fpadovani/heb-hebr-100mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
heb-hebr-100mb-100mb_seed3407/checkpoint-9000/trainer_state.json
ModelHub XC 5952c974ef 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/heb-hebr-100mb-100mb_seed3407
Source: Original Platform
2026-07-17 23:57:58 +08:00

18068 lines
495 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9631333939750656,
"eval_steps": 3000,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.691989517211914,
"epoch": 0.0005350741077639254,
"grad_norm": 10.75,
"learning_rate": 2e-06,
"loss": 10.8201,
"mean_token_accuracy": 0.0,
"num_tokens": 10038.0,
"step": 5
},
{
"entropy": 10.69193468093872,
"epoch": 0.0010701482155278507,
"grad_norm": 11.9375,
"learning_rate": 4.5e-06,
"loss": 10.789,
"mean_token_accuracy": 0.00010810811072587967,
"num_tokens": 21916.0,
"step": 10
},
{
"entropy": 10.690756225585938,
"epoch": 0.0016052223232917758,
"grad_norm": 8.875,
"learning_rate": 7e-06,
"loss": 10.5855,
"mean_token_accuracy": 0.029181174421682953,
"num_tokens": 33434.0,
"step": 15
},
{
"entropy": 10.660381984710693,
"epoch": 0.0021402964310557014,
"grad_norm": 5.71875,
"learning_rate": 9.5e-06,
"loss": 10.3113,
"mean_token_accuracy": 0.04592931531369686,
"num_tokens": 44216.0,
"step": 20
},
{
"entropy": 10.51073760986328,
"epoch": 0.0026753705388196263,
"grad_norm": 3.71875,
"learning_rate": 1.2e-05,
"loss": 10.1036,
"mean_token_accuracy": 0.04351861346513033,
"num_tokens": 54874.0,
"step": 25
},
{
"entropy": 10.5190824508667,
"epoch": 0.0032104446465835517,
"grad_norm": 3.09375,
"learning_rate": 1.4500000000000002e-05,
"loss": 9.9872,
"mean_token_accuracy": 0.041112912446260454,
"num_tokens": 65085.0,
"step": 30
},
{
"entropy": 10.547387790679931,
"epoch": 0.003745518754347477,
"grad_norm": 2.859375,
"learning_rate": 1.7000000000000003e-05,
"loss": 9.904,
"mean_token_accuracy": 0.04395376648753881,
"num_tokens": 75601.0,
"step": 35
},
{
"entropy": 10.474061203002929,
"epoch": 0.004280592862111403,
"grad_norm": 2.515625,
"learning_rate": 1.95e-05,
"loss": 9.8879,
"mean_token_accuracy": 0.04307105913758278,
"num_tokens": 86193.0,
"step": 40
},
{
"entropy": 10.531705379486084,
"epoch": 0.004815666969875327,
"grad_norm": 2.40625,
"learning_rate": 2.2e-05,
"loss": 9.8417,
"mean_token_accuracy": 0.043867765367031096,
"num_tokens": 96957.0,
"step": 45
},
{
"entropy": 10.478270435333252,
"epoch": 0.005350741077639253,
"grad_norm": 2.171875,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.8114,
"mean_token_accuracy": 0.046108495071530343,
"num_tokens": 108408.0,
"step": 50
},
{
"entropy": 10.473760986328125,
"epoch": 0.005885815185403178,
"grad_norm": 1.7578125,
"learning_rate": 2.7e-05,
"loss": 9.7686,
"mean_token_accuracy": 0.05027508921921253,
"num_tokens": 119273.0,
"step": 55
},
{
"entropy": 10.462549209594727,
"epoch": 0.006420889293167103,
"grad_norm": 1.921875,
"learning_rate": 2.95e-05,
"loss": 9.6954,
"mean_token_accuracy": 0.05541543699800968,
"num_tokens": 130817.0,
"step": 60
},
{
"entropy": 10.302016353607177,
"epoch": 0.006955963400931029,
"grad_norm": 1.8671875,
"learning_rate": 3.2e-05,
"loss": 9.6448,
"mean_token_accuracy": 0.05586838573217392,
"num_tokens": 142599.0,
"step": 65
},
{
"entropy": 10.404022407531738,
"epoch": 0.007491037508694954,
"grad_norm": 1.9609375,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.5697,
"mean_token_accuracy": 0.05663685463368893,
"num_tokens": 153617.0,
"step": 70
},
{
"entropy": 10.37060546875,
"epoch": 0.00802611161645888,
"grad_norm": 1.6953125,
"learning_rate": 3.7e-05,
"loss": 9.5239,
"mean_token_accuracy": 0.05575243458151817,
"num_tokens": 165280.0,
"step": 75
},
{
"entropy": 10.341149616241456,
"epoch": 0.008561185724222806,
"grad_norm": 1.546875,
"learning_rate": 3.95e-05,
"loss": 9.4643,
"mean_token_accuracy": 0.055300182476639745,
"num_tokens": 176896.0,
"step": 80
},
{
"entropy": 10.367101669311523,
"epoch": 0.00909625983198673,
"grad_norm": 1.5859375,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.4019,
"mean_token_accuracy": 0.05629164129495621,
"num_tokens": 188000.0,
"step": 85
},
{
"entropy": 10.27455005645752,
"epoch": 0.009631333939750655,
"grad_norm": 1.6953125,
"learning_rate": 4.45e-05,
"loss": 9.3493,
"mean_token_accuracy": 0.057158011198043826,
"num_tokens": 199360.0,
"step": 90
},
{
"entropy": 10.319551277160645,
"epoch": 0.01016640804751458,
"grad_norm": 1.5703125,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.2957,
"mean_token_accuracy": 0.052791529521346095,
"num_tokens": 210214.0,
"step": 95
},
{
"entropy": 10.196423625946045,
"epoch": 0.010701482155278505,
"grad_norm": 1.5078125,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.1506,
"mean_token_accuracy": 0.06025413721799851,
"num_tokens": 220567.0,
"step": 100
},
{
"entropy": 10.14912462234497,
"epoch": 0.011236556263042432,
"grad_norm": 1.578125,
"learning_rate": 5.2e-05,
"loss": 9.057,
"mean_token_accuracy": 0.06051998771727085,
"num_tokens": 230585.0,
"step": 105
},
{
"entropy": 10.099914455413819,
"epoch": 0.011771630370806356,
"grad_norm": 1.53125,
"learning_rate": 5.45e-05,
"loss": 8.9907,
"mean_token_accuracy": 0.06339392885565757,
"num_tokens": 241996.0,
"step": 110
},
{
"entropy": 10.083064746856689,
"epoch": 0.012306704478570282,
"grad_norm": 1.3671875,
"learning_rate": 5.7e-05,
"loss": 8.9558,
"mean_token_accuracy": 0.05828723199665546,
"num_tokens": 253348.0,
"step": 115
},
{
"entropy": 9.981218147277833,
"epoch": 0.012841778586334207,
"grad_norm": 1.421875,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.8376,
"mean_token_accuracy": 0.05671278983354568,
"num_tokens": 264239.0,
"step": 120
},
{
"entropy": 9.828715324401855,
"epoch": 0.013376852694098133,
"grad_norm": 1.2578125,
"learning_rate": 6.2e-05,
"loss": 8.7502,
"mean_token_accuracy": 0.056628919392824176,
"num_tokens": 274522.0,
"step": 125
},
{
"entropy": 9.77067985534668,
"epoch": 0.013911926801862057,
"grad_norm": 1.34375,
"learning_rate": 6.450000000000001e-05,
"loss": 8.6474,
"mean_token_accuracy": 0.06091504544019699,
"num_tokens": 284492.0,
"step": 130
},
{
"entropy": 9.592880916595458,
"epoch": 0.014447000909625984,
"grad_norm": 1.40625,
"learning_rate": 6.7e-05,
"loss": 8.6035,
"mean_token_accuracy": 0.05946716032922268,
"num_tokens": 294796.0,
"step": 135
},
{
"entropy": 9.45958080291748,
"epoch": 0.014982075017389908,
"grad_norm": 1.0,
"learning_rate": 6.950000000000001e-05,
"loss": 8.574,
"mean_token_accuracy": 0.05749143101274967,
"num_tokens": 305252.0,
"step": 140
},
{
"entropy": 9.359578514099121,
"epoch": 0.015517149125153834,
"grad_norm": 1.0859375,
"learning_rate": 7.2e-05,
"loss": 8.4804,
"mean_token_accuracy": 0.056437050178647044,
"num_tokens": 315505.0,
"step": 145
},
{
"entropy": 9.158096408843994,
"epoch": 0.01605222323291776,
"grad_norm": 1.5,
"learning_rate": 7.45e-05,
"loss": 8.4819,
"mean_token_accuracy": 0.05990221984684467,
"num_tokens": 326997.0,
"step": 150
},
{
"entropy": 9.094014644622803,
"epoch": 0.016587297340681685,
"grad_norm": 1.109375,
"learning_rate": 7.7e-05,
"loss": 8.4165,
"mean_token_accuracy": 0.05523004196584225,
"num_tokens": 338089.0,
"step": 155
},
{
"entropy": 8.880174732208252,
"epoch": 0.01712237144844561,
"grad_norm": 1.234375,
"learning_rate": 7.950000000000001e-05,
"loss": 8.4077,
"mean_token_accuracy": 0.05682347491383553,
"num_tokens": 348536.0,
"step": 160
},
{
"entropy": 8.93060073852539,
"epoch": 0.017657445556209534,
"grad_norm": 0.9296875,
"learning_rate": 8.2e-05,
"loss": 8.4181,
"mean_token_accuracy": 0.058796605467796324,
"num_tokens": 359776.0,
"step": 165
},
{
"entropy": 8.809493732452392,
"epoch": 0.01819251966397346,
"grad_norm": 1.015625,
"learning_rate": 8.450000000000001e-05,
"loss": 8.3472,
"mean_token_accuracy": 0.06165213547646999,
"num_tokens": 370810.0,
"step": 170
},
{
"entropy": 8.747115898132325,
"epoch": 0.018727593771737387,
"grad_norm": 0.94140625,
"learning_rate": 8.7e-05,
"loss": 8.3511,
"mean_token_accuracy": 0.05888295993208885,
"num_tokens": 381788.0,
"step": 175
},
{
"entropy": 8.742884063720703,
"epoch": 0.01926266787950131,
"grad_norm": 1.0390625,
"learning_rate": 8.95e-05,
"loss": 8.3376,
"mean_token_accuracy": 0.0602983683347702,
"num_tokens": 391984.0,
"step": 180
},
{
"entropy": 8.610486602783203,
"epoch": 0.019797741987265235,
"grad_norm": 1.0,
"learning_rate": 9.2e-05,
"loss": 8.3311,
"mean_token_accuracy": 0.058925900235772134,
"num_tokens": 403430.0,
"step": 185
},
{
"entropy": 8.670144844055176,
"epoch": 0.02033281609502916,
"grad_norm": 1.0234375,
"learning_rate": 9.45e-05,
"loss": 8.318,
"mean_token_accuracy": 0.05627542734146118,
"num_tokens": 414467.0,
"step": 190
},
{
"entropy": 8.56268243789673,
"epoch": 0.020867890202793088,
"grad_norm": 0.921875,
"learning_rate": 9.7e-05,
"loss": 8.2838,
"mean_token_accuracy": 0.06462269835174084,
"num_tokens": 425752.0,
"step": 195
},
{
"entropy": 8.562543296813965,
"epoch": 0.02140296431055701,
"grad_norm": 0.94921875,
"learning_rate": 9.95e-05,
"loss": 8.2618,
"mean_token_accuracy": 0.06471644267439843,
"num_tokens": 435908.0,
"step": 200
},
{
"entropy": 8.58950891494751,
"epoch": 0.021938038418320937,
"grad_norm": 1.0390625,
"learning_rate": 0.000102,
"loss": 8.3051,
"mean_token_accuracy": 0.05973440445959568,
"num_tokens": 446424.0,
"step": 205
},
{
"entropy": 8.575794887542724,
"epoch": 0.022473112526084863,
"grad_norm": 1.03125,
"learning_rate": 0.00010449999999999999,
"loss": 8.2691,
"mean_token_accuracy": 0.0648487914353609,
"num_tokens": 456854.0,
"step": 210
},
{
"entropy": 8.506811428070069,
"epoch": 0.02300818663384879,
"grad_norm": 1.078125,
"learning_rate": 0.000107,
"loss": 8.2861,
"mean_token_accuracy": 0.06673729792237282,
"num_tokens": 467899.0,
"step": 215
},
{
"entropy": 8.497941493988037,
"epoch": 0.023543260741612712,
"grad_norm": 1.0078125,
"learning_rate": 0.0001095,
"loss": 8.2297,
"mean_token_accuracy": 0.06340378113090991,
"num_tokens": 478683.0,
"step": 220
},
{
"entropy": 8.574148082733155,
"epoch": 0.02407833484937664,
"grad_norm": 1.046875,
"learning_rate": 0.000112,
"loss": 8.2648,
"mean_token_accuracy": 0.06170066706836223,
"num_tokens": 488745.0,
"step": 225
},
{
"entropy": 8.4907470703125,
"epoch": 0.024613408957140565,
"grad_norm": 1.0703125,
"learning_rate": 0.0001145,
"loss": 8.3431,
"mean_token_accuracy": 0.059930100291967395,
"num_tokens": 499448.0,
"step": 230
},
{
"entropy": 8.599916458129883,
"epoch": 0.02514848306490449,
"grad_norm": 0.95703125,
"learning_rate": 0.00011700000000000001,
"loss": 8.27,
"mean_token_accuracy": 0.0650645636022091,
"num_tokens": 510148.0,
"step": 235
},
{
"entropy": 8.483675956726074,
"epoch": 0.025683557172668414,
"grad_norm": 1.0078125,
"learning_rate": 0.00011949999999999999,
"loss": 8.1965,
"mean_token_accuracy": 0.06112907975912094,
"num_tokens": 522093.0,
"step": 240
},
{
"entropy": 8.460341262817384,
"epoch": 0.02621863128043234,
"grad_norm": 1.015625,
"learning_rate": 0.000122,
"loss": 8.2277,
"mean_token_accuracy": 0.06530554480850696,
"num_tokens": 532598.0,
"step": 245
},
{
"entropy": 8.49462070465088,
"epoch": 0.026753705388196266,
"grad_norm": 1.171875,
"learning_rate": 0.0001245,
"loss": 8.1956,
"mean_token_accuracy": 0.06578520573675632,
"num_tokens": 542652.0,
"step": 250
},
{
"entropy": 8.460633659362793,
"epoch": 0.027288779495960192,
"grad_norm": 1.015625,
"learning_rate": 0.000127,
"loss": 8.2328,
"mean_token_accuracy": 0.06693280301988125,
"num_tokens": 552526.0,
"step": 255
},
{
"entropy": 8.399180793762207,
"epoch": 0.027823853603724115,
"grad_norm": 0.98046875,
"learning_rate": 0.0001295,
"loss": 8.2517,
"mean_token_accuracy": 0.06344069167971611,
"num_tokens": 563143.0,
"step": 260
},
{
"entropy": 8.48379201889038,
"epoch": 0.02835892771148804,
"grad_norm": 0.94921875,
"learning_rate": 0.000132,
"loss": 8.2149,
"mean_token_accuracy": 0.06581672765314579,
"num_tokens": 573814.0,
"step": 265
},
{
"entropy": 8.457896327972412,
"epoch": 0.028894001819251967,
"grad_norm": 0.890625,
"learning_rate": 0.00013450000000000002,
"loss": 8.1733,
"mean_token_accuracy": 0.07218663729727268,
"num_tokens": 585367.0,
"step": 270
},
{
"entropy": 8.507985496520996,
"epoch": 0.02942907592701589,
"grad_norm": 3.078125,
"learning_rate": 0.00013700000000000002,
"loss": 8.2318,
"mean_token_accuracy": 0.06521918512880802,
"num_tokens": 595902.0,
"step": 275
},
{
"entropy": 8.41034107208252,
"epoch": 0.029964150034779816,
"grad_norm": 1.4375,
"learning_rate": 0.0001395,
"loss": 8.138,
"mean_token_accuracy": 0.0642226304858923,
"num_tokens": 607178.0,
"step": 280
},
{
"entropy": 8.431483745574951,
"epoch": 0.030499224142543743,
"grad_norm": 0.94140625,
"learning_rate": 0.00014199999999999998,
"loss": 8.1309,
"mean_token_accuracy": 0.06817701198160649,
"num_tokens": 618604.0,
"step": 285
},
{
"entropy": 8.404507827758788,
"epoch": 0.03103429825030767,
"grad_norm": 0.9921875,
"learning_rate": 0.0001445,
"loss": 8.2,
"mean_token_accuracy": 0.0719299040734768,
"num_tokens": 628111.0,
"step": 290
},
{
"entropy": 8.490602397918702,
"epoch": 0.03156937235807159,
"grad_norm": 1.453125,
"learning_rate": 0.000147,
"loss": 8.2289,
"mean_token_accuracy": 0.06933129876852036,
"num_tokens": 639358.0,
"step": 295
},
{
"entropy": 8.35900354385376,
"epoch": 0.03210444646583552,
"grad_norm": 1.0078125,
"learning_rate": 0.0001495,
"loss": 8.0903,
"mean_token_accuracy": 0.07086234539747238,
"num_tokens": 649504.0,
"step": 300
},
{
"entropy": 8.400328350067138,
"epoch": 0.032639520573599444,
"grad_norm": 1.1171875,
"learning_rate": 0.000152,
"loss": 8.1111,
"mean_token_accuracy": 0.0681851863861084,
"num_tokens": 659924.0,
"step": 305
},
{
"entropy": 8.422282886505126,
"epoch": 0.03317459468136337,
"grad_norm": 1.328125,
"learning_rate": 0.00015450000000000001,
"loss": 8.15,
"mean_token_accuracy": 0.06455899253487588,
"num_tokens": 671120.0,
"step": 310
},
{
"entropy": 8.319589424133301,
"epoch": 0.033709668789127296,
"grad_norm": 1.0390625,
"learning_rate": 0.000157,
"loss": 8.1985,
"mean_token_accuracy": 0.06430512070655822,
"num_tokens": 681956.0,
"step": 315
},
{
"entropy": 8.48956241607666,
"epoch": 0.03424474289689122,
"grad_norm": 0.953125,
"learning_rate": 0.0001595,
"loss": 8.182,
"mean_token_accuracy": 0.06503869034349918,
"num_tokens": 693226.0,
"step": 320
},
{
"entropy": 8.33993911743164,
"epoch": 0.03477981700465514,
"grad_norm": 0.94140625,
"learning_rate": 0.000162,
"loss": 8.1812,
"mean_token_accuracy": 0.07134119421243668,
"num_tokens": 704691.0,
"step": 325
},
{
"entropy": 8.362220954895019,
"epoch": 0.03531489111241907,
"grad_norm": 0.94921875,
"learning_rate": 0.00016450000000000001,
"loss": 8.1304,
"mean_token_accuracy": 0.062357674539089206,
"num_tokens": 716000.0,
"step": 330
},
{
"entropy": 8.332426834106446,
"epoch": 0.035849965220182994,
"grad_norm": 0.9296875,
"learning_rate": 0.00016700000000000002,
"loss": 8.0529,
"mean_token_accuracy": 0.06962560564279556,
"num_tokens": 726723.0,
"step": 335
},
{
"entropy": 8.32654685974121,
"epoch": 0.03638503932794692,
"grad_norm": 1.0390625,
"learning_rate": 0.00016950000000000003,
"loss": 8.0646,
"mean_token_accuracy": 0.07072275690734386,
"num_tokens": 736740.0,
"step": 340
},
{
"entropy": 8.333600425720215,
"epoch": 0.03692011343571085,
"grad_norm": 1.859375,
"learning_rate": 0.00017199999999999998,
"loss": 8.1131,
"mean_token_accuracy": 0.068069913610816,
"num_tokens": 747567.0,
"step": 345
},
{
"entropy": 8.398985290527344,
"epoch": 0.03745518754347477,
"grad_norm": 1.03125,
"learning_rate": 0.00017449999999999999,
"loss": 8.0626,
"mean_token_accuracy": 0.06914932392537594,
"num_tokens": 757937.0,
"step": 350
},
{
"entropy": 8.163579368591309,
"epoch": 0.0379902616512387,
"grad_norm": 1.0234375,
"learning_rate": 0.000177,
"loss": 8.0955,
"mean_token_accuracy": 0.07165054567158222,
"num_tokens": 768003.0,
"step": 355
},
{
"entropy": 8.49596710205078,
"epoch": 0.03852533575900262,
"grad_norm": 0.9453125,
"learning_rate": 0.0001795,
"loss": 8.1288,
"mean_token_accuracy": 0.06314317509531975,
"num_tokens": 779098.0,
"step": 360
},
{
"entropy": 8.285042762756348,
"epoch": 0.039060409866766545,
"grad_norm": 1.1796875,
"learning_rate": 0.000182,
"loss": 8.099,
"mean_token_accuracy": 0.06513171158730983,
"num_tokens": 790002.0,
"step": 365
},
{
"entropy": 8.28923921585083,
"epoch": 0.03959548397453047,
"grad_norm": 1.0234375,
"learning_rate": 0.0001845,
"loss": 8.0887,
"mean_token_accuracy": 0.06843880005180836,
"num_tokens": 800444.0,
"step": 370
},
{
"entropy": 8.375462627410888,
"epoch": 0.0401305580822944,
"grad_norm": 1.0390625,
"learning_rate": 0.000187,
"loss": 8.0323,
"mean_token_accuracy": 0.0713271751999855,
"num_tokens": 810912.0,
"step": 375
},
{
"entropy": 8.272585010528564,
"epoch": 0.04066563219005832,
"grad_norm": 0.9375,
"learning_rate": 0.0001895,
"loss": 8.1198,
"mean_token_accuracy": 0.0620170421898365,
"num_tokens": 822809.0,
"step": 380
},
{
"entropy": 8.293894195556641,
"epoch": 0.04120070629782225,
"grad_norm": 1.0078125,
"learning_rate": 0.000192,
"loss": 8.0433,
"mean_token_accuracy": 0.07052243873476982,
"num_tokens": 833362.0,
"step": 385
},
{
"entropy": 8.276584053039551,
"epoch": 0.041735780405586176,
"grad_norm": 1.109375,
"learning_rate": 0.0001945,
"loss": 8.0465,
"mean_token_accuracy": 0.07357921116054059,
"num_tokens": 844239.0,
"step": 390
},
{
"entropy": 8.217984676361084,
"epoch": 0.0422708545133501,
"grad_norm": 1.0546875,
"learning_rate": 0.00019700000000000002,
"loss": 8.0493,
"mean_token_accuracy": 0.06498224660754204,
"num_tokens": 855178.0,
"step": 395
},
{
"entropy": 8.19730110168457,
"epoch": 0.04280592862111402,
"grad_norm": 1.15625,
"learning_rate": 0.00019950000000000002,
"loss": 8.0369,
"mean_token_accuracy": 0.07206248976290226,
"num_tokens": 866120.0,
"step": 400
},
{
"entropy": 8.337959480285644,
"epoch": 0.04334100272887795,
"grad_norm": 0.9453125,
"learning_rate": 0.000202,
"loss": 8.0361,
"mean_token_accuracy": 0.07243039160966873,
"num_tokens": 876569.0,
"step": 405
},
{
"entropy": 8.284227752685547,
"epoch": 0.043876076836641874,
"grad_norm": 0.9921875,
"learning_rate": 0.00020449999999999998,
"loss": 8.0951,
"mean_token_accuracy": 0.06462065242230892,
"num_tokens": 887580.0,
"step": 410
},
{
"entropy": 8.235477828979493,
"epoch": 0.0444111509444058,
"grad_norm": 1.046875,
"learning_rate": 0.000207,
"loss": 8.0165,
"mean_token_accuracy": 0.0666106827557087,
"num_tokens": 897922.0,
"step": 415
},
{
"entropy": 8.222909069061279,
"epoch": 0.044946225052169726,
"grad_norm": 1.0234375,
"learning_rate": 0.0002095,
"loss": 7.9827,
"mean_token_accuracy": 0.07159848175942898,
"num_tokens": 908522.0,
"step": 420
},
{
"entropy": 8.27180290222168,
"epoch": 0.04548129915993365,
"grad_norm": 1.078125,
"learning_rate": 0.000212,
"loss": 7.9946,
"mean_token_accuracy": 0.06654324233531952,
"num_tokens": 918253.0,
"step": 425
},
{
"entropy": 8.194285202026368,
"epoch": 0.04601637326769758,
"grad_norm": 0.91015625,
"learning_rate": 0.0002145,
"loss": 8.0034,
"mean_token_accuracy": 0.07207662984728813,
"num_tokens": 929150.0,
"step": 430
},
{
"entropy": 8.239371013641357,
"epoch": 0.0465514473754615,
"grad_norm": 0.94140625,
"learning_rate": 0.00021700000000000002,
"loss": 8.0511,
"mean_token_accuracy": 0.07106035277247429,
"num_tokens": 941041.0,
"step": 435
},
{
"entropy": 8.180944156646728,
"epoch": 0.047086521483225424,
"grad_norm": 1.59375,
"learning_rate": 0.0002195,
"loss": 7.9757,
"mean_token_accuracy": 0.07760139927268028,
"num_tokens": 952714.0,
"step": 440
},
{
"entropy": 8.150847339630127,
"epoch": 0.04762159559098935,
"grad_norm": 1.109375,
"learning_rate": 0.000222,
"loss": 8.0514,
"mean_token_accuracy": 0.06907343231141568,
"num_tokens": 964330.0,
"step": 445
},
{
"entropy": 8.246770858764648,
"epoch": 0.04815666969875328,
"grad_norm": 1.1796875,
"learning_rate": 0.0002245,
"loss": 7.9805,
"mean_token_accuracy": 0.06862297914922237,
"num_tokens": 975257.0,
"step": 450
},
{
"entropy": 8.260761165618897,
"epoch": 0.0486917438065172,
"grad_norm": 1.09375,
"learning_rate": 0.00022700000000000002,
"loss": 7.9876,
"mean_token_accuracy": 0.07001541927456856,
"num_tokens": 986335.0,
"step": 455
},
{
"entropy": 8.069970512390137,
"epoch": 0.04922681791428113,
"grad_norm": 0.859375,
"learning_rate": 0.00022950000000000002,
"loss": 7.8883,
"mean_token_accuracy": 0.07471956759691238,
"num_tokens": 997521.0,
"step": 460
},
{
"entropy": 8.172566795349121,
"epoch": 0.049761892022045055,
"grad_norm": 0.921875,
"learning_rate": 0.00023200000000000003,
"loss": 7.941,
"mean_token_accuracy": 0.0669164728373289,
"num_tokens": 1008383.0,
"step": 465
},
{
"entropy": 8.124609184265136,
"epoch": 0.05029696612980898,
"grad_norm": 1.09375,
"learning_rate": 0.00023449999999999998,
"loss": 7.998,
"mean_token_accuracy": 0.06806416064500809,
"num_tokens": 1020552.0,
"step": 470
},
{
"entropy": 8.196929836273194,
"epoch": 0.0508320402375729,
"grad_norm": 0.90625,
"learning_rate": 0.000237,
"loss": 7.9718,
"mean_token_accuracy": 0.07053619101643563,
"num_tokens": 1031127.0,
"step": 475
},
{
"entropy": 8.022743558883667,
"epoch": 0.05136711434533683,
"grad_norm": 0.890625,
"learning_rate": 0.0002395,
"loss": 7.8837,
"mean_token_accuracy": 0.07386223524808884,
"num_tokens": 1043209.0,
"step": 480
},
{
"entropy": 8.2311222076416,
"epoch": 0.05190218845310075,
"grad_norm": 0.96875,
"learning_rate": 0.000242,
"loss": 8.0182,
"mean_token_accuracy": 0.06782832555472851,
"num_tokens": 1054032.0,
"step": 485
},
{
"entropy": 8.030213356018066,
"epoch": 0.05243726256086468,
"grad_norm": 1.03125,
"learning_rate": 0.0002445,
"loss": 7.8628,
"mean_token_accuracy": 0.07288614809513091,
"num_tokens": 1064565.0,
"step": 490
},
{
"entropy": 8.178169107437133,
"epoch": 0.052972336668628606,
"grad_norm": 0.94140625,
"learning_rate": 0.000247,
"loss": 7.9975,
"mean_token_accuracy": 0.0721826508641243,
"num_tokens": 1076352.0,
"step": 495
},
{
"entropy": 8.112149047851563,
"epoch": 0.05350741077639253,
"grad_norm": 1.1484375,
"learning_rate": 0.0002495,
"loss": 7.9381,
"mean_token_accuracy": 0.07542271986603737,
"num_tokens": 1087697.0,
"step": 500
},
{
"entropy": 8.137825202941894,
"epoch": 0.05404248488415646,
"grad_norm": 0.99609375,
"learning_rate": 0.000252,
"loss": 7.876,
"mean_token_accuracy": 0.07182962782680988,
"num_tokens": 1098056.0,
"step": 505
},
{
"entropy": 8.017078876495361,
"epoch": 0.054577558991920384,
"grad_norm": 1.0,
"learning_rate": 0.0002545,
"loss": 7.8477,
"mean_token_accuracy": 0.07691125422716141,
"num_tokens": 1109219.0,
"step": 510
},
{
"entropy": 8.091530561447144,
"epoch": 0.055112633099684304,
"grad_norm": 0.94140625,
"learning_rate": 0.000257,
"loss": 7.9312,
"mean_token_accuracy": 0.0700716007500887,
"num_tokens": 1120248.0,
"step": 515
},
{
"entropy": 8.096777963638306,
"epoch": 0.05564770720744823,
"grad_norm": 1.1171875,
"learning_rate": 0.0002595,
"loss": 7.9366,
"mean_token_accuracy": 0.07186660170555115,
"num_tokens": 1131278.0,
"step": 520
},
{
"entropy": 8.089863491058349,
"epoch": 0.056182781315212156,
"grad_norm": 1.1875,
"learning_rate": 0.000262,
"loss": 7.9537,
"mean_token_accuracy": 0.06224438659846783,
"num_tokens": 1143147.0,
"step": 525
},
{
"entropy": 8.162030553817749,
"epoch": 0.05671785542297608,
"grad_norm": 0.89453125,
"learning_rate": 0.00026450000000000003,
"loss": 7.84,
"mean_token_accuracy": 0.0739860113710165,
"num_tokens": 1154316.0,
"step": 530
},
{
"entropy": 7.977855777740478,
"epoch": 0.05725292953074001,
"grad_norm": 1.0234375,
"learning_rate": 0.00026700000000000004,
"loss": 7.8967,
"mean_token_accuracy": 0.06953345276415349,
"num_tokens": 1164696.0,
"step": 535
},
{
"entropy": 8.048009157180786,
"epoch": 0.057788003638503935,
"grad_norm": 1.0,
"learning_rate": 0.00026950000000000005,
"loss": 7.8769,
"mean_token_accuracy": 0.07776758968830108,
"num_tokens": 1176129.0,
"step": 540
},
{
"entropy": 8.093161487579346,
"epoch": 0.05832307774626786,
"grad_norm": 1.0625,
"learning_rate": 0.00027200000000000005,
"loss": 7.8881,
"mean_token_accuracy": 0.07334664314985276,
"num_tokens": 1186499.0,
"step": 545
},
{
"entropy": 8.04116907119751,
"epoch": 0.05885815185403178,
"grad_norm": 0.98046875,
"learning_rate": 0.0002745,
"loss": 7.8557,
"mean_token_accuracy": 0.07062512040138244,
"num_tokens": 1196841.0,
"step": 550
},
{
"entropy": 8.01404366493225,
"epoch": 0.059393225961795706,
"grad_norm": 0.91796875,
"learning_rate": 0.000277,
"loss": 7.8108,
"mean_token_accuracy": 0.07467551231384277,
"num_tokens": 1207647.0,
"step": 555
},
{
"entropy": 7.952892255783081,
"epoch": 0.05992830006955963,
"grad_norm": 1.1015625,
"learning_rate": 0.0002795,
"loss": 7.801,
"mean_token_accuracy": 0.07650438733398915,
"num_tokens": 1217072.0,
"step": 560
},
{
"entropy": 8.061871433258057,
"epoch": 0.06046337417732356,
"grad_norm": 1.0859375,
"learning_rate": 0.00028199999999999997,
"loss": 7.9129,
"mean_token_accuracy": 0.0698221705853939,
"num_tokens": 1228564.0,
"step": 565
},
{
"entropy": 7.987078428268433,
"epoch": 0.060998448285087485,
"grad_norm": 0.99609375,
"learning_rate": 0.0002845,
"loss": 7.8458,
"mean_token_accuracy": 0.07497354932129383,
"num_tokens": 1238755.0,
"step": 570
},
{
"entropy": 7.964116954803467,
"epoch": 0.06153352239285141,
"grad_norm": 1.1484375,
"learning_rate": 0.000287,
"loss": 7.8004,
"mean_token_accuracy": 0.0752273216843605,
"num_tokens": 1248968.0,
"step": 575
},
{
"entropy": 7.962353372573853,
"epoch": 0.06206859650061534,
"grad_norm": 1.0,
"learning_rate": 0.0002895,
"loss": 7.7714,
"mean_token_accuracy": 0.07172017395496369,
"num_tokens": 1259256.0,
"step": 580
},
{
"entropy": 7.952462911605835,
"epoch": 0.06260367060837926,
"grad_norm": 1.0859375,
"learning_rate": 0.000292,
"loss": 7.8067,
"mean_token_accuracy": 0.07487296611070633,
"num_tokens": 1270469.0,
"step": 585
},
{
"entropy": 7.963806390762329,
"epoch": 0.06313874471614318,
"grad_norm": 1.0078125,
"learning_rate": 0.0002945,
"loss": 7.7694,
"mean_token_accuracy": 0.07279012463986874,
"num_tokens": 1280558.0,
"step": 590
},
{
"entropy": 7.986854600906372,
"epoch": 0.06367381882390712,
"grad_norm": 1.046875,
"learning_rate": 0.000297,
"loss": 7.8307,
"mean_token_accuracy": 0.0718784749507904,
"num_tokens": 1291372.0,
"step": 595
},
{
"entropy": 8.02498688697815,
"epoch": 0.06420889293167104,
"grad_norm": 0.9921875,
"learning_rate": 0.0002995,
"loss": 7.9061,
"mean_token_accuracy": 0.0674042422324419,
"num_tokens": 1302521.0,
"step": 600
},
{
"entropy": 7.8433699131011965,
"epoch": 0.06474396703943497,
"grad_norm": 1.109375,
"learning_rate": 0.000302,
"loss": 7.7274,
"mean_token_accuracy": 0.07444682195782662,
"num_tokens": 1313632.0,
"step": 605
},
{
"entropy": 8.005569124221802,
"epoch": 0.06527904114719889,
"grad_norm": 0.96484375,
"learning_rate": 0.0003045,
"loss": 7.8118,
"mean_token_accuracy": 0.07571293860673904,
"num_tokens": 1325247.0,
"step": 610
},
{
"entropy": 7.937390947341919,
"epoch": 0.06581411525496281,
"grad_norm": 0.9453125,
"learning_rate": 0.000307,
"loss": 7.8015,
"mean_token_accuracy": 0.075490290671587,
"num_tokens": 1336026.0,
"step": 615
},
{
"entropy": 7.93022665977478,
"epoch": 0.06634918936272674,
"grad_norm": 0.93359375,
"learning_rate": 0.0003095,
"loss": 7.7122,
"mean_token_accuracy": 0.0769732192158699,
"num_tokens": 1346920.0,
"step": 620
},
{
"entropy": 7.828203010559082,
"epoch": 0.06688426347049066,
"grad_norm": 1.21875,
"learning_rate": 0.000312,
"loss": 7.8585,
"mean_token_accuracy": 0.0724661260843277,
"num_tokens": 1359752.0,
"step": 625
},
{
"entropy": 8.050937986373901,
"epoch": 0.06741933757825459,
"grad_norm": 0.95703125,
"learning_rate": 0.0003145,
"loss": 7.8615,
"mean_token_accuracy": 0.07357175201177597,
"num_tokens": 1370874.0,
"step": 630
},
{
"entropy": 7.897751331329346,
"epoch": 0.06795441168601851,
"grad_norm": 0.9140625,
"learning_rate": 0.000317,
"loss": 7.7624,
"mean_token_accuracy": 0.07756051793694496,
"num_tokens": 1382316.0,
"step": 635
},
{
"entropy": 7.840310096740723,
"epoch": 0.06848948579378245,
"grad_norm": 1.0,
"learning_rate": 0.0003195,
"loss": 7.7878,
"mean_token_accuracy": 0.07577594220638276,
"num_tokens": 1394656.0,
"step": 640
},
{
"entropy": 7.928963756561279,
"epoch": 0.06902455990154636,
"grad_norm": 1.0,
"learning_rate": 0.000322,
"loss": 7.7522,
"mean_token_accuracy": 0.07719769552350045,
"num_tokens": 1405061.0,
"step": 645
},
{
"entropy": 7.838529491424561,
"epoch": 0.06955963400931028,
"grad_norm": 0.9140625,
"learning_rate": 0.00032450000000000003,
"loss": 7.7183,
"mean_token_accuracy": 0.08189134374260902,
"num_tokens": 1415573.0,
"step": 650
},
{
"entropy": 7.810403823852539,
"epoch": 0.07009470811707422,
"grad_norm": 1.015625,
"learning_rate": 0.00032700000000000003,
"loss": 7.7436,
"mean_token_accuracy": 0.07435674965381622,
"num_tokens": 1426103.0,
"step": 655
},
{
"entropy": 7.886046600341797,
"epoch": 0.07062978222483814,
"grad_norm": 1.6796875,
"learning_rate": 0.00032950000000000004,
"loss": 7.7599,
"mean_token_accuracy": 0.07234186008572578,
"num_tokens": 1437256.0,
"step": 660
},
{
"entropy": 7.909786796569824,
"epoch": 0.07116485633260207,
"grad_norm": 1.0859375,
"learning_rate": 0.00033200000000000005,
"loss": 7.7241,
"mean_token_accuracy": 0.0786049198359251,
"num_tokens": 1446975.0,
"step": 665
},
{
"entropy": 7.831602573394775,
"epoch": 0.07169993044036599,
"grad_norm": 0.99609375,
"learning_rate": 0.00033450000000000005,
"loss": 7.7425,
"mean_token_accuracy": 0.07444198578596115,
"num_tokens": 1457751.0,
"step": 670
},
{
"entropy": 7.829366731643677,
"epoch": 0.07223500454812992,
"grad_norm": 1.046875,
"learning_rate": 0.000337,
"loss": 7.7049,
"mean_token_accuracy": 0.0754860796034336,
"num_tokens": 1468186.0,
"step": 675
},
{
"entropy": 7.825147390365601,
"epoch": 0.07277007865589384,
"grad_norm": 0.83203125,
"learning_rate": 0.0003395,
"loss": 7.7508,
"mean_token_accuracy": 0.07570655383169651,
"num_tokens": 1479780.0,
"step": 680
},
{
"entropy": 7.8603636741638185,
"epoch": 0.07330515276365776,
"grad_norm": 0.94921875,
"learning_rate": 0.000342,
"loss": 7.6712,
"mean_token_accuracy": 0.0832133986055851,
"num_tokens": 1490565.0,
"step": 685
},
{
"entropy": 7.892835903167724,
"epoch": 0.0738402268714217,
"grad_norm": 0.96875,
"learning_rate": 0.00034449999999999997,
"loss": 7.7227,
"mean_token_accuracy": 0.07809135504066944,
"num_tokens": 1501552.0,
"step": 690
},
{
"entropy": 7.798111057281494,
"epoch": 0.07437530097918561,
"grad_norm": 0.9375,
"learning_rate": 0.000347,
"loss": 7.7035,
"mean_token_accuracy": 0.07376831583678722,
"num_tokens": 1513311.0,
"step": 695
},
{
"entropy": 7.795570516586304,
"epoch": 0.07491037508694955,
"grad_norm": 0.98046875,
"learning_rate": 0.0003495,
"loss": 7.7428,
"mean_token_accuracy": 0.08057990670204163,
"num_tokens": 1524142.0,
"step": 700
},
{
"entropy": 7.8165356636047365,
"epoch": 0.07544544919471347,
"grad_norm": 1.03125,
"learning_rate": 0.000352,
"loss": 7.7188,
"mean_token_accuracy": 0.07599906846880913,
"num_tokens": 1535571.0,
"step": 705
},
{
"entropy": 7.865922546386718,
"epoch": 0.0759805233024774,
"grad_norm": 0.89453125,
"learning_rate": 0.0003545,
"loss": 7.7225,
"mean_token_accuracy": 0.07798022851347923,
"num_tokens": 1546528.0,
"step": 710
},
{
"entropy": 7.727042579650879,
"epoch": 0.07651559741024132,
"grad_norm": 0.8828125,
"learning_rate": 0.000357,
"loss": 7.7386,
"mean_token_accuracy": 0.07202219888567925,
"num_tokens": 1557264.0,
"step": 715
},
{
"entropy": 7.867890882492065,
"epoch": 0.07705067151800524,
"grad_norm": 1.0625,
"learning_rate": 0.0003595,
"loss": 7.6785,
"mean_token_accuracy": 0.08276100754737854,
"num_tokens": 1566773.0,
"step": 720
},
{
"entropy": 7.699678039550781,
"epoch": 0.07758574562576917,
"grad_norm": 0.96875,
"learning_rate": 0.000362,
"loss": 7.6459,
"mean_token_accuracy": 0.08662735894322396,
"num_tokens": 1578173.0,
"step": 725
},
{
"entropy": 7.779191970825195,
"epoch": 0.07812081973353309,
"grad_norm": 0.98046875,
"learning_rate": 0.0003645,
"loss": 7.6221,
