracy": 0.08556340485811234,
"num_tokens": 598715.0,
"step": 325
},
{
"entropy": 7.329131650924682,
"epoch": 0.027564316739057803,
"grad_norm": 1.3203125,
"learning_rate": 0.00016450000000000001,
"loss": 6.8909,
"mean_token_accuracy": 0.09089908376336098,
"num_tokens": 606939.0,
"step": 330
},
{
"entropy": 7.301250839233399,
"epoch": 0.0279819579017708,
"grad_norm": 1.484375,
"learning_rate": 0.00016700000000000002,
"loss": 6.9723,
"mean_token_accuracy": 0.09220347255468368,
"num_tokens": 615803.0,
"step": 335
},
{
"entropy": 7.307532596588135,
"epoch": 0.028399599064483794,
"grad_norm": 1.140625,
"learning_rate": 0.00016950000000000003,
"loss": 6.843,
"mean_token_accuracy": 0.08909467533230782,
"num_tokens": 624730.0,
"step": 340
},
{
"entropy": 7.364404821395874,
"epoch": 0.028817240227196794,
"grad_norm": 1.0625,
"learning_rate": 0.00017199999999999998,
"loss": 6.9289,
"mean_token_accuracy": 0.09335248172283173,
"num_tokens": 634082.0,
"step": 345
},
{
"entropy": 7.287901830673218,
"epoch": 0.02923488138990979,
"grad_norm": 1.203125,
"learning_rate": 0.00017449999999999999,
"loss": 6.9168,
"mean_token_accuracy": 0.0900125928223133,
"num_tokens": 643367.0,
"step": 350
},
{
"entropy": 7.12314624786377,
"epoch": 0.029652522552622785,
"grad_norm": 1.1171875,
"learning_rate": 0.000177,
"loss": 6.7417,
"mean_token_accuracy": 0.09290036335587501,
"num_tokens": 651861.0,
"step": 355
},
{
"entropy": 7.289297485351563,
"epoch": 0.030070163715335784,
"grad_norm": 1.0625,
"learning_rate": 0.0001795,
"loss": 6.8694,
"mean_token_accuracy": 0.08788876459002495,
"num_tokens": 661064.0,
"step": 360
},
{
"entropy": 7.208283281326294,
"epoch": 0.03048780487804878,
"grad_norm": 1.1171875,
"learning_rate": 0.000182,
"loss": 6.8389,
"mean_token_accuracy": 0.09632247537374497,
"num_tokens": 669299.0,
"step": 365
},
{
"entropy": 7.210040044784546,
"epoch": 0.03090544604076178,
"grad_norm": 1.1875,
"learning_rate": 0.0001845,
"loss": 6.917,
"mean_token_accuracy": 0.0862058375030756,
"num_tokens": 678631.0,
"step": 370
},
{
"entropy": 7.237060689926148,
"epoch": 0.03132308720347477,
"grad_norm": 1.125,
"learning_rate": 0.000187,
"loss": 6.906,
"mean_token_accuracy": 0.08874215260148048,
"num_tokens": 688511.0,
"step": 375
},
{
"entropy": 7.259998083114624,
"epoch": 0.03174072836618777,
"grad_norm": 1.515625,
"learning_rate": 0.0001895,
"loss": 6.8535,
"mean_token_accuracy": 0.08548418506979942,
"num_tokens": 697956.0,
"step": 380
},
{
"entropy": 7.210371494293213,
"epoch": 0.03215836952890077,
"grad_norm": 1.2421875,
"learning_rate": 0.000192,
"loss": 6.772,
"mean_token_accuracy": 0.09457366839051247,
"num_tokens": 706779.0,
"step": 385
},
{
"entropy": 7.116398477554322,
"epoch": 0.03257601069161377,
"grad_norm": 1.078125,
"learning_rate": 0.0001945,
"loss": 6.7441,
"mean_token_accuracy": 0.0957478053867817,
"num_tokens": 715749.0,
"step": 390
},
{
"entropy": 7.1532073497772215,
"epoch": 0.03299365185432676,
"grad_norm": 1.3046875,
"learning_rate": 0.00019700000000000002,
"loss": 6.7309,
"mean_token_accuracy": 0.08776799142360688,
"num_tokens": 724756.0,
"step": 395
},
{
"entropy": 7.086204910278321,
"epoch": 0.03341129301703976,
"grad_norm": 1.03125,
"learning_rate": 0.00019950000000000002,
"loss": 6.7542,
"mean_token_accuracy": 0.09009818658232689,
"num_tokens": 733288.0,
"step": 400
},
{
"entropy": 7.221790742874146,
"epoch": 0.03382893417975276,
"grad_norm": 1.390625,
"learning_rate": 0.000202,
"loss": 6.8585,
"mean_token_accuracy": 0.0874378077685833,
"num_tokens": 741629.0,
"step": 405
},
{
"entropy": 7.075778245925903,
"epoch": 0.03424657534246575,
"grad_norm": 1.21875,
"learning_rate": 0.00020449999999999998,
"loss": 6.805,
"mean_token_accuracy": 0.09457797557115555,
"num_tokens": 751336.0,
"step": 410
},
{
"entropy": 7.2090919494628904,
"epoch": 0.03466421650517875,
"grad_norm": 1.203125,
"learning_rate": 0.000207,
"loss": 6.8688,
"mean_token_accuracy": 0.08673702701926231,
"num_tokens": 759725.0,
"step": 415
},
{
"entropy": 7.149182081222534,
"epoch": 0.03508185766789175,
"grad_norm": 1.3828125,
"learning_rate": 0.0002095,
"loss": 6.8821,
"mean_token_accuracy": 0.08862207159399986,
"num_tokens": 769752.0,
"step": 420
},
{
"entropy": 7.189090824127197,
"epoch": 0.03549949883060474,
"grad_norm": 1.328125,
"learning_rate": 0.000212,
"loss": 6.764,
"mean_token_accuracy": 0.0942901387810707,
"num_tokens": 779124.0,
"step": 425
},
{
"entropy": 7.04871621131897,
"epoch": 0.03591713999331774,
"grad_norm": 1.390625,
"learning_rate": 0.0002145,
"loss": 6.7259,
"mean_token_accuracy": 0.09670177772641182,
"num_tokens": 788892.0,
"step": 430
},
{
"entropy": 6.927881717681885,
"epoch": 0.03633478115603074,
"grad_norm": 1.4453125,
"learning_rate": 0.00021700000000000002,
"loss": 6.6789,
"mean_token_accuracy": 0.09807197973132134,
"num_tokens": 798265.0,
"step": 435
},
{
"entropy": 7.044727849960327,
"epoch": 0.03675242231874373,
"grad_norm": 1.3125,
"learning_rate": 0.0002195,
"loss": 6.6357,
"mean_token_accuracy": 0.09784042909741401,
"num_tokens": 806896.0,
"step": 440
},
{
"entropy": 7.097750091552735,
"epoch": 0.03717006348145673,
"grad_norm": 1.25,
"learning_rate": 0.000222,
"loss": 6.7061,
"mean_token_accuracy": 0.09837085008621216,
"num_tokens": 815540.0,
"step": 445
},
{
"entropy": 6.942305183410644,
"epoch": 0.03758770464416973,
"grad_norm": 1.0390625,
"learning_rate": 0.0002245,
"loss": 6.7667,
"mean_token_accuracy": 0.0919320560991764,
"num_tokens": 824555.0,
"step": 450
},
{
"entropy": 7.130709552764893,
"epoch": 0.03800534580688272,
"grad_norm": 1.2421875,
"learning_rate": 0.00022700000000000002,
"loss": 6.7973,
"mean_token_accuracy": 0.09710298180580139,
"num_tokens": 833414.0,
"step": 455
},
{
"entropy": 6.919841003417969,
"epoch": 0.03842298696959572,
"grad_norm": 1.09375,
"learning_rate": 0.00022950000000000002,
"loss": 6.6649,
"mean_token_accuracy": 0.09833707362413406,
"num_tokens": 842644.0,
"step": 460
},
{
"entropy": 7.071603155136108,
"epoch": 0.03884062813230872,
"grad_norm": 1.234375,
"learning_rate": 0.00023200000000000003,
"loss": 6.6819,
"mean_token_accuracy": 0.09334187656641006,
"num_tokens": 851786.0,
"step": 465
},
{
"entropy": 6.958913516998291,
"epoch": 0.03925826929502172,
"grad_norm": 1.15625,
"learning_rate": 0.00023449999999999998,
"loss": 6.7444,
"mean_token_accuracy": 0.09533607885241509,
"num_tokens": 861508.0,
"step": 470
},
{
"entropy": 7.00027494430542,
"epoch": 0.03967591045773471,
"grad_norm": 1.1484375,
"learning_rate": 0.000237,
"loss": 6.6109,
"mean_token_accuracy": 0.10132770463824273,
"num_tokens": 870433.0,
"step": 475
},
{
"entropy": 7.001135778427124,
"epoch": 0.04009355162044771,
"grad_norm": 1.2265625,
"learning_rate": 0.0002395,
"loss": 6.7219,
"mean_token_accuracy": 0.09730773493647575,
"num_tokens": 879059.0,
"step": 480
},
{
"entropy": 7.048274087905884,
"epoch": 0.04051119278316071,
"grad_norm": 1.484375,
"learning_rate": 0.000242,
"loss": 6.7824,
"mean_token_accuracy": 0.09053769409656524,
"num_tokens": 888009.0,
"step": 485
},
{
"entropy": 6.907274293899536,
"epoch": 0.040928833945873704,
"grad_norm": 1.15625,
"learning_rate": 0.0002445,
"loss": 6.6314,
"mean_token_accuracy": 0.09549420028924942,
"num_tokens": 897335.0,
"step": 490
},
{
"entropy": 7.023768520355224,
"epoch": 0.0413464751085867,
"grad_norm": 1.1875,
"learning_rate": 0.000247,
"loss": 6.637,
"mean_token_accuracy": 0.10220221057534218,
"num_tokens": 905802.0,
"step": 495
},
{
"entropy": 6.961791563034057,
"epoch": 0.0417641162712997,
"grad_norm": 1.2734375,
"learning_rate": 0.0002495,
"loss": 6.6318,
"mean_token_accuracy": 0.10365633964538574,
"num_tokens": 914836.0,
"step": 500
},
{
"entropy": 6.8752366065979,
"epoch": 0.042181757434012694,
"grad_norm": 1.2734375,
"learning_rate": 0.000252,
"loss": 6.5863,
"mean_token_accuracy": 0.10103005766868592,
"num_tokens": 924254.0,
"step": 505
},
{
"entropy": 6.950608396530152,
"epoch": 0.04259939859672569,
"grad_norm": 1.1484375,
"learning_rate": 0.0002545,
"loss": 6.6791,
"mean_token_accuracy": 0.09583078920841218,
"num_tokens": 933841.0,
"step": 510
},
{
"entropy": 7.015104198455811,
"epoch": 0.04301703975943869,
"grad_norm": 1.25,
"learning_rate": 0.000257,
"loss": 6.7416,
"mean_token_accuracy": 0.09457004889845848,
"num_tokens": 943304.0,
"step": 515
},
{
"entropy": 6.826920413970948,
"epoch": 0.043434680922151685,
"grad_norm": 1.0078125,
"learning_rate": 0.0002595,
"loss": 6.5701,
"mean_token_accuracy": 0.09870983734726906,
"num_tokens": 952915.0,
"step": 520
},
{
"entropy": 6.92740478515625,
"epoch": 0.043852322084864684,
"grad_norm": 1.109375,
"learning_rate": 0.000262,
"loss": 6.6264,
"mean_token_accuracy": 0.10174370929598808,
"num_tokens": 962752.0,
"step": 525
},
{
"entropy": 6.86022162437439,
"epoch": 0.04426996324757768,
"grad_norm": 1.203125,
"learning_rate": 0.00026450000000000003,
"loss": 6.6449,
"mean_token_accuracy": 0.0926640897989273,
"num_tokens": 971240.0,
"step": 530
},
{
"entropy": 6.899795818328857,
"epoch": 0.044687604410290675,
"grad_norm": 1.5234375,
"learning_rate": 0.00026700000000000004,
"loss": 6.6886,
"mean_token_accuracy": 0.09312456697225571,
"num_tokens": 981391.0,
"step": 535
},
{
"entropy": 6.915267086029052,
"epoch": 0.045105245573003674,
"grad_norm": 1.0390625,
"learning_rate": 0.00026950000000000005,
"loss": 6.6941,
"mean_token_accuracy": 0.10025263205170631,
"num_tokens": 990551.0,
"step": 540
},
{
"entropy": 6.9169872283935545,
"epoch": 0.045522886735716674,
"grad_norm": 1.25,
"learning_rate": 0.00027200000000000005,
"loss": 6.6148,
"mean_token_accuracy": 0.09887195006012917,
"num_tokens": 999593.0,
"step": 545
},
{
"entropy": 6.888961887359619,
"epoch": 0.045940527898429666,
"grad_norm": 1.1640625,
"learning_rate": 0.0002745,
"loss": 6.6766,
"mean_token_accuracy": 0.09649211689829826,
"num_tokens": 1008785.0,
"step": 550
},
{
"entropy": 6.853534841537476,
"epoch": 0.046358169061142665,
"grad_norm": 1.1953125,
"learning_rate": 0.000277,
"loss": 6.651,
"mean_token_accuracy": 0.09519294798374175,
"num_tokens": 1016815.0,
"step": 555
},
{
"entropy": 6.857504081726074,
"epoch": 0.046775810223855664,
"grad_norm": 1.2265625,
"learning_rate": 0.0002795,
"loss": 6.5615,
"mean_token_accuracy": 0.10218018293380737,
"num_tokens": 1026182.0,
"step": 560
},
{
"entropy": 6.799322652816772,
"epoch": 0.04719345138656866,
"grad_norm": 1.4375,
"learning_rate": 0.00028199999999999997,
"loss": 6.4846,
"mean_token_accuracy": 0.10682575106620788,
"num_tokens": 1034032.0,
"step": 565
},
{
"entropy": 6.839007759094239,
"epoch": 0.047611092549281656,
"grad_norm": 1.078125,
"learning_rate": 0.0002845,
"loss": 6.6722,
"mean_token_accuracy": 0.10130207389593124,
"num_tokens": 1043493.0,
"step": 570
},
{
"entropy": 6.8931989669799805,
"epoch": 0.048028733711994655,
"grad_norm": 1.3203125,
"learning_rate": 0.000287,
"loss": 6.5889,
"mean_token_accuracy": 0.10233498141169547,
"num_tokens": 1052989.0,
"step": 575
},
{
"entropy": 6.765599155426026,
"epoch": 0.048446374874707654,
"grad_norm": 1.171875,
"learning_rate": 0.0002895,
"loss": 6.5573,
"mean_token_accuracy": 0.10355089157819748,
"num_tokens": 1061803.0,
"step": 580
},
{
"entropy": 6.836974954605102,
"epoch": 0.048864016037420646,
"grad_norm": 0.9921875,
"learning_rate": 0.000292,
"loss": 6.6617,
"mean_token_accuracy": 0.09715791568160057,
"num_tokens": 1071227.0,
"step": 585
},
{
"entropy": 6.887206649780273,
"epoch": 0.049281657200133645,
"grad_norm": 1.078125,
"learning_rate": 0.0002945,
"loss": 6.6496,
"mean_token_accuracy": 0.10116483271121979,
"num_tokens": 1080082.0,
"step": 590
},
{
"entropy": 6.793925666809082,
"epoch": 0.049699298362846644,
"grad_norm": 1.234375,
"learning_rate": 0.000297,
"loss": 6.5013,
"mean_token_accuracy": 0.10670637711882591,
"num_tokens": 1089750.0,
"step": 595
},
{
"entropy": 6.714077949523926,
"epoch": 0.05011693952555964,
"grad_norm": 1.2734375,
"learning_rate": 0.0002995,
"loss": 6.4937,
"mean_token_accuracy": 0.1003747045993805,
"num_tokens": 1098691.0,
"step": 600
},
{
"entropy": 6.748677492141724,
"epoch": 0.050534580688272636,
"grad_norm": 1.1484375,
"learning_rate": 0.000302,
"loss": 6.5418,
"mean_token_accuracy": 0.10015128180384636,
"num_tokens": 1107602.0,
"step": 605
},
{
"entropy": 6.73736834526062,
"epoch": 0.050952221850985635,
"grad_norm": 1.3125,
"learning_rate": 0.0003045,
"loss": 6.4533,
"mean_token_accuracy": 0.10660258829593658,
"num_tokens": 1116426.0,
"step": 610
},
{
"entropy": 6.826512193679809,
"epoch": 0.05136986301369863,
"grad_norm": 1.0703125,
"learning_rate": 0.000307,
"loss": 6.6419,
"mean_token_accuracy": 0.10342860966920853,
"num_tokens": 1126221.0,
"step": 615
},
{
"entropy": 6.683390331268311,
"epoch": 0.051787504176411626,
"grad_norm": 1.3515625,
"learning_rate": 0.0003095,
"loss": 6.5224,
"mean_token_accuracy": 0.10383139550685883,
"num_tokens": 1135662.0,
"step": 620
},
{
"entropy": 6.7781775951385494,
"epoch": 0.052205145339124626,
"grad_norm": 1.2109375,
"learning_rate": 0.000312,
"loss": 6.6037,
"mean_token_accuracy": 0.10372941642999649,
"num_tokens": 1143963.0,
"step": 625
},
{
"entropy": 6.685908460617066,
"epoch": 0.05262278650183762,
"grad_norm": 1.15625,
"learning_rate": 0.0003145,
"loss": 6.4911,
"mean_token_accuracy": 0.11205898746848106,
"num_tokens": 1152470.0,
"step": 630
},
{
"entropy": 6.785271120071411,
"epoch": 0.05304042766455062,
"grad_norm": 1.15625,
"learning_rate": 0.000317,
"loss": 6.4578,
"mean_token_accuracy": 0.10582235008478165,
"num_tokens": 1161257.0,
"step": 635
},
{
"entropy": 6.792138624191284,
"epoch": 0.053458068827263616,
"grad_norm": 1.171875,
"learning_rate": 0.0003195,
"loss": 6.5567,
"mean_token_accuracy": 0.10366798266768455,
"num_tokens": 1170969.0,
"step": 640
},
{
"entropy": 6.781137418746948,
"epoch": 0.053875709989976615,
"grad_norm": 1.0859375,
"learning_rate": 0.000322,
"loss": 6.6947,
"mean_token_accuracy": 0.10238798782229423,
"num_tokens": 1181127.0,
"step": 645
},
{
"entropy": 6.742985105514526,
"epoch": 0.05429335115268961,
"grad_norm": 1.1640625,
"learning_rate": 0.00032450000000000003,
"loss": 6.5046,
"mean_token_accuracy": 0.10330399274826049,
"num_tokens": 1190260.0,
"step": 650
},
{
"entropy": 6.6638226985931395,
"epoch": 0.05471099231540261,
"grad_norm": 1.125,
"learning_rate": 0.00032700000000000003,
"loss": 6.5777,
"mean_token_accuracy": 0.10619365200400352,
"num_tokens": 1200013.0,
"step": 655
},
{
"entropy": 6.780775451660157,
"epoch": 0.055128633478115606,
"grad_norm": 1.4765625,
"learning_rate": 0.00032950000000000004,
"loss": 6.518,
"mean_token_accuracy": 0.11185077950358391,
"num_tokens": 1209239.0,
"step": 660
},
{
"entropy": 6.6537374496459964,
"epoch": 0.0555462746408286,
"grad_norm": 1.09375,
"learning_rate": 0.00033200000000000005,
"loss": 6.4537,
"mean_token_accuracy": 0.104566191136837,
"num_tokens": 1218552.0,
"step": 665
},
{
"entropy": 6.640014934539795,
"epoch": 0.0559639158035416,
"grad_norm": 1.1953125,
"learning_rate": 0.00033450000000000005,
"loss": 6.5097,
"mean_token_accuracy": 0.10526655986905098,
"num_tokens": 1227632.0,
"step": 670
},
{
"entropy": 6.731128692626953,
"epoch": 0.056381556966254597,
"grad_norm": 1.1640625,
"learning_rate": 0.000337,
"loss": 6.5372,
"mean_token_accuracy": 0.10736859366297721,
"num_tokens": 1236433.0,
"step": 675
},
{
"entropy": 6.618252611160278,
"epoch": 0.05679919812896759,
"grad_norm": 1.3203125,
"learning_rate": 0.0003395,
"loss": 6.5768,
"mean_token_accuracy": 0.09474495351314545,
"num_tokens": 1245700.0,
"step": 680
},
{
"entropy": 6.727372455596924,
"epoch": 0.05721683929168059,
"grad_norm": 1.0625,
"learning_rate": 0.000342,
"loss": 6.3969,
"mean_token_accuracy": 0.10844348147511482,
"num_tokens": 1254206.0,
"step": 685
},
{
"entropy": 6.64897894859314,
"epoch": 0.05763448045439359,
"grad_norm": 1.0625,
"learning_rate": 0.00034449999999999997,
"loss": 6.4587,
"mean_token_accuracy": 0.10715654492378235,
"num_tokens": 1263302.0,
"step": 690
},
{
"entropy": 6.695715236663818,
"epoch": 0.05805212161710658,
"grad_norm": 1.15625,
"learning_rate": 0.000347,
"loss": 6.509,
"mean_token_accuracy": 0.1054201953113079,
"num_tokens": 1272998.0,
"step": 695
},
{
"entropy": 6.5739256858825685,
"epoch": 0.05846976277981958,
"grad_norm": 1.1875,
"learning_rate": 0.0003495,
"loss": 6.4765,
"mean_token_accuracy": 0.10761839896440506,
"num_tokens": 1281955.0,
"step": 700
},
{
"entropy": 6.726444578170776,
"epoch": 0.05888740394253258,
"grad_norm": 1.078125,
"learning_rate": 0.000352,
"loss": 6.4905,
"mean_token_accuracy": 0.10599962621927261,
"num_tokens": 1291219.0,
"step": 705
},
{
"entropy": 6.621089363098145,
"epoch": 0.05930504510524557,
"grad_norm": 1.1484375,
"learning_rate": 0.0003545,
"loss": 6.5137,
"mean_token_accuracy": 0.10294601619243622,
"num_tokens": 1300963.0,
"step": 710
},
{
"entropy": 6.677341604232788,
"epoch": 0.05972268626795857,
"grad_norm": 1.0703125,
"learning_rate": 0.000357,
"loss": 6.5359,
"mean_token_accuracy": 0.1052409827709198,
"num_tokens": 1310112.0,
"step": 715
},
{
"entropy": 6.599183893203735,
"epoch": 0.06014032743067157,
"grad_norm": 1.1875,
"learning_rate": 0.0003595,
"loss": 6.4664,
"mean_token_accuracy": 0.11334097385406494,
"num_tokens": 1319309.0,
"step": 720
},
{
"entropy": 6.576204824447632,
"epoch": 0.06055796859338457,
"grad_norm": 1.0859375,
"learning_rate": 0.000362,
"loss": 6.4372,
"mean_token_accuracy": 0.10994711145758629,
"num_tokens": 1329049.0,
"step": 725
},
{
"entropy": 6.535724353790283,
"epoch": 0.06097560975609756,
"grad_norm": 1.2421875,
"learning_rate": 0.0003645,
"loss": 6.2879,
"mean_token_accuracy": 0.11354263052344322,
"num_tokens": 1337770.0,
"step": 730
},
{
"entropy": 6.510106229782105,
"epoch": 0.06139325091881056,
"grad_norm": 1.1953125,
"learning_rate": 0.000367,
"loss": 6.3909,
"mean_token_accuracy": 0.11583516970276833,
"num_tokens": 1346705.0,
"step": 735
},
{
"entropy": 6.545791149139404,
"epoch": 0.06181089208152356,
"grad_norm": 1.2421875,
"learning_rate": 0.0003695,
"loss": 6.4397,
"mean_token_accuracy": 0.10958988890051842,
"num_tokens": 1354860.0,
"step": 740
},
{
"entropy": 6.596643924713135,
"epoch": 0.06222853324423655,
"grad_norm": 1.3203125,
"learning_rate": 0.000372,
"loss": 6.4185,
"mean_token_accuracy": 0.11004089862108231,
"num_tokens": 1364140.0,
"step": 745
},
{
"entropy": 6.651843929290772,
"epoch": 0.06264617440694954,
"grad_norm": 1.0859375,
"learning_rate": 0.0003745,
"loss": 6.5647,
"mean_token_accuracy": 0.10021075382828712,
"num_tokens": 1373645.0,
"step": 750
},
{
"entropy": 6.609305429458618,
"epoch": 0.06306381556966255,
"grad_norm": 1.0859375,
"learning_rate": 0.000377,
"loss": 6.4247,
"mean_token_accuracy": 0.11530799940228462,
"num_tokens": 1383291.0,
"step": 755
},
{
"entropy": 6.4960479736328125,
"epoch": 0.06348145673237554,
"grad_norm": 1.171875,
"learning_rate": 0.0003795,
"loss": 6.3551,
"mean_token_accuracy": 0.11109194234013557,
"num_tokens": 1391172.0,
"step": 760
},
{
"entropy": 6.569580030441284,
"epoch": 0.06389909789508853,
"grad_norm": 1.0234375,
"learning_rate": 0.000382,
"loss": 6.3463,
"mean_token_accuracy": 0.11715650334954261,
"num_tokens": 1400864.0,
"step": 765
},
{
"entropy": 6.5577380657196045,
"epoch": 0.06431673905780154,
"grad_norm": 1.046875,
"learning_rate": 0.0003845,
"loss": 6.4101,
"mean_token_accuracy": 0.11468178406357765,
"num_tokens": 1410570.0,
"step": 770
},
{
"entropy": 6.469339036941529,
"epoch": 0.06473438022051453,
"grad_norm": 1.0859375,
"learning_rate": 0.00038700000000000003,
"loss": 6.3944,
"mean_token_accuracy": 0.11272586584091186,
"num_tokens": 1419536.0,
"step": 775
},
{
"entropy": 6.4584187984466555,
"epoch": 0.06515202138322754,
"grad_norm": 1.140625,
"learning_rate": 0.00038950000000000003,
"loss": 6.3037,
"mean_token_accuracy": 0.11476944163441657,
"num_tokens": 1428000.0,
"step": 780
},
{
"entropy": 6.520781803131103,
"epoch": 0.06556966254594053,
"grad_norm": 1.1796875,
"learning_rate": 0.00039200000000000004,
"loss": 6.5022,
"mean_token_accuracy": 0.11618321910500526,
"num_tokens": 1437789.0,
"step": 785
},
{
"entropy": 6.464061641693116,
"epoch": 0.06598730370865352,
"grad_norm": 1.1015625,
"learning_rate": 0.00039450000000000005,
"loss": 6.3112,
"mean_token_accuracy": 0.11233830228447914,
"num_tokens": 1446848.0,
"step": 790
},
{
"entropy": 6.465522575378418,
"epoch": 0.06640494487136653,
"grad_norm": 1.1484375,
"learning_rate": 0.00039700000000000005,
"loss": 6.3565,
"mean_token_accuracy": 0.1105519786477089,
"num_tokens": 1455749.0,
"step": 795
},
{
"entropy": 6.481429529190064,
"epoch": 0.06682258603407952,
"grad_norm": 1.1796875,
"learning_rate": 0.0003995,
"loss": 6.2822,
"mean_token_accuracy": 0.1205395795404911,
"num_tokens": 1464556.0,
"step": 800
},
{
"entropy": 6.4752460479736325,
"epoch": 0.06724022719679251,
"grad_norm": 1.21875,
"learning_rate": 0.000402,
"loss": 6.4169,
"mean_token_accuracy": 0.1133482426404953,
"num_tokens": 1473617.0,
"step": 805
},
{
"entropy": 6.538545083999634,
"epoch": 0.06765786835950552,
"grad_norm": 1.1875,
"learning_rate": 0.0004045,
"loss": 6.4181,
"mean_token_accuracy": 0.11163719817996025,
"num_tokens": 1483006.0,
"step": 810
},
{
"entropy": 6.553357219696045,
"epoch": 0.06807550952221851,
"grad_norm": 1.1796875,
"learning_rate": 0.00040699999999999997,
"loss": 6.4388,
"mean_token_accuracy": 0.10800940170884132,
"num_tokens": 1493009.0,
"step": 815
},
{
"entropy": 6.474475574493408,
"epoch": 0.0684931506849315,
"grad_norm": 1.2578125,
"learning_rate": 0.0004095,
"loss": 6.3283,
"mean_token_accuracy": 0.11251509189605713,
"num_tokens": 1501261.0,
"step": 820
},
{
"entropy": 6.469084453582764,
"epoch": 0.06891079184764451,
"grad_norm": 0.984375,
"learning_rate": 0.000412,
"loss": 6.4256,
"mean_token_accuracy": 0.10885349065065383,
"num_tokens": 1511623.0,
"step": 825
},
{
"entropy": 6.471516227722168,
"epoch": 0.0693284330103575,
"grad_norm": 1.0625,
"learning_rate": 0.0004145,
"loss": 6.4375,
"mean_token_accuracy": 0.10788336396217346,
"num_tokens": 1521741.0,
"step": 830
},
{
"entropy": 6.568622636795044,
"epoch": 0.0697460741730705,
"grad_norm": 0.9765625,
"learning_rate": 0.000417,
"loss": 6.3519,
"mean_token_accuracy": 0.11642592176795005,
"num_tokens": 1530988.0,
"step": 835
},
{
"entropy": 6.395647430419922,
"epoch": 0.0701637153357835,
"grad_norm": 1.125,
"learning_rate": 0.0004195,
"loss": 6.2078,
"mean_token_accuracy": 0.11523091495037079,
"num_tokens": 1539415.0,
"step": 840
},
{
"entropy": 6.517097806930542,
"epoch": 0.07058135649849649,
"grad_norm": 1.1796875,
"learning_rate": 0.000422,
"loss": 6.4095,
"mean_token_accuracy": 0.1168409302830696,
"num_tokens": 1547969.0,
"step": 845
},
{
"entropy": 6.352308654785157,
"epoch": 0.07099899766120948,
"grad_norm": 1.0234375,
"learning_rate": 0.0004245,
"loss": 6.2502,
"mean_token_accuracy": 0.1196600504219532,
"num_tokens": 1557526.0,
"step": 850
},
{
"entropy": 6.366084146499634,
"epoch": 0.07141663882392249,
"grad_norm": 1.25,
"learning_rate": 0.000427,
"loss": 6.2262,
"mean_token_accuracy": 0.1241718977689743,
"num_tokens": 1565982.0,
"step": 855
},
{
"entropy": 6.376620531082153,
"epoch": 0.07183427998663548,
"grad_norm": 1.1484375,
"learning_rate": 0.0004295,
"loss": 6.2966,
"mean_token_accuracy": 0.12094038277864456,
"num_tokens": 1574492.0,
"step": 860
},
{
"entropy": 6.346592712402344,
"epoch": 0.07225192114934847,
"grad_norm": 1.0625,
"learning_rate": 0.000432,
"loss": 6.2522,
"mean_token_accuracy": 0.12188317254185677,
"num_tokens": 1583726.0,
"step": 865
},
{
"entropy": 6.353613567352295,
"epoch": 0.07266956231206148,
"grad_norm": 1.125,
"learning_rate": 0.0004345,
"loss": 6.2754,
"mean_token_accuracy": 0.10796950832009315,
"num_tokens": 1593780.0,
"step": 870
},
{
"entropy": 6.4174741268157955,
"epoch": 0.07308720347477447,
"grad_norm": 1.125,
"learning_rate": 0.000437,
"loss": 6.2736,
"mean_token_accuracy": 0.11892682611942292,
"num_tokens": 1602746.0,
"step": 875
},
{
"entropy": 6.391238355636597,
"epoch": 0.07350484463748747,
"grad_norm": 1.0703125,
"learning_rate": 0.0004395,
"loss": 6.303,
"mean_token_accuracy": 0.11670434921979904,
"num_tokens": 1611767.0,
"step": 880
},
{
"entropy": 6.4701165676116945,
"epoch": 0.07392248580020047,
"grad_norm": 0.94140625,
"learning_rate": 0.000442,
"loss": 6.3862,
"mean_token_accuracy": 0.11985925436019898,
"num_tokens": 1621568.0,
"step": 885
},
{
"entropy": 6.4399733543396,
"epoch": 0.07434012696291346,
"grad_norm": 1.03125,
"learning_rate": 0.0004445,
"loss": 6.3134,
"mean_token_accuracy": 0.12187910750508309,
"num_tokens": 1630650.0,
"step": 890
},
{
"entropy": 6.359479999542236,
"epoch": 0.07475776812562646,
"grad_norm": 1.0546875,
"learning_rate": 0.000447,
"loss": 6.3295,
"mean_token_accuracy": 0.11347245573997497,
"num_tokens": 1640482.0,
"step": 895
},
{
"entropy": 6.448725175857544,
"epoch": 0.07517540928833946,
"grad_norm": 1.1953125,
"learning_rate": 0.00044950000000000003,
"loss": 6.3712,
"mean_token_accuracy": 0.11579439342021942,
"num_tokens": 1649344.0,
"step": 900
},
{
"entropy": 6.3760090351104735,
"epoch": 0.07559305045105245,
"grad_norm": 1.0546875,
"learning_rate": 0.00045200000000000004,
"loss": 6.3006,
"mean_token_accuracy": 0.12386192455887794,
"num_tokens": 1658962.0,
"step": 905
},
{
"entropy": 6.3498400211334225,
"epoch": 0.07601069161376545,
"grad_norm": 1.046875,
"learning_rate": 0.00045450000000000004,
"loss": 6.3114,
"mean_token_accuracy": 0.11954073309898376,
"num_tokens": 1668615.0,
"step": 910
},
{
"entropy": 6.432513904571533,
"epoch": 0.07642833277647845,
"grad_norm": 1.0,
"learning_rate": 0.00045700000000000005,
"loss": 6.2568,
"mean_token_accuracy": 0.12355147078633308,
"num_tokens": 1679282.0,
"step": 915
},
{
"entropy": 6.294129085540772,
"epoch": 0.07684597393919144,
"grad_norm": 1.015625,
"learning_rate": 0.00045950000000000006,
"loss": 6.3482,
"mean_token_accuracy": 0.11906261742115021,
"num_tokens": 1688362.0,
"step": 920
},
{
"entropy": 6.388503551483154,
"epoch": 0.07726361510190444,
"grad_norm": 0.92578125,
"learning_rate": 0.000462,
"loss": 6.2429,
"mean_token_accuracy": 0.11470115780830384,
"num_tokens": 1697593.0,
"step": 925
},
{
"entropy": 6.37017035484314,
"epoch": 0.07768125626461744,
"grad_norm": 1.09375,
"learning_rate": 0.0004645,
"loss": 6.3006,
"mean_token_accuracy": 0.11924741566181182,
"num_tokens": 1706961.0,
"step": 930
},
{
"entropy": 6.479405403137207,
"epoch": 0.07809889742733044,
"grad_norm": 1.078125,
"learning_rate": 0.000467,
"loss": 6.3381,
"mean_token_accuracy": 0.12333182618021965,
"num_tokens": 1716222.0,
"step": 935
},
{
"entropy": 6.206763172149659,
"epoch": 0.07851653859004344,
"grad_norm": 1.1328125,
"learning_rate": 0.0004695,
"loss": 6.2058,
"mean_token_accuracy": 0.11398206949234009,
"num_tokens": 1726750.0,
"step": 940
},
{
"entropy": 6.3856278419494625,
"epoch": 0.07893417975275643,
"grad_norm": 0.95703125,
"learning_rate": 0.000472,
"loss": 6.3057,
"mean_token_accuracy": 0.11221355125308037,
"num_tokens": 1736262.0,
"step": 945
},
{
"entropy": 6.335583257675171,
"epoch": 0.07935182091546943,
"grad_norm": 1.078125,
"learning_rate": 0.0004745,
"loss": 6.3252,
"mean_token_accuracy": 0.1209489643573761,
"num_tokens": 1745555.0,
"step": 950
},
{
"entropy": 6.279627895355224,
"epoch": 0.07976946207818243,
"grad_norm": 1.109375,
"learning_rate": 0.000477,
"loss": 6.1916,
"mean_token_accuracy": 0.12183400169014931,
"num_tokens": 1754107.0,
"step": 955
},
{
"entropy": 6.405155515670776,
"epoch": 0.08018710324089542,
"grad_norm": 1.078125,
"learning_rate": 0.0004795,
"loss": 6.3739,
"mean_token_accuracy": 0.11344922855496406,
"num_tokens": 1762982.0,
"step": 960
},
{
"entropy": 6.309114837646485,
"epoch": 0.08060474440360842,
"grad_norm": 1.09375,
"learning_rate": 0.000482,
"loss": 6.2691,
"mean_token_accuracy": 0.12115986794233322,
"num_tokens": 1772545.0,
"step": 965
},
{
"entropy": 6.5117112159729,
"epoch": 0.08102238556632142,
"grad_norm": 0.9921875,
"learning_rate": 0.0004845,
"loss": 6.3958,
"mean_token_accuracy": 0.11408328413963317,
"num_tokens": 1782052.0,
"step": 970
},
{
"entropy": 6.217897701263428,
"epoch": 0.08144002672903441,
"grad_norm": 1.046875,
"learning_rate": 0.000487,
"loss": 6.1938,
"mean_token_accuracy": 0.12337508872151375,
"num_tokens": 1791682.0,
"step": 975
},
{
"entropy": 6.3203489780426025,
"epoch": 0.08185766789174741,
"grad_norm": 1.0390625,
"learning_rate": 0.0004895,
"loss": 6.2117,
"mean_token_accuracy": 0.1230445772409439,
"num_tokens": 1800526.0,
"step": 980
},
{
"entropy": 6.3890846252441404,
"epoch": 0.08227530905446041,
"grad_norm": 1.125,
"learning_rate": 0.000492,
"loss": 6.3486,
"mean_token_accuracy": 0.12185624241828918,
"num_tokens": 1810062.0,
"step": 985
},
{
"entropy": 6.2288158416748045,
"epoch": 0.0826929502171734,
"grad_norm": 1.171875,
"learning_rate": 0.0004945,
"loss": 6.3146,
"mean_token_accuracy": 0.12449721470475197,
"num_tokens": 1818909.0,
"step": 990
},
{
"entropy": 6.3776692867279055,
"epoch": 0.0831105913798864,
"grad_norm": 1.140625,
"learning_rate": 0.000497,
"loss": 6.2739,
"mean_token_accuracy": 0.12884685695171355,
"num_tokens": 1827978.0,
"step": 995
},
{
"entropy": 6.25860481262207,
"epoch": 0.0835282325425994,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995,
"loss": 6.2488,
"mean_token_accuracy": 0.12477765381336212,
"num_tokens": 1837955.0,
"step": 1000
},
{
"entropy": 6.30791540145874,
"epoch": 0.0839458737053124,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999999987395534,
"loss": 6.3305,
"mean_token_accuracy": 0.1183539167046547,
"num_tokens": 1847699.0,
"step": 1005
},
{
"entropy": 6.37754864692688,
"epoch": 0.08436351486802539,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999936189888,
"loss": 6.2603,
"mean_token_accuracy": 0.12056818008422851,
"num_tokens": 1856490.0,
"step": 1010
},
{
"entropy": 6.28082389831543,
"epoch": 0.0847811560307384,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999999845595285,
"loss": 6.2708,
"mean_token_accuracy": 0.12282358333468438,
"num_tokens": 1865644.0,
"step": 1015
},
{
"entropy": 6.315361738204956,
"epoch": 0.08519879719345139,
"grad_norm": 1.203125,
"learning_rate": 0.0004999999715611727,
"loss": 6.2108,
"mean_token_accuracy": 0.12240432277321815,
"num_tokens": 1874450.0,
"step": 1020
},
{
"entropy": 6.259311962127685,
"epoch": 0.08561643835616438,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999546239215,
"loss": 6.2188,
"mean_token_accuracy": 0.12367575094103814,
"num_tokens": 1883259.0,
"step": 1025
},
{
"entropy": 6.258958435058593,
"epoch": 0.08603407951887738,
"grad_norm": 1.0625,
"learning_rate": 0.0004999999337477753,
"loss": 6.2918,
"mean_token_accuracy": 0.12324620634317399,
"num_tokens": 1891897.0,
"step": 1030
},
{
"entropy": 6.379085683822632,
"epoch": 0.08645172068159038,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999999089327344,
"loss": 6.2741,
"mean_token_accuracy": 0.11758544445037841,
"num_tokens": 1901011.0,
"step": 1035
},
{
"entropy": 6.152502918243409,
"epoch": 0.08686936184430337,
"grad_norm": 1.171875,
"learning_rate": 0.0004999998801787993,
"loss": 6.1785,
"mean_token_accuracy": 0.12819710671901702,
"num_tokens": 1910132.0,
"step": 1040
},
{
"entropy": 6.345338630676269,
"epoch": 0.08728700300701638,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999998474859704,
"loss": 6.3036,
"mean_token_accuracy": 0.1242813840508461,
"num_tokens": 1919938.0,
"step": 1045
},
{
"entropy": 6.223077201843262,
"epoch": 0.08770464416972937,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999998108542483,
"loss": 6.1585,
"mean_token_accuracy": 0.12859595492482184,
"num_tokens": 1928199.0,
"step": 1050
},
{
"entropy": 6.240682601928711,
"epoch": 0.08812228533244236,
"grad_norm": 1.125,
"learning_rate": 0.0004999997702836339,
"loss": 6.1582,
"mean_token_accuracy": 0.1269869789481163,
"num_tokens": 1937133.0,
"step": 1055
},
{
"entropy": 6.340724658966065,
"epoch": 0.08853992649515537,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999997257741274,
"loss": 6.328,
"mean_token_accuracy": 0.12459098696708679,
"num_tokens": 1946580.0,
"step": 1060
},
{
"entropy": 6.245114326477051,
"epoch": 0.08895756765786836,
"grad_norm": 1.0703125,
"learning_rate": 0.00049999967732573,
"loss": 6.2332,
"mean_token_accuracy": 0.12014665976166725,