"mean_token_accuracy": 0.08249625936150551,
"num_tokens": 1588537.0,
"step": 730
},
{
"entropy": 7.7231770038604735,
"epoch": 0.07865589384129702,
"grad_norm": 0.98046875,
"learning_rate": 0.000367,
"loss": 7.6355,
"mean_token_accuracy": 0.08746174871921539,
"num_tokens": 1599159.0,
"step": 735
},
{
"entropy": 7.7618378639221195,
"epoch": 0.07919096794906094,
"grad_norm": 0.9921875,
"learning_rate": 0.0003695,
"loss": 7.6989,
"mean_token_accuracy": 0.07925031632184983,
"num_tokens": 1610378.0,
"step": 740
},
{
"entropy": 7.720492362976074,
"epoch": 0.07972604205682488,
"grad_norm": 1.1328125,
"learning_rate": 0.000372,
"loss": 7.705,
"mean_token_accuracy": 0.07612368240952491,
"num_tokens": 1621215.0,
"step": 745
},
{
"entropy": 7.786748743057251,
"epoch": 0.0802611161645888,
"grad_norm": 1.0625,
"learning_rate": 0.0003745,
"loss": 7.6622,
"mean_token_accuracy": 0.0790612380951643,
"num_tokens": 1631170.0,
"step": 750
},
{
"entropy": 7.78691840171814,
"epoch": 0.08079619027235273,
"grad_norm": 0.98828125,
"learning_rate": 0.000377,
"loss": 7.6574,
"mean_token_accuracy": 0.07428538724780083,
"num_tokens": 1641897.0,
"step": 755
},
{
"entropy": 7.766008996963501,
"epoch": 0.08133126438011665,
"grad_norm": 0.95703125,
"learning_rate": 0.0003795,
"loss": 7.7353,
"mean_token_accuracy": 0.07569916620850563,
"num_tokens": 1653592.0,
"step": 760
},
{
"entropy": 7.616559743881226,
"epoch": 0.08186633848788057,
"grad_norm": 0.9375,
"learning_rate": 0.000382,
"loss": 7.6528,
"mean_token_accuracy": 0.07474047541618348,
"num_tokens": 1665760.0,
"step": 765
},
{
"entropy": 7.830330228805542,
"epoch": 0.0824014125956445,
"grad_norm": 0.9609375,
"learning_rate": 0.0003845,
"loss": 7.6454,
"mean_token_accuracy": 0.08570240736007691,
"num_tokens": 1676211.0,
"step": 770
},
{
"entropy": 7.737586927413941,
"epoch": 0.08293648670340842,
"grad_norm": 1.0,
"learning_rate": 0.00038700000000000003,
"loss": 7.6666,
"mean_token_accuracy": 0.0797475405037403,
"num_tokens": 1686405.0,
"step": 775
},
{
"entropy": 7.75378794670105,
"epoch": 0.08347156081117235,
"grad_norm": 1.0546875,
"learning_rate": 0.00038950000000000003,
"loss": 7.6699,
"mean_token_accuracy": 0.07950277514755726,
"num_tokens": 1697340.0,
"step": 780
},
{
"entropy": 7.726286888122559,
"epoch": 0.08400663491893627,
"grad_norm": 1.046875,
"learning_rate": 0.00039200000000000004,
"loss": 7.5787,
"mean_token_accuracy": 0.0861763522028923,
"num_tokens": 1707250.0,
"step": 785
},
{
"entropy": 7.63549427986145,
"epoch": 0.0845417090267002,
"grad_norm": 1.0078125,
"learning_rate": 0.00039450000000000005,
"loss": 7.5653,
"mean_token_accuracy": 0.0784637302160263,
"num_tokens": 1718333.0,
"step": 790
},
{
"entropy": 7.709765481948852,
"epoch": 0.08507678313446412,
"grad_norm": 1.0625,
"learning_rate": 0.00039700000000000005,
"loss": 7.6627,
"mean_token_accuracy": 0.07198781482875347,
"num_tokens": 1730520.0,
"step": 795
},
{
"entropy": 7.7255151748657225,
"epoch": 0.08561185724222804,
"grad_norm": 1.0625,
"learning_rate": 0.0003995,
"loss": 7.7149,
"mean_token_accuracy": 0.07679420076310635,
"num_tokens": 1741898.0,
"step": 800
},
{
"entropy": 7.718593311309815,
"epoch": 0.08614693134999198,
"grad_norm": 1.09375,
"learning_rate": 0.000402,
"loss": 7.6508,
"mean_token_accuracy": 0.08017718270421029,
"num_tokens": 1751671.0,
"step": 805
},
{
"entropy": 7.577520227432251,
"epoch": 0.0866820054577559,
"grad_norm": 0.91015625,
"learning_rate": 0.0004045,
"loss": 7.6669,
"mean_token_accuracy": 0.08265335634350776,
"num_tokens": 1762076.0,
"step": 810
},
{
"entropy": 7.796188259124756,
"epoch": 0.08721707956551983,
"grad_norm": 0.98046875,
"learning_rate": 0.00040699999999999997,
"loss": 7.642,
"mean_token_accuracy": 0.07752098590135574,
"num_tokens": 1772666.0,
"step": 815
},
{
"entropy": 7.724516534805298,
"epoch": 0.08775215367328375,
"grad_norm": 0.94921875,
"learning_rate": 0.0004095,
"loss": 7.6572,
"mean_token_accuracy": 0.08227546997368336,
"num_tokens": 1783621.0,
"step": 820
},
{
"entropy": 7.700971364974976,
"epoch": 0.08828722778104768,
"grad_norm": 1.5234375,
"learning_rate": 0.000412,
"loss": 7.7147,
"mean_token_accuracy": 0.07807004414498805,
"num_tokens": 1794654.0,
"step": 825
},
{
"entropy": 7.651264762878418,
"epoch": 0.0888223018888116,
"grad_norm": 0.88671875,
"learning_rate": 0.0004145,
"loss": 7.6008,
"mean_token_accuracy": 0.0848226711153984,
"num_tokens": 1805853.0,
"step": 830
},
{
"entropy": 7.536796283721924,
"epoch": 0.08935737599657552,
"grad_norm": 0.9375,
"learning_rate": 0.000417,
"loss": 7.5896,
"mean_token_accuracy": 0.0828265130519867,
"num_tokens": 1817006.0,
"step": 835
},
{
"entropy": 7.733750724792481,
"epoch": 0.08989245010433945,
"grad_norm": 1.0,
"learning_rate": 0.0004195,
"loss": 7.583,
"mean_token_accuracy": 0.07734453342854977,
"num_tokens": 1827166.0,
"step": 840
},
{
"entropy": 7.7373639106750485,
"epoch": 0.09042752421210337,
"grad_norm": 1.140625,
"learning_rate": 0.000422,
"loss": 7.6809,
"mean_token_accuracy": 0.07751412987709046,
"num_tokens": 1837182.0,
"step": 845
},
{
"entropy": 7.567351484298706,
"epoch": 0.0909625983198673,
"grad_norm": 0.94140625,
"learning_rate": 0.0004245,
"loss": 7.6388,
"mean_token_accuracy": 0.07895995602011681,
"num_tokens": 1849034.0,
"step": 850
},
{
"entropy": 7.595164203643799,
"epoch": 0.09149767242763122,
"grad_norm": 0.98828125,
"learning_rate": 0.000427,
"loss": 7.5075,
"mean_token_accuracy": 0.08833360373973846,
"num_tokens": 1859033.0,
"step": 855
},
{
"entropy": 7.610419750213623,
"epoch": 0.09203274653539516,
"grad_norm": 0.9765625,
"learning_rate": 0.0004295,
"loss": 7.6121,
"mean_token_accuracy": 0.08088177517056465,
"num_tokens": 1870274.0,
"step": 860
},
{
"entropy": 7.5731220722198485,
"epoch": 0.09256782064315908,
"grad_norm": 1.265625,
"learning_rate": 0.000432,
"loss": 7.5611,
"mean_token_accuracy": 0.08315053284168243,
"num_tokens": 1881306.0,
"step": 865
},
{
"entropy": 7.561473751068116,
"epoch": 0.093102894750923,
"grad_norm": 1.0078125,
"learning_rate": 0.0004345,
"loss": 7.5356,
"mean_token_accuracy": 0.08015808910131454,
"num_tokens": 1892464.0,
"step": 870
},
{
"entropy": 7.661314296722412,
"epoch": 0.09363796885868693,
"grad_norm": 0.9609375,
"learning_rate": 0.000437,
"loss": 7.6219,
"mean_token_accuracy": 0.07585933655500413,
"num_tokens": 1903743.0,
"step": 875
},
{
"entropy": 7.607653522491455,
"epoch": 0.09417304296645085,
"grad_norm": 1.0,
"learning_rate": 0.0004395,
"loss": 7.5902,
"mean_token_accuracy": 0.08468717783689499,
"num_tokens": 1915253.0,
"step": 880
},
{
"entropy": 7.622669744491577,
"epoch": 0.09470811707421478,
"grad_norm": 1.2421875,
"learning_rate": 0.000442,
"loss": 7.5473,
"mean_token_accuracy": 0.08187699615955353,
"num_tokens": 1926113.0,
"step": 885
},
{
"entropy": 7.57468581199646,
"epoch": 0.0952431911819787,
"grad_norm": 0.98046875,
"learning_rate": 0.0004445,
"loss": 7.6456,
"mean_token_accuracy": 0.08173620626330376,
"num_tokens": 1937718.0,
"step": 890
},
{
"entropy": 7.625470066070557,
"epoch": 0.09577826528974263,
"grad_norm": 1.0234375,
"learning_rate": 0.000447,
"loss": 7.6334,
"mean_token_accuracy": 0.0791581004858017,
"num_tokens": 1948307.0,
"step": 895
},
{
"entropy": 7.575066232681275,
"epoch": 0.09631333939750655,
"grad_norm": 0.96484375,
"learning_rate": 0.00044950000000000003,
"loss": 7.4358,
"mean_token_accuracy": 0.09009880647063255,
"num_tokens": 1957861.0,
"step": 900
},
{
"entropy": 7.480437755584717,
"epoch": 0.09684841350527049,
"grad_norm": 0.92578125,
"learning_rate": 0.00045200000000000004,
"loss": 7.4911,
"mean_token_accuracy": 0.08970233350992203,
"num_tokens": 1967984.0,
"step": 905
},
{
"entropy": 7.55365891456604,
"epoch": 0.0973834876130344,
"grad_norm": 1.0,
"learning_rate": 0.00045450000000000004,
"loss": 7.4885,
"mean_token_accuracy": 0.08413884043693542,
"num_tokens": 1978608.0,
"step": 910
},
{
"entropy": 7.550483322143554,
"epoch": 0.09791856172079832,
"grad_norm": 0.98828125,
"learning_rate": 0.00045700000000000005,
"loss": 7.547,
"mean_token_accuracy": 0.08248279690742492,
"num_tokens": 1989355.0,
"step": 915
},
{
"entropy": 7.514042758941651,
"epoch": 0.09845363582856226,
"grad_norm": 0.96484375,
"learning_rate": 0.00045950000000000006,
"loss": 7.5303,
"mean_token_accuracy": 0.08344742506742478,
"num_tokens": 1999754.0,
"step": 920
},
{
"entropy": 7.540969705581665,
"epoch": 0.09898870993632618,
"grad_norm": 0.99609375,
"learning_rate": 0.000462,
"loss": 7.5232,
"mean_token_accuracy": 0.0832397285848856,
"num_tokens": 2010877.0,
"step": 925
},
{
"entropy": 7.635095930099487,
"epoch": 0.09952378404409011,
"grad_norm": 0.9140625,
"learning_rate": 0.0004645,
"loss": 7.6286,
"mean_token_accuracy": 0.08003128916025162,
"num_tokens": 2022301.0,
"step": 930
},
{
"entropy": 7.549521160125733,
"epoch": 0.10005885815185403,
"grad_norm": 0.99609375,
"learning_rate": 0.000467,
"loss": 7.5119,
"mean_token_accuracy": 0.08251530304551125,
"num_tokens": 2033537.0,
"step": 935
},
{
"entropy": 7.477602958679199,
"epoch": 0.10059393225961796,
"grad_norm": 1.0390625,
"learning_rate": 0.0004695,
"loss": 7.4502,
"mean_token_accuracy": 0.0781034879386425,
"num_tokens": 2044360.0,
"step": 940
},
{
"entropy": 7.583854675292969,
"epoch": 0.10112900636738188,
"grad_norm": 0.9296875,
"learning_rate": 0.000472,
"loss": 7.5601,
"mean_token_accuracy": 0.08421726487576961,
"num_tokens": 2055000.0,
"step": 945
},
{
"entropy": 7.492854499816895,
"epoch": 0.1016640804751458,
"grad_norm": 0.9609375,
"learning_rate": 0.0004745,
"loss": 7.5467,
"mean_token_accuracy": 0.08681007251143455,
"num_tokens": 2064807.0,
"step": 950
},
{
"entropy": 7.5536168098449705,
"epoch": 0.10219915458290973,
"grad_norm": 0.96484375,
"learning_rate": 0.000477,
"loss": 7.5102,
"mean_token_accuracy": 0.08256960362195968,
"num_tokens": 2075618.0,
"step": 955
},
{
"entropy": 7.505359792709351,
"epoch": 0.10273422869067365,
"grad_norm": 0.953125,
"learning_rate": 0.0004795,
"loss": 7.5116,
"mean_token_accuracy": 0.08244576305150986,
"num_tokens": 2086226.0,
"step": 960
},
{
"entropy": 7.541278457641601,
"epoch": 0.10326930279843759,
"grad_norm": 0.89453125,
"learning_rate": 0.000482,
"loss": 7.5116,
"mean_token_accuracy": 0.08005514740943909,
"num_tokens": 2097222.0,
"step": 965
},
{
"entropy": 7.481921100616455,
"epoch": 0.1038043769062015,
"grad_norm": 0.90625,
"learning_rate": 0.0004845,
"loss": 7.541,
"mean_token_accuracy": 0.08661947920918464,
"num_tokens": 2108919.0,
"step": 970
},
{
"entropy": 7.46715989112854,
"epoch": 0.10433945101396544,
"grad_norm": 0.96875,
"learning_rate": 0.000487,
"loss": 7.4939,
"mean_token_accuracy": 0.08317564725875855,
"num_tokens": 2120215.0,
"step": 975
},
{
"entropy": 7.5228071212768555,
"epoch": 0.10487452512172936,
"grad_norm": 1.015625,
"learning_rate": 0.0004895,
"loss": 7.5188,
"mean_token_accuracy": 0.0818349651992321,
"num_tokens": 2131503.0,
"step": 980
},
{
"entropy": 7.487857866287231,
"epoch": 0.10540959922949328,
"grad_norm": 0.96484375,
"learning_rate": 0.000492,
"loss": 7.4588,
"mean_token_accuracy": 0.08936587274074555,
"num_tokens": 2141551.0,
"step": 985
},
{
"entropy": 7.4559633255004885,
"epoch": 0.10594467333725721,
"grad_norm": 0.92578125,
"learning_rate": 0.0004945,
"loss": 7.4615,
"mean_token_accuracy": 0.08929954171180725,
"num_tokens": 2151821.0,
"step": 990
},
{
"entropy": 7.451591444015503,
"epoch": 0.10647974744502113,
"grad_norm": 1.03125,
"learning_rate": 0.000497,
"loss": 7.5065,
"mean_token_accuracy": 0.08539062738418579,
"num_tokens": 2162474.0,
"step": 995
},
{
"entropy": 7.53413724899292,
"epoch": 0.10701482155278506,
"grad_norm": 0.96875,
"learning_rate": 0.0004995,
"loss": 7.5602,
"mean_token_accuracy": 0.08457510098814965,
"num_tokens": 2173985.0,
"step": 1000
},
{
"entropy": 7.5422203063964846,
"epoch": 0.10754989566054898,
"grad_norm": 0.90625,
"learning_rate": 0.0004999999979210101,
"loss": 7.5944,
"mean_token_accuracy": 0.07782966643571854,
"num_tokens": 2186432.0,
"step": 1005
},
{
"entropy": 7.4759071350097654,
"epoch": 0.10808496976831292,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999999894751138,
"loss": 7.4463,
"mean_token_accuracy": 0.0834141232073307,
"num_tokens": 2197085.0,
"step": 1010
},
{
"entropy": 7.424153137207031,
"epoch": 0.10862004387607684,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999999745323744,
"loss": 7.4486,
"mean_token_accuracy": 0.08975227214396,
"num_tokens": 2207162.0,
"step": 1015
},
{
"entropy": 7.464861583709717,
"epoch": 0.10915511798384077,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999999530927923,
"loss": 7.5469,
"mean_token_accuracy": 0.07938519641757011,
"num_tokens": 2218714.0,
"step": 1020
},
{
"entropy": 7.515410757064819,
"epoch": 0.10969019209160469,
"grad_norm": 0.96875,
"learning_rate": 0.0004999999251563682,
"loss": 7.4879,
"mean_token_accuracy": 0.08599804565310479,
"num_tokens": 2229711.0,
"step": 1025
},
{
"entropy": 7.45625114440918,
"epoch": 0.11022526619936861,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999998907231029,
"loss": 7.4912,
"mean_token_accuracy": 0.08116911202669144,
"num_tokens": 2239885.0,
"step": 1030
},
{
"entropy": 7.481421899795532,
"epoch": 0.11076034030713254,
"grad_norm": 1.0,
"learning_rate": 0.0004999998497929974,
"loss": 7.4923,
"mean_token_accuracy": 0.08827452957630158,
"num_tokens": 2250221.0,
"step": 1035
},
{
"entropy": 7.4760716438293455,
"epoch": 0.11129541441489646,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999998023660527,
"loss": 7.5526,
"mean_token_accuracy": 0.07815601006150245,
"num_tokens": 2261059.0,
"step": 1040
},
{
"entropy": 7.454020214080811,
"epoch": 0.11183048852266039,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999997484422705,
"loss": 7.4273,
"mean_token_accuracy": 0.08656688034534454,
"num_tokens": 2271174.0,
"step": 1045
},
{
"entropy": 7.404057025909424,
"epoch": 0.11236556263042431,
"grad_norm": 1.0625,
"learning_rate": 0.0004999996880216521,
"loss": 7.5037,
"mean_token_accuracy": 0.08764296993613244,
"num_tokens": 2282431.0,
"step": 1050
},
{
"entropy": 7.506435585021973,
"epoch": 0.11290063673818825,
"grad_norm": 1.109375,
"learning_rate": 0.0004999996211041994,
"loss": 7.4922,
"mean_token_accuracy": 0.08441931083798408,
"num_tokens": 2292891.0,
"step": 1055
},
{
"entropy": 7.472446441650391,
"epoch": 0.11343571084595216,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999995476899141,
"loss": 7.5049,
"mean_token_accuracy": 0.08200813457369804,
"num_tokens": 2303764.0,
"step": 1060
},
{
"entropy": 7.319696569442749,
"epoch": 0.11397078495371608,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999994677787987,
"loss": 7.4002,
"mean_token_accuracy": 0.09418007209897042,
"num_tokens": 2314172.0,
"step": 1065
},
{
"entropy": 7.449784660339356,
"epoch": 0.11450585906148002,
"grad_norm": 1.046875,
"learning_rate": 0.0004999993813708551,
"loss": 7.3939,
"mean_token_accuracy": 0.08950763419270516,
"num_tokens": 2324848.0,
"step": 1070
},
{
"entropy": 7.320974636077881,
"epoch": 0.11504093316924394,
"grad_norm": 0.89453125,
"learning_rate": 0.000499999288466086,
"loss": 7.4332,
"mean_token_accuracy": 0.08251803517341613,
"num_tokens": 2336601.0,
"step": 1075
},
{
"entropy": 7.46982364654541,
"epoch": 0.11557600727700787,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999991890644941,
"loss": 7.4167,
"mean_token_accuracy": 0.0846226740628481,
"num_tokens": 2347281.0,
"step": 1080
},
{
"entropy": 7.453939628601074,
"epoch": 0.11611108138477179,
"grad_norm": 1.0,
"learning_rate": 0.0004999990831660822,
"loss": 7.4224,
"mean_token_accuracy": 0.0908889777958393,
"num_tokens": 2358199.0,
"step": 1085
},
{
"entropy": 7.446761226654052,
"epoch": 0.11664615549253572,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999989707708534,
"loss": 7.4877,
"mean_token_accuracy": 0.08379835970699787,
"num_tokens": 2369092.0,
"step": 1090
},
{
"entropy": 7.3978382587432865,
"epoch": 0.11718122960029964,
"grad_norm": 0.984375,
"learning_rate": 0.0004999988518788111,
"loss": 7.458,
"mean_token_accuracy": 0.08341244608163834,
"num_tokens": 2378849.0,
"step": 1095
},
{
"entropy": 7.428147649765014,
"epoch": 0.11771630370806356,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999987264899583,
"loss": 7.4385,
"mean_token_accuracy": 0.09159169495105743,
"num_tokens": 2389587.0,
"step": 1100
},
{
"entropy": 7.492716360092163,
"epoch": 0.1182513778158275,
"grad_norm": 0.97265625,
"learning_rate": 0.000499998594604299,
"loss": 7.434,
"mean_token_accuracy": 0.08649549037218093,
"num_tokens": 2400442.0,
"step": 1105
},
{
"entropy": 7.284411334991455,
"epoch": 0.11878645192359141,
"grad_norm": 0.890625,
"learning_rate": 0.000499998456221837,
"loss": 7.4154,
"mean_token_accuracy": 0.08677608072757721,
"num_tokens": 2410771.0,
"step": 1110
},
{
"entropy": 7.476975584030152,
"epoch": 0.11932152603135535,
"grad_norm": 1.0,
"learning_rate": 0.0004999983113425762,
"loss": 7.4299,
"mean_token_accuracy": 0.09064068272709846,
"num_tokens": 2421112.0,
"step": 1115
},
{
"entropy": 7.400354480743408,
"epoch": 0.11985660013911927,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999981599665207,
"loss": 7.4163,
"mean_token_accuracy": 0.08564619868993759,
"num_tokens": 2431891.0,
"step": 1120
},
{
"entropy": 7.315796279907227,
"epoch": 0.1203916742468832,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999980020936748,
"loss": 7.3479,
"mean_token_accuracy": 0.08722568973898888,
"num_tokens": 2442987.0,
"step": 1125
},
{
"entropy": 7.354978418350219,
"epoch": 0.12092674835464712,
"grad_norm": 1.09375,
"learning_rate": 0.0004999978377240434,
"loss": 7.4002,
"mean_token_accuracy": 0.08564592450857163,
"num_tokens": 2452703.0,
"step": 1130
},
{
"entropy": 7.375062036514282,
"epoch": 0.12146182246241104,
"grad_norm": 0.984375,
"learning_rate": 0.000499997666857631,
"loss": 7.4598,
"mean_token_accuracy": 0.08786297589540482,
"num_tokens": 2462975.0,
"step": 1135
},
{
"entropy": 7.394401025772095,
"epoch": 0.12199689657017497,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999974894944426,
"loss": 7.4003,
"mean_token_accuracy": 0.08598766103386879,
"num_tokens": 2473853.0,
"step": 1140
},
{
"entropy": 7.384897661209107,
"epoch": 0.12253197067793889,
"grad_norm": 0.90625,
"learning_rate": 0.0004999973056344832,
"loss": 7.3822,
"mean_token_accuracy": 0.08470982015132904,
"num_tokens": 2485633.0,
"step": 1145
},
{
"entropy": 7.337519359588623,
"epoch": 0.12306704478570282,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999971152777583,
"loss": 7.3414,
"mean_token_accuracy": 0.08668373227119446,
"num_tokens": 2496922.0,
"step": 1150
},
{
"entropy": 7.443722057342529,
"epoch": 0.12360211889346674,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999969184242733,
"loss": 7.4147,
"mean_token_accuracy": 0.0896741084754467,
"num_tokens": 2507890.0,
"step": 1155
},
{
"entropy": 7.319026279449463,
"epoch": 0.12413719300123068,
"grad_norm": 1.0390625,
"learning_rate": 0.000499996715074034,
"loss": 7.4076,
"mean_token_accuracy": 0.09130077436566353,
"num_tokens": 2519749.0,
"step": 1160
},
{
"entropy": 7.3418297290802,
"epoch": 0.1246722671089946,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999965052270461,
"loss": 7.4121,
"mean_token_accuracy": 0.09187431186437607,
"num_tokens": 2530666.0,
"step": 1165
},
{
"entropy": 7.364394998550415,
"epoch": 0.12520734121675853,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999962888833157,
"loss": 7.4155,
"mean_token_accuracy": 0.08915347754955291,
"num_tokens": 2541594.0,
"step": 1170
},
{
"entropy": 7.363469171524048,
"epoch": 0.12574241532452243,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999960660428491,
"loss": 7.3946,
"mean_token_accuracy": 0.0930194191634655,
"num_tokens": 2552330.0,
"step": 1175
},
{
"entropy": 7.449467992782592,
"epoch": 0.12627748943228637,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999958367056526,
"loss": 7.4494,
"mean_token_accuracy": 0.08116971924901009,
"num_tokens": 2564601.0,
"step": 1180
},
{
"entropy": 7.422946262359619,
"epoch": 0.1268125635400503,
"grad_norm": 0.9921875,
"learning_rate": 0.000499995600871733,
"loss": 7.4242,
"mean_token_accuracy": 0.08717223256826401,
"num_tokens": 2575539.0,
"step": 1185
},
{
"entropy": 7.284857511520386,
"epoch": 0.12734763764781423,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999953585410971,
"loss": 7.3588,
"mean_token_accuracy": 0.08616523109376431,
"num_tokens": 2586635.0,
"step": 1190
},
{
"entropy": 7.405594778060913,
"epoch": 0.12788271175557814,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999951097137518,
"loss": 7.388,
"mean_token_accuracy": 0.08991223797202111,
"num_tokens": 2596814.0,
"step": 1195
},
{
"entropy": 7.325789880752564,
"epoch": 0.12841778586334207,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999948543897044,
"loss": 7.4089,
"mean_token_accuracy": 0.08616380542516708,
"num_tokens": 2608087.0,
"step": 1200
},
{
"entropy": 7.498183393478394,
"epoch": 0.128952859971106,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999945925689621,
"loss": 7.41,
"mean_token_accuracy": 0.08784973174333573,
"num_tokens": 2619377.0,
"step": 1205
},
{
"entropy": 7.285972166061401,
"epoch": 0.12948793407886994,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999943242515325,
"loss": 7.314,
"mean_token_accuracy": 0.08880106881260871,
"num_tokens": 2629089.0,
"step": 1210
},
{
"entropy": 7.320152521133423,
"epoch": 0.13002300818663384,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999940494374236,
"loss": 7.3835,
"mean_token_accuracy": 0.08867338374257087,
"num_tokens": 2639743.0,
"step": 1215
},
{
"entropy": 7.45218596458435,
"epoch": 0.13055808229439778,
"grad_norm": 1.015625,
"learning_rate": 0.000499993768126643,
"loss": 7.4254,
"mean_token_accuracy": 0.082533248513937,
"num_tokens": 2650631.0,
"step": 1220
},
{
"entropy": 7.268480491638184,
"epoch": 0.1310931564021617,
"grad_norm": 0.96484375,
"learning_rate": 0.000499993480319199,
"loss": 7.3811,
"mean_token_accuracy": 0.0892926201224327,
"num_tokens": 2660197.0,
"step": 1225
},
{
"entropy": 7.394870710372925,
"epoch": 0.13162823050992561,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999931860150999,
"loss": 7.3556,
"mean_token_accuracy": 0.08826216235756874,
"num_tokens": 2670162.0,
"step": 1230
},
{
"entropy": 7.390807247161865,
"epoch": 0.13216330461768955,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999928852143541,
"loss": 7.3554,
"mean_token_accuracy": 0.08857011944055557,
"num_tokens": 2681233.0,
"step": 1235
},
{
"entropy": 7.200170421600342,
"epoch": 0.13269837872545348,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999925779169703,
"loss": 7.3187,
"mean_token_accuracy": 0.09283828288316727,
"num_tokens": 2692145.0,
"step": 1240
},
{
"entropy": 7.413662004470825,
"epoch": 0.13323345283321741,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999922641229575,
"loss": 7.3447,
"mean_token_accuracy": 0.09158495515584945,
"num_tokens": 2703192.0,
"step": 1245
},
{
"entropy": 7.324686527252197,
"epoch": 0.13376852694098132,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999919438323248,
"loss": 7.3597,
"mean_token_accuracy": 0.08890319019556045,
"num_tokens": 2714804.0,
"step": 1250
},
{
"entropy": 7.333380794525146,
"epoch": 0.13430360104874525,
"grad_norm": 1.046875,
"learning_rate": 0.0004999916170450812,
"loss": 7.3259,
"mean_token_accuracy": 0.09193522408604622,
"num_tokens": 2724340.0,
"step": 1255
},
{
"entropy": 7.259483861923218,
"epoch": 0.13483867515650919,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999912837612364,
"loss": 7.3341,
"mean_token_accuracy": 0.09227988123893738,
"num_tokens": 2735767.0,
"step": 1260
},
{
"entropy": 7.455391693115234,
"epoch": 0.1353737492642731,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999909439807997,
"loss": 7.3851,
"mean_token_accuracy": 0.08482631966471672,
"num_tokens": 2747658.0,
"step": 1265
},
{
"entropy": 7.264850521087647,
"epoch": 0.13590882337203702,
"grad_norm": 1.015625,
"learning_rate": 0.0004999905977037813,
"loss": 7.2465,
"mean_token_accuracy": 0.09336961582303047,
"num_tokens": 2757901.0,
"step": 1270
},
{
"entropy": 7.284139823913574,
"epoch": 0.13644389747980096,
"grad_norm": 1.015625,
"learning_rate": 0.0004999902449301909,
"loss": 7.2989,
"mean_token_accuracy": 0.0874991238117218,
"num_tokens": 2768945.0,
"step": 1275
},
{
"entropy": 7.24199595451355,
"epoch": 0.1369789715875649,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999898856600387,
"loss": 7.271,
"mean_token_accuracy": 0.103138717263937,
"num_tokens": 2780276.0,
"step": 1280
},
{
"entropy": 7.350469398498535,
"epoch": 0.1375140456953288,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999895198933354,
"loss": 7.3416,
"mean_token_accuracy": 0.09537098631262779,
"num_tokens": 2791183.0,
"step": 1285
},
{
"entropy": 7.342680644989014,
"epoch": 0.13804911980309273,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999891476300911,
"loss": 7.2607,
"mean_token_accuracy": 0.08925738111138344,
"num_tokens": 2801928.0,
"step": 1290
},
{
"entropy": 7.263091659545898,
"epoch": 0.13858419391085666,
"grad_norm": 0.890625,
"learning_rate": 0.000499988768870317,
"loss": 7.3525,
"mean_token_accuracy": 0.08583549410104752,
"num_tokens": 2813415.0,
"step": 1295
},
{
"entropy": 7.329908657073974,
"epoch": 0.13911926801862057,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999883836140236,
"loss": 7.3533,
"mean_token_accuracy": 0.08882890194654465,
"num_tokens": 2825288.0,
"step": 1300
},
{
"entropy": 7.312420797348023,
"epoch": 0.1396543421263845,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999879918612224,
"loss": 7.3068,
"mean_token_accuracy": 0.09304987639188766,
"num_tokens": 2836298.0,
"step": 1305
},
{
"entropy": 7.313601350784301,
"epoch": 0.14018941623414843,
"grad_norm": 1.03125,
"learning_rate": 0.0004999875936119245,
"loss": 7.2898,
"mean_token_accuracy": 0.0945392332971096,
"num_tokens": 2846367.0,
"step": 1310
},
{
"entropy": 7.240101766586304,
"epoch": 0.14072449034191237,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999871888661415,
"loss": 7.3047,
"mean_token_accuracy": 0.08923600688576698,
"num_tokens": 2856832.0,
"step": 1315
},
{
"entropy": 7.309039974212647,
"epoch": 0.14125956444967627,
"grad_norm": 0.9140625,
"learning_rate": 0.000499986777623885,
"loss": 7.3245,
"mean_token_accuracy": 0.08692366257309914,
"num_tokens": 2867954.0,
"step": 1320
},
{
"entropy": 7.439533376693726,
"epoch": 0.1417946385574402,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999863598851669,
"loss": 7.3611,
"mean_token_accuracy": 0.09297569468617439,
"num_tokens": 2878714.0,
"step": 1325
},
{
"entropy": 7.053972816467285,
"epoch": 0.14232971266520414,
"grad_norm": 1.3359375,
"learning_rate": 0.0004999859356499992,
"loss": 7.298,
"mean_token_accuracy": 0.10052636489272118,
"num_tokens": 2890697.0,
"step": 1330
},
{
"entropy": 7.382122421264649,
"epoch": 0.14286478677296804,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999855049183943,
"loss": 7.3219,
"mean_token_accuracy": 0.08626203685998916,
"num_tokens": 2901468.0,
"step": 1335
},
{
"entropy": 7.25765290260315,
"epoch": 0.14339986088073198,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999850676903646,
"loss": 7.2671,
"mean_token_accuracy": 0.09408968985080719,
"num_tokens": 2913466.0,
"step": 1340
},
{
"entropy": 7.231286239624024,
"epoch": 0.1439349349884959,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999846239659226,
"loss": 7.2505,
"mean_token_accuracy": 0.09706522151827812,
"num_tokens": 2923638.0,
"step": 1345
},
{
"entropy": 7.269954109191895,
"epoch": 0.14447000909625984,
"grad_norm": 1.015625,
"learning_rate": 0.0004999841737450812,
"loss": 7.3357,
"mean_token_accuracy": 0.09442631006240845,
"num_tokens": 2934790.0,
"step": 1350
},
{
"entropy": 7.268723440170288,
"epoch": 0.14500508320402375,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999837170278535,
"loss": 7.2757,
"mean_token_accuracy": 0.08723422512412071,
"num_tokens": 2946320.0,
"step": 1355
},
{
"entropy": 7.3566694259643555,
"epoch": 0.14554015731178768,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999832538142526,
"loss": 7.2369,
"mean_token_accuracy": 0.09091865047812461,
"num_tokens": 2957077.0,
"step": 1360
},
{
"entropy": 7.177463865280151,
"epoch": 0.14607523141955162,
"grad_norm": 0.953125,
"learning_rate": 0.0004999827841042917,
"loss": 7.2678,
"mean_token_accuracy": 0.08816000148653984,
"num_tokens": 2967286.0,
"step": 1365
},
{
"entropy": 7.2939177513122555,
"epoch": 0.14661030552731552,
"grad_norm": 1.34375,
"learning_rate": 0.0004999823078979846,
"loss": 7.3057,
"mean_token_accuracy": 0.09016561508178711,
"num_tokens": 2979516.0,
"step": 1370
},
{
"entropy": 7.379639339447022,
"epoch": 0.14714537963507945,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999818251953449,
"loss": 7.419,
"mean_token_accuracy": 0.08294185698032379,
"num_tokens": 2991594.0,
"step": 1375
},
{
"entropy": 7.246149206161499,
"epoch": 0.1476804537428434,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999813359963867,
"loss": 7.2972,
"mean_token_accuracy": 0.08775367885828018,
"num_tokens": 3002487.0,
"step": 1380
},
{
"entropy": 7.372300243377685,
"epoch": 0.14821552785060732,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999808403011241,
"loss": 7.3157,
"mean_token_accuracy": 0.09396110102534294,
"num_tokens": 3013197.0,
"step": 1385
},
{
"entropy": 7.142706060409546,
"epoch": 0.14875060195837123,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999803381095712,
"loss": 7.3451,
"mean_token_accuracy": 0.09063652530312538,
"num_tokens": 3024512.0,
"step": 1390
},
{
"entropy": 7.415092658996582,
"epoch": 0.14928567606613516,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999798294217428,
"loss": 7.3046,
"mean_token_accuracy": 0.08789012357592582,
"num_tokens": 3035621.0,
"step": 1395
},
{
"entropy": 7.229216432571411,
"epoch": 0.1498207501738991,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999793142376533,
"loss": 7.2876,
"mean_token_accuracy": 0.09253833740949631,
"num_tokens": 3046492.0,
"step": 1400
},
{
"entropy": 7.263148021697998,
"epoch": 0.150355824281663,
"grad_norm": 0.90625,
"learning_rate": 0.0004999787925573177,
"loss": 7.3086,
"mean_token_accuracy": 0.09204896241426468,
"num_tokens": 3057715.0,
"step": 1405
},
{
"entropy": 7.3932922840118405,
"epoch": 0.15089089838942693,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999782643807513,
"loss": 7.2654,
"mean_token_accuracy": 0.09031828343868256,
"num_tokens": 3068344.0,
"step": 1410
},
{
"entropy": 7.193799686431885,
"epoch": 0.15142597249719086,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999777297079689,
"loss": 7.2547,
"mean_token_accuracy": 0.09031877219676972,
"num_tokens": 3079576.0,
"step": 1415
},
{
"entropy": 7.342895412445069,
"epoch": 0.1519610466049548,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999771885389863,
"loss": 7.2153,
"mean_token_accuracy": 0.09486196860671044,
"num_tokens": 3090201.0,
"step": 1420
},
{
"entropy": 7.1453712463378904,
"epoch": 0.1524961207127187,
"grad_norm": 1.03125,
"learning_rate": 0.0004999766408738189,
"loss": 7.1936,
"mean_token_accuracy": 0.10047192424535752,
"num_tokens": 3099824.0,
"step": 1425
},
{
"entropy": 7.24136872291565,
"epoch": 0.15303119482048264,
"grad_norm": 0.90625,
"learning_rate": 0.0004999760867124827,
"loss": 7.2738,
"mean_token_accuracy": 0.09082015976309776,
"num_tokens": 3109948.0,
"step": 1430
},
{
"entropy": 7.299263906478882,
"epoch": 0.15356626892824657,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999755260549937,
"loss": 7.2626,
"mean_token_accuracy": 0.0974075585603714,
"num_tokens": 3121492.0,
"step": 1435
},
{
"entropy": 7.21744818687439,
"epoch": 0.15410134303601047,
"grad_norm": 1.671875,
"learning_rate": 0.0004999749589013677,
"loss": 7.0545,
"mean_token_accuracy": 0.10996845588088036,
"num_tokens": 3131990.0,
"step": 1440
},
{
"entropy": 7.1235129833221436,
"epoch": 0.1546364171437744,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999743852516217,
"loss": 7.2409,
"mean_token_accuracy": 0.0921150028705597,
"num_tokens": 3142567.0,
"step": 1445
},
{
"entropy": 7.32010293006897,
"epoch": 0.15517149125153834,
"grad_norm": 1.328125,
"learning_rate": 0.0004999738051057717,
"loss": 7.2522,
"mean_token_accuracy": 0.09435679912567138,
"num_tokens": 3153083.0,
"step": 1450
},
{
"entropy": 7.188344669342041,
"epoch": 0.15570656535930227,
"grad_norm": 0.984375,
"learning_rate": 0.0004999732184638347,
"loss": 7.2555,
"mean_token_accuracy": 0.09004457890987397,
"num_tokens": 3163985.0,
"step": 1455
},
{
"entropy": 7.262719058990479,
"epoch": 0.15624163946706618,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999726253258278,
"loss": 7.2392,
"mean_token_accuracy": 0.09183276668190957,
"num_tokens": 3175259.0,
"step": 1460
},
{
"entropy": 7.254732751846314,
"epoch": 0.1567767135748301,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999720256917678,
"loss": 7.206,
"mean_token_accuracy": 0.0977108657360077,
"num_tokens": 3185487.0,
"step": 1465
},
{
"entropy": 7.127510118484497,
"epoch": 0.15731178768259405,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999714195616723,
"loss": 7.1831,
"mean_token_accuracy": 0.09826227650046349,
"num_tokens": 3196200.0,
"step": 1470
},
{
"entropy": 7.333015441894531,
"epoch": 0.15784686179035795,
"grad_norm": 0.96875,
"learning_rate": 0.0004999708069355586,
"loss": 7.3054,
"mean_token_accuracy": 0.0843054249882698,
"num_tokens": 3207675.0,
"step": 1475
},
{
"entropy": 7.17571177482605,
"epoch": 0.15838193589812188,
"grad_norm": 1.078125,
"learning_rate": 0.0004999701878134445,
"loss": 7.1815,
"mean_token_accuracy": 0.09501941055059433,
"num_tokens": 3218117.0,
"step": 1480
},
{
"entropy": 7.304254627227783,
"epoch": 0.15891701000588582,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999695621953477,
"loss": 7.2505,
"mean_token_accuracy": 0.10208047479391098,
"num_tokens": 3229208.0,
"step": 1485
},
{
"entropy": 7.173984098434448,
"epoch": 0.15945208411364975,
"grad_norm": 0.96875,
"learning_rate": 0.0004999689300812866,
"loss": 7.1584,
"mean_token_accuracy": 0.101242895424366,
"num_tokens": 3239586.0,
"step": 1490
},
{
"entropy": 7.24401364326477,
"epoch": 0.15998715822141366,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999682914712791,
"loss": 7.2254,
"mean_token_accuracy": 0.09680869728326798,
"num_tokens": 3250103.0,
"step": 1495
},
{
"entropy": 7.228427124023438,
"epoch": 0.1605222323291776,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999676463653439,
"loss": 7.2027,
"mean_token_accuracy": 0.10232653468847275,
"num_tokens": 3261844.0,
"step": 1500
},
{
"entropy": 7.227877807617188,
"epoch": 0.16105730643694152,
"grad_norm": 1.09375,
"learning_rate": 0.0004999669947634996,
"loss": 7.1639,
"mean_token_accuracy": 0.09760255515575408,
"num_tokens": 3272073.0,
"step": 1505
},
{
"entropy": 7.178577184677124,
"epoch": 0.16159238054470546,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999663366657647,
"loss": 7.1667,
"mean_token_accuracy": 0.09620498046278954,
"num_tokens": 3281725.0,
"step": 1510
},
{
"entropy": 7.214940547943115,
"epoch": 0.16212745465246936,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999656720721586,
"loss": 7.3059,
"mean_token_accuracy": 0.094396660476923,
"num_tokens": 3292708.0,
"step": 1515
},
{
"entropy": 7.296448183059693,
"epoch": 0.1626625287602333,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999650009827004,
"loss": 7.2773,
"mean_token_accuracy": 0.09127842709422111,
"num_tokens": 3303548.0,
"step": 1520
},
{
"entropy": 7.19203143119812,
"epoch": 0.16319760286799723,
"grad_norm": 0.921875,
"learning_rate": 0.0004999643233974092,
"loss": 7.1486,
"mean_token_accuracy": 0.09975013583898544,
"num_tokens": 3314261.0,
"step": 1525
},
{
"entropy": 7.1947274684906,
"epoch": 0.16373267697576113,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999636393163049,
"loss": 7.1714,
"mean_token_accuracy": 0.09445247575640678,
"num_tokens": 3324833.0,
"step": 1530
},
{
"entropy": 7.053837537765503,
"epoch": 0.16426775108352507,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999629487394071,
"loss": 7.0568,
"mean_token_accuracy": 0.10237468853592872,
"num_tokens": 3334790.0,
"step": 1535
},
{
"entropy": 7.266930866241455,
"epoch": 0.164802825191289,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999622516667358,
"loss": 7.205,
"mean_token_accuracy": 0.09691800698637962,
"num_tokens": 3345115.0,
"step": 1540
},
{
"entropy": 7.133483982086181,
"epoch": 0.16533789929905293,
"grad_norm": 0.875,
"learning_rate": 0.000499961548098311,
"loss": 7.089,
"mean_token_accuracy": 0.10441275909543038,
"num_tokens": 3356187.0,
"step": 1545
},
{
"entropy": 7.117019701004028,
"epoch": 0.16587297340681684,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999608380341533,
"loss": 7.1282,
"mean_token_accuracy": 0.09887576475739479,
"num_tokens": 3367218.0,
"step": 1550
},
{
"entropy": 7.160897636413575,
"epoch": 0.16640804751458077,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999601214742829,
"loss": 7.27,
"mean_token_accuracy": 0.08936030119657516,
"num_tokens": 3377557.0,
"step": 1555
},
{
"entropy": 7.371627759933472,
"epoch": 0.1669431216223447,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999593984187206,
"loss": 7.2482,
"mean_token_accuracy": 0.09751093834638595,
"num_tokens": 3387402.0,
"step": 1560
},
{
"entropy": 7.0916835308074955,
"epoch": 0.1674781957301086,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999586688674872,
"loss": 7.1346,
"mean_token_accuracy": 0.09648581743240356,
"num_tokens": 3397924.0,
"step": 1565
},
{
"entropy": 7.216659879684448,
"epoch": 0.16801326983787254,
"grad_norm": 0.94921875,
"learning_rate": 0.000499957932820604,
"loss": 7.2118,
"mean_token_accuracy": 0.08759412840008736,
"num_tokens": 3408622.0,
"step": 1570
},
{
"entropy": 7.262946891784668,
"epoch": 0.16854834394563648,
"grad_norm": 0.98046875,
"learning_rate": 0.000499957190278092,
"loss": 7.1622,
"mean_token_accuracy": 0.09248490408062934,
"num_tokens": 3419692.0,
"step": 1575
},
{
"entropy": 7.098281383514404,
"epoch": 0.1690834180534004,
"grad_norm": 1.015625,
"learning_rate": 0.0004999564412399728,
"loss": 7.2154,
"mean_token_accuracy": 0.09459864124655723,
"num_tokens": 3430986.0,
"step": 1580
},
{
"entropy": 7.176830434799195,
"epoch": 0.1696184921611643,
"grad_norm": 1.0390625,
"learning_rate": 0.000499955685706268,
"loss": 7.1354,
"mean_token_accuracy": 0.09504674524068832,
"num_tokens": 3442211.0,
"step": 1585
},
{
"entropy": 7.03007402420044,
"epoch": 0.17015356626892825,
"grad_norm": 1.140625,
"learning_rate": 0.0004999549236769993,
"loss": 7.1229,
"mean_token_accuracy": 0.10297105386853218,
"num_tokens": 3453585.0,
"step": 1590
},
{
"entropy": 7.225273466110229,
"epoch": 0.17068864037669218,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999541551521888,
"loss": 7.1695,
"mean_token_accuracy": 0.0996770367026329,
"num_tokens": 3464156.0,
"step": 1595
},
{
"entropy": 7.249066162109375,
"epoch": 0.17122371448445609,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999533801318587,
"loss": 7.2238,
"mean_token_accuracy": 0.08952004536986351,
"num_tokens": 3474853.0,
"step": 1600
},
{
"entropy": 7.262774181365967,
"epoch": 0.17175878859222002,
"grad_norm": 1.03125,
"learning_rate": 0.0004999525986160313,
"loss": 7.194,
"mean_token_accuracy": 0.10073793306946754,
"num_tokens": 3486841.0,
"step": 1605
},
{
"entropy": 7.140778827667236,
"epoch": 0.17229386269998395,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999518106047293,
"loss": 7.2169,
"mean_token_accuracy": 0.09210691601037979,
"num_tokens": 3498735.0,
"step": 1610
},
{
"entropy": 7.175544404983521,
"epoch": 0.17282893680774788,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999510160979754,
"loss": 7.2352,
"mean_token_accuracy": 0.10104434639215469,
"num_tokens": 3508829.0,
"step": 1615
},
{
"entropy": 7.1884317874908445,
"epoch": 0.1733640109155118,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999502150957925,
"loss": 7.141,
"mean_token_accuracy": 0.09438429847359657,
"num_tokens": 3519465.0,
"step": 1620
},
{
"entropy": 7.188577127456665,
"epoch": 0.17389908502327572,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999494075982037,
"loss": 7.1086,
"mean_token_accuracy": 0.10255491808056831,
"num_tokens": 3530401.0,
"step": 1625
},
{
"entropy": 7.12376561164856,
"epoch": 0.17443415913103966,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999485936052324,
"loss": 7.1856,
"mean_token_accuracy": 0.09849482700228691,
"num_tokens": 3541136.0,
"step": 1630
},
{
"entropy": 7.288824605941772,
"epoch": 0.17496923323880356,
"grad_norm": 0.95703125,
"learning_rate": 0.000499947773116902,
"loss": 7.1753,
"mean_token_accuracy": 0.09568259268999099,
"num_tokens": 3552525.0,
"step": 1635
},
{
"entropy": 7.090163278579712,
"epoch": 0.1755043073465675,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999469461332365,
"loss": 7.1414,
"mean_token_accuracy": 0.09975111782550812,
"num_tokens": 3563028.0,
"step": 1640
},
{
"entropy": 7.199271249771118,
"epoch": 0.17603938145433143,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999461126542592,
"loss": 7.0912,
"mean_token_accuracy": 0.10047454982995987,
"num_tokens": 3573354.0,
"step": 1645
},
{
"entropy": 6.979637813568115,
"epoch": 0.17657445556209536,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999452726799947,
"loss": 7.0802,
"mean_token_accuracy": 0.09618928134441376,
"num_tokens": 3585576.0,
"step": 1650
},
{
"entropy": 7.297661590576172,
"epoch": 0.17710952966985927,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999444262104671,
"loss": 7.2058,
"mean_token_accuracy": 0.08774002119898797,
"num_tokens": 3596478.0,
"step": 1655
},
{
"entropy": 7.110372161865234,
"epoch": 0.1776446037776232,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999435732457007,
"loss": 7.153,
"mean_token_accuracy": 0.09257297441363335,
"num_tokens": 3608014.0,
"step": 1660
},
{
"entropy": 7.221296691894532,
"epoch": 0.17817967788538713,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999427137857203,
"loss": 7.2102,
"mean_token_accuracy": 0.09210594072937965,
"num_tokens": 3620120.0,
"step": 1665
},
{
"entropy": 7.107110977172852,
"epoch": 0.17871475199315104,
"grad_norm": 0.953125,
"learning_rate": 0.0004999418478305506,
"loss": 7.1096,
"mean_token_accuracy": 0.09546202942728996,
"num_tokens": 3632578.0,
"step": 1670
},
{
"entropy": 7.156751394271851,
"epoch": 0.17924982610091497,
"grad_norm": 0.984375,
"learning_rate": 0.0004999409753802167,
"loss": 7.131,
"mean_token_accuracy": 0.10362366437911988,
"num_tokens": 3643245.0,
"step": 1675
},
{
"entropy": 7.167398452758789,
"epoch": 0.1797849002086789,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999400964347437,
"loss": 7.1251,
"mean_token_accuracy": 0.09799086153507233,
"num_tokens": 3655832.0,
"step": 1680
},
{
"entropy": 7.151060581207275,
"epoch": 0.18031997431644284,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999392109941571,
"loss": 7.1001,
"mean_token_accuracy": 0.09830698594450951,
"num_tokens": 3666566.0,
"step": 1685
},
{
"entropy": 7.172510385513306,
"epoch": 0.18085504842420674,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999383190584822,
"loss": 7.1298,
"mean_token_accuracy": 0.10040193051099777,
"num_tokens": 3677572.0,
"step": 1690
},
{
"entropy": 7.112444877624512,
"epoch": 0.18139012253197068,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999374206277451,
"loss": 7.0549,
"mean_token_accuracy": 0.09736391827464104,
"num_tokens": 3687116.0,
"step": 1695
},
{
"entropy": 7.057082557678223,
"epoch": 0.1819251966397346,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999365157019717,
"loss": 7.1658,
"mean_token_accuracy": 0.09722192957997322,
"num_tokens": 3697961.0,
"step": 1700
},
{
"entropy": 7.233665180206299,
"epoch": 0.18246027074749852,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999356042811878,
"loss": 7.0802,
"mean_token_accuracy": 0.10210576206445694,
"num_tokens": 3708834.0,
"step": 1705
},
{
"entropy": 7.087706089019775,
"epoch": 0.18299534485526245,
"grad_norm": 0.93359375,
"learning_rate": 0.00049993468636542,
"loss": 7.1472,
"mean_token_accuracy": 0.09556594043970108,
"num_tokens": 3719835.0,
"step": 1710
},
{
"entropy": 7.116140556335449,
"epoch": 0.18353041896302638,
"grad_norm": 1.03125,
"learning_rate": 0.0004999337619546948,
"loss": 7.0339,
"mean_token_accuracy": 0.10466360598802567,
"num_tokens": 3730408.0,
"step": 1715
},
{
"entropy": 7.085261631011963,
"epoch": 0.18406549307079031,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999328310490387,
"loss": 7.1275,
"mean_token_accuracy": 0.09871943295001984,
"num_tokens": 3741257.0,
"step": 1720
},
{
"entropy": 7.190493202209472,
"epoch": 0.18460056717855422,
"grad_norm": 0.96875,
"learning_rate": 0.0004999318936484789,
"loss": 7.1175,
"mean_token_accuracy": 0.10177163109183311,
"num_tokens": 3752615.0,
"step": 1725
},
{
"entropy": 7.106344223022461,
"epoch": 0.18513564128631815,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999309497530422,
"loss": 7.1395,
"mean_token_accuracy": 0.09094236344099045,
"num_tokens": 3763656.0,
"step": 1730
},
{
"entropy": 7.083023738861084,
"epoch": 0.1856707153940821,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999299993627558,
"loss": 7.1148,
"mean_token_accuracy": 0.10153383314609528,
"num_tokens": 3775257.0,
"step": 1735
},
{
"entropy": 7.0514672756195065,
"epoch": 0.186205789501846,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999290424776475,
"loss": 7.0758,
"mean_token_accuracy": 0.1056319996714592,
"num_tokens": 3786094.0,
"step": 1740
},
{
"entropy": 7.330498886108399,
"epoch": 0.18674086360960993,
"grad_norm": 1.015625,
"learning_rate": 0.0004999280790977445,
"loss": 7.1657,
"mean_token_accuracy": 0.09170655235648155,
"num_tokens": 3796672.0,
"step": 1745
},
{
"entropy": 7.131537580490113,
"epoch": 0.18727593771737386,
"grad_norm": 0.984375,
"learning_rate": 0.000499927109223075,
"loss": 7.1508,
"mean_token_accuracy": 0.09670756980776787,
"num_tokens": 3807910.0,
"step": 1750
},
{
"entropy": 7.103532934188843,
"epoch": 0.1878110118251378,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999261328536667,
"loss": 7.1028,
"mean_token_accuracy": 0.1024494618177414,
"num_tokens": 3820792.0,
"step": 1755
},
{
"entropy": 7.02842001914978,
"epoch": 0.1883460859329017,
"grad_norm": 0.921875,
"learning_rate": 0.0004999251499895479,
"loss": 7.0234,
"mean_token_accuracy": 0.1013932503759861,
"num_tokens": 3831126.0,
"step": 1760
},
{
"entropy": 7.213843870162964,
"epoch": 0.18888116004066563,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999241606307471,
"loss": 7.1353,
"mean_token_accuracy": 0.10057510435581207,
"num_tokens": 3841923.0,
"step": 1765
},
{
"entropy": 7.119020175933838,
"epoch": 0.18941623414842956,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999231647772927,
"loss": 7.0491,
"mean_token_accuracy": 0.10542590543627739,
"num_tokens": 3852504.0,
"step": 1770
},
{
"entropy": 6.992453765869141,
"epoch": 0.1899513082561935,
"grad_norm": 0.96875,
"learning_rate": 0.0004999221624292136,
"loss": 7.0559,
"mean_token_accuracy": 0.10485710874199868,
"num_tokens": 3862761.0,
"step": 1775
},
{
"entropy": 7.134696340560913,
"epoch": 0.1904863823639574,
"grad_norm": 0.96875,
"learning_rate": 0.0004999211535865385,
"loss": 7.0556,
"mean_token_accuracy": 0.09678238332271576,
"num_tokens": 3873662.0,
"step": 1780
},
{
"entropy": 7.063473129272461,
"epoch": 0.19102145647172133,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999201382492969,
"loss": 7.0088,
"mean_token_accuracy": 0.10047155022621154,
"num_tokens": 3884835.0,
"step": 1785
},
{
"entropy": 7.110291051864624,
"epoch": 0.19155653057948527,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999191164175178,
"loss": 7.154,
"mean_token_accuracy": 0.09878607764840126,
"num_tokens": 3895073.0,
"step": 1790
},
{
"entropy": 7.209331798553467,
"epoch": 0.19209160468724917,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999180880912309,
"loss": 7.1115,
"mean_token_accuracy": 0.09884128645062447,
"num_tokens": 3905440.0,
"step": 1795
},
{
"entropy": 7.072672271728516,
"epoch": 0.1926266787950131,
"grad_norm": 1.078125,
"learning_rate": 0.0004999170532704657,
"loss": 7.1145,
"mean_token_accuracy": 0.0968818336725235,
"num_tokens": 3917135.0,
"step": 1800
},
{
"entropy": 7.13597002029419,
"epoch": 0.19316175290277704,
"grad_norm": 0.84375,
"learning_rate": 0.0004999160119552523,
"loss": 7.117,
"mean_token_accuracy": 0.10278327167034149,
"num_tokens": 3928247.0,
"step": 1805
},
{
"entropy": 7.091355848312378,
"epoch": 0.19369682701054097,
"grad_norm": 0.9375,
"learning_rate": 0.0004999149641456206,
"loss": 7.1009,
"mean_token_accuracy": 0.09872358813881874,
"num_tokens": 3938609.0,
"step": 1810
},
{
"entropy": 7.21585783958435,
"epoch": 0.19423190111830488,
"grad_norm": 1.046875,
"learning_rate": 0.000499913909841601,
"loss": 7.0861,
"mean_token_accuracy": 0.10127425193786621,
"num_tokens": 3949221.0,
"step": 1815
},
{
"entropy": 7.009944200515747,
"epoch": 0.1947669752260688,
"grad_norm": 1.015625,
"learning_rate": 0.0004999128490432238,
"loss": 7.0484,
"mean_token_accuracy": 0.09548230320215226,
"num_tokens": 3959475.0,
"step": 1820
},
{
"entropy": 7.083078098297119,
"epoch": 0.19530204933383274,
"grad_norm": 1.046875,
"learning_rate": 0.0004999117817505196,
"loss": 7.0515,
"mean_token_accuracy": 0.1084510900080204,
"num_tokens": 3969638.0,
"step": 1825
},
{
"entropy": 7.053667259216309,
"epoch": 0.19583712344159665,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999107079635194,
"loss": 7.1079,
"mean_token_accuracy": 0.10137292668223381,
"num_tokens": 3980543.0,
"step": 1830
},
{
"entropy": 7.355273485183716,
"epoch": 0.19637219754936058,
"grad_norm": 0.921875,
"learning_rate": 0.000499909627682254,
"loss": 7.162,
"mean_token_accuracy": 0.08682658635079861,
"num_tokens": 3993518.0,
"step": 1835
},
{
"entropy": 7.068727111816406,
"epoch": 0.19690727165712452,
"grad_norm": 1.046875,
"learning_rate": 0.0004999085409067549,
"loss": 7.1487,
"mean_token_accuracy": 0.09789205938577653,
"num_tokens": 4002940.0,
"step": 1840
},
{
"entropy": 7.013206481933594,
"epoch": 0.19744234576488845,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999074476370531,
"loss": 7.0792,
"mean_token_accuracy": 0.09683184549212456,
"num_tokens": 4013982.0,
"step": 1845
},
{
"entropy": 7.100980806350708,
"epoch": 0.19797741987265235,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999063478731805,
"loss": 7.0638,
"mean_token_accuracy": 0.10558338239789009,
"num_tokens": 4025607.0,
"step": 1850
},
{
"entropy": 7.0951310157775875,
"epoch": 0.1985124939804163,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999052416151687,
"loss": 7.1104,
"mean_token_accuracy": 0.10141080170869828,
"num_tokens": 4037467.0,
"step": 1855
},
{
"entropy": 7.149590158462525,
"epoch": 0.19904756808818022,
"grad_norm": 0.890625,
"learning_rate": 0.0004999041288630496,
"loss": 7.1127,
"mean_token_accuracy": 0.10126180425286294,
"num_tokens": 4049146.0,
"step": 1860
},
{
"entropy": 7.181366395950318,
"epoch": 0.19958264219594413,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999030096168553,
"loss": 7.1485,
"mean_token_accuracy": 0.09781184569001197,
"num_tokens": 4060500.0,
"step": 1865
},
{
"entropy": 7.157667875289917,
"epoch": 0.20011771630370806,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999018838766182,
"loss": 7.0987,
"mean_token_accuracy": 0.0985781691968441,
"num_tokens": 4070553.0,
"step": 1870
},
{
"entropy": 7.14162769317627,
"epoch": 0.200652790411472,
"grad_norm": 0.98046875,
"learning_rate": 0.000499900751642371,
"loss": 7.0343,
"mean_token_accuracy": 0.11338024884462357,
"num_tokens": 4079653.0,
"step": 1875
},
{
"entropy": 7.015338134765625,
"epoch": 0.20118786451923593,
"grad_norm": 0.921875,
"learning_rate": 0.0004998996129141461,
"loss": 7.123,
"mean_token_accuracy": 0.09923376441001892,
"num_tokens": 4089912.0,
"step": 1880
},
{
"entropy": 7.212585592269898,
"epoch": 0.20172293862699983,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998984676919764,
"loss": 7.0772,
"mean_token_accuracy": 0.09714810699224471,
"num_tokens": 4100577.0,
"step": 1885
},
{
"entropy": 7.073903751373291,
"epoch": 0.20225801273476376,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998973159758952,
"loss": 7.1033,
"mean_token_accuracy": 0.09901956990361213,
"num_tokens": 4111752.0,
"step": 1890
},
{
"entropy": 7.06131501197815,
"epoch": 0.2027930868425277,
"grad_norm": 1.015625,
"learning_rate": 0.0004998961577659355,
"loss": 7.0616,
"mean_token_accuracy": 0.10182834565639495,
"num_tokens": 4122616.0,
"step": 1895
},
{
"entropy": 7.192514705657959,
"epoch": 0.2033281609502916,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998949930621309,
"loss": 7.1547,
"mean_token_accuracy": 0.10196215733885765,
"num_tokens": 4134275.0,
"step": 1900
},
{
"entropy": 7.112344741821289,
"epoch": 0.20386323505805554,
"grad_norm": 1.046875,
"learning_rate": 0.0004998938218645149,
"loss": 7.0506,
"mean_token_accuracy": 0.10170819535851479,
"num_tokens": 4146412.0,
"step": 1905
},
{
"entropy": 7.029837560653687,
"epoch": 0.20439830916581947,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998926441731215,
"loss": 7.0512,
"mean_token_accuracy": 0.1002343863248825,
"num_tokens": 4157592.0,
"step": 1910
},
{
"entropy": 7.06212477684021,
"epoch": 0.2049333832735834,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998914599879845,
"loss": 7.0269,
"mean_token_accuracy": 0.09835589304566383,
"num_tokens": 4168637.0,
"step": 1915
},
{
"entropy": 7.0664630889892575,
"epoch": 0.2054684573813473,
"grad_norm": 0.96875,
"learning_rate": 0.0004998902693091382,
"loss": 7.0736,
"mean_token_accuracy": 0.09925776198506356,
"num_tokens": 4179951.0,
"step": 1920
},
{
"entropy": 7.0451087474823,
"epoch": 0.20600353148911124,
"grad_norm": 0.96875,
"learning_rate": 0.0004998890721366171,
"loss": 6.9616,
"mean_token_accuracy": 0.10666931569576263,
"num_tokens": 4189895.0,
"step": 1925
},
{
"entropy": 7.097507953643799,
"epoch": 0.20653860559687517,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998878684704556,
"loss": 7.0214,
"mean_token_accuracy": 0.10524726510047913,
"num_tokens": 4199796.0,
"step": 1930
},
{
"entropy": 7.036600160598755,
"epoch": 0.20707367970463908,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998866583106884,
"loss": 7.0461,
"mean_token_accuracy": 0.09883956611156464,
"num_tokens": 4210428.0,
"step": 1935
},
{
"entropy": 7.218980216979981,
"epoch": 0.207608753812403,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998854416573506,
"loss": 7.1246,
"mean_token_accuracy": 0.10078881680965424,
"num_tokens": 4221515.0,
"step": 1940
},
{
"entropy": 7.1100975513458256,
"epoch": 0.20814382792016695,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998842185104772,
"loss": 7.1387,
"mean_token_accuracy": 0.09548087790608406,
"num_tokens": 4232580.0,
"step": 1945
},
{
"entropy": 7.041958284378052,
"epoch": 0.20867890202793088,
"grad_norm": 0.96875,
"learning_rate": 0.0004998829888701036,
"loss": 7.0484,
"mean_token_accuracy": 0.10664152652025223,
"num_tokens": 4242871.0,
"step": 1950
},
{
"entropy": 7.147633504867554,
"epoch": 0.20921397613569478,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998817527362654,
"loss": 7.0773,
"mean_token_accuracy": 0.09752313494682312,
"num_tokens": 4253419.0,
"step": 1955
},
{
"entropy": 7.032255792617798,
"epoch": 0.20974905024345872,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998805101089982,
"loss": 7.0528,
"mean_token_accuracy": 0.10345240533351899,
"num_tokens": 4264327.0,
"step": 1960
},
{
"entropy": 7.170184564590454,
"epoch": 0.21028412435122265,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998792609883378,
"loss": 7.0428,
"mean_token_accuracy": 0.10343464836478233,
"num_tokens": 4274568.0,
"step": 1965
},
{
"entropy": 7.056762933731079,
"epoch": 0.21081919845898656,
"grad_norm": 1.171875,
"learning_rate": 0.0004998780053743203,
"loss": 7.0324,
"mean_token_accuracy": 0.10682544857263565,
"num_tokens": 4284895.0,
"step": 1970
},
{
"entropy": 7.12225079536438,
"epoch": 0.2113542725667505,
"grad_norm": 0.953125,
"learning_rate": 0.0004998767432669822,
"loss": 7.0637,
"mean_token_accuracy": 0.0994573712348938,
"num_tokens": 4297258.0,
"step": 1975
},
{
"entropy": 7.026572132110596,
"epoch": 0.21188934667451442,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998754746663595,
"loss": 6.9842,
"mean_token_accuracy": 0.10585973560810089,
"num_tokens": 4308307.0,
"step": 1980
},
{
"entropy": 7.050811910629273,
"epoch": 0.21242442078227836,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998741995724892,
"loss": 7.0491,
"mean_token_accuracy": 0.10263351649045944,
"num_tokens": 4318801.0,
"step": 1985
},
{
"entropy": 7.070459413528442,
"epoch": 0.21295949489004226,
"grad_norm": 0.984375,
"learning_rate": 0.0004998729179854079,
"loss": 7.0252,
"mean_token_accuracy": 0.10016985535621643,
"num_tokens": 4329531.0,
"step": 1990
},
{
"entropy": 7.054882478713989,
"epoch": 0.2134945689978062,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998716299051527,
"loss": 7.0253,
"mean_token_accuracy": 0.10470812693238259,
"num_tokens": 4341203.0,
"step": 1995
},
{
"entropy": 7.026881456375122,
"epoch": 0.21402964310557013,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998703353317609,
"loss": 7.0788,
"mean_token_accuracy": 0.0973646655678749,
"num_tokens": 4352426.0,
"step": 2000
},
{
"entropy": 7.11639552116394,
"epoch": 0.21456471721333403,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998690342652697,
"loss": 7.0103,
"mean_token_accuracy": 0.10563170909881592,
"num_tokens": 4363413.0,
"step": 2005
},
{
"entropy": 7.054438829421997,
"epoch": 0.21509979132109797,
"grad_norm": 0.890625,
"learning_rate": 0.0004998677267057166,
"loss": 6.951,
"mean_token_accuracy": 0.10644056648015976,
"num_tokens": 4373470.0,
"step": 2010
},
{
"entropy": 6.9533806324005125,
"epoch": 0.2156348654288619,
"grad_norm": 1.015625,
"learning_rate": 0.0004998664126531397,
"loss": 7.0164,
"mean_token_accuracy": 0.10371329113841057,
"num_tokens": 4384438.0,
"step": 2015
},
{
"entropy": 7.1470112800598145,
"epoch": 0.21616993953662583,
"grad_norm": 0.953125,
"learning_rate": 0.0004998650921075766,
"loss": 6.9978,
"mean_token_accuracy": 0.10406329333782197,
"num_tokens": 4395100.0,
"step": 2020
},
{
"entropy": 6.9813416481018065,
"epoch": 0.21670501364438974,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998637650690656,
"loss": 7.1179,
"mean_token_accuracy": 0.09519804865121842,
"num_tokens": 4407131.0,
"step": 2025
},
{
"entropy": 7.10913496017456,
"epoch": 0.21724008775215367,
"grad_norm": 1.046875,
"learning_rate": 0.000499862431537645,
"loss": 6.9569,
"mean_token_accuracy": 0.10057392492890357,
"num_tokens": 4417701.0,
"step": 2030
},
{
"entropy": 6.981591272354126,
"epoch": 0.2177751618599176,
"grad_norm": 1.109375,
"learning_rate": 0.0004998610915133533,
"loss": 7.0173,
"mean_token_accuracy": 0.10014636293053628,
"num_tokens": 4429309.0,
"step": 2035
},
{
"entropy": 7.007551097869873,
"epoch": 0.21831023596768154,
"grad_norm": 0.9375,
"learning_rate": 0.0004998597449962292,
"loss": 7.0663,
"mean_token_accuracy": 0.10122283399105073,
"num_tokens": 4440850.0,
"step": 2040
},
{
"entropy": 7.092380857467651,
"epoch": 0.21884531007544544,
"grad_norm": 0.953125,
"learning_rate": 0.0004998583919863115,
"loss": 7.0501,
"mean_token_accuracy": 0.09975898936390877,
"num_tokens": 4451995.0,
"step": 2045
},
{
"entropy": 6.99484076499939,
"epoch": 0.21938038418320938,
"grad_norm": 1.046875,
"learning_rate": 0.0004998570324836393,
"loss": 6.959,
"mean_token_accuracy": 0.10816834419965744,
"num_tokens": 4461916.0,
"step": 2050
},
{
"entropy": 7.084143257141113,
"epoch": 0.2199154582909733,
"grad_norm": 1.0546875,
"learning_rate": 0.000499855666488252,
"loss": 7.0571,
"mean_token_accuracy": 0.10147327557206154,
"num_tokens": 4471603.0,
"step": 2055
},
{
"entropy": 7.105906391143799,
"epoch": 0.22045053239873721,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998542940001889,
"loss": 7.083,
"mean_token_accuracy": 0.10265670791268348,
"num_tokens": 4483166.0,
"step": 2060
},
{
"entropy": 7.020576333999633,
"epoch": 0.22098560650650115,
"grad_norm": 1.0,
"learning_rate": 0.0004998529150194895,
"loss": 7.0022,
"mean_token_accuracy": 0.09936799854040146,
"num_tokens": 4493824.0,
"step": 2065
},
{
"entropy": 7.097088718414307,
"epoch": 0.22152068061426508,
"grad_norm": 0.953125,
"learning_rate": 0.0004998515295461939,
"loss": 6.9225,
"mean_token_accuracy": 0.1188635177910328,
"num_tokens": 4505674.0,
"step": 2070
},
{
"entropy": 6.968977689743042,
"epoch": 0.22205575472202901,
"grad_norm": 0.97265625,
"learning_rate": 0.000499850137580342,
"loss": 7.0049,
"mean_token_accuracy": 0.10718596279621125,
"num_tokens": 4516516.0,
"step": 2075
},
{
"entropy": 7.039025449752808,
"epoch": 0.22259082882979292,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998487391219739,
"loss": 6.9963,
"mean_token_accuracy": 0.09949744194746017,
"num_tokens": 4527811.0,
"step": 2080
},
{
"entropy": 7.0311308860778805,
"epoch": 0.22312590293755685,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998473341711301,
"loss": 6.9388,
"mean_token_accuracy": 0.09995686337351799,
"num_tokens": 4538988.0,
"step": 2085
},
{
"entropy": 7.018882703781128,
"epoch": 0.22366097704532079,
"grad_norm": 1.03125,
"learning_rate": 0.0004998459227278512,
"loss": 6.9733,
"mean_token_accuracy": 0.1059156596660614,
"num_tokens": 4550049.0,
"step": 2090
},
{
"entropy": 7.072498369216919,
"epoch": 0.2241960511530847,
"grad_norm": 1.0625,
"learning_rate": 0.0004998445047921777,
"loss": 7.0719,
"mean_token_accuracy": 0.10480737760663032,
"num_tokens": 4559784.0,
"step": 2095
},
{
"entropy": 7.031924390792847,
"epoch": 0.22473112526084862,
"grad_norm": 1.0625,
"learning_rate": 0.0004998430803641509,
"loss": 6.9337,
"mean_token_accuracy": 0.10361162722110748,
"num_tokens": 4569982.0,
"step": 2100
},
{
"entropy": 6.939174699783325,
"epoch": 0.22526619936861256,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998416494438118,
"loss": 6.9534,
"mean_token_accuracy": 0.10496753230690956,
"num_tokens": 4581117.0,
"step": 2105
},
{
"entropy": 7.033869123458862,
"epoch": 0.2258012734763765,
"grad_norm": 1.109375,
"learning_rate": 0.0004998402120312014,
"loss": 7.0198,
"mean_token_accuracy": 0.10706626474857331,
"num_tokens": 4592012.0,
"step": 2110
},
{
"entropy": 7.034424209594727,
"epoch": 0.2263363475841404,
"grad_norm": 0.96875,
"learning_rate": 0.0004998387681263616,
"loss": 7.0581,
"mean_token_accuracy": 0.10076023638248444,
"num_tokens": 4602591.0,
"step": 2115
},
{
"entropy": 7.038776493072509,
"epoch": 0.22687142169190433,
"grad_norm": 1.015625,
"learning_rate": 0.000499837317729334,
"loss": 6.985,
"mean_token_accuracy": 0.10343138575553894,
"num_tokens": 4613418.0,
"step": 2120
},
{
"entropy": 6.944168281555176,
"epoch": 0.22740649579966826,
"grad_norm": 0.90625,
"learning_rate": 0.0004998358608401603,
"loss": 6.9051,
"mean_token_accuracy": 0.11338085606694222,
"num_tokens": 4625156.0,
"step": 2125
},
{
"entropy": 7.1117607116699215,
"epoch": 0.22794156990743217,
"grad_norm": 0.94140625,
"learning_rate": 0.000499834397458883,
"loss": 6.937,
"mean_token_accuracy": 0.10821644887328148,
"num_tokens": 4634523.0,
"step": 2130
},
{
"entropy": 6.958322238922119,
"epoch": 0.2284766440151961,
"grad_norm": 1.0,
"learning_rate": 0.0004998329275855438,
"loss": 6.929,
"mean_token_accuracy": 0.11250298544764518,
"num_tokens": 4646292.0,
"step": 2135
},
{
"entropy": 6.985101842880249,
"epoch": 0.22901171812296003,
"grad_norm": 0.984375,
"learning_rate": 0.0004998314512201855,