"num_tokens": 1955413.0,
"step": 1065
},
{
"entropy": 6.345558071136475,
"epoch": 0.08937520882058135,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999996249384425,
"loss": 6.3084,
"mean_token_accuracy": 0.12168630361557006,
"num_tokens": 1964716.0,
"step": 1070
},
{
"entropy": 6.22707314491272,
"epoch": 0.08979284998329436,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999995686122657,
"loss": 6.1647,
"mean_token_accuracy": 0.13024285659193993,
"num_tokens": 1974081.0,
"step": 1075
},
{
"entropy": 6.095598602294922,
"epoch": 0.09021049114600735,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999995083472006,
"loss": 6.135,
"mean_token_accuracy": 0.12958285063505173,
"num_tokens": 1983035.0,
"step": 1080
},
{
"entropy": 6.304395771026611,
"epoch": 0.09062813230872034,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999994441432481,
"loss": 6.1679,
"mean_token_accuracy": 0.13268934264779092,
"num_tokens": 1992475.0,
"step": 1085
},
{
"entropy": 6.179671812057495,
"epoch": 0.09104577347143335,
"grad_norm": 1.015625,
"learning_rate": 0.0004999993760004097,
"loss": 6.2474,
"mean_token_accuracy": 0.1233457788825035,
"num_tokens": 2002729.0,
"step": 1090
},
{
"entropy": 6.167952680587769,
"epoch": 0.09146341463414634,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999993039186864,
"loss": 6.0768,
"mean_token_accuracy": 0.1327129513025284,
"num_tokens": 2011319.0,
"step": 1095
},
{
"entropy": 6.2613760471344,
"epoch": 0.09188105579685933,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999992278980794,
"loss": 6.1096,
"mean_token_accuracy": 0.12951697111129762,
"num_tokens": 2020456.0,
"step": 1100
},
{
"entropy": 6.096225452423096,
"epoch": 0.09229869695957234,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999991479385901,
"loss": 6.21,
"mean_token_accuracy": 0.13094062879681587,
"num_tokens": 2029417.0,
"step": 1105
},
{
"entropy": 6.244110536575318,
"epoch": 0.09271633812228533,
"grad_norm": 1.125,
"learning_rate": 0.0004999990640402198,
"loss": 6.2909,
"mean_token_accuracy": 0.1225017212331295,
"num_tokens": 2039290.0,
"step": 1110
},
{
"entropy": 6.264288282394409,
"epoch": 0.09313397928499834,
"grad_norm": 1.109375,
"learning_rate": 0.0004999989762029701,
"loss": 6.1195,
"mean_token_accuracy": 0.13131056353449821,
"num_tokens": 2048485.0,
"step": 1115
},
{
"entropy": 6.046679449081421,
"epoch": 0.09355162044771133,
"grad_norm": 1.046875,
"learning_rate": 0.0004999988844268425,
"loss": 6.0778,
"mean_token_accuracy": 0.13759607523679734,
"num_tokens": 2057131.0,
"step": 1120
},
{
"entropy": 6.173616600036621,
"epoch": 0.09396926161042432,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999987887118387,
"loss": 6.088,
"mean_token_accuracy": 0.12637848854064943,
"num_tokens": 2066295.0,
"step": 1125
},
{
"entropy": 6.116608095169068,
"epoch": 0.09438690277313733,
"grad_norm": 1.078125,
"learning_rate": 0.0004999986890579602,
"loss": 6.0789,
"mean_token_accuracy": 0.13630023896694182,
"num_tokens": 2075656.0,
"step": 1130
},
{
"entropy": 6.201944017410279,
"epoch": 0.09480454393585032,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999985854652087,
"loss": 6.2334,
"mean_token_accuracy": 0.12537547051906586,
"num_tokens": 2086046.0,
"step": 1135
},
{
"entropy": 6.252005624771118,
"epoch": 0.09522218509856331,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999984779335862,
"loss": 6.24,
"mean_token_accuracy": 0.1255707710981369,
"num_tokens": 2094808.0,
"step": 1140
},
{
"entropy": 6.133102750778198,
"epoch": 0.09563982626127632,
"grad_norm": 0.9375,
"learning_rate": 0.0004999983664630945,
"loss": 6.1696,
"mean_token_accuracy": 0.13095618411898613,
"num_tokens": 2104155.0,
"step": 1145
},
{
"entropy": 6.21075177192688,
"epoch": 0.09605746742398931,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999982510537357,
"loss": 6.0633,
"mean_token_accuracy": 0.13114586398005484,
"num_tokens": 2112517.0,
"step": 1150
},
{
"entropy": 6.032001113891601,
"epoch": 0.0964751085867023,
"grad_norm": 1.09375,
"learning_rate": 0.0004999981317055116,
"loss": 6.0206,
"mean_token_accuracy": 0.13395098671317102,
"num_tokens": 2121355.0,
"step": 1155
},
{
"entropy": 6.203887701034546,
"epoch": 0.09689274974941531,
"grad_norm": 1.109375,
"learning_rate": 0.0004999980084184244,
"loss": 6.1515,
"mean_token_accuracy": 0.13314087614417075,
"num_tokens": 2129278.0,
"step": 1160
},
{
"entropy": 6.228727769851685,
"epoch": 0.0973103909121283,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999978811924762,
"loss": 6.2993,
"mean_token_accuracy": 0.12339182645082473,
"num_tokens": 2139548.0,
"step": 1165
},
{
"entropy": 6.1066957950592045,
"epoch": 0.09772803207484129,
"grad_norm": 1.078125,
"learning_rate": 0.0004999977500276693,
"loss": 6.2067,
"mean_token_accuracy": 0.12968555092811584,
"num_tokens": 2148765.0,
"step": 1170
},
{
"entropy": 6.299677038192749,
"epoch": 0.0981456732375543,
"grad_norm": 1.046875,
"learning_rate": 0.0004999976149240058,
"loss": 6.1138,
"mean_token_accuracy": 0.1367984600365162,
"num_tokens": 2157482.0,
"step": 1175
},
{
"entropy": 6.137264728546143,
"epoch": 0.09856331440026729,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999974758814884,
"loss": 6.2185,
"mean_token_accuracy": 0.12485746145248414,
"num_tokens": 2167320.0,
"step": 1180
},
{
"entropy": 6.206802892684936,
"epoch": 0.09898095556298028,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999973329001193,
"loss": 6.16,
"mean_token_accuracy": 0.1285059556365013,
"num_tokens": 2175454.0,
"step": 1185
},
{
"entropy": 6.2009917259216305,
"epoch": 0.09939859672569329,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999971859799009,
"loss": 6.101,
"mean_token_accuracy": 0.1268243044614792,
"num_tokens": 2184391.0,
"step": 1190
},
{
"entropy": 6.065373229980469,
"epoch": 0.09981623788840628,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999970351208361,
"loss": 5.9825,
"mean_token_accuracy": 0.13687402829527856,
"num_tokens": 2193309.0,
"step": 1195
},
{
"entropy": 6.1508321285247805,
"epoch": 0.10023387905111927,
"grad_norm": 1.0,
"learning_rate": 0.0004999968803229275,
"loss": 6.1157,
"mean_token_accuracy": 0.13073571994900704,
"num_tokens": 2202057.0,
"step": 1200
},
{
"entropy": 6.1929127216339115,
"epoch": 0.10065152021383228,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999967215861775,
"loss": 6.2279,
"mean_token_accuracy": 0.12980359345674514,
"num_tokens": 2212384.0,
"step": 1205
},
{
"entropy": 6.194336175918579,
"epoch": 0.10106916137654527,
"grad_norm": 1.078125,
"learning_rate": 0.000499996558910589,
"loss": 6.1376,
"mean_token_accuracy": 0.1303119406104088,
"num_tokens": 2221451.0,
"step": 1210
},
{
"entropy": 6.04097547531128,
"epoch": 0.10148680253925826,
"grad_norm": 1.078125,
"learning_rate": 0.000499996392296165,
"loss": 6.0355,
"mean_token_accuracy": 0.13070774227380752,
"num_tokens": 2230440.0,
"step": 1215
},
{
"entropy": 6.18285231590271,
"epoch": 0.10190444370197127,
"grad_norm": 1.0625,
"learning_rate": 0.0004999962217429083,
"loss": 6.2252,
"mean_token_accuracy": 0.12367333248257636,
"num_tokens": 2239985.0,
"step": 1220
},
{
"entropy": 6.184245300292969,
"epoch": 0.10232208486468426,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999960472508219,
"loss": 6.0915,
"mean_token_accuracy": 0.12686059027910232,
"num_tokens": 2250391.0,
"step": 1225
},
{
"entropy": 6.161716556549072,
"epoch": 0.10273972602739725,
"grad_norm": 1.03125,
"learning_rate": 0.000499995868819909,
"loss": 6.1891,
"mean_token_accuracy": 0.12735605537891387,
"num_tokens": 2260620.0,
"step": 1230
},
{
"entropy": 6.183428859710693,
"epoch": 0.10315736719011026,
"grad_norm": 0.921875,
"learning_rate": 0.0004999956864501725,
"loss": 6.1651,
"mean_token_accuracy": 0.1268216535449028,
"num_tokens": 2271176.0,
"step": 1235
},
{
"entropy": 6.244471549987793,
"epoch": 0.10357500835282325,
"grad_norm": 1.0625,
"learning_rate": 0.0004999955001416156,
"loss": 6.0675,
"mean_token_accuracy": 0.13557338193058968,
"num_tokens": 2280371.0,
"step": 1240
},
{
"entropy": 5.979556226730347,
"epoch": 0.10399264951553625,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999953098942417,
"loss": 6.0192,
"mean_token_accuracy": 0.13581715449690818,
"num_tokens": 2289015.0,
"step": 1245
},
{
"entropy": 6.061905384063721,
"epoch": 0.10441029067824925,
"grad_norm": 1.140625,
"learning_rate": 0.0004999951157080541,
"loss": 6.1043,
"mean_token_accuracy": 0.13418104201555253,
"num_tokens": 2297409.0,
"step": 1250
},
{
"entropy": 6.081577634811401,
"epoch": 0.10482793184096224,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999949175830561,
"loss": 6.1405,
"mean_token_accuracy": 0.12714434340596198,
"num_tokens": 2307017.0,
"step": 1255
},
{
"entropy": 6.15795488357544,
"epoch": 0.10524557300367524,
"grad_norm": 1.0625,
"learning_rate": 0.0004999947155192514,
"loss": 5.9629,
"mean_token_accuracy": 0.14337851256132125,
"num_tokens": 2316014.0,
"step": 1260
},
{
"entropy": 6.072656536102295,
"epoch": 0.10566321416638824,
"grad_norm": 1.09375,
"learning_rate": 0.0004999945095166433,
"loss": 6.0679,
"mean_token_accuracy": 0.13084115758538245,
"num_tokens": 2326326.0,
"step": 1265
},
{
"entropy": 6.264087438583374,
"epoch": 0.10608085532910123,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999942995752354,
"loss": 6.1629,
"mean_token_accuracy": 0.12950412258505822,
"num_tokens": 2335108.0,
"step": 1270
},
{
"entropy": 6.085091543197632,
"epoch": 0.10649849649181423,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999940856950315,
"loss": 6.0949,
"mean_token_accuracy": 0.1381733551621437,
"num_tokens": 2343971.0,
"step": 1275
},
{
"entropy": 6.127774620056153,
"epoch": 0.10691613765452723,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999938678760354,
"loss": 6.1451,
"mean_token_accuracy": 0.12672161757946016,
"num_tokens": 2353742.0,
"step": 1280
},
{
"entropy": 6.142896223068237,
"epoch": 0.10733377881724022,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999936461182507,
"loss": 6.0859,
"mean_token_accuracy": 0.1363840654492378,
"num_tokens": 2363629.0,
"step": 1285
},
{
"entropy": 5.952917051315308,
"epoch": 0.10775141997995323,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999934204216814,
"loss": 5.9475,
"mean_token_accuracy": 0.13911077901721,
"num_tokens": 2372705.0,
"step": 1290
},
{
"entropy": 6.111120700836182,
"epoch": 0.10816906114266622,
"grad_norm": 1.109375,
"learning_rate": 0.0004999931907863315,
"loss": 6.0691,
"mean_token_accuracy": 0.13577704951167108,
"num_tokens": 2382136.0,
"step": 1295
},
{
"entropy": 5.975588846206665,
"epoch": 0.10858670230537922,
"grad_norm": 1.0859375,
"learning_rate": 0.000499992957212205,
"loss": 6.0286,
"mean_token_accuracy": 0.1295459158718586,
"num_tokens": 2391060.0,
"step": 1300
},
{
"entropy": 6.16234655380249,
"epoch": 0.10900434346809222,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999927196993059,
"loss": 6.1797,
"mean_token_accuracy": 0.12658073604106904,
"num_tokens": 2399885.0,
"step": 1305
},
{
"entropy": 6.155429744720459,
"epoch": 0.10942198463080521,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999924782476384,
"loss": 6.0237,
"mean_token_accuracy": 0.13930752426385878,
"num_tokens": 2409654.0,
"step": 1310
},
{
"entropy": 6.133743381500244,
"epoch": 0.1098396257935182,
"grad_norm": 1.0625,
"learning_rate": 0.0004999922328572068,
"loss": 6.1013,
"mean_token_accuracy": 0.1349839001893997,
"num_tokens": 2419342.0,
"step": 1315
},
{
"entropy": 5.979508876800537,
"epoch": 0.11025726695623121,
"grad_norm": 1.0,
"learning_rate": 0.0004999919835280154,
"loss": 5.9811,
"mean_token_accuracy": 0.1363253116607666,
"num_tokens": 2428845.0,
"step": 1320
},
{
"entropy": 6.172998523712158,
"epoch": 0.1106749081189442,
"grad_norm": 1.140625,
"learning_rate": 0.0004999917302600684,
"loss": 5.9856,
"mean_token_accuracy": 0.14307858645915986,
"num_tokens": 2437134.0,
"step": 1325
},
{
"entropy": 5.883855867385864,
"epoch": 0.1110925492816572,
"grad_norm": 1.046875,
"learning_rate": 0.0004999914730533704,
"loss": 5.8203,
"mean_token_accuracy": 0.14599396288394928,
"num_tokens": 2445438.0,
"step": 1330
},
{
"entropy": 6.0954326629638675,
"epoch": 0.1115101904443702,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999912119079258,
"loss": 6.1545,
"mean_token_accuracy": 0.13536207303404807,
"num_tokens": 2455039.0,
"step": 1335
},
{
"entropy": 6.078612756729126,
"epoch": 0.1119278316070832,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999909468237393,
"loss": 6.0842,
"mean_token_accuracy": 0.13297579810023308,
"num_tokens": 2465170.0,
"step": 1340
},
{
"entropy": 5.97861909866333,
"epoch": 0.11234547276979619,
"grad_norm": 1.03125,
"learning_rate": 0.0004999906778008153,
"loss": 5.8295,
"mean_token_accuracy": 0.14431577324867248,
"num_tokens": 2473688.0,
"step": 1345
},
{
"entropy": 5.943620347976685,
"epoch": 0.11276311393250919,
"grad_norm": 1.015625,
"learning_rate": 0.0004999904048391588,
"loss": 6.093,
"mean_token_accuracy": 0.1311701536178589,
"num_tokens": 2482335.0,
"step": 1350
},
{
"entropy": 6.156413507461548,
"epoch": 0.11318075509522219,
"grad_norm": 1.0625,
"learning_rate": 0.0004999901279387743,
"loss": 6.0151,
"mean_token_accuracy": 0.1401954136788845,
"num_tokens": 2491808.0,
"step": 1355
},
{
"entropy": 6.1728456020355225,
"epoch": 0.11359839625793518,
"grad_norm": 0.9375,
"learning_rate": 0.0004999898470996669,
"loss": 6.1564,
"mean_token_accuracy": 0.1268514558672905,
"num_tokens": 2501422.0,
"step": 1360
},
{
"entropy": 6.035900592803955,
"epoch": 0.11401603742064818,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999895623218415,
"loss": 5.9552,
"mean_token_accuracy": 0.1372804895043373,
"num_tokens": 2511201.0,
"step": 1365
},
{
"entropy": 6.070261144638062,
"epoch": 0.11443367858336118,
"grad_norm": 1.140625,
"learning_rate": 0.0004999892736053027,
"loss": 5.9907,
"mean_token_accuracy": 0.13410537466406822,
"num_tokens": 2519798.0,
"step": 1370
},
{
"entropy": 6.010339260101318,
"epoch": 0.11485131974607417,
"grad_norm": 0.96875,
"learning_rate": 0.0004999889809500561,
"loss": 5.9764,
"mean_token_accuracy": 0.14055416211485863,
"num_tokens": 2529177.0,
"step": 1375
},
{
"entropy": 6.112850570678711,
"epoch": 0.11526896090878717,
"grad_norm": 1.0625,
"learning_rate": 0.0004999886843561065,
"loss": 5.9685,
"mean_token_accuracy": 0.1412968449294567,
"num_tokens": 2537571.0,
"step": 1380
},
{
"entropy": 5.973147058486939,
"epoch": 0.11568660207150017,
"grad_norm": 0.96875,
"learning_rate": 0.0004999883838234592,
"loss": 5.8842,
"mean_token_accuracy": 0.1444699227809906,
"num_tokens": 2546258.0,
"step": 1385
},
{
"entropy": 5.95906081199646,
"epoch": 0.11610424323421316,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999880793521194,
"loss": 5.913,
"mean_token_accuracy": 0.13197560086846352,
"num_tokens": 2555862.0,
"step": 1390
},
{
"entropy": 5.864916181564331,
"epoch": 0.11652188439692616,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999877709420924,
"loss": 5.9164,
"mean_token_accuracy": 0.14239994511008264,
"num_tokens": 2564583.0,
"step": 1395
},
{
"entropy": 6.036420249938965,
"epoch": 0.11693952555963916,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999874585933837,
"loss": 6.0084,
"mean_token_accuracy": 0.14236027300357817,
"num_tokens": 2574217.0,
"step": 1400
},
{
"entropy": 6.026413488388061,
"epoch": 0.11735716672235215,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999871423059987,
"loss": 5.9206,
"mean_token_accuracy": 0.1449699342250824,
"num_tokens": 2582821.0,
"step": 1405
},
{
"entropy": 5.989486932754517,
"epoch": 0.11777480788506516,
"grad_norm": 0.9765625,
"learning_rate": 0.000499986822079943,
"loss": 6.0776,
"mean_token_accuracy": 0.13281765207648277,
"num_tokens": 2592748.0,
"step": 1410
},
{
"entropy": 6.07455530166626,
"epoch": 0.11819244904777815,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999864979152221,
"loss": 5.9816,
"mean_token_accuracy": 0.14064599573612213,
"num_tokens": 2602053.0,
"step": 1415
},
{
"entropy": 5.99542646408081,
"epoch": 0.11861009021049114,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999861698118417,
"loss": 6.054,
"mean_token_accuracy": 0.13785783499479293,
"num_tokens": 2610557.0,
"step": 1420
},
{
"entropy": 6.072488784790039,
"epoch": 0.11902773137320415,
"grad_norm": 1.0,
"learning_rate": 0.0004999858377698076,
"loss": 5.9501,
"mean_token_accuracy": 0.13496977612376213,
"num_tokens": 2619628.0,
"step": 1425
},
{
"entropy": 6.129187536239624,
"epoch": 0.11944537253591714,
"grad_norm": 1.109375,
"learning_rate": 0.0004999855017891256,
"loss": 6.1036,
"mean_token_accuracy": 0.134950502961874,
"num_tokens": 2629284.0,
"step": 1430
},
{
"entropy": 5.965825462341309,
"epoch": 0.11986301369863013,
"grad_norm": 1.09375,
"learning_rate": 0.0004999851618698017,
"loss": 6.0273,
"mean_token_accuracy": 0.14388199001550675,
"num_tokens": 2638001.0,
"step": 1435
},
{
"entropy": 6.06349925994873,
"epoch": 0.12028065486134314,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999848180118416,
"loss": 6.0182,
"mean_token_accuracy": 0.14972263872623442,
"num_tokens": 2647071.0,
"step": 1440
},
{
"entropy": 6.035121726989746,
"epoch": 0.12069829602405613,
"grad_norm": 1.015625,
"learning_rate": 0.0004999844702152516,
"loss": 5.9926,
"mean_token_accuracy": 0.14700488597154618,
"num_tokens": 2655826.0,
"step": 1445
},
{
"entropy": 6.019858503341675,
"epoch": 0.12111593718676913,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999841184800374,
"loss": 5.8717,
"mean_token_accuracy": 0.1444035328924656,
"num_tokens": 2664431.0,
"step": 1450
},
{
"entropy": 5.853117942810059,
"epoch": 0.12153357834948213,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999837628062055,
"loss": 5.7793,
"mean_token_accuracy": 0.14595547914505005,
"num_tokens": 2673631.0,
"step": 1455
},
{
"entropy": 5.931141424179077,
"epoch": 0.12195121951219512,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999834031937621,
"loss": 6.067,
"mean_token_accuracy": 0.1372426927089691,
"num_tokens": 2683363.0,
"step": 1460
},
{
"entropy": 6.112060260772705,
"epoch": 0.12236886067490813,
"grad_norm": 0.984375,
"learning_rate": 0.0004999830396427133,
"loss": 6.015,
"mean_token_accuracy": 0.13856782242655755,
"num_tokens": 2692533.0,
"step": 1465
},
{
"entropy": 5.99221658706665,
"epoch": 0.12278650183762112,
"grad_norm": 1.109375,
"learning_rate": 0.0004999826721530657,
"loss": 5.9993,
"mean_token_accuracy": 0.14568220674991608,
"num_tokens": 2701211.0,
"step": 1470
},
{
"entropy": 6.068982028961182,
"epoch": 0.12320414300033411,
"grad_norm": 1.03125,
"learning_rate": 0.0004999823007248255,
"loss": 5.9933,
"mean_token_accuracy": 0.14043085724115373,
"num_tokens": 2709543.0,
"step": 1475
},
{
"entropy": 6.061050462722778,
"epoch": 0.12362178416304712,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999819253579992,
"loss": 6.0117,
"mean_token_accuracy": 0.13802490904927253,
"num_tokens": 2718336.0,
"step": 1480
},
{
"entropy": 5.954245090484619,
"epoch": 0.12403942532576011,
"grad_norm": 1.03125,
"learning_rate": 0.0004999815460525938,
"loss": 5.9627,
"mean_token_accuracy": 0.14610837548971176,
"num_tokens": 2726504.0,
"step": 1485
},
{
"entropy": 6.027486419677734,
"epoch": 0.1244570664884731,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999811628086155,
"loss": 5.9638,
"mean_token_accuracy": 0.1369314007461071,
"num_tokens": 2736076.0,
"step": 1490
},
{
"entropy": 5.88101716041565,
"epoch": 0.1248747076511861,
"grad_norm": 0.9921875,
"learning_rate": 0.000499980775626071,
"loss": 5.8929,
"mean_token_accuracy": 0.15019335895776748,
"num_tokens": 2745497.0,
"step": 1495
},
{
"entropy": 5.989426612854004,
"epoch": 0.12529234881389908,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999803845049674,
"loss": 6.0232,
"mean_token_accuracy": 0.1404777944087982,
"num_tokens": 2754455.0,
"step": 1500
},
{
"entropy": 6.1398618698120115,
"epoch": 0.1257099899766121,
"grad_norm": 0.953125,
"learning_rate": 0.0004999799894453112,
"loss": 6.0603,
"mean_token_accuracy": 0.1374159500002861,
"num_tokens": 2763603.0,
"step": 1505
},
{
"entropy": 5.951066875457764,
"epoch": 0.1261276311393251,
"grad_norm": 1.0625,
"learning_rate": 0.0004999795904471095,
"loss": 5.9195,
"mean_token_accuracy": 0.1388728216290474,
"num_tokens": 2772152.0,
"step": 1510
},
{
"entropy": 6.14337387084961,
"epoch": 0.12654527230203808,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999791875103694,
"loss": 6.0086,
"mean_token_accuracy": 0.1470729224383831,
"num_tokens": 2781459.0,
"step": 1515
},
{
"entropy": 6.0223418235778805,
"epoch": 0.12696291346475108,
"grad_norm": 1.046875,
"learning_rate": 0.0004999787806350977,
"loss": 5.9746,
"mean_token_accuracy": 0.13922247141599656,
"num_tokens": 2791177.0,
"step": 1520
},
{
"entropy": 5.888947582244873,
"epoch": 0.1273805546274641,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999783698213016,
"loss": 5.9436,
"mean_token_accuracy": 0.14956361576914787,
"num_tokens": 2800543.0,
"step": 1525
},
{
"entropy": 5.90831561088562,
"epoch": 0.12779819579017707,
"grad_norm": 1.09375,
"learning_rate": 0.0004999779550689885,
"loss": 5.7998,
"mean_token_accuracy": 0.147525854408741,
"num_tokens": 2809301.0,
"step": 1530
},
{
"entropy": 6.075090217590332,
"epoch": 0.12821583695289007,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999775363781654,
"loss": 6.0128,
"mean_token_accuracy": 0.13644266203045846,
"num_tokens": 2818269.0,
"step": 1535
},
{
"entropy": 5.899623394012451,
"epoch": 0.12863347811560308,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999771137488395,
"loss": 5.8811,
"mean_token_accuracy": 0.14791993498802186,
"num_tokens": 2827170.0,
"step": 1540
},
{
"entropy": 5.894683170318603,
"epoch": 0.12905111927831608,
"grad_norm": 1.09375,
"learning_rate": 0.0004999766871810188,
"loss": 5.8697,
"mean_token_accuracy": 0.14779433608055115,
"num_tokens": 2835320.0,
"step": 1545
},
{
"entropy": 5.993678569793701,
"epoch": 0.12946876044102906,
"grad_norm": 1.125,
"learning_rate": 0.0004999762566747101,
"loss": 5.9535,
"mean_token_accuracy": 0.14536867663264275,
"num_tokens": 2843755.0,
"step": 1550
},
{
"entropy": 6.051535272598267,
"epoch": 0.12988640160374207,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999758222299214,
"loss": 5.9713,
"mean_token_accuracy": 0.13523470386862754,
"num_tokens": 2853462.0,
"step": 1555
},
{
"entropy": 5.948653268814087,
"epoch": 0.13030404276645507,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999753838466601,
"loss": 5.887,
"mean_token_accuracy": 0.14211906045675277,
"num_tokens": 2863041.0,
"step": 1560
},
{
"entropy": 6.120198440551758,
"epoch": 0.13072168392916805,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999749415249338,
"loss": 6.1009,
"mean_token_accuracy": 0.1411801263689995,
"num_tokens": 2873019.0,
"step": 1565
},
{
"entropy": 5.8590634822845455,
"epoch": 0.13113932509188106,
"grad_norm": 1.078125,
"learning_rate": 0.0004999744952647505,
"loss": 5.8654,
"mean_token_accuracy": 0.1394391618669033,
"num_tokens": 2881467.0,
"step": 1570
},
{
"entropy": 6.0079615116119385,
"epoch": 0.13155696625459407,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999740450661176,
"loss": 5.9389,
"mean_token_accuracy": 0.14143842533230783,
"num_tokens": 2890914.0,
"step": 1575
},
{
"entropy": 5.827870559692383,
"epoch": 0.13197460741730704,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999735909290436,
"loss": 5.8623,
"mean_token_accuracy": 0.14539902061223983,
"num_tokens": 2901164.0,
"step": 1580
},
{
"entropy": 6.032009553909302,
"epoch": 0.13239224858002005,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999731328535358,
"loss": 5.9985,
"mean_token_accuracy": 0.1358782656490803,
"num_tokens": 2910560.0,
"step": 1585
},
{
"entropy": 5.878810119628906,
"epoch": 0.13280988974273306,
"grad_norm": 1.09375,
"learning_rate": 0.0004999726708396026,
"loss": 5.8151,
"mean_token_accuracy": 0.14725780636072158,
"num_tokens": 2918980.0,
"step": 1590
},
{
"entropy": 6.005957126617432,
"epoch": 0.13322753090544603,
"grad_norm": 0.91796875,
"learning_rate": 0.000499972204887252,
"loss": 6.0063,
"mean_token_accuracy": 0.13966943323612213,
"num_tokens": 2929516.0,
"step": 1595
},
{
"entropy": 5.949875402450561,
"epoch": 0.13364517206815904,
"grad_norm": 1.0,
"learning_rate": 0.0004999717349964922,
"loss": 6.0912,
"mean_token_accuracy": 0.1308370091021061,
"num_tokens": 2940603.0,
"step": 1600
},
{
"entropy": 6.106319904327393,
"epoch": 0.13406281323087205,
"grad_norm": 1.03125,
"learning_rate": 0.0004999712611673313,
"loss": 5.9359,
"mean_token_accuracy": 0.14399992525577546,
"num_tokens": 2950082.0,
"step": 1605
},
{
"entropy": 5.812672138214111,
"epoch": 0.13448045439358502,
"grad_norm": 1.09375,
"learning_rate": 0.0004999707833997777,
"loss": 5.8116,
"mean_token_accuracy": 0.154158915579319,
"num_tokens": 2959098.0,
"step": 1610
},
{
"entropy": 6.02878098487854,
"epoch": 0.13489809555629803,
"grad_norm": 1.015625,
"learning_rate": 0.0004999703016938398,
"loss": 5.9241,
"mean_token_accuracy": 0.14911588430404663,
"num_tokens": 2968585.0,
"step": 1615
},
{
"entropy": 5.894638872146606,
"epoch": 0.13531573671901104,
"grad_norm": 0.9453125,
"learning_rate": 0.000499969816049526,
"loss": 5.9672,
"mean_token_accuracy": 0.14088482335209845,
"num_tokens": 2978430.0,
"step": 1620
},
{
"entropy": 6.1249841213226315,
"epoch": 0.13573337788172402,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999693264668446,
"loss": 5.9966,
"mean_token_accuracy": 0.13702865466475486,
"num_tokens": 2988398.0,
"step": 1625
},
{
"entropy": 5.973216581344604,
"epoch": 0.13615101904443702,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999688329458045,
"loss": 6.031,
"mean_token_accuracy": 0.13957799673080445,
"num_tokens": 2999255.0,
"step": 1630
},
{
"entropy": 6.0353431224823,
"epoch": 0.13656866020715003,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999683354864141,
"loss": 5.9203,
"mean_token_accuracy": 0.14319872185587884,
"num_tokens": 3008638.0,
"step": 1635
},
{
"entropy": 5.878868436813354,
"epoch": 0.136986301369863,
"grad_norm": 1.109375,
"learning_rate": 0.0004999678340886823,
"loss": 5.9299,
"mean_token_accuracy": 0.14424165040254594,
"num_tokens": 3018070.0,
"step": 1640
},
{
"entropy": 5.857062816619873,
"epoch": 0.137403942532576,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999673287526176,
"loss": 5.8894,
"mean_token_accuracy": 0.14291651770472527,
"num_tokens": 3027878.0,
"step": 1645
},
{
"entropy": 6.106118249893188,
"epoch": 0.13782158369528902,
"grad_norm": 1.0625,
"learning_rate": 0.0004999668194782291,
"loss": 6.0733,
"mean_token_accuracy": 0.14018596187233925,
"num_tokens": 3037031.0,
"step": 1650
},
{
"entropy": 6.135175895690918,
"epoch": 0.138239224858002,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999663062655257,
"loss": 6.0063,
"mean_token_accuracy": 0.13246337324380875,
"num_tokens": 3046311.0,
"step": 1655
},
{
"entropy": 5.940270566940308,
"epoch": 0.138656866020715,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999657891145162,
"loss": 5.8304,
"mean_token_accuracy": 0.14716822504997254,
"num_tokens": 3055108.0,
"step": 1660
},
{
"entropy": 5.826205778121948,
"epoch": 0.139074507183428,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999652680252099,
"loss": 5.8513,
"mean_token_accuracy": 0.14699818193912506,
"num_tokens": 3063492.0,
"step": 1665
},
{
"entropy": 6.048397064208984,
"epoch": 0.139492148346141,
"grad_norm": 1.015625,
"learning_rate": 0.0004999647429976157,
"loss": 5.9303,
"mean_token_accuracy": 0.1480340465903282,
"num_tokens": 3072572.0,
"step": 1670
},
{
"entropy": 5.992930126190186,
"epoch": 0.139909789508854,
"grad_norm": 0.9375,
"learning_rate": 0.0004999642140317429,
"loss": 5.9678,
"mean_token_accuracy": 0.14264177456498145,
"num_tokens": 3082942.0,
"step": 1675
},
{
"entropy": 5.970309495925903,
"epoch": 0.140327430671567,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999636811276007,
"loss": 5.9357,
"mean_token_accuracy": 0.14391134530305863,
"num_tokens": 3091410.0,
"step": 1680
},
{
"entropy": 6.008646440505982,
"epoch": 0.14074507183427998,
"grad_norm": 1.0,
"learning_rate": 0.0004999631442851986,
"loss": 5.9969,
"mean_token_accuracy": 0.1431201532483101,
"num_tokens": 3100626.0,
"step": 1685
},
{
"entropy": 5.941483640670777,
"epoch": 0.14116271299699298,
"grad_norm": 1.0625,
"learning_rate": 0.0004999626035045458,
"loss": 5.7996,
"mean_token_accuracy": 0.1511163219809532,
"num_tokens": 3109339.0,
"step": 1690
},
{
"entropy": 5.984173202514649,
"epoch": 0.141580354159706,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999620587856519,
"loss": 5.9358,
"mean_token_accuracy": 0.1376422181725502,
"num_tokens": 3118636.0,
"step": 1695
},
{
"entropy": 5.928273868560791,
"epoch": 0.14199799532241897,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999615101285262,
"loss": 5.9019,
"mean_token_accuracy": 0.14366057217121125,
"num_tokens": 3128162.0,
"step": 1700
},
{
"entropy": 5.889788866043091,
"epoch": 0.14241563648513197,
"grad_norm": 1.03125,
"learning_rate": 0.0004999609575331786,
"loss": 5.8986,
"mean_token_accuracy": 0.1415964312851429,
"num_tokens": 3136451.0,
"step": 1705
},
{
"entropy": 6.013433027267456,
"epoch": 0.14283327764784498,
"grad_norm": 1.09375,
"learning_rate": 0.0004999604009996186,
"loss": 5.8999,
"mean_token_accuracy": 0.14590146243572236,
"num_tokens": 3146599.0,
"step": 1710
},
{
"entropy": 5.959149312973023,
"epoch": 0.14325091881055796,
"grad_norm": 1.0234375,
"learning_rate": 0.000499959840527856,
"loss": 5.9603,
"mean_token_accuracy": 0.14268366768956184,
"num_tokens": 3155941.0,
"step": 1715
},
{
"entropy": 5.918374681472779,
"epoch": 0.14366855997327097,
"grad_norm": 1.15625,
"learning_rate": 0.0004999592761179007,
"loss": 5.8897,
"mean_token_accuracy": 0.14959860891103743,
"num_tokens": 3164641.0,
"step": 1720
},
{
"entropy": 5.9503552436828615,
"epoch": 0.14408620113598397,
"grad_norm": 1.046875,
"learning_rate": 0.0004999587077697624,
"loss": 5.9235,
"mean_token_accuracy": 0.15331145748496056,
"num_tokens": 3173701.0,
"step": 1725
},
{
"entropy": 5.859464836120606,
"epoch": 0.14450384229869695,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999581354834512,
"loss": 5.7683,
"mean_token_accuracy": 0.14914821982383727,
"num_tokens": 3182268.0,
"step": 1730
},
{
"entropy": 5.7966564178466795,
"epoch": 0.14492148346140996,
"grad_norm": 1.0546875,
"learning_rate": 0.000499957559258977,
"loss": 5.8545,
"mean_token_accuracy": 0.1475864365696907,
"num_tokens": 3191513.0,
"step": 1735
},
{
"entropy": 5.931589937210083,
"epoch": 0.14533912462412296,
"grad_norm": 1.0390625,
"learning_rate": 0.00049995697909635,
"loss": 5.8051,
"mean_token_accuracy": 0.14466859847307206,
"num_tokens": 3200580.0,
"step": 1740
},
{
"entropy": 5.861790657043457,
"epoch": 0.14575676578683594,
"grad_norm": 1.03125,
"learning_rate": 0.0004999563949955803,
"loss": 5.8985,
"mean_token_accuracy": 0.15266793817281724,
"num_tokens": 3209053.0,
"step": 1745
},
{
"entropy": 5.94061427116394,
"epoch": 0.14617440694954895,
"grad_norm": 1.015625,
"learning_rate": 0.000499955806956678,
"loss": 5.8802,
"mean_token_accuracy": 0.14815177470445634,
"num_tokens": 3218241.0,
"step": 1750
},
{
"entropy": 5.9239448547363285,
"epoch": 0.14659204811226195,
"grad_norm": 0.984375,
"learning_rate": 0.0004999552149796536,
"loss": 5.9388,
"mean_token_accuracy": 0.1445421226322651,
"num_tokens": 3227887.0,
"step": 1755
},
{
"entropy": 5.891859674453736,
"epoch": 0.14700968927497493,
"grad_norm": 1.046875,
"learning_rate": 0.0004999546190645175,
"loss": 5.7544,
"mean_token_accuracy": 0.15687984526157378,
"num_tokens": 3237281.0,
"step": 1760
},
{
"entropy": 5.83390998840332,
"epoch": 0.14742733043768794,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999540192112799,
"loss": 5.7996,
"mean_token_accuracy": 0.14778247475624084,
"num_tokens": 3246907.0,
"step": 1765
},
{
"entropy": 5.751151847839355,
"epoch": 0.14784497160040094,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999534154199514,
"loss": 5.796,
"mean_token_accuracy": 0.14501683712005614,
"num_tokens": 3256312.0,
"step": 1770
},
{
"entropy": 6.118484783172607,
"epoch": 0.14826261276311392,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999528076905426,
"loss": 5.8851,
"mean_token_accuracy": 0.14510790407657623,
"num_tokens": 3265554.0,
"step": 1775
},
{
"entropy": 5.698562574386597,
"epoch": 0.14868025392582693,
"grad_norm": 1.109375,
"learning_rate": 0.0004999521960230641,
"loss": 5.7285,
"mean_token_accuracy": 0.1552291989326477,
"num_tokens": 3274731.0,
"step": 1780
},
{
"entropy": 5.851112747192383,
"epoch": 0.14909789508853993,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999515804175268,
"loss": 5.8785,
"mean_token_accuracy": 0.14295923113822936,
"num_tokens": 3283518.0,
"step": 1785
},
{
"entropy": 5.983470010757446,
"epoch": 0.1495155362512529,
"grad_norm": 0.953125,
"learning_rate": 0.000499950960873941,
"loss": 5.8957,
"mean_token_accuracy": 0.15030095726251602,
"num_tokens": 3292244.0,
"step": 1790
},
{
"entropy": 5.942580986022949,
"epoch": 0.14993317741396592,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999503373923182,
"loss": 5.8523,
"mean_token_accuracy": 0.15247018486261368,
"num_tokens": 3301198.0,
"step": 1795
},
{
"entropy": 5.859848976135254,
"epoch": 0.15035081857667892,
"grad_norm": 1.0625,
"learning_rate": 0.0004999497099726687,
"loss": 5.8375,
"mean_token_accuracy": 0.15560614168643952,
"num_tokens": 3309983.0,
"step": 1800
},
{
"entropy": 5.930718183517456,
"epoch": 0.1507684597393919,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999490786150039,
"loss": 5.9202,
"mean_token_accuracy": 0.14689518436789512,
"num_tokens": 3319313.0,
"step": 1805
},
{
"entropy": 5.904070234298706,
"epoch": 0.1511861009021049,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999484433193348,
"loss": 5.7746,
"mean_token_accuracy": 0.14794460088014602,
"num_tokens": 3327497.0,
"step": 1810
},
{
"entropy": 5.807149267196655,
"epoch": 0.15160374206481791,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999478040856722,
"loss": 5.794,
"mean_token_accuracy": 0.1535494938492775,
"num_tokens": 3337029.0,
"step": 1815
},
{
"entropy": 5.730625915527344,
"epoch": 0.1520213832275309,
"grad_norm": 1.046875,
"learning_rate": 0.0004999471609140276,
"loss": 5.7624,
"mean_token_accuracy": 0.15014600828289987,
"num_tokens": 3346243.0,
"step": 1820
},
{
"entropy": 6.032606983184815,
"epoch": 0.1524390243902439,
"grad_norm": 1.09375,
"learning_rate": 0.0004999465138044122,
"loss": 5.9295,
"mean_token_accuracy": 0.14537826776504517,
"num_tokens": 3354846.0,
"step": 1825
},
{
"entropy": 5.818017435073853,
"epoch": 0.1528566655529569,
"grad_norm": 1.03125,
"learning_rate": 0.0004999458627568373,
"loss": 5.8785,
"mean_token_accuracy": 0.14605555981397628,
"num_tokens": 3364683.0,
"step": 1830
},
{
"entropy": 5.95707950592041,
"epoch": 0.15327430671566988,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999452077713144,
"loss": 5.8113,
"mean_token_accuracy": 0.15545088350772857,
"num_tokens": 3374652.0,
"step": 1835
},
{
"entropy": 5.90073070526123,
"epoch": 0.1536919478783829,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999445488478547,
"loss": 5.8302,
"mean_token_accuracy": 0.1472937285900116,
"num_tokens": 3384040.0,
"step": 1840
},
{
"entropy": 5.8382641792297365,
"epoch": 0.1541095890410959,
"grad_norm": 1.1484375,
"learning_rate": 0.00049994388598647,
"loss": 5.9108,
"mean_token_accuracy": 0.1450721114873886,
"num_tokens": 3393185.0,
"step": 1845
},
{
"entropy": 5.88736367225647,
"epoch": 0.15452723020380887,
"grad_norm": 1.03125,
"learning_rate": 0.0004999432191871719,
"loss": 5.8423,
"mean_token_accuracy": 0.1527671605348587,
"num_tokens": 3403077.0,
"step": 1850
},
{
"entropy": 5.835742855072022,
"epoch": 0.15494487136652188,
"grad_norm": 1.015625,
"learning_rate": 0.0004999425484499718,
"loss": 5.7927,
"mean_token_accuracy": 0.15965959578752517,
"num_tokens": 3412165.0,
"step": 1855
},
{
"entropy": 5.928402662277222,
"epoch": 0.1553625125292349,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999418737748817,
"loss": 5.863,
"mean_token_accuracy": 0.14788993522524835,
"num_tokens": 3421162.0,
"step": 1860
},
{
"entropy": 5.9775951385498045,
"epoch": 0.15578015369194786,
"grad_norm": 1.0625,
"learning_rate": 0.0004999411951619134,
"loss": 5.9244,
"mean_token_accuracy": 0.14430140256881713,
"num_tokens": 3430363.0,
"step": 1865
},
{
"entropy": 5.951111555099487,
"epoch": 0.15619779485466087,
"grad_norm": 1.09375,
"learning_rate": 0.0004999405126110786,
"loss": 5.8616,
"mean_token_accuracy": 0.1452996090054512,
"num_tokens": 3439199.0,
"step": 1870
},
{
"entropy": 6.038522577285766,
"epoch": 0.15661543601737388,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999398261223895,
"loss": 5.9711,
"mean_token_accuracy": 0.13953697234392165,
"num_tokens": 3448720.0,
"step": 1875
},
{
"entropy": 5.901465845108032,
"epoch": 0.15703307718008688,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999391356958579,
"loss": 5.932,
"mean_token_accuracy": 0.1455860674381256,
"num_tokens": 3458472.0,
"step": 1880
},
{
"entropy": 5.898933792114258,
"epoch": 0.15745071834279986,
"grad_norm": 1.0859375,
"learning_rate": 0.000499938441331496,
"loss": 5.8795,
"mean_token_accuracy": 0.14417899698019027,
"num_tokens": 3467140.0,
"step": 1885
},
{
"entropy": 5.983265781402588,
"epoch": 0.15786835950551287,
"grad_norm": 1.078125,
"learning_rate": 0.0004999377430293159,
"loss": 5.8692,
"mean_token_accuracy": 0.15023969635367393,
"num_tokens": 3476431.0,
"step": 1890
},
{
"entropy": 5.839737033843994,
"epoch": 0.15828600066822587,
"grad_norm": 1.140625,
"learning_rate": 0.0004999370407893299,
"loss": 5.7882,
"mean_token_accuracy": 0.14888408929109573,
"num_tokens": 3485768.0,
"step": 1895
},
{
"entropy": 5.889467048645019,
"epoch": 0.15870364183093885,
"grad_norm": 1.03125,
"learning_rate": 0.0004999363346115501,
"loss": 5.9281,
"mean_token_accuracy": 0.14230190366506576,
"num_tokens": 3495678.0,
"step": 1900
},
{
"entropy": 5.933831214904785,
"epoch": 0.15912128299365186,
"grad_norm": 1.03125,
"learning_rate": 0.0004999356244959893,
"loss": 5.91,
"mean_token_accuracy": 0.14573577046394348,
"num_tokens": 3505168.0,
"step": 1905
},
{
"entropy": 5.84587984085083,
"epoch": 0.15953892415636486,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999349104426594,
"loss": 5.8737,
"mean_token_accuracy": 0.14372649043798447,
"num_tokens": 3514498.0,
"step": 1910
},
{
"entropy": 5.9566321849823,
"epoch": 0.15995656531907784,
"grad_norm": 1.046875,
"learning_rate": 0.0004999341924515732,
"loss": 5.7918,
"mean_token_accuracy": 0.14322776719927788,
"num_tokens": 3524617.0,
"step": 1915
},
{
"entropy": 5.8233592987060545,
"epoch": 0.16037420648179085,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999334705227431,
"loss": 5.7657,
"mean_token_accuracy": 0.15299297571182252,
"num_tokens": 3533523.0,
"step": 1920
},
{
"entropy": 5.8879499435424805,
"epoch": 0.16079184764450385,
"grad_norm": 0.95703125,
"learning_rate": 0.000499932744656182,
"loss": 5.8459,
"mean_token_accuracy": 0.14985390305519103,
"num_tokens": 3542922.0,
"step": 1925
},
{
"entropy": 5.916223335266113,
"epoch": 0.16120948880721683,
"grad_norm": 1.046875,
"learning_rate": 0.0004999320148519024,
"loss": 5.8423,
"mean_token_accuracy": 0.15080732256174087,
"num_tokens": 3551827.0,
"step": 1930
},
{
"entropy": 5.802985715866089,
"epoch": 0.16162712996992984,
"grad_norm": 1.078125,
"learning_rate": 0.0004999312811099171,
"loss": 5.7995,
"mean_token_accuracy": 0.15028103291988373,
"num_tokens": 3560911.0,
"step": 1935
},
{
"entropy": 5.859633922576904,
"epoch": 0.16204477113264285,
"grad_norm": 1.078125,
"learning_rate": 0.0004999305434302391,
"loss": 5.8243,
"mean_token_accuracy": 0.14473793730139733,
"num_tokens": 3569473.0,
"step": 1940
},
{
"entropy": 5.864997625350952,
"epoch": 0.16246241229535582,
"grad_norm": 1.0546875,
"learning_rate": 0.000499929801812881,
"loss": 5.8141,
"mean_token_accuracy": 0.15167805999517442,
"num_tokens": 3578843.0,
"step": 1945
},
{
"entropy": 5.896296644210816,
"epoch": 0.16288005345806883,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999290562578561,
"loss": 5.8362,
"mean_token_accuracy": 0.14523487836122512,
"num_tokens": 3588304.0,
"step": 1950
},
{
"entropy": 5.866140937805175,
"epoch": 0.16329769462078184,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999283067651773,
"loss": 5.8962,
"mean_token_accuracy": 0.14130446910858155,
"num_tokens": 3598354.0,
"step": 1955
},
{
"entropy": 5.998507690429688,
"epoch": 0.16371533578349481,
"grad_norm": 1.015625,
"learning_rate": 0.0004999275533348577,
"loss": 5.9347,
"mean_token_accuracy": 0.13961449563503264,
"num_tokens": 3607503.0,
"step": 1960
},
{
"entropy": 5.763048839569092,
"epoch": 0.16413297694620782,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999267959669106,
"loss": 5.737,
"mean_token_accuracy": 0.15360360890626906,
"num_tokens": 3617467.0,
"step": 1965
},
{
"entropy": 5.88459005355835,
"epoch": 0.16455061810892083,
"grad_norm": 1.078125,
"learning_rate": 0.0004999260346613491,
"loss": 5.7477,
"mean_token_accuracy": 0.14444309696555138,
"num_tokens": 3627048.0,
"step": 1970
},
{
"entropy": 5.772797679901123,
"epoch": 0.1649682592716338,
"grad_norm": 1.0,
"learning_rate": 0.0004999252694181867,
"loss": 5.7743,
"mean_token_accuracy": 0.14830282256007193,
"num_tokens": 3636683.0,
"step": 1975
},
{
"entropy": 5.829623556137085,
"epoch": 0.1653859004343468,
"grad_norm": 1.078125,
"learning_rate": 0.0004999245002374367,
"loss": 5.7055,
"mean_token_accuracy": 0.15671087950468063,
"num_tokens": 3645231.0,
"step": 1980
},
{
"entropy": 5.799342775344849,
"epoch": 0.16580354159705982,
"grad_norm": 1.140625,
"learning_rate": 0.0004999237271191125,
"loss": 5.7774,
"mean_token_accuracy": 0.1466456539928913,
"num_tokens": 3653851.0,
"step": 1985
},
{
"entropy": 5.80361499786377,
"epoch": 0.1662211827597728,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999229500632279,
"loss": 5.7423,
"mean_token_accuracy": 0.15951442569494248,
"num_tokens": 3662720.0,
"step": 1990
},
{
"entropy": 5.723167181015015,
"epoch": 0.1666388239224858,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999221690697961,
"loss": 5.6775,
"mean_token_accuracy": 0.15539188981056212,
"num_tokens": 3671777.0,
"step": 1995
},
{
"entropy": 5.76633882522583,
"epoch": 0.1670564650851988,
"grad_norm": 1.078125,
"learning_rate": 0.0004999213841388312,
"loss": 5.7838,
"mean_token_accuracy": 0.15459920316934586,
"num_tokens": 3681390.0,
"step": 2000
},
{
"entropy": 5.979287242889404,
"epoch": 0.16747410624791179,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999205952703466,
"loss": 5.8338,
"mean_token_accuracy": 0.1480877362191677,
"num_tokens": 3690899.0,
"step": 2005
},
{
"entropy": 5.9035234451293945,
"epoch": 0.1678917474106248,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999198024643563,
"loss": 5.9237,
"mean_token_accuracy": 0.14184789806604386,
"num_tokens": 3699891.0,
"step": 2010
},
{
"entropy": 5.689746618270874,
"epoch": 0.1683093885733378,
"grad_norm": 1.03125,
"learning_rate": 0.0004999190057208741,
"loss": 5.7049,
"mean_token_accuracy": 0.1579154908657074,
"num_tokens": 3709860.0,
"step": 2015
},
{
"entropy": 5.858812427520752,
"epoch": 0.16872702973605078,
"grad_norm": 1.015625,
"learning_rate": 0.000499918205039914,
"loss": 5.8474,
"mean_token_accuracy": 0.14867348447442055,
"num_tokens": 3718896.0,
"step": 2020
},
{
"entropy": 5.816157627105713,
"epoch": 0.16914467089876378,
"grad_norm": 1.03125,
"learning_rate": 0.0004999174004214899,
"loss": 5.648,
"mean_token_accuracy": 0.15528853088617325,
"num_tokens": 3727355.0,
"step": 2025
},
{
"entropy": 5.764490699768066,
"epoch": 0.1695623120614768,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999165918656161,
"loss": 5.7618,
"mean_token_accuracy": 0.14644922390580178,
"num_tokens": 3736191.0,
"step": 2030
},
{
"entropy": 5.8880585670471195,
"epoch": 0.16997995322418977,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999157793723065,
"loss": 5.7863,
"mean_token_accuracy": 0.1557353988289833,
"num_tokens": 3744832.0,
"step": 2035
},
{
"entropy": 5.739016819000244,
"epoch": 0.17039759438690277,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999149629415755,
"loss": 5.7553,
"mean_token_accuracy": 0.1509310707449913,
"num_tokens": 3753835.0,
"step": 2040
},
{
"entropy": 5.95294942855835,
"epoch": 0.17081523554961578,
"grad_norm": 1.03125,
"learning_rate": 0.0004999141425734374,
"loss": 5.833,
"mean_token_accuracy": 0.14866042733192444,
"num_tokens": 3762816.0,
"step": 2045
},
{
"entropy": 5.80229434967041,
"epoch": 0.17123287671232876,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999133182679064,
"loss": 5.8724,
"mean_token_accuracy": 0.14491825997829438,
"num_tokens": 3773244.0,
"step": 2050
},
{
"entropy": 5.860693883895874,
"epoch": 0.17165051787504176,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999124900249971,
"loss": 5.7368,
"mean_token_accuracy": 0.15648135989904405,
"num_tokens": 3782122.0,
"step": 2055
},
{
"entropy": 5.810638523101806,
"epoch": 0.17206815903775477,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999116578447239,
"loss": 5.7313,
"mean_token_accuracy": 0.15645205229520798,
"num_tokens": 3791303.0,
"step": 2060
},
{
"entropy": 5.854610586166382,
"epoch": 0.17248580020046775,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999108217271015,
"loss": 5.802,
"mean_token_accuracy": 0.15410538166761398,
"num_tokens": 3800991.0,
"step": 2065
},
{
"entropy": 5.82975959777832,
"epoch": 0.17290344136318075,
"grad_norm": 1.0,
"learning_rate": 0.0004999099816721444,
"loss": 5.7015,
"mean_token_accuracy": 0.15028277784585953,
"num_tokens": 3811293.0,
"step": 2070
},
{
"entropy": 5.861966800689697,
"epoch": 0.17332108252589376,
"grad_norm": 1.03125,
"learning_rate": 0.0004999091376798673,
"loss": 5.8035,
"mean_token_accuracy": 0.14866990596055984,
"num_tokens": 3820013.0,
"step": 2075
},
{
"entropy": 5.761788749694825,
"epoch": 0.17373872368860674,
"grad_norm": 0.953125,
"learning_rate": 0.0004999082897502849,
"loss": 5.7137,
"mean_token_accuracy": 0.15295772403478622,
"num_tokens": 3829099.0,
"step": 2080
},
{
"entropy": 5.803824234008789,
"epoch": 0.17415636485131974,
"grad_norm": 0.984375,
"learning_rate": 0.0004999074378834123,
"loss": 5.8006,
"mean_token_accuracy": 0.15180691108107566,
"num_tokens": 3838308.0,
"step": 2085
},
{
"entropy": 5.818635940551758,
"epoch": 0.17457400601403275,
"grad_norm": 0.984375,
"learning_rate": 0.0004999065820792644,
"loss": 5.7584,
"mean_token_accuracy": 0.1558691903948784,
"num_tokens": 3847650.0,
"step": 2090
},
{
"entropy": 5.951139116287232,
"epoch": 0.17499164717674573,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999057223378559,
"loss": 5.925,
"mean_token_accuracy": 0.14143869131803513,
"num_tokens": 3857252.0,
"step": 2095
},
{
"entropy": 5.830401134490967,
"epoch": 0.17540928833945874,
"grad_norm": 1.0546875,
"learning_rate": 0.000499904858659202,
"loss": 5.7856,
"mean_token_accuracy": 0.1516982689499855,
"num_tokens": 3865747.0,
"step": 2100
},
{
"entropy": 5.821669197082519,
"epoch": 0.17582692950217174,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999039910433179,
"loss": 5.748,
"mean_token_accuracy": 0.15975648611783982,
"num_tokens": 3874958.0,
"step": 2105
},
{
"entropy": 5.709894752502441,
"epoch": 0.17624457066488472,
"grad_norm": 0.921875,
"learning_rate": 0.0004999031194902187,
"loss": 5.6917,
"mean_token_accuracy": 0.15944382697343826,
"num_tokens": 3885216.0,
"step": 2110
},
{
"entropy": 5.8136450290679935,
"epoch": 0.17666221182759773,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999022439999198,
"loss": 5.7441,
"mean_token_accuracy": 0.15798072814941405,
"num_tokens": 3893973.0,
"step": 2115
},
{
"entropy": 5.852199506759644,
"epoch": 0.17707985299031073,
"grad_norm": 1.0625,
"learning_rate": 0.0004999013645724362,
"loss": 5.8281,
"mean_token_accuracy": 0.15254421532154083,
"num_tokens": 3903532.0,
"step": 2120
},
{
"entropy": 5.854474306106567,
"epoch": 0.1774974941530237,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999004812077836,
"loss": 5.7335,
"mean_token_accuracy": 0.15467983186244966,
"num_tokens": 3913427.0,
"step": 2125
},
{
"entropy": 5.822262239456177,
"epoch": 0.17791513531573672,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998995939059773,
"loss": 5.8745,
"mean_token_accuracy": 0.14384034276008606,
"num_tokens": 3923870.0,
"step": 2130
},
{
"entropy": 5.7962089538574215,
"epoch": 0.17833277647844972,
"grad_norm": 1.0078125,
"learning_rate": 0.000499898702667033,
"loss": 5.6851,
"mean_token_accuracy": 0.1585500419139862,
"num_tokens": 3933469.0,
"step": 2135
},
{
"entropy": 5.739233779907226,
"epoch": 0.1787504176411627,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998978074909661,
"loss": 5.7204,
"mean_token_accuracy": 0.16389980912208557,
"num_tokens": 3942427.0,
"step": 2140
},
{
"entropy": 5.78730525970459,
"epoch": 0.1791680588038757,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998969083777925,
"loss": 5.7606,
"mean_token_accuracy": 0.16544813811779022,
"num_tokens": 3951250.0,
"step": 2145
},
{
"entropy": 5.78267993927002,
"epoch": 0.1795856999665887,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998960053275277,
"loss": 5.7401,
"mean_token_accuracy": 0.15243376642465592,
"num_tokens": 3959969.0,
"step": 2150
},
{
"entropy": 5.798359203338623,
"epoch": 0.1800033411293017,
"grad_norm": 1.140625,
"learning_rate": 0.0004998950983401875,
"loss": 5.7764,
"mean_token_accuracy": 0.15054931342601777,
"num_tokens": 3968171.0,
"step": 2155
},
{
"entropy": 5.7987181663513185,
"epoch": 0.1804209822920147,
"grad_norm": 1.140625,
"learning_rate": 0.0004998941874157882,
"loss": 5.6984,
"mean_token_accuracy": 0.15999143272638322,
"num_tokens": 3976686.0,
"step": 2160
},
{
"entropy": 5.768506765365601,
"epoch": 0.1808386234547277,
"grad_norm": 1.015625,
"learning_rate": 0.0004998932725543453,
"loss": 5.8083,
"mean_token_accuracy": 0.148274065554142,
"num_tokens": 3986552.0,
"step": 2165
},
{
"entropy": 5.7735827445983885,
"epoch": 0.18125626461744068,
"grad_norm": 1.140625,
"learning_rate": 0.0004998923537558749,
"loss": 5.6559,
"mean_token_accuracy": 0.1490070179104805,
"num_tokens": 3995501.0,
"step": 2170
},
{
"entropy": 5.652433967590332,
"epoch": 0.1816739057801537,
"grad_norm": 1.03125,
"learning_rate": 0.0004998914310203932,
"loss": 5.6342,
"mean_token_accuracy": 0.15843954980373381,
"num_tokens": 4004423.0,
"step": 2175
},
{
"entropy": 5.809165954589844,
"epoch": 0.1820915469428667,
"grad_norm": 1.078125,
"learning_rate": 0.0004998905043479163,
"loss": 5.8373,
"mean_token_accuracy": 0.1496340163052082,
"num_tokens": 4013324.0,
"step": 2180
},
{
"entropy": 5.831840944290161,
"epoch": 0.18250918810557967,
"grad_norm": 1.078125,
"learning_rate": 0.0004998895737384605,
"loss": 5.7442,
"mean_token_accuracy": 0.1489358127117157,
"num_tokens": 4021896.0,
"step": 2185
},
{
"entropy": 5.756416845321655,
"epoch": 0.18292682926829268,
"grad_norm": 1.0,
"learning_rate": 0.0004998886391920418,
"loss": 5.7289,
"mean_token_accuracy": 0.14855716228485108,
"num_tokens": 4030590.0,
"step": 2190
},
{
"entropy": 5.781473064422608,
"epoch": 0.18334447043100568,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998877007086769,
"loss": 5.6028,
"mean_token_accuracy": 0.16275189369916915,
"num_tokens": 4040041.0,
"step": 2195
},
{
"entropy": 5.759913444519043,
"epoch": 0.18376211159371866,
"grad_norm": 1.0,
"learning_rate": 0.0004998867582883821,
"loss": 5.7829,
"mean_token_accuracy": 0.15073932483792304,
"num_tokens": 4049119.0,
"step": 2200
},
{
"entropy": 5.84478497505188,
"epoch": 0.18417975275643167,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998858119311739,
"loss": 5.71,
"mean_token_accuracy": 0.1516206756234169,
"num_tokens": 4059265.0,
"step": 2205
},
{
"entropy": 5.831651401519776,
"epoch": 0.18459739391914468,
"grad_norm": 1.015625,
"learning_rate": 0.0004998848616370689,
"loss": 5.765,
"mean_token_accuracy": 0.15168515592813492,
"num_tokens": 4068363.0,
"step": 2210
},
{
"entropy": 5.692407274246216,
"epoch": 0.18501503508185768,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998839074060837,
"loss": 5.6543,
"mean_token_accuracy": 0.16224435716867447,
"num_tokens": 4077390.0,
"step": 2215
},
{
"entropy": 5.759458827972412,
"epoch": 0.18543267624457066,
"grad_norm": 1.078125,
"learning_rate": 0.000499882949238235,
"loss": 5.7242,
"mean_token_accuracy": 0.15047064572572708,
"num_tokens": 4087331.0,
"step": 2220
},
{
"entropy": 5.9659055233001705,
"epoch": 0.18585031740728367,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998819871335396,
"loss": 5.9135,
"mean_token_accuracy": 0.14774233102798462,
"num_tokens": 4096425.0,
"step": 2225
},
{
"entropy": 5.717609214782715,
"epoch": 0.18626795856999667,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998810210920142,
"loss": 5.6762,
"mean_token_accuracy": 0.15252107828855516,
"num_tokens": 4106056.0,
"step": 2230
},
{
"entropy": 5.782451534271241,
"epoch": 0.18668559973270965,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998800511136759,
"loss": 5.7798,
"mean_token_accuracy": 0.1493629887700081,
"num_tokens": 4115133.0,
"step": 2235
},
{
"entropy": 5.825572729110718,
"epoch": 0.18710324089542266,
"grad_norm": 1.03125,
"learning_rate": 0.0004998790771985417,
"loss": 5.6638,
"mean_token_accuracy": 0.1552652359008789,
"num_tokens": 4123630.0,
"step": 2240
},
{
"entropy": 5.7272449970245365,
"epoch": 0.18752088205813566,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998780993466284,
"loss": 5.8069,
"mean_token_accuracy": 0.15545787811279296,
"num_tokens": 4133337.0,
"step": 2245
},
{
"entropy": 5.86578254699707,
"epoch": 0.18793852322084864,
"grad_norm": 1.0625,
"learning_rate": 0.0004998771175579534,
"loss": 5.754,
"mean_token_accuracy": 0.15167360827326776,
"num_tokens": 4142240.0,
"step": 2250
},
{
"entropy": 5.743827629089355,
"epoch": 0.18835616438356165,
"grad_norm": 1.0,
"learning_rate": 0.0004998761318325337,
"loss": 5.721,
"mean_token_accuracy": 0.155141282081604,
"num_tokens": 4151297.0,
"step": 2255
},
{
"entropy": 5.773589324951172,
"epoch": 0.18877380554627465,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998751421703868,
"loss": 5.6829,
"mean_token_accuracy": 0.15771111100912094,
"num_tokens": 4160193.0,
"step": 2260
},
{
"entropy": 5.784664821624756,
"epoch": 0.18919144670898763,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998741485715297,
"loss": 5.826,
"mean_token_accuracy": 0.15172395929694177,
"num_tokens": 4169948.0,
"step": 2265
},
{
"entropy": 6.013588809967041,
"epoch": 0.18960908787170064,
"grad_norm": 1.0234375,
"learning_rate": 0.00049987315103598,
"loss": 5.8425,
"mean_token_accuracy": 0.14488094672560692,
"num_tokens": 4179981.0,
"step": 2270
},
{
"entropy": 5.801202297210693,
"epoch": 0.19002672903441364,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998721495637551,
"loss": 5.8662,
"mean_token_accuracy": 0.1497806191444397,
"num_tokens": 4189172.0,
"step": 2275
},
{
"entropy": 5.8167328357696535,
"epoch": 0.19044437019712662,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998711441548726,
"loss": 5.6645,
"mean_token_accuracy": 0.1592305213212967,
"num_tokens": 4197312.0,
"step": 2280
},
{
"entropy": 5.8211297512054445,
"epoch": 0.19086201135983963,
"grad_norm": 1.171875,
"learning_rate": 0.0004998701348093499,
"loss": 5.7304,
"mean_token_accuracy": 0.15834505409002303,
"num_tokens": 4206828.0,
"step": 2285
},
{
"entropy": 5.894831991195678,
"epoch": 0.19127965252255263,
"grad_norm": 1.015625,
"learning_rate": 0.000499869121527205,
"loss": 5.6956,
"mean_token_accuracy": 0.15966909676790236,
"num_tokens": 4214835.0,
"step": 2290
},
{
"entropy": 5.723678064346314,
"epoch": 0.1916972936852656,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998681043084553,
"loss": 5.8746,
"mean_token_accuracy": 0.14336028322577477,
"num_tokens": 4225119.0,
"step": 2295
},
{
"entropy": 5.7763958930969235,
"epoch": 0.19211493484797862,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998670831531188,
"loss": 5.7067,
"mean_token_accuracy": 0.15475979149341584,
"num_tokens": 4235257.0,
"step": 2300
},
{
"entropy": 5.793825387954712,
"epoch": 0.19253257601069163,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998660580612135,
"loss": 5.7976,
"mean_token_accuracy": 0.1541542060673237,
"num_tokens": 4244211.0,
"step": 2305
},
{
"entropy": 5.899015283584594,
"epoch": 0.1929502171734046,
"grad_norm": 1.0625,
"learning_rate": 0.0004998650290327569,
"loss": 5.7991,
"mean_token_accuracy": 0.15465924888849258,
"num_tokens": 4253149.0,
"step": 2310
},
{
"entropy": 5.734692716598511,
"epoch": 0.1933678583361176,
"grad_norm": 1.09375,
"learning_rate": 0.0004998639960677675,
"loss": 5.6727,
"mean_token_accuracy": 0.1618822380900383,
"num_tokens": 4262557.0,
"step": 2315
},
{
"entropy": 5.741332149505615,
"epoch": 0.19378549949883062,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998629591662631,
"loss": 5.7096,
"mean_token_accuracy": 0.1537215679883957,
"num_tokens": 4272881.0,
"step": 2320
},
{
"entropy": 5.7456244945526125,
"epoch": 0.1942031406615436,
"grad_norm": 1.171875,
"learning_rate": 0.000499861918328262,
"loss": 5.7074,
"mean_token_accuracy": 0.15838230550289153,
"num_tokens": 4281158.0,
"step": 2325
},
{
"entropy": 5.758973026275635,
"epoch": 0.1946207818242566,
"grad_norm": 1.03125,
"learning_rate": 0.0004998608735537822,
"loss": 5.7041,
"mean_token_accuracy": 0.15414268523454666,
"num_tokens": 4290319.0,
"step": 2330
},
{
"entropy": 5.740860414505005,
"epoch": 0.1950384229869696,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998598248428423,
"loss": 5.7807,
"mean_token_accuracy": 0.1590170294046402,
"num_tokens": 4299276.0,
"step": 2335
},
{
"entropy": 5.873537826538086,
"epoch": 0.19545606414968258,
"grad_norm": 1.046875,
"learning_rate": 0.0004998587721954604,
"loss": 5.7551,
"mean_token_accuracy": 0.15692680478096008,
"num_tokens": 4308291.0,
"step": 2340
},
{
"entropy": 5.695183038711548,
"epoch": 0.1958737053123956,
"grad_norm": 1.125,
"learning_rate": 0.0004998577156116551,
"loss": 5.7046,
"mean_token_accuracy": 0.15472691208124162,
"num_tokens": 4316654.0,
"step": 2345
},
{
"entropy": 5.789601373672485,
"epoch": 0.1962913464751086,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998566550914448,
"loss": 5.6544,
"mean_token_accuracy": 0.15280546993017197,
"num_tokens": 4325911.0,
"step": 2350
},
{
"entropy": 5.821873140335083,
"epoch": 0.19670898763782158,
"grad_norm": 1.046875,
"learning_rate": 0.0004998555906348481,
"loss": 5.8266,
"mean_token_accuracy": 0.15374174118041992,
"num_tokens": 4334846.0,
"step": 2355
},
{
"entropy": 5.770111083984375,
"epoch": 0.19712662880053458,
"grad_norm": 1.078125,
"learning_rate": 0.0004998545222418836,
"loss": 5.7134,
"mean_token_accuracy": 0.15767551213502884,
"num_tokens": 4343728.0,
"step": 2360
},
{
"entropy": 5.79700288772583,
"epoch": 0.1975442699632476,
"grad_norm": 1.0625,
"learning_rate": 0.0004998534499125701,
"loss": 5.7072,
"mean_token_accuracy": 0.1550982862710953,
"num_tokens": 4352774.0,
"step": 2365
},
{
"entropy": 5.86985936164856,
"epoch": 0.19796191112596057,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998523736469263,
"loss": 5.8227,
"mean_token_accuracy": 0.15387885868549347,
"num_tokens": 4361663.0,
"step": 2370
},
{
"entropy": 5.73446249961853,
"epoch": 0.19837955228867357,
"grad_norm": 1.078125,
"learning_rate": 0.000499851293444971,
"loss": 5.6621,
"mean_token_accuracy": 0.1654502682387829,
"num_tokens": 4370278.0,
"step": 2375
},
{
"entropy": 5.795806503295898,
"epoch": 0.19879719345138658,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998502093067231,
"loss": 5.7328,
"mean_token_accuracy": 0.16133349984884263,
"num_tokens": 4379010.0,
"step": 2380
},
{
"entropy": 5.768770170211792,
"epoch": 0.19921483461409956,
"grad_norm": 1.046875,
"learning_rate": 0.0004998491212322017,
"loss": 5.7087,
"mean_token_accuracy": 0.15297210067510605,
"num_tokens": 4388145.0,
"step": 2385
},
{
"entropy": 5.765263605117798,
"epoch": 0.19963247577681256,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998480292214258,
"loss": 5.8348,
"mean_token_accuracy": 0.15209052711725235,
"num_tokens": 4396321.0,
"step": 2390
},
{
"entropy": 5.829952001571655,
"epoch": 0.20005011693952557,
"grad_norm": 1.15625,
"learning_rate": 0.0004998469332744144,
"loss": 5.6203,
"mean_token_accuracy": 0.15987894535064698,
"num_tokens": 4404854.0,
"step": 2395
},
{
"entropy": 5.549867105484009,
"epoch": 0.20046775810223855,
"grad_norm": 1.1953125,
"learning_rate": 0.0004998458333911869,
"loss": 5.6054,
"mean_token_accuracy": 0.16445660591125488,
"num_tokens": 4413890.0,
"step": 2400
},
{
"entropy": 5.793780994415283,
"epoch": 0.20088539926495155,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998447295717624,
"loss": 5.766,
"mean_token_accuracy": 0.15457075983285903,
"num_tokens": 4423019.0,
"step": 2405
},
{
"entropy": 5.814533615112305,
"epoch": 0.20130304042766456,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998436218161602,
"loss": 5.6992,
"mean_token_accuracy": 0.1568400338292122,
"num_tokens": 4432340.0,
"step": 2410
},
{
"entropy": 5.715889596939087,
"epoch": 0.20172068159037754,
"grad_norm": 1.015625,
"learning_rate": 0.0004998425101243999,
"loss": 5.6767,
"mean_token_accuracy": 0.1556083530187607,
"num_tokens": 4441834.0,
"step": 2415
},
{
"entropy": 5.73554949760437,
"epoch": 0.20213832275309054,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998413944965007,
"loss": 5.7037,
"mean_token_accuracy": 0.15672216266393663,
"num_tokens": 4450864.0,
"step": 2420
},
{
"entropy": 5.813930034637451,
"epoch": 0.20255596391580355,
"grad_norm": 1.03125,
"learning_rate": 0.0004998402749324823,
"loss": 5.7822,
"mean_token_accuracy": 0.14905955046415328,
"num_tokens": 4460697.0,
"step": 2425
},
{
"entropy": 5.903818655014038,
"epoch": 0.20297360507851653,
"grad_norm": 1.015625,
"learning_rate": 0.0004998391514323642,
"loss": 5.7705,
"mean_token_accuracy": 0.15528278648853303,
"num_tokens": 4469926.0,
"step": 2430
},
{
"entropy": 5.695542097091675,
"epoch": 0.20339124624122953,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998380239961661,
"loss": 5.7188,
"mean_token_accuracy": 0.15469388663768768,
"num_tokens": 4478818.0,
"step": 2435
},
{
"entropy": 5.840320062637329,
"epoch": 0.20380888740394254,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998368926239078,
"loss": 5.6545,
"mean_token_accuracy": 0.1583703339099884,
"num_tokens": 4487451.0,
"step": 2440
},
{
"entropy": 5.759117460250854,
"epoch": 0.20422652856665552,
"grad_norm": 1.0390625,
"learning_rate": 0.000499835757315609,
"loss": 5.7263,
"mean_token_accuracy": 0.1547158733010292,
"num_tokens": 4497674.0,
"step": 2445
},
{
"entropy": 5.7022154331207275,
"epoch": 0.20464416972936852,
"grad_norm": 1.109375,
"learning_rate": 0.0004998346180712898,
"loss": 5.6802,
"mean_token_accuracy": 0.1570851117372513,
"num_tokens": 4507666.0,
"step": 2450
},
{
"entropy": 5.789335775375366,
"epoch": 0.20506181089208153,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998334748909698,
"loss": 5.8022,
"mean_token_accuracy": 0.15129673779010772,
"num_tokens": 4517409.0,
"step": 2455
},
{
"entropy": 5.738123893737793,
"epoch": 0.2054794520547945,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998323277746694,
"loss": 5.59,
"mean_token_accuracy": 0.156346096098423,
"num_tokens": 4526495.0,
"step": 2460
},
{
"entropy": 5.719364356994629,
"epoch": 0.20589709321750752,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998311767224081,
"loss": 5.6809,
"mean_token_accuracy": 0.16575086265802383,
"num_tokens": 4535240.0,
"step": 2465
},
{
"entropy": 5.811622524261475,
"epoch": 0.20631473438022052,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998300217342067,
"loss": 5.7619,
"mean_token_accuracy": 0.15527621805667877,
"num_tokens": 4544716.0,
"step": 2470
},
{
"entropy": 5.673983573913574,
"epoch": 0.2067323755429335,
"grad_norm": 1.03125,
"learning_rate": 0.0004998288628100851,
"loss": 5.6993,