"loss": 6.9099,
"mean_token_accuracy": 0.10238043665885925,
"num_tokens": 4656417.0,
"step": 2140
},
{
"entropy": 6.924033832550049,
"epoch": 0.22954679223072397,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998299683628507,
"loss": 6.9282,
"mean_token_accuracy": 0.10534189045429229,
"num_tokens": 4666933.0,
"step": 2145
},
{
"entropy": 7.059316158294678,
"epoch": 0.23008186633848787,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998284790135821,
"loss": 7.1084,
"mean_token_accuracy": 0.09782184064388275,
"num_tokens": 4678482.0,
"step": 2150
},
{
"entropy": 7.053033971786499,
"epoch": 0.2306169404462518,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998269831724228,
"loss": 6.9481,
"mean_token_accuracy": 0.1064787782728672,
"num_tokens": 4688454.0,
"step": 2155
},
{
"entropy": 6.9519795894622805,
"epoch": 0.23115201455401574,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998254808394159,
"loss": 6.9839,
"mean_token_accuracy": 0.1061179980635643,
"num_tokens": 4699545.0,
"step": 2160
},
{
"entropy": 7.083680486679077,
"epoch": 0.23168708866177964,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998239720146048,
"loss": 6.9986,
"mean_token_accuracy": 0.10272667407989503,
"num_tokens": 4711149.0,
"step": 2165
},
{
"entropy": 6.961511278152466,
"epoch": 0.23222216276954358,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998224566980332,
"loss": 6.9935,
"mean_token_accuracy": 0.10200226753950119,
"num_tokens": 4722651.0,
"step": 2170
},
{
"entropy": 7.031105852127075,
"epoch": 0.2327572368773075,
"grad_norm": 1.046875,
"learning_rate": 0.0004998209348897447,
"loss": 6.9926,
"mean_token_accuracy": 0.09458313658833503,
"num_tokens": 4733883.0,
"step": 2175
},
{
"entropy": 7.135159015655518,
"epoch": 0.23329231098507144,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998194065897833,
"loss": 6.955,
"mean_token_accuracy": 0.10691621899604797,
"num_tokens": 4744517.0,
"step": 2180
},
{
"entropy": 6.9776819229125975,
"epoch": 0.23382738509283535,
"grad_norm": 1.0390625,
"learning_rate": 0.000499817871798193,
"loss": 6.9804,
"mean_token_accuracy": 0.11214952319860458,
"num_tokens": 4755481.0,
"step": 2185
},
{
"entropy": 6.972070980072021,
"epoch": 0.23436245920059928,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998163305150185,
"loss": 6.9039,
"mean_token_accuracy": 0.11068079173564911,
"num_tokens": 4765535.0,
"step": 2190
},
{
"entropy": 6.973287200927734,
"epoch": 0.23489753330836322,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998147827403038,
"loss": 7.0273,
"mean_token_accuracy": 0.10471541583538055,
"num_tokens": 4777045.0,
"step": 2195
},
{
"entropy": 7.077732086181641,
"epoch": 0.23543260741612712,
"grad_norm": 0.9921875,
"learning_rate": 0.000499813228474094,
"loss": 6.9064,
"mean_token_accuracy": 0.10825864002108573,
"num_tokens": 4787958.0,
"step": 2200
},
{
"entropy": 6.917084503173828,
"epoch": 0.23596768152389105,
"grad_norm": 0.953125,
"learning_rate": 0.0004998116677164338,
"loss": 6.9445,
"mean_token_accuracy": 0.10730226412415504,
"num_tokens": 4798502.0,
"step": 2205
},
{
"entropy": 6.931448936462402,
"epoch": 0.236502755631655,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998101004673682,
"loss": 6.9413,
"mean_token_accuracy": 0.11088823229074478,
"num_tokens": 4809179.0,
"step": 2210
},
{
"entropy": 7.135161733627319,
"epoch": 0.23703782973941892,
"grad_norm": 1.015625,
"learning_rate": 0.0004998085267269426,
"loss": 7.0215,
"mean_token_accuracy": 0.09467339664697647,
"num_tokens": 4819803.0,
"step": 2215
},
{
"entropy": 7.004361200332641,
"epoch": 0.23757290384718283,
"grad_norm": 0.96875,
"learning_rate": 0.0004998069464952025,
"loss": 6.9916,
"mean_token_accuracy": 0.10000614672899247,
"num_tokens": 4831226.0,
"step": 2220
},
{
"entropy": 7.049906778335571,
"epoch": 0.23810797795494676,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998053597721932,
"loss": 7.0124,
"mean_token_accuracy": 0.10538152158260346,
"num_tokens": 4842168.0,
"step": 2225
},
{
"entropy": 7.019608783721924,
"epoch": 0.2386430520627107,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998037665579608,
"loss": 6.9621,
"mean_token_accuracy": 0.10858180820941925,
"num_tokens": 4852655.0,
"step": 2230
},
{
"entropy": 6.915230321884155,
"epoch": 0.2391781261704746,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998021668525513,
"loss": 6.8626,
"mean_token_accuracy": 0.10984659045934678,
"num_tokens": 4862164.0,
"step": 2235
},
{
"entropy": 7.046167945861816,
"epoch": 0.23971320027823853,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998005606560107,
"loss": 6.9964,
"mean_token_accuracy": 0.10990610420703888,
"num_tokens": 4874233.0,
"step": 2240
},
{
"entropy": 6.992388391494751,
"epoch": 0.24024827438600246,
"grad_norm": 0.953125,
"learning_rate": 0.0004997989479683856,
"loss": 6.9876,
"mean_token_accuracy": 0.1066631942987442,
"num_tokens": 4885948.0,
"step": 2245
},
{
"entropy": 6.963246202468872,
"epoch": 0.2407833484937664,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997973287897224,
"loss": 6.9667,
"mean_token_accuracy": 0.10817388147115707,
"num_tokens": 4897128.0,
"step": 2250
},
{
"entropy": 6.980781888961792,
"epoch": 0.2413184226015303,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997957031200681,
"loss": 6.939,
"mean_token_accuracy": 0.11121672838926315,
"num_tokens": 4907699.0,
"step": 2255
},
{
"entropy": 6.887354183197021,
"epoch": 0.24185349670929424,
"grad_norm": 0.984375,
"learning_rate": 0.0004997940709594692,
"loss": 6.8796,
"mean_token_accuracy": 0.10674800127744674,
"num_tokens": 4918772.0,
"step": 2260
},
{
"entropy": 7.091795778274536,
"epoch": 0.24238857081705817,
"grad_norm": 1.078125,
"learning_rate": 0.0004997924323079733,
"loss": 6.9368,
"mean_token_accuracy": 0.10729291215538979,
"num_tokens": 4930046.0,
"step": 2265
},
{
"entropy": 6.9101183891296385,
"epoch": 0.24292364492482207,
"grad_norm": 1.078125,
"learning_rate": 0.0004997907871656273,
"loss": 6.9568,
"mean_token_accuracy": 0.10801581889390946,
"num_tokens": 4940609.0,
"step": 2270
},
{
"entropy": 7.002806997299194,
"epoch": 0.243458719032586,
"grad_norm": 1.0,
"learning_rate": 0.0004997891355324791,
"loss": 6.9658,
"mean_token_accuracy": 0.10598139539361,
"num_tokens": 4951339.0,
"step": 2275
},
{
"entropy": 6.983994722366333,
"epoch": 0.24399379314034994,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997874774085761,
"loss": 6.878,
"mean_token_accuracy": 0.10262224972248077,
"num_tokens": 4961083.0,
"step": 2280
},
{
"entropy": 6.895718050003052,
"epoch": 0.24452886724811387,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997858127939662,
"loss": 6.8908,
"mean_token_accuracy": 0.10761211439967155,
"num_tokens": 4971399.0,
"step": 2285
},
{
"entropy": 7.041612005233764,
"epoch": 0.24506394135587778,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997841416886976,
"loss": 6.9715,
"mean_token_accuracy": 0.1061974436044693,
"num_tokens": 4982429.0,
"step": 2290
},
{
"entropy": 7.0298323154449465,
"epoch": 0.2455990154636417,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997824640928186,
"loss": 7.0184,
"mean_token_accuracy": 0.10754595324397087,
"num_tokens": 4993706.0,
"step": 2295
},
{
"entropy": 6.988226366043091,
"epoch": 0.24613408957140565,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997807800063774,
"loss": 6.9127,
"mean_token_accuracy": 0.10858980864286423,
"num_tokens": 5005411.0,
"step": 2300
},
{
"entropy": 7.023472309112549,
"epoch": 0.24666916367916958,
"grad_norm": 0.97265625,
"learning_rate": 0.000499779089429423,
"loss": 7.0413,
"mean_token_accuracy": 0.10208934471011162,
"num_tokens": 5016011.0,
"step": 2305
},
{
"entropy": 7.013689231872559,
"epoch": 0.24720423778693348,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997773923620037,
"loss": 6.9954,
"mean_token_accuracy": 0.10329191759228706,
"num_tokens": 5027841.0,
"step": 2310
},
{
"entropy": 7.100406980514526,
"epoch": 0.24773931189469742,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997756888041689,
"loss": 6.9853,
"mean_token_accuracy": 0.09798330888152122,
"num_tokens": 5038640.0,
"step": 2315
},
{
"entropy": 6.95166802406311,
"epoch": 0.24827438600246135,
"grad_norm": 1.0,
"learning_rate": 0.0004997739787559677,
"loss": 6.9356,
"mean_token_accuracy": 0.10357877835631371,
"num_tokens": 5050423.0,
"step": 2320
},
{
"entropy": 6.978884506225586,
"epoch": 0.24880946011022526,
"grad_norm": 1.140625,
"learning_rate": 0.0004997722622174494,
"loss": 6.954,
"mean_token_accuracy": 0.10356095060706139,
"num_tokens": 5060654.0,
"step": 2325
},
{
"entropy": 6.934212970733642,
"epoch": 0.2493445342179892,
"grad_norm": 1.015625,
"learning_rate": 0.0004997705391886636,
"loss": 6.8582,
"mean_token_accuracy": 0.11322998628020287,
"num_tokens": 5070165.0,
"step": 2330
},
{
"entropy": 6.913174200057983,
"epoch": 0.24987960832575312,
"grad_norm": 1.0,
"learning_rate": 0.0004997688096696601,
"loss": 6.8995,
"mean_token_accuracy": 0.10103722065687179,
"num_tokens": 5080937.0,
"step": 2335
},
{
"entropy": 6.911728572845459,
"epoch": 0.25041468243351706,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997670736604888,
"loss": 6.8744,
"mean_token_accuracy": 0.11288927420973778,
"num_tokens": 5090955.0,
"step": 2340
},
{
"entropy": 7.074804210662842,
"epoch": 0.250949756541281,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997653311611998,
"loss": 7.0112,
"mean_token_accuracy": 0.09920887500047684,
"num_tokens": 5102201.0,
"step": 2345
},
{
"entropy": 7.0164875984191895,
"epoch": 0.25148483064904487,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997635821718435,
"loss": 6.9312,
"mean_token_accuracy": 0.11026841551065444,
"num_tokens": 5112235.0,
"step": 2350
},
{
"entropy": 6.8925614833831785,
"epoch": 0.2520199047568088,
"grad_norm": 1.40625,
"learning_rate": 0.0004997618266924704,
"loss": 6.8219,
"mean_token_accuracy": 0.10299434512853622,
"num_tokens": 5122360.0,
"step": 2355
},
{
"entropy": 7.002241230010986,
"epoch": 0.25255497886457273,
"grad_norm": 1.015625,
"learning_rate": 0.000499760064723131,
"loss": 6.9801,
"mean_token_accuracy": 0.1024145670235157,
"num_tokens": 5133084.0,
"step": 2360
},
{
"entropy": 6.938703870773315,
"epoch": 0.25309005297233667,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997582962638764,
"loss": 6.9133,
"mean_token_accuracy": 0.09837969020009041,
"num_tokens": 5144495.0,
"step": 2365
},
{
"entropy": 7.031873512268066,
"epoch": 0.2536251270801006,
"grad_norm": 1.0,
"learning_rate": 0.0004997565213147575,
"loss": 6.9238,
"mean_token_accuracy": 0.10646962001919746,
"num_tokens": 5154824.0,
"step": 2370
},
{
"entropy": 6.939056253433227,
"epoch": 0.25416020118786453,
"grad_norm": 1.03125,
"learning_rate": 0.0004997547398758257,
"loss": 6.9111,
"mean_token_accuracy": 0.10193499475717545,
"num_tokens": 5165061.0,
"step": 2375
},
{
"entropy": 6.981441783905029,
"epoch": 0.25469527529562846,
"grad_norm": 1.0,
"learning_rate": 0.0004997529519471323,
"loss": 6.8679,
"mean_token_accuracy": 0.11417916044592857,
"num_tokens": 5174964.0,
"step": 2380
},
{
"entropy": 6.884857940673828,
"epoch": 0.25523034940339234,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997511575287291,
"loss": 6.9267,
"mean_token_accuracy": 0.10812475010752678,
"num_tokens": 5185458.0,
"step": 2385
},
{
"entropy": 6.991512250900269,
"epoch": 0.2557654235111563,
"grad_norm": 1.03125,
"learning_rate": 0.0004997493566206677,
"loss": 6.9025,
"mean_token_accuracy": 0.10954718366265297,
"num_tokens": 5196193.0,
"step": 2390
},
{
"entropy": 6.966927862167358,
"epoch": 0.2563004976189202,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997475492230003,
"loss": 7.0198,
"mean_token_accuracy": 0.09881971776485443,
"num_tokens": 5208541.0,
"step": 2395
},
{
"entropy": 6.989241456985473,
"epoch": 0.25683557172668414,
"grad_norm": 1.0234375,
"learning_rate": 0.000499745735335779,
"loss": 6.8501,
"mean_token_accuracy": 0.10620964914560319,
"num_tokens": 5218567.0,
"step": 2400
},
{
"entropy": 6.943647241592407,
"epoch": 0.2573706458344481,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997439149590561,
"loss": 6.9666,
"mean_token_accuracy": 0.10247603580355644,
"num_tokens": 5228293.0,
"step": 2405
},
{
"entropy": 6.982013940811157,
"epoch": 0.257905719942212,
"grad_norm": 1.09375,
"learning_rate": 0.0004997420880928843,
"loss": 6.913,
"mean_token_accuracy": 0.10574470162391662,
"num_tokens": 5238880.0,
"step": 2410
},
{
"entropy": 6.9576988697052,
"epoch": 0.25844079404997594,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997402547373162,
"loss": 6.9947,
"mean_token_accuracy": 0.10212295204401016,
"num_tokens": 5250104.0,
"step": 2415
},
{
"entropy": 7.100748538970947,
"epoch": 0.2589758681577399,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997384148924049,
"loss": 7.0243,
"mean_token_accuracy": 0.10289878994226456,
"num_tokens": 5260355.0,
"step": 2420
},
{
"entropy": 6.962227201461792,
"epoch": 0.25951094226550375,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997365685582035,
"loss": 6.9296,
"mean_token_accuracy": 0.10283196046948433,
"num_tokens": 5272117.0,
"step": 2425
},
{
"entropy": 6.9883177280426025,
"epoch": 0.2600460163732677,
"grad_norm": 0.890625,
"learning_rate": 0.0004997347157347651,
"loss": 7.0307,
"mean_token_accuracy": 0.10037739872932434,
"num_tokens": 5284583.0,
"step": 2430
},
{
"entropy": 7.049625587463379,
"epoch": 0.2605810904810316,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997328564221435,
"loss": 6.8241,
"mean_token_accuracy": 0.12059795558452606,
"num_tokens": 5295846.0,
"step": 2435
},
{
"entropy": 6.871410655975342,
"epoch": 0.26111616458879555,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997309906203922,
"loss": 6.9662,
"mean_token_accuracy": 0.10526571422815323,
"num_tokens": 5307802.0,
"step": 2440
},
{
"entropy": 7.033421468734741,
"epoch": 0.2616512386965595,
"grad_norm": 1.046875,
"learning_rate": 0.0004997291183295652,
"loss": 6.8913,
"mean_token_accuracy": 0.10737156346440316,
"num_tokens": 5318263.0,
"step": 2445
},
{
"entropy": 6.998748159408569,
"epoch": 0.2621863128043234,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997272395497164,
"loss": 6.9388,
"mean_token_accuracy": 0.10455406904220581,
"num_tokens": 5328555.0,
"step": 2450
},
{
"entropy": 6.912611675262451,
"epoch": 0.26272138691208735,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997253542809002,
"loss": 6.9152,
"mean_token_accuracy": 0.10682450905442238,
"num_tokens": 5339954.0,
"step": 2455
},
{
"entropy": 7.079062366485596,
"epoch": 0.26325646101985123,
"grad_norm": 0.984375,
"learning_rate": 0.0004997234625231709,
"loss": 6.9637,
"mean_token_accuracy": 0.10808332860469819,
"num_tokens": 5350919.0,
"step": 2460
},
{
"entropy": 6.93997278213501,
"epoch": 0.26379153512761516,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997215642765832,
"loss": 6.925,
"mean_token_accuracy": 0.10664721727371215,
"num_tokens": 5361844.0,
"step": 2465
},
{
"entropy": 6.987752532958984,
"epoch": 0.2643266092353791,
"grad_norm": 0.95703125,
"learning_rate": 0.000499719659541192,
"loss": 6.9562,
"mean_token_accuracy": 0.10334968268871307,
"num_tokens": 5372026.0,
"step": 2470
},
{
"entropy": 6.954167985916138,
"epoch": 0.26486168334314303,
"grad_norm": 0.890625,
"learning_rate": 0.0004997177483170521,
"loss": 6.9927,
"mean_token_accuracy": 0.10936494544148445,
"num_tokens": 5382965.0,
"step": 2475
},
{
"entropy": 7.007135391235352,
"epoch": 0.26539675745090696,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997158306042188,
"loss": 6.9458,
"mean_token_accuracy": 0.10711643844842911,
"num_tokens": 5393738.0,
"step": 2480
},
{
"entropy": 6.897781848907471,
"epoch": 0.2659318315586709,
"grad_norm": 1.0625,
"learning_rate": 0.0004997139064027476,
"loss": 6.8237,
"mean_token_accuracy": 0.1119477018713951,
"num_tokens": 5404059.0,
"step": 2485
},
{
"entropy": 6.912676382064819,
"epoch": 0.26646690566643483,
"grad_norm": 1.046875,
"learning_rate": 0.0004997119757126938,
"loss": 6.855,
"mean_token_accuracy": 0.10871710032224655,
"num_tokens": 5414152.0,
"step": 2490
},
{
"entropy": 7.022440433502197,
"epoch": 0.2670019797741987,
"grad_norm": 1.015625,
"learning_rate": 0.0004997100385341133,
"loss": 7.0235,
"mean_token_accuracy": 0.10726772770285606,
"num_tokens": 5425227.0,
"step": 2495
},
{
"entropy": 7.004488468170166,
"epoch": 0.26753705388196264,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997080948670621,
"loss": 6.8663,
"mean_token_accuracy": 0.11122208759188652,
"num_tokens": 5435716.0,
"step": 2500
},
{
"entropy": 6.936031723022461,
"epoch": 0.26807212798972657,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997061447115962,
"loss": 6.9347,
"mean_token_accuracy": 0.10375397354364395,
"num_tokens": 5446519.0,
"step": 2505
},
{
"entropy": 6.975105237960816,
"epoch": 0.2686072020974905,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997041880677719,
"loss": 6.925,
"mean_token_accuracy": 0.10548185482621193,
"num_tokens": 5457496.0,
"step": 2510
},
{
"entropy": 6.880755996704101,
"epoch": 0.26914227620525444,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997022249356457,
"loss": 6.823,
"mean_token_accuracy": 0.11008255481719971,
"num_tokens": 5469059.0,
"step": 2515
},
{
"entropy": 6.923517560958862,
"epoch": 0.26967735031301837,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997002553152745,
"loss": 6.9138,
"mean_token_accuracy": 0.10938023030757904,
"num_tokens": 5480324.0,
"step": 2520
},
{
"entropy": 6.9742035388946535,
"epoch": 0.2702124244207823,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996982792067148,
"loss": 6.9339,
"mean_token_accuracy": 0.09864719212055206,
"num_tokens": 5491513.0,
"step": 2525
},
{
"entropy": 6.8970160484313965,
"epoch": 0.2707474985285462,
"grad_norm": 0.95703125,
"learning_rate": 0.000499696296610024,
"loss": 6.8394,
"mean_token_accuracy": 0.1051232136785984,
"num_tokens": 5502414.0,
"step": 2530
},
{
"entropy": 6.9252660274505615,
"epoch": 0.2712825726363101,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996943075252592,
"loss": 6.9715,
"mean_token_accuracy": 0.1062911570072174,
"num_tokens": 5512962.0,
"step": 2535
},
{
"entropy": 6.977799654006958,
"epoch": 0.27181764674407405,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996923119524777,
"loss": 6.8355,
"mean_token_accuracy": 0.10766077339649201,
"num_tokens": 5524648.0,
"step": 2540
},
{
"entropy": 6.855292940139771,
"epoch": 0.272352720851838,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996903098917375,
"loss": 6.9214,
"mean_token_accuracy": 0.0998130515217781,
"num_tokens": 5536158.0,
"step": 2545
},
{
"entropy": 6.939780187606812,
"epoch": 0.2728877949596019,
"grad_norm": 1.0078125,
"learning_rate": 0.000499688301343096,
"loss": 6.8405,
"mean_token_accuracy": 0.10747457072138786,
"num_tokens": 5547374.0,
"step": 2550
},
{
"entropy": 7.020946025848389,
"epoch": 0.27342286906736585,
"grad_norm": 1.0,
"learning_rate": 0.0004996862863066114,
"loss": 6.9137,
"mean_token_accuracy": 0.10850636437535285,
"num_tokens": 5559212.0,
"step": 2555
},
{
"entropy": 6.907329320907593,
"epoch": 0.2739579431751298,
"grad_norm": 1.078125,
"learning_rate": 0.0004996842647823419,
"loss": 6.8823,
"mean_token_accuracy": 0.10653341263532638,
"num_tokens": 5569302.0,
"step": 2560
},
{
"entropy": 6.910869598388672,
"epoch": 0.27449301728289366,
"grad_norm": 1.140625,
"learning_rate": 0.0004996822367703458,
"loss": 6.8905,
"mean_token_accuracy": 0.10348111093044281,
"num_tokens": 5581688.0,
"step": 2565
},
{
"entropy": 6.955097198486328,
"epoch": 0.2750280913906576,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996802022706817,
"loss": 6.8211,
"mean_token_accuracy": 0.10833865478634834,
"num_tokens": 5591376.0,
"step": 2570
},
{
"entropy": 6.861729669570923,
"epoch": 0.2755631654984215,
"grad_norm": 1.046875,
"learning_rate": 0.0004996781612834083,
"loss": 6.8798,
"mean_token_accuracy": 0.10732169449329376,
"num_tokens": 5601510.0,
"step": 2575
},
{
"entropy": 7.005447006225586,
"epoch": 0.27609823960618546,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996761138085846,
"loss": 6.9316,
"mean_token_accuracy": 0.1021797627210617,
"num_tokens": 5611436.0,
"step": 2580
},
{
"entropy": 6.92574257850647,
"epoch": 0.2766333137139494,
"grad_norm": 1.21875,
"learning_rate": 0.0004996740598462697,
"loss": 6.9341,
"mean_token_accuracy": 0.10267650112509727,
"num_tokens": 5620470.0,
"step": 2585
},
{
"entropy": 6.860880899429321,
"epoch": 0.2771683878217133,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996719993965229,
"loss": 6.8297,
"mean_token_accuracy": 0.10769467577338218,
"num_tokens": 5630740.0,
"step": 2590
},
{
"entropy": 6.8753687858581545,
"epoch": 0.27770346192947726,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996699324594036,
"loss": 6.8303,
"mean_token_accuracy": 0.11734023317694664,
"num_tokens": 5641138.0,
"step": 2595
},
{
"entropy": 6.985844898223877,
"epoch": 0.27823853603724114,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996678590349717,
"loss": 6.974,
"mean_token_accuracy": 0.1085183672606945,
"num_tokens": 5652474.0,
"step": 2600
},
{
"entropy": 6.9662446022033695,
"epoch": 0.27877361014500507,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996657791232868,
"loss": 6.8709,
"mean_token_accuracy": 0.11153203919529915,
"num_tokens": 5663178.0,
"step": 2605
},
{
"entropy": 6.891348218917846,
"epoch": 0.279308684252769,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996636927244092,
"loss": 6.8963,
"mean_token_accuracy": 0.10582477301359176,
"num_tokens": 5673565.0,
"step": 2610
},
{
"entropy": 7.040777587890625,
"epoch": 0.27984375836053293,
"grad_norm": 1.1171875,
"learning_rate": 0.000499661599838399,
"loss": 6.9389,
"mean_token_accuracy": 0.10204790085554123,
"num_tokens": 5684651.0,
"step": 2615
},
{
"entropy": 6.929867219924927,
"epoch": 0.28037883246829687,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996595004653167,
"loss": 6.8204,
"mean_token_accuracy": 0.10971119105815888,
"num_tokens": 5696081.0,
"step": 2620
},
{
"entropy": 6.925084161758423,
"epoch": 0.2809139065760608,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996573946052229,
"loss": 6.855,
"mean_token_accuracy": 0.10883698239922523,
"num_tokens": 5706830.0,
"step": 2625
},
{
"entropy": 6.910267782211304,
"epoch": 0.28144898068382473,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996552822581783,
"loss": 6.8815,
"mean_token_accuracy": 0.11361035332083702,
"num_tokens": 5717261.0,
"step": 2630
},
{
"entropy": 6.91580810546875,
"epoch": 0.2819840547915886,
"grad_norm": 1.1640625,
"learning_rate": 0.000499653163424244,
"loss": 6.8481,
"mean_token_accuracy": 0.10841184929013252,
"num_tokens": 5727280.0,
"step": 2635
},
{
"entropy": 6.938942956924438,
"epoch": 0.28251912889935255,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996510381034814,
"loss": 6.9083,
"mean_token_accuracy": 0.10522805228829384,
"num_tokens": 5737485.0,
"step": 2640
},
{
"entropy": 6.926429605484008,
"epoch": 0.2830542030071165,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996489062959515,
"loss": 6.8735,
"mean_token_accuracy": 0.10467999801039696,
"num_tokens": 5747683.0,
"step": 2645
},
{
"entropy": 6.867719650268555,
"epoch": 0.2835892771148804,
"grad_norm": 1.046875,
"learning_rate": 0.0004996467680017159,
"loss": 6.9325,
"mean_token_accuracy": 0.10280923023819924,
"num_tokens": 5758148.0,
"step": 2650
},
{
"entropy": 6.917259216308594,
"epoch": 0.28412435122264434,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996446232208365,
"loss": 6.8627,
"mean_token_accuracy": 0.11170612499117852,
"num_tokens": 5769109.0,
"step": 2655
},
{
"entropy": 6.9372947216033936,
"epoch": 0.2846594253304083,
"grad_norm": 1.03125,
"learning_rate": 0.0004996424719533753,
"loss": 6.8939,
"mean_token_accuracy": 0.1052110217511654,
"num_tokens": 5780344.0,
"step": 2660
},
{
"entropy": 6.913125228881836,
"epoch": 0.2851944994381722,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996403141993941,
"loss": 6.6962,
"mean_token_accuracy": 0.11824454516172409,
"num_tokens": 5790167.0,
"step": 2665
},
{
"entropy": 6.92453293800354,
"epoch": 0.2857295735459361,
"grad_norm": 1.0625,
"learning_rate": 0.0004996381499589555,
"loss": 6.8624,
"mean_token_accuracy": 0.10613423809409142,
"num_tokens": 5800076.0,
"step": 2670
},
{
"entropy": 6.944131517410279,
"epoch": 0.2862646476537,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996359792321219,
"loss": 6.8113,
"mean_token_accuracy": 0.11141969487071038,
"num_tokens": 5810547.0,
"step": 2675
},
{
"entropy": 6.8161341667175295,
"epoch": 0.28679972176146395,
"grad_norm": 1.0625,
"learning_rate": 0.000499633802018956,
"loss": 6.8478,
"mean_token_accuracy": 0.1100471280515194,
"num_tokens": 5821210.0,
"step": 2680
},
{
"entropy": 6.917617702484131,
"epoch": 0.2873347958692279,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996316183195204,
"loss": 6.8869,
"mean_token_accuracy": 0.10786092206835747,
"num_tokens": 5832361.0,
"step": 2685
},
{
"entropy": 6.962878274917602,
"epoch": 0.2878698699769918,
"grad_norm": 1.03125,
"learning_rate": 0.0004996294281338786,
"loss": 6.8717,
"mean_token_accuracy": 0.10681624412536621,
"num_tokens": 5843242.0,
"step": 2690
},
{
"entropy": 6.976676988601684,
"epoch": 0.28840494408475575,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996272314620936,
"loss": 6.9215,
"mean_token_accuracy": 0.10194345191121101,
"num_tokens": 5855163.0,
"step": 2695
},
{
"entropy": 6.837163925170898,
"epoch": 0.2889400181925197,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996250283042288,
"loss": 6.8839,
"mean_token_accuracy": 0.11228486746549607,
"num_tokens": 5866908.0,
"step": 2700
},
{
"entropy": 6.974035596847534,
"epoch": 0.28947509230028357,
"grad_norm": 1.0546875,
"learning_rate": 0.000499622818660348,
"loss": 6.9045,
"mean_token_accuracy": 0.10939288139343262,
"num_tokens": 5877342.0,
"step": 2705
},
{
"entropy": 6.942460250854492,
"epoch": 0.2900101664080475,
"grad_norm": 1.078125,
"learning_rate": 0.0004996206025305147,
"loss": 6.8573,
"mean_token_accuracy": 0.10582418739795685,
"num_tokens": 5887960.0,
"step": 2710
},
{
"entropy": 6.87635645866394,
"epoch": 0.29054524051581143,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996183799147932,
"loss": 6.8397,
"mean_token_accuracy": 0.1095063179731369,
"num_tokens": 5899887.0,
"step": 2715
},
{
"entropy": 6.917806816101074,
"epoch": 0.29108031462357536,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996161508132475,
"loss": 6.8583,
"mean_token_accuracy": 0.10231703668832778,
"num_tokens": 5911022.0,
"step": 2720
},
{
"entropy": 7.016377210617065,
"epoch": 0.2916153887313393,
"grad_norm": 0.99609375,
"learning_rate": 0.000499613915225942,
"loss": 6.9222,
"mean_token_accuracy": 0.11102863773703575,
"num_tokens": 5922780.0,
"step": 2725
},
{
"entropy": 6.881858253479004,
"epoch": 0.29215046283910323,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996116731529412,
"loss": 6.9251,
"mean_token_accuracy": 0.10592872574925423,
"num_tokens": 5934112.0,
"step": 2730
},
{
"entropy": 6.8940752983093265,
"epoch": 0.29268553694686716,
"grad_norm": 1.0546875,
"learning_rate": 0.00049960942459431,
"loss": 6.8866,
"mean_token_accuracy": 0.1067856416106224,
"num_tokens": 5944386.0,
"step": 2735
},
{
"entropy": 6.870763492584229,
"epoch": 0.29322061105463104,
"grad_norm": 1.03125,
"learning_rate": 0.0004996071695501132,
"loss": 6.813,
"mean_token_accuracy": 0.10948670953512192,
"num_tokens": 5955552.0,
"step": 2740
},
{
"entropy": 6.921037864685059,
"epoch": 0.293755685162395,
"grad_norm": 1.0,
"learning_rate": 0.0004996049080204159,
"loss": 6.8137,
"mean_token_accuracy": 0.1098986655473709,
"num_tokens": 5966790.0,
"step": 2745
},
{
"entropy": 6.925626182556153,
"epoch": 0.2942907592701589,
"grad_norm": 1.0625,
"learning_rate": 0.0004996026400052835,
"loss": 6.8214,
"mean_token_accuracy": 0.10917853116989136,
"num_tokens": 5977704.0,
"step": 2750
},
{
"entropy": 6.853207015991211,
"epoch": 0.29482583337792284,
"grad_norm": 0.953125,
"learning_rate": 0.0004996003655047814,
"loss": 6.8364,
"mean_token_accuracy": 0.11162171512842178,
"num_tokens": 5989221.0,
"step": 2755
},
{
"entropy": 6.919997644424439,
"epoch": 0.2953609074856868,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995980845189753,
"loss": 6.871,
"mean_token_accuracy": 0.10931915193796157,
"num_tokens": 6000906.0,
"step": 2760
},
{
"entropy": 6.931192445755005,
"epoch": 0.2958959815934507,
"grad_norm": 1.015625,
"learning_rate": 0.0004995957970479312,
"loss": 6.8945,
"mean_token_accuracy": 0.10708941668272018,
"num_tokens": 6012039.0,
"step": 2765
},
{
"entropy": 6.8547321319580075,
"epoch": 0.29643105570121464,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995935030917148,
"loss": 6.776,
"mean_token_accuracy": 0.11066783592104912,
"num_tokens": 6024383.0,
"step": 2770
},
{
"entropy": 6.994161558151245,
"epoch": 0.2969661298089785,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995912026503926,
"loss": 6.8508,
"mean_token_accuracy": 0.10267031714320182,
"num_tokens": 6034904.0,
"step": 2775
},
{
"entropy": 6.838885879516601,
"epoch": 0.29750120391674245,
"grad_norm": 1.0234375,
"learning_rate": 0.000499588895724031,
"loss": 6.8488,
"mean_token_accuracy": 0.10749915912747383,
"num_tokens": 6044449.0,
"step": 2780
},
{
"entropy": 6.962449741363526,
"epoch": 0.2980362780245064,
"grad_norm": 1.109375,
"learning_rate": 0.0004995865823126968,
"loss": 6.8765,
"mean_token_accuracy": 0.10825972631573677,
"num_tokens": 6054465.0,
"step": 2785
},
{
"entropy": 6.8947694301605225,
"epoch": 0.2985713521322703,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995842624164564,
"loss": 6.8962,
"mean_token_accuracy": 0.10824007838964463,
"num_tokens": 6065539.0,
"step": 2790
},
{
"entropy": 6.909287405014038,
"epoch": 0.29910642624003425,
"grad_norm": 1.0859375,
"learning_rate": 0.000499581936035377,
"loss": 6.8459,
"mean_token_accuracy": 0.10840294510126114,
"num_tokens": 6076135.0,
"step": 2795
},
{
"entropy": 6.903677034378052,
"epoch": 0.2996415003477982,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995796031695257,
"loss": 6.8641,
"mean_token_accuracy": 0.10931005999445915,
"num_tokens": 6086604.0,
"step": 2800
},
{
"entropy": 6.90281114578247,
"epoch": 0.3001765744555621,
"grad_norm": 1.125,
"learning_rate": 0.00049957726381897,
"loss": 6.8781,
"mean_token_accuracy": 0.10950748696923256,
"num_tokens": 6095672.0,
"step": 2805
},
{
"entropy": 6.90290675163269,
"epoch": 0.300711648563326,
"grad_norm": 1.0,
"learning_rate": 0.0004995749179837773,
"loss": 6.7498,
"mean_token_accuracy": 0.11466853022575378,
"num_tokens": 6105652.0,
"step": 2810
},
{
"entropy": 6.826155042648315,
"epoch": 0.30124672267108993,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995725656640155,
"loss": 6.7856,
"mean_token_accuracy": 0.11099454537034034,
"num_tokens": 6116930.0,
"step": 2815
},
{
"entropy": 6.844819593429565,
"epoch": 0.30178179677885386,
"grad_norm": 1.03125,
"learning_rate": 0.0004995702068597524,
"loss": 6.7894,
"mean_token_accuracy": 0.11423054486513137,
"num_tokens": 6126659.0,
"step": 2820
},
{
"entropy": 6.890478515625,
"epoch": 0.3023168708866178,
"grad_norm": 1.015625,
"learning_rate": 0.0004995678415710561,
"loss": 6.8288,
"mean_token_accuracy": 0.11060848459601402,
"num_tokens": 6137227.0,
"step": 2825
},
{
"entropy": 6.962517929077149,
"epoch": 0.3028519449943817,
"grad_norm": 0.98046875,
"learning_rate": 0.000499565469797995,
"loss": 6.8821,
"mean_token_accuracy": 0.10931277051568031,
"num_tokens": 6147800.0,
"step": 2830
},
{
"entropy": 6.822924327850342,
"epoch": 0.30338701910214566,
"grad_norm": 1.0,
"learning_rate": 0.0004995630915406374,
"loss": 6.7326,
"mean_token_accuracy": 0.12161283493041992,
"num_tokens": 6157671.0,
"step": 2835
},
{
"entropy": 6.918770694732666,