"mean_token_accuracy": 0.1583945244550705,
"num_tokens": 4553867.0,
"step": 2475
},
{
"entropy": 5.760221529006958,
"epoch": 0.2071500167056465,
"grad_norm": 0.984375,
"learning_rate": 0.0004998276999500636,
"loss": 5.7359,
"mean_token_accuracy": 0.15061172246932983,
"num_tokens": 4562529.0,
"step": 2480
},
{
"entropy": 5.663964939117432,
"epoch": 0.2075676578683595,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998265331541627,
"loss": 5.5193,
"mean_token_accuracy": 0.1735447093844414,
"num_tokens": 4571041.0,
"step": 2485
},
{
"entropy": 5.6653406620025635,
"epoch": 0.2079852990310725,
"grad_norm": 1.109375,
"learning_rate": 0.0004998253624224026,
"loss": 5.6775,
"mean_token_accuracy": 0.15766091048717498,
"num_tokens": 4579914.0,
"step": 2490
},
{
"entropy": 5.758085489273071,
"epoch": 0.2084029401937855,
"grad_norm": 1.1875,
"learning_rate": 0.000499824187754804,
"loss": 5.7562,
"mean_token_accuracy": 0.15930444449186326,
"num_tokens": 4590277.0,
"step": 2495
},
{
"entropy": 5.822220373153686,
"epoch": 0.2088205813564985,
"grad_norm": 1.140625,
"learning_rate": 0.0004998230091513872,
"loss": 5.67,
"mean_token_accuracy": 0.16226856559515,
"num_tokens": 4598719.0,
"step": 2500
},
{
"entropy": 5.6369061946868895,
"epoch": 0.20923822251921148,
"grad_norm": 1.015625,
"learning_rate": 0.0004998218266121731,
"loss": 5.6358,
"mean_token_accuracy": 0.15583296865224838,
"num_tokens": 4607804.0,
"step": 2505
},
{
"entropy": 5.705494403839111,
"epoch": 0.2096558636819245,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998206401371823,
"loss": 5.6148,
"mean_token_accuracy": 0.16443610489368438,
"num_tokens": 4616529.0,
"step": 2510
},
{
"entropy": 5.741948223114013,
"epoch": 0.2100735048446375,
"grad_norm": 1.15625,
"learning_rate": 0.0004998194497264357,
"loss": 5.6813,
"mean_token_accuracy": 0.15554540753364562,
"num_tokens": 4624545.0,
"step": 2515
},
{
"entropy": 5.7241405010223385,
"epoch": 0.21049114600735047,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998182553799538,
"loss": 5.6417,
"mean_token_accuracy": 0.1507075920701027,
"num_tokens": 4633316.0,
"step": 2520
},
{
"entropy": 5.6320202350616455,
"epoch": 0.21090878717006348,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998170570977579,
"loss": 5.597,
"mean_token_accuracy": 0.1645556092262268,
"num_tokens": 4643353.0,
"step": 2525
},
{
"entropy": 5.701850748062133,
"epoch": 0.21132642833277648,
"grad_norm": 1.015625,
"learning_rate": 0.0004998158548798687,
"loss": 5.6999,
"mean_token_accuracy": 0.15542778819799424,
"num_tokens": 4651998.0,
"step": 2530
},
{
"entropy": 5.7899423122406,
"epoch": 0.21174406949548946,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998146487263074,
"loss": 5.6929,
"mean_token_accuracy": 0.15939732789993286,
"num_tokens": 4659937.0,
"step": 2535
},
{
"entropy": 5.644034767150879,
"epoch": 0.21216171065820247,
"grad_norm": 1.015625,
"learning_rate": 0.0004998134386370951,
"loss": 5.6649,
"mean_token_accuracy": 0.1551835395395756,
"num_tokens": 4669396.0,
"step": 2540
},
{
"entropy": 5.848598384857178,
"epoch": 0.21257935182091547,
"grad_norm": 1.015625,
"learning_rate": 0.0004998122246122529,
"loss": 5.7452,
"mean_token_accuracy": 0.15422798991203307,
"num_tokens": 4679320.0,
"step": 2545
},
{
"entropy": 5.740176057815551,
"epoch": 0.21299699298362845,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998110066518021,
"loss": 5.6689,
"mean_token_accuracy": 0.15735814869403839,
"num_tokens": 4688807.0,
"step": 2550
},
{
"entropy": 5.746449422836304,
"epoch": 0.21341463414634146,
"grad_norm": 1.09375,
"learning_rate": 0.000499809784755764,
"loss": 5.6618,
"mean_token_accuracy": 0.16325941383838655,
"num_tokens": 4697819.0,
"step": 2555
},
{
"entropy": 5.671551179885864,
"epoch": 0.21383227530905446,
"grad_norm": 1.0,
"learning_rate": 0.00049980855892416,
"loss": 5.651,
"mean_token_accuracy": 0.1612872451543808,
"num_tokens": 4706931.0,
"step": 2560
},
{
"entropy": 5.6566770553588865,
"epoch": 0.21424991647176747,
"grad_norm": 0.984375,
"learning_rate": 0.0004998073291570116,
"loss": 5.6237,
"mean_token_accuracy": 0.1609420210123062,
"num_tokens": 4716262.0,
"step": 2565
},
{
"entropy": 5.776531457901001,
"epoch": 0.21466755763448045,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998060954543404,
"loss": 5.6447,
"mean_token_accuracy": 0.16187569946050645,
"num_tokens": 4725415.0,
"step": 2570
},
{
"entropy": 5.642414903640747,
"epoch": 0.21508519879719346,
"grad_norm": 1.125,
"learning_rate": 0.0004998048578161678,
"loss": 5.7127,
"mean_token_accuracy": 0.15629575103521348,
"num_tokens": 4734580.0,
"step": 2575
},
{
"entropy": 5.740717124938965,
"epoch": 0.21550283995990646,
"grad_norm": 1.15625,
"learning_rate": 0.0004998036162425155,
"loss": 5.5339,
"mean_token_accuracy": 0.16538190245628356,
"num_tokens": 4742966.0,
"step": 2580
},
{
"entropy": 5.672183084487915,
"epoch": 0.21592048112261944,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998023707334055,
"loss": 5.6798,
"mean_token_accuracy": 0.1618649870157242,
"num_tokens": 4751696.0,
"step": 2585
},
{
"entropy": 5.812973880767823,
"epoch": 0.21633812228533245,
"grad_norm": 1.0625,
"learning_rate": 0.0004998011212888592,
"loss": 5.7348,
"mean_token_accuracy": 0.16206271648406984,
"num_tokens": 4760460.0,
"step": 2590
},
{
"entropy": 5.73599271774292,
"epoch": 0.21675576344804545,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997998679088988,
"loss": 5.6601,
"mean_token_accuracy": 0.15789793580770492,
"num_tokens": 4769004.0,
"step": 2595
},
{
"entropy": 5.651835107803345,
"epoch": 0.21717340461075843,
"grad_norm": 1.0625,
"learning_rate": 0.0004997986105935461,
"loss": 5.6462,
"mean_token_accuracy": 0.16610565036535263,
"num_tokens": 4777204.0,
"step": 2600
},
{
"entropy": 5.839091444015503,
"epoch": 0.21759104577347144,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997973493428231,
"loss": 5.7613,
"mean_token_accuracy": 0.15818434655666352,
"num_tokens": 4786766.0,
"step": 2605
},
{
"entropy": 5.6804766178131105,
"epoch": 0.21800868693618444,
"grad_norm": 1.0625,
"learning_rate": 0.0004997960841567519,
"loss": 5.6571,
"mean_token_accuracy": 0.16050463169813156,
"num_tokens": 4796676.0,
"step": 2610
},
{
"entropy": 5.868012285232544,
"epoch": 0.21842632809889742,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997948150353548,
"loss": 5.6617,
"mean_token_accuracy": 0.15644014179706572,
"num_tokens": 4805514.0,
"step": 2615
},
{
"entropy": 5.665460681915283,
"epoch": 0.21884396926161043,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997935419786537,
"loss": 5.7374,
"mean_token_accuracy": 0.15640380531549453,
"num_tokens": 4815487.0,
"step": 2620
},
{
"entropy": 5.803626251220703,
"epoch": 0.21926161042432343,
"grad_norm": 0.96875,
"learning_rate": 0.0004997922649866712,
"loss": 5.6948,
"mean_token_accuracy": 0.1630088746547699,
"num_tokens": 4825044.0,
"step": 2625
},
{
"entropy": 5.620978498458863,
"epoch": 0.2196792515870364,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997909840594294,
"loss": 5.486,
"mean_token_accuracy": 0.1689409002661705,
"num_tokens": 4833875.0,
"step": 2630
},
{
"entropy": 5.659580278396606,
"epoch": 0.22009689274974942,
"grad_norm": 1.0,
"learning_rate": 0.0004997896991969509,
"loss": 5.7134,
"mean_token_accuracy": 0.1507007360458374,
"num_tokens": 4843317.0,
"step": 2635
},
{
"entropy": 5.801263904571533,
"epoch": 0.22051453391246242,
"grad_norm": 0.984375,
"learning_rate": 0.0004997884103992581,
"loss": 5.695,
"mean_token_accuracy": 0.15063075274229049,
"num_tokens": 4852514.0,
"step": 2640
},
{
"entropy": 5.705005836486817,
"epoch": 0.2209321750751754,
"grad_norm": 1.03125,
"learning_rate": 0.0004997871176663736,
"loss": 5.6512,
"mean_token_accuracy": 0.16415767073631288,
"num_tokens": 4861260.0,
"step": 2645
},
{
"entropy": 5.70496301651001,
"epoch": 0.2213498162378884,
"grad_norm": 1.09375,
"learning_rate": 0.0004997858209983201,
"loss": 5.6218,
"mean_token_accuracy": 0.15602522864937782,
"num_tokens": 4869654.0,
"step": 2650
},
{
"entropy": 5.69069185256958,
"epoch": 0.22176745740060141,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997845203951202,
"loss": 5.6501,
"mean_token_accuracy": 0.15217012390494347,
"num_tokens": 4878311.0,
"step": 2655
},
{
"entropy": 5.796048784255982,
"epoch": 0.2221850985633144,
"grad_norm": 1.09375,
"learning_rate": 0.0004997832158567967,
"loss": 5.7153,
"mean_token_accuracy": 0.15374061986804008,
"num_tokens": 4887251.0,
"step": 2660
},
{
"entropy": 5.704570245742798,
"epoch": 0.2226027397260274,
"grad_norm": 1.015625,
"learning_rate": 0.0004997819073833724,
"loss": 5.6061,
"mean_token_accuracy": 0.16006359308958054,
"num_tokens": 4896528.0,
"step": 2665
},
{
"entropy": 5.679030656814575,
"epoch": 0.2230203808887404,
"grad_norm": 1.09375,
"learning_rate": 0.0004997805949748702,
"loss": 5.6365,
"mean_token_accuracy": 0.15641110390424728,
"num_tokens": 4904962.0,
"step": 2670
},
{
"entropy": 5.712535619735718,
"epoch": 0.22343802205145338,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997792786313132,
"loss": 5.6527,
"mean_token_accuracy": 0.15834657847881317,
"num_tokens": 4914735.0,
"step": 2675
},
{
"entropy": 5.577741289138794,
"epoch": 0.2238556632141664,
"grad_norm": 1.03125,
"learning_rate": 0.0004997779583527243,
"loss": 5.519,
"mean_token_accuracy": 0.15640904307365416,
"num_tokens": 4924271.0,
"step": 2680
},
{
"entropy": 5.790355157852173,
"epoch": 0.2242733043768794,
"grad_norm": 1.078125,
"learning_rate": 0.0004997766341391268,
"loss": 5.7583,
"mean_token_accuracy": 0.15324506014585496,
"num_tokens": 4933928.0,
"step": 2685
},
{
"entropy": 5.668966484069824,
"epoch": 0.22469094553959237,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997753059905437,
"loss": 5.4782,
"mean_token_accuracy": 0.1679123729467392,
"num_tokens": 4942714.0,
"step": 2690
},
{
"entropy": 5.584737396240234,
"epoch": 0.22510858670230538,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997739739069981,
"loss": 5.5993,
"mean_token_accuracy": 0.16191962063312532,
"num_tokens": 4952297.0,
"step": 2695
},
{
"entropy": 5.652891540527344,
"epoch": 0.22552622786501839,
"grad_norm": 1.046875,
"learning_rate": 0.0004997726378885138,
"loss": 5.54,
"mean_token_accuracy": 0.1690146654844284,
"num_tokens": 4960947.0,
"step": 2700
},
{
"entropy": 5.7303954601287845,
"epoch": 0.22594386902773136,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997712979351137,
"loss": 5.6991,
"mean_token_accuracy": 0.15585016310214997,
"num_tokens": 4969705.0,
"step": 2705
},
{
"entropy": 5.763876342773438,
"epoch": 0.22636151019044437,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997699540468216,
"loss": 5.6717,
"mean_token_accuracy": 0.15742296129465103,
"num_tokens": 4979112.0,
"step": 2710
},
{
"entropy": 5.6515544891357425,
"epoch": 0.22677915135315738,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997686062236609,
"loss": 5.5907,
"mean_token_accuracy": 0.16135782301425933,
"num_tokens": 4987894.0,
"step": 2715
},
{
"entropy": 5.711192798614502,
"epoch": 0.22719679251587035,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997672544656553,
"loss": 5.7182,
"mean_token_accuracy": 0.15377283990383148,
"num_tokens": 4997774.0,
"step": 2720
},
{
"entropy": 5.736680126190185,
"epoch": 0.22761443367858336,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997658987728283,
"loss": 5.6583,
"mean_token_accuracy": 0.16393161863088607,
"num_tokens": 5006183.0,
"step": 2725
},
{
"entropy": 5.659987592697144,
"epoch": 0.22803207484129637,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997645391452035,
"loss": 5.6492,
"mean_token_accuracy": 0.15915142148733138,
"num_tokens": 5016909.0,
"step": 2730
},
{
"entropy": 5.789839601516723,
"epoch": 0.22844971600400935,
"grad_norm": 1.0390625,
"learning_rate": 0.000499763175582805,
"loss": 5.6891,
"mean_token_accuracy": 0.16223524361848832,
"num_tokens": 5027326.0,
"step": 2735
},
{
"entropy": 5.682078647613525,
"epoch": 0.22886735716672235,
"grad_norm": 0.96875,
"learning_rate": 0.0004997618080856566,
"loss": 5.6343,
"mean_token_accuracy": 0.15808423310518266,
"num_tokens": 5037043.0,
"step": 2740
},
{
"entropy": 5.623650789260864,
"epoch": 0.22928499832943536,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997604366537823,
"loss": 5.6393,
"mean_token_accuracy": 0.15592788010835648,
"num_tokens": 5045992.0,
"step": 2745
},
{
"entropy": 5.673027992248535,
"epoch": 0.22970263949214834,
"grad_norm": 1.046875,
"learning_rate": 0.0004997590612872058,
"loss": 5.5983,
"mean_token_accuracy": 0.16179318875074386,
"num_tokens": 5055568.0,
"step": 2750
},
{
"entropy": 5.760726261138916,
"epoch": 0.23012028065486134,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997576819859515,
"loss": 5.6889,
"mean_token_accuracy": 0.15718334317207336,
"num_tokens": 5065512.0,
"step": 2755
},
{
"entropy": 5.578152847290039,
"epoch": 0.23053792181757435,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997562987500434,
"loss": 5.5388,
"mean_token_accuracy": 0.166161248087883,
"num_tokens": 5074776.0,
"step": 2760
},
{
"entropy": 5.694838762283325,
"epoch": 0.23095556298028733,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997549115795058,
"loss": 5.5842,
"mean_token_accuracy": 0.16133284717798232,
"num_tokens": 5084743.0,
"step": 2765
},
{
"entropy": 5.77802414894104,
"epoch": 0.23137320414300033,
"grad_norm": 1.0390625,
"learning_rate": 0.000499753520474363,
"loss": 5.7451,
"mean_token_accuracy": 0.15966019704937934,
"num_tokens": 5095387.0,
"step": 2770
},
{
"entropy": 5.68107967376709,
"epoch": 0.23179084530571334,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997521254346391,
"loss": 5.651,
"mean_token_accuracy": 0.16418414637446405,
"num_tokens": 5104628.0,
"step": 2775
},
{
"entropy": 5.721260213851929,
"epoch": 0.23220848646842632,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997507264603589,
"loss": 5.5944,
"mean_token_accuracy": 0.1611643835902214,
"num_tokens": 5113169.0,
"step": 2780
},
{
"entropy": 5.683739376068115,
"epoch": 0.23262612763113932,
"grad_norm": 1.015625,
"learning_rate": 0.0004997493235515466,
"loss": 5.6716,
"mean_token_accuracy": 0.15972235798835754,
"num_tokens": 5122294.0,
"step": 2785
},
{
"entropy": 5.627259922027588,
"epoch": 0.23304376879385233,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997479167082269,
"loss": 5.5023,
"mean_token_accuracy": 0.1682959020137787,
"num_tokens": 5131191.0,
"step": 2790
},
{
"entropy": 5.659307527542114,
"epoch": 0.2334614099565653,
"grad_norm": 1.03125,
"learning_rate": 0.0004997465059304243,
"loss": 5.7164,
"mean_token_accuracy": 0.1605928435921669,
"num_tokens": 5141974.0,
"step": 2795
},
{
"entropy": 5.716783761978149,
"epoch": 0.23387905111927831,
"grad_norm": 1.0,
"learning_rate": 0.0004997450912181637,
"loss": 5.6504,
"mean_token_accuracy": 0.16146593242883683,
"num_tokens": 5151412.0,
"step": 2800
},
{
"entropy": 5.660967063903809,
"epoch": 0.23429669228199132,
"grad_norm": 1.0,
"learning_rate": 0.0004997436725714696,
"loss": 5.5774,
"mean_token_accuracy": 0.16266391277313233,
"num_tokens": 5160989.0,
"step": 2805
},
{
"entropy": 5.701135158538818,
"epoch": 0.2347143334447043,
"grad_norm": 1.046875,
"learning_rate": 0.0004997422499903671,
"loss": 5.6485,
"mean_token_accuracy": 0.15926841497421265,
"num_tokens": 5170241.0,
"step": 2810
},
{
"entropy": 5.7674877643585205,
"epoch": 0.2351319746074173,
"grad_norm": 1.0234375,
"learning_rate": 0.000499740823474881,
"loss": 5.771,
"mean_token_accuracy": 0.1533608093857765,
"num_tokens": 5180247.0,
"step": 2815
},
{
"entropy": 5.670247268676758,
"epoch": 0.2355496157701303,
"grad_norm": 1.078125,
"learning_rate": 0.0004997393930250363,
"loss": 5.6428,
"mean_token_accuracy": 0.16462425887584686,
"num_tokens": 5189144.0,
"step": 2820
},
{
"entropy": 5.711422109603882,
"epoch": 0.2359672569328433,
"grad_norm": 1.015625,
"learning_rate": 0.000499737958640858,
"loss": 5.7033,
"mean_token_accuracy": 0.15771541371941566,
"num_tokens": 5199410.0,
"step": 2825
},
{
"entropy": 5.740068626403809,
"epoch": 0.2363848980955563,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997365203223711,
"loss": 5.5803,
"mean_token_accuracy": 0.16387100964784623,
"num_tokens": 5209032.0,
"step": 2830
},
{
"entropy": 5.492058753967285,
"epoch": 0.2368025392582693,
"grad_norm": 1.0625,
"learning_rate": 0.000499735078069601,
"loss": 5.4773,
"mean_token_accuracy": 0.16404626965522767,
"num_tokens": 5217787.0,
"step": 2835
},
{
"entropy": 5.574188566207885,
"epoch": 0.23722018042098228,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997336318825728,
"loss": 5.601,
"mean_token_accuracy": 0.1620022639632225,
"num_tokens": 5226962.0,
"step": 2840
},
{
"entropy": 5.649361181259155,
"epoch": 0.23763782158369529,
"grad_norm": 1.0,
"learning_rate": 0.0004997321817613119,
"loss": 5.5662,
"mean_token_accuracy": 0.1645295113325119,
"num_tokens": 5236111.0,
"step": 2845
},
{
"entropy": 5.703363037109375,
"epoch": 0.2380554627464083,
"grad_norm": 0.9375,
"learning_rate": 0.0004997307277058436,
"loss": 5.6886,
"mean_token_accuracy": 0.15562707632780076,
"num_tokens": 5246628.0,
"step": 2850
},
{
"entropy": 5.72263731956482,
"epoch": 0.23847310390912127,
"grad_norm": 1.015625,
"learning_rate": 0.0004997292697161935,
"loss": 5.641,
"mean_token_accuracy": 0.15522865802049637,
"num_tokens": 5255543.0,
"step": 2855
},
{
"entropy": 5.555987739562989,
"epoch": 0.23889074507183428,
"grad_norm": 1.0546875,
"learning_rate": 0.000499727807792387,
"loss": 5.4705,
"mean_token_accuracy": 0.17789942026138306,
"num_tokens": 5264559.0,
"step": 2860
},
{
"entropy": 5.5421061515808105,
"epoch": 0.23930838623454728,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997263419344495,
"loss": 5.5572,
"mean_token_accuracy": 0.1629979580640793,
"num_tokens": 5274599.0,
"step": 2865
},
{
"entropy": 5.606746578216553,
"epoch": 0.23972602739726026,
"grad_norm": 1.109375,
"learning_rate": 0.0004997248721424071,
"loss": 5.5493,
"mean_token_accuracy": 0.1658317342400551,
"num_tokens": 5283495.0,
"step": 2870
},
{
"entropy": 5.6887702465057375,
"epoch": 0.24014366855997327,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997233984162853,
"loss": 5.5925,
"mean_token_accuracy": 0.16624960899353028,
"num_tokens": 5292064.0,
"step": 2875
},
{
"entropy": 5.646259498596192,
"epoch": 0.24056130972268627,
"grad_norm": 0.984375,
"learning_rate": 0.0004997219207561099,
"loss": 5.5957,
"mean_token_accuracy": 0.1569068178534508,
"num_tokens": 5301696.0,
"step": 2880
},
{
"entropy": 5.735414361953735,
"epoch": 0.24097895088539925,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997204391619068,
"loss": 5.7093,
"mean_token_accuracy": 0.1573967769742012,
"num_tokens": 5311427.0,
"step": 2885
},
{
"entropy": 5.792837524414063,
"epoch": 0.24139659204811226,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997189536337018,
"loss": 5.6568,
"mean_token_accuracy": 0.15813831835985184,
"num_tokens": 5321406.0,
"step": 2890
},
{
"entropy": 5.703797006607056,
"epoch": 0.24181423321082526,
"grad_norm": 1.171875,
"learning_rate": 0.0004997174641715211,
"loss": 5.7185,
"mean_token_accuracy": 0.15567933470010759,
"num_tokens": 5332257.0,
"step": 2895
},
{
"entropy": 5.589183568954468,
"epoch": 0.24223187437353827,
"grad_norm": 1.015625,
"learning_rate": 0.0004997159707753906,
"loss": 5.4653,
"mean_token_accuracy": 0.16655567437410354,
"num_tokens": 5341857.0,
"step": 2900
},
{
"entropy": 5.568355941772461,
"epoch": 0.24264951553625125,
"grad_norm": 1.015625,
"learning_rate": 0.0004997144734453367,
"loss": 5.6049,
"mean_token_accuracy": 0.1704968884587288,
"num_tokens": 5351367.0,
"step": 2905
},
{
"entropy": 5.63011474609375,
"epoch": 0.24306715669896425,
"grad_norm": 1.046875,
"learning_rate": 0.0004997129721813853,
"loss": 5.4481,
"mean_token_accuracy": 0.17252209633588791,
"num_tokens": 5360627.0,
"step": 2910
},
{
"entropy": 5.588594675064087,
"epoch": 0.24348479786167726,
"grad_norm": 1.125,
"learning_rate": 0.0004997114669835629,
"loss": 5.5418,
"mean_token_accuracy": 0.17313383519649506,
"num_tokens": 5369703.0,
"step": 2915
},
{
"entropy": 5.64463095664978,
"epoch": 0.24390243902439024,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997099578518959,
"loss": 5.6075,
"mean_token_accuracy": 0.16318197697401046,
"num_tokens": 5378432.0,
"step": 2920
},
{
"entropy": 5.603151750564575,
"epoch": 0.24432008018710324,
"grad_norm": 1.15625,
"learning_rate": 0.0004997084447864105,
"loss": 5.4732,
"mean_token_accuracy": 0.16346939504146576,
"num_tokens": 5386489.0,
"step": 2925
},
{
"entropy": 5.722626972198486,
"epoch": 0.24473772134981625,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997069277871333,
"loss": 5.6873,
"mean_token_accuracy": 0.1580083653330803,
"num_tokens": 5395908.0,
"step": 2930
},
{
"entropy": 5.77180233001709,
"epoch": 0.24515536251252923,
"grad_norm": 1.03125,
"learning_rate": 0.0004997054068540908,
"loss": 5.6112,
"mean_token_accuracy": 0.16377026438713074,
"num_tokens": 5404822.0,
"step": 2935
},
{
"entropy": 5.547445392608642,
"epoch": 0.24557300367524224,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997038819873098,
"loss": 5.5305,
"mean_token_accuracy": 0.16950465887784957,
"num_tokens": 5413980.0,
"step": 2940
},
{
"entropy": 5.642219066619873,
"epoch": 0.24599064483795524,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997023531868167,
"loss": 5.5896,
"mean_token_accuracy": 0.15958906412124635,
"num_tokens": 5423538.0,
"step": 2945
},
{
"entropy": 5.587050342559815,
"epoch": 0.24640828600066822,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997008204526386,
"loss": 5.5178,
"mean_token_accuracy": 0.16479797512292862,
"num_tokens": 5431930.0,
"step": 2950
},
{
"entropy": 5.580323791503906,
"epoch": 0.24682592716338123,
"grad_norm": 1.015625,
"learning_rate": 0.0004996992837848021,
"loss": 5.5136,
"mean_token_accuracy": 0.16964638084173203,
"num_tokens": 5440721.0,
"step": 2955
},
{
"entropy": 5.584440279006958,
"epoch": 0.24724356832609423,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996977431833342,
"loss": 5.5884,
"mean_token_accuracy": 0.16492316871881485,
"num_tokens": 5449617.0,
"step": 2960
},
{
"entropy": 5.661376190185547,
"epoch": 0.2476612094888072,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996961986482618,
"loss": 5.625,
"mean_token_accuracy": 0.16289934664964675,
"num_tokens": 5458827.0,
"step": 2965
},
{
"entropy": 5.654081535339356,
"epoch": 0.24807885065152022,
"grad_norm": 1.109375,
"learning_rate": 0.0004996946501796119,
"loss": 5.4954,
"mean_token_accuracy": 0.1681208238005638,
"num_tokens": 5467348.0,
"step": 2970
},
{
"entropy": 5.662711048126221,
"epoch": 0.24849649181423322,
"grad_norm": 1.015625,
"learning_rate": 0.0004996930977774118,
"loss": 5.5766,
"mean_token_accuracy": 0.15531770139932632,
"num_tokens": 5476645.0,
"step": 2975
},
{
"entropy": 5.694996976852417,
"epoch": 0.2489141329769462,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996915414416885,
"loss": 5.6111,
"mean_token_accuracy": 0.16002191007137298,
"num_tokens": 5486120.0,
"step": 2980
},
{
"entropy": 5.694322109222412,
"epoch": 0.2493317741396592,
"grad_norm": 1.078125,
"learning_rate": 0.0004996899811724693,
"loss": 5.5809,
"mean_token_accuracy": 0.16251296252012254,
"num_tokens": 5495313.0,
"step": 2985
},
{
"entropy": 5.6362920761108395,
"epoch": 0.2497494153023722,
"grad_norm": 1.0,
"learning_rate": 0.0004996884169697816,
"loss": 5.5924,
"mean_token_accuracy": 0.16137247085571288,
"num_tokens": 5504587.0,
"step": 2990
},
{
"entropy": 5.584541702270508,
"epoch": 0.2501670564650852,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996868488336526,
"loss": 5.6032,
"mean_token_accuracy": 0.16095196455717087,
"num_tokens": 5514188.0,
"step": 2995
},
{
"entropy": 5.6625102996826175,
"epoch": 0.25058469762779817,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996852767641099,
"loss": 5.6634,
"mean_token_accuracy": 0.15933622270822526,
"num_tokens": 5524075.0,
"step": 3000
},
{
"epoch": 0.25058469762779817,
"eval_entropy": 5.569995272288356,
"eval_loss": 5.59145975112915,
"eval_mean_token_accuracy": 0.17057215184718214,
"eval_num_tokens": 5524075.0,
"eval_runtime": 27.3414,
"eval_samples_per_second": 1381.167,
"eval_steps_per_second": 172.669,
"step": 3000
},
{
"entropy": 5.607466077804565,
"epoch": 0.2510023387905112,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996837007611811,
"loss": 5.431,
"mean_token_accuracy": 0.16868284195661545,
"num_tokens": 5533323.0,
"step": 3005
},
{
"entropy": 5.568152904510498,
"epoch": 0.2514199799532242,
"grad_norm": 1.03125,
"learning_rate": 0.0004996821208248935,
"loss": 5.4497,
"mean_token_accuracy": 0.1743725836277008,
"num_tokens": 5542288.0,
"step": 3010
},
{
"entropy": 5.700623035430908,
"epoch": 0.25183762111593716,
"grad_norm": 1.1875,
"learning_rate": 0.0004996805369552749,
"loss": 5.6288,
"mean_token_accuracy": 0.15931995809078217,
"num_tokens": 5550869.0,
"step": 3015
},
{
"entropy": 5.568376159667968,
"epoch": 0.2522552622786502,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996789491523532,
"loss": 5.5462,
"mean_token_accuracy": 0.1728607803583145,
"num_tokens": 5560147.0,
"step": 3020
},
{
"entropy": 5.533891201019287,
"epoch": 0.2526729034413632,
"grad_norm": 1.078125,
"learning_rate": 0.0004996773574161559,
"loss": 5.4473,
"mean_token_accuracy": 0.17114736139774323,
"num_tokens": 5568358.0,
"step": 3025
},
{
"entropy": 5.59341607093811,
"epoch": 0.25309054460407615,
"grad_norm": 1.03125,
"learning_rate": 0.0004996757617467112,
"loss": 5.5637,
"mean_token_accuracy": 0.16135943233966826,
"num_tokens": 5577755.0,
"step": 3030
},
{
"entropy": 5.690317440032959,
"epoch": 0.2535081857667892,
"grad_norm": 1.015625,
"learning_rate": 0.0004996741621440466,
"loss": 5.5915,
"mean_token_accuracy": 0.15686252117156982,
"num_tokens": 5587515.0,
"step": 3035
},
{
"entropy": 5.739242792129517,
"epoch": 0.25392582692950216,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996725586081906,
"loss": 5.743,
"mean_token_accuracy": 0.15502790808677674,
"num_tokens": 5597860.0,
"step": 3040
},
{
"entropy": 5.684181785583496,
"epoch": 0.25434346809221514,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996709511391709,
"loss": 5.611,
"mean_token_accuracy": 0.1606859192252159,
"num_tokens": 5607302.0,
"step": 3045
},
{
"entropy": 5.653680467605591,
"epoch": 0.2547611092549282,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996693397370158,
"loss": 5.6661,
"mean_token_accuracy": 0.1618943840265274,
"num_tokens": 5616345.0,
"step": 3050
},
{
"entropy": 5.65850076675415,
"epoch": 0.25517875041764115,
"grad_norm": 1.1875,
"learning_rate": 0.0004996677244017534,
"loss": 5.4552,
"mean_token_accuracy": 0.16565490514039993,
"num_tokens": 5624745.0,
"step": 3055
},
{
"entropy": 5.620255422592163,
"epoch": 0.25559639158035413,
"grad_norm": 1.078125,
"learning_rate": 0.0004996661051334121,
"loss": 5.5859,
"mean_token_accuracy": 0.15683500468730927,
"num_tokens": 5634022.0,
"step": 3060
},
{
"entropy": 5.679241800308228,
"epoch": 0.25601403274306717,
"grad_norm": 1.015625,
"learning_rate": 0.0004996644819320201,
"loss": 5.529,
"mean_token_accuracy": 0.16621628701686858,
"num_tokens": 5643631.0,
"step": 3065
},
{
"entropy": 5.661149644851685,
"epoch": 0.25643167390578014,
"grad_norm": 1.125,
"learning_rate": 0.0004996628547976061,
"loss": 5.6308,
"mean_token_accuracy": 0.15741878598928452,
"num_tokens": 5653155.0,
"step": 3070
},
{
"entropy": 5.610762977600098,
"epoch": 0.2568493150684932,
"grad_norm": 1.015625,
"learning_rate": 0.0004996612237301983,
"loss": 5.6105,
"mean_token_accuracy": 0.16782927215099336,
"num_tokens": 5663091.0,
"step": 3075
},
{
"entropy": 5.592285108566284,
"epoch": 0.25726695623120616,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996595887298254,
"loss": 5.5236,
"mean_token_accuracy": 0.16966763138771057,
"num_tokens": 5671885.0,
"step": 3080
},
{
"entropy": 5.6391643524169925,
"epoch": 0.25768459739391913,
"grad_norm": 1.15625,
"learning_rate": 0.000499657949796516,
"loss": 5.5548,
"mean_token_accuracy": 0.16405507028102875,
"num_tokens": 5680225.0,
"step": 3085
},
{
"entropy": 5.484503316879272,
"epoch": 0.25810223855663217,
"grad_norm": 1.046875,
"learning_rate": 0.0004996563069302986,
"loss": 5.4109,
"mean_token_accuracy": 0.17345360964536666,
"num_tokens": 5689456.0,
"step": 3090
},
{
"entropy": 5.612054824829102,
"epoch": 0.25851987971934515,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996546601312023,
"loss": 5.5153,
"mean_token_accuracy": 0.16717812567949294,
"num_tokens": 5698127.0,
"step": 3095
},
{
"entropy": 5.540970039367676,
"epoch": 0.2589375208820581,
"grad_norm": 1.125,
"learning_rate": 0.0004996530093992558,
"loss": 5.5582,
"mean_token_accuracy": 0.1660047322511673,
"num_tokens": 5706825.0,
"step": 3100
},
{
"entropy": 5.6462705612182615,
"epoch": 0.25935516204477116,
"grad_norm": 1.015625,
"learning_rate": 0.0004996513547344877,
"loss": 5.5099,
"mean_token_accuracy": 0.16422566175460815,
"num_tokens": 5715677.0,
"step": 3105
},
{
"entropy": 5.534139490127563,
"epoch": 0.25977280320748414,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996496961369274,
"loss": 5.4365,
"mean_token_accuracy": 0.16353990882635117,
"num_tokens": 5725371.0,
"step": 3110
},
{
"entropy": 5.5601325035095215,
"epoch": 0.2601904443701971,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996480336066037,
"loss": 5.5351,
"mean_token_accuracy": 0.16860325932502745,
"num_tokens": 5734664.0,
"step": 3115
},
{
"entropy": 5.640633153915405,
"epoch": 0.26060808553291015,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996463671435458,
"loss": 5.5432,
"mean_token_accuracy": 0.16731960326433182,
"num_tokens": 5743870.0,
"step": 3120
},
{
"entropy": 5.521408987045288,
"epoch": 0.26102572669562313,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996446967477829,
"loss": 5.545,
"mean_token_accuracy": 0.16755045354366302,
"num_tokens": 5752855.0,
"step": 3125
},
{
"entropy": 5.6759865283966064,
"epoch": 0.2614433678583361,
"grad_norm": 0.9375,
"learning_rate": 0.0004996430224193441,
"loss": 5.6123,
"mean_token_accuracy": 0.160480996966362,
"num_tokens": 5762435.0,
"step": 3130
},
{
"entropy": 5.653593683242798,
"epoch": 0.26186100902104914,
"grad_norm": 1.125,
"learning_rate": 0.0004996413441582589,
"loss": 5.5916,
"mean_token_accuracy": 0.16109202653169633,
"num_tokens": 5771463.0,
"step": 3135
},
{
"entropy": 5.657980346679688,
"epoch": 0.2622786501837621,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996396619645564,
"loss": 5.6074,
"mean_token_accuracy": 0.15687217712402343,
"num_tokens": 5780530.0,
"step": 3140
},
{
"entropy": 5.585852479934692,
"epoch": 0.2626962913464751,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996379758382662,
"loss": 5.4728,
"mean_token_accuracy": 0.17368066161870957,
"num_tokens": 5789199.0,
"step": 3145
},
{
"entropy": 5.7254638671875,
"epoch": 0.26311393250918813,
"grad_norm": 0.90625,
"learning_rate": 0.000499636285779418,
"loss": 5.6648,
"mean_token_accuracy": 0.1626562237739563,
"num_tokens": 5799170.0,
"step": 3150
},
{
"entropy": 5.670967721939087,
"epoch": 0.2635315736719011,
"grad_norm": 1.046875,
"learning_rate": 0.0004996345917880412,
"loss": 5.576,
"mean_token_accuracy": 0.16182495653629303,
"num_tokens": 5807662.0,
"step": 3155
},
{
"entropy": 5.527741003036499,
"epoch": 0.2639492148346141,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996328938641654,
"loss": 5.4795,
"mean_token_accuracy": 0.16211243867874145,
"num_tokens": 5815790.0,
"step": 3160
},
{
"entropy": 5.652725696563721,
"epoch": 0.2643668559973271,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996311920078204,
"loss": 5.4905,
"mean_token_accuracy": 0.169982647895813,
"num_tokens": 5824746.0,
"step": 3165
},
{
"entropy": 5.504961204528809,
"epoch": 0.2647844971600401,
"grad_norm": 1.046875,
"learning_rate": 0.0004996294862190361,
"loss": 5.5063,
"mean_token_accuracy": 0.16816410273313523,
"num_tokens": 5833678.0,
"step": 3170
},
{
"entropy": 5.624357748031616,
"epoch": 0.2652021383227531,
"grad_norm": 1.109375,
"learning_rate": 0.000499627776497842,
"loss": 5.5739,
"mean_token_accuracy": 0.16289925575256348,
"num_tokens": 5842218.0,
"step": 3175
},
{
"entropy": 5.754259538650513,
"epoch": 0.2656197794854661,
"grad_norm": 1.09375,
"learning_rate": 0.0004996260628442686,
"loss": 5.7347,
"mean_token_accuracy": 0.15551001206040382,
"num_tokens": 5853250.0,
"step": 3180
},
{
"entropy": 5.6128898620605465,
"epoch": 0.2660374206481791,
"grad_norm": 1.046875,
"learning_rate": 0.0004996243452583454,
"loss": 5.4719,
"mean_token_accuracy": 0.17064643502235413,
"num_tokens": 5862661.0,
"step": 3185
},
{
"entropy": 5.589026355743409,
"epoch": 0.26645506181089207,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996226237401028,
"loss": 5.6404,
"mean_token_accuracy": 0.15764035731554032,
"num_tokens": 5872327.0,
"step": 3190
},
{
"entropy": 5.574653387069702,
"epoch": 0.2668727029736051,
"grad_norm": 1.0625,
"learning_rate": 0.0004996208982895706,
"loss": 5.4664,
"mean_token_accuracy": 0.16750174760818481,
"num_tokens": 5881539.0,
"step": 3195
},
{
"entropy": 5.666677093505859,
"epoch": 0.2672903441363181,
"grad_norm": 1.109375,
"learning_rate": 0.0004996191689067794,
"loss": 5.5889,
"mean_token_accuracy": 0.1603172555565834,
"num_tokens": 5891602.0,
"step": 3200
},
{
"entropy": 5.566559886932373,
"epoch": 0.26770798529903106,
"grad_norm": 1.0625,
"learning_rate": 0.0004996174355917591,
"loss": 5.5752,
"mean_token_accuracy": 0.16306904405355455,
"num_tokens": 5899831.0,
"step": 3205
},
{
"entropy": 5.741880655288696,
"epoch": 0.2681256264617441,
"grad_norm": 1.015625,
"learning_rate": 0.0004996156983445404,
"loss": 5.6795,
"mean_token_accuracy": 0.15481222420930862,
"num_tokens": 5909193.0,
"step": 3210
},
{
"entropy": 5.702956914901733,
"epoch": 0.26854326762445707,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996139571651534,
"loss": 5.5792,
"mean_token_accuracy": 0.16287053525447845,
"num_tokens": 5919353.0,
"step": 3215
},
{
"entropy": 5.530245161056518,
"epoch": 0.26896090878717005,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996122120536288,
"loss": 5.5154,
"mean_token_accuracy": 0.16863827109336854,
"num_tokens": 5927853.0,
"step": 3220
},
{
"entropy": 5.595543336868286,
"epoch": 0.2693785499498831,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996104630099969,
"loss": 5.6132,
"mean_token_accuracy": 0.16393956243991853,
"num_tokens": 5937943.0,
"step": 3225
},
{
"entropy": 5.6530517578125,
"epoch": 0.26979619111259606,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996087100342885,
"loss": 5.5182,
"mean_token_accuracy": 0.17117328196763992,
"num_tokens": 5946213.0,
"step": 3230
},
{
"entropy": 5.550307941436768,
"epoch": 0.27021383227530904,
"grad_norm": 1.078125,
"learning_rate": 0.0004996069531265345,
"loss": 5.4276,
"mean_token_accuracy": 0.16582605391740798,
"num_tokens": 5954997.0,
"step": 3235
},
{
"entropy": 5.548944425582886,
"epoch": 0.2706314734380221,
"grad_norm": 1.0625,
"learning_rate": 0.0004996051922867653,
"loss": 5.4735,
"mean_token_accuracy": 0.17404431700706482,
"num_tokens": 5963947.0,
"step": 3240
},
{
"entropy": 5.595423603057862,
"epoch": 0.27104911460073505,
"grad_norm": 1.078125,
"learning_rate": 0.0004996034275150118,
"loss": 5.5175,
"mean_token_accuracy": 0.16747236698865892,
"num_tokens": 5972077.0,
"step": 3245
},
{
"entropy": 5.555712890625,
"epoch": 0.27146675576344803,
"grad_norm": 1.0234375,
"learning_rate": 0.000499601658811305,
"loss": 5.507,
"mean_token_accuracy": 0.1668444722890854,
"num_tokens": 5982649.0,
"step": 3250
},
{
"entropy": 5.624109172821045,
"epoch": 0.27188439692616106,
"grad_norm": 1.0625,
"learning_rate": 0.0004995998861756758,
"loss": 5.523,
"mean_token_accuracy": 0.16545868515968323,
"num_tokens": 5991349.0,
"step": 3255
},
{
"entropy": 5.545545864105224,
"epoch": 0.27230203808887404,
"grad_norm": 1.0,
"learning_rate": 0.0004995981096081552,
"loss": 5.5057,
"mean_token_accuracy": 0.1664276748895645,
"num_tokens": 6000555.0,
"step": 3260
},
{
"entropy": 5.717396688461304,
"epoch": 0.272719679251587,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995963291087744,
"loss": 5.6208,
"mean_token_accuracy": 0.15836774557828903,
"num_tokens": 6009554.0,
"step": 3265
},
{
"entropy": 5.7361825466156,
"epoch": 0.27313732041430006,
"grad_norm": 1.0,
"learning_rate": 0.0004995945446775645,
"loss": 5.7257,
"mean_token_accuracy": 0.1579975128173828,
"num_tokens": 6019330.0,
"step": 3270
},
{
"entropy": 5.727784776687622,
"epoch": 0.27355496157701303,
"grad_norm": 1.0625,
"learning_rate": 0.0004995927563145566,
"loss": 5.6603,
"mean_token_accuracy": 0.15948612093925477,
"num_tokens": 6028527.0,
"step": 3275
},
{
"entropy": 5.565222692489624,
"epoch": 0.273972602739726,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995909640197825,
"loss": 5.5236,
"mean_token_accuracy": 0.16562965959310533,
"num_tokens": 6038223.0,
"step": 3280
},
{
"entropy": 5.5374757766723635,
"epoch": 0.27439024390243905,
"grad_norm": 1.0,
"learning_rate": 0.0004995891677932729,
"loss": 5.4592,
"mean_token_accuracy": 0.16831437796354293,
"num_tokens": 6047459.0,
"step": 3285
},
{
"entropy": 5.751672601699829,
"epoch": 0.274807885065152,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995873676350597,
"loss": 5.6572,
"mean_token_accuracy": 0.161651611328125,
"num_tokens": 6057322.0,
"step": 3290
},
{
"entropy": 5.653816175460816,
"epoch": 0.275225526227865,
"grad_norm": 1.0,
"learning_rate": 0.0004995855635451743,
"loss": 5.7703,
"mean_token_accuracy": 0.1549335740506649,
"num_tokens": 6067520.0,
"step": 3295
},
{
"entropy": 5.689868640899658,
"epoch": 0.27564316739057804,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995837555236484,
"loss": 5.6557,
"mean_token_accuracy": 0.15675333738327027,
"num_tokens": 6076729.0,
"step": 3300
},
{
"entropy": 5.683741521835327,
"epoch": 0.276060808553291,
"grad_norm": 1.0,
"learning_rate": 0.0004995819435705133,
"loss": 5.5808,
"mean_token_accuracy": 0.1602635681629181,
"num_tokens": 6086529.0,
"step": 3305
},
{
"entropy": 5.68086347579956,
"epoch": 0.276478449716004,
"grad_norm": 1.046875,
"learning_rate": 0.0004995801276858011,
"loss": 5.6091,
"mean_token_accuracy": 0.16630065739154815,
"num_tokens": 6095784.0,
"step": 3310
},
{
"entropy": 5.7539050579071045,
"epoch": 0.276896090878717,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995783078695434,
"loss": 5.6169,
"mean_token_accuracy": 0.16327303051948547,
"num_tokens": 6104626.0,
"step": 3315
},
{
"entropy": 5.603176927566528,
"epoch": 0.27731373204143,
"grad_norm": 1.0625,
"learning_rate": 0.000499576484121772,
"loss": 5.5308,
"mean_token_accuracy": 0.16884438395500184,
"num_tokens": 6113758.0,
"step": 3320
},
{
"entropy": 5.658589696884155,
"epoch": 0.277731373204143,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995746564425191,
"loss": 5.5663,
"mean_token_accuracy": 0.16707850992679596,
"num_tokens": 6123265.0,
"step": 3325
},
{
"entropy": 5.62137656211853,
"epoch": 0.278149014366856,
"grad_norm": 1.015625,
"learning_rate": 0.0004995728248318163,
"loss": 5.6163,
"mean_token_accuracy": 0.15987214148044587,
"num_tokens": 6132855.0,
"step": 3330
},
{
"entropy": 5.573479270935058,
"epoch": 0.278566655529569,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995709892896961,
"loss": 5.4933,
"mean_token_accuracy": 0.16592755168676376,
"num_tokens": 6142739.0,
"step": 3335
},
{
"entropy": 5.52806396484375,
"epoch": 0.278984296692282,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995691498161903,
"loss": 5.4458,
"mean_token_accuracy": 0.16964126229286194,
"num_tokens": 6152169.0,
"step": 3340
},
{
"entropy": 5.465058279037476,
"epoch": 0.279401937854995,
"grad_norm": 1.09375,
"learning_rate": 0.0004995673064113313,
"loss": 5.4693,
"mean_token_accuracy": 0.17196338325738908,
"num_tokens": 6160235.0,
"step": 3345
},
{
"entropy": 5.574095296859741,
"epoch": 0.279819579017708,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995654590751513,
"loss": 5.4651,
"mean_token_accuracy": 0.16654023975133897,
"num_tokens": 6169440.0,
"step": 3350
},
{
"entropy": 5.548044109344483,
"epoch": 0.28023722018042097,
"grad_norm": 1.03125,
"learning_rate": 0.0004995636078076825,
"loss": 5.4853,
"mean_token_accuracy": 0.16782020777463913,
"num_tokens": 6178456.0,
"step": 3355
},
{
"entropy": 5.558378124237061,
"epoch": 0.280654861343134,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995617526089575,
"loss": 5.5636,
"mean_token_accuracy": 0.15922029614448546,
"num_tokens": 6186844.0,
"step": 3360
},
{
"entropy": 5.6190282821655275,
"epoch": 0.281072502505847,
"grad_norm": 1.03125,
"learning_rate": 0.0004995598934790087,
"loss": 5.4815,
"mean_token_accuracy": 0.16773421615362166,
"num_tokens": 6196117.0,
"step": 3365
},
{
"entropy": 5.659291076660156,
"epoch": 0.28149014366855996,
"grad_norm": 1.09375,
"learning_rate": 0.0004995580304178687,
"loss": 5.4587,
"mean_token_accuracy": 0.17263172715902328,
"num_tokens": 6204636.0,
"step": 3370
},
{
"entropy": 5.435099983215332,
"epoch": 0.281907784831273,
"grad_norm": 1.0,
"learning_rate": 0.0004995561634255701,
"loss": 5.5031,
"mean_token_accuracy": 0.1672251433134079,
"num_tokens": 6214537.0,
"step": 3375
},
{
"entropy": 5.607173585891724,
"epoch": 0.28232542599398597,
"grad_norm": 1.015625,
"learning_rate": 0.0004995542925021455,
"loss": 5.5677,
"mean_token_accuracy": 0.16606256663799285,
"num_tokens": 6224753.0,
"step": 3380
},
{
"entropy": 5.86768217086792,
"epoch": 0.28274306715669895,
"grad_norm": 1.046875,
"learning_rate": 0.0004995524176476277,
"loss": 5.7346,
"mean_token_accuracy": 0.1616437777876854,
"num_tokens": 6233842.0,
"step": 3385
},
{
"entropy": 5.680369091033936,
"epoch": 0.283160708319412,
"grad_norm": 1.046875,
"learning_rate": 0.0004995505388620497,
"loss": 5.57,
"mean_token_accuracy": 0.16978590339422225,
"num_tokens": 6243467.0,
"step": 3390
},
{
"entropy": 5.504652309417724,
"epoch": 0.28357834948212496,
"grad_norm": 1.046875,
"learning_rate": 0.000499548656145444,
"loss": 5.6123,
"mean_token_accuracy": 0.15932482331991196,
"num_tokens": 6253263.0,
"step": 3395
},
{
"entropy": 5.652474069595337,
"epoch": 0.28399599064483794,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995467694978439,
"loss": 5.4742,
"mean_token_accuracy": 0.17390806525945662,
"num_tokens": 6262249.0,
"step": 3400
},
{
"entropy": 5.500980186462402,
"epoch": 0.28441363180755097,
"grad_norm": 1.125,
"learning_rate": 0.0004995448789192823,
"loss": 5.4878,
"mean_token_accuracy": 0.1721876159310341,
"num_tokens": 6271154.0,
"step": 3405
},
{
"entropy": 5.558666467666626,
"epoch": 0.28483127297026395,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995429844097922,
"loss": 5.5588,
"mean_token_accuracy": 0.16892617344856262,
"num_tokens": 6280377.0,
"step": 3410
},
{
"entropy": 5.660373592376709,
"epoch": 0.2852489141329769,
"grad_norm": 1.09375,
"learning_rate": 0.0004995410859694071,
"loss": 5.5567,
"mean_token_accuracy": 0.16368617564439775,
"num_tokens": 6289941.0,
"step": 3415
},
{
"entropy": 5.639143180847168,
"epoch": 0.28566655529568996,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995391835981599,
"loss": 5.5982,
"mean_token_accuracy": 0.16234036535024643,
"num_tokens": 6299281.0,
"step": 3420
},
{
"entropy": 5.633917808532715,
"epoch": 0.28608419645840294,
"grad_norm": 1.046875,
"learning_rate": 0.000499537277296084,
"loss": 5.5704,
"mean_token_accuracy": 0.16069992482662201,
"num_tokens": 6309408.0,
"step": 3425
},
{
"entropy": 5.566310119628906,
"epoch": 0.2865018376211159,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995353670632127,
"loss": 5.5802,
"mean_token_accuracy": 0.1667129948735237,
"num_tokens": 6318979.0,
"step": 3430
},
{
"entropy": 5.632243919372558,
"epoch": 0.28691947878382895,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995334528995797,
"loss": 5.4096,
"mean_token_accuracy": 0.171867673099041,
"num_tokens": 6327616.0,
"step": 3435
},
{
"entropy": 5.460249996185302,
"epoch": 0.28733711994654193,
"grad_norm": 1.0625,
"learning_rate": 0.0004995315348052182,
"loss": 5.4549,
"mean_token_accuracy": 0.17647364735603333,
"num_tokens": 6336486.0,
"step": 3440
},
{
"entropy": 5.595021772384643,
"epoch": 0.2877547611092549,
"grad_norm": 1.1328125,
"learning_rate": 0.000499529612780162,
"loss": 5.503,
"mean_token_accuracy": 0.16232585459947585,
"num_tokens": 6345671.0,
"step": 3445
},
{
"entropy": 5.663350009918213,
"epoch": 0.28817240227196794,
"grad_norm": 1.125,
"learning_rate": 0.0004995276868244446,
"loss": 5.6095,
"mean_token_accuracy": 0.16484829932451248,
"num_tokens": 6354534.0,
"step": 3450
},
{
"entropy": 5.5742090225219725,
"epoch": 0.2885900434346809,
"grad_norm": 0.9453125,
"learning_rate": 0.0004995257569380998,
"loss": 5.4436,
"mean_token_accuracy": 0.16887350082397462,
"num_tokens": 6364052.0,
"step": 3455
},
{
"entropy": 5.630385494232177,
"epoch": 0.2890076845973939,
"grad_norm": 1.078125,
"learning_rate": 0.0004995238231211614,
"loss": 5.5746,
"mean_token_accuracy": 0.16773510575294495,
"num_tokens": 6373305.0,
"step": 3460
},
{
"entropy": 5.5635439395904545,
"epoch": 0.28942532576010693,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995218853736631,
"loss": 5.4859,
"mean_token_accuracy": 0.16553768813610076,
"num_tokens": 6382673.0,
"step": 3465
},
{
"entropy": 5.706551933288575,
"epoch": 0.2898429669228199,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995199436956391,
"loss": 5.6028,
"mean_token_accuracy": 0.15886935740709304,
"num_tokens": 6392122.0,
"step": 3470
},
{
"entropy": 5.684014892578125,
"epoch": 0.2902606080855329,
"grad_norm": 1.046875,
"learning_rate": 0.0004995179980871231,
"loss": 5.6048,
"mean_token_accuracy": 0.16065345853567123,
"num_tokens": 6401130.0,
"step": 3475
},
{
"entropy": 5.601016521453857,
"epoch": 0.2906782492482459,
"grad_norm": 1.125,
"learning_rate": 0.0004995160485481493,
"loss": 5.4693,
"mean_token_accuracy": 0.16418304592370986,
"num_tokens": 6409471.0,
"step": 3480
},
{
"entropy": 5.565523386001587,
"epoch": 0.2910958904109589,
"grad_norm": 0.99609375,
"learning_rate": 0.000499514095078752,
"loss": 5.4835,
"mean_token_accuracy": 0.16717326939105986,
"num_tokens": 6418902.0,
"step": 3485
},
{
"entropy": 5.587199735641479,
"epoch": 0.2915135315736719,
"grad_norm": 1.0,
"learning_rate": 0.000499512137678965,
"loss": 5.5862,
"mean_token_accuracy": 0.15920206159353256,
"num_tokens": 6428529.0,
"step": 3490
},
{
"entropy": 5.634693336486817,
"epoch": 0.2919311727363849,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995101763488229,
"loss": 5.4355,
"mean_token_accuracy": 0.16721768975257872,
"num_tokens": 6437460.0,
"step": 3495
},
{
"entropy": 5.578807163238525,
"epoch": 0.2923488138990979,
"grad_norm": 1.09375,
"learning_rate": 0.0004995082110883599,
"loss": 5.4875,
"mean_token_accuracy": 0.1697748214006424,
"num_tokens": 6446005.0,
"step": 3500
},
{
"entropy": 5.5946056842803955,
"epoch": 0.29276645506181087,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995062418976104,
"loss": 5.5272,
"mean_token_accuracy": 0.1632330521941185,
"num_tokens": 6455241.0,
"step": 3505
},
{
"entropy": 5.615413665771484,
"epoch": 0.2931840962245239,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995042687766089,
"loss": 5.5068,
"mean_token_accuracy": 0.17199014723300934,
"num_tokens": 6464748.0,
"step": 3510
},
{
"entropy": 5.603789901733398,
"epoch": 0.2936017373872369,
"grad_norm": 1.03125,
"learning_rate": 0.0004995022917253899,
"loss": 5.5727,
"mean_token_accuracy": 0.16414013504981995,
"num_tokens": 6473834.0,
"step": 3515
},
{
"entropy": 5.587377595901489,
"epoch": 0.29401937854994986,
"grad_norm": 1.015625,
"learning_rate": 0.0004995003107439882,
"loss": 5.5077,
"mean_token_accuracy": 0.1692238926887512,
"num_tokens": 6482169.0,
"step": 3520
},
{
"entropy": 5.498481750488281,
"epoch": 0.2944370197126629,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994983258324383,
"loss": 5.4608,
"mean_token_accuracy": 0.17181194126605986,
"num_tokens": 6491798.0,
"step": 3525
},
{
"entropy": 5.463162231445312,
"epoch": 0.2948546608753759,
"grad_norm": 1.140625,
"learning_rate": 0.0004994963369907748,
"loss": 5.3877,
"mean_token_accuracy": 0.17222898602485656,
"num_tokens": 6500143.0,
"step": 3530
},
{
"entropy": 5.472366762161255,
"epoch": 0.29527230203808885,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994943442190328,
"loss": 5.533,
"mean_token_accuracy": 0.16390003114938737,
"num_tokens": 6509745.0,
"step": 3535
},
{
"entropy": 5.751410436630249,
"epoch": 0.2956899432008019,
"grad_norm": 0.95703125,
"learning_rate": 0.0004994923475172471,
"loss": 5.617,
"mean_token_accuracy": 0.16012435555458068,
"num_tokens": 6520931.0,
"step": 3540
},
{
"entropy": 5.605918979644775,
"epoch": 0.29610758436351486,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994903468854527,
"loss": 5.5381,
"mean_token_accuracy": 0.16189739108085632,
"num_tokens": 6529774.0,
"step": 3545
},
{
"entropy": 5.493441534042359,
"epoch": 0.29652522552622784,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994883423236843,
"loss": 5.4599,
"mean_token_accuracy": 0.16605947762727738,
"num_tokens": 6538922.0,
"step": 3550
},
{
"entropy": 5.556844329833984,
"epoch": 0.2969428666889409,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994863338319775,
"loss": 5.4911,
"mean_token_accuracy": 0.17253804653882981,
"num_tokens": 6548891.0,
"step": 3555
},
{
"entropy": 5.445564317703247,
"epoch": 0.29736050785165385,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994843214103672,
"loss": 5.3205,
"mean_token_accuracy": 0.17109228670597076,
"num_tokens": 6557025.0,
"step": 3560
},
{
"entropy": 5.546029663085937,
"epoch": 0.29777814901436683,
"grad_norm": 1.03125,
"learning_rate": 0.0004994823050588886,
"loss": 5.5046,
"mean_token_accuracy": 0.1655927926301956,
"num_tokens": 6567037.0,
"step": 3565
},
{
"entropy": 5.6032483100891115,
"epoch": 0.29819579017707987,
"grad_norm": 1.03125,
"learning_rate": 0.000499480284777577,
"loss": 5.5363,
"mean_token_accuracy": 0.167324697971344,
"num_tokens": 6576766.0,
"step": 3570
},
{
"entropy": 5.607816743850708,
"epoch": 0.29861343133979285,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994782605664677,
"loss": 5.5051,
"mean_token_accuracy": 0.16765168458223342,
"num_tokens": 6585576.0,
"step": 3575
},
{
"entropy": 5.629946756362915,
"epoch": 0.2990310725025058,
"grad_norm": 1.15625,
"learning_rate": 0.0004994762324255965,
"loss": 5.5717,
"mean_token_accuracy": 0.16156770884990693,
"num_tokens": 6594463.0,
"step": 3580
},
{
"entropy": 5.542790174484253,
"epoch": 0.29944871366521886,
"grad_norm": 1.0625,
"learning_rate": 0.0004994742003549986,
"loss": 5.4513,
"mean_token_accuracy": 0.16937345415353774,
"num_tokens": 6603651.0,
"step": 3585
},
{
"entropy": 5.59451060295105,
"epoch": 0.29986635482793184,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994721643547095,
"loss": 5.523,
"mean_token_accuracy": 0.17147031277418137,
"num_tokens": 6612950.0,
"step": 3590
},
{
"entropy": 5.594419384002686,
"epoch": 0.3002839959906448,
"grad_norm": 1.0390625,
"learning_rate": 0.000499470124424765,
"loss": 5.456,
"mean_token_accuracy": 0.1732052370905876,
"num_tokens": 6622013.0,
"step": 3595
},
{
"entropy": 5.572998094558716,
"epoch": 0.30070163715335785,
"grad_norm": 1.046875,
"learning_rate": 0.0004994680805652008,
"loss": 5.5265,
"mean_token_accuracy": 0.16979334205389024,
"num_tokens": 6632046.0,
"step": 3600
},
{
"entropy": 5.54917950630188,
"epoch": 0.3011192783160708,
"grad_norm": 1.078125,
"learning_rate": 0.0004994660327760527,
"loss": 5.5947,
"mean_token_accuracy": 0.17079425603151321,
"num_tokens": 6641169.0,
"step": 3605
},
{
"entropy": 5.5686663627624515,
"epoch": 0.3015369194787838,
"grad_norm": 1.015625,
"learning_rate": 0.0004994639810573565,
"loss": 5.4001,
"mean_token_accuracy": 0.17856993675231933,
"num_tokens": 6649663.0,
"step": 3610
},
{
"entropy": 5.638326120376587,
"epoch": 0.30195456064149684,
"grad_norm": 1.0703125,
"learning_rate": 0.000499461925409148,
"loss": 5.5769,
"mean_token_accuracy": 0.16235045343637466,
"num_tokens": 6659448.0,
"step": 3615
},
{
"entropy": 5.658876085281372,
"epoch": 0.3023722018042098,
"grad_norm": 1.03125,
"learning_rate": 0.0004994598658314634,
"loss": 5.5319,
"mean_token_accuracy": 0.1667335346341133,
"num_tokens": 6670314.0,
"step": 3620
},
{
"entropy": 5.52158145904541,
"epoch": 0.3027898429669228,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994578023243386,
"loss": 5.49,
"mean_token_accuracy": 0.1768211081624031,
"num_tokens": 6679413.0,
"step": 3625
},
{
"entropy": 5.5182188034057615,
"epoch": 0.30320748412963583,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994557348878098,
"loss": 5.4711,
"mean_token_accuracy": 0.17201100587844848,
"num_tokens": 6688439.0,
"step": 3630
},
{
"entropy": 5.570156574249268,
"epoch": 0.3036251252923488,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994536635219133,
"loss": 5.4552,
"mean_token_accuracy": 0.17211078852415085,
"num_tokens": 6697471.0,
"step": 3635
},
{
"entropy": 5.649779844284057,
"epoch": 0.3040427664550618,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994515882266852,
"loss": 5.5259,
"mean_token_accuracy": 0.1673920840024948,
"num_tokens": 6706698.0,
"step": 3640
},
{
"entropy": 5.439210033416748,
"epoch": 0.3044604076177748,
"grad_norm": 1.109375,
"learning_rate": 0.0004994495090021618,
"loss": 5.4351,
"mean_token_accuracy": 0.1794208511710167,
"num_tokens": 6715676.0,
"step": 3645
},
{
"entropy": 5.5099845886230465,
"epoch": 0.3048780487804878,
"grad_norm": 1.15625,
"learning_rate": 0.0004994474258483796,
"loss": 5.4778,
"mean_token_accuracy": 0.17199191451072693,
"num_tokens": 6723818.0,
"step": 3650
},
{
"entropy": 5.608595657348633,
"epoch": 0.3052956899432008,
"grad_norm": 1.046875,
"learning_rate": 0.0004994453387653751,
"loss": 5.4945,
"mean_token_accuracy": 0.16601282060146333,
"num_tokens": 6733450.0,
"step": 3655
},
{
"entropy": 5.546295404434204,
"epoch": 0.3057133311059138,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994432477531847,
"loss": 5.4255,
"mean_token_accuracy": 0.1695643410086632,
"num_tokens": 6743142.0,
"step": 3660
},
{
"entropy": 5.480252599716186,
"epoch": 0.3061309722686268,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994411528118452,
"loss": 5.3809,
"mean_token_accuracy": 0.1699554443359375,
"num_tokens": 6752185.0,
"step": 3665
},
{
"entropy": 5.451222324371338,
"epoch": 0.30654861343133977,
"grad_norm": 1.015625,
"learning_rate": 0.0004994390539413931,
"loss": 5.4721,
"mean_token_accuracy": 0.17315813302993774,
"num_tokens": 6761549.0,
"step": 3670
},
{
"entropy": 5.60224027633667,
"epoch": 0.3069662545940528,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994369511418652,
"loss": 5.4971,
"mean_token_accuracy": 0.1704183340072632,
"num_tokens": 6770395.0,
"step": 3675
},
{
"entropy": 5.684746551513672,
"epoch": 0.3073838957567658,
"grad_norm": 1.109375,
"learning_rate": 0.0004994348444132983,
"loss": 5.5669,
"mean_token_accuracy": 0.1692931741476059,
"num_tokens": 6779154.0,
"step": 3680
},
{
"entropy": 5.476674604415893,
"epoch": 0.30780153691947876,
"grad_norm": 1.0625,
"learning_rate": 0.0004994327337557293,
"loss": 5.4648,
"mean_token_accuracy": 0.17590045183897018,
"num_tokens": 6789263.0,
"step": 3685
},
{
"entropy": 5.602137660980224,
"epoch": 0.3082191780821918,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994306191691952,
"loss": 5.5512,
"mean_token_accuracy": 0.16789348274469376,
"num_tokens": 6797985.0,
"step": 3690
},
{
"entropy": 5.550741529464721,
"epoch": 0.30863681924490477,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994285006537329,
"loss": 5.4257,
"mean_token_accuracy": 0.1740383103489876,
"num_tokens": 6807566.0,
"step": 3695
},
{
"entropy": 5.488746929168701,
"epoch": 0.30905446040761775,
"grad_norm": 1.09375,
"learning_rate": 0.0004994263782093796,
"loss": 5.4969,
"mean_token_accuracy": 0.16826924532651902,
"num_tokens": 6815637.0,
"step": 3700
},
{
"entropy": 5.506741046905518,
"epoch": 0.3094721015703308,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994242518361723,
"loss": 5.3749,
"mean_token_accuracy": 0.1795153483748436,
"num_tokens": 6823721.0,
"step": 3705
},
{
"entropy": 5.6497314929962155,
"epoch": 0.30988974273304376,
"grad_norm": 0.953125,
"learning_rate": 0.0004994221215341484,
"loss": 5.5426,
"mean_token_accuracy": 0.1678917244076729,
"num_tokens": 6833437.0,
"step": 3710
},
{
"entropy": 5.545580911636352,
"epoch": 0.31030738389575674,
"grad_norm": 1.0625,
"learning_rate": 0.0004994199873033452,
"loss": 5.5965,
"mean_token_accuracy": 0.16947946548461915,
"num_tokens": 6842287.0,
"step": 3715
},
{
"entropy": 5.63468747138977,
"epoch": 0.3107250250584698,
"grad_norm": 0.96875,
"learning_rate": 0.0004994178491437999,
"loss": 5.4328,
"mean_token_accuracy": 0.16542082726955415,
"num_tokens": 6851520.0,
"step": 3720
},
{
"entropy": 5.4503031253814695,
"epoch": 0.31114266622118275,
"grad_norm": 0.953125,
"learning_rate": 0.0004994157070555501,
"loss": 5.4902,
"mean_token_accuracy": 0.17241426259279252,
"num_tokens": 6861219.0,
"step": 3725
},
{
"entropy": 5.613488674163818,
"epoch": 0.31156030738389573,
"grad_norm": 1.109375,
"learning_rate": 0.0004994135610386331,
"loss": 5.6055,
"mean_token_accuracy": 0.16569115072488785,
"num_tokens": 6870525.0,
"step": 3730
},
{
"entropy": 5.681842231750489,
"epoch": 0.31197794854660876,
"grad_norm": 0.95703125,
"learning_rate": 0.0004994114110930867,
"loss": 5.5379,
"mean_token_accuracy": 0.1655278518795967,
"num_tokens": 6880183.0,
"step": 3735
},
{
"entropy": 5.537979507446289,
"epoch": 0.31239558970932174,
"grad_norm": 1.0,
"learning_rate": 0.0004994092572189483,
"loss": 5.5689,
"mean_token_accuracy": 0.161477729678154,
"num_tokens": 6890442.0,
"step": 3740
},
{
"entropy": 5.574763774871826,
"epoch": 0.3128132308720348,
"grad_norm": 1.09375,
"learning_rate": 0.0004994070994162558,
"loss": 5.4693,
"mean_token_accuracy": 0.17136416137218474,
"num_tokens": 6900091.0,
"step": 3745
},
{
"entropy": 5.473384141921997,
"epoch": 0.31323087203474775,
"grad_norm": 1.0,
"learning_rate": 0.0004994049376850471,
"loss": 5.5162,
"mean_token_accuracy": 0.16881705969572067,
"num_tokens": 6910489.0,
"step": 3750
},
{
"entropy": 5.612695598602295,
"epoch": 0.31364851319746073,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994027720253596,
"loss": 5.4984,
"mean_token_accuracy": 0.16930713802576064,
"num_tokens": 6919895.0,
"step": 3755
},
{
"entropy": 5.765688180923462,
"epoch": 0.31406615436017377,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994006024372315,
"loss": 5.6344,
"mean_token_accuracy": 0.16401913166046142,
"num_tokens": 6929511.0,
"step": 3760
},
{
"entropy": 5.4873370170593265,
"epoch": 0.31448379552288674,
"grad_norm": 1.0625,
"learning_rate": 0.0004993984289207007,
"loss": 5.4073,
"mean_token_accuracy": 0.17526261806488036,
"num_tokens": 6938059.0,
"step": 3765
},
{
"entropy": 5.518824672698974,
"epoch": 0.3149014366855997,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993962514758054,
"loss": 5.4281,
"mean_token_accuracy": 0.16512711942195893,
"num_tokens": 6946894.0,
"step": 3770
},
{
"entropy": 5.6265708923339846,
"epoch": 0.31531907784831276,
"grad_norm": 1.140625,
"learning_rate": 0.0004993940701025836,
"loss": 5.4361,
"mean_token_accuracy": 0.1635988935828209,
"num_tokens": 6955494.0,
"step": 3775
},
{
"entropy": 5.5025022506713865,
"epoch": 0.31573671901102573,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993918848010734,
"loss": 5.5429,
"mean_token_accuracy": 0.16656569838523866,
"num_tokens": 6964347.0,
"step": 3780
},
{
"entropy": 5.569203281402588,
"epoch": 0.3161543601737387,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993896955713133,
"loss": 5.5004,
"mean_token_accuracy": 0.1626885414123535,
"num_tokens": 6974750.0,
"step": 3785
},
{
"entropy": 5.593154859542847,
"epoch": 0.31657200133645175,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993875024133414,
"loss": 5.5097,
"mean_token_accuracy": 0.16821067929267883,
"num_tokens": 6984228.0,
"step": 3790
},
{
"entropy": 5.488545370101929,
"epoch": 0.3169896424991647,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993853053271962,
"loss": 5.4962,
"mean_token_accuracy": 0.16494379788637162,
"num_tokens": 6993173.0,
"step": 3795
},
{
"entropy": 5.77180233001709,
"epoch": 0.3174072836618777,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993831043129161,
"loss": 5.6863,
"mean_token_accuracy": 0.15461417585611342,
"num_tokens": 7003486.0,
"step": 3800
},
{
"entropy": 5.521976280212402,
"epoch": 0.31782492482459074,
"grad_norm": 1.1875,
"learning_rate": 0.0004993808993705397,
"loss": 5.4328,
"mean_token_accuracy": 0.17399922311306,
"num_tokens": 7011985.0,
"step": 3805
},
{
"entropy": 5.511684131622315,
"epoch": 0.3182425659873037,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993786905001055,
"loss": 5.3938,
"mean_token_accuracy": 0.1734773874282837,
"num_tokens": 7021148.0,
"step": 3810
},
{
"entropy": 5.458965539932251,
"epoch": 0.3186602071500167,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993764777016524,
"loss": 5.4387,
"mean_token_accuracy": 0.17023057043552398,
"num_tokens": 7030110.0,
"step": 3815
},
{
"entropy": 5.603944492340088,
"epoch": 0.31907784831272973,
"grad_norm": 1.109375,
"learning_rate": 0.0004993742609752188,
"loss": 5.519,
"mean_token_accuracy": 0.1731454461812973,
"num_tokens": 7039428.0,
"step": 3820
},
{
"entropy": 5.653282690048218,
"epoch": 0.3194954894754427,
"grad_norm": 1.046875,
"learning_rate": 0.0004993720403208438,
"loss": 5.5768,
"mean_token_accuracy": 0.16301795542240144,
"num_tokens": 7048703.0,
"step": 3825
},
{
"entropy": 5.518553829193115,
"epoch": 0.3199131306381557,
"grad_norm": 1.1171875,
"learning_rate": 0.000499369815738566,
"loss": 5.4644,
"mean_token_accuracy": 0.1645129919052124,
"num_tokens": 7057612.0,
"step": 3830
},
{
"entropy": 5.581686782836914,
"epoch": 0.3203307718008687,
"grad_norm": 1.03125,
"learning_rate": 0.0004993675872284248,
"loss": 5.431,
"mean_token_accuracy": 0.16800006628036498,
"num_tokens": 7066660.0,
"step": 3835
},
{
"entropy": 5.465031623840332,
"epoch": 0.3207484129635817,
"grad_norm": 1.015625,
"learning_rate": 0.0004993653547904586,
"loss": 5.4178,
"mean_token_accuracy": 0.1650697574019432,
"num_tokens": 7075368.0,
"step": 3840
},
{
"entropy": 5.546475076675415,
"epoch": 0.3211660541262947,
"grad_norm": 0.984375,
"learning_rate": 0.000499363118424707,
"loss": 5.5111,
"mean_token_accuracy": 0.17623901069164277,
"num_tokens": 7085276.0,
"step": 3845
},
{
"entropy": 5.562846660614014,
"epoch": 0.3215836952890077,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993608781312088,
"loss": 5.4967,
"mean_token_accuracy": 0.16285595893859864,
"num_tokens": 7094292.0,
"step": 3850
},
{
"entropy": 5.673039293289184,
"epoch": 0.3220013364517207,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993586339100035,
"loss": 5.685,
"mean_token_accuracy": 0.1596235029399395,
"num_tokens": 7104712.0,
"step": 3855
},
{
"entropy": 5.697730255126953,
"epoch": 0.32241897761443367,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993563857611301,
"loss": 5.4939,
"mean_token_accuracy": 0.1786169320344925,
"num_tokens": 7113297.0,
"step": 3860
},
{
"entropy": 5.455618762969971,
"epoch": 0.3228366187771467,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993541336846283,
"loss": 5.5003,
"mean_token_accuracy": 0.1605915516614914,
"num_tokens": 7122706.0,
"step": 3865
},
{
"entropy": 5.571385478973388,
"epoch": 0.3232542599398597,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993518776805372,
"loss": 5.4617,
"mean_token_accuracy": 0.16935173273086548,
"num_tokens": 7131939.0,
"step": 3870
},
{
"entropy": 5.454518365859985,
"epoch": 0.32367190110257266,
"grad_norm": 1.09375,
"learning_rate": 0.0004993496177488965,
"loss": 5.3575,
"mean_token_accuracy": 0.17998364865779876,
"num_tokens": 7141016.0,
"step": 3875
},
{
"entropy": 5.409318161010742,
"epoch": 0.3240895422652857,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993473538897457,
"loss": 5.3123,
"mean_token_accuracy": 0.17959339469671248,
"num_tokens": 7149177.0,
"step": 3880
},
{