"epoch": 0.3039220932099096,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995607067990521,
"loss": 6.8698,
"mean_token_accuracy": 0.10645673274993897,
"num_tokens": 6168651.0,
"step": 2840
},
{
"entropy": 6.873465251922608,
"epoch": 0.30445716731767347,
"grad_norm": 1.453125,
"learning_rate": 0.000499558315573308,
"loss": 6.8487,
"mean_token_accuracy": 0.1086043007671833,
"num_tokens": 6179925.0,
"step": 2845
},
{
"entropy": 6.818239831924439,
"epoch": 0.3049922414254374,
"grad_norm": 1.234375,
"learning_rate": 0.0004995559178634741,
"loss": 6.8438,
"mean_token_accuracy": 0.10887812077999115,
"num_tokens": 6190541.0,
"step": 2850
},
{
"entropy": 7.041949510574341,
"epoch": 0.30552731553320134,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995535136696196,
"loss": 6.8866,
"mean_token_accuracy": 0.09955592602491378,
"num_tokens": 6201491.0,
"step": 2855
},
{
"entropy": 6.8139080047607425,
"epoch": 0.30606238964096527,
"grad_norm": 1.0078125,
"learning_rate": 0.000499551102991814,
"loss": 6.6991,
"mean_token_accuracy": 0.11249880269169807,
"num_tokens": 6212443.0,
"step": 2860
},
{
"entropy": 6.817849111557007,
"epoch": 0.3065974637487292,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995486858301268,
"loss": 6.8981,
"mean_token_accuracy": 0.1030767761170864,
"num_tokens": 6222623.0,
"step": 2865
},
{
"entropy": 6.925403165817261,
"epoch": 0.30713253785649314,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995462621846279,
"loss": 6.7852,
"mean_token_accuracy": 0.10933116525411606,
"num_tokens": 6233712.0,
"step": 2870
},
{
"entropy": 6.854152202606201,
"epoch": 0.30766761196425707,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995438320553871,
"loss": 6.8534,
"mean_token_accuracy": 0.10412434935569763,
"num_tokens": 6244254.0,
"step": 2875
},
{
"entropy": 6.823256492614746,
"epoch": 0.30820268607202095,
"grad_norm": 1.2265625,
"learning_rate": 0.0004995413954424748,
"loss": 6.7504,
"mean_token_accuracy": 0.116096231341362,
"num_tokens": 6254202.0,
"step": 2880
},
{
"entropy": 6.877709770202637,
"epoch": 0.3087377601797849,
"grad_norm": 1.140625,
"learning_rate": 0.0004995389523459612,
"loss": 6.8231,
"mean_token_accuracy": 0.10538254305720329,
"num_tokens": 6264496.0,
"step": 2885
},
{
"entropy": 6.839356851577759,
"epoch": 0.3092728342875488,
"grad_norm": 1.078125,
"learning_rate": 0.000499536502765917,
"loss": 6.828,
"mean_token_accuracy": 0.1125398851931095,
"num_tokens": 6275835.0,
"step": 2890
},
{
"entropy": 6.893553924560547,
"epoch": 0.30980790839531275,
"grad_norm": 1.2421875,
"learning_rate": 0.0004995340467024128,
"loss": 6.7842,
"mean_token_accuracy": 0.10834744572639465,
"num_tokens": 6286805.0,
"step": 2895
},
{
"entropy": 6.906386661529541,
"epoch": 0.3103429825030767,
"grad_norm": 0.984375,
"learning_rate": 0.0004995315841555195,
"loss": 6.7916,
"mean_token_accuracy": 0.10366136953234673,
"num_tokens": 6298444.0,
"step": 2900
},
{
"entropy": 6.783759641647339,
"epoch": 0.3108780566108406,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995291151253083,
"loss": 6.7348,
"mean_token_accuracy": 0.11380124613642692,
"num_tokens": 6308259.0,
"step": 2905
},
{
"entropy": 6.846135902404785,
"epoch": 0.31141313071860455,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995266396118505,
"loss": 6.8013,
"mean_token_accuracy": 0.10161566883325576,
"num_tokens": 6320058.0,
"step": 2910
},
{
"entropy": 6.900366973876953,
"epoch": 0.3119482048263684,
"grad_norm": 0.9140625,
"learning_rate": 0.0004995241576152174,
"loss": 6.7723,
"mean_token_accuracy": 0.10676236972212791,
"num_tokens": 6331203.0,
"step": 2915
},
{
"entropy": 6.897499227523804,
"epoch": 0.31248327893413236,
"grad_norm": 1.078125,
"learning_rate": 0.000499521669135481,
"loss": 6.8882,
"mean_token_accuracy": 0.10588050186634064,
"num_tokens": 6342027.0,
"step": 2920
},
{
"entropy": 6.894981956481933,
"epoch": 0.3130183530418963,
"grad_norm": 1.015625,
"learning_rate": 0.0004995191741727128,
"loss": 6.7873,
"mean_token_accuracy": 0.10809844583272935,
"num_tokens": 6352580.0,
"step": 2925
},
{
"entropy": 6.887385463714599,
"epoch": 0.3135534271496602,
"grad_norm": 0.94140625,
"learning_rate": 0.000499516672726985,
"loss": 6.8106,
"mean_token_accuracy": 0.11217835023999215,
"num_tokens": 6363997.0,
"step": 2930
},
{
"entropy": 6.83791708946228,
"epoch": 0.31408850125742416,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995141647983699,
"loss": 6.8464,
"mean_token_accuracy": 0.11019290536642075,
"num_tokens": 6375284.0,
"step": 2935
},
{
"entropy": 6.8710222244262695,
"epoch": 0.3146235753651881,
"grad_norm": 0.984375,
"learning_rate": 0.0004995116503869398,
"loss": 6.7437,
"mean_token_accuracy": 0.11758178621530532,
"num_tokens": 6386698.0,
"step": 2940
},
{
"entropy": 6.864429473876953,
"epoch": 0.315158649472952,
"grad_norm": 1.3203125,
"learning_rate": 0.0004995091294927673,
"loss": 6.818,
"mean_token_accuracy": 0.10886727124452591,
"num_tokens": 6397237.0,
"step": 2945
},
{
"entropy": 6.973788785934448,
"epoch": 0.3156937235807159,
"grad_norm": 1.0,
"learning_rate": 0.0004995066021159253,
"loss": 6.9364,
"mean_token_accuracy": 0.10483678579330444,
"num_tokens": 6408266.0,
"step": 2950
},
{
"entropy": 6.842577266693115,
"epoch": 0.31622879768847983,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995040682564867,
"loss": 6.8199,
"mean_token_accuracy": 0.11405593305826187,
"num_tokens": 6419563.0,
"step": 2955
},
{
"entropy": 6.8359404563903805,
"epoch": 0.31676387179624377,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995015279145247,
"loss": 6.7863,
"mean_token_accuracy": 0.11180907115340233,
"num_tokens": 6430220.0,
"step": 2960
},
{
"entropy": 6.857833385467529,
"epoch": 0.3172989459040077,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994989810901126,
"loss": 6.7961,
"mean_token_accuracy": 0.10685641467571258,
"num_tokens": 6440157.0,
"step": 2965
},
{
"entropy": 6.856301593780517,
"epoch": 0.31783402001177163,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994964277833239,
"loss": 6.8463,
"mean_token_accuracy": 0.11110184714198112,
"num_tokens": 6449511.0,
"step": 2970
},
{
"entropy": 6.923713493347168,
"epoch": 0.31836909411953557,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994938679942324,
"loss": 6.8365,
"mean_token_accuracy": 0.10282818749547004,
"num_tokens": 6460608.0,
"step": 2975
},
{
"entropy": 6.919619989395142,
"epoch": 0.3189041682272995,
"grad_norm": 1.046875,
"learning_rate": 0.0004994913017229121,
"loss": 6.8372,
"mean_token_accuracy": 0.1131968080997467,
"num_tokens": 6472737.0,
"step": 2980
},
{
"entropy": 6.817192554473877,
"epoch": 0.31943924233506343,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994887289694371,
"loss": 6.7184,
"mean_token_accuracy": 0.11385754495859146,
"num_tokens": 6482993.0,
"step": 2985
},
{
"entropy": 6.960534572601318,
"epoch": 0.3199743164428273,
"grad_norm": 1.078125,
"learning_rate": 0.0004994861497338814,
"loss": 6.9207,
"mean_token_accuracy": 0.11038238108158112,
"num_tokens": 6494530.0,
"step": 2990
},
{
"entropy": 6.894915056228638,
"epoch": 0.32050939055059124,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994835640163198,
"loss": 6.7626,
"mean_token_accuracy": 0.10633266866207122,
"num_tokens": 6504691.0,
"step": 2995
},
{
"entropy": 6.885722494125366,
"epoch": 0.3210444646583552,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994809718168267,
"loss": 6.7865,
"mean_token_accuracy": 0.10957603380084038,
"num_tokens": 6514924.0,
"step": 3000
},
{
"epoch": 0.3210444646583552,
"eval_entropy": 6.6982880289021205,
"eval_loss": 6.9056596755981445,
"eval_mean_token_accuracy": 0.11213191206946045,
"eval_num_tokens": 6514924.0,
"eval_runtime": 22.6381,
"eval_samples_per_second": 1311.022,
"eval_steps_per_second": 163.883,
"step": 3000
},
{
"entropy": 6.774571943283081,
"epoch": 0.3215795387661191,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994783731354771,
"loss": 6.7524,
"mean_token_accuracy": 0.11166785731911659,
"num_tokens": 6526097.0,
"step": 3005
},
{
"entropy": 6.929418516159058,
"epoch": 0.32211461287388304,
"grad_norm": 1.0625,
"learning_rate": 0.0004994757679723461,
"loss": 6.8149,
"mean_token_accuracy": 0.10639956593513489,
"num_tokens": 6536384.0,
"step": 3010
},
{
"entropy": 6.919300746917725,
"epoch": 0.322649686981647,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994731563275088,
"loss": 6.8694,
"mean_token_accuracy": 0.1046457588672638,
"num_tokens": 6547161.0,
"step": 3015
},
{
"entropy": 6.785147857666016,
"epoch": 0.3231847610894109,
"grad_norm": 1.046875,
"learning_rate": 0.0004994705382010406,
"loss": 6.7936,
"mean_token_accuracy": 0.10720081478357316,
"num_tokens": 6556751.0,
"step": 3020
},
{
"entropy": 6.861992263793946,
"epoch": 0.3237198351971748,
"grad_norm": 1.109375,
"learning_rate": 0.0004994679135930174,
"loss": 6.8486,
"mean_token_accuracy": 0.1134506493806839,
"num_tokens": 6567655.0,
"step": 3025
},
{
"entropy": 6.987712669372558,
"epoch": 0.3242549093049387,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994652825035144,
"loss": 6.8797,
"mean_token_accuracy": 0.10770419910550118,
"num_tokens": 6579127.0,
"step": 3030
},
{
"entropy": 6.9072469711303714,
"epoch": 0.32478998341270265,
"grad_norm": 1.265625,
"learning_rate": 0.0004994626449326081,
"loss": 6.8376,
"mean_token_accuracy": 0.11200311556458473,
"num_tokens": 6590779.0,
"step": 3035
},
{
"entropy": 6.858027362823487,
"epoch": 0.3253250575204666,
"grad_norm": 1.09375,
"learning_rate": 0.0004994600008803745,
"loss": 6.7456,
"mean_token_accuracy": 0.11969574466347695,
"num_tokens": 6600849.0,
"step": 3040
},
{
"entropy": 6.831806564331055,
"epoch": 0.3258601316282305,
"grad_norm": 1.09375,
"learning_rate": 0.0004994573503468899,
"loss": 6.7547,
"mean_token_accuracy": 0.11736322343349456,
"num_tokens": 6610633.0,
"step": 3045
},
{
"entropy": 6.787566423416138,
"epoch": 0.32639520573599445,
"grad_norm": 0.953125,
"learning_rate": 0.0004994546933322308,
"loss": 6.7953,
"mean_token_accuracy": 0.10811478719115257,
"num_tokens": 6621327.0,
"step": 3050
},
{
"entropy": 6.93750228881836,
"epoch": 0.3269302798437584,
"grad_norm": 0.984375,
"learning_rate": 0.000499452029836474,
"loss": 6.7415,
"mean_token_accuracy": 0.11278096809983254,
"num_tokens": 6632267.0,
"step": 3055
},
{
"entropy": 6.810693454742432,
"epoch": 0.32746535395152226,
"grad_norm": 1.1796875,
"learning_rate": 0.0004994493598596963,
"loss": 6.7705,
"mean_token_accuracy": 0.10761533975601197,
"num_tokens": 6642714.0,
"step": 3060
},
{
"entropy": 6.849345254898071,
"epoch": 0.3280004280592862,
"grad_norm": 1.078125,
"learning_rate": 0.0004994466834019749,
"loss": 6.7758,
"mean_token_accuracy": 0.11640546843409538,
"num_tokens": 6652353.0,
"step": 3065
},
{
"entropy": 6.842514944076538,
"epoch": 0.32853550216705013,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994440004633871,
"loss": 6.7517,
"mean_token_accuracy": 0.11130421757698059,
"num_tokens": 6663133.0,
"step": 3070
},
{
"entropy": 6.801913261413574,
"epoch": 0.32907057627481406,
"grad_norm": 1.1953125,
"learning_rate": 0.0004994413110440103,
"loss": 6.7354,
"mean_token_accuracy": 0.11714180111885071,
"num_tokens": 6673984.0,
"step": 3075
},
{
"entropy": 6.8637457370758055,
"epoch": 0.329605650382578,
"grad_norm": 0.875,
"learning_rate": 0.0004994386151439222,
"loss": 6.8076,
"mean_token_accuracy": 0.1080280676484108,
"num_tokens": 6686171.0,
"step": 3080
},
{
"entropy": 6.933955192565918,
"epoch": 0.33014072449034193,
"grad_norm": 1.109375,
"learning_rate": 0.0004994359127632006,
"loss": 6.8475,
"mean_token_accuracy": 0.10514529719948769,
"num_tokens": 6696609.0,
"step": 3085
},
{
"entropy": 6.89633002281189,
"epoch": 0.33067579859810586,
"grad_norm": 1.1875,
"learning_rate": 0.0004994332039019235,
"loss": 6.7854,
"mean_token_accuracy": 0.11049062609672547,
"num_tokens": 6706188.0,
"step": 3090
},
{
"entropy": 6.708978939056396,
"epoch": 0.33121087270586974,
"grad_norm": 1.125,
"learning_rate": 0.0004994304885601694,
"loss": 6.7667,
"mean_token_accuracy": 0.11125922203063965,
"num_tokens": 6716276.0,
"step": 3095
},
{
"entropy": 6.774325132369995,
"epoch": 0.3317459468136337,
"grad_norm": 1.2890625,
"learning_rate": 0.0004994277667380162,
"loss": 6.737,
"mean_token_accuracy": 0.1192335270345211,
"num_tokens": 6727176.0,
"step": 3100
},
{
"entropy": 6.887723302841186,
"epoch": 0.3322810209213976,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994250384355429,
"loss": 6.8119,
"mean_token_accuracy": 0.11386096626520156,
"num_tokens": 6737768.0,
"step": 3105
},
{
"entropy": 6.840688323974609,
"epoch": 0.33281609502916154,
"grad_norm": 6.46875,
"learning_rate": 0.0004994223036528282,
"loss": 6.7618,
"mean_token_accuracy": 0.11079908534884453,
"num_tokens": 6747962.0,
"step": 3110
},
{
"entropy": 6.8406102657318115,
"epoch": 0.3333511691369255,
"grad_norm": 1.1171875,
"learning_rate": 0.000499419562389951,
"loss": 6.7548,
"mean_token_accuracy": 0.1147709958255291,
"num_tokens": 6758400.0,
"step": 3115
},
{
"entropy": 6.851141119003296,
"epoch": 0.3338862432446894,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994168146469904,
"loss": 6.7693,
"mean_token_accuracy": 0.11088330820202827,
"num_tokens": 6768855.0,
"step": 3120
},
{
"entropy": 6.863172101974487,
"epoch": 0.33442131735245334,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994140604240259,
"loss": 6.7514,
"mean_token_accuracy": 0.1078583225607872,
"num_tokens": 6778656.0,
"step": 3125
},
{
"entropy": 6.882100915908813,
"epoch": 0.3349563914602172,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994112997211369,
"loss": 6.8182,
"mean_token_accuracy": 0.11499066054821014,
"num_tokens": 6789960.0,
"step": 3130
},
{
"entropy": 6.730247640609742,
"epoch": 0.33549146556798115,
"grad_norm": 1.1875,
"learning_rate": 0.0004994085325384031,
"loss": 6.8076,
"mean_token_accuracy": 0.10851202681660652,
"num_tokens": 6800188.0,
"step": 3135
},
{
"entropy": 6.906764650344849,
"epoch": 0.3360265396757451,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994057588759045,
"loss": 6.7393,
"mean_token_accuracy": 0.11358117088675498,
"num_tokens": 6811032.0,
"step": 3140
},
{
"entropy": 6.824139022827149,
"epoch": 0.336561613783509,
"grad_norm": 1.0625,
"learning_rate": 0.0004994029787337212,
"loss": 6.7666,
"mean_token_accuracy": 0.10829171538352966,
"num_tokens": 6823060.0,
"step": 3145
},
{
"entropy": 6.8112482070922855,
"epoch": 0.33709668789127295,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994001921119333,
"loss": 6.8496,
"mean_token_accuracy": 0.10913093164563178,
"num_tokens": 6833559.0,
"step": 3150
},
{
"entropy": 6.894295167922974,
"epoch": 0.3376317619990369,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993973990106215,
"loss": 6.7974,
"mean_token_accuracy": 0.11234046593308449,
"num_tokens": 6843657.0,
"step": 3155
},
{
"entropy": 6.814738702774048,
"epoch": 0.3381668361068008,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993945994298662,
"loss": 6.7267,
"mean_token_accuracy": 0.10636094883084297,
"num_tokens": 6853615.0,
"step": 3160
},
{
"entropy": 6.858942174911499,
"epoch": 0.3387019102145647,
"grad_norm": 1.03125,
"learning_rate": 0.0004993917933697484,
"loss": 6.8354,
"mean_token_accuracy": 0.10942788496613502,
"num_tokens": 6865256.0,
"step": 3165
},
{
"entropy": 7.015973472595215,
"epoch": 0.3392369843223286,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993889808303493,
"loss": 6.8711,
"mean_token_accuracy": 0.11036419197916984,
"num_tokens": 6875358.0,
"step": 3170
},
{
"entropy": 6.757581901550293,
"epoch": 0.33977205843009256,
"grad_norm": 1.015625,
"learning_rate": 0.0004993861618117498,
"loss": 6.7711,
"mean_token_accuracy": 0.1102716788649559,
"num_tokens": 6886584.0,
"step": 3175
},
{
"entropy": 6.8233020305633545,
"epoch": 0.3403071325378565,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993833363140314,
"loss": 6.7109,
"mean_token_accuracy": 0.11896613985300064,
"num_tokens": 6896958.0,
"step": 3180
},
{
"entropy": 6.830180215835571,
"epoch": 0.3408422066456204,
"grad_norm": 1.03125,
"learning_rate": 0.0004993805043372756,
"loss": 6.775,
"mean_token_accuracy": 0.1163598895072937,
"num_tokens": 6907944.0,
"step": 3185
},
{
"entropy": 6.831008625030518,
"epoch": 0.34137728075338436,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993776658815644,
"loss": 6.8739,
"mean_token_accuracy": 0.1049034409224987,
"num_tokens": 6919486.0,
"step": 3190
},
{
"entropy": 6.957132577896118,
"epoch": 0.3419123548611483,
"grad_norm": 1.2578125,
"learning_rate": 0.0004993748209469796,
"loss": 6.8414,
"mean_token_accuracy": 0.11248302012681961,
"num_tokens": 6930650.0,
"step": 3195
},
{
"entropy": 6.833132362365722,
"epoch": 0.34244742896891217,
"grad_norm": 1.015625,
"learning_rate": 0.0004993719695336033,
"loss": 6.7926,
"mean_token_accuracy": 0.11160960346460343,
"num_tokens": 6941519.0,
"step": 3200
},
{
"entropy": 6.775629329681396,
"epoch": 0.3429825030766761,
"grad_norm": 1.5078125,
"learning_rate": 0.0004993691116415179,
"loss": 6.8265,
"mean_token_accuracy": 0.10980288833379745,
"num_tokens": 6952952.0,
"step": 3205
},
{
"entropy": 6.945135641098022,
"epoch": 0.34351757718444004,
"grad_norm": 1.0078125,
"learning_rate": 0.000499366247270806,
"loss": 6.7907,
"mean_token_accuracy": 0.11398276537656785,
"num_tokens": 6962867.0,
"step": 3210
},
{
"entropy": 6.850128173828125,
"epoch": 0.34405265129220397,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993633764215502,
"loss": 6.7128,
"mean_token_accuracy": 0.1149346224963665,
"num_tokens": 6973434.0,
"step": 3215
},
{
"entropy": 6.805971765518189,
"epoch": 0.3445877253999679,
"grad_norm": 1.125,
"learning_rate": 0.0004993604990938336,
"loss": 6.7242,
"mean_token_accuracy": 0.12372536882758141,
"num_tokens": 6984161.0,
"step": 3220
},
{
"entropy": 6.831466388702393,
"epoch": 0.34512279950773184,
"grad_norm": 1.078125,
"learning_rate": 0.0004993576152877389,
"loss": 6.7023,
"mean_token_accuracy": 0.11155548840761184,
"num_tokens": 6994700.0,
"step": 3225
},
{
"entropy": 6.859937858581543,
"epoch": 0.34565787361549577,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993547250033496,
"loss": 6.8702,
"mean_token_accuracy": 0.10222596898674965,
"num_tokens": 7006497.0,
"step": 3230
},
{
"entropy": 6.907996559143067,
"epoch": 0.34619294772325965,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993518282407493,
"loss": 6.7762,
"mean_token_accuracy": 0.11640164703130722,
"num_tokens": 7016415.0,
"step": 3235
},
{
"entropy": 6.780554246902466,
"epoch": 0.3467280218310236,
"grad_norm": 1.2109375,
"learning_rate": 0.0004993489250000213,
"loss": 6.7293,
"mean_token_accuracy": 0.11351820677518845,
"num_tokens": 7026610.0,
"step": 3240
},
{
"entropy": 6.797022676467895,
"epoch": 0.3472630959387875,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993460152812497,
"loss": 6.7409,
"mean_token_accuracy": 0.11043249741196633,
"num_tokens": 7038236.0,
"step": 3245
},
{
"entropy": 6.871045017242432,
"epoch": 0.34779817004655145,
"grad_norm": 1.140625,
"learning_rate": 0.0004993430990845184,
"loss": 6.7892,
"mean_token_accuracy": 0.12099697217345237,
"num_tokens": 7049438.0,
"step": 3250
},
{
"entropy": 6.795158195495605,
"epoch": 0.3483332441543154,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993401764099118,
"loss": 6.75,
"mean_token_accuracy": 0.10678878873586654,
"num_tokens": 7061001.0,
"step": 3255
},
{
"entropy": 6.869133567810058,
"epoch": 0.3488683182620793,
"grad_norm": 1.09375,
"learning_rate": 0.0004993372472575139,
"loss": 6.7464,
"mean_token_accuracy": 0.11511967554688454,
"num_tokens": 7071328.0,
"step": 3260
},
{
"entropy": 6.9016501903533936,
"epoch": 0.34940339236984325,
"grad_norm": 1.140625,
"learning_rate": 0.0004993343116274096,
"loss": 6.93,
"mean_token_accuracy": 0.10523641705513001,
"num_tokens": 7082597.0,
"step": 3265
},
{
"entropy": 6.860869073867798,
"epoch": 0.3499384664776071,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993313695196836,
"loss": 6.759,
"mean_token_accuracy": 0.116279286891222,
"num_tokens": 7093463.0,
"step": 3270
},
{
"entropy": 6.816839218139648,
"epoch": 0.35047354058537106,
"grad_norm": 0.984375,
"learning_rate": 0.0004993284209344208,
"loss": 6.8225,
"mean_token_accuracy": 0.11397869065403939,
"num_tokens": 7104350.0,
"step": 3275
},
{
"entropy": 6.929663181304932,
"epoch": 0.351008614693135,
"grad_norm": 1.2265625,
"learning_rate": 0.0004993254658717065,
"loss": 6.78,
"mean_token_accuracy": 0.11524501666426659,
"num_tokens": 7114851.0,
"step": 3280
},
{
"entropy": 6.865924549102783,
"epoch": 0.3515436888008989,
"grad_norm": 1.078125,
"learning_rate": 0.0004993225043316257,
"loss": 6.8189,
"mean_token_accuracy": 0.10788183063268661,
"num_tokens": 7126440.0,
"step": 3285
},
{
"entropy": 6.94174747467041,
"epoch": 0.35207876290866286,
"grad_norm": 1.125,
"learning_rate": 0.0004993195363142643,
"loss": 6.8435,
"mean_token_accuracy": 0.10635844394564628,
"num_tokens": 7137884.0,
"step": 3290
},
{
"entropy": 6.8471081256866455,
"epoch": 0.3526138370164268,
"grad_norm": 1.2265625,
"learning_rate": 0.0004993165618197076,
"loss": 6.8169,
"mean_token_accuracy": 0.10543133914470673,
"num_tokens": 7147930.0,
"step": 3295
},
{
"entropy": 6.814928293228149,
"epoch": 0.3531489111241907,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993135808480418,
"loss": 6.6758,
"mean_token_accuracy": 0.12192264795303345,
"num_tokens": 7159200.0,
"step": 3300
},
{
"entropy": 6.856072282791137,
"epoch": 0.3536839852319546,
"grad_norm": 1.0625,
"learning_rate": 0.0004993105933993529,
"loss": 6.7738,
"mean_token_accuracy": 0.11219395026564598,
"num_tokens": 7169601.0,
"step": 3305
},
{
"entropy": 6.731938648223877,
"epoch": 0.35421905933971853,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993075994737271,
"loss": 6.7014,
"mean_token_accuracy": 0.12335732132196427,
"num_tokens": 7180465.0,
"step": 3310
},
{
"entropy": 6.827739572525024,
"epoch": 0.35475413344748247,
"grad_norm": 1.0625,
"learning_rate": 0.0004993045990712508,
"loss": 6.7214,
"mean_token_accuracy": 0.1124566912651062,
"num_tokens": 7190603.0,
"step": 3315
},
{
"entropy": 6.744693279266357,
"epoch": 0.3552892075552464,
"grad_norm": 1.015625,
"learning_rate": 0.0004993015921920109,
"loss": 6.6822,
"mean_token_accuracy": 0.1178012415766716,
"num_tokens": 7201017.0,
"step": 3320
},
{
"entropy": 6.700474643707276,
"epoch": 0.35582428166301033,
"grad_norm": 1.109375,
"learning_rate": 0.000499298578836094,
"loss": 6.6851,
"mean_token_accuracy": 0.11364478841423989,
"num_tokens": 7210908.0,
"step": 3325
},
{
"entropy": 6.913705682754516,
"epoch": 0.35635935577077427,
"grad_norm": 1.078125,
"learning_rate": 0.000499295559003587,
"loss": 6.7313,
"mean_token_accuracy": 0.11469347327947617,
"num_tokens": 7221306.0,
"step": 3330
},
{
"entropy": 6.857224035263061,
"epoch": 0.3568944298785382,
"grad_norm": 0.984375,
"learning_rate": 0.0004992925326945774,
"loss": 6.8446,
"mean_token_accuracy": 0.11008940190076828,
"num_tokens": 7233076.0,
"step": 3335
},
{
"entropy": 6.836943006515503,
"epoch": 0.3574295039863021,
"grad_norm": 1.1875,
"learning_rate": 0.0004992894999091524,
"loss": 6.833,
"mean_token_accuracy": 0.11023118868470191,
"num_tokens": 7244703.0,
"step": 3340
},
{
"entropy": 6.865760707855225,
"epoch": 0.357964578094066,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992864606473996,
"loss": 6.8127,
"mean_token_accuracy": 0.10562887638807297,
"num_tokens": 7255438.0,
"step": 3345
},
{
"entropy": 6.873836708068848,
"epoch": 0.35849965220182994,
"grad_norm": 1.1171875,
"learning_rate": 0.0004992834149094069,
"loss": 6.7746,
"mean_token_accuracy": 0.10823791846632957,
"num_tokens": 7265720.0,
"step": 3350
},
{
"entropy": 6.854056453704834,
"epoch": 0.3590347263095939,
"grad_norm": 1.0625,
"learning_rate": 0.0004992803626952619,
"loss": 6.8105,
"mean_token_accuracy": 0.10965728536248207,
"num_tokens": 7277791.0,
"step": 3355
},
{
"entropy": 6.743177652359009,
"epoch": 0.3595698004173578,
"grad_norm": 1.0625,
"learning_rate": 0.000499277304005053,
"loss": 6.6904,
"mean_token_accuracy": 0.11896344646811485,
"num_tokens": 7288521.0,
"step": 3360
},
{
"entropy": 6.8449372291564945,
"epoch": 0.36010487452512174,
"grad_norm": 1.03125,
"learning_rate": 0.0004992742388388686,
"loss": 6.8681,
"mean_token_accuracy": 0.10312251970171929,
"num_tokens": 7299298.0,
"step": 3365
},
{
"entropy": 6.848871850967408,
"epoch": 0.3606399486328857,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992711671967968,
"loss": 6.6907,
"mean_token_accuracy": 0.11500442028045654,
"num_tokens": 7310271.0,
"step": 3370
},
{
"entropy": 6.832868242263794,
"epoch": 0.36117502274064955,
"grad_norm": 1.046875,
"learning_rate": 0.0004992680890789268,
"loss": 6.7535,
"mean_token_accuracy": 0.11115506291389465,
"num_tokens": 7321112.0,
"step": 3375
},
{
"entropy": 6.744096136093139,
"epoch": 0.3617100968484135,
"grad_norm": 1.125,
"learning_rate": 0.0004992650044853471,
"loss": 6.7111,
"mean_token_accuracy": 0.11423057690262794,
"num_tokens": 7331976.0,
"step": 3380
},
{
"entropy": 6.881365203857422,
"epoch": 0.3622451709561774,
"grad_norm": 1.15625,
"learning_rate": 0.000499261913416147,
"loss": 6.8157,
"mean_token_accuracy": 0.11083743125200271,
"num_tokens": 7342800.0,
"step": 3385
},
{
"entropy": 6.820743465423584,
"epoch": 0.36278024506394135,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992588158714157,
"loss": 6.6508,
"mean_token_accuracy": 0.11105694174766541,
"num_tokens": 7354213.0,
"step": 3390
},
{
"entropy": 6.789371490478516,
"epoch": 0.3633153191717053,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992557118512424,
"loss": 6.7069,
"mean_token_accuracy": 0.11869031190872192,
"num_tokens": 7363250.0,
"step": 3395
},
{
"entropy": 6.7624798774719235,
"epoch": 0.3638503932794692,
"grad_norm": 1.078125,
"learning_rate": 0.0004992526013557171,
"loss": 6.7605,
"mean_token_accuracy": 0.11203641667962075,
"num_tokens": 7374185.0,
"step": 3400
},
{
"entropy": 6.752663087844849,
"epoch": 0.36438546738723315,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992494843849294,
"loss": 6.7287,
"mean_token_accuracy": 0.10924962237477302,
"num_tokens": 7384579.0,
"step": 3405
},
{
"entropy": 6.896695327758789,
"epoch": 0.36492054149499703,
"grad_norm": 1.1171875,
"learning_rate": 0.0004992463609389694,
"loss": 6.7422,
"mean_token_accuracy": 0.10970545783638955,
"num_tokens": 7395277.0,
"step": 3410
},
{
"entropy": 6.726968002319336,
"epoch": 0.36545561560276096,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992432310179272,
"loss": 6.7332,
"mean_token_accuracy": 0.1160312682390213,
"num_tokens": 7406688.0,
"step": 3415
},
{
"entropy": 6.766181993484497,
"epoch": 0.3659906897105249,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992400946218932,
"loss": 6.795,
"mean_token_accuracy": 0.11444359049201011,
"num_tokens": 7417378.0,
"step": 3420
},
{
"entropy": 6.967504692077637,
"epoch": 0.36652576381828883,
"grad_norm": 1.046875,
"learning_rate": 0.0004992369517509581,
"loss": 6.8179,
"mean_token_accuracy": 0.1145808644592762,
"num_tokens": 7427675.0,
"step": 3425
},
{
"entropy": 6.834393882751465,
"epoch": 0.36706083792605276,
"grad_norm": 1.0625,
"learning_rate": 0.0004992338024052123,
"loss": 6.8405,
"mean_token_accuracy": 0.10944146141409875,
"num_tokens": 7438083.0,
"step": 3430
},
{
"entropy": 6.772138452529907,
"epoch": 0.3675959120338167,
"grad_norm": 1.0859375,
"learning_rate": 0.0004992306465847473,
"loss": 6.6866,
"mean_token_accuracy": 0.11544951871037483,
"num_tokens": 7449051.0,
"step": 3435
},
{
"entropy": 6.803123044967651,
"epoch": 0.36813098614158063,
"grad_norm": 1.03125,
"learning_rate": 0.0004992274842896536,
"loss": 6.7313,
"mean_token_accuracy": 0.11575418040156364,
"num_tokens": 7459074.0,
"step": 3440
},
{
"entropy": 6.810998582839966,
"epoch": 0.3686660602493445,
"grad_norm": 1.0625,
"learning_rate": 0.0004992243155200229,
"loss": 6.7908,
"mean_token_accuracy": 0.11545799896121026,
"num_tokens": 7470472.0,
"step": 3445
},
{
"entropy": 6.810088348388672,
"epoch": 0.36920113435710844,
"grad_norm": 1.0,
"learning_rate": 0.0004992211402759466,
"loss": 6.7589,
"mean_token_accuracy": 0.11591348797082901,
"num_tokens": 7481477.0,
"step": 3450
},
{
"entropy": 6.811910390853882,
"epoch": 0.3697362084648724,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992179585575164,
"loss": 6.7166,
"mean_token_accuracy": 0.11710453182458877,
"num_tokens": 7491906.0,
"step": 3455
},
{
"entropy": 6.801189374923706,
"epoch": 0.3702712825726363,
"grad_norm": 1.1328125,
"learning_rate": 0.000499214770364824,
"loss": 6.6567,
"mean_token_accuracy": 0.10969180539250374,
"num_tokens": 7502447.0,
"step": 3460
},
{
"entropy": 6.731618452072143,
"epoch": 0.37080635668040024,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992115756979617,
"loss": 6.6488,
"mean_token_accuracy": 0.12150704562664032,
"num_tokens": 7513505.0,
"step": 3465
},
{
"entropy": 6.763904094696045,
"epoch": 0.3713414307881642,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992083745570218,
"loss": 6.6731,
"mean_token_accuracy": 0.11672577038407325,
"num_tokens": 7524884.0,
"step": 3470
},
{
"entropy": 6.764589834213257,
"epoch": 0.3718765048959281,
"grad_norm": 1.078125,
"learning_rate": 0.0004992051669420964,
"loss": 6.7534,
"mean_token_accuracy": 0.11659367606043816,
"num_tokens": 7535942.0,
"step": 3475
},
{
"entropy": 6.77218713760376,
"epoch": 0.372411579003692,
"grad_norm": 1.0859375,
"learning_rate": 0.0004992019528532785,
"loss": 6.7354,
"mean_token_accuracy": 0.12092937007546425,
"num_tokens": 7545506.0,
"step": 3480
},
{
"entropy": 6.874322080612183,
"epoch": 0.3729466531114559,
"grad_norm": 1.25,
"learning_rate": 0.0004991987322906605,
"loss": 6.801,
"mean_token_accuracy": 0.10959522724151612,
"num_tokens": 7555328.0,
"step": 3485
},
{
"entropy": 6.811362361907959,
"epoch": 0.37348172721921985,
"grad_norm": 1.1640625,
"learning_rate": 0.0004991955052543357,
"loss": 6.6899,
"mean_token_accuracy": 0.1134427696466446,
"num_tokens": 7565762.0,
"step": 3490
},
{
"entropy": 6.830954933166504,
"epoch": 0.3740168013269838,
"grad_norm": 1.09375,
"learning_rate": 0.0004991922717443972,
"loss": 6.7216,
"mean_token_accuracy": 0.12056689336895943,
"num_tokens": 7575884.0,
"step": 3495
},
{
"entropy": 6.742701625823974,
"epoch": 0.3745518754347477,
"grad_norm": 1.046875,
"learning_rate": 0.0004991890317609384,
"loss": 6.7499,
"mean_token_accuracy": 0.11369063630700112,
"num_tokens": 7586100.0,
"step": 3500
},
{
"entropy": 6.859546327590943,
"epoch": 0.37508694954251165,
"grad_norm": 1.171875,
"learning_rate": 0.0004991857853040528,
"loss": 6.7591,
"mean_token_accuracy": 0.11486376821994781,
"num_tokens": 7597137.0,
"step": 3505
},
{
"entropy": 6.82744083404541,
"epoch": 0.3756220236502756,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991825323738341,
"loss": 6.7932,
"mean_token_accuracy": 0.11721144542098046,
"num_tokens": 7607640.0,
"step": 3510
},
{
"entropy": 6.840407514572144,
"epoch": 0.3761570977580395,
"grad_norm": 1.1953125,
"learning_rate": 0.0004991792729703763,
"loss": 6.851,
"mean_token_accuracy": 0.1160910353064537,
"num_tokens": 7618319.0,
"step": 3515
},
{
"entropy": 6.877895498275757,
"epoch": 0.3766921718658034,
"grad_norm": 1.046875,
"learning_rate": 0.0004991760070937734,
"loss": 6.7208,
"mean_token_accuracy": 0.11491079404950141,
"num_tokens": 7628350.0,
"step": 3520
},
{
"entropy": 6.842350673675537,
"epoch": 0.3772272459735673,
"grad_norm": 1.203125,
"learning_rate": 0.0004991727347441199,
"loss": 6.7361,
"mean_token_accuracy": 0.11011843830347061,
"num_tokens": 7637857.0,
"step": 3525
},
{
"entropy": 6.858046007156372,
"epoch": 0.37776232008133126,
"grad_norm": 1.125,
"learning_rate": 0.0004991694559215102,
"loss": 6.7602,
"mean_token_accuracy": 0.11095951348543168,
"num_tokens": 7648242.0,
"step": 3530
},
{
"entropy": 6.830993270874023,
"epoch": 0.3782973941890952,
"grad_norm": 0.9453125,