"entropy": 5.540026378631592,
"epoch": 0.32450718342799867,
"grad_norm": 1.0625,
"learning_rate": 0.0004993450861031243,
"loss": 5.43,
"mean_token_accuracy": 0.1784561589360237,
"num_tokens": 7157839.0,
"step": 3885
},
{
"entropy": 5.634092378616333,
"epoch": 0.32492482459071165,
"grad_norm": 1.03125,
"learning_rate": 0.0004993428143890722,
"loss": 5.575,
"mean_token_accuracy": 0.16033362746238708,
"num_tokens": 7167759.0,
"step": 3890
},
{
"entropy": 5.643723487854004,
"epoch": 0.3253424657534247,
"grad_norm": 1.125,
"learning_rate": 0.0004993405387476291,
"loss": 5.5168,
"mean_token_accuracy": 0.1591552436351776,
"num_tokens": 7177141.0,
"step": 3895
},
{
"entropy": 5.439709424972534,
"epoch": 0.32576010691613766,
"grad_norm": 1.125,
"learning_rate": 0.0004993382591788349,
"loss": 5.4251,
"mean_token_accuracy": 0.1731685906648636,
"num_tokens": 7185350.0,
"step": 3900
},
{
"entropy": 5.4100688934326175,
"epoch": 0.32617774807885064,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993359756827293,
"loss": 5.3738,
"mean_token_accuracy": 0.173531411588192,
"num_tokens": 7194073.0,
"step": 3905
},
{
"entropy": 5.550141429901123,
"epoch": 0.32659538924156367,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993336882593524,
"loss": 5.533,
"mean_token_accuracy": 0.1637530192732811,
"num_tokens": 7203421.0,
"step": 3910
},
{
"entropy": 5.485321187973023,
"epoch": 0.32701303040427665,
"grad_norm": 1.140625,
"learning_rate": 0.0004993313969087444,
"loss": 5.321,
"mean_token_accuracy": 0.17406193614006044,
"num_tokens": 7212548.0,
"step": 3915
},
{
"entropy": 5.566910362243652,
"epoch": 0.32743067156698963,
"grad_norm": 1.078125,
"learning_rate": 0.0004993291016309452,
"loss": 5.5545,
"mean_token_accuracy": 0.16739266812801362,
"num_tokens": 7221668.0,
"step": 3920
},
{
"entropy": 5.486429595947266,
"epoch": 0.32784831272970266,
"grad_norm": 0.99609375,
"learning_rate": 0.000499326802425995,
"loss": 5.4428,
"mean_token_accuracy": 0.1691237285733223,
"num_tokens": 7231775.0,
"step": 3925
},
{
"entropy": 5.569035339355469,
"epoch": 0.32826595389241564,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993244992939342,
"loss": 5.4755,
"mean_token_accuracy": 0.16648643761873244,
"num_tokens": 7241264.0,
"step": 3930
},
{
"entropy": 5.548462629318237,
"epoch": 0.3286835950551286,
"grad_norm": 1.0859375,
"learning_rate": 0.000499322192234803,
"loss": 5.5032,
"mean_token_accuracy": 0.1648609831929207,
"num_tokens": 7250519.0,
"step": 3935
},
{
"entropy": 5.657499980926514,
"epoch": 0.32910123621784165,
"grad_norm": 1.1953125,
"learning_rate": 0.0004993198812486418,
"loss": 5.5489,
"mean_token_accuracy": 0.16496647745370865,
"num_tokens": 7259533.0,
"step": 3940
},
{
"entropy": 5.471747827529907,
"epoch": 0.32951887738055463,
"grad_norm": 1.03125,
"learning_rate": 0.000499317566335491,
"loss": 5.3754,
"mean_token_accuracy": 0.16652699708938598,
"num_tokens": 7268980.0,
"step": 3945
},
{
"entropy": 5.540273571014405,
"epoch": 0.3299365185432676,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993152474953913,
"loss": 5.4767,
"mean_token_accuracy": 0.1645531475543976,
"num_tokens": 7277126.0,
"step": 3950
},
{
"entropy": 5.538077116012573,
"epoch": 0.33035415970598064,
"grad_norm": 1.09375,
"learning_rate": 0.0004993129247283831,
"loss": 5.4535,
"mean_token_accuracy": 0.1744188755750656,
"num_tokens": 7286223.0,
"step": 3955
},
{
"entropy": 5.496949481964111,
"epoch": 0.3307718008686936,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993105980345071,
"loss": 5.4623,
"mean_token_accuracy": 0.17597895860671997,
"num_tokens": 7295054.0,
"step": 3960
},
{
"entropy": 5.541554498672485,
"epoch": 0.3311894420314066,
"grad_norm": 1.0625,
"learning_rate": 0.0004993082674138042,
"loss": 5.4763,
"mean_token_accuracy": 0.1700817570090294,
"num_tokens": 7304502.0,
"step": 3965
},
{
"entropy": 5.541930103302002,
"epoch": 0.33160708319411963,
"grad_norm": 0.99609375,
"learning_rate": 0.000499305932866315,
"loss": 5.4328,
"mean_token_accuracy": 0.17302184104919432,
"num_tokens": 7314093.0,
"step": 3970
},
{
"entropy": 5.55930118560791,
"epoch": 0.3320247243568326,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993035943920805,
"loss": 5.4756,
"mean_token_accuracy": 0.17047226876020433,
"num_tokens": 7323059.0,
"step": 3975
},
{
"entropy": 5.656615447998047,
"epoch": 0.3324423655195456,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993012519911416,
"loss": 5.6124,
"mean_token_accuracy": 0.16392915248870848,
"num_tokens": 7332596.0,
"step": 3980
},
{
"entropy": 5.517297315597534,
"epoch": 0.3328600066822586,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992989056635392,
"loss": 5.4105,
"mean_token_accuracy": 0.17303418666124343,
"num_tokens": 7341671.0,
"step": 3985
},
{
"entropy": 5.464655780792237,
"epoch": 0.3332776478449716,
"grad_norm": 1.09375,
"learning_rate": 0.0004992965554093146,
"loss": 5.5355,
"mean_token_accuracy": 0.16607269048690795,
"num_tokens": 7350344.0,
"step": 3990
},
{
"entropy": 5.601719188690185,
"epoch": 0.3336952890076846,
"grad_norm": 1.046875,
"learning_rate": 0.0004992942012285086,
"loss": 5.5448,
"mean_token_accuracy": 0.1706473007798195,
"num_tokens": 7359844.0,
"step": 3995
},
{
"entropy": 5.595786094665527,
"epoch": 0.3341129301703976,
"grad_norm": 1.09375,
"learning_rate": 0.0004992918431211629,
"loss": 5.4703,
"mean_token_accuracy": 0.1695278689265251,
"num_tokens": 7369377.0,
"step": 4000
},
{
"entropy": 5.504941749572754,
"epoch": 0.3345305713331106,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992894810873183,
"loss": 5.4409,
"mean_token_accuracy": 0.16942341327667237,
"num_tokens": 7378814.0,
"step": 4005
},
{
"entropy": 5.4811779975891115,
"epoch": 0.33494821249582357,
"grad_norm": 1.1171875,
"learning_rate": 0.0004992871151270165,
"loss": 5.4364,
"mean_token_accuracy": 0.17195167988538743,
"num_tokens": 7387333.0,
"step": 4010
},
{
"entropy": 5.524077558517456,
"epoch": 0.3353658536585366,
"grad_norm": 0.96875,
"learning_rate": 0.0004992847452402986,
"loss": 5.476,
"mean_token_accuracy": 0.17165167927742003,
"num_tokens": 7396895.0,
"step": 4015
},
{
"entropy": 5.610746717453003,
"epoch": 0.3357834948212496,
"grad_norm": 1.1171875,
"learning_rate": 0.0004992823714272064,
"loss": 5.464,
"mean_token_accuracy": 0.17307177931070328,
"num_tokens": 7406430.0,
"step": 4020
},
{
"entropy": 5.4938287258148195,
"epoch": 0.33620113598396256,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992799936877813,
"loss": 5.3164,
"mean_token_accuracy": 0.17947741448879242,
"num_tokens": 7415224.0,
"step": 4025
},
{
"entropy": 5.556525421142578,
"epoch": 0.3366187771466756,
"grad_norm": 1.1953125,
"learning_rate": 0.0004992776120220649,
"loss": 5.6357,
"mean_token_accuracy": 0.15191856622695923,
"num_tokens": 7424250.0,
"step": 4030
},
{
"entropy": 5.5899182796478275,
"epoch": 0.3370364183093886,
"grad_norm": 1.1640625,
"learning_rate": 0.0004992752264300989,
"loss": 5.4769,
"mean_token_accuracy": 0.17585989683866501,
"num_tokens": 7433579.0,
"step": 4035
},
{
"entropy": 5.438165283203125,
"epoch": 0.33745405947210155,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992728369119252,
"loss": 5.4135,
"mean_token_accuracy": 0.17460322380065918,
"num_tokens": 7442621.0,
"step": 4040
},
{
"entropy": 5.605597305297851,
"epoch": 0.3378717006348146,
"grad_norm": 0.9765625,
"learning_rate": 0.0004992704434675855,
"loss": 5.5241,
"mean_token_accuracy": 0.16902994662523269,
"num_tokens": 7452077.0,
"step": 4045
},
{
"entropy": 5.458940124511718,
"epoch": 0.33828934179752757,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992680460971218,
"loss": 5.4524,
"mean_token_accuracy": 0.1666106477379799,
"num_tokens": 7461146.0,
"step": 4050
},
{
"entropy": 5.5238037109375,
"epoch": 0.33870698296024054,
"grad_norm": 1.0859375,
"learning_rate": 0.0004992656448005758,
"loss": 5.4314,
"mean_token_accuracy": 0.16991481184959412,
"num_tokens": 7470335.0,
"step": 4055
},
{
"entropy": 5.5821552753448485,
"epoch": 0.3391246241229536,
"grad_norm": 1.09375,
"learning_rate": 0.0004992632395779899,
"loss": 5.4328,
"mean_token_accuracy": 0.17112898975610732,
"num_tokens": 7479458.0,
"step": 4060
},
{
"entropy": 5.386154508590698,
"epoch": 0.33954226528566656,
"grad_norm": 1.046875,
"learning_rate": 0.0004992608304294061,
"loss": 5.3887,
"mean_token_accuracy": 0.1781470522284508,
"num_tokens": 7488005.0,
"step": 4065
},
{
"entropy": 5.44560980796814,
"epoch": 0.33995990644837953,
"grad_norm": 1.140625,
"learning_rate": 0.0004992584173548664,
"loss": 5.3879,
"mean_token_accuracy": 0.1756664276123047,
"num_tokens": 7496807.0,
"step": 4070
},
{
"entropy": 5.542823123931885,
"epoch": 0.34037754761109257,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992560003544131,
"loss": 5.4191,
"mean_token_accuracy": 0.17410115152597427,
"num_tokens": 7505383.0,
"step": 4075
},
{
"entropy": 5.58421835899353,
"epoch": 0.34079518877380555,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992535794280887,
"loss": 5.4816,
"mean_token_accuracy": 0.17036136388778686,
"num_tokens": 7515093.0,
"step": 4080
},
{
"entropy": 5.5146726131439205,
"epoch": 0.3412128299365185,
"grad_norm": 1.0625,
"learning_rate": 0.0004992511545759354,
"loss": 5.477,
"mean_token_accuracy": 0.1660609781742096,
"num_tokens": 7524067.0,
"step": 4085
},
{
"entropy": 5.444987487792969,
"epoch": 0.34163047109923156,
"grad_norm": 0.984375,
"learning_rate": 0.0004992487257979958,
"loss": 5.4073,
"mean_token_accuracy": 0.16477025300264359,
"num_tokens": 7533197.0,
"step": 4090
},
{
"entropy": 5.587007236480713,
"epoch": 0.34204811226194454,
"grad_norm": 1.078125,
"learning_rate": 0.0004992462930943123,
"loss": 5.4163,
"mean_token_accuracy": 0.1667389005422592,
"num_tokens": 7543030.0,
"step": 4095
},
{
"entropy": 5.480115938186645,
"epoch": 0.3424657534246575,
"grad_norm": 1.09375,
"learning_rate": 0.0004992438564649275,
"loss": 5.4692,
"mean_token_accuracy": 0.16808066219091417,
"num_tokens": 7551556.0,
"step": 4100
},
{
"entropy": 5.49069881439209,
"epoch": 0.34288339458737055,
"grad_norm": 1.125,
"learning_rate": 0.0004992414159098841,
"loss": 5.4075,
"mean_token_accuracy": 0.172200308740139,
"num_tokens": 7559284.0,
"step": 4105
},
{
"entropy": 5.606996965408325,
"epoch": 0.3433010357500835,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992389714292247,
"loss": 5.522,
"mean_token_accuracy": 0.17131585925817489,
"num_tokens": 7568750.0,
"step": 4110
},
{
"entropy": 5.417969465255737,
"epoch": 0.3437186769127965,
"grad_norm": 1.046875,
"learning_rate": 0.0004992365230229922,
"loss": 5.3881,
"mean_token_accuracy": 0.17342173904180527,
"num_tokens": 7578309.0,
"step": 4115
},
{
"entropy": 5.429501152038574,
"epoch": 0.34413631807550954,
"grad_norm": 1.0859375,
"learning_rate": 0.0004992340706912295,
"loss": 5.4034,
"mean_token_accuracy": 0.1707387700676918,
"num_tokens": 7586760.0,
"step": 4120
},
{
"entropy": 5.435112714767456,
"epoch": 0.3445539592382225,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992316144339794,
"loss": 5.3514,
"mean_token_accuracy": 0.17216803580522538,
"num_tokens": 7595472.0,
"step": 4125
},
{
"entropy": 5.537144660949707,
"epoch": 0.3449716004009355,
"grad_norm": 1.09375,
"learning_rate": 0.0004992291542512852,
"loss": 5.5529,
"mean_token_accuracy": 0.15992954969406128,
"num_tokens": 7605663.0,
"step": 4130
},
{
"entropy": 5.558428335189819,
"epoch": 0.34538924156364853,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992266901431895,
"loss": 5.4503,
"mean_token_accuracy": 0.1700645625591278,
"num_tokens": 7616110.0,
"step": 4135
},
{
"entropy": 5.448402690887451,
"epoch": 0.3458068827263615,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992242221097358,
"loss": 5.4086,
"mean_token_accuracy": 0.16535117626190185,
"num_tokens": 7625311.0,
"step": 4140
},
{
"entropy": 5.458443975448608,
"epoch": 0.3462245238890745,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992217501509672,
"loss": 5.4375,
"mean_token_accuracy": 0.16637078374624253,
"num_tokens": 7634562.0,
"step": 4145
},
{
"entropy": 5.504283428192139,
"epoch": 0.3466421650517875,
"grad_norm": 1.015625,
"learning_rate": 0.000499219274266927,
"loss": 5.3727,
"mean_token_accuracy": 0.16918073147535323,
"num_tokens": 7643231.0,
"step": 4150
},
{
"entropy": 5.541957664489746,
"epoch": 0.3470598062145005,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992167944576584,
"loss": 5.5044,
"mean_token_accuracy": 0.16804981082677842,
"num_tokens": 7653556.0,
"step": 4155
},
{
"entropy": 5.536978912353516,
"epoch": 0.3474774473772135,
"grad_norm": 1.0546875,
"learning_rate": 0.000499214310723205,
"loss": 5.4208,
"mean_token_accuracy": 0.17377680838108062,
"num_tokens": 7662295.0,
"step": 4160
},
{
"entropy": 5.3876220703125,
"epoch": 0.3478950885399265,
"grad_norm": 1.09375,
"learning_rate": 0.0004992118230636102,
"loss": 5.4348,
"mean_token_accuracy": 0.17047763615846634,
"num_tokens": 7671429.0,
"step": 4165
},
{
"entropy": 5.467368650436401,
"epoch": 0.3483127297026395,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992093314789176,
"loss": 5.3443,
"mean_token_accuracy": 0.16915934681892394,
"num_tokens": 7680589.0,
"step": 4170
},
{
"entropy": 5.486794424057007,
"epoch": 0.34873037086535247,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992068359691707,
"loss": 5.3973,
"mean_token_accuracy": 0.1751487910747528,
"num_tokens": 7689005.0,
"step": 4175
},
{
"entropy": 5.376707363128662,
"epoch": 0.3491480120280655,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992043365344132,
"loss": 5.323,
"mean_token_accuracy": 0.17639176547527313,
"num_tokens": 7698231.0,
"step": 4180
},
{
"entropy": 5.527692222595215,
"epoch": 0.3495656531907785,
"grad_norm": 1.03125,
"learning_rate": 0.0004992018331746891,
"loss": 5.4262,
"mean_token_accuracy": 0.16878757178783416,
"num_tokens": 7706678.0,
"step": 4185
},
{
"entropy": 5.557464504241944,
"epoch": 0.34998329435349146,
"grad_norm": 1.109375,
"learning_rate": 0.0004991993258900418,
"loss": 5.4804,
"mean_token_accuracy": 0.16915973871946335,
"num_tokens": 7715506.0,
"step": 4190
},
{
"entropy": 5.564558506011963,
"epoch": 0.3504009355162045,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991968146805156,
"loss": 5.4536,
"mean_token_accuracy": 0.1686076983809471,
"num_tokens": 7726549.0,
"step": 4195
},
{
"entropy": 5.505348777770996,
"epoch": 0.35081857667891747,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991942995461542,
"loss": 5.544,
"mean_token_accuracy": 0.16439460217952728,
"num_tokens": 7735563.0,
"step": 4200
},
{
"entropy": 5.556121873855591,
"epoch": 0.35123621784163045,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991917804870018,
"loss": 5.465,
"mean_token_accuracy": 0.17040056139230728,
"num_tokens": 7746144.0,
"step": 4205
},
{
"entropy": 5.569299364089966,
"epoch": 0.3516538590043435,
"grad_norm": 1.125,
"learning_rate": 0.0004991892575031024,
"loss": 5.4816,
"mean_token_accuracy": 0.17106465697288514,
"num_tokens": 7755394.0,
"step": 4210
},
{
"entropy": 5.375949335098267,
"epoch": 0.35207150016705646,
"grad_norm": 1.03125,
"learning_rate": 0.0004991867305945001,
"loss": 5.433,
"mean_token_accuracy": 0.17580727934837342,
"num_tokens": 7764855.0,
"step": 4215
},
{
"entropy": 5.438043546676636,
"epoch": 0.35248914132976944,
"grad_norm": 1.0625,
"learning_rate": 0.0004991841997612394,
"loss": 5.3575,
"mean_token_accuracy": 0.16997540295124053,
"num_tokens": 7773394.0,
"step": 4220
},
{
"entropy": 5.546854400634766,
"epoch": 0.3529067824924825,
"grad_norm": 1.09375,
"learning_rate": 0.0004991816650033643,
"loss": 5.4021,
"mean_token_accuracy": 0.17029064893722534,
"num_tokens": 7783111.0,
"step": 4225
},
{
"entropy": 5.56868200302124,
"epoch": 0.35332442365519545,
"grad_norm": 1.046875,
"learning_rate": 0.0004991791263209194,
"loss": 5.5515,
"mean_token_accuracy": 0.16735957562923431,
"num_tokens": 7792569.0,
"step": 4230
},
{
"entropy": 5.5392766952514645,
"epoch": 0.35374206481790843,
"grad_norm": 1.140625,
"learning_rate": 0.0004991765837139491,
"loss": 5.4424,
"mean_token_accuracy": 0.17671339213848114,
"num_tokens": 7801483.0,
"step": 4235
},
{
"entropy": 5.52734785079956,
"epoch": 0.35415970598062146,
"grad_norm": 0.9375,
"learning_rate": 0.0004991740371824977,
"loss": 5.5575,
"mean_token_accuracy": 0.16014997959136962,
"num_tokens": 7812274.0,
"step": 4240
},
{
"entropy": 5.49487271308899,
"epoch": 0.35457734714333444,
"grad_norm": 1.03125,
"learning_rate": 0.00049917148672661,
"loss": 5.381,
"mean_token_accuracy": 0.17165984213352203,
"num_tokens": 7821101.0,
"step": 4245
},
{
"entropy": 5.540127801895141,
"epoch": 0.3549949883060474,
"grad_norm": 1.0546875,
"learning_rate": 0.0004991689323463306,
"loss": 5.367,
"mean_token_accuracy": 0.17645271718502045,
"num_tokens": 7829805.0,
"step": 4250
},
{
"entropy": 5.476155138015747,
"epoch": 0.35541262946876045,
"grad_norm": 1.0,
"learning_rate": 0.0004991663740417042,
"loss": 5.4683,
"mean_token_accuracy": 0.16883495151996614,
"num_tokens": 7839772.0,
"step": 4255
},
{
"entropy": 5.475458765029908,
"epoch": 0.35583027063147343,
"grad_norm": 1.0546875,
"learning_rate": 0.0004991638118127757,
"loss": 5.5447,
"mean_token_accuracy": 0.1596977412700653,
"num_tokens": 7850290.0,
"step": 4260
},
{
"entropy": 5.6490367412567135,
"epoch": 0.3562479117941864,
"grad_norm": 1.09375,
"learning_rate": 0.0004991612456595896,
"loss": 5.5249,
"mean_token_accuracy": 0.16274113059043885,
"num_tokens": 7860180.0,
"step": 4265
},
{
"entropy": 5.38519434928894,
"epoch": 0.35666555295689945,
"grad_norm": 1.125,
"learning_rate": 0.0004991586755821913,
"loss": 5.3772,
"mean_token_accuracy": 0.18272387385368347,
"num_tokens": 7869625.0,
"step": 4270
},
{
"entropy": 5.384925889968872,
"epoch": 0.3570831941196124,
"grad_norm": 1.078125,
"learning_rate": 0.0004991561015806254,
"loss": 5.3308,
"mean_token_accuracy": 0.17186231166124344,
"num_tokens": 7878671.0,
"step": 4275
},
{
"entropy": 5.542608261108398,
"epoch": 0.3575008352823254,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991535236549372,
"loss": 5.4423,
"mean_token_accuracy": 0.17534592151641845,
"num_tokens": 7887225.0,
"step": 4280
},
{
"entropy": 5.417961120605469,
"epoch": 0.35791847644503844,
"grad_norm": 1.09375,
"learning_rate": 0.0004991509418051718,
"loss": 5.3389,
"mean_token_accuracy": 0.18268680572509766,
"num_tokens": 7895504.0,
"step": 4285
},
{
"entropy": 5.5606138706207275,
"epoch": 0.3583361176077514,
"grad_norm": 1.078125,
"learning_rate": 0.0004991483560313743,
"loss": 5.4095,
"mean_token_accuracy": 0.16898203194141387,
"num_tokens": 7904875.0,
"step": 4290
},
{
"entropy": 5.543120527267456,
"epoch": 0.3587537587704644,
"grad_norm": 0.97265625,
"learning_rate": 0.00049914576633359,
"loss": 5.4702,
"mean_token_accuracy": 0.16795135140419007,
"num_tokens": 7914747.0,
"step": 4295
},
{
"entropy": 5.499494552612305,
"epoch": 0.3591713999331774,
"grad_norm": 1.1328125,
"learning_rate": 0.0004991431727118643,
"loss": 5.4477,
"mean_token_accuracy": 0.16878911405801772,
"num_tokens": 7923858.0,
"step": 4300
},
{
"entropy": 5.480336284637451,
"epoch": 0.3595890410958904,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991405751662426,
"loss": 5.4103,
"mean_token_accuracy": 0.1708666056394577,
"num_tokens": 7932396.0,
"step": 4305
},
{
"entropy": 5.502483367919922,
"epoch": 0.3600066822586034,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991379736967703,
"loss": 5.4087,
"mean_token_accuracy": 0.17307820171117783,
"num_tokens": 7940763.0,
"step": 4310
},
{
"entropy": 5.463760662078857,
"epoch": 0.3604243234213164,
"grad_norm": 1.15625,
"learning_rate": 0.0004991353683034929,
"loss": 5.411,
"mean_token_accuracy": 0.1683210924267769,
"num_tokens": 7949424.0,
"step": 4315
},
{
"entropy": 5.552070283889771,
"epoch": 0.3608419645840294,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991327589864563,
"loss": 5.3825,
"mean_token_accuracy": 0.17842664420604706,
"num_tokens": 7958805.0,
"step": 4320
},
{
"entropy": 5.557480144500732,
"epoch": 0.3612596057467424,
"grad_norm": 1.1328125,
"learning_rate": 0.0004991301457457057,
"loss": 5.4905,
"mean_token_accuracy": 0.17496787160634994,
"num_tokens": 7967859.0,
"step": 4325
},
{
"entropy": 5.5827103614807125,
"epoch": 0.3616772469094554,
"grad_norm": 1.015625,
"learning_rate": 0.0004991275285812873,
"loss": 5.6177,
"mean_token_accuracy": 0.1633064940571785,
"num_tokens": 7978357.0,
"step": 4330
},
{
"entropy": 5.5931725025177,
"epoch": 0.3620948880721684,
"grad_norm": 1.1484375,
"learning_rate": 0.0004991249074932466,
"loss": 5.4658,
"mean_token_accuracy": 0.17505218535661698,
"num_tokens": 7986637.0,
"step": 4335
},
{
"entropy": 5.3996623992919925,
"epoch": 0.36251252923488136,
"grad_norm": 1.078125,
"learning_rate": 0.0004991222824816296,
"loss": 5.3885,
"mean_token_accuracy": 0.1762543112039566,
"num_tokens": 7995954.0,
"step": 4340
},
{
"entropy": 5.530343246459961,
"epoch": 0.3629301703975944,
"grad_norm": 1.1640625,
"learning_rate": 0.0004991196535464824,
"loss": 5.4574,
"mean_token_accuracy": 0.16591205894947053,
"num_tokens": 8005193.0,
"step": 4345
},
{
"entropy": 5.6107710838317875,
"epoch": 0.3633478115603074,
"grad_norm": 1.09375,
"learning_rate": 0.0004991170206878508,
"loss": 5.4877,
"mean_token_accuracy": 0.17001755386590958,
"num_tokens": 8014419.0,
"step": 4350
},
{
"entropy": 5.515951633453369,
"epoch": 0.36376545272302035,
"grad_norm": 1.03125,
"learning_rate": 0.000499114383905781,
"loss": 5.4214,
"mean_token_accuracy": 0.16749041378498078,
"num_tokens": 8024133.0,
"step": 4355
},
{
"entropy": 5.398339080810547,
"epoch": 0.3641830938857334,
"grad_norm": 1.03125,
"learning_rate": 0.0004991117432003192,
"loss": 5.3842,
"mean_token_accuracy": 0.1772811308503151,
"num_tokens": 8033059.0,
"step": 4360
},
{
"entropy": 5.566193628311157,
"epoch": 0.36460073504844637,
"grad_norm": 1.078125,
"learning_rate": 0.0004991090985715116,
"loss": 5.4566,
"mean_token_accuracy": 0.17722202092409134,
"num_tokens": 8041866.0,
"step": 4365
},
{
"entropy": 5.528580141067505,
"epoch": 0.36501837621115935,
"grad_norm": 1.09375,
"learning_rate": 0.0004991064500194043,
"loss": 5.4814,
"mean_token_accuracy": 0.16703143566846848,
"num_tokens": 8051541.0,
"step": 4370
},
{
"entropy": 5.619556570053101,
"epoch": 0.3654360173738724,
"grad_norm": 1.140625,
"learning_rate": 0.0004991037975440439,
"loss": 5.5565,
"mean_token_accuracy": 0.16549645811319352,
"num_tokens": 8061964.0,
"step": 4375
},
{
"entropy": 5.416561555862427,
"epoch": 0.36585365853658536,
"grad_norm": 1.140625,
"learning_rate": 0.0004991011411454769,
"loss": 5.3785,
"mean_token_accuracy": 0.17025724649429322,
"num_tokens": 8071006.0,
"step": 4380
},
{
"entropy": 5.4514140605926515,
"epoch": 0.36627129969929834,
"grad_norm": 1.109375,
"learning_rate": 0.0004990984808237495,
"loss": 5.3757,
"mean_token_accuracy": 0.17259391695261,
"num_tokens": 8080104.0,
"step": 4385
},
{
"entropy": 5.575825357437134,
"epoch": 0.36668894086201137,
"grad_norm": 1.03125,
"learning_rate": 0.0004990958165789086,
"loss": 5.4242,
"mean_token_accuracy": 0.1789444699883461,
"num_tokens": 8089397.0,
"step": 4390
},
{
"entropy": 5.476417112350464,
"epoch": 0.36710658202472435,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990931484110005,
"loss": 5.4537,
"mean_token_accuracy": 0.1721838116645813,
"num_tokens": 8099055.0,
"step": 4395
},
{
"entropy": 5.479823112487793,
"epoch": 0.3675242231874373,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990904763200722,
"loss": 5.3974,
"mean_token_accuracy": 0.17528146207332612,
"num_tokens": 8107609.0,
"step": 4400
},
{
"entropy": 5.551629161834716,
"epoch": 0.36794186435015036,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990878003061705,
"loss": 5.4218,
"mean_token_accuracy": 0.17829948663711548,
"num_tokens": 8116905.0,
"step": 4405
},
{
"entropy": 5.438847637176513,
"epoch": 0.36835950551286334,
"grad_norm": 1.140625,
"learning_rate": 0.0004990851203693419,
"loss": 5.3899,
"mean_token_accuracy": 0.17322081178426743,
"num_tokens": 8125580.0,
"step": 4410
},
{
"entropy": 5.396791076660156,
"epoch": 0.3687771466755763,
"grad_norm": 1.03125,
"learning_rate": 0.0004990824365096336,
"loss": 5.322,
"mean_token_accuracy": 0.17096327543258666,
"num_tokens": 8134568.0,
"step": 4415
},
{
"entropy": 5.369096040725708,
"epoch": 0.36919478783828935,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990797487270926,
"loss": 5.3196,
"mean_token_accuracy": 0.18015577644109726,
"num_tokens": 8143044.0,
"step": 4420
},
{
"entropy": 5.49015703201294,
"epoch": 0.36961242900100233,
"grad_norm": 1.1328125,
"learning_rate": 0.0004990770570217659,
"loss": 5.4231,
"mean_token_accuracy": 0.1667809158563614,
"num_tokens": 8153701.0,
"step": 4425
},
{
"entropy": 5.449046087265015,
"epoch": 0.37003007016371536,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990743613937005,
"loss": 5.3311,
"mean_token_accuracy": 0.17967470586299897,
"num_tokens": 8163264.0,
"step": 4430
},
{
"entropy": 5.439912414550781,
"epoch": 0.37044771132642834,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990716618429436,
"loss": 5.4401,
"mean_token_accuracy": 0.16971767246723174,
"num_tokens": 8173332.0,
"step": 4435
},
{
"entropy": 5.568769931793213,
"epoch": 0.3708653524891413,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990689583695428,
"loss": 5.4332,
"mean_token_accuracy": 0.16823667883872986,
"num_tokens": 8181881.0,
"step": 4440
},
{
"entropy": 5.517188310623169,
"epoch": 0.37128299365185435,
"grad_norm": 1.1328125,
"learning_rate": 0.000499066250973545,
"loss": 5.4702,
"mean_token_accuracy": 0.1688990131020546,
"num_tokens": 8190771.0,
"step": 4445
},
{
"entropy": 5.4473552227020265,
"epoch": 0.37170063481456733,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990635396549978,
"loss": 5.4019,
"mean_token_accuracy": 0.16778893321752547,
"num_tokens": 8200740.0,
"step": 4450
},
{
"entropy": 5.453600025177002,
"epoch": 0.3721182759772803,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990608244139486,
"loss": 5.3895,
"mean_token_accuracy": 0.17340001314878464,
"num_tokens": 8209935.0,
"step": 4455
},
{
"entropy": 5.4877028465271,
"epoch": 0.37253591713999334,
"grad_norm": 1.0625,
"learning_rate": 0.000499058105250445,
"loss": 5.471,
"mean_token_accuracy": 0.16836313307285308,
"num_tokens": 8219137.0,
"step": 4460
},
{
"entropy": 5.586174011230469,
"epoch": 0.3729535583027063,
"grad_norm": 1.0,
"learning_rate": 0.0004990553821645346,
"loss": 5.4352,
"mean_token_accuracy": 0.17806873768568038,
"num_tokens": 8228076.0,
"step": 4465
},
{
"entropy": 5.4609466075897215,
"epoch": 0.3733711994654193,
"grad_norm": 1.0390625,
"learning_rate": 0.000499052655156265,
"loss": 5.3966,
"mean_token_accuracy": 0.17221251577138902,
"num_tokens": 8237581.0,
"step": 4470
},
{
"entropy": 5.519050741195679,
"epoch": 0.37378884062813234,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990499242256839,
"loss": 5.4422,
"mean_token_accuracy": 0.16933617889881133,
"num_tokens": 8247384.0,
"step": 4475
},
{
"entropy": 5.552171897888184,
"epoch": 0.3742064817908453,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990471893728394,
"loss": 5.5096,
"mean_token_accuracy": 0.16877915859222412,
"num_tokens": 8256383.0,
"step": 4480
},
{
"entropy": 5.492855978012085,
"epoch": 0.3746241229535583,
"grad_norm": 1.1953125,
"learning_rate": 0.000499044450597779,
"loss": 5.3645,
"mean_token_accuracy": 0.17729802876710893,
"num_tokens": 8264848.0,
"step": 4485
},
{
"entropy": 5.472712516784668,
"epoch": 0.3750417641162713,
"grad_norm": 1.0703125,
"learning_rate": 0.0004990417079005508,
"loss": 5.3991,
"mean_token_accuracy": 0.17253781408071517,
"num_tokens": 8274403.0,
"step": 4490
},
{
"entropy": 5.432583808898926,
"epoch": 0.3754594052789843,
"grad_norm": 1.1484375,
"learning_rate": 0.000499038961281203,
"loss": 5.3301,
"mean_token_accuracy": 0.17931564450263976,
"num_tokens": 8282700.0,
"step": 4495
},
{
"entropy": 5.424058294296264,
"epoch": 0.3758770464416973,
"grad_norm": 1.140625,
"learning_rate": 0.0004990362107397834,
"loss": 5.3543,
"mean_token_accuracy": 0.18285386115312577,
"num_tokens": 8291718.0,
"step": 4500
},
{
"entropy": 5.458253526687622,
"epoch": 0.3762946876044103,
"grad_norm": 1.0625,
"learning_rate": 0.0004990334562763401,
"loss": 5.4142,
"mean_token_accuracy": 0.16921038329601287,
"num_tokens": 8301113.0,
"step": 4505
},
{
"entropy": 5.572368526458741,
"epoch": 0.3767123287671233,
"grad_norm": 1.1328125,
"learning_rate": 0.0004990306978909216,
"loss": 5.5148,
"mean_token_accuracy": 0.17073766738176346,
"num_tokens": 8310357.0,
"step": 4510
},
{
"entropy": 5.561983442306518,
"epoch": 0.3771299699298363,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990279355835761,
"loss": 5.4489,
"mean_token_accuracy": 0.17559705078601837,
"num_tokens": 8319680.0,
"step": 4515
},
{
"entropy": 5.52631139755249,
"epoch": 0.3775476110925493,
"grad_norm": 1.09375,
"learning_rate": 0.0004990251693543519,
"loss": 5.4536,
"mean_token_accuracy": 0.16949882954359055,
"num_tokens": 8329127.0,
"step": 4520
},
{
"entropy": 5.382966136932373,
"epoch": 0.3779652522552623,
"grad_norm": 1.078125,
"learning_rate": 0.0004990223992032975,
"loss": 5.3037,
"mean_token_accuracy": 0.18054991960525513,
"num_tokens": 8338308.0,
"step": 4525
},
{
"entropy": 5.454804182052612,
"epoch": 0.37838289341797526,
"grad_norm": 1.1640625,
"learning_rate": 0.0004990196251304612,
"loss": 5.3412,
"mean_token_accuracy": 0.17313064336776735,
"num_tokens": 8347491.0,
"step": 4530
},
{
"entropy": 5.341402626037597,
"epoch": 0.3788005345806883,
"grad_norm": 1.0625,
"learning_rate": 0.000499016847135892,
"loss": 5.3429,
"mean_token_accuracy": 0.1704577088356018,
"num_tokens": 8357112.0,
"step": 4535
},
{
"entropy": 5.521670484542847,
"epoch": 0.3792181757434013,
"grad_norm": 1.015625,
"learning_rate": 0.000499014065219638,
"loss": 5.3607,
"mean_token_accuracy": 0.17628980427980423,
"num_tokens": 8366285.0,
"step": 4540
},
{
"entropy": 5.438895797729492,
"epoch": 0.37963581690611425,
"grad_norm": 1.1875,
"learning_rate": 0.0004990112793817481,
"loss": 5.2526,
"mean_token_accuracy": 0.1850603237748146,
"num_tokens": 8374362.0,
"step": 4545
},
{
"entropy": 5.407459306716919,
"epoch": 0.3800534580688273,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990084896222713,
"loss": 5.4892,
"mean_token_accuracy": 0.17154589891433716,
"num_tokens": 8383297.0,
"step": 4550
},
{
"entropy": 5.5045722961425785,
"epoch": 0.38047109923154027,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990056959412562,
"loss": 5.387,
"mean_token_accuracy": 0.17965543121099473,
"num_tokens": 8391926.0,
"step": 4555
},
{
"entropy": 5.535581254959107,
"epoch": 0.38088874039425324,
"grad_norm": 1.1328125,
"learning_rate": 0.0004990028983387517,
"loss": 5.3559,
"mean_token_accuracy": 0.1749042421579361,
"num_tokens": 8401226.0,
"step": 4560
},
{
"entropy": 5.455212688446045,
"epoch": 0.3813063815569663,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990000968148068,
"loss": 5.394,
"mean_token_accuracy": 0.1762553095817566,
"num_tokens": 8409966.0,
"step": 4565
},
{
"entropy": 5.386058282852173,
"epoch": 0.38172402271967926,
"grad_norm": 1.0625,
"learning_rate": 0.0004989972913694707,
"loss": 5.2452,
"mean_token_accuracy": 0.18678999394178392,
"num_tokens": 8418281.0,
"step": 4570
},
{
"entropy": 5.34035096168518,
"epoch": 0.38214166388239224,
"grad_norm": 1.046875,
"learning_rate": 0.0004989944820027923,
"loss": 5.3565,
"mean_token_accuracy": 0.17558874040842057,
"num_tokens": 8427436.0,
"step": 4575
},
{
"entropy": 5.478307771682739,
"epoch": 0.38255930504510527,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989916687148209,
"loss": 5.3986,
"mean_token_accuracy": 0.1782542034983635,
"num_tokens": 8436066.0,
"step": 4580
},
{
"entropy": 5.454951810836792,
"epoch": 0.38297694620781825,
"grad_norm": 1.109375,
"learning_rate": 0.0004989888515056057,
"loss": 5.3924,
"mean_token_accuracy": 0.17499198764562607,
"num_tokens": 8445572.0,
"step": 4585
},
{
"entropy": 5.558172369003296,
"epoch": 0.3833945873705312,
"grad_norm": 1.0546875,
"learning_rate": 0.000498986030375196,
"loss": 5.4444,