"learning_rate": 0.000499166170626039,
"loss": 6.7743,
"mean_token_accuracy": 0.10925297141075134,
"num_tokens": 7660565.0,
"step": 3535
},
{
"entropy": 6.797787475585937,
"epoch": 0.3788324682968591,
"grad_norm": 1.21875,
"learning_rate": 0.000499162878857801,
"loss": 6.7711,
"mean_token_accuracy": 0.1141252376139164,
"num_tokens": 7671600.0,
"step": 3540
},
{
"entropy": 6.854355716705323,
"epoch": 0.37936754240462306,
"grad_norm": 1.03125,
"learning_rate": 0.0004991595806168915,
"loss": 6.7761,
"mean_token_accuracy": 0.10867656469345092,
"num_tokens": 7682453.0,
"step": 3545
},
{
"entropy": 6.801757192611694,
"epoch": 0.379902616512387,
"grad_norm": 1.0,
"learning_rate": 0.0004991562759034055,
"loss": 6.7221,
"mean_token_accuracy": 0.11367509812116623,
"num_tokens": 7693490.0,
"step": 3550
},
{
"entropy": 6.828986072540284,
"epoch": 0.38043769062015087,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991529647174386,
"loss": 6.6761,
"mean_token_accuracy": 0.1118717484176159,
"num_tokens": 7703478.0,
"step": 3555
},
{
"entropy": 6.7392528533935545,
"epoch": 0.3809727647279148,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991496470590864,
"loss": 6.7761,
"mean_token_accuracy": 0.1103611208498478,
"num_tokens": 7714647.0,
"step": 3560
},
{
"entropy": 6.863647413253784,
"epoch": 0.38150783883567874,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991463229284445,
"loss": 6.7173,
"mean_token_accuracy": 0.11717814281582832,
"num_tokens": 7725528.0,
"step": 3565
},
{
"entropy": 6.771598768234253,
"epoch": 0.38204291294344267,
"grad_norm": 1.109375,
"learning_rate": 0.0004991429923256091,
"loss": 6.7756,
"mean_token_accuracy": 0.11080185323953629,
"num_tokens": 7737708.0,
"step": 3570
},
{
"entropy": 6.878248071670532,
"epoch": 0.3825779870512066,
"grad_norm": 1.015625,
"learning_rate": 0.0004991396552506762,
"loss": 6.764,
"mean_token_accuracy": 0.11186526715755463,
"num_tokens": 7749093.0,
"step": 3575
},
{
"entropy": 6.7422984600067135,
"epoch": 0.38311306115897054,
"grad_norm": 1.0546875,
"learning_rate": 0.0004991363117037423,
"loss": 6.7004,
"mean_token_accuracy": 0.11901898533105851,
"num_tokens": 7760024.0,
"step": 3580
},
{
"entropy": 6.767995119094849,
"epoch": 0.38364813526673447,
"grad_norm": 1.03125,
"learning_rate": 0.0004991329616849039,
"loss": 6.7284,
"mean_token_accuracy": 0.11957356333732605,
"num_tokens": 7770801.0,
"step": 3585
},
{
"entropy": 6.866620016098023,
"epoch": 0.38418320937449835,
"grad_norm": 1.046875,
"learning_rate": 0.0004991296051942578,
"loss": 6.7847,
"mean_token_accuracy": 0.1116973415017128,
"num_tokens": 7782257.0,
"step": 3590
},
{
"entropy": 6.763524389266967,
"epoch": 0.3847182834822623,
"grad_norm": 1.015625,
"learning_rate": 0.0004991262422319008,
"loss": 6.7091,
"mean_token_accuracy": 0.11752377822995186,
"num_tokens": 7793016.0,
"step": 3595
},
{
"entropy": 6.7549824714660645,
"epoch": 0.3852533575900262,
"grad_norm": 1.25,
"learning_rate": 0.00049912287279793,
"loss": 6.6492,
"mean_token_accuracy": 0.1228921391069889,
"num_tokens": 7802207.0,
"step": 3600
},
{
"entropy": 6.812817192077636,
"epoch": 0.38578843169779015,
"grad_norm": 1.171875,
"learning_rate": 0.0004991194968924429,
"loss": 6.7834,
"mean_token_accuracy": 0.11178419142961502,
"num_tokens": 7813487.0,
"step": 3605
},
{
"entropy": 6.858286571502686,
"epoch": 0.3863235058055541,
"grad_norm": 1.109375,
"learning_rate": 0.0004991161145155367,
"loss": 6.7563,
"mean_token_accuracy": 0.11035445258021355,
"num_tokens": 7824587.0,
"step": 3610
},
{
"entropy": 6.882280063629151,
"epoch": 0.386858579913318,
"grad_norm": 1.0625,
"learning_rate": 0.0004991127256673091,
"loss": 6.7368,
"mean_token_accuracy": 0.11012360826134682,
"num_tokens": 7836240.0,
"step": 3615
},
{
"entropy": 6.753285121917725,
"epoch": 0.38739365402108195,
"grad_norm": 1.15625,
"learning_rate": 0.0004991093303478582,
"loss": 6.6861,
"mean_token_accuracy": 0.11604151576757431,
"num_tokens": 7847014.0,
"step": 3620
},
{
"entropy": 6.788820505142212,
"epoch": 0.3879287281288458,
"grad_norm": 1.046875,
"learning_rate": 0.0004991059285572818,
"loss": 6.807,
"mean_token_accuracy": 0.11351005062460899,
"num_tokens": 7857899.0,
"step": 3625
},
{
"entropy": 6.857131767272949,
"epoch": 0.38846380223660976,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991025202956781,
"loss": 6.6902,
"mean_token_accuracy": 0.11639388576149941,
"num_tokens": 7868987.0,
"step": 3630
},
{
"entropy": 6.850029611587525,
"epoch": 0.3889988763443737,
"grad_norm": 1.1875,
"learning_rate": 0.0004990991055631457,
"loss": 6.81,
"mean_token_accuracy": 0.11189388036727906,
"num_tokens": 7879547.0,
"step": 3635
},
{
"entropy": 6.762338304519654,
"epoch": 0.3895339504521376,
"grad_norm": 1.0625,
"learning_rate": 0.0004990956843597832,
"loss": 6.6763,
"mean_token_accuracy": 0.1170013889670372,
"num_tokens": 7889824.0,
"step": 3640
},
{
"entropy": 6.673375177383423,
"epoch": 0.39006902455990156,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990922566856892,
"loss": 6.669,
"mean_token_accuracy": 0.11331403329968452,
"num_tokens": 7901156.0,
"step": 3645
},
{
"entropy": 6.770735454559326,
"epoch": 0.3906040986676655,
"grad_norm": 1.296875,
"learning_rate": 0.0004990888225409629,
"loss": 6.7076,
"mean_token_accuracy": 0.11252901628613472,
"num_tokens": 7911695.0,
"step": 3650
},
{
"entropy": 6.755655097961426,
"epoch": 0.3911391727754294,
"grad_norm": 0.984375,
"learning_rate": 0.0004990853819257033,
"loss": 6.7026,
"mean_token_accuracy": 0.11720905601978301,
"num_tokens": 7923559.0,
"step": 3655
},
{
"entropy": 6.84619779586792,
"epoch": 0.3916742468831933,
"grad_norm": 1.09375,
"learning_rate": 0.0004990819348400097,
"loss": 6.7214,
"mean_token_accuracy": 0.11208191886544228,
"num_tokens": 7935792.0,
"step": 3660
},
{
"entropy": 6.793423795700074,
"epoch": 0.39220932099095723,
"grad_norm": 1.359375,
"learning_rate": 0.0004990784812839818,
"loss": 6.7092,
"mean_token_accuracy": 0.1186470128595829,
"num_tokens": 7946987.0,
"step": 3665
},
{
"entropy": 6.699849939346313,
"epoch": 0.39274439509872117,
"grad_norm": 1.125,
"learning_rate": 0.0004990750212577192,
"loss": 6.6369,
"mean_token_accuracy": 0.12066235840320587,
"num_tokens": 7957160.0,
"step": 3670
},
{
"entropy": 6.781460571289062,
"epoch": 0.3932794692064851,
"grad_norm": 1.09375,
"learning_rate": 0.0004990715547613219,
"loss": 6.7333,
"mean_token_accuracy": 0.10956480428576469,
"num_tokens": 7968833.0,
"step": 3675
},
{
"entropy": 6.873096227645874,
"epoch": 0.39381454331424903,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990680817948898,
"loss": 6.7822,
"mean_token_accuracy": 0.10927629321813584,
"num_tokens": 7980661.0,
"step": 3680
},
{
"entropy": 6.861059999465942,
"epoch": 0.39434961742201297,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990646023585234,
"loss": 6.7065,
"mean_token_accuracy": 0.11303527727723121,
"num_tokens": 7992677.0,
"step": 3685
},
{
"entropy": 6.700655746459961,
"epoch": 0.3948846915297769,
"grad_norm": 1.125,
"learning_rate": 0.0004990611164523231,
"loss": 6.6919,
"mean_token_accuracy": 0.1178669311106205,
"num_tokens": 8002796.0,
"step": 3690
},
{
"entropy": 6.6995728492736815,
"epoch": 0.3954197656375408,
"grad_norm": 1.171875,
"learning_rate": 0.0004990576240763896,
"loss": 6.7181,
"mean_token_accuracy": 0.11917829737067223,
"num_tokens": 8012779.0,
"step": 3695
},
{
"entropy": 6.79206166267395,
"epoch": 0.3959548397453047,
"grad_norm": 1.015625,
"learning_rate": 0.0004990541252308237,
"loss": 6.6986,
"mean_token_accuracy": 0.11830845102667809,
"num_tokens": 8023442.0,
"step": 3700
},
{
"entropy": 6.777542209625244,
"epoch": 0.39648991385306864,
"grad_norm": 1.109375,
"learning_rate": 0.0004990506199157263,
"loss": 6.7844,
"mean_token_accuracy": 0.11193263530731201,
"num_tokens": 8035163.0,
"step": 3705
},
{
"entropy": 6.826216363906861,
"epoch": 0.3970249879608326,
"grad_norm": 1.1484375,
"learning_rate": 0.0004990471081311988,
"loss": 6.7221,
"mean_token_accuracy": 0.11202551424503326,
"num_tokens": 8045330.0,
"step": 3710
},
{
"entropy": 6.826126527786255,
"epoch": 0.3975600620685965,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990435898773426,
"loss": 6.7341,
"mean_token_accuracy": 0.11674164608120918,
"num_tokens": 8056154.0,
"step": 3715
},
{
"entropy": 6.795881938934326,
"epoch": 0.39809513617636044,
"grad_norm": 1.015625,
"learning_rate": 0.0004990400651542592,
"loss": 6.6916,
"mean_token_accuracy": 0.1208396390080452,
"num_tokens": 8066101.0,
"step": 3720
},
{
"entropy": 6.82635269165039,
"epoch": 0.3986302102841244,
"grad_norm": 1.09375,
"learning_rate": 0.0004990365339620504,
"loss": 6.778,
"mean_token_accuracy": 0.11029551848769188,
"num_tokens": 8077249.0,
"step": 3725
},
{
"entropy": 6.791606903076172,
"epoch": 0.39916528439188825,
"grad_norm": 1.2734375,
"learning_rate": 0.000499032996300818,
"loss": 6.7108,
"mean_token_accuracy": 0.11683286353945732,
"num_tokens": 8088532.0,
"step": 3730
},
{
"entropy": 6.7761242389678955,
"epoch": 0.3997003584996522,
"grad_norm": 1.265625,
"learning_rate": 0.0004990294521706645,
"loss": 6.7513,
"mean_token_accuracy": 0.11043450236320496,
"num_tokens": 8099541.0,
"step": 3735
},
{
"entropy": 6.778212356567383,
"epoch": 0.4002354326074161,
"grad_norm": 1.0,
"learning_rate": 0.000499025901571692,
"loss": 6.7925,
"mean_token_accuracy": 0.11460375785827637,
"num_tokens": 8110532.0,
"step": 3740
},
{
"entropy": 6.860037422180175,
"epoch": 0.40077050671518005,
"grad_norm": 1.265625,
"learning_rate": 0.0004990223445040031,
"loss": 6.7246,
"mean_token_accuracy": 0.12051974311470985,
"num_tokens": 8120991.0,
"step": 3745
},
{
"entropy": 6.796218538284302,
"epoch": 0.401305580822944,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990187809677004,
"loss": 6.6332,
"mean_token_accuracy": 0.11498752385377883,
"num_tokens": 8130737.0,
"step": 3750
},
{
"entropy": 6.687275981903076,
"epoch": 0.4018406549307079,
"grad_norm": 1.3203125,
"learning_rate": 0.0004990152109628869,
"loss": 6.6614,
"mean_token_accuracy": 0.11621819585561752,
"num_tokens": 8142315.0,
"step": 3755
},
{
"entropy": 6.747700786590576,
"epoch": 0.40237572903847185,
"grad_norm": 1.2734375,
"learning_rate": 0.0004990116344896657,
"loss": 6.7604,
"mean_token_accuracy": 0.10932756140828133,
"num_tokens": 8153618.0,
"step": 3760
},
{
"entropy": 6.786961269378662,
"epoch": 0.40291080314623573,
"grad_norm": 1.109375,
"learning_rate": 0.00049900805154814,
"loss": 6.6973,
"mean_token_accuracy": 0.11853857338428497,
"num_tokens": 8164304.0,
"step": 3765
},
{
"entropy": 6.784421157836914,
"epoch": 0.40344587725399966,
"grad_norm": 1.1328125,
"learning_rate": 0.0004990044621384134,
"loss": 6.694,
"mean_token_accuracy": 0.11464422270655632,
"num_tokens": 8175289.0,
"step": 3770
},
{
"entropy": 6.834812927246094,
"epoch": 0.4039809513617636,
"grad_norm": 1.125,
"learning_rate": 0.0004990008662605893,
"loss": 6.7375,
"mean_token_accuracy": 0.11080463156104088,
"num_tokens": 8186274.0,
"step": 3775
},
{
"entropy": 6.850270223617554,
"epoch": 0.40451602546952753,
"grad_norm": 1.1640625,
"learning_rate": 0.0004989972639147717,
"loss": 6.6819,
"mean_token_accuracy": 0.11826486811041832,
"num_tokens": 8196380.0,
"step": 3780
},
{
"entropy": 6.745476388931275,
"epoch": 0.40505109957729146,
"grad_norm": 1.15625,
"learning_rate": 0.0004989936551010646,
"loss": 6.7327,
"mean_token_accuracy": 0.11468399837613105,
"num_tokens": 8207204.0,
"step": 3785
},
{
"entropy": 6.715826034545898,
"epoch": 0.4055861736850554,
"grad_norm": 1.1796875,
"learning_rate": 0.0004989900398195722,
"loss": 6.6512,
"mean_token_accuracy": 0.11534782201051712,
"num_tokens": 8217745.0,
"step": 3790
},
{
"entropy": 6.774829626083374,
"epoch": 0.40612124779281933,
"grad_norm": 1.140625,
"learning_rate": 0.0004989864180703988,
"loss": 6.6424,
"mean_token_accuracy": 0.11919590383768082,
"num_tokens": 8228171.0,
"step": 3795
},
{
"entropy": 6.758113431930542,
"epoch": 0.4066563219005832,
"grad_norm": 1.2421875,
"learning_rate": 0.0004989827898536491,
"loss": 6.6973,
"mean_token_accuracy": 0.11042807400226592,
"num_tokens": 8239220.0,
"step": 3800
},
{
"entropy": 6.832132291793823,
"epoch": 0.40719139600834714,
"grad_norm": 1.1640625,
"learning_rate": 0.0004989791551694279,
"loss": 6.7518,
"mean_token_accuracy": 0.109524205327034,
"num_tokens": 8250363.0,
"step": 3805
},
{
"entropy": 6.776221656799317,
"epoch": 0.4077264701161111,
"grad_norm": 1.1171875,
"learning_rate": 0.0004989755140178398,
"loss": 6.7123,
"mean_token_accuracy": 0.11326616704463958,
"num_tokens": 8261207.0,
"step": 3810
},
{
"entropy": 6.741102790832519,
"epoch": 0.408261544223875,
"grad_norm": 1.109375,
"learning_rate": 0.0004989718663989905,
"loss": 6.6174,
"mean_token_accuracy": 0.11389720663428307,
"num_tokens": 8272343.0,
"step": 3815
},
{
"entropy": 6.711847400665283,
"epoch": 0.40879661833163894,
"grad_norm": 1.0625,
"learning_rate": 0.0004989682123129849,
"loss": 6.6662,
"mean_token_accuracy": 0.11849249228835106,
"num_tokens": 8282677.0,
"step": 3820
},
{
"entropy": 6.69567084312439,
"epoch": 0.40933169243940287,
"grad_norm": 1.03125,
"learning_rate": 0.0004989645517599286,
"loss": 6.6745,
"mean_token_accuracy": 0.11427896246314048,
"num_tokens": 8292753.0,
"step": 3825
},
{
"entropy": 6.774503755569458,
"epoch": 0.4098667665471668,
"grad_norm": 1.046875,
"learning_rate": 0.0004989608847399275,
"loss": 6.6625,
"mean_token_accuracy": 0.11602176800370216,
"num_tokens": 8302747.0,
"step": 3830
},
{
"entropy": 6.774828100204468,
"epoch": 0.4104018406549307,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989572112530871,
"loss": 6.7446,
"mean_token_accuracy": 0.10802446901798249,
"num_tokens": 8314766.0,
"step": 3835
},
{
"entropy": 6.862928152084351,
"epoch": 0.4109369147626946,
"grad_norm": 1.0625,
"learning_rate": 0.0004989535312995139,
"loss": 6.7682,
"mean_token_accuracy": 0.11004860624670983,
"num_tokens": 8325791.0,
"step": 3840
},
{
"entropy": 6.746032285690307,
"epoch": 0.41147198887045855,
"grad_norm": 1.1328125,
"learning_rate": 0.0004989498448793139,
"loss": 6.6939,
"mean_token_accuracy": 0.11601744443178177,
"num_tokens": 8335809.0,
"step": 3845
},
{
"entropy": 6.742163610458374,
"epoch": 0.4120070629782225,
"grad_norm": 1.0546875,
"learning_rate": 0.0004989461519925935,
"loss": 6.6923,
"mean_token_accuracy": 0.11100121140480042,
"num_tokens": 8347505.0,
"step": 3850
},
{
"entropy": 6.8387268543243405,
"epoch": 0.4125421370859864,
"grad_norm": 1.0625,
"learning_rate": 0.0004989424526394596,
"loss": 6.7162,
"mean_token_accuracy": 0.1083609253168106,
"num_tokens": 8358363.0,
"step": 3855
},
{
"entropy": 6.726605081558228,
"epoch": 0.41307721119375035,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989387468200187,
"loss": 6.7182,
"mean_token_accuracy": 0.1165882483124733,
"num_tokens": 8369076.0,
"step": 3860
},
{
"entropy": 6.849684762954712,
"epoch": 0.4136122853015143,
"grad_norm": 1.109375,
"learning_rate": 0.000498935034534378,
"loss": 6.7567,
"mean_token_accuracy": 0.11488589569926262,
"num_tokens": 8380071.0,
"step": 3865
},
{
"entropy": 6.779253625869751,
"epoch": 0.41414735940927816,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989313157826446,
"loss": 6.7043,
"mean_token_accuracy": 0.11365948691964149,
"num_tokens": 8390574.0,
"step": 3870
},
{
"entropy": 6.762129831314087,
"epoch": 0.4146824335170421,
"grad_norm": 1.0859375,
"learning_rate": 0.000498927590564926,
"loss": 6.7016,
"mean_token_accuracy": 0.11860018000006675,
"num_tokens": 8401154.0,
"step": 3875
},
{
"entropy": 6.805259132385254,
"epoch": 0.415217507624806,
"grad_norm": 1.0546875,
"learning_rate": 0.0004989238588813296,
"loss": 6.7816,
"mean_token_accuracy": 0.10864789858460426,
"num_tokens": 8412113.0,
"step": 3880
},
{
"entropy": 6.893309640884399,
"epoch": 0.41575258173256996,
"grad_norm": 1.203125,
"learning_rate": 0.0004989201207319634,
"loss": 6.7582,
"mean_token_accuracy": 0.11186528205871582,
"num_tokens": 8422734.0,
"step": 3885
},
{
"entropy": 6.758945322036743,
"epoch": 0.4162876558403339,
"grad_norm": 1.125,
"learning_rate": 0.0004989163761169351,
"loss": 6.6912,
"mean_token_accuracy": 0.11084168404340744,
"num_tokens": 8433647.0,
"step": 3890
},
{
"entropy": 6.744187116622925,
"epoch": 0.4168227299480978,
"grad_norm": 1.203125,
"learning_rate": 0.0004989126250363529,
"loss": 6.7234,
"mean_token_accuracy": 0.11423909962177277,
"num_tokens": 8444726.0,
"step": 3895
},
{
"entropy": 6.83604621887207,
"epoch": 0.41735780405586176,
"grad_norm": 1.09375,
"learning_rate": 0.0004989088674903251,
"loss": 6.7264,
"mean_token_accuracy": 0.11136697009205818,
"num_tokens": 8455351.0,
"step": 3900
},
{
"entropy": 6.788657855987549,
"epoch": 0.41789287816362564,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989051034789602,
"loss": 6.7073,
"mean_token_accuracy": 0.11919761151075363,
"num_tokens": 8466423.0,
"step": 3905
},
{
"entropy": 6.769079303741455,
"epoch": 0.41842795227138957,
"grad_norm": 1.28125,
"learning_rate": 0.000498901333002367,
"loss": 6.6522,
"mean_token_accuracy": 0.11701465100049972,
"num_tokens": 8475841.0,
"step": 3910
},
{
"entropy": 6.814233303070068,
"epoch": 0.4189630263791535,
"grad_norm": 1.03125,
"learning_rate": 0.0004988975560606541,
"loss": 6.762,
"mean_token_accuracy": 0.11229455918073654,
"num_tokens": 8487656.0,
"step": 3915
},
{
"entropy": 6.776022148132324,
"epoch": 0.41949810048691744,
"grad_norm": 1.1640625,
"learning_rate": 0.0004988937726539308,
"loss": 6.7744,
"mean_token_accuracy": 0.10670395717024803,
"num_tokens": 8497959.0,
"step": 3920
},
{
"entropy": 6.763610315322876,
"epoch": 0.42003317459468137,
"grad_norm": 1.078125,
"learning_rate": 0.0004988899827823062,
"loss": 6.6751,
"mean_token_accuracy": 0.11493971273303032,
"num_tokens": 8508228.0,
"step": 3925
},
{
"entropy": 6.782846927642822,
"epoch": 0.4205682487024453,
"grad_norm": 1.171875,
"learning_rate": 0.00049888618644589,
"loss": 6.7229,
"mean_token_accuracy": 0.11063806042075157,
"num_tokens": 8518999.0,
"step": 3930
},
{
"entropy": 6.691450452804565,
"epoch": 0.42110332281020924,
"grad_norm": 1.2265625,
"learning_rate": 0.0004988823836447916,
"loss": 6.7519,
"mean_token_accuracy": 0.11179999187588692,
"num_tokens": 8529583.0,
"step": 3935
},
{
"entropy": 6.894530963897705,
"epoch": 0.4216383969179731,
"grad_norm": 1.4609375,
"learning_rate": 0.0004988785743791207,
"loss": 6.8223,
"mean_token_accuracy": 0.10763506144285202,
"num_tokens": 8540069.0,
"step": 3940
},
{
"entropy": 6.875021266937256,
"epoch": 0.42217347102573705,
"grad_norm": 1.1953125,
"learning_rate": 0.0004988747586489876,
"loss": 6.7009,
"mean_token_accuracy": 0.11498932242393493,
"num_tokens": 8550441.0,
"step": 3945
},
{
"entropy": 6.709805822372436,
"epoch": 0.422708545133501,
"grad_norm": 1.0546875,
"learning_rate": 0.000498870936454502,
"loss": 6.6653,
"mean_token_accuracy": 0.11834602132439613,
"num_tokens": 8560257.0,
"step": 3950
},
{
"entropy": 6.828940343856812,
"epoch": 0.4232436192412649,
"grad_norm": 1.078125,
"learning_rate": 0.0004988671077957749,
"loss": 6.7485,
"mean_token_accuracy": 0.1175241857767105,
"num_tokens": 8570991.0,
"step": 3955
},
{
"entropy": 6.891437435150147,
"epoch": 0.42377869334902885,
"grad_norm": 1.609375,
"learning_rate": 0.0004988632726729164,
"loss": 6.8051,
"mean_token_accuracy": 0.10697821751236916,
"num_tokens": 8583668.0,
"step": 3960
},
{
"entropy": 6.802229166030884,
"epoch": 0.4243137674567928,
"grad_norm": 1.109375,
"learning_rate": 0.0004988594310860374,
"loss": 6.7184,
"mean_token_accuracy": 0.1156295083463192,
"num_tokens": 8593727.0,
"step": 3965
},
{
"entropy": 6.714612770080566,
"epoch": 0.4248488415645567,
"grad_norm": 1.0625,
"learning_rate": 0.0004988555830352485,
"loss": 6.7209,
"mean_token_accuracy": 0.11398242264986039,
"num_tokens": 8605470.0,
"step": 3970
},
{
"entropy": 6.7963508605957035,
"epoch": 0.4253839156723206,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988517285206614,
"loss": 6.6687,
"mean_token_accuracy": 0.11099848076701165,
"num_tokens": 8617176.0,
"step": 3975
},
{
"entropy": 6.803634119033814,
"epoch": 0.4259189897800845,
"grad_norm": 1.0703125,
"learning_rate": 0.0004988478675423871,
"loss": 6.7558,
"mean_token_accuracy": 0.11225240230560303,
"num_tokens": 8627756.0,
"step": 3980
},
{
"entropy": 6.793984985351562,
"epoch": 0.42645406388784846,
"grad_norm": 1.234375,
"learning_rate": 0.0004988440001005369,
"loss": 6.6853,
"mean_token_accuracy": 0.11842627823352814,
"num_tokens": 8637766.0,
"step": 3985
},
{
"entropy": 6.800572538375855,
"epoch": 0.4269891379956124,
"grad_norm": 1.1640625,
"learning_rate": 0.0004988401261952227,
"loss": 6.7096,
"mean_token_accuracy": 0.11486690863966942,
"num_tokens": 8648562.0,
"step": 3990
},
{
"entropy": 6.808977746963501,
"epoch": 0.4275242121033763,
"grad_norm": 1.1875,
"learning_rate": 0.0004988362458265565,
"loss": 6.7412,
"mean_token_accuracy": 0.10799469202756881,
"num_tokens": 8658393.0,
"step": 3995
},
{
"entropy": 6.730482482910157,
"epoch": 0.42805928621114026,
"grad_norm": 1.125,
"learning_rate": 0.0004988323589946499,
"loss": 6.6543,
"mean_token_accuracy": 0.1162540040910244,
"num_tokens": 8668365.0,
"step": 4000
},
{
"entropy": 6.777298212051392,
"epoch": 0.4285943603189042,
"grad_norm": 1.1953125,
"learning_rate": 0.0004988284656996156,
"loss": 6.7009,
"mean_token_accuracy": 0.1099359154701233,
"num_tokens": 8679228.0,
"step": 4005
},
{
"entropy": 6.780287981033325,
"epoch": 0.42912943442666807,
"grad_norm": 1.234375,
"learning_rate": 0.0004988245659415656,
"loss": 6.5866,
"mean_token_accuracy": 0.12446512654423714,
"num_tokens": 8688940.0,
"step": 4010
},
{
"entropy": 6.810698747634888,
"epoch": 0.429664508534432,
"grad_norm": 1.1796875,
"learning_rate": 0.0004988206597206129,
"loss": 6.7438,
"mean_token_accuracy": 0.10872282534837723,
"num_tokens": 8699923.0,
"step": 4015
},
{
"entropy": 6.720220708847046,
"epoch": 0.43019958264219593,
"grad_norm": 1.0546875,
"learning_rate": 0.00049881674703687,
"loss": 6.58,
"mean_token_accuracy": 0.12176239043474198,
"num_tokens": 8710445.0,
"step": 4020
},
{
"entropy": 6.721858978271484,
"epoch": 0.43073465674995987,
"grad_norm": 1.3046875,
"learning_rate": 0.0004988128278904499,
"loss": 6.7083,
"mean_token_accuracy": 0.11447286382317542,
"num_tokens": 8720686.0,
"step": 4025
},
{
"entropy": 6.834508752822876,
"epoch": 0.4312697308577238,
"grad_norm": 1.140625,
"learning_rate": 0.000498808902281466,
"loss": 6.7282,
"mean_token_accuracy": 0.11510429829359055,
"num_tokens": 8732122.0,
"step": 4030
},
{
"entropy": 6.744071006774902,
"epoch": 0.43180480496548773,
"grad_norm": 1.125,
"learning_rate": 0.0004988049702100315,
"loss": 6.6825,
"mean_token_accuracy": 0.1182553879916668,
"num_tokens": 8744122.0,
"step": 4035
},
{
"entropy": 6.787398195266723,
"epoch": 0.43233987907325166,
"grad_norm": 0.984375,
"learning_rate": 0.0004988010316762598,
"loss": 6.6911,
"mean_token_accuracy": 0.115293999761343,
"num_tokens": 8754758.0,
"step": 4040
},
{
"entropy": 6.7287671089172365,
"epoch": 0.4328749531810156,
"grad_norm": 1.1953125,
"learning_rate": 0.0004987970866802649,
"loss": 6.7121,
"mean_token_accuracy": 0.10785665214061738,
"num_tokens": 8766577.0,
"step": 4045
},
{
"entropy": 6.773381233215332,
"epoch": 0.4334100272887795,
"grad_norm": 1.1171875,
"learning_rate": 0.0004987931352221605,
"loss": 6.7083,
"mean_token_accuracy": 0.11141579747200012,
"num_tokens": 8777698.0,
"step": 4050
},
{
"entropy": 6.788628673553466,
"epoch": 0.4339451013965434,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987891773020606,
"loss": 6.7231,
"mean_token_accuracy": 0.11151274070143699,
"num_tokens": 8788637.0,
"step": 4055
},
{
"entropy": 6.799765920639038,
"epoch": 0.43448017550430734,
"grad_norm": 1.265625,
"learning_rate": 0.0004987852129200799,
"loss": 6.7009,
"mean_token_accuracy": 0.11956919953227044,
"num_tokens": 8799709.0,
"step": 4060
},
{
"entropy": 6.795716381072998,
"epoch": 0.4350152496120713,
"grad_norm": 1.171875,
"learning_rate": 0.0004987812420763326,
"loss": 6.6905,
"mean_token_accuracy": 0.11553208827972412,
"num_tokens": 8810469.0,
"step": 4065
},
{
"entropy": 6.7736915111541744,
"epoch": 0.4355503237198352,
"grad_norm": 1.171875,
"learning_rate": 0.0004987772647709333,
"loss": 6.6708,
"mean_token_accuracy": 0.1123408019542694,
"num_tokens": 8821142.0,
"step": 4070
},
{
"entropy": 6.70412278175354,
"epoch": 0.43608539782759914,
"grad_norm": 1.1796875,
"learning_rate": 0.000498773281003997,
"loss": 6.6226,
"mean_token_accuracy": 0.12270026281476021,
"num_tokens": 8832951.0,
"step": 4075
},
{
"entropy": 6.725498294830322,
"epoch": 0.4366204719353631,
"grad_norm": 1.3203125,
"learning_rate": 0.0004987692907756385,
"loss": 6.7344,
"mean_token_accuracy": 0.10939609631896019,
"num_tokens": 8843579.0,
"step": 4080
},
{
"entropy": 6.6969774723052975,
"epoch": 0.43715554604312695,
"grad_norm": 1.125,
"learning_rate": 0.0004987652940859733,
"loss": 6.6606,
"mean_token_accuracy": 0.10758041813969613,
"num_tokens": 8854612.0,
"step": 4085
},
{
"entropy": 6.766866064071655,
"epoch": 0.4376906201508909,
"grad_norm": 1.0390625,
"learning_rate": 0.0004987612909351167,
"loss": 6.6101,
"mean_token_accuracy": 0.12067801505327225,
"num_tokens": 8865584.0,
"step": 4090
},
{
"entropy": 6.688352346420288,
"epoch": 0.4382256942586548,
"grad_norm": 1.109375,
"learning_rate": 0.0004987572813231842,
"loss": 6.66,
"mean_token_accuracy": 0.12287814989686012,
"num_tokens": 8876662.0,
"step": 4095
},
{
"entropy": 6.756010103225708,
"epoch": 0.43876076836641875,
"grad_norm": 1.09375,
"learning_rate": 0.0004987532652502917,
"loss": 6.6831,
"mean_token_accuracy": 0.11270663142204285,
"num_tokens": 8887249.0,
"step": 4100
},
{
"entropy": 6.713475131988526,
"epoch": 0.4392958424741827,
"grad_norm": 1.2109375,
"learning_rate": 0.0004987492427165551,
"loss": 6.531,
"mean_token_accuracy": 0.1301178902387619,
"num_tokens": 8898635.0,
"step": 4105
},
{
"entropy": 6.678007888793945,
"epoch": 0.4398309165819466,
"grad_norm": 1.09375,
"learning_rate": 0.0004987452137220906,
"loss": 6.6895,
"mean_token_accuracy": 0.11399767398834229,
"num_tokens": 8908794.0,
"step": 4110
},
{
"entropy": 6.847482490539551,
"epoch": 0.44036599068971055,
"grad_norm": 1.171875,
"learning_rate": 0.0004987411782670144,
"loss": 6.7013,
"mean_token_accuracy": 0.10850507244467736,
"num_tokens": 8919015.0,
"step": 4115
},
{
"entropy": 6.8152250289917,
"epoch": 0.44090106479747443,
"grad_norm": 1.140625,
"learning_rate": 0.0004987371363514431,
"loss": 6.7349,
"mean_token_accuracy": 0.10842297747731208,
"num_tokens": 8929028.0,
"step": 4120
},
{
"entropy": 6.664592504501343,
"epoch": 0.44143613890523836,
"grad_norm": 1.125,
"learning_rate": 0.0004987330879754933,
"loss": 6.6382,
"mean_token_accuracy": 0.11036456525325775,
"num_tokens": 8939789.0,
"step": 4125
},
{
"entropy": 6.707959175109863,
"epoch": 0.4419712130130023,
"grad_norm": 1.015625,
"learning_rate": 0.0004987290331392823,
"loss": 6.6225,
"mean_token_accuracy": 0.11801224276423454,
"num_tokens": 8950657.0,
"step": 4130
},
{
"entropy": 6.828504610061645,
"epoch": 0.44250628712076623,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987249718429269,
"loss": 6.7236,
"mean_token_accuracy": 0.11803532913327217,
"num_tokens": 8960744.0,
"step": 4135
},
{
"entropy": 6.705205869674683,
"epoch": 0.44304136122853016,
"grad_norm": 1.09375,
"learning_rate": 0.0004987209040865441,
"loss": 6.5768,
"mean_token_accuracy": 0.12112260162830353,
"num_tokens": 8972411.0,
"step": 4140
},
{
"entropy": 6.801696109771728,
"epoch": 0.4435764353362941,
"grad_norm": 1.2265625,
"learning_rate": 0.0004987168298702519,
"loss": 6.7465,
"mean_token_accuracy": 0.11419489085674286,
"num_tokens": 8982818.0,
"step": 4145
},
{
"entropy": 6.745963048934937,
"epoch": 0.44411150944405803,
"grad_norm": 1.234375,
"learning_rate": 0.0004987127491941675,
"loss": 6.6657,
"mean_token_accuracy": 0.1177960142493248,
"num_tokens": 8993227.0,
"step": 4150
},
{
"entropy": 6.669366073608399,
"epoch": 0.4446465835518219,
"grad_norm": 1.25,
"learning_rate": 0.0004987086620584091,
"loss": 6.6037,
"mean_token_accuracy": 0.11799841225147248,
"num_tokens": 9003796.0,
"step": 4155
},
{
"entropy": 6.772842884063721,
"epoch": 0.44518165765958584,
"grad_norm": 1.421875,
"learning_rate": 0.0004987045684630944,
"loss": 6.6369,
"mean_token_accuracy": 0.12047947198152542,
"num_tokens": 9014931.0,
"step": 4160
},
{
"entropy": 6.774410676956177,
"epoch": 0.44571673176734977,
"grad_norm": 1.15625,
"learning_rate": 0.0004987004684083417,
"loss": 6.6984,
"mean_token_accuracy": 0.1164324201643467,
"num_tokens": 9025312.0,
"step": 4165
},
{
"entropy": 6.7285847663879395,
"epoch": 0.4462518058751137,
"grad_norm": 1.1015625,
"learning_rate": 0.0004986963618942694,
"loss": 6.6642,
"mean_token_accuracy": 0.12318786159157753,
"num_tokens": 9035563.0,
"step": 4170
},
{
"entropy": 6.795755052566529,
"epoch": 0.44678687998287764,
"grad_norm": 1.2578125,
"learning_rate": 0.0004986922489209961,
"loss": 6.7039,
"mean_token_accuracy": 0.1175026074051857,
"num_tokens": 9046065.0,
"step": 4175
},
{
"entropy": 6.751359653472901,
"epoch": 0.44732195409064157,
"grad_norm": 1.0625,
"learning_rate": 0.0004986881294886406,
"loss": 6.6057,
"mean_token_accuracy": 0.1231963075697422,
"num_tokens": 9056696.0,
"step": 4180
},
{
"entropy": 6.676720142364502,
"epoch": 0.4478570281984055,
"grad_norm": 1.171875,
"learning_rate": 0.0004986840035973218,
"loss": 6.6808,
"mean_token_accuracy": 0.11223538517951966,
"num_tokens": 9067035.0,
"step": 4185
},
{
"entropy": 6.6606361865997314,
"epoch": 0.4483921023061694,
"grad_norm": 1.140625,
"learning_rate": 0.0004986798712471588,
"loss": 6.6574,
"mean_token_accuracy": 0.11127996742725373,
"num_tokens": 9078253.0,
"step": 4190
},
{
"entropy": 6.8346106052398685,
"epoch": 0.4489271764139333,
"grad_norm": 1.296875,
"learning_rate": 0.0004986757324382709,
"loss": 6.7174,
"mean_token_accuracy": 0.10800155103206635,
"num_tokens": 9088184.0,
"step": 4195
},
{
"entropy": 6.7804230690002445,
"epoch": 0.44946225052169725,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986715871707777,
"loss": 6.7777,
"mean_token_accuracy": 0.1087512344121933,
"num_tokens": 9099321.0,
"step": 4200
},
{
"entropy": 6.843128871917725,
"epoch": 0.4499973246294612,
"grad_norm": 1.140625,
"learning_rate": 0.0004986674354447988,
"loss": 6.8009,
"mean_token_accuracy": 0.11305928155779839,
"num_tokens": 9111410.0,
"step": 4205
},
{
"entropy": 6.831840133666992,
"epoch": 0.4505323987372251,
"grad_norm": 1.1796875,
"learning_rate": 0.0004986632772604543,
"loss": 6.6256,
"mean_token_accuracy": 0.11788237541913986,
"num_tokens": 9121932.0,
"step": 4210
},
{
"entropy": 6.651640558242798,
"epoch": 0.45106747284498905,
"grad_norm": 1.6640625,
"learning_rate": 0.0004986591126178639,
"loss": 6.7181,
"mean_token_accuracy": 0.1195308357477188,
"num_tokens": 9132686.0,
"step": 4215
},
{
"entropy": 6.678121089935303,
"epoch": 0.451602546952753,
"grad_norm": 1.4375,
"learning_rate": 0.0004986549415171482,
"loss": 6.6445,
"mean_token_accuracy": 0.11687054708600045,
"num_tokens": 9143582.0,
"step": 4220
},
{
"entropy": 6.877950763702392,
"epoch": 0.45213762106051686,
"grad_norm": 1.09375,
"learning_rate": 0.0004986507639584274,
"loss": 6.7232,
"mean_token_accuracy": 0.11007717251777649,
"num_tokens": 9156173.0,
"step": 4225
},
{
"entropy": 6.7473616123199465,
"epoch": 0.4526726951682808,
"grad_norm": 1.2265625,
"learning_rate": 0.0004986465799418223,
"loss": 6.7062,
"mean_token_accuracy": 0.11616904065012931,
"num_tokens": 9166882.0,
"step": 4230
},
{
"entropy": 6.726114082336426,
"epoch": 0.4532077692760447,
"grad_norm": 1.28125,
"learning_rate": 0.0004986423894674534,
"loss": 6.7034,
"mean_token_accuracy": 0.11051197499036788,
"num_tokens": 9177525.0,
"step": 4235
},
{
"entropy": 6.824288320541382,
"epoch": 0.45374284338380866,
"grad_norm": 1.21875,
"learning_rate": 0.0004986381925354422,
"loss": 6.6713,