"mean_token_accuracy": 0.171583916246891,
"num_tokens": 8454781.0,
"step": 4590
},
{
"entropy": 5.50851674079895,
"epoch": 0.38381222853324426,
"grad_norm": 1.109375,
"learning_rate": 0.0004989832053236413,
"loss": 5.5158,
"mean_token_accuracy": 0.16855873316526412,
"num_tokens": 8463239.0,
"step": 4595
},
{
"entropy": 5.510396289825439,
"epoch": 0.38422986969595724,
"grad_norm": 1.1484375,
"learning_rate": 0.000498980376350991,
"loss": 5.4084,
"mean_token_accuracy": 0.1754463106393814,
"num_tokens": 8472664.0,
"step": 4600
},
{
"entropy": 5.504710292816162,
"epoch": 0.3846475108586702,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989775434572945,
"loss": 5.47,
"mean_token_accuracy": 0.16816986054182054,
"num_tokens": 8483149.0,
"step": 4605
},
{
"entropy": 5.582529878616333,
"epoch": 0.38506515202138325,
"grad_norm": 1.140625,
"learning_rate": 0.0004989747066426015,
"loss": 5.4105,
"mean_token_accuracy": 0.17679672837257385,
"num_tokens": 8492297.0,
"step": 4610
},
{
"entropy": 5.401932001113892,
"epoch": 0.38548279318409623,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989718659069618,
"loss": 5.4906,
"mean_token_accuracy": 0.16797342747449875,
"num_tokens": 8502092.0,
"step": 4615
},
{
"entropy": 5.54914174079895,
"epoch": 0.3859004343468092,
"grad_norm": 1.0546875,
"learning_rate": 0.000498969021250425,
"loss": 5.3977,
"mean_token_accuracy": 0.17605937272310257,
"num_tokens": 8510816.0,
"step": 4620
},
{
"entropy": 5.546402645111084,
"epoch": 0.38631807550952224,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989661726730407,
"loss": 5.4126,
"mean_token_accuracy": 0.17610876709222795,
"num_tokens": 8520055.0,
"step": 4625
},
{
"entropy": 5.488043260574341,
"epoch": 0.3867357166722352,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989633201748591,
"loss": 5.4087,
"mean_token_accuracy": 0.17527229338884354,
"num_tokens": 8528271.0,
"step": 4630
},
{
"entropy": 5.422880077362061,
"epoch": 0.3871533578349482,
"grad_norm": 1.171875,
"learning_rate": 0.0004989604637559299,
"loss": 5.3117,
"mean_token_accuracy": 0.1829973429441452,
"num_tokens": 8537360.0,
"step": 4635
},
{
"entropy": 5.318050050735474,
"epoch": 0.38757099899766123,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989576034163033,
"loss": 5.2864,
"mean_token_accuracy": 0.18681524097919464,
"num_tokens": 8546574.0,
"step": 4640
},
{
"entropy": 5.426437568664551,
"epoch": 0.3879886401603742,
"grad_norm": 1.0625,
"learning_rate": 0.0004989547391560293,
"loss": 5.3921,
"mean_token_accuracy": 0.17277875542640686,
"num_tokens": 8555581.0,
"step": 4645
},
{
"entropy": 5.473151540756225,
"epoch": 0.3884062813230872,
"grad_norm": 0.984375,
"learning_rate": 0.000498951870975158,
"loss": 5.4301,
"mean_token_accuracy": 0.17651775926351548,
"num_tokens": 8564539.0,
"step": 4650
},
{
"entropy": 5.522072887420654,
"epoch": 0.3888239224858002,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989489988737395,
"loss": 5.4537,
"mean_token_accuracy": 0.17094668000936508,
"num_tokens": 8573804.0,
"step": 4655
},
{
"entropy": 5.454622507095337,
"epoch": 0.3892415636485132,
"grad_norm": 1.09375,
"learning_rate": 0.0004989461228518243,
"loss": 5.3791,
"mean_token_accuracy": 0.1732862874865532,
"num_tokens": 8583262.0,
"step": 4660
},
{
"entropy": 5.444647312164307,
"epoch": 0.3896592048112262,
"grad_norm": 1.1484375,
"learning_rate": 0.0004989432429094627,
"loss": 5.4554,
"mean_token_accuracy": 0.1736641988158226,
"num_tokens": 8592763.0,
"step": 4665
},
{
"entropy": 5.5642941951751705,
"epoch": 0.3900768459739392,
"grad_norm": 1.1484375,
"learning_rate": 0.0004989403590467049,
"loss": 5.3178,
"mean_token_accuracy": 0.17884466499090196,
"num_tokens": 8601884.0,
"step": 4670
},
{
"entropy": 5.434857130050659,
"epoch": 0.3904944871366522,
"grad_norm": 1.140625,
"learning_rate": 0.0004989374712636017,
"loss": 5.4648,
"mean_token_accuracy": 0.16872687935829161,
"num_tokens": 8611207.0,
"step": 4675
},
{
"entropy": 5.355410432815551,
"epoch": 0.39091212829936517,
"grad_norm": 1.09375,
"learning_rate": 0.0004989345795602036,
"loss": 5.1821,
"mean_token_accuracy": 0.19296537935733796,
"num_tokens": 8620104.0,
"step": 4680
},
{
"entropy": 5.489849519729614,
"epoch": 0.3913297694620782,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989316839365609,
"loss": 5.4073,
"mean_token_accuracy": 0.17335034757852555,
"num_tokens": 8629552.0,
"step": 4685
},
{
"entropy": 5.439398717880249,
"epoch": 0.3917474106247912,
"grad_norm": 1.28125,
"learning_rate": 0.0004989287843927246,
"loss": 5.3277,
"mean_token_accuracy": 0.17526139616966246,
"num_tokens": 8638961.0,
"step": 4690
},
{
"entropy": 5.398167371749878,
"epoch": 0.39216505178750416,
"grad_norm": 1.125,
"learning_rate": 0.0004989258809287455,
"loss": 5.3624,
"mean_token_accuracy": 0.1763260081410408,
"num_tokens": 8648051.0,
"step": 4695
},
{
"entropy": 5.485393619537353,
"epoch": 0.3925826929502172,
"grad_norm": 1.1171875,
"learning_rate": 0.0004989229735446743,
"loss": 5.3692,
"mean_token_accuracy": 0.17140538841485978,
"num_tokens": 8656863.0,
"step": 4700
},
{
"entropy": 5.3344724655151365,
"epoch": 0.39300033411293017,
"grad_norm": 1.1796875,
"learning_rate": 0.0004989200622405619,
"loss": 5.1929,
"mean_token_accuracy": 0.18634516447782518,
"num_tokens": 8665455.0,
"step": 4705
},
{
"entropy": 5.338350009918213,
"epoch": 0.39341797527564315,
"grad_norm": 1.09375,
"learning_rate": 0.0004989171470164592,
"loss": 5.3068,
"mean_token_accuracy": 0.17981014400720596,
"num_tokens": 8674119.0,
"step": 4710
},
{
"entropy": 5.464507055282593,
"epoch": 0.3938356164383562,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989142278724173,
"loss": 5.4137,
"mean_token_accuracy": 0.17251340299844742,
"num_tokens": 8682979.0,
"step": 4715
},
{
"entropy": 5.472417497634888,
"epoch": 0.39425325760106916,
"grad_norm": 1.1328125,
"learning_rate": 0.0004989113048084874,
"loss": 5.3048,
"mean_token_accuracy": 0.16820131987333298,
"num_tokens": 8692392.0,
"step": 4720
},
{
"entropy": 5.393416547775269,
"epoch": 0.39467089876378214,
"grad_norm": 1.125,
"learning_rate": 0.0004989083778247206,
"loss": 5.3766,
"mean_token_accuracy": 0.17461734712123872,
"num_tokens": 8700932.0,
"step": 4725
},
{
"entropy": 5.5903651237487795,
"epoch": 0.3950885399264952,
"grad_norm": 1.1171875,
"learning_rate": 0.000498905446921168,
"loss": 5.5322,
"mean_token_accuracy": 0.1624947041273117,
"num_tokens": 8710683.0,
"step": 4730
},
{
"entropy": 5.3998308181762695,
"epoch": 0.39550618108920815,
"grad_norm": 1.078125,
"learning_rate": 0.0004989025120978812,
"loss": 5.362,
"mean_token_accuracy": 0.17818318009376527,
"num_tokens": 8719939.0,
"step": 4735
},
{
"entropy": 5.559629631042481,
"epoch": 0.39592382225192113,
"grad_norm": 1.0859375,
"learning_rate": 0.0004988995733549113,
"loss": 5.4132,
"mean_token_accuracy": 0.17718918472528458,
"num_tokens": 8729275.0,
"step": 4740
},
{
"entropy": 5.364532518386841,
"epoch": 0.39634146341463417,
"grad_norm": 1.125,
"learning_rate": 0.0004988966306923099,
"loss": 5.2911,
"mean_token_accuracy": 0.183379165828228,
"num_tokens": 8738048.0,
"step": 4745
},
{
"entropy": 5.417474317550659,
"epoch": 0.39675910457734714,
"grad_norm": 1.1875,
"learning_rate": 0.0004988936841101284,
"loss": 5.3583,
"mean_token_accuracy": 0.17376182675361634,
"num_tokens": 8746967.0,
"step": 4750
},
{
"entropy": 5.413084220886231,
"epoch": 0.3971767457400601,
"grad_norm": 1.3359375,
"learning_rate": 0.0004988907336084186,
"loss": 5.3121,
"mean_token_accuracy": 0.17945242077112197,
"num_tokens": 8754908.0,
"step": 4755
},
{
"entropy": 5.43807692527771,
"epoch": 0.39759438690277316,
"grad_norm": 1.03125,
"learning_rate": 0.0004988877791872319,
"loss": 5.3396,
"mean_token_accuracy": 0.18502106070518493,
"num_tokens": 8764489.0,
"step": 4760
},
{
"entropy": 5.603342771530151,
"epoch": 0.39801202806548613,
"grad_norm": 1.15625,
"learning_rate": 0.0004988848208466202,
"loss": 5.5068,
"mean_token_accuracy": 0.15701909065246583,
"num_tokens": 8774365.0,
"step": 4765
},
{
"entropy": 5.460353755950928,
"epoch": 0.3984296692281991,
"grad_norm": 1.1875,
"learning_rate": 0.0004988818585866351,
"loss": 5.3104,
"mean_token_accuracy": 0.17841035425662993,
"num_tokens": 8783538.0,
"step": 4770
},
{
"entropy": 5.364460277557373,
"epoch": 0.39884731039091215,
"grad_norm": 1.1640625,
"learning_rate": 0.0004988788924073288,
"loss": 5.4081,
"mean_token_accuracy": 0.17358129620552062,
"num_tokens": 8792171.0,
"step": 4775
},
{
"entropy": 5.486460971832275,
"epoch": 0.3992649515536251,
"grad_norm": 1.09375,
"learning_rate": 0.0004988759223087528,
"loss": 5.4263,
"mean_token_accuracy": 0.17069908380508422,
"num_tokens": 8801325.0,
"step": 4780
},
{
"entropy": 5.507883977890015,
"epoch": 0.3996825927163381,
"grad_norm": 1.046875,
"learning_rate": 0.0004988729482909595,
"loss": 5.3494,
"mean_token_accuracy": 0.17493296712636947,
"num_tokens": 8810566.0,
"step": 4785
},
{
"entropy": 5.370326042175293,
"epoch": 0.40010023387905114,
"grad_norm": 1.0,
"learning_rate": 0.0004988699703540006,
"loss": 5.2915,
"mean_token_accuracy": 0.18314679563045502,
"num_tokens": 8819735.0,
"step": 4790
},
{
"entropy": 5.405260324478149,
"epoch": 0.4005178750417641,
"grad_norm": 1.0625,
"learning_rate": 0.0004988669884979284,
"loss": 5.432,
"mean_token_accuracy": 0.17118214964866638,
"num_tokens": 8828820.0,
"step": 4795
},
{
"entropy": 5.5137152671813965,
"epoch": 0.4009355162044771,
"grad_norm": 1.046875,
"learning_rate": 0.0004988640027227951,
"loss": 5.3147,
"mean_token_accuracy": 0.17574127465486528,
"num_tokens": 8837858.0,
"step": 4800
},
{
"entropy": 5.440872716903686,
"epoch": 0.4013531573671901,
"grad_norm": 1.015625,
"learning_rate": 0.000498861013028653,
"loss": 5.2989,
"mean_token_accuracy": 0.17621319442987443,
"num_tokens": 8847172.0,
"step": 4805
},
{
"entropy": 5.434757995605469,
"epoch": 0.4017707985299031,
"grad_norm": 1.03125,
"learning_rate": 0.0004988580194155544,
"loss": 5.3119,
"mean_token_accuracy": 0.17747572362422942,
"num_tokens": 8856700.0,
"step": 4810
},
{
"entropy": 5.306950616836548,
"epoch": 0.4021884396926161,
"grad_norm": 1.109375,
"learning_rate": 0.0004988550218835517,
"loss": 5.2215,
"mean_token_accuracy": 0.18293379694223405,
"num_tokens": 8866108.0,
"step": 4815
},
{
"entropy": 5.347055053710937,
"epoch": 0.4026060808553291,
"grad_norm": 1.078125,
"learning_rate": 0.0004988520204326975,
"loss": 5.3244,
"mean_token_accuracy": 0.1850435584783554,
"num_tokens": 8874833.0,
"step": 4820
},
{
"entropy": 5.511721134185791,
"epoch": 0.4030237220180421,
"grad_norm": 1.078125,
"learning_rate": 0.0004988490150630441,
"loss": 5.389,
"mean_token_accuracy": 0.18025784492492675,
"num_tokens": 8884211.0,
"step": 4825
},
{
"entropy": 5.546769905090332,
"epoch": 0.4034413631807551,
"grad_norm": 1.1171875,
"learning_rate": 0.0004988460057746443,
"loss": 5.4906,
"mean_token_accuracy": 0.17315116077661513,
"num_tokens": 8893443.0,
"step": 4830
},
{
"entropy": 5.399085950851441,
"epoch": 0.4038590043434681,
"grad_norm": 1.0390625,
"learning_rate": 0.0004988429925675507,
"loss": 5.3561,
"mean_token_accuracy": 0.1843828722834587,
"num_tokens": 8903544.0,
"step": 4835
},
{
"entropy": 5.510149908065796,
"epoch": 0.4042766455061811,
"grad_norm": 1.125,
"learning_rate": 0.0004988399754418161,
"loss": 5.3695,
"mean_token_accuracy": 0.18004968166351318,
"num_tokens": 8912270.0,
"step": 4840
},
{
"entropy": 5.542214012145996,
"epoch": 0.40469428666889407,
"grad_norm": 1.015625,
"learning_rate": 0.0004988369543974932,
"loss": 5.4706,
"mean_token_accuracy": 0.16505088955163955,
"num_tokens": 8922155.0,
"step": 4845
},
{
"entropy": 5.466346979141235,
"epoch": 0.4051119278316071,
"grad_norm": 1.0546875,
"learning_rate": 0.0004988339294346351,
"loss": 5.4701,
"mean_token_accuracy": 0.17315089404582978,
"num_tokens": 8931844.0,
"step": 4850
},
{
"entropy": 5.517349576950073,
"epoch": 0.4055295689943201,
"grad_norm": 1.0703125,
"learning_rate": 0.0004988309005532946,
"loss": 5.3173,
"mean_token_accuracy": 0.18095384240150453,
"num_tokens": 8940874.0,
"step": 4855
},
{
"entropy": 5.416971731185913,
"epoch": 0.40594721015703306,
"grad_norm": 1.140625,
"learning_rate": 0.0004988278677535247,
"loss": 5.3917,
"mean_token_accuracy": 0.18220391720533372,
"num_tokens": 8950108.0,
"step": 4860
},
{
"entropy": 5.492914772033691,
"epoch": 0.4063648513197461,
"grad_norm": 1.0625,
"learning_rate": 0.0004988248310353788,
"loss": 5.4044,
"mean_token_accuracy": 0.16736428141593934,
"num_tokens": 8960610.0,
"step": 4865
},
{
"entropy": 5.37611083984375,
"epoch": 0.40678249248245907,
"grad_norm": 1.2265625,
"learning_rate": 0.0004988217903989095,
"loss": 5.3359,
"mean_token_accuracy": 0.17368675768375397,
"num_tokens": 8969853.0,
"step": 4870
},
{
"entropy": 5.4439164161682125,
"epoch": 0.40720013364517205,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988187458441704,
"loss": 5.324,
"mean_token_accuracy": 0.17724681198596953,
"num_tokens": 8979800.0,
"step": 4875
},
{
"entropy": 5.5503577709198,
"epoch": 0.4076177748078851,
"grad_norm": 1.203125,
"learning_rate": 0.0004988156973712149,
"loss": 5.3856,
"mean_token_accuracy": 0.1728046178817749,
"num_tokens": 8988667.0,
"step": 4880
},
{
"entropy": 5.364498662948608,
"epoch": 0.40803541597059806,
"grad_norm": 1.15625,
"learning_rate": 0.0004988126449800961,
"loss": 5.2687,
"mean_token_accuracy": 0.1794217899441719,
"num_tokens": 8997385.0,
"step": 4885
},
{
"entropy": 5.403355121612549,
"epoch": 0.40845305713331104,
"grad_norm": 1.1953125,
"learning_rate": 0.0004988095886708676,
"loss": 5.2799,
"mean_token_accuracy": 0.18214777261018752,
"num_tokens": 9005999.0,
"step": 4890
},
{
"entropy": 5.437708711624145,
"epoch": 0.40887069829602407,
"grad_norm": 1.1484375,
"learning_rate": 0.0004988065284435828,
"loss": 5.4167,
"mean_token_accuracy": 0.1787819132208824,
"num_tokens": 9014837.0,
"step": 4895
},
{
"entropy": 5.547063875198364,
"epoch": 0.40928833945873705,
"grad_norm": 1.09375,
"learning_rate": 0.0004988034642982954,
"loss": 5.4656,
"mean_token_accuracy": 0.1768844336271286,
"num_tokens": 9023956.0,
"step": 4900
},
{
"entropy": 5.538508605957031,
"epoch": 0.40970598062145,
"grad_norm": 1.03125,
"learning_rate": 0.0004988003962350589,
"loss": 5.4103,
"mean_token_accuracy": 0.1739673912525177,
"num_tokens": 9032945.0,
"step": 4905
},
{
"entropy": 5.448762273788452,
"epoch": 0.41012362178416306,
"grad_norm": 1.0859375,
"learning_rate": 0.000498797324253927,
"loss": 5.3368,
"mean_token_accuracy": 0.17977107018232347,
"num_tokens": 9042109.0,
"step": 4910
},
{
"entropy": 5.322491264343261,
"epoch": 0.41054126294687604,
"grad_norm": 1.140625,
"learning_rate": 0.0004987942483549537,
"loss": 5.3225,
"mean_token_accuracy": 0.1740400969982147,
"num_tokens": 9052210.0,
"step": 4915
},
{
"entropy": 5.456244850158692,
"epoch": 0.410958904109589,
"grad_norm": 1.1875,
"learning_rate": 0.0004987911685381926,
"loss": 5.2955,
"mean_token_accuracy": 0.1807815983891487,
"num_tokens": 9060089.0,
"step": 4920
},
{
"entropy": 5.554550838470459,
"epoch": 0.41137654527230205,
"grad_norm": 1.109375,
"learning_rate": 0.0004987880848036977,
"loss": 5.3934,
"mean_token_accuracy": 0.17276457697153091,
"num_tokens": 9069235.0,
"step": 4925
},
{
"entropy": 5.416343069076538,
"epoch": 0.41179418643501503,
"grad_norm": 1.140625,
"learning_rate": 0.0004987849971515232,
"loss": 5.3597,
"mean_token_accuracy": 0.17957380264997483,
"num_tokens": 9077509.0,
"step": 4930
},
{
"entropy": 5.449953413009643,
"epoch": 0.412211827597728,
"grad_norm": 1.109375,
"learning_rate": 0.0004987819055817227,
"loss": 5.2672,
"mean_token_accuracy": 0.1885462522506714,
"num_tokens": 9085493.0,
"step": 4935
},
{
"entropy": 5.2786545753479,
"epoch": 0.41262946876044104,
"grad_norm": 1.046875,
"learning_rate": 0.0004987788100943507,
"loss": 5.2434,
"mean_token_accuracy": 0.17916715145111084,
"num_tokens": 9094748.0,
"step": 4940
},
{
"entropy": 5.398625183105469,
"epoch": 0.413047109923154,
"grad_norm": 1.09375,
"learning_rate": 0.0004987757106894613,
"loss": 5.3082,
"mean_token_accuracy": 0.18082510977983474,
"num_tokens": 9103678.0,
"step": 4945
},
{
"entropy": 5.460435628890991,
"epoch": 0.413464751085867,
"grad_norm": 1.1953125,
"learning_rate": 0.0004987726073671086,
"loss": 5.373,
"mean_token_accuracy": 0.17700842022895813,
"num_tokens": 9112851.0,
"step": 4950
},
{
"entropy": 5.534562349319458,
"epoch": 0.41388239224858003,
"grad_norm": 0.96484375,
"learning_rate": 0.0004987695001273471,
"loss": 5.5833,
"mean_token_accuracy": 0.15996475592255593,
"num_tokens": 9122106.0,
"step": 4955
},
{
"entropy": 5.446104288101196,
"epoch": 0.414300033411293,
"grad_norm": 1.0234375,
"learning_rate": 0.0004987663889702313,
"loss": 5.2942,
"mean_token_accuracy": 0.18406701385974883,
"num_tokens": 9131580.0,
"step": 4960
},
{
"entropy": 5.494612503051758,
"epoch": 0.414717674574006,
"grad_norm": 1.0859375,
"learning_rate": 0.0004987632738958155,
"loss": 5.4341,
"mean_token_accuracy": 0.17709944397211075,
"num_tokens": 9141642.0,
"step": 4965
},
{
"entropy": 5.6023462295532225,
"epoch": 0.415135315736719,
"grad_norm": 1.046875,
"learning_rate": 0.0004987601549041541,
"loss": 5.4765,
"mean_token_accuracy": 0.16715555340051652,
"num_tokens": 9151192.0,
"step": 4970
},
{
"entropy": 5.439905166625977,
"epoch": 0.415552956899432,
"grad_norm": 1.203125,
"learning_rate": 0.0004987570319953018,
"loss": 5.3728,
"mean_token_accuracy": 0.1760926589369774,
"num_tokens": 9161316.0,
"step": 4975
},
{
"entropy": 5.491254663467407,
"epoch": 0.415970598062145,
"grad_norm": 1.078125,
"learning_rate": 0.0004987539051693134,
"loss": 5.3794,
"mean_token_accuracy": 0.17287528663873672,
"num_tokens": 9169535.0,
"step": 4980
},
{
"entropy": 5.487295722961425,
"epoch": 0.416388239224858,
"grad_norm": 1.203125,
"learning_rate": 0.0004987507744262436,
"loss": 5.4041,
"mean_token_accuracy": 0.17094085663557052,
"num_tokens": 9178187.0,
"step": 4985
},
{
"entropy": 5.377679872512817,
"epoch": 0.416805880387571,
"grad_norm": 1.078125,
"learning_rate": 0.0004987476397661472,
"loss": 5.2955,
"mean_token_accuracy": 0.1785726800560951,
"num_tokens": 9186342.0,
"step": 4990
},
{
"entropy": 5.47825608253479,
"epoch": 0.41722352155028397,
"grad_norm": 1.125,
"learning_rate": 0.0004987445011890789,
"loss": 5.4193,
"mean_token_accuracy": 0.17057800740003587,
"num_tokens": 9196470.0,
"step": 4995
},
{
"entropy": 5.393553590774536,
"epoch": 0.417641162712997,
"grad_norm": 1.125,
"learning_rate": 0.000498741358695094,
"loss": 5.3357,
"mean_token_accuracy": 0.18022921085357665,
"num_tokens": 9206587.0,
"step": 5000
},
{
"entropy": 5.378795957565307,
"epoch": 0.41805880387571,
"grad_norm": 1.0625,
"learning_rate": 0.000498738212284247,
"loss": 5.3333,
"mean_token_accuracy": 0.17070587873458862,
"num_tokens": 9216459.0,
"step": 5005
},
{
"entropy": 5.604639482498169,
"epoch": 0.41847644503842296,
"grad_norm": 1.0390625,
"learning_rate": 0.0004987350619565934,
"loss": 5.4398,
"mean_token_accuracy": 0.17089735865592956,
"num_tokens": 9227301.0,
"step": 5010
},
{
"entropy": 5.508641576766967,
"epoch": 0.418894086201136,
"grad_norm": 1.0390625,
"learning_rate": 0.0004987319077121884,
"loss": 5.369,
"mean_token_accuracy": 0.16725013107061387,
"num_tokens": 9236577.0,
"step": 5015
},
{
"entropy": 5.332752132415772,
"epoch": 0.419311727363849,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987287495510869,
"loss": 5.3723,
"mean_token_accuracy": 0.17771873474121094,
"num_tokens": 9246451.0,
"step": 5020
},
{
"entropy": 5.426217317581177,
"epoch": 0.41972936852656195,
"grad_norm": 1.078125,
"learning_rate": 0.0004987255874733444,
"loss": 5.2586,
"mean_token_accuracy": 0.1744895100593567,
"num_tokens": 9256108.0,
"step": 5025
},
{
"entropy": 5.353275871276855,
"epoch": 0.420147009689275,
"grad_norm": 1.1328125,
"learning_rate": 0.0004987224214790161,
"loss": 5.3174,
"mean_token_accuracy": 0.17657518833875657,
"num_tokens": 9265173.0,
"step": 5030
},
{
"entropy": 5.472219324111938,
"epoch": 0.42056465085198796,
"grad_norm": 1.1484375,
"learning_rate": 0.0004987192515681575,
"loss": 5.4175,
"mean_token_accuracy": 0.17183869034051896,
"num_tokens": 9273979.0,
"step": 5035
},
{
"entropy": 5.436709642410278,
"epoch": 0.42098229201470094,
"grad_norm": 1.15625,
"learning_rate": 0.0004987160777408243,
"loss": 5.3531,
"mean_token_accuracy": 0.17467418760061265,
"num_tokens": 9281893.0,
"step": 5040
},
{
"entropy": 5.537257528305053,
"epoch": 0.421399933177414,
"grad_norm": 1.171875,
"learning_rate": 0.0004987128999970717,
"loss": 5.4657,
"mean_token_accuracy": 0.1758879691362381,
"num_tokens": 9290799.0,
"step": 5045
},
{
"entropy": 5.526646900177002,
"epoch": 0.42181757434012696,
"grad_norm": 1.109375,
"learning_rate": 0.0004987097183369555,
"loss": 5.439,
"mean_token_accuracy": 0.16732749193906785,
"num_tokens": 9300562.0,
"step": 5050
},
{
"entropy": 5.344728708267212,
"epoch": 0.42223521550283993,
"grad_norm": 1.09375,
"learning_rate": 0.0004987065327605315,
"loss": 5.2176,
"mean_token_accuracy": 0.18489586859941481,
"num_tokens": 9308889.0,
"step": 5055
},
{
"entropy": 5.336235618591308,
"epoch": 0.42265285666555297,
"grad_norm": 1.0703125,
"learning_rate": 0.0004987033432678553,
"loss": 5.3115,
"mean_token_accuracy": 0.17783447057008744,
"num_tokens": 9317957.0,
"step": 5060
},
{
"entropy": 5.538064289093017,
"epoch": 0.42307049782826595,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987001498589828,
"loss": 5.4978,
"mean_token_accuracy": 0.16852868497371673,
"num_tokens": 9327849.0,
"step": 5065
},
{
"entropy": 5.499323081970215,
"epoch": 0.4234881389909789,
"grad_norm": 1.140625,
"learning_rate": 0.00049869695253397,
"loss": 5.3588,
"mean_token_accuracy": 0.17048564702272415,
"num_tokens": 9337165.0,
"step": 5070
},
{
"entropy": 5.382080745697022,
"epoch": 0.42390578015369196,
"grad_norm": 1.1015625,
"learning_rate": 0.0004986937512928727,
"loss": 5.271,
"mean_token_accuracy": 0.17790974378585817,
"num_tokens": 9346817.0,
"step": 5075
},
{
"entropy": 5.524003171920777,
"epoch": 0.42432342131640494,
"grad_norm": 1.1171875,
"learning_rate": 0.000498690546135747,
"loss": 5.383,
"mean_token_accuracy": 0.1751638501882553,
"num_tokens": 9356420.0,
"step": 5080
},
{
"entropy": 5.344154596328735,
"epoch": 0.4247410624791179,
"grad_norm": 1.171875,
"learning_rate": 0.0004986873370626491,
"loss": 5.3191,
"mean_token_accuracy": 0.17939028441905974,
"num_tokens": 9365580.0,
"step": 5085
},
{
"entropy": 5.391177749633789,
"epoch": 0.42515870364183095,
"grad_norm": 1.0859375,
"learning_rate": 0.000498684124073635,
"loss": 5.3769,
"mean_token_accuracy": 0.17133576571941375,
"num_tokens": 9375116.0,
"step": 5090
},
{
"entropy": 5.4978883266448975,
"epoch": 0.4255763448045439,
"grad_norm": 1.125,
"learning_rate": 0.0004986809071687612,
"loss": 5.353,
"mean_token_accuracy": 0.17932742089033127,
"num_tokens": 9384129.0,
"step": 5095
},
{
"entropy": 5.398127508163452,
"epoch": 0.4259939859672569,
"grad_norm": 1.1328125,
"learning_rate": 0.0004986776863480838,
"loss": 5.2946,
"mean_token_accuracy": 0.1778986394405365,
"num_tokens": 9392757.0,
"step": 5100
},
{
"entropy": 5.340671062469482,
"epoch": 0.42641162712996994,
"grad_norm": 1.09375,
"learning_rate": 0.0004986744616116593,
"loss": 5.2664,
"mean_token_accuracy": 0.17629901617765426,
"num_tokens": 9401753.0,
"step": 5105
},
{
"entropy": 5.452417516708374,
"epoch": 0.4268292682926829,
"grad_norm": 1.0859375,
"learning_rate": 0.000498671232959544,
"loss": 5.4771,
"mean_token_accuracy": 0.16760873049497604,
"num_tokens": 9411400.0,
"step": 5110
},
{
"entropy": 5.488298177719116,
"epoch": 0.42724690945539595,
"grad_norm": 1.140625,
"learning_rate": 0.0004986680003917947,
"loss": 5.3168,
"mean_token_accuracy": 0.1760762795805931,
"num_tokens": 9420206.0,
"step": 5115
},
{
"entropy": 5.440007305145263,
"epoch": 0.42766455061810893,
"grad_norm": 1.09375,
"learning_rate": 0.0004986647639084677,
"loss": 5.3231,
"mean_token_accuracy": 0.1782320261001587,
"num_tokens": 9428418.0,
"step": 5120
},
{
"entropy": 5.436130094528198,
"epoch": 0.4280821917808219,
"grad_norm": 1.1796875,
"learning_rate": 0.00049866152350962,
"loss": 5.3551,
"mean_token_accuracy": 0.17875337302684785,
"num_tokens": 9437624.0,
"step": 5125
},
{
"entropy": 5.354599046707153,
"epoch": 0.42849983294353494,
"grad_norm": 1.109375,
"learning_rate": 0.0004986582791953078,
"loss": 5.281,
"mean_token_accuracy": 0.18172071129083633,
"num_tokens": 9446857.0,
"step": 5130
},
{
"entropy": 5.390440177917481,
"epoch": 0.4289174741062479,
"grad_norm": 1.109375,
"learning_rate": 0.0004986550309655884,
"loss": 5.3842,
"mean_token_accuracy": 0.17395807951688766,
"num_tokens": 9456024.0,
"step": 5135
},
{
"entropy": 5.389064693450928,
"epoch": 0.4293351152689609,
"grad_norm": 1.2578125,
"learning_rate": 0.0004986517788205184,
"loss": 5.2789,
"mean_token_accuracy": 0.18302677869796752,
"num_tokens": 9464018.0,
"step": 5140
},
{
"entropy": 5.397318744659424,
"epoch": 0.42975275643167393,
"grad_norm": 1.140625,
"learning_rate": 0.0004986485227601548,
"loss": 5.3924,
"mean_token_accuracy": 0.17411963343620301,
"num_tokens": 9472430.0,
"step": 5145
},
{
"entropy": 5.411116313934326,
"epoch": 0.4301703975943869,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986452627845547,
"loss": 5.3192,
"mean_token_accuracy": 0.18268029391765594,
"num_tokens": 9480592.0,
"step": 5150
},
{
"entropy": 5.389275979995728,
"epoch": 0.4305880387570999,
"grad_norm": 1.1796875,
"learning_rate": 0.000498641998893775,
"loss": 5.266,
"mean_token_accuracy": 0.18702370822429656,
"num_tokens": 9489162.0,
"step": 5155
},
{
"entropy": 5.399213218688965,
"epoch": 0.4310056799198129,
"grad_norm": 1.1015625,
"learning_rate": 0.0004986387310878728,
"loss": 5.3496,
"mean_token_accuracy": 0.17896925210952758,
"num_tokens": 9499210.0,
"step": 5160
},
{
"entropy": 5.490306854248047,
"epoch": 0.4314233210825259,
"grad_norm": 1.078125,
"learning_rate": 0.0004986354593669056,
"loss": 5.3873,
"mean_token_accuracy": 0.18292965292930602,
"num_tokens": 9508062.0,
"step": 5165
},
{
"entropy": 5.457595777511597,
"epoch": 0.4318409622452389,
"grad_norm": 1.09375,
"learning_rate": 0.0004986321837309304,
"loss": 5.3113,
"mean_token_accuracy": 0.18360127359628678,
"num_tokens": 9516756.0,
"step": 5170
},
{
"entropy": 5.379010248184204,
"epoch": 0.4322586034079519,
"grad_norm": 1.1328125,
"learning_rate": 0.0004986289041800046,
"loss": 5.306,
"mean_token_accuracy": 0.1830867722630501,
"num_tokens": 9525305.0,
"step": 5175
},
{
"entropy": 5.366206789016724,
"epoch": 0.4326762445706649,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986256207141857,
"loss": 5.3112,
"mean_token_accuracy": 0.17962215691804886,
"num_tokens": 9534310.0,
"step": 5180
},
{
"entropy": 5.554233837127685,
"epoch": 0.43309388573337787,
"grad_norm": 1.0625,
"learning_rate": 0.0004986223333335311,
"loss": 5.3906,
"mean_token_accuracy": 0.17555053606629373,
"num_tokens": 9543648.0,
"step": 5185
},
{
"entropy": 5.3237368106842045,
"epoch": 0.4335115268960909,
"grad_norm": 1.1171875,
"learning_rate": 0.0004986190420380985,
"loss": 5.3183,
"mean_token_accuracy": 0.17989316135644912,
"num_tokens": 9553081.0,
"step": 5190
},
{
"entropy": 5.449706602096557,
"epoch": 0.4339291680588039,
"grad_norm": 1.1171875,
"learning_rate": 0.0004986157468279453,
"loss": 5.3588,
"mean_token_accuracy": 0.17973058968782424,
"num_tokens": 9563184.0,
"step": 5195
},
{
"entropy": 5.484891700744629,
"epoch": 0.43434680922151686,
"grad_norm": 1.0703125,
"learning_rate": 0.0004986124477031293,
"loss": 5.3176,
"mean_token_accuracy": 0.17933139204978943,
"num_tokens": 9572550.0,
"step": 5200
},
{
"entropy": 5.35568790435791,
"epoch": 0.4347644503842299,
"grad_norm": 1.0703125,
"learning_rate": 0.0004986091446637083,
"loss": 5.3747,
"mean_token_accuracy": 0.17584731429815292,
"num_tokens": 9580900.0,
"step": 5205
},
{
"entropy": 5.485783100128174,
"epoch": 0.4351820915469429,
"grad_norm": 1.21875,
"learning_rate": 0.00049860583770974,
"loss": 5.3463,
"mean_token_accuracy": 0.18211012482643127,
"num_tokens": 9589340.0,
"step": 5210
},
{
"entropy": 5.475605583190918,
"epoch": 0.43559973270965585,
"grad_norm": 1.046875,
"learning_rate": 0.0004986025268412823,
"loss": 5.4713,
"mean_token_accuracy": 0.17878260910511018,
"num_tokens": 9599316.0,
"step": 5215
},
{
"entropy": 5.4975409507751465,
"epoch": 0.4360173738723689,
"grad_norm": 1.0546875,
"learning_rate": 0.0004985992120583933,
"loss": 5.3714,
"mean_token_accuracy": 0.17172240614891052,
"num_tokens": 9609350.0,
"step": 5220
},
{
"entropy": 5.503689765930176,
"epoch": 0.43643501503508186,
"grad_norm": 1.1328125,
"learning_rate": 0.000498595893361131,
"loss": 5.4071,
"mean_token_accuracy": 0.17903128117322922,
"num_tokens": 9618491.0,
"step": 5225
},
{
"entropy": 5.449696588516235,
"epoch": 0.43685265619779484,
"grad_norm": 1.0234375,
"learning_rate": 0.0004985925707495533,
"loss": 5.4811,
"mean_token_accuracy": 0.16969321966171264,
"num_tokens": 9629613.0,
"step": 5230
},
{
"entropy": 5.409396409988403,
"epoch": 0.4372702973605079,
"grad_norm": 1.21875,
"learning_rate": 0.0004985892442237186,
"loss": 5.3401,
"mean_token_accuracy": 0.18231789022684097,
"num_tokens": 9638035.0,
"step": 5235
},
{
"entropy": 5.519572019577026,
"epoch": 0.43768793852322085,
"grad_norm": 1.0546875,
"learning_rate": 0.0004985859137836851,
"loss": 5.4313,
"mean_token_accuracy": 0.16765984743833542,
"num_tokens": 9647378.0,
"step": 5240
},
{
"entropy": 5.504799222946167,
"epoch": 0.43810557968593383,
"grad_norm": 1.125,
"learning_rate": 0.0004985825794295109,
"loss": 5.3126,
"mean_token_accuracy": 0.18056171983480454,
"num_tokens": 9655728.0,
"step": 5245
},
{
"entropy": 5.4076361656188965,
"epoch": 0.43852322084864687,
"grad_norm": 1.125,
"learning_rate": 0.0004985792411612546,
"loss": 5.3732,
"mean_token_accuracy": 0.1801733359694481,
"num_tokens": 9665063.0,
"step": 5250
},
{
"entropy": 5.47503399848938,
"epoch": 0.43894086201135984,
"grad_norm": 1.078125,
"learning_rate": 0.0004985758989789746,
"loss": 5.3572,
"mean_token_accuracy": 0.17598394453525543,
"num_tokens": 9674406.0,
"step": 5255
},
{
"entropy": 5.367693471908569,
"epoch": 0.4393585031740728,
"grad_norm": 1.0859375,
"learning_rate": 0.0004985725528827293,
"loss": 5.3287,
"mean_token_accuracy": 0.18045067936182022,
"num_tokens": 9683303.0,
"step": 5260
},
{
"entropy": 5.367869329452515,
"epoch": 0.43977614433678586,
"grad_norm": 1.125,
"learning_rate": 0.0004985692028725774,
"loss": 5.3207,
"mean_token_accuracy": 0.17502983659505844,
"num_tokens": 9692938.0,
"step": 5265
},
{
"entropy": 5.492369079589844,
"epoch": 0.44019378549949884,
"grad_norm": 1.1640625,
"learning_rate": 0.0004985658489485774,
"loss": 5.3955,
"mean_token_accuracy": 0.17700106352567674,
"num_tokens": 9702192.0,
"step": 5270
},
{
"entropy": 5.5023651123046875,
"epoch": 0.4406114266622118,
"grad_norm": 1.0234375,
"learning_rate": 0.0004985624911107882,
"loss": 5.355,
"mean_token_accuracy": 0.16955212503671646,
"num_tokens": 9711301.0,
"step": 5275
},
{
"entropy": 5.394012784957885,
"epoch": 0.44102906782492485,
"grad_norm": 1.1171875,
"learning_rate": 0.0004985591293592685,
"loss": 5.3491,
"mean_token_accuracy": 0.17134528905153273,
"num_tokens": 9720768.0,
"step": 5280
},
{
"entropy": 5.480498743057251,
"epoch": 0.4414467089876378,
"grad_norm": 1.1796875,
"learning_rate": 0.0004985557636940771,
"loss": 5.3936,
"mean_token_accuracy": 0.17649945765733718,
"num_tokens": 9730007.0,
"step": 5285
},
{
"entropy": 5.498113584518433,
"epoch": 0.4418643501503508,
"grad_norm": 1.1953125,
"learning_rate": 0.000498552394115273,
"loss": 5.4179,
"mean_token_accuracy": 0.17401509433984758,
"num_tokens": 9739546.0,
"step": 5290
},
{
"entropy": 5.352558422088623,
"epoch": 0.44228199131306384,
"grad_norm": 1.2421875,
"learning_rate": 0.0004985490206229151,
"loss": 5.333,
"mean_token_accuracy": 0.17644752115011214,
"num_tokens": 9749615.0,
"step": 5295
},
{
"entropy": 5.49708194732666,
"epoch": 0.4426996324757768,
"grad_norm": 1.0859375,
"learning_rate": 0.0004985456432170626,
"loss": 5.4166,
"mean_token_accuracy": 0.172282674908638,
"num_tokens": 9759285.0,
"step": 5300