"mean_token_accuracy": 0.11391364932060241,
"num_tokens": 9188586.0,
"step": 4240
},
{
"entropy": 6.741261911392212,
"epoch": 0.4542779174915726,
"grad_norm": 1.328125,
"learning_rate": 0.0004986339891459094,
"loss": 6.5797,
"mean_token_accuracy": 0.11786932200193405,
"num_tokens": 9199920.0,
"step": 4245
},
{
"entropy": 6.590327644348145,
"epoch": 0.4548129915993365,
"grad_norm": 1.21875,
"learning_rate": 0.0004986297792989768,
"loss": 6.583,
"mean_token_accuracy": 0.12098821178078652,
"num_tokens": 9210923.0,
"step": 4250
},
{
"entropy": 6.832316637039185,
"epoch": 0.45534806570710046,
"grad_norm": 1.2421875,
"learning_rate": 0.0004986255629947655,
"loss": 6.7539,
"mean_token_accuracy": 0.10964171513915062,
"num_tokens": 9221459.0,
"step": 4255
},
{
"entropy": 6.773038291931153,
"epoch": 0.45588313981486434,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986213402333978,
"loss": 6.6935,
"mean_token_accuracy": 0.11024248600006104,
"num_tokens": 9231768.0,
"step": 4260
},
{
"entropy": 6.713382816314697,
"epoch": 0.45641821392262827,
"grad_norm": 1.09375,
"learning_rate": 0.0004986171110149951,
"loss": 6.6848,
"mean_token_accuracy": 0.11179919168353081,
"num_tokens": 9242230.0,
"step": 4265
},
{
"entropy": 6.807206058502198,
"epoch": 0.4569532880303922,
"grad_norm": 1.0859375,
"learning_rate": 0.0004986128753396798,
"loss": 6.5794,
"mean_token_accuracy": 0.1219303086400032,
"num_tokens": 9252485.0,
"step": 4270
},
{
"entropy": 6.657087945938111,
"epoch": 0.45748836213815613,
"grad_norm": 1.3125,
"learning_rate": 0.0004986086332075742,
"loss": 6.6527,
"mean_token_accuracy": 0.12027468383312226,
"num_tokens": 9263014.0,
"step": 4275
},
{
"entropy": 6.737786865234375,
"epoch": 0.45802343624592007,
"grad_norm": 1.3046875,
"learning_rate": 0.0004986043846188006,
"loss": 6.6932,
"mean_token_accuracy": 0.12059752494096757,
"num_tokens": 9273777.0,
"step": 4280
},
{
"entropy": 6.658463954925537,
"epoch": 0.458558510353684,
"grad_norm": 1.3203125,
"learning_rate": 0.000498600129573482,
"loss": 6.6266,
"mean_token_accuracy": 0.11395588889718056,
"num_tokens": 9284738.0,
"step": 4285
},
{
"entropy": 6.793663120269775,
"epoch": 0.45909358446144793,
"grad_norm": 1.125,
"learning_rate": 0.000498595868071741,
"loss": 6.6807,
"mean_token_accuracy": 0.11528572961688041,
"num_tokens": 9295921.0,
"step": 4290
},
{
"entropy": 6.744870233535766,
"epoch": 0.4596286585692118,
"grad_norm": 1.2421875,
"learning_rate": 0.0004985916001137006,
"loss": 6.6609,
"mean_token_accuracy": 0.11062274277210235,
"num_tokens": 9306568.0,
"step": 4295
},
{
"entropy": 6.757999467849731,
"epoch": 0.46016373267697575,
"grad_norm": 1.1875,
"learning_rate": 0.0004985873256994843,
"loss": 6.6262,
"mean_token_accuracy": 0.11653959006071091,
"num_tokens": 9316468.0,
"step": 4300
},
{
"entropy": 6.737951755523682,
"epoch": 0.4606988067847397,
"grad_norm": 1.21875,
"learning_rate": 0.0004985830448292154,
"loss": 6.7008,
"mean_token_accuracy": 0.1143273763358593,
"num_tokens": 9328185.0,
"step": 4305
},
{
"entropy": 6.743026876449585,
"epoch": 0.4612338808925036,
"grad_norm": 1.234375,
"learning_rate": 0.0004985787575030175,
"loss": 6.6547,
"mean_token_accuracy": 0.11528828516602516,
"num_tokens": 9338591.0,
"step": 4310
},
{
"entropy": 6.768393468856812,
"epoch": 0.46176895500026754,
"grad_norm": 1.078125,
"learning_rate": 0.0004985744637210144,
"loss": 6.6761,
"mean_token_accuracy": 0.11174369007349014,
"num_tokens": 9349286.0,
"step": 4315
},
{
"entropy": 6.783293533325195,
"epoch": 0.4623040291080315,
"grad_norm": 1.375,
"learning_rate": 0.00049857016348333,
"loss": 6.658,
"mean_token_accuracy": 0.11993650272488594,
"num_tokens": 9359752.0,
"step": 4320
},
{
"entropy": 6.78243350982666,
"epoch": 0.4628391032157954,
"grad_norm": 1.1953125,
"learning_rate": 0.0004985658567900886,
"loss": 6.7129,
"mean_token_accuracy": 0.1122040569782257,
"num_tokens": 9371823.0,
"step": 4325
},
{
"entropy": 6.643386602401733,
"epoch": 0.4633741773235593,
"grad_norm": 1.0703125,
"learning_rate": 0.0004985615436414145,
"loss": 6.5839,
"mean_token_accuracy": 0.11392313092947007,
"num_tokens": 9383127.0,
"step": 4330
},
{
"entropy": 6.734666919708252,
"epoch": 0.4639092514313232,
"grad_norm": 1.1875,
"learning_rate": 0.0004985572240374321,
"loss": 6.6771,
"mean_token_accuracy": 0.11793937757611275,
"num_tokens": 9394296.0,
"step": 4335
},
{
"entropy": 6.754271030426025,
"epoch": 0.46444432553908715,
"grad_norm": 1.1640625,
"learning_rate": 0.0004985528979782665,
"loss": 6.6688,
"mean_token_accuracy": 0.11072949096560478,
"num_tokens": 9405988.0,
"step": 4340
},
{
"entropy": 6.822707033157348,
"epoch": 0.4649793996468511,
"grad_norm": 1.1015625,
"learning_rate": 0.0004985485654640423,
"loss": 6.6693,
"mean_token_accuracy": 0.1139429748058319,
"num_tokens": 9417479.0,
"step": 4345
},
{
"entropy": 6.696816492080688,
"epoch": 0.465514473754615,
"grad_norm": 1.375,
"learning_rate": 0.0004985442264948847,
"loss": 6.6197,
"mean_token_accuracy": 0.11986967846751213,
"num_tokens": 9428050.0,
"step": 4350
},
{
"entropy": 6.653148794174195,
"epoch": 0.46604954786237895,
"grad_norm": 1.1484375,
"learning_rate": 0.0004985398810709189,
"loss": 6.5722,
"mean_token_accuracy": 0.12251258864998818,
"num_tokens": 9438229.0,
"step": 4355
},
{
"entropy": 6.654054975509643,
"epoch": 0.4665846219701429,
"grad_norm": 1.171875,
"learning_rate": 0.0004985355291922704,
"loss": 6.6493,
"mean_token_accuracy": 0.11258513107895851,
"num_tokens": 9449611.0,
"step": 4360
},
{
"entropy": 6.709498739242553,
"epoch": 0.46711969607790677,
"grad_norm": 1.515625,
"learning_rate": 0.000498531170859065,
"loss": 6.614,
"mean_token_accuracy": 0.11630048006772994,
"num_tokens": 9460234.0,
"step": 4365
},
{
"entropy": 6.828418874740601,
"epoch": 0.4676547701856707,
"grad_norm": 1.625,
"learning_rate": 0.0004985268060714285,
"loss": 6.7023,
"mean_token_accuracy": 0.1152671605348587,
"num_tokens": 9472136.0,
"step": 4370
},
{
"entropy": 6.6818465232849125,
"epoch": 0.46818984429343463,
"grad_norm": 1.3515625,
"learning_rate": 0.0004985224348294868,
"loss": 6.6403,
"mean_token_accuracy": 0.12047104090452194,
"num_tokens": 9482602.0,
"step": 4375
},
{
"entropy": 6.785189914703369,
"epoch": 0.46872491840119856,
"grad_norm": 1.28125,
"learning_rate": 0.0004985180571333663,
"loss": 6.6442,
"mean_token_accuracy": 0.11965756639838218,
"num_tokens": 9493782.0,
"step": 4380
},
{
"entropy": 6.763046312332153,
"epoch": 0.4692599925089625,
"grad_norm": 1.5078125,
"learning_rate": 0.0004985136729831932,
"loss": 6.6249,
"mean_token_accuracy": 0.11808878481388092,
"num_tokens": 9504389.0,
"step": 4385
},
{
"entropy": 6.648394393920898,
"epoch": 0.46979506661672643,
"grad_norm": 1.5390625,
"learning_rate": 0.0004985092823790942,
"loss": 6.7447,
"mean_token_accuracy": 0.1183752790093422,
"num_tokens": 9515166.0,
"step": 4390
},
{
"entropy": 6.747145318984986,
"epoch": 0.47033014072449036,
"grad_norm": 1.140625,
"learning_rate": 0.0004985048853211961,
"loss": 6.6434,
"mean_token_accuracy": 0.116434495896101,
"num_tokens": 9525566.0,
"step": 4395
},
{
"entropy": 6.7929564952850345,
"epoch": 0.47086521483225424,
"grad_norm": 1.21875,
"learning_rate": 0.000498500481809626,
"loss": 6.5549,
"mean_token_accuracy": 0.1208685427904129,
"num_tokens": 9536021.0,
"step": 4400
},
{
"entropy": 6.649615907669068,
"epoch": 0.4714002889400182,
"grad_norm": 1.3984375,
"learning_rate": 0.0004984960718445107,
"loss": 6.6641,
"mean_token_accuracy": 0.11327528953552246,
"num_tokens": 9546827.0,
"step": 4405
},
{
"entropy": 6.756602811813354,
"epoch": 0.4719353630477821,
"grad_norm": 1.28125,
"learning_rate": 0.0004984916554259778,
"loss": 6.7758,
"mean_token_accuracy": 0.10899243578314781,
"num_tokens": 9558104.0,
"step": 4410
},
{
"entropy": 6.866197681427002,
"epoch": 0.47247043715554604,
"grad_norm": 1.1796875,
"learning_rate": 0.0004984872325541547,
"loss": 6.668,
"mean_token_accuracy": 0.11704885140061379,
"num_tokens": 9568487.0,
"step": 4415
},
{
"entropy": 6.716640186309815,
"epoch": 0.47300551126331,
"grad_norm": 1.3125,
"learning_rate": 0.0004984828032291692,
"loss": 6.6107,
"mean_token_accuracy": 0.12084364518523216,
"num_tokens": 9579509.0,
"step": 4420
},
{
"entropy": 6.742200565338135,
"epoch": 0.4735405853710739,
"grad_norm": 1.96875,
"learning_rate": 0.0004984783674511492,
"loss": 6.64,
"mean_token_accuracy": 0.1198382891714573,
"num_tokens": 9589931.0,
"step": 4425
},
{
"entropy": 6.761376428604126,
"epoch": 0.47407565947883784,
"grad_norm": 1.328125,
"learning_rate": 0.0004984739252202227,
"loss": 6.6733,
"mean_token_accuracy": 0.11940053328871728,
"num_tokens": 9599764.0,
"step": 4430
},
{
"entropy": 6.7573480129241945,
"epoch": 0.4746107335866017,
"grad_norm": 1.2578125,
"learning_rate": 0.000498469476536518,
"loss": 6.6472,
"mean_token_accuracy": 0.1184098556637764,
"num_tokens": 9611145.0,
"step": 4435
},
{
"entropy": 6.6948751449584964,
"epoch": 0.47514580769436565,
"grad_norm": 1.4453125,
"learning_rate": 0.0004984650214001636,
"loss": 6.6436,
"mean_token_accuracy": 0.12355822175741196,
"num_tokens": 9621441.0,
"step": 4440
},
{
"entropy": 6.741222667694092,
"epoch": 0.4756808818021296,
"grad_norm": 1.59375,
"learning_rate": 0.0004984605598112881,
"loss": 6.6857,
"mean_token_accuracy": 0.12146021127700805,
"num_tokens": 9631874.0,
"step": 4445
},
{
"entropy": 6.717041254043579,
"epoch": 0.4762159559098935,
"grad_norm": 1.1484375,
"learning_rate": 0.0004984560917700204,
"loss": 6.6614,
"mean_token_accuracy": 0.11106858849525451,
"num_tokens": 9643006.0,
"step": 4450
},
{
"entropy": 6.698492193222046,
"epoch": 0.47675103001765745,
"grad_norm": 1.328125,
"learning_rate": 0.0004984516172764894,
"loss": 6.5533,
"mean_token_accuracy": 0.12116582468152046,
"num_tokens": 9653897.0,
"step": 4455
},
{
"entropy": 6.691738748550415,
"epoch": 0.4772861041254214,
"grad_norm": 1.2734375,
"learning_rate": 0.0004984471363308243,
"loss": 6.6263,
"mean_token_accuracy": 0.1176533468067646,
"num_tokens": 9664460.0,
"step": 4460
},
{
"entropy": 6.711106204986573,
"epoch": 0.4778211782331853,
"grad_norm": 1.1015625,
"learning_rate": 0.0004984426489331546,
"loss": 6.6385,
"mean_token_accuracy": 0.12148366495966911,
"num_tokens": 9674534.0,
"step": 4465
},
{
"entropy": 6.8625555515289305,
"epoch": 0.4783562523409492,
"grad_norm": 1.2109375,
"learning_rate": 0.0004984381550836097,
"loss": 6.732,
"mean_token_accuracy": 0.11358912587165833,
"num_tokens": 9684828.0,
"step": 4470
},
{
"entropy": 6.851891565322876,
"epoch": 0.47889132644871313,
"grad_norm": 1.6015625,
"learning_rate": 0.0004984336547823196,
"loss": 6.6838,
"mean_token_accuracy": 0.11679960638284684,
"num_tokens": 9694536.0,
"step": 4475
},
{
"entropy": 6.712016582489014,
"epoch": 0.47942640055647706,
"grad_norm": 1.0703125,
"learning_rate": 0.000498429148029414,
"loss": 6.5548,
"mean_token_accuracy": 0.12640432715415956,
"num_tokens": 9705715.0,
"step": 4480
},
{
"entropy": 6.654127883911133,
"epoch": 0.479961474664241,
"grad_norm": 1.125,
"learning_rate": 0.0004984246348250233,
"loss": 6.6293,
"mean_token_accuracy": 0.12291777729988099,
"num_tokens": 9717505.0,
"step": 4485
},
{
"entropy": 6.698559045791626,
"epoch": 0.4804965487720049,
"grad_norm": 1.15625,
"learning_rate": 0.0004984201151692775,
"loss": 6.5005,
"mean_token_accuracy": 0.11582913845777512,
"num_tokens": 9727910.0,
"step": 4490
},
{
"entropy": 6.681623268127441,
"epoch": 0.48103162287976886,
"grad_norm": 1.2578125,
"learning_rate": 0.0004984155890623074,
"loss": 6.6164,
"mean_token_accuracy": 0.12003767117857933,
"num_tokens": 9738528.0,
"step": 4495
},
{
"entropy": 6.688356590270996,
"epoch": 0.4815666969875328,
"grad_norm": 1.1484375,
"learning_rate": 0.0004984110565042435,
"loss": 6.6433,
"mean_token_accuracy": 0.11511807218194008,
"num_tokens": 9749744.0,
"step": 4500
},
{
"entropy": 6.787782335281372,
"epoch": 0.48210177109529667,
"grad_norm": 1.1796875,
"learning_rate": 0.0004984065174952168,
"loss": 6.6415,
"mean_token_accuracy": 0.11788514703512191,
"num_tokens": 9759968.0,
"step": 4505
},
{
"entropy": 6.7138995170593265,
"epoch": 0.4826368452030606,
"grad_norm": 1.3046875,
"learning_rate": 0.0004984019720353582,
"loss": 6.6483,
"mean_token_accuracy": 0.11984834149479866,
"num_tokens": 9769791.0,
"step": 4510
},
{
"entropy": 6.669849443435669,
"epoch": 0.48317191931082454,
"grad_norm": 1.1953125,
"learning_rate": 0.0004983974201247991,
"loss": 6.6168,
"mean_token_accuracy": 0.12382967993617058,
"num_tokens": 9780369.0,
"step": 4515
},
{
"entropy": 6.718169641494751,
"epoch": 0.48370699341858847,
"grad_norm": 1.0703125,
"learning_rate": 0.0004983928617636709,
"loss": 6.6649,
"mean_token_accuracy": 0.11595848128199578,
"num_tokens": 9791125.0,
"step": 4520
},
{
"entropy": 6.822065162658691,
"epoch": 0.4842420675263524,
"grad_norm": 1.296875,
"learning_rate": 0.0004983882969521052,
"loss": 6.6623,
"mean_token_accuracy": 0.11691764071583748,
"num_tokens": 9801697.0,
"step": 4525
},
{
"entropy": 6.784085750579834,
"epoch": 0.48477714163411634,
"grad_norm": 1.046875,
"learning_rate": 0.0004983837256902338,
"loss": 6.7085,
"mean_token_accuracy": 0.1206382617354393,
"num_tokens": 9812171.0,
"step": 4530
},
{
"entropy": 6.700609016418457,
"epoch": 0.48531221574188027,
"grad_norm": 1.1171875,
"learning_rate": 0.0004983791479781887,
"loss": 6.5452,
"mean_token_accuracy": 0.12644470036029815,
"num_tokens": 9821596.0,
"step": 4535
},
{
"entropy": 6.76994104385376,
"epoch": 0.48584728984964415,
"grad_norm": 1.296875,
"learning_rate": 0.0004983745638161021,
"loss": 6.6662,
"mean_token_accuracy": 0.10886719152331352,
"num_tokens": 9831634.0,
"step": 4540
},
{
"entropy": 6.702042055130005,
"epoch": 0.4863823639574081,
"grad_norm": 1.4296875,
"learning_rate": 0.0004983699732041064,
"loss": 6.6297,
"mean_token_accuracy": 0.12174258530139923,
"num_tokens": 9841662.0,
"step": 4545
},
{
"entropy": 6.773559427261352,
"epoch": 0.486917438065172,
"grad_norm": 1.109375,
"learning_rate": 0.000498365376142334,
"loss": 6.618,
"mean_token_accuracy": 0.11437310129404069,
"num_tokens": 9853797.0,
"step": 4550
},
{
"entropy": 6.712430953979492,
"epoch": 0.48745251217293595,
"grad_norm": 1.1484375,
"learning_rate": 0.0004983607726309179,
"loss": 6.6691,
"mean_token_accuracy": 0.1182169571518898,
"num_tokens": 9865245.0,
"step": 4555
},
{
"entropy": 6.814268589019775,
"epoch": 0.4879875862806999,
"grad_norm": 1.234375,
"learning_rate": 0.0004983561626699908,
"loss": 6.7454,
"mean_token_accuracy": 0.10475116744637489,
"num_tokens": 9876207.0,
"step": 4560
},
{
"entropy": 6.760530710220337,
"epoch": 0.4885226603884638,
"grad_norm": 1.234375,
"learning_rate": 0.0004983515462596859,
"loss": 6.5712,
"mean_token_accuracy": 0.12076753973960877,
"num_tokens": 9887210.0,
"step": 4565
},
{
"entropy": 6.743468475341797,
"epoch": 0.48905773449622775,
"grad_norm": 1.2578125,
"learning_rate": 0.0004983469234001365,
"loss": 6.746,
"mean_token_accuracy": 0.11424284428358078,
"num_tokens": 9899607.0,
"step": 4570
},
{
"entropy": 6.662120580673218,
"epoch": 0.4895928086039917,
"grad_norm": 1.3515625,
"learning_rate": 0.000498342294091476,
"loss": 6.5284,
"mean_token_accuracy": 0.1255207620561123,
"num_tokens": 9909910.0,
"step": 4575
},
{
"entropy": 6.648574686050415,
"epoch": 0.49012788271175556,
"grad_norm": 1.25,
"learning_rate": 0.0004983376583338382,
"loss": 6.6563,
"mean_token_accuracy": 0.11490331664681434,
"num_tokens": 9922137.0,
"step": 4580
},
{
"entropy": 6.735975837707519,
"epoch": 0.4906629568195195,
"grad_norm": 1.1953125,
"learning_rate": 0.0004983330161273569,
"loss": 6.6451,
"mean_token_accuracy": 0.11815119236707687,
"num_tokens": 9932514.0,
"step": 4585
},
{
"entropy": 6.717578268051147,
"epoch": 0.4911980309272834,
"grad_norm": 1.5390625,
"learning_rate": 0.0004983283674721662,
"loss": 6.6847,
"mean_token_accuracy": 0.11441330313682556,
"num_tokens": 9943054.0,
"step": 4590
},
{
"entropy": 6.8306371688842775,
"epoch": 0.49173310503504736,
"grad_norm": 1.1796875,
"learning_rate": 0.0004983237123684002,
"loss": 6.6517,
"mean_token_accuracy": 0.11424371600151062,
"num_tokens": 9953791.0,
"step": 4595
},
{
"entropy": 6.725871992111206,
"epoch": 0.4922681791428113,
"grad_norm": 1.234375,
"learning_rate": 0.0004983190508161934,
"loss": 6.6071,
"mean_token_accuracy": 0.1176756463944912,
"num_tokens": 9964782.0,
"step": 4600
},
{
"entropy": 6.776705169677735,
"epoch": 0.4928032532505752,
"grad_norm": 1.6953125,
"learning_rate": 0.0004983143828156804,
"loss": 6.6759,
"mean_token_accuracy": 0.1201729841530323,
"num_tokens": 9975843.0,
"step": 4605
},
{
"entropy": 6.766890525817871,
"epoch": 0.49333832735833916,
"grad_norm": 1.5859375,
"learning_rate": 0.0004983097083669959,
"loss": 6.6515,
"mean_token_accuracy": 0.12076375931501389,
"num_tokens": 9987060.0,
"step": 4610
},
{
"entropy": 6.729199981689453,
"epoch": 0.49387340146610303,
"grad_norm": 1.296875,
"learning_rate": 0.0004983050274702751,
"loss": 6.6204,
"mean_token_accuracy": 0.11678859367966651,
"num_tokens": 9997380.0,
"step": 4615
},
{
"entropy": 6.704911518096924,
"epoch": 0.49440847557386697,
"grad_norm": 1.1171875,
"learning_rate": 0.0004983003401256529,
"loss": 6.6666,
"mean_token_accuracy": 0.11961752027273179,
"num_tokens": 10007955.0,
"step": 4620
},
{
"entropy": 6.858521175384522,
"epoch": 0.4949435496816309,
"grad_norm": 1.390625,
"learning_rate": 0.0004982956463332647,
"loss": 6.7414,
"mean_token_accuracy": 0.11539428681135178,
"num_tokens": 10017645.0,
"step": 4625
},
{
"entropy": 6.7818702220916744,
"epoch": 0.49547862378939483,
"grad_norm": 2.328125,
"learning_rate": 0.0004982909460932463,
"loss": 6.6603,
"mean_token_accuracy": 0.11978099420666695,
"num_tokens": 10027619.0,
"step": 4630
},
{
"entropy": 6.654353475570678,
"epoch": 0.49601369789715877,
"grad_norm": 1.15625,
"learning_rate": 0.000498286239405733,
"loss": 6.5454,
"mean_token_accuracy": 0.12450932785868644,
"num_tokens": 10037551.0,
"step": 4635
},
{
"entropy": 6.672497034072876,
"epoch": 0.4965487720049227,
"grad_norm": 1.1171875,
"learning_rate": 0.000498281526270861,
"loss": 6.5851,
"mean_token_accuracy": 0.13364579305052757,
"num_tokens": 10047771.0,
"step": 4640
},
{
"entropy": 6.803449487686157,
"epoch": 0.49708384611268663,
"grad_norm": 1.203125,
"learning_rate": 0.0004982768066887663,
"loss": 6.7076,
"mean_token_accuracy": 0.112837415933609,
"num_tokens": 10058771.0,
"step": 4645
},
{
"entropy": 6.723805093765259,
"epoch": 0.4976189202204505,
"grad_norm": 1.2109375,
"learning_rate": 0.0004982720806595851,
"loss": 6.5841,
"mean_token_accuracy": 0.12232825830578804,
"num_tokens": 10070318.0,
"step": 4650
},
{
"entropy": 6.68502402305603,
"epoch": 0.49815399432821444,
"grad_norm": 1.1171875,
"learning_rate": 0.0004982673481834541,
"loss": 6.614,
"mean_token_accuracy": 0.11796658039093018,
"num_tokens": 10081810.0,
"step": 4655
},
{
"entropy": 6.6698534965515135,
"epoch": 0.4986890684359784,
"grad_norm": 1.1953125,
"learning_rate": 0.0004982626092605097,
"loss": 6.6297,
"mean_token_accuracy": 0.11533761844038963,
"num_tokens": 10091724.0,
"step": 4660
},
{
"entropy": 6.763433408737183,
"epoch": 0.4992241425437423,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982578638908888,
"loss": 6.6159,
"mean_token_accuracy": 0.11898914724588394,
"num_tokens": 10102510.0,
"step": 4665
},
{
"entropy": 6.755692815780639,
"epoch": 0.49975921665150624,
"grad_norm": 1.2734375,
"learning_rate": 0.0004982531120747286,
"loss": 6.6781,
"mean_token_accuracy": 0.11363844275474548,
"num_tokens": 10112636.0,
"step": 4670
},
{
"entropy": 6.679724454879761,
"epoch": 0.5002942907592701,
"grad_norm": 1.546875,
"learning_rate": 0.000498248353812166,
"loss": 6.684,
"mean_token_accuracy": 0.11922202631831169,
"num_tokens": 10123562.0,
"step": 4675
},
{
"entropy": 6.720367956161499,
"epoch": 0.5008293648670341,
"grad_norm": 1.234375,
"learning_rate": 0.0004982435891033387,
"loss": 6.6344,
"mean_token_accuracy": 0.11208936050534249,
"num_tokens": 10134731.0,
"step": 4680
},
{
"entropy": 6.773916530609131,
"epoch": 0.501364438974798,
"grad_norm": 1.2265625,
"learning_rate": 0.000498238817948384,
"loss": 6.6307,
"mean_token_accuracy": 0.11377799212932586,
"num_tokens": 10145232.0,
"step": 4685
},
{
"entropy": 6.715243864059448,
"epoch": 0.501899513082562,
"grad_norm": 1.3671875,
"learning_rate": 0.0004982340403474398,
"loss": 6.601,
"mean_token_accuracy": 0.11478810831904411,
"num_tokens": 10155857.0,
"step": 4690
},
{
"entropy": 6.7384484767913815,
"epoch": 0.5024345871903259,
"grad_norm": 1.1796875,
"learning_rate": 0.0004982292563006441,
"loss": 6.6107,
"mean_token_accuracy": 0.12016694247722626,
"num_tokens": 10166145.0,
"step": 4695
},
{
"entropy": 6.8165631771087645,
"epoch": 0.5029696612980897,
"grad_norm": 1.2421875,
"learning_rate": 0.000498224465808135,
"loss": 6.7341,
"mean_token_accuracy": 0.10747307166457176,
"num_tokens": 10177086.0,
"step": 4700
},
{
"entropy": 6.807919645309449,
"epoch": 0.5035047354058537,
"grad_norm": 1.6015625,
"learning_rate": 0.0004982196688700509,
"loss": 6.6168,
"mean_token_accuracy": 0.11652338951826095,
"num_tokens": 10187120.0,
"step": 4705
},
{
"entropy": 6.746653747558594,
"epoch": 0.5040398095136176,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982148654865302,
"loss": 6.6816,
"mean_token_accuracy": 0.1173696868121624,
"num_tokens": 10197784.0,
"step": 4710
},
{
"entropy": 6.769271898269653,
"epoch": 0.5045748836213816,
"grad_norm": 1.1875,
"learning_rate": 0.0004982100556577115,
"loss": 6.5533,
"mean_token_accuracy": 0.12568439543247223,
"num_tokens": 10209373.0,
"step": 4715
},
{
"entropy": 6.758709573745728,
"epoch": 0.5051099577291455,
"grad_norm": 1.328125,
"learning_rate": 0.0004982052393837338,
"loss": 6.669,
"mean_token_accuracy": 0.11871311962604522,
"num_tokens": 10220316.0,
"step": 4720
},
{
"entropy": 6.6824239730834964,
"epoch": 0.5056450318369095,
"grad_norm": 1.2265625,
"learning_rate": 0.0004982004166647363,
"loss": 6.5571,
"mean_token_accuracy": 0.12401916012167931,
"num_tokens": 10230478.0,
"step": 4725
},
{
"entropy": 6.623062801361084,
"epoch": 0.5061801059446733,
"grad_norm": 1.1796875,
"learning_rate": 0.0004981955875008582,
"loss": 6.6518,
"mean_token_accuracy": 0.11519326344132423,
"num_tokens": 10240641.0,
"step": 4730
},
{
"entropy": 6.744868040084839,
"epoch": 0.5067151800524372,
"grad_norm": 1.2578125,
"learning_rate": 0.0004981907518922387,
"loss": 6.6428,
"mean_token_accuracy": 0.12281770035624504,
"num_tokens": 10250676.0,
"step": 4735
},
{
"entropy": 6.859378337860107,
"epoch": 0.5072502541602012,
"grad_norm": 1.9765625,
"learning_rate": 0.0004981859098390177,
"loss": 6.7032,
"mean_token_accuracy": 0.11514592617750168,
"num_tokens": 10262016.0,
"step": 4740
},
{
"entropy": 6.7457503318786625,
"epoch": 0.5077853282679651,
"grad_norm": 1.3046875,
"learning_rate": 0.0004981810613413348,
"loss": 6.6448,
"mean_token_accuracy": 0.12012371569871902,
"num_tokens": 10271985.0,
"step": 4745
},
{
"entropy": 6.745742273330689,
"epoch": 0.5083204023757291,
"grad_norm": 1.2734375,
"learning_rate": 0.0004981762063993302,
"loss": 6.6133,
"mean_token_accuracy": 0.12136726751923561,
"num_tokens": 10282156.0,
"step": 4750
},
{
"entropy": 6.716922092437744,
"epoch": 0.5088554764834929,
"grad_norm": 1.34375,
"learning_rate": 0.000498171345013144,
"loss": 6.611,
"mean_token_accuracy": 0.11782191470265388,
"num_tokens": 10292576.0,
"step": 4755
},
{
"entropy": 6.661800575256348,
"epoch": 0.5093905505912569,
"grad_norm": 1.6640625,
"learning_rate": 0.0004981664771829165,
"loss": 6.6254,
"mean_token_accuracy": 0.1142365463078022,
"num_tokens": 10303648.0,
"step": 4760
},
{
"entropy": 6.715044784545898,
"epoch": 0.5099256246990208,
"grad_norm": 1.2734375,
"learning_rate": 0.0004981616029087884,
"loss": 6.5102,
"mean_token_accuracy": 0.12722289934754372,
"num_tokens": 10314275.0,
"step": 4765
},
{
"entropy": 6.675028467178345,
"epoch": 0.5104606988067847,
"grad_norm": 1.3515625,
"learning_rate": 0.0004981567221909004,
"loss": 6.6398,
"mean_token_accuracy": 0.11122743040323257,
"num_tokens": 10325386.0,
"step": 4770
},
{
"entropy": 6.712424802780151,
"epoch": 0.5109957729145487,
"grad_norm": 1.8828125,
"learning_rate": 0.0004981518350293935,
"loss": 6.6132,
"mean_token_accuracy": 0.11537306159734725,
"num_tokens": 10335041.0,
"step": 4775
},
{
"entropy": 6.743242788314819,
"epoch": 0.5115308470223126,
"grad_norm": 1.6796875,
"learning_rate": 0.0004981469414244086,
"loss": 6.5693,
"mean_token_accuracy": 0.12268173769116401,
"num_tokens": 10345746.0,
"step": 4780
},
{
"entropy": 6.624886989593506,
"epoch": 0.5120659211300765,
"grad_norm": 1.25,
"learning_rate": 0.000498142041376087,
"loss": 6.6945,
"mean_token_accuracy": 0.11829338818788529,
"num_tokens": 10356995.0,
"step": 4785
},
{
"entropy": 6.7250689506530765,
"epoch": 0.5126009952378404,
"grad_norm": 1.359375,
"learning_rate": 0.0004981371348845705,
"loss": 6.5939,
"mean_token_accuracy": 0.1246476911008358,
"num_tokens": 10368445.0,
"step": 4790
},
{
"entropy": 6.810773944854736,
"epoch": 0.5131360693456044,
"grad_norm": 1.1875,
"learning_rate": 0.0004981322219500006,
"loss": 6.5364,
"mean_token_accuracy": 0.1264880530536175,
"num_tokens": 10378256.0,
"step": 4795
},
{
"entropy": 6.698590278625488,
"epoch": 0.5136711434533683,
"grad_norm": 1.328125,
"learning_rate": 0.000498127302572519,
"loss": 6.6306,
"mean_token_accuracy": 0.11948861032724381,
"num_tokens": 10388698.0,
"step": 4800
},
{
"entropy": 6.626500415802002,
"epoch": 0.5142062175611322,
"grad_norm": 1.3203125,
"learning_rate": 0.0004981223767522679,
"loss": 6.6314,
"mean_token_accuracy": 0.12543342038989067,
"num_tokens": 10399975.0,
"step": 4805
},
{
"entropy": 6.73018741607666,
"epoch": 0.5147412916688962,
"grad_norm": 1.234375,
"learning_rate": 0.0004981174444893896,
"loss": 6.632,
"mean_token_accuracy": 0.11657147109508514,
"num_tokens": 10410480.0,
"step": 4810
},
{
"entropy": 6.754246234893799,
"epoch": 0.51527636577666,
"grad_norm": 1.484375,
"learning_rate": 0.0004981125057840264,
"loss": 6.5903,
"mean_token_accuracy": 0.12315746694803238,
"num_tokens": 10421023.0,
"step": 4815
},
{
"entropy": 6.691860723495483,
"epoch": 0.515811439884424,
"grad_norm": 1.34375,
"learning_rate": 0.000498107560636321,
"loss": 6.6376,
"mean_token_accuracy": 0.11937036588788033,
"num_tokens": 10431152.0,
"step": 4820
},
{
"entropy": 6.669547367095947,
"epoch": 0.5163465139921879,
"grad_norm": 1.375,
"learning_rate": 0.0004981026090464161,
"loss": 6.5395,
"mean_token_accuracy": 0.11537841185927392,
"num_tokens": 10441986.0,
"step": 4825
},
{
"entropy": 6.76296591758728,
"epoch": 0.5168815880999519,
"grad_norm": 1.515625,
"learning_rate": 0.0004980976510144548,
"loss": 6.6294,
"mean_token_accuracy": 0.11330693066120148,
"num_tokens": 10452932.0,
"step": 4830
},
{
"entropy": 6.730077743530273,
"epoch": 0.5174166622077158,
"grad_norm": 1.1875,
"learning_rate": 0.0004980926865405801,
"loss": 6.7059,
"mean_token_accuracy": 0.1165323257446289,
"num_tokens": 10463411.0,
"step": 4835
},
{
"entropy": 6.755215454101562,
"epoch": 0.5179517363154797,
"grad_norm": 1.3203125,
"learning_rate": 0.0004980877156249354,
"loss": 6.5955,
"mean_token_accuracy": 0.12119976133108139,
"num_tokens": 10474820.0,
"step": 4840
},
{
"entropy": 6.693712377548218,
"epoch": 0.5184868104232436,
"grad_norm": 1.3671875,
"learning_rate": 0.0004980827382676643,
"loss": 6.5956,
"mean_token_accuracy": 0.12424970418214798,
"num_tokens": 10485701.0,
"step": 4845
},
{
"entropy": 6.697637557983398,
"epoch": 0.5190218845310075,
"grad_norm": 1.375,
"learning_rate": 0.0004980777544689105,
"loss": 6.5925,
"mean_token_accuracy": 0.12146328687667847,
"num_tokens": 10495825.0,
"step": 4850
},
{
"entropy": 6.6988880157470705,
"epoch": 0.5195569586387715,
"grad_norm": 1.265625,
"learning_rate": 0.0004980727642288178,
"loss": 6.5759,
"mean_token_accuracy": 0.12025154009461403,
"num_tokens": 10505847.0,
"step": 4855
},
{
"entropy": 6.712225770950317,
"epoch": 0.5200920327465354,
"grad_norm": 1.359375,
"learning_rate": 0.0004980677675475305,
"loss": 6.6367,
"mean_token_accuracy": 0.11677327230572701,
"num_tokens": 10516596.0,
"step": 4860
},
{
"entropy": 6.733681726455688,
"epoch": 0.5206271068542994,
"grad_norm": 1.3203125,
"learning_rate": 0.0004980627644251926,
"loss": 6.5497,
"mean_token_accuracy": 0.1293226294219494,
"num_tokens": 10528175.0,
"step": 4865
},
{
"entropy": 6.651979970932007,
"epoch": 0.5211621809620632,
"grad_norm": 1.3203125,
"learning_rate": 0.0004980577548619489,
"loss": 6.6329,
"mean_token_accuracy": 0.11742270663380623,
"num_tokens": 10540076.0,
"step": 4870
},
{
"entropy": 6.686881160736084,
"epoch": 0.5216972550698272,
"grad_norm": 1.25,
"learning_rate": 0.0004980527388579437,
"loss": 6.6555,
"mean_token_accuracy": 0.11610461995005608,
"num_tokens": 10552025.0,
"step": 4875
},
{
"entropy": 6.746124219894409,
"epoch": 0.5222323291775911,
"grad_norm": 1.1875,
"learning_rate": 0.0004980477164133221,
"loss": 6.6528,
"mean_token_accuracy": 0.11348175257444382,
"num_tokens": 10562535.0,
"step": 4880
},
{
"entropy": 6.723954057693481,
"epoch": 0.522767403285355,
"grad_norm": 1.2421875,
"learning_rate": 0.000498042687528229,
"loss": 6.574,
"mean_token_accuracy": 0.12090551033616066,
"num_tokens": 10572989.0,
"step": 4885
},
{
"entropy": 6.658336591720581,
"epoch": 0.523302477393119,
"grad_norm": 1.234375,
"learning_rate": 0.0004980376522028098,
"loss": 6.6058,
"mean_token_accuracy": 0.12160259932279587,
"num_tokens": 10583033.0,
"step": 4890
},
{
"entropy": 6.734746789932251,
"epoch": 0.5238375515008828,
"grad_norm": 1.6015625,
"learning_rate": 0.0004980326104372095,
"loss": 6.6853,
"mean_token_accuracy": 0.11474591270089149,
"num_tokens": 10595205.0,
"step": 4895
},
{
"entropy": 6.786480188369751,
"epoch": 0.5243726256086468,
"grad_norm": 1.4609375,
"learning_rate": 0.000498027562231574,
"loss": 6.5392,
"mean_token_accuracy": 0.12003358751535416,
"num_tokens": 10605244.0,
"step": 4900
},
{
"entropy": 6.718193483352661,
"epoch": 0.5249076997164107,
"grad_norm": 1.21875,
"learning_rate": 0.000498022507586049,
"loss": 6.6214,
"mean_token_accuracy": 0.11722708195447921,
"num_tokens": 10615817.0,
"step": 4905
},
{
"entropy": 6.725398159027099,
"epoch": 0.5254427738241747,
"grad_norm": 1.2265625,
"learning_rate": 0.0004980174465007804,
"loss": 6.6366,
"mean_token_accuracy": 0.11559919267892838,
"num_tokens": 10627160.0,
"step": 4910
},
{
"entropy": 6.766883087158203,
"epoch": 0.5259778479319386,
"grad_norm": 1.390625,
"learning_rate": 0.0004980123789759145,
"loss": 6.631,
"mean_token_accuracy": 0.12066573053598403,
"num_tokens": 10637552.0,
"step": 4915
},
{
"entropy": 6.666731262207032,
"epoch": 0.5265129220397025,
"grad_norm": 1.7734375,
"learning_rate": 0.0004980073050115973,
"loss": 6.6807,
"mean_token_accuracy": 0.11911093890666961,
"num_tokens": 10648773.0,
"step": 4920
},
{
"entropy": 6.761361503601075,
"epoch": 0.5270479961474664,
"grad_norm": 1.40625,
"learning_rate": 0.0004980022246079755,
"loss": 6.6387,
"mean_token_accuracy": 0.1175591915845871,
"num_tokens": 10659537.0,
"step": 4925
},
{
"entropy": 6.820301675796509,
"epoch": 0.5275830702552303,
"grad_norm": 1.234375,
"learning_rate": 0.000497997137765196,
"loss": 6.6932,
"mean_token_accuracy": 0.11372315660119056,
"num_tokens": 10671390.0,
"step": 4930
},
{
"entropy": 6.718183946609497,
"epoch": 0.5281181443629943,
"grad_norm": 1.609375,
"learning_rate": 0.0004979920444834054,
"loss": 6.5349,
"mean_token_accuracy": 0.12883347496390343,
"num_tokens": 10682389.0,
"step": 4935
},
{
"entropy": 6.594164085388184,
"epoch": 0.5286532184707582,
"grad_norm": 1.296875,
"learning_rate": 0.0004979869447627508,
"loss": 6.606,
"mean_token_accuracy": 0.12021337449550629,
"num_tokens": 10692400.0,
"step": 4940
},
{
"entropy": 6.691273975372314,
"epoch": 0.5291882925785222,
"grad_norm": 1.2734375,
"learning_rate": 0.0004979818386033795,
"loss": 6.526,
"mean_token_accuracy": 0.12336401715874672,
"num_tokens": 10702396.0,
"step": 4945
},
{
"entropy": 6.723889923095703,
"epoch": 0.5297233666862861,
"grad_norm": 1.7265625,
"learning_rate": 0.000497976726005439,
"loss": 6.6549,
"mean_token_accuracy": 0.11776790469884872,
"num_tokens": 10712863.0,
"step": 4950
},
{
"entropy": 6.707800340652466,
"epoch": 0.5302584407940499,
"grad_norm": 1.2109375,
"learning_rate": 0.0004979716069690768,
"loss": 6.6556,
"mean_token_accuracy": 0.11826993152499199,
"num_tokens": 10724234.0,
"step": 4955
},
{
"entropy": 6.8028627872467045,
"epoch": 0.5307935149018139,
"grad_norm": 2.09375,
"learning_rate": 0.0004979664814944409,
"loss": 6.6898,
"mean_token_accuracy": 0.11730713248252869,
"num_tokens": 10734405.0,
"step": 4960
},
{
"entropy": 6.796906423568726,
"epoch": 0.5313285890095778,
"grad_norm": 1.40625,
"learning_rate": 0.0004979613495816791,
"loss": 6.6269,
"mean_token_accuracy": 0.11673597991466522,
"num_tokens": 10744573.0,
"step": 4965
},
{
"entropy": 6.81073899269104,
"epoch": 0.5318636631173418,
"grad_norm": 1.6171875,
"learning_rate": 0.0004979562112309397,
"loss": 6.6701,
"mean_token_accuracy": 0.12054512575268746,
"num_tokens": 10755555.0,
"step": 4970
},
{
"entropy": 6.69157977104187,
"epoch": 0.5323987372251057,
"grad_norm": 1.1640625,
"learning_rate": 0.000497951066442371,
"loss": 6.6797,
"mean_token_accuracy": 0.11503533869981766,
"num_tokens": 10766493.0,
"step": 4975
},
{
"entropy": 6.757311391830444,
"epoch": 0.5329338113328697,
"grad_norm": 1.3984375,
"learning_rate": 0.0004979459152161215,
"loss": 6.6385,
"mean_token_accuracy": 0.11236128360033035,
"num_tokens": 10776735.0,
"step": 4980
},
{
"entropy": 6.700515174865723,
"epoch": 0.5334688854406335,
"grad_norm": 1.2578125,
"learning_rate": 0.0004979407575523402,
"loss": 6.6145,
"mean_token_accuracy": 0.12050373330712319,
"num_tokens": 10787500.0,
"step": 4985
},
{
"entropy": 6.703393936157227,
"epoch": 0.5340039595483974,
"grad_norm": 1.609375,
"learning_rate": 0.0004979355934511757,
"loss": 6.6344,
"mean_token_accuracy": 0.12052067667245865,
"num_tokens": 10798002.0,
"step": 4990
},
{
"entropy": 6.75645604133606,
"epoch": 0.5345390336561614,
"grad_norm": 1.375,
"learning_rate": 0.0004979304229127773,
"loss": 6.6112,
"mean_token_accuracy": 0.11992594972252846,
"num_tokens": 10808982.0,
"step": 4995
},
{
"entropy": 6.736630868911743,
"epoch": 0.5350741077639253,
"grad_norm": 1.2421875,
"learning_rate": 0.0004979252459372944,
"loss": 6.5706,
"mean_token_accuracy": 0.11807732656598091,
"num_tokens": 10821619.0,
"step": 5000
},
{
"entropy": 6.699893808364868,
"epoch": 0.5356091818716893,
"grad_norm": 1.390625,
"learning_rate": 0.0004979200625248761,
"loss": 6.5904,
"mean_token_accuracy": 0.12047486156225204,
"num_tokens": 10832344.0,
"step": 5005
},
{
"entropy": 6.670212745666504,
"epoch": 0.5361442559794531,
"grad_norm": 1.4296875,
"learning_rate": 0.0004979148726756725,
"loss": 6.5571,
"mean_token_accuracy": 0.11860336735844612,
"num_tokens": 10843149.0,
"step": 5010
},
{
"entropy": 6.675074863433838,
"epoch": 0.5366793300872171,
"grad_norm": 1.6171875,
"learning_rate": 0.0004979096763898333,
"loss": 6.5998,
"mean_token_accuracy": 0.12452303320169449,
"num_tokens": 10854590.0,
"step": 5015
},
{
"entropy": 6.791008424758911,
"epoch": 0.537214404194981,
"grad_norm": 1.78125,
"learning_rate": 0.0004979044736675084,
"loss": 6.5671,
"mean_token_accuracy": 0.11767032742500305,
"num_tokens": 10865500.0,
"step": 5020
},
{
"entropy": 6.680721569061279,
"epoch": 0.5377494783027449,
"grad_norm": 1.9375,
"learning_rate": 0.0004978992645088483,
"loss": 6.6021,
"mean_token_accuracy": 0.12172115072607995,
"num_tokens": 10876866.0,
"step": 5025
},
{
"entropy": 6.5734340190887455,
"epoch": 0.5382845524105089,
"grad_norm": 1.265625,
"learning_rate": 0.000497894048914003,
"loss": 6.5612,
"mean_token_accuracy": 0.11841940134763718,
"num_tokens": 10888627.0,
"step": 5030
},
{
"entropy": 6.746210193634033,
"epoch": 0.5388196265182728,
"grad_norm": 1.9921875,
"learning_rate": 0.0004978888268831236,
"loss": 6.6696,
"mean_token_accuracy": 0.11132773160934448,
"num_tokens": 10900992.0,
"step": 5035
},
{
"entropy": 6.842794561386109,
"epoch": 0.5393547006260367,
"grad_norm": 1.40625,
"learning_rate": 0.0004978835984163606,
"loss": 6.7276,
"mean_token_accuracy": 0.10932021215558052,
"num_tokens": 10912323.0,
"step": 5040
},
{
"entropy": 6.800081539154053,
"epoch": 0.5398897747338006,
"grad_norm": 1.5078125,
"learning_rate": 0.0004978783635138649,
"loss": 6.6771,
"mean_token_accuracy": 0.11647720038890838,
"num_tokens": 10923991.0,
"step": 5045
},
{
"entropy": 6.639073657989502,
"epoch": 0.5404248488415646,
"grad_norm": 1.34375,
"learning_rate": 0.0004978731221757878,
"loss": 6.5041,
"mean_token_accuracy": 0.12417407482862472,
"num_tokens": 10934081.0,
"step": 5050
},
{
"entropy": 6.727223587036133,
"epoch": 0.5409599229493285,
"grad_norm": 1.5546875,
"learning_rate": 0.0004978678744022808,
"loss": 6.6006,
"mean_token_accuracy": 0.117810919880867,
"num_tokens": 10944729.0,
"step": 5055
},
{
"entropy": 6.709056615829468,
"epoch": 0.5414949970570924,
"grad_norm": 1.296875,
"learning_rate": 0.000497862620193495,
"loss": 6.6635,
"mean_token_accuracy": 0.11320299580693245,
"num_tokens": 10955708.0,
"step": 5060
},
{
"entropy": 6.72125997543335,
"epoch": 0.5420300711648564,
"grad_norm": 1.40625,
"learning_rate": 0.0004978573595495826,
"loss": 6.6002,
"mean_token_accuracy": 0.11878297626972198,
"num_tokens": 10966427.0,
"step": 5065
},
{
"entropy": 6.740125894546509,
"epoch": 0.5425651452726202,
"grad_norm": 1.9140625,
"learning_rate": 0.000497852092470695,
"loss": 6.6485,
"mean_token_accuracy": 0.11773342937231064,
"num_tokens": 10977741.0,
"step": 5070
},
{
"entropy": 6.7118854999542235,
"epoch": 0.5431002193803842,
"grad_norm": 1.5390625,
"learning_rate": 0.0004978468189569847,
"loss": 6.5956,
"mean_token_accuracy": 0.12482826039195061,
"num_tokens": 10989127.0,
"step": 5075
},
{
"entropy": 6.6592412948608395,
"epoch": 0.5436352934881481,
"grad_norm": 1.9140625,
"learning_rate": 0.0004978415390086038,
"loss": 6.6217,
"mean_token_accuracy": 0.11322500184178352,
"num_tokens": 10999791.0,
"step": 5080
},
{
"entropy": 6.783138751983643,
"epoch": 0.5441703675959121,
"grad_norm": 1.3515625,
"learning_rate": 0.0004978362526257047,
"loss": 6.6783,
"mean_token_accuracy": 0.11360784098505974,
"num_tokens": 11010045.0,
"step": 5085
},
{
"entropy": 6.735276794433593,
"epoch": 0.544705441703676,
"grad_norm": 1.515625,
"learning_rate": 0.0004978309598084403,
"loss": 6.6067,
"mean_token_accuracy": 0.11822862327098846,
"num_tokens": 11020729.0,
"step": 5090
},
{
"entropy": 6.708194351196289,
"epoch": 0.5452405158114398,
"grad_norm": 1.40625,
"learning_rate": 0.0004978256605569631,
"loss": 6.6289,
"mean_token_accuracy": 0.1177783451974392,
"num_tokens": 11030641.0,
"step": 5095
},
{
"entropy": 6.784500598907471,
"epoch": 0.5457755899192038,
"grad_norm": 1.3515625,
"learning_rate": 0.0004978203548714262,
"loss": 6.6366,
"mean_token_accuracy": 0.12379909828305244,
"num_tokens": 11041533.0,
"step": 5100
},
{
"entropy": 6.726690483093262,
"epoch": 0.5463106640269677,
"grad_norm": 1.265625,
"learning_rate": 0.000497815042751983,
"loss": 6.58,
"mean_token_accuracy": 0.11729968935251237,
"num_tokens": 11052170.0,
"step": 5105
},
{
"entropy": 6.652199029922485,
"epoch": 0.5468457381347317,
"grad_norm": 1.2421875,
"learning_rate": 0.0004978097241987868,
"loss": 6.5646,
"mean_token_accuracy": 0.1203967772424221,
"num_tokens": 11062872.0,
"step": 5110
},
{
"entropy": 6.719713973999023,
"epoch": 0.5473808122424956,
"grad_norm": 1.390625,
"learning_rate": 0.000497804399211991,
"loss": 6.6657,
"mean_token_accuracy": 0.11651256605982781,
"num_tokens": 11074199.0,
"step": 5115
},
{
"entropy": 6.7464769840240475,
"epoch": 0.5479158863502596,
"grad_norm": 1.3046875,
"learning_rate": 0.0004977990677917495,
"loss": 6.6129,
"mean_token_accuracy": 0.11894905865192414,
"num_tokens": 11083881.0,
"step": 5120
},
{
"entropy": 6.6402746677398685,
"epoch": 0.5484509604580234,
"grad_norm": 1.328125,
"learning_rate": 0.0004977937299382162,
"loss": 6.6584,
"mean_token_accuracy": 0.1151296466588974,
"num_tokens": 11094374.0,
"step": 5125
},
{
"entropy": 6.679858827590943,
"epoch": 0.5489860345657873,
"grad_norm": 3.578125,
"learning_rate": 0.0004977883856515453,
"loss": 6.4958,
"mean_token_accuracy": 0.1267719380557537,
"num_tokens": 11105174.0,
"step": 5130
},
{
"entropy": 6.751749181747437,
"epoch": 0.5495211086735513,
"grad_norm": 1.3984375,
"learning_rate": 0.000497783034931891,
"loss": 6.656,
"mean_token_accuracy": 0.11584192886948586,
"num_tokens": 11117223.0,
"step": 5135
},
{
"entropy": 6.7202818393707275,
"epoch": 0.5500561827813152,
"grad_norm": 1.5234375,
"learning_rate": 0.000497777677779408,
"loss": 6.579,
"mean_token_accuracy": 0.12366545721888542,
"num_tokens": 11127240.0,
"step": 5140
},
{
"entropy": 6.676993894577026,
"epoch": 0.5505912568890792,
"grad_norm": 1.3828125,
"learning_rate": 0.0004977723141942509,
"loss": 6.5678,
"mean_token_accuracy": 0.1281234510242939,
"num_tokens": 11137287.0,
"step": 5145
},
{
"entropy": 6.7371532917022705,
"epoch": 0.551126330996843,
"grad_norm": 1.2578125,
"learning_rate": 0.0004977669441765743,
"loss": 6.7007,
"mean_token_accuracy": 0.11768926754593849,
"num_tokens": 11148988.0,
"step": 5150
},
{
"entropy": 6.705612564086914,
"epoch": 0.551661405104607,
"grad_norm": 1.2734375,
"learning_rate": 0.0004977615677265336,
"loss": 6.6389,
"mean_token_accuracy": 0.12069165110588073,
"num_tokens": 11159874.0,
"step": 5155
},
{
"entropy": 6.73426923751831,
"epoch": 0.5521964792123709,
"grad_norm": 1.671875,
"learning_rate": 0.000497756184844284,
"loss": 6.6121,
"mean_token_accuracy": 0.12401502057909966,
"num_tokens": 11170703.0,
"step": 5160
},
{
"entropy": 6.744797801971435,
"epoch": 0.5527315533201348,
"grad_norm": 1.21875,
"learning_rate": 0.0004977507955299809,
"loss": 6.5001,
"mean_token_accuracy": 0.1257123127579689,
"num_tokens": 11182230.0,
"step": 5165
},
{
"entropy": 6.698776006698608,
"epoch": 0.5532666274278988,
"grad_norm": 1.4140625,
"learning_rate": 0.0004977453997837797,
"loss": 6.5677,
"mean_token_accuracy": 0.10981405228376388,
"num_tokens": 11192897.0,
"step": 5170
},
{
"entropy": 6.682893180847168,
"epoch": 0.5538017015356627,
"grad_norm": 1.2578125,
"learning_rate": 0.0004977399976058366,
"loss": 6.6521,
"mean_token_accuracy": 0.11399049237370491,
"num_tokens": 11203942.0,
"step": 5175
},
{
"entropy": 6.743161678314209,
"epoch": 0.5543367756434266,
"grad_norm": 1.203125,
"learning_rate": 0.0004977345889963072,
"loss": 6.7014,
"mean_token_accuracy": 0.11261942982673645,
"num_tokens": 11214361.0,
"step": 5180
},
{
"entropy": 6.6744081497192385,
"epoch": 0.5548718497511905,
"grad_norm": 1.2890625,
"learning_rate": 0.0004977291739553479,
"loss": 6.5902,
"mean_token_accuracy": 0.12366337925195695,
"num_tokens": 11225928.0,
"step": 5185
},
{
"entropy": 6.682669401168823,
"epoch": 0.5554069238589545,
"grad_norm": 1.03125,
"learning_rate": 0.0004977237524831149,
"loss": 6.5444,
"mean_token_accuracy": 0.11170822679996491,
"num_tokens": 11236660.0,
"step": 5190
},
{
"entropy": 6.714070177078247,
"epoch": 0.5559419979667184,
"grad_norm": 1.2734375,
"learning_rate": 0.0004977183245797649,
"loss": 6.5723,
"mean_token_accuracy": 0.1271256498992443,
"num_tokens": 11247272.0,
"step": 5195
},
{
"entropy": 6.77830867767334,
"epoch": 0.5564770720744823,
"grad_norm": 1.2734375,
"learning_rate": 0.0004977128902454547,
"loss": 6.6722,
"mean_token_accuracy": 0.11327713802456855,
"num_tokens": 11258472.0,
"step": 5200
},
{
"entropy": 6.7154069423675535,
"epoch": 0.5570121461822463,
"grad_norm": 1.3046875,
"learning_rate": 0.0004977074494803409,
"loss": 6.5841,
"mean_token_accuracy": 0.1255582146346569,
"num_tokens": 11268344.0,
"step": 5205
},
{
"entropy": 6.722737216949463,
"epoch": 0.5575472202900101,
"grad_norm": 1.2734375,
"learning_rate": 0.0004977020022845809,
"loss": 6.6762,
"mean_token_accuracy": 0.11227439492940902,
"num_tokens": 11279931.0,
"step": 5210
},
{
"entropy": 6.80668683052063,
"epoch": 0.5580822943977741,
"grad_norm": 1.3125,
"learning_rate": 0.0004976965486583318,
"loss": 6.6255,
"mean_token_accuracy": 0.11421727836132049,
"num_tokens": 11290595.0,
"step": 5215
},
{
"entropy": 6.7050305843353275,
"epoch": 0.558617368505538,
"grad_norm": 1.6015625,
"learning_rate": 0.0004976910886017513,
"loss": 6.6411,
"mean_token_accuracy": 0.12072176486253738,
"num_tokens": 11300774.0,
"step": 5220
},
{
"entropy": 6.6693981170654295,
"epoch": 0.559152442613302,
"grad_norm": 1.75,
"learning_rate": 0.0004976856221149969,
"loss": 6.5309,
"mean_token_accuracy": 0.12517891749739646,
"num_tokens": 11310288.0,
"step": 5225
},
{
"entropy": 6.700547170639038,
"epoch": 0.5596875167210659,
"grad_norm": 1.1484375,
"learning_rate": 0.0004976801491982263,
"loss": 6.6109,
"mean_token_accuracy": 0.11751497983932495,
"num_tokens": 11321599.0,
"step": 5230
},
{
"entropy": 6.786514472961426,
"epoch": 0.5602225908288297,
"grad_norm": 1.3828125,
"learning_rate": 0.0004976746698515977,
"loss": 6.6606,
"mean_token_accuracy": 0.11887592300772667,
"num_tokens": 11331383.0,
"step": 5235
},
{
"entropy": 6.712525081634522,
"epoch": 0.5607576649365937,
"grad_norm": 1.2421875,
"learning_rate": 0.0004976691840752694,
"loss": 6.6195,
"mean_token_accuracy": 0.11564537510275841,
"num_tokens": 11342085.0,
"step": 5240
},
{
"entropy": 6.718255138397216,
"epoch": 0.5612927390443576,
"grad_norm": 1.171875,
"learning_rate": 0.0004976636918693996,
"loss": 6.6574,
"mean_token_accuracy": 0.11577147841453553,
"num_tokens": 11354061.0,
"step": 5245
},
{
"entropy": 6.779119157791138,
"epoch": 0.5618278131521216,
"grad_norm": 1.3125,
"learning_rate": 0.0004976581932341471,
"loss": 6.6136,
"mean_token_accuracy": 0.12151364013552665,
"num_tokens": 11364448.0,
"step": 5250
},
{
"entropy": 6.769805908203125,
"epoch": 0.5623628872598855,
"grad_norm": 1.3046875,
"learning_rate": 0.0004976526881696706,
"loss": 6.5791,
"mean_token_accuracy": 0.12037570253014565,
"num_tokens": 11374838.0,
"step": 5255
},
{
"entropy": 6.629622316360473,
"epoch": 0.5628979613676495,
"grad_norm": 1.25,
"learning_rate": 0.0004976471766761288,
"loss": 6.6105,
"mean_token_accuracy": 0.12379851192235947,
"num_tokens": 11385310.0,
"step": 5260
},
{
"entropy": 6.691438817977906,
"epoch": 0.5634330354754133,
"grad_norm": 1.6328125,
"learning_rate": 0.0004976416587536812,
"loss": 6.6142,
"mean_token_accuracy": 0.11430059969425202,
"num_tokens": 11396590.0,
"step": 5265
},
{
"entropy": 6.742701053619385,
"epoch": 0.5639681095831772,
"grad_norm": 1.1875,
"learning_rate": 0.0004976361344024871,
"loss": 6.5707,
"mean_token_accuracy": 0.12226603850722313,
"num_tokens": 11407218.0,
"step": 5270
},
{
"entropy": 6.652452564239502,
"epoch": 0.5645031836909412,
"grad_norm": 1.5078125,
"learning_rate": 0.0004976306036227056,
"loss": 6.5931,
"mean_token_accuracy": 0.11696631386876107,
"num_tokens": 11417716.0,
"step": 5275
},
{
"entropy": 6.721119165420532,
"epoch": 0.5650382577987051,
"grad_norm": 1.3515625,
"learning_rate": 0.000497625066414497,
"loss": 6.6823,
"mean_token_accuracy": 0.11784773990511895,
"num_tokens": 11428295.0,
"step": 5280
},
{
"entropy": 6.797392892837524,
"epoch": 0.5655733319064691,
"grad_norm": 1.21875,
"learning_rate": 0.0004976195227780207,
"loss": 6.6173,
"mean_token_accuracy": 0.11284109130501747,
"num_tokens": 11439044.0,
"step": 5285
},
{
"entropy": 6.779853343963623,
"epoch": 0.566108406014233,
"grad_norm": 1.3515625,
"learning_rate": 0.0004976139727134371,
"loss": 6.5497,
"mean_token_accuracy": 0.11817987859249116,
"num_tokens": 11450029.0,
"step": 5290
},
{
"entropy": 6.715458774566651,
"epoch": 0.566643480121997,
"grad_norm": 1.21875,
"learning_rate": 0.0004976084162209062,
"loss": 6.7264,
"mean_token_accuracy": 0.10616886764764785,
"num_tokens": 11461190.0,
"step": 5295
},
{
"entropy": 6.745611715316772,
"epoch": 0.5671785542297608,
"grad_norm": 1.2421875,
"learning_rate": 0.0004976028533005886,
"loss": 6.5772,
"mean_token_accuracy": 0.11882699206471443,
"num_tokens": 11472143.0,
"step": 5300
},
{
"entropy": 6.771622133255005,
"epoch": 0.5677136283375247,
"grad_norm": 1.4765625,
"learning_rate": 0.0004975972839526449,
"loss": 6.7163,
"mean_token_accuracy": 0.1069483369588852,
"num_tokens": 11483745.0,
"step": 5305
},
{
"entropy": 6.7719972133636475,
"epoch": 0.5682487024452887,
"grad_norm": 1.265625,
"learning_rate": 0.0004975917081772358,
"loss": 6.5791,
"mean_token_accuracy": 0.1188987985253334,
"num_tokens": 11495695.0,
"step": 5310
},
{
"entropy": 6.718594884872436,
"epoch": 0.5687837765530526,
"grad_norm": 1.4140625,
"learning_rate": 0.0004975861259745225,
"loss": 6.6224,
"mean_token_accuracy": 0.11705151796340943,
"num_tokens": 11505576.0,
"step": 5315
},
{
"entropy": 6.661882305145264,
"epoch": 0.5693188506608166,
"grad_norm": 1.34375,
"learning_rate": 0.0004975805373446662,
"loss": 6.479,
"mean_token_accuracy": 0.12648551389575005,
"num_tokens": 11516614.0,
"step": 5320
},
{
"entropy": 6.7097954750061035,
"epoch": 0.5698539247685804,
"grad_norm": 1.359375,
"learning_rate": 0.000497574942287828,
"loss": 6.6111,
"mean_token_accuracy": 0.11843594536185265,
"num_tokens": 11528032.0,
"step": 5325
},
{
"entropy": 6.664940547943115,
"epoch": 0.5703889988763444,
"grad_norm": 1.609375,
"learning_rate": 0.0004975693408041697,
"loss": 6.6154,
"mean_token_accuracy": 0.11421292722225189,
"num_tokens": 11538822.0,
"step": 5330
},
{
"entropy": 6.699612236022949,
"epoch": 0.5709240729841083,
"grad_norm": 1.34375,
"learning_rate": 0.0004975637328938529,
"loss": 6.6399,
"mean_token_accuracy": 0.11097949370741844,
"num_tokens": 11548895.0,
"step": 5335
},
{
"entropy": 6.714142751693726,
"epoch": 0.5714591470918722,
"grad_norm": 1.328125,
"learning_rate": 0.0004975581185570398,
"loss": 6.5215,
"mean_token_accuracy": 0.11886597573757171,
"num_tokens": 11559613.0,
"step": 5340
},
{
"entropy": 6.713043832778931,
"epoch": 0.5719942211996362,
"grad_norm": 1.1640625,
"learning_rate": 0.0004975524977938921,
"loss": 6.5537,
"mean_token_accuracy": 0.12163764908909798,
"num_tokens": 11571092.0,
"step": 5345
},
{
"entropy": 6.692736101150513,
"epoch": 0.5725292953074,
"grad_norm": 1.4296875,
"learning_rate": 0.0004975468706045722,
"loss": 6.6814,
"mean_token_accuracy": 0.11672013476490975,
"num_tokens": 11581448.0,
"step": 5350
},
{
"entropy": 6.728510904312134,
"epoch": 0.573064369415164,
"grad_norm": 1.4375,
"learning_rate": 0.0004975412369892429,
"loss": 6.58,
"mean_token_accuracy": 0.12553823739290237,
"num_tokens": 11592273.0,
"step": 5355
},
{
"entropy": 6.757077264785766,
"epoch": 0.5735994435229279,
"grad_norm": 1.25,
"learning_rate": 0.0004975355969480665,
"loss": 6.5555,
"mean_token_accuracy": 0.125929856300354,
"num_tokens": 11602759.0,
"step": 5360
},
{
"entropy": 6.640834856033325,
"epoch": 0.5741345176306919,
"grad_norm": 1.3359375,
"learning_rate": 0.0004975299504812061,
"loss": 6.608,
"mean_token_accuracy": 0.11442179679870605,
"num_tokens": 11613977.0,
"step": 5365
},
{
"entropy": 6.71792254447937,
"epoch": 0.5746695917384558,
"grad_norm": 1.2734375,
"learning_rate": 0.0004975242975888246,
"loss": 6.636,
"mean_token_accuracy": 0.12084050849080086,
"num_tokens": 11625425.0,
"step": 5370
},
{
"entropy": 6.744046688079834,
"epoch": 0.5752046658462197,
"grad_norm": 1.3671875,
"learning_rate": 0.0004975186382710854,
"loss": 6.6071,
"mean_token_accuracy": 0.11423679888248443,
"num_tokens": 11636530.0,
"step": 5375
},
{
"entropy": 6.741583633422851,
"epoch": 0.5757397399539836,
"grad_norm": 1.3125,
"learning_rate": 0.0004975129725281516,
"loss": 6.5856,
"mean_token_accuracy": 0.12342779189348221,
"num_tokens": 11646516.0,
"step": 5380
},
{
"entropy": 6.701775741577149,
"epoch": 0.5762748140617475,
"grad_norm": 1.2890625,
"learning_rate": 0.0004975073003601871,
"loss": 6.5849,
"mean_token_accuracy": 0.11874212771654129,
"num_tokens": 11657424.0,
"step": 5385
},
{
"entropy": 6.704905319213867,
"epoch": 0.5768098881695115,
"grad_norm": 1.3125,
"learning_rate": 0.0004975016217673556,
"loss": 6.599,
"mean_token_accuracy": 0.12148372679948807,
"num_tokens": 11669447.0,
"step": 5390
},
{
"entropy": 6.697764110565186,
"epoch": 0.5773449622772754,
"grad_norm": 1.4140625,
"learning_rate": 0.000497495936749821,
"loss": 6.5675,
"mean_token_accuracy": 0.12611041441559792,
"num_tokens": 11679612.0,
"step": 5395
},
{
"entropy": 6.683344936370849,
"epoch": 0.5778800363850394,
"grad_norm": 1.3046875,
"learning_rate": 0.0004974902453077474,
"loss": 6.5514,
"mean_token_accuracy": 0.1218061052262783,
"num_tokens": 11689527.0,
"step": 5400
},
{
"entropy": 6.765136861801148,
"epoch": 0.5784151104928033,
"grad_norm": 1.3671875,
"learning_rate": 0.0004974845474412993,
"loss": 6.5856,
"mean_token_accuracy": 0.12035764157772064,
"num_tokens": 11700452.0,
"step": 5405
},
{
"entropy": 6.700014257431031,
"epoch": 0.5789501846005671,
"grad_norm": 1.625,
"learning_rate": 0.0004974788431506412,
"loss": 6.5971,
"mean_token_accuracy": 0.1191385418176651,
"num_tokens": 11710902.0,
"step": 5410
},
{
"entropy": 6.735801887512207,
"epoch": 0.5794852587083311,
"grad_norm": 1.484375,
"learning_rate": 0.0004974731324359378,
"loss": 6.6328,
"mean_token_accuracy": 0.10948319584131241,
"num_tokens": 11721901.0,
"step": 5415
},
{
"entropy": 6.765271472930908,
"epoch": 0.580020332816095,
"grad_norm": 1.2421875,
"learning_rate": 0.0004974674152973538,
"loss": 6.6146,
"mean_token_accuracy": 0.11682233065366746,
"num_tokens": 11733622.0,
"step": 5420
},
{
"entropy": 6.789870738983154,
"epoch": 0.580555406923859,
"grad_norm": 1.3359375,
"learning_rate": 0.0004974616917350545,
"loss": 6.6101,
"mean_token_accuracy": 0.11982190161943436,
"num_tokens": 11745116.0,
"step": 5425
},
{
"entropy": 6.661038208007812,
"epoch": 0.5810904810316229,
"grad_norm": 1.28125,
"learning_rate": 0.0004974559617492052,
"loss": 6.5446,
"mean_token_accuracy": 0.12379105761647224,
"num_tokens": 11756203.0,
"step": 5430
},
{
"entropy": 6.719126272201538,
"epoch": 0.5816255551393869,
"grad_norm": 1.375,
"learning_rate": 0.0004974502253399712,
"loss": 6.6041,
"mean_token_accuracy": 0.12229030430316926,
"num_tokens": 11766797.0,
"step": 5435
},
{
"entropy": 6.752104377746582,
"epoch": 0.5821606292471507,
"grad_norm": 1.3359375,
"learning_rate": 0.0004974444825075182,
"loss": 6.6896,
"mean_token_accuracy": 0.11470270082354546,
"num_tokens": 11778110.0,
"step": 5440
},
{
"entropy": 6.6274865627288815,
"epoch": 0.5826957033549146,
"grad_norm": 1.59375,
"learning_rate": 0.000497438733252012,
"loss": 6.551,
"mean_token_accuracy": 0.12010784074664116,
"num_tokens": 11788396.0,
"step": 5445
},
{
"entropy": 6.7564613819122314,
"epoch": 0.5832307774626786,
"grad_norm": 2.53125,
"learning_rate": 0.0004974329775736188,
"loss": 6.6412,
"mean_token_accuracy": 0.1130734585225582,
"num_tokens": 11798691.0,
"step": 5450
},
{
"entropy": 6.8176110744476315,
"epoch": 0.5837658515704425,
"grad_norm": 1.421875,
"learning_rate": 0.0004974272154725044,
"loss": 6.6809,
"mean_token_accuracy": 0.1145077645778656,
"num_tokens": 11808965.0,
"step": 5455
},
{
"entropy": 6.655050134658813,
"epoch": 0.5843009256782065,
"grad_norm": 1.3359375,
"learning_rate": 0.0004974214469488355,
"loss": 6.5144,
"mean_token_accuracy": 0.12288807928562165,
"num_tokens": 11819932.0,
"step": 5460
},
{
"entropy": 6.678020906448364,
"epoch": 0.5848359997859703,
"grad_norm": 1.5078125,
"learning_rate": 0.0004974156720027786,
"loss": 6.6115,
"mean_token_accuracy": 0.11556534245610237,
"num_tokens": 11831332.0,
"step": 5465
},
{
"entropy": 6.728974628448486,
"epoch": 0.5853710738937343,
"grad_norm": 1.53125,
"learning_rate": 0.0004974098906345004,
"loss": 6.5811,
"mean_token_accuracy": 0.11275865510106087,
"num_tokens": 11840947.0,
"step": 5470
},
{
"entropy": 6.7000159740448,
"epoch": 0.5859061480014982,
"grad_norm": 1.3046875,
"learning_rate": 0.0004974041028441679,
"loss": 6.5377,
"mean_token_accuracy": 0.11802567318081855,
"num_tokens": 11851640.0,
"step": 5475
},
{
"entropy": 6.733285045623779,
"epoch": 0.5864412221092621,
"grad_norm": 1.375,
"learning_rate": 0.0004973983086319481,
"loss": 6.6649,
"mean_token_accuracy": 0.11307286396622658,
"num_tokens": 11863608.0,
"step": 5480
},
{
"entropy": 6.722926664352417,
"epoch": 0.5869762962170261,
"grad_norm": 1.125,
"learning_rate": 0.0004973925079980085,
"loss": 6.5586,
"mean_token_accuracy": 0.11841953471302986,
"num_tokens": 11874707.0,
"step": 5485
},
{
"entropy": 6.687315177917481,
"epoch": 0.58751137032479,
"grad_norm": 1.59375,
"learning_rate": 0.0004973867009425164,
"loss": 6.659,
"mean_token_accuracy": 0.11953919008374214,
"num_tokens": 11885118.0,
"step": 5490
},
{
"entropy": 6.783655786514283,
"epoch": 0.5880464444325539,
"grad_norm": 1.2578125,
"learning_rate": 0.0004973808874656396,
"loss": 6.6231,
"mean_token_accuracy": 0.11821292191743851,
"num_tokens": 11895859.0,
"step": 5495
},
{
"entropy": 6.7643838882446286,
"epoch": 0.5885815185403178,
"grad_norm": 1.28125,
"learning_rate": 0.0004973750675675458,
"loss": 6.64,
"mean_token_accuracy": 0.11871552392840386,
"num_tokens": 11907019.0,
"step": 5500
},
{
"entropy": 6.729001569747925,
"epoch": 0.5891165926480818,
"grad_norm": 1.125,
"learning_rate": 0.0004973692412484034,
"loss": 6.5712,
"mean_token_accuracy": 0.1146469995379448,
"num_tokens": 11918257.0,
"step": 5505
},
{
"entropy": 6.721626091003418,
"epoch": 0.5896516667558457,
"grad_norm": 1.6015625,
"learning_rate": 0.0004973634085083802,
"loss": 6.5975,
"mean_token_accuracy": 0.12574032694101334,
"num_tokens": 11929003.0,
"step": 5510
},
{
"entropy": 6.757476425170898,
"epoch": 0.5901867408636096,
"grad_norm": 1.1640625,
"learning_rate": 0.000497357569347645,
"loss": 6.7213,
"mean_token_accuracy": 0.11348426267504692,
"num_tokens": 11939786.0,
"step": 5515
},
{
"entropy": 6.751842021942139,
"epoch": 0.5907218149713735,
"grad_norm": 1.1640625,
"learning_rate": 0.0004973517237663661,
"loss": 6.6311,
"mean_token_accuracy": 0.11161758229136468,
"num_tokens": 11951042.0,
"step": 5520
},
{
"entropy": 6.710853910446167,
"epoch": 0.5912568890791374,
"grad_norm": 1.953125,
"learning_rate": 0.0004973458717647124,
"loss": 6.5283,
"mean_token_accuracy": 0.12709658294916154,
"num_tokens": 11961982.0,
"step": 5525
},
{
"entropy": 6.703456449508667,
"epoch": 0.5917919631869014,
"grad_norm": 1.2890625,
"learning_rate": 0.000497340013342853,
"loss": 6.5923,
"mean_token_accuracy": 0.12106578648090363,
"num_tokens": 11972078.0,
"step": 5530
},
{
"entropy": 6.719264888763428,
"epoch": 0.5923270372946653,
"grad_norm": 1.609375,
"learning_rate": 0.000497334148500957,
"loss": 6.6446,
"mean_token_accuracy": 0.11806860640645027,
"num_tokens": 11984037.0,
"step": 5535
},
{
"entropy": 6.7681842803955075,
"epoch": 0.5928621114024293,
"grad_norm": 1.2890625,
"learning_rate": 0.0004973282772391936,
"loss": 6.5852,
"mean_token_accuracy": 0.11801839917898178,
"num_tokens": 11993688.0,
"step": 5540
},
{
"entropy": 6.667469120025634,
"epoch": 0.5933971855101932,
"grad_norm": 1.328125,
"learning_rate": 0.0004973223995577324,
"loss": 6.5477,
"mean_token_accuracy": 0.11981470957398414,
"num_tokens": 12003759.0,
"step": 5545
},
{
"entropy": 6.715129756927491,
"epoch": 0.593932259617957,
"grad_norm": 1.3125,
"learning_rate": 0.0004973165154567432,
"loss": 6.5858,
"mean_token_accuracy": 0.1203451156616211,
"num_tokens": 12015030.0,
"step": 5550
},
{
"entropy": 6.72750506401062,
"epoch": 0.594467333725721,
"grad_norm": 1.140625,
"learning_rate": 0.0004973106249363959,
"loss": 6.6509,
"mean_token_accuracy": 0.11390956789255142,
"num_tokens": 12026181.0,
"step": 5555
},
{
"entropy": 6.618847751617432,
"epoch": 0.5950024078334849,
"grad_norm": 1.28125,
"learning_rate": 0.0004973047279968606,
"loss": 6.4185,
"mean_token_accuracy": 0.13287391439080237,
"num_tokens": 12037336.0,
"step": 5560
},
{
"entropy": 6.65236268043518,
"epoch": 0.5955374819412489,
"grad_norm": 1.234375,
"learning_rate": 0.0004972988246383074,
"loss": 6.4918,
"mean_token_accuracy": 0.12340264916419982,
"num_tokens": 12048425.0,
"step": 5565
},
{
"entropy": 6.557363128662109,
"epoch": 0.5960725560490128,
"grad_norm": 1.65625,
"learning_rate": 0.0004972929148609068,
"loss": 6.4594,
"mean_token_accuracy": 0.13635959550738336,
"num_tokens": 12059814.0,
"step": 5570
},
{
"entropy": 6.68271279335022,
"epoch": 0.5966076301567768,
"grad_norm": 1.59375,
"learning_rate": 0.0004972869986648296,
"loss": 6.54,
"mean_token_accuracy": 0.12527887672185897,
"num_tokens": 12069377.0,
"step": 5575
},
{
"entropy": 6.707939195632934,
"epoch": 0.5971427042645406,
"grad_norm": 1.2890625,
"learning_rate": 0.0004972810760502467,
"loss": 6.666,
"mean_token_accuracy": 0.11514274403452873,
"num_tokens": 12080277.0,
"step": 5580
},
{
"entropy": 6.727867317199707,
"epoch": 0.5976777783723045,
"grad_norm": 1.21875,
"learning_rate": 0.0004972751470173287,
"loss": 6.5453,
"mean_token_accuracy": 0.12022876664996147,
"num_tokens": 12091375.0,
"step": 5585
},
{
"entropy": 6.676542520523071,
"epoch": 0.5982128524800685,
"grad_norm": 1.203125,
"learning_rate": 0.0004972692115662473,
"loss": 6.5893,
"mean_token_accuracy": 0.11981748417019844,
"num_tokens": 12103110.0,
"step": 5590
},
{
"entropy": 6.751420688629151,
"epoch": 0.5987479265878324,
"grad_norm": 1.7421875,
"learning_rate": 0.0004972632696971735,
"loss": 6.6451,
"mean_token_accuracy": 0.11599904671311378,
"num_tokens": 12112454.0,
"step": 5595
},
{
"entropy": 6.6340423107147215,
"epoch": 0.5992830006955964,
"grad_norm": 1.28125,
"learning_rate": 0.0004972573214102791,
"loss": 6.5246,
"mean_token_accuracy": 0.11687616854906083,
"num_tokens": 12125140.0,
"step": 5600
},
{
"entropy": 6.698513317108154,
"epoch": 0.5998180748033602,
"grad_norm": 1.8046875,
"learning_rate": 0.0004972513667057358,
"loss": 6.5691,
"mean_token_accuracy": 0.12039392963051795,
"num_tokens": 12136061.0,
"step": 5605
},
{
"entropy": 6.791307878494263,
"epoch": 0.6003531489111242,
"grad_norm": 1.171875,
"learning_rate": 0.0004972454055837155,
"loss": 6.6003,
"mean_token_accuracy": 0.12150818780064583,
"num_tokens": 12146654.0,
"step": 5610
},
{
"entropy": 6.723158931732177,
"epoch": 0.6008882230188881,
"grad_norm": 1.390625,
"learning_rate": 0.0004972394380443903,
"loss": 6.5878,
"mean_token_accuracy": 0.11757106855511665,
"num_tokens": 12157881.0,
"step": 5615
},
{
"entropy": 6.675697088241577,
"epoch": 0.601423297126652,
"grad_norm": 1.671875,
"learning_rate": 0.0004972334640879325,
"loss": 6.6085,
"mean_token_accuracy": 0.11922492906451225,
"num_tokens": 12168534.0,
"step": 5620
},
{
"entropy": 6.661304664611817,
"epoch": 0.601958371234416,
"grad_norm": 1.2421875,
"learning_rate": 0.0004972274837145147,
"loss": 6.5432,
"mean_token_accuracy": 0.11988529115915299,
"num_tokens": 12178773.0,
"step": 5625
},
{
"entropy": 6.739487743377685,
"epoch": 0.6024934453421799,
"grad_norm": 1.1875,
"learning_rate": 0.0004972214969243096,
"loss": 6.5808,
"mean_token_accuracy": 0.11824664920568466,
"num_tokens": 12188819.0,
"step": 5630
},
{
"entropy": 6.660564947128296,
"epoch": 0.6030285194499438,
"grad_norm": 1.2265625,
"learning_rate": 0.00049721550371749,
"loss": 6.5607,
"mean_token_accuracy": 0.12463830485939979,
"num_tokens": 12199198.0,
"step": 5635
},
{
"entropy": 6.6687784671783445,
"epoch": 0.6035635935577077,
"grad_norm": 1.1015625,
"learning_rate": 0.0004972095040942289,
"loss": 6.5644,
"mean_token_accuracy": 0.1162213332951069,
"num_tokens": 12210765.0,
"step": 5640
},
{
"entropy": 6.709344291687012,
"epoch": 0.6040986676654717,
"grad_norm": 1.71875,
"learning_rate": 0.0004972034980546996,
"loss": 6.6061,
"mean_token_accuracy": 0.11544334441423416,
"num_tokens": 12221067.0,
"step": 5645
},
{
"entropy": 6.653585481643677,
"epoch": 0.6046337417732356,
"grad_norm": 1.1875,
"learning_rate": 0.0004971974855990755,
"loss": 6.6023,
"mean_token_accuracy": 0.11971136629581451,
"num_tokens": 12232994.0,
"step": 5650
},
{
"entropy": 6.738733911514283,
"epoch": 0.6051