},
{
"entropy": 5.434144401550293,
"epoch": 0.4431172736384898,
"grad_norm": 1.09375,
"learning_rate": 0.0004985422618977744,
"loss": 5.2869,
"mean_token_accuracy": 0.17822668999433516,
"num_tokens": 9769396.0,
"step": 5305
},
{
"entropy": 5.4409651279449465,
"epoch": 0.44353491480120283,
"grad_norm": 1.0859375,
"learning_rate": 0.00049853887666511,
"loss": 5.4028,
"mean_token_accuracy": 0.1721086859703064,
"num_tokens": 9779632.0,
"step": 5310
},
{
"entropy": 5.508889245986938,
"epoch": 0.4439525559639158,
"grad_norm": 1.1875,
"learning_rate": 0.0004985354875191283,
"loss": 5.3215,
"mean_token_accuracy": 0.18220975399017333,
"num_tokens": 9788090.0,
"step": 5315
},
{
"entropy": 5.302156400680542,
"epoch": 0.4443701971266288,
"grad_norm": 1.1171875,
"learning_rate": 0.0004985320944598889,
"loss": 5.2432,
"mean_token_accuracy": 0.18811671137809755,
"num_tokens": 9795844.0,
"step": 5320
},
{
"entropy": 5.274960374832153,
"epoch": 0.4447878382893418,
"grad_norm": 1.15625,
"learning_rate": 0.0004985286974874512,
"loss": 5.2352,
"mean_token_accuracy": 0.1914828434586525,
"num_tokens": 9805119.0,
"step": 5325
},
{
"entropy": 5.494675588607788,
"epoch": 0.4452054794520548,
"grad_norm": 1.1328125,
"learning_rate": 0.0004985252966018745,
"loss": 5.3511,
"mean_token_accuracy": 0.18178332448005677,
"num_tokens": 9813842.0,
"step": 5330
},
{
"entropy": 5.447766351699829,
"epoch": 0.4456231206147678,
"grad_norm": 1.140625,
"learning_rate": 0.0004985218918032184,
"loss": 5.3406,
"mean_token_accuracy": 0.1682691529393196,
"num_tokens": 9822872.0,
"step": 5335
},
{
"entropy": 5.52358775138855,
"epoch": 0.4460407617774808,
"grad_norm": 1.0703125,
"learning_rate": 0.0004985184830915426,
"loss": 5.4632,
"mean_token_accuracy": 0.17231479287147522,
"num_tokens": 9832710.0,
"step": 5340
},
{
"entropy": 5.490021181106568,
"epoch": 0.4464584029401938,
"grad_norm": 1.1796875,
"learning_rate": 0.0004985150704669067,
"loss": 5.3562,
"mean_token_accuracy": 0.17534402161836624,
"num_tokens": 9841047.0,
"step": 5345
},
{
"entropy": 5.416197633743286,
"epoch": 0.44687604410290677,
"grad_norm": 1.046875,
"learning_rate": 0.0004985116539293704,
"loss": 5.3323,
"mean_token_accuracy": 0.17485671788454055,
"num_tokens": 9850168.0,
"step": 5350
},
{
"entropy": 5.391839218139649,
"epoch": 0.4472936852656198,
"grad_norm": 1.125,
"learning_rate": 0.0004985082334789936,
"loss": 5.3799,
"mean_token_accuracy": 0.16806098073720932,
"num_tokens": 9859550.0,
"step": 5355
},
{
"entropy": 5.511943435668945,
"epoch": 0.4477113264283328,
"grad_norm": 1.1171875,
"learning_rate": 0.0004985048091158361,
"loss": 5.3348,
"mean_token_accuracy": 0.177691850066185,
"num_tokens": 9868317.0,
"step": 5360
},
{
"entropy": 5.55558443069458,
"epoch": 0.44812896759104576,
"grad_norm": 1.03125,
"learning_rate": 0.0004985013808399578,
"loss": 5.5465,
"mean_token_accuracy": 0.16269573122262954,
"num_tokens": 9878421.0,
"step": 5365
},
{
"entropy": 5.43976583480835,
"epoch": 0.4485466087537588,
"grad_norm": 1.1640625,
"learning_rate": 0.0004984979486514189,
"loss": 5.2943,
"mean_token_accuracy": 0.18231809586286546,
"num_tokens": 9887213.0,
"step": 5370
},
{
"entropy": 5.406460857391357,
"epoch": 0.44896424991647177,
"grad_norm": 1.0625,
"learning_rate": 0.0004984945125502795,
"loss": 5.4074,
"mean_token_accuracy": 0.17153139263391495,
"num_tokens": 9897274.0,
"step": 5375
},
{
"entropy": 5.436111354827881,
"epoch": 0.44938189107918475,
"grad_norm": 1.25,
"learning_rate": 0.0004984910725365994,
"loss": 5.3318,
"mean_token_accuracy": 0.17394720166921615,
"num_tokens": 9906765.0,
"step": 5380
},
{
"entropy": 5.515875196456909,
"epoch": 0.4497995322418978,
"grad_norm": 1.15625,
"learning_rate": 0.0004984876286104392,
"loss": 5.3697,
"mean_token_accuracy": 0.16384685039520264,
"num_tokens": 9916149.0,
"step": 5385
},
{
"entropy": 5.460874795913696,
"epoch": 0.45021717340461076,
"grad_norm": 1.09375,
"learning_rate": 0.0004984841807718591,
"loss": 5.3469,
"mean_token_accuracy": 0.17650602459907533,
"num_tokens": 9925657.0,
"step": 5390
},
{
"entropy": 5.382658290863037,
"epoch": 0.45063481456732374,
"grad_norm": 1.0625,
"learning_rate": 0.0004984807290209193,
"loss": 5.3234,
"mean_token_accuracy": 0.17409603595733641,
"num_tokens": 9934740.0,
"step": 5395
},
{
"entropy": 5.584508943557739,
"epoch": 0.45105245573003677,
"grad_norm": 1.0625,
"learning_rate": 0.0004984772733576803,
"loss": 5.484,
"mean_token_accuracy": 0.17520519942045212,
"num_tokens": 9943516.0,
"step": 5400
},
{
"entropy": 5.500282096862793,
"epoch": 0.45147009689274975,
"grad_norm": 1.0625,
"learning_rate": 0.0004984738137822028,
"loss": 5.4232,
"mean_token_accuracy": 0.16446612477302552,
"num_tokens": 9952998.0,
"step": 5405
},
{
"entropy": 5.397160243988037,
"epoch": 0.45188773805546273,
"grad_norm": 1.0859375,
"learning_rate": 0.000498470350294547,
"loss": 5.244,
"mean_token_accuracy": 0.18151458501815795,
"num_tokens": 9961698.0,
"step": 5410
},
{
"entropy": 5.424762487411499,
"epoch": 0.45230537921817576,
"grad_norm": 1.078125,
"learning_rate": 0.0004984668828947739,
"loss": 5.3735,
"mean_token_accuracy": 0.17093538343906403,
"num_tokens": 9972015.0,
"step": 5415
},
{
"entropy": 5.461427021026611,
"epoch": 0.45272302038088874,
"grad_norm": 1.0546875,
"learning_rate": 0.0004984634115829439,
"loss": 5.4568,
"mean_token_accuracy": 0.16634009331464766,
"num_tokens": 9981834.0,
"step": 5420
},
{
"entropy": 5.384671831130982,
"epoch": 0.4531406615436017,
"grad_norm": 1.1015625,
"learning_rate": 0.000498459936359118,
"loss": 5.2961,
"mean_token_accuracy": 0.18410439640283585,
"num_tokens": 9991113.0,
"step": 5425
},
{
"entropy": 5.419455718994141,
"epoch": 0.45355830270631475,
"grad_norm": 1.1328125,
"learning_rate": 0.0004984564572233568,
"loss": 5.3002,
"mean_token_accuracy": 0.17880427986383438,
"num_tokens": 10000048.0,
"step": 5430
},
{
"entropy": 5.479193782806396,
"epoch": 0.45397594386902773,
"grad_norm": 1.0703125,
"learning_rate": 0.0004984529741757216,
"loss": 5.3432,
"mean_token_accuracy": 0.1738032504916191,
"num_tokens": 10009389.0,
"step": 5435
},
{
"entropy": 5.342263317108154,
"epoch": 0.4543935850317407,
"grad_norm": 1.1953125,
"learning_rate": 0.0004984494872162728,
"loss": 5.1899,
"mean_token_accuracy": 0.18017552942037582,
"num_tokens": 10018342.0,
"step": 5440
},
{
"entropy": 5.3943102836608885,
"epoch": 0.45481122619445374,
"grad_norm": 1.1328125,
"learning_rate": 0.000498445996345072,
"loss": 5.3701,
"mean_token_accuracy": 0.1769678920507431,
"num_tokens": 10027219.0,
"step": 5445
},
{
"entropy": 5.3587939739227295,
"epoch": 0.4552288673571667,
"grad_norm": 1.15625,
"learning_rate": 0.0004984425015621799,
"loss": 5.2766,
"mean_token_accuracy": 0.17801198810338975,
"num_tokens": 10036635.0,
"step": 5450
},
{
"entropy": 5.362944602966309,
"epoch": 0.4556465085198797,
"grad_norm": 1.2265625,
"learning_rate": 0.000498439002867658,
"loss": 5.312,
"mean_token_accuracy": 0.18080106973648072,
"num_tokens": 10045655.0,
"step": 5455
},
{
"entropy": 5.542423963546753,
"epoch": 0.45606414968259273,
"grad_norm": 1.046875,
"learning_rate": 0.0004984355002615673,
"loss": 5.3854,
"mean_token_accuracy": 0.16903369277715682,
"num_tokens": 10054366.0,
"step": 5460
},
{
"entropy": 5.436358547210693,
"epoch": 0.4564817908453057,
"grad_norm": 1.109375,
"learning_rate": 0.0004984319937439693,
"loss": 5.3342,
"mean_token_accuracy": 0.18257999867200853,
"num_tokens": 10062548.0,
"step": 5465
},
{
"entropy": 5.392876482009887,
"epoch": 0.4568994320080187,
"grad_norm": 1.0859375,
"learning_rate": 0.0004984284833149252,
"loss": 5.3283,
"mean_token_accuracy": 0.1774166539311409,
"num_tokens": 10071333.0,
"step": 5470
},
{
"entropy": 5.346408748626709,
"epoch": 0.4573170731707317,
"grad_norm": 1.1171875,
"learning_rate": 0.0004984249689744966,
"loss": 5.2248,
"mean_token_accuracy": 0.186882945895195,
"num_tokens": 10079502.0,
"step": 5475
},
{
"entropy": 5.352870750427246,
"epoch": 0.4577347143334447,
"grad_norm": 1.0703125,
"learning_rate": 0.000498421450722745,
"loss": 5.2398,
"mean_token_accuracy": 0.18825229704380037,
"num_tokens": 10088648.0,
"step": 5480
},
{
"entropy": 5.340742397308349,
"epoch": 0.4581523554961577,
"grad_norm": 1.09375,
"learning_rate": 0.000498417928559732,
"loss": 5.2626,
"mean_token_accuracy": 0.18324540555477142,
"num_tokens": 10097742.0,
"step": 5485
},
{
"entropy": 5.477860593795777,
"epoch": 0.4585699966588707,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984144024855191,
"loss": 5.357,
"mean_token_accuracy": 0.17452765852212906,
"num_tokens": 10107228.0,
"step": 5490
},
{
"entropy": 5.433778238296509,
"epoch": 0.4589876378215837,
"grad_norm": 1.1484375,
"learning_rate": 0.0004984108725001683,
"loss": 5.2775,
"mean_token_accuracy": 0.17918587923049928,
"num_tokens": 10115488.0,
"step": 5495
},
{
"entropy": 5.304645681381226,
"epoch": 0.45940527898429667,
"grad_norm": 1.1640625,
"learning_rate": 0.0004984073386037413,
"loss": 5.2124,
"mean_token_accuracy": 0.18658908903598787,
"num_tokens": 10124235.0,
"step": 5500
},
{
"entropy": 5.318438005447388,
"epoch": 0.4598229201470097,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984038007962997,
"loss": 5.3139,
"mean_token_accuracy": 0.179317207634449,
"num_tokens": 10133208.0,
"step": 5505
},
{
"entropy": 5.377500009536743,
"epoch": 0.4602405613097227,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984002590779059,
"loss": 5.2342,
"mean_token_accuracy": 0.18262568712234498,
"num_tokens": 10142754.0,
"step": 5510
},
{
"entropy": 5.447999000549316,
"epoch": 0.46065820247243566,
"grad_norm": 1.1875,
"learning_rate": 0.0004983967134486215,
"loss": 5.3543,
"mean_token_accuracy": 0.17982084006071092,
"num_tokens": 10152114.0,
"step": 5515
},
{
"entropy": 5.385389184951782,
"epoch": 0.4610758436351487,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983931639085088,
"loss": 5.2651,
"mean_token_accuracy": 0.18124637454748155,
"num_tokens": 10161562.0,
"step": 5520
},
{
"entropy": 5.396249008178711,
"epoch": 0.4614934847978617,
"grad_norm": 1.1875,
"learning_rate": 0.0004983896104576298,
"loss": 5.3886,
"mean_token_accuracy": 0.17525294572114944,
"num_tokens": 10170461.0,
"step": 5525
},
{
"entropy": 5.423574161529541,
"epoch": 0.46191112596057465,
"grad_norm": 1.265625,
"learning_rate": 0.000498386053096047,
"loss": 5.3709,
"mean_token_accuracy": 0.1749160885810852,
"num_tokens": 10179243.0,
"step": 5530
},
{
"entropy": 5.407944679260254,
"epoch": 0.4623287671232877,
"grad_norm": 1.265625,
"learning_rate": 0.0004983824918238221,
"loss": 5.2918,
"mean_token_accuracy": 0.18273720741271973,
"num_tokens": 10187935.0,
"step": 5535
},
{
"entropy": 5.44966983795166,
"epoch": 0.46274640828600067,
"grad_norm": 1.109375,
"learning_rate": 0.000498378926641018,
"loss": 5.3862,
"mean_token_accuracy": 0.17387618869543076,
"num_tokens": 10196185.0,
"step": 5540
},
{
"entropy": 5.453420782089234,
"epoch": 0.46316404944871364,
"grad_norm": 1.1953125,
"learning_rate": 0.0004983753575476969,
"loss": 5.3573,
"mean_token_accuracy": 0.17224089950323104,
"num_tokens": 10205373.0,
"step": 5545
},
{
"entropy": 5.411218500137329,
"epoch": 0.4635816906114267,
"grad_norm": 1.015625,
"learning_rate": 0.0004983717845439212,
"loss": 5.3272,
"mean_token_accuracy": 0.1835055485367775,
"num_tokens": 10214464.0,
"step": 5550
},
{
"entropy": 5.4495831489562985,
"epoch": 0.46399933177413966,
"grad_norm": 1.125,
"learning_rate": 0.0004983682076297536,
"loss": 5.4208,
"mean_token_accuracy": 0.17140191793441772,
"num_tokens": 10224462.0,
"step": 5555
},
{
"entropy": 5.473630428314209,
"epoch": 0.46441697293685263,
"grad_norm": 1.046875,
"learning_rate": 0.0004983646268052567,
"loss": 5.3438,
"mean_token_accuracy": 0.17826244235038757,
"num_tokens": 10233123.0,
"step": 5560
},
{
"entropy": 5.329004001617432,
"epoch": 0.46483461409956567,
"grad_norm": 1.0859375,
"learning_rate": 0.0004983610420704931,
"loss": 5.1861,
"mean_token_accuracy": 0.184926638007164,
"num_tokens": 10241613.0,
"step": 5565
},
{
"entropy": 5.450223112106324,
"epoch": 0.46525225526227865,
"grad_norm": 1.1015625,
"learning_rate": 0.0004983574534255256,
"loss": 5.5353,
"mean_token_accuracy": 0.16888171583414077,
"num_tokens": 10251809.0,
"step": 5570
},
{
"entropy": 5.375366687774658,
"epoch": 0.4656698964249916,
"grad_norm": 1.1328125,
"learning_rate": 0.000498353860870417,
"loss": 5.2204,
"mean_token_accuracy": 0.18266345113515853,
"num_tokens": 10260198.0,
"step": 5575
},
{
"entropy": 5.472739410400391,
"epoch": 0.46608753758770466,
"grad_norm": 1.015625,
"learning_rate": 0.0004983502644052303,
"loss": 5.2818,
"mean_token_accuracy": 0.1775073826313019,
"num_tokens": 10268906.0,
"step": 5580
},
{
"entropy": 5.328040933609008,
"epoch": 0.46650517875041764,
"grad_norm": 1.0859375,
"learning_rate": 0.0004983466640300282,
"loss": 5.2027,
"mean_token_accuracy": 0.17830956280231475,
"num_tokens": 10278526.0,
"step": 5585
},
{
"entropy": 5.310849571228028,
"epoch": 0.4669228199131306,
"grad_norm": 1.21875,
"learning_rate": 0.0004983430597448741,
"loss": 5.328,
"mean_token_accuracy": 0.17870931774377824,
"num_tokens": 10287428.0,
"step": 5590
},
{
"entropy": 5.519404554367066,
"epoch": 0.46734046107584365,
"grad_norm": 1.140625,
"learning_rate": 0.0004983394515498308,
"loss": 5.3218,
"mean_token_accuracy": 0.1714861884713173,
"num_tokens": 10297136.0,
"step": 5595
},
{
"entropy": 5.426532888412476,
"epoch": 0.46775810223855663,
"grad_norm": 1.1015625,
"learning_rate": 0.0004983358394449616,
"loss": 5.3954,
"mean_token_accuracy": 0.17515646517276764,
"num_tokens": 10305678.0,
"step": 5600
},
{
"entropy": 5.312557744979858,
"epoch": 0.4681757434012696,
"grad_norm": 1.109375,
"learning_rate": 0.0004983322234303297,
"loss": 5.3057,
"mean_token_accuracy": 0.18001095950603485,
"num_tokens": 10314690.0,
"step": 5605
},
{
"entropy": 5.415669727325439,
"epoch": 0.46859338456398264,
"grad_norm": 1.0859375,
"learning_rate": 0.0004983286035059984,
"loss": 5.2547,
"mean_token_accuracy": 0.1789726808667183,
"num_tokens": 10323521.0,
"step": 5610
},
{
"entropy": 5.4074794292449955,
"epoch": 0.4690110257266956,
"grad_norm": 1.1171875,
"learning_rate": 0.0004983249796720311,
"loss": 5.2938,
"mean_token_accuracy": 0.1871481016278267,
"num_tokens": 10333077.0,
"step": 5615
},
{
"entropy": 5.332143354415893,
"epoch": 0.4694286668894086,
"grad_norm": 1.1953125,
"learning_rate": 0.0004983213519284912,
"loss": 5.3141,
"mean_token_accuracy": 0.1771883860230446,
"num_tokens": 10342194.0,
"step": 5620
},
{
"entropy": 5.361347913742065,
"epoch": 0.46984630805212163,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983177202754421,
"loss": 5.1707,
"mean_token_accuracy": 0.18755336254835128,
"num_tokens": 10351009.0,
"step": 5625
},
{
"entropy": 5.417886972427368,
"epoch": 0.4702639492148346,
"grad_norm": 1.25,
"learning_rate": 0.0004983140847129476,
"loss": 5.3638,
"mean_token_accuracy": 0.1762734591960907,
"num_tokens": 10360872.0,
"step": 5630
},
{
"entropy": 5.390548324584961,
"epoch": 0.4706815903775476,
"grad_norm": 1.1171875,
"learning_rate": 0.0004983104452410713,
"loss": 5.2872,
"mean_token_accuracy": 0.1783614620566368,
"num_tokens": 10370159.0,
"step": 5635
},
{
"entropy": 5.37326455116272,
"epoch": 0.4710992315402606,
"grad_norm": 1.1796875,
"learning_rate": 0.0004983068018598768,
"loss": 5.2842,
"mean_token_accuracy": 0.18557757437229155,
"num_tokens": 10379219.0,
"step": 5640
},
{
"entropy": 5.396330547332764,
"epoch": 0.4715168727029736,
"grad_norm": 1.09375,
"learning_rate": 0.000498303154569428,
"loss": 5.3156,
"mean_token_accuracy": 0.17928726822137833,
"num_tokens": 10389425.0,
"step": 5645
},
{
"entropy": 5.451020097732544,
"epoch": 0.4719345138656866,
"grad_norm": 1.234375,
"learning_rate": 0.0004982995033697887,
"loss": 5.3388,
"mean_token_accuracy": 0.18085601776838303,
"num_tokens": 10398023.0,
"step": 5650
},
{
"entropy": 5.383209705352783,
"epoch": 0.4723521550283996,
"grad_norm": 1.203125,
"learning_rate": 0.0004982958482610226,
"loss": 5.2622,
"mean_token_accuracy": 0.17927923053503036,
"num_tokens": 10406578.0,
"step": 5655
},
{
"entropy": 5.318174505233765,
"epoch": 0.4727697961911126,
"grad_norm": 1.0703125,
"learning_rate": 0.000498292189243194,
"loss": 5.2988,
"mean_token_accuracy": 0.182905176281929,
"num_tokens": 10416448.0,
"step": 5660
},
{
"entropy": 5.467951917648316,
"epoch": 0.47318743735382557,
"grad_norm": 1.171875,
"learning_rate": 0.000498288526316367,
"loss": 5.3088,
"mean_token_accuracy": 0.1799843803048134,
"num_tokens": 10425414.0,
"step": 5665
},
{
"entropy": 5.414143800735474,
"epoch": 0.4736050785165386,
"grad_norm": 1.0546875,
"learning_rate": 0.0004982848594806054,
"loss": 5.2907,
"mean_token_accuracy": 0.18458415120840072,
"num_tokens": 10433900.0,
"step": 5670
},
{
"entropy": 5.433807611465454,
"epoch": 0.4740227196792516,
"grad_norm": 1.125,
"learning_rate": 0.0004982811887359736,
"loss": 5.3993,
"mean_token_accuracy": 0.17519030272960662,
"num_tokens": 10443077.0,
"step": 5675
},
{
"entropy": 5.422829675674438,
"epoch": 0.47444036084196456,
"grad_norm": 1.1484375,
"learning_rate": 0.0004982775140825359,
"loss": 5.3329,
"mean_token_accuracy": 0.1719972684979439,
"num_tokens": 10452791.0,
"step": 5680
},
{
"entropy": 5.518968534469605,
"epoch": 0.4748580020046776,
"grad_norm": 1.125,
"learning_rate": 0.0004982738355203565,
"loss": 5.4792,
"mean_token_accuracy": 0.17520434260368348,
"num_tokens": 10462087.0,
"step": 5685
},
{
"entropy": 5.536628866195679,
"epoch": 0.47527564316739057,
"grad_norm": 1.109375,
"learning_rate": 0.0004982701530494998,
"loss": 5.4855,
"mean_token_accuracy": 0.1645841434597969,
"num_tokens": 10472576.0,
"step": 5690
},
{
"entropy": 5.432694959640503,
"epoch": 0.47569328433010355,
"grad_norm": 1.1796875,
"learning_rate": 0.0004982664666700304,
"loss": 5.2081,
"mean_token_accuracy": 0.19180862009525299,
"num_tokens": 10481176.0,
"step": 5695
},
{
"entropy": 5.354813432693481,
"epoch": 0.4761109254928166,
"grad_norm": 1.21875,
"learning_rate": 0.0004982627763820128,
"loss": 5.3057,
"mean_token_accuracy": 0.1755675345659256,
"num_tokens": 10490558.0,
"step": 5700
},
{
"entropy": 5.507250213623047,
"epoch": 0.47652856665552956,
"grad_norm": 1.0625,
"learning_rate": 0.0004982590821855115,
"loss": 5.3753,
"mean_token_accuracy": 0.17909520715475083,
"num_tokens": 10499543.0,
"step": 5705
},
{
"entropy": 5.3077422142028805,
"epoch": 0.47694620781824254,
"grad_norm": 1.1875,
"learning_rate": 0.0004982553840805912,
"loss": 5.2653,
"mean_token_accuracy": 0.18212928324937822,
"num_tokens": 10508162.0,
"step": 5710
},
{
"entropy": 5.328194046020508,
"epoch": 0.4773638489809556,
"grad_norm": 1.0859375,
"learning_rate": 0.0004982516820673167,
"loss": 5.1672,
"mean_token_accuracy": 0.1851506218314171,
"num_tokens": 10516837.0,
"step": 5715
},
{
"entropy": 5.358450841903687,
"epoch": 0.47778149014366855,
"grad_norm": 1.140625,
"learning_rate": 0.0004982479761457528,
"loss": 5.3287,
"mean_token_accuracy": 0.17948042899370192,
"num_tokens": 10526208.0,
"step": 5720
},
{
"entropy": 5.447207117080689,
"epoch": 0.47819913130638153,
"grad_norm": 1.1875,
"learning_rate": 0.0004982442663159643,
"loss": 5.3654,
"mean_token_accuracy": 0.17470479011535645,
"num_tokens": 10536002.0,
"step": 5725
},
{
"entropy": 5.443183565139771,
"epoch": 0.47861677246909456,
"grad_norm": 1.140625,
"learning_rate": 0.0004982405525780162,
"loss": 5.3299,
"mean_token_accuracy": 0.17736094146966935,
"num_tokens": 10545861.0,
"step": 5730
},
{
"entropy": 5.413560009002685,
"epoch": 0.47903441363180754,
"grad_norm": 1.1640625,
"learning_rate": 0.0004982368349319736,
"loss": 5.2533,
"mean_token_accuracy": 0.18133245706558226,
"num_tokens": 10555357.0,
"step": 5735
},
{
"entropy": 5.405882930755615,
"epoch": 0.4794520547945205,
"grad_norm": 1.21875,
"learning_rate": 0.0004982331133779014,
"loss": 5.3266,
"mean_token_accuracy": 0.17830179631710052,
"num_tokens": 10564744.0,
"step": 5740
},
{
"entropy": 5.345563840866089,
"epoch": 0.47986969595723356,
"grad_norm": 1.0390625,
"learning_rate": 0.0004982293879158649,
"loss": 5.2906,
"mean_token_accuracy": 0.18240951001644135,
"num_tokens": 10573736.0,
"step": 5745
},
{
"entropy": 5.358388376235962,
"epoch": 0.48028733711994653,
"grad_norm": 1.109375,
"learning_rate": 0.0004982256585459293,
"loss": 5.3502,
"mean_token_accuracy": 0.17719824910163878,
"num_tokens": 10583238.0,
"step": 5750
},
{
"entropy": 5.459206628799438,
"epoch": 0.4807049782826595,
"grad_norm": 1.125,
"learning_rate": 0.0004982219252681597,
"loss": 5.391,
"mean_token_accuracy": 0.17387733608484268,
"num_tokens": 10592316.0,
"step": 5755
},
{
"entropy": 5.451083660125732,
"epoch": 0.48112261944537255,
"grad_norm": 1.1328125,
"learning_rate": 0.0004982181880826218,
"loss": 5.4167,
"mean_token_accuracy": 0.18136786818504333,
"num_tokens": 10601418.0,
"step": 5760
},
{
"entropy": 5.356658220291138,
"epoch": 0.4815402606080855,
"grad_norm": 1.1015625,
"learning_rate": 0.0004982144469893807,
"loss": 5.2817,
"mean_token_accuracy": 0.18482076227664948,
"num_tokens": 10610513.0,
"step": 5765
},
{
"entropy": 5.395135927200317,
"epoch": 0.4819579017707985,
"grad_norm": 1.09375,
"learning_rate": 0.000498210701988502,
"loss": 5.2398,
"mean_token_accuracy": 0.19010515362024308,
"num_tokens": 10619529.0,
"step": 5770
},
{
"entropy": 5.456553173065186,
"epoch": 0.48237554293351154,
"grad_norm": 1.078125,
"learning_rate": 0.0004982069530800514,
"loss": 5.3362,
"mean_token_accuracy": 0.18640314787626266,
"num_tokens": 10628263.0,
"step": 5775
},
{
"entropy": 5.393055582046509,
"epoch": 0.4827931840962245,
"grad_norm": 1.0703125,
"learning_rate": 0.0004982032002640943,
"loss": 5.3513,
"mean_token_accuracy": 0.18499072194099425,
"num_tokens": 10637605.0,
"step": 5780
},
{
"entropy": 5.448045825958252,
"epoch": 0.48321082525893755,
"grad_norm": 1.1328125,
"learning_rate": 0.0004981994435406965,
"loss": 5.4179,
"mean_token_accuracy": 0.17729529738426208,
"num_tokens": 10647113.0,
"step": 5785
},
{
"entropy": 5.42731237411499,
"epoch": 0.4836284664216505,
"grad_norm": 1.125,
"learning_rate": 0.0004981956829099239,
"loss": 5.181,
"mean_token_accuracy": 0.19317345321178436,
"num_tokens": 10655464.0,
"step": 5790
},
{
"entropy": 5.357528352737427,
"epoch": 0.4840461075843635,
"grad_norm": 1.171875,
"learning_rate": 0.0004981919183718422,
"loss": 5.3106,
"mean_token_accuracy": 0.1787724867463112,
"num_tokens": 10664689.0,
"step": 5795
},
{
"entropy": 5.2970787525177006,
"epoch": 0.48446374874707654,
"grad_norm": 1.09375,
"learning_rate": 0.0004981881499265173,
"loss": 5.1987,
"mean_token_accuracy": 0.1789809450507164,
"num_tokens": 10673440.0,
"step": 5800
},
{
"entropy": 5.339220428466797,
"epoch": 0.4848813899097895,
"grad_norm": 1.0625,
"learning_rate": 0.0004981843775740151,
"loss": 5.3297,
"mean_token_accuracy": 0.17820059955120088,
"num_tokens": 10683075.0,
"step": 5805
},
{
"entropy": 5.440635728836059,
"epoch": 0.4852990310725025,
"grad_norm": 1.1171875,
"learning_rate": 0.0004981806013144017,
"loss": 5.2914,
"mean_token_accuracy": 0.18571672439575196,
"num_tokens": 10692663.0,
"step": 5810
},
{
"entropy": 5.3674538135528564,
"epoch": 0.48571667223521553,
"grad_norm": 1.1015625,
"learning_rate": 0.0004981768211477434,
"loss": 5.1849,
"mean_token_accuracy": 0.18739022463560104,
"num_tokens": 10702068.0,
"step": 5815
},
{
"entropy": 5.376219892501831,
"epoch": 0.4861343133979285,
"grad_norm": 1.0703125,
"learning_rate": 0.000498173037074106,
"loss": 5.3354,
"mean_token_accuracy": 0.17257352769374848,
"num_tokens": 10711230.0,
"step": 5820
},
{
"entropy": 5.347614574432373,
"epoch": 0.4865519545606415,
"grad_norm": 1.1171875,
"learning_rate": 0.0004981692490935561,
"loss": 5.2974,
"mean_token_accuracy": 0.18375744223594664,
"num_tokens": 10719651.0,
"step": 5825
},
{
"entropy": 5.424826049804688,
"epoch": 0.4869695957233545,
"grad_norm": 1.171875,
"learning_rate": 0.0004981654572061597,
"loss": 5.3198,
"mean_token_accuracy": 0.1827986091375351,
"num_tokens": 10729342.0,
"step": 5830
},
{
"entropy": 5.467437267303467,
"epoch": 0.4873872368860675,
"grad_norm": 1.1484375,
"learning_rate": 0.0004981616614119836,
"loss": 5.3074,
"mean_token_accuracy": 0.18250320851802826,
"num_tokens": 10738617.0,
"step": 5835
},
{
"entropy": 5.373071098327637,
"epoch": 0.4878048780487805,
"grad_norm": 1.125,
"learning_rate": 0.0004981578617110938,
"loss": 5.3431,
"mean_token_accuracy": 0.18125945031642915,
"num_tokens": 10746881.0,
"step": 5840
},
{
"entropy": 5.387403392791748,
"epoch": 0.4882225192114935,
"grad_norm": 1.1875,
"learning_rate": 0.0004981540581035571,
"loss": 5.2165,
"mean_token_accuracy": 0.188943549990654,
"num_tokens": 10755848.0,
"step": 5845
},
{
"entropy": 5.367232418060302,
"epoch": 0.4886401603742065,
"grad_norm": 1.125,
"learning_rate": 0.00049815025058944,
"loss": 5.2142,
"mean_token_accuracy": 0.18217775225639343,
"num_tokens": 10765519.0,
"step": 5850
},
{
"entropy": 5.339392805099488,
"epoch": 0.48905780153691947,
"grad_norm": 1.0703125,
"learning_rate": 0.0004981464391688092,
"loss": 5.3335,
"mean_token_accuracy": 0.17536888122558594,
"num_tokens": 10776408.0,
"step": 5855
},
{
"entropy": 5.382868862152099,
"epoch": 0.4894754426996325,
"grad_norm": 1.109375,
"learning_rate": 0.0004981426238417312,
"loss": 5.3149,
"mean_token_accuracy": 0.18250253051519394,
"num_tokens": 10785499.0,
"step": 5860
},
{
"entropy": 5.427073383331299,
"epoch": 0.4898930838623455,
"grad_norm": 1.2734375,
"learning_rate": 0.0004981388046082731,
"loss": 5.2954,
"mean_token_accuracy": 0.17718665450811386,
"num_tokens": 10794736.0,
"step": 5865
},
{
"entropy": 5.393600988388061,
"epoch": 0.49031072502505846,
"grad_norm": 1.1640625,
"learning_rate": 0.0004981349814685016,
"loss": 5.3661,
"mean_token_accuracy": 0.17548820674419402,
"num_tokens": 10804125.0,
"step": 5870
},
{
"entropy": 5.498532390594482,
"epoch": 0.4907283661877715,
"grad_norm": 1.1953125,
"learning_rate": 0.0004981311544224838,
"loss": 5.4037,
"mean_token_accuracy": 0.18298432230949402,
"num_tokens": 10813465.0,
"step": 5875
},
{
"entropy": 5.412685346603394,
"epoch": 0.49114600735048447,
"grad_norm": 1.078125,
"learning_rate": 0.0004981273234702864,
"loss": 5.2333,
"mean_token_accuracy": 0.1874050587415695,
"num_tokens": 10821550.0,
"step": 5880
},
{
"entropy": 5.317541265487671,
"epoch": 0.49156364851319745,
"grad_norm": 1.1640625,
"learning_rate": 0.0004981234886119767,
"loss": 5.2514,
"mean_token_accuracy": 0.18231901824474334,
"num_tokens": 10830683.0,
"step": 5885
},
{
"entropy": 5.430666828155518,
"epoch": 0.4919812896759105,
"grad_norm": 1.1875,
"learning_rate": 0.0004981196498476217,
"loss": 5.2758,
"mean_token_accuracy": 0.1870675653219223,
"num_tokens": 10839492.0,
"step": 5890
},
{
"entropy": 5.377346181869507,
"epoch": 0.49239893083862346,
"grad_norm": 1.09375,
"learning_rate": 0.0004981158071772887,
"loss": 5.1955,
"mean_token_accuracy": 0.1918774351477623,
"num_tokens": 10847998.0,
"step": 5895
},
{
"entropy": 5.1997452735900875,
"epoch": 0.49281657200133644,
"grad_norm": 1.1484375,
"learning_rate": 0.0004981119606010449,
"loss": 5.1961,
"mean_token_accuracy": 0.19257864654064177,
"num_tokens": 10856438.0,
"step": 5900
},
{
"entropy": 5.3496339321136475,
"epoch": 0.4932342131640495,
"grad_norm": 1.21875,
"learning_rate": 0.0004981081101189577,
"loss": 5.2095,
"mean_token_accuracy": 0.18233095705509186,
"num_tokens": 10865936.0,
"step": 5905
},
{
"entropy": 5.452116775512695,
"epoch": 0.49365185432676245,
"grad_norm": 1.109375,
"learning_rate": 0.0004981042557310945,
"loss": 5.2084,
"mean_token_accuracy": 0.18771637827157975,
"num_tokens": 10874621.0,
"step": 5910
},
{
"entropy": 5.283743953704834,
"epoch": 0.49406949548947543,
"grad_norm": 1.125,
"learning_rate": 0.0004981003974375227,
"loss": 5.1845,
"mean_token_accuracy": 0.18220340311527253,
"num_tokens": 10883516.0,
"step": 5915
},
{
"entropy": 5.36220965385437,
"epoch": 0.49448713665218846,
"grad_norm": 1.125,
"learning_rate": 0.0004980965352383099,
"loss": 5.3306,
"mean_token_accuracy": 0.17834787219762802,
"num_tokens": 10892580.0,
"step": 5920
},
{
"entropy": 5.394275522232055,
"epoch": 0.49490477781490144,
"grad_norm": 1.109375,
"learning_rate": 0.0004980926691335238,
"loss": 5.2476,
"mean_token_accuracy": 0.18676111698150635,
"num_tokens": 10901227.0,
"step": 5925
},
{
"entropy": 5.384476232528686,
"epoch": 0.4953224189776144,
"grad_norm": 1.03125,
"learning_rate": 0.0004980887991232318,
"loss": 5.2979,
"mean_token_accuracy": 0.17742019891738892,
"num_tokens": 10909454.0,
"step": 5930
},
{
"entropy": 5.514995479583741,
"epoch": 0.49574006014032745,
"grad_norm": 1.125,
"learning_rate": 0.000498084925207502,
"loss": 5.3588,
"mean_token_accuracy": 0.1789746552705765,
"num_tokens": 10919792.0,
"step": 5935
},
{
"entropy": 5.384832334518433,
"epoch": 0.49615770130304043,
"grad_norm": 1.046875,
"learning_rate": 0.000498081047386402,
"loss": 5.4152,
"mean_token_accuracy": 0.17762281745672226,
"num_tokens": 10930523.0,
"step": 5940
},
{
"entropy": 5.328708362579346,
"epoch": 0.4965753424657534,
"grad_norm": 1.1640625,
"learning_rate": 0.0004980771656599998,
"loss": 5.2657,
"mean_token_accuracy": 0.1854361653327942,
"num_tokens": 10940075.0,
"step": 5945
},
{
"entropy": 5.378875827789306,
"epoch": 0.49699298362846644,
"grad_norm": 1.1640625,
"learning_rate": 0.0004980732800283633,
"loss": 5.2718,
"mean_token_accuracy": 0.17992978692054748,
"num_tokens": 10949059.0,
"step": 5950
},
{
"entropy": 5.3756019115448,
"epoch": 0.4974106247911794,
"grad_norm": 1.1796875,
"learning_rate": 0.0004980693904915604,
"loss": 5.2595,
"mean_token_accuracy": 0.18561322540044783,
"num_tokens": 10958313.0,
"step": 5955
},
{
"entropy": 5.35467562675476,
"epoch": 0.4978282659538924,
"grad_norm": 1.0703125,
"learning_rate": 0.0004980654970496595,
"loss": 5.3093,
"mean_token_accuracy": 0.17678902298212051,
"num_tokens": 10967666.0,
"step": 5960
},
{
"entropy": 5.470755290985108,
"epoch": 0.49824590711660544,
"grad_norm": 1.1875,
"learning_rate": 0.0004980615997027284,
"loss": 5.3382,
"mean_token_accuracy": 0.17992876917123796,
"num_tokens": 10976519.0,
"step": 5965
},
{
"entropy": 5.51383957862854,
"epoch": 0.4986635482793184,
"grad_norm": 1.09375,
"learning_rate": 0.0004980576984508356,
"loss": 5.4453,
"mean_token_accuracy": 0.17184604555368424,
"num_tokens": 10987138.0,
"step": 5970
},
{
"entropy": 5.366135168075561,
"epoch": 0.4990811894420314,
"grad_norm": 1.1640625,
"learning_rate": 0.0004980537932940492,
"loss": 5.3078,
"mean_token_accuracy": 0.18276798576116562,
"num_tokens": 10996517.0,
"step": 5975
},
{
"entropy": 5.419854784011841,
"epoch": 0.4994988306047444,
"grad_norm": 1.203125,
"learning_rate": 0.0004980498842324377,
"loss": 5.1893,
"mean_token_accuracy": 0.17991814911365508,
"num_tokens": 11006312.0,
"step": 5980
},
{
"entropy": 5.345807027816773,
"epoch": 0.4999164717674574,
"grad_norm": 1.15625,
"learning_rate": 0.0004980459712660694,
"loss": 5.3058,
"mean_token_accuracy": 0.18420714586973191,
"num_tokens": 11015535.0,
"step": 5985
},
{
"entropy": 5.349178981781006,
"epoch": 0.5003341129301704,
"grad_norm": 1.046875,
"learning_rate": 0.0004980420543950131,
"loss": 5.2916,
"mean_token_accuracy": 0.18334818333387376,
"num_tokens": 11024887.0,
"step": 5990
},
{
"entropy": 5.326980066299439,
"epoch": 0.5007517540928834,
"grad_norm": 1.109375,
"learning_rate": 0.0004980381336193371,
"loss": 5.1616,
"mean_token_accuracy": 0.1905719518661499,
"num_tokens": 11033147.0,
"step": 5995
},
{
"entropy": 5.317165517807007,
"epoch": 0.5011693952555963,
"grad_norm": 1.0859375,
"learning_rate": 0.0004980342089391099,
"loss": 5.3119,
"mean_token_accuracy": 0.18629232347011565,
"num_tokens": 11042726.0,
"step": 6000
},
{
"epoch": 0.5011693952555963,
"eval_entropy": 5.158226667736274,
"eval_loss": 5.28607177734375,
"eval_mean_token_accuracy": 0.19066548403182845,
"eval_num_tokens": 11042726.0,
"eval_runtime": 27.4205,
"eval_samples_per_second": 1377.183,
"eval_steps_per_second": 172.171,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 119720,
"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": 1.632250497024e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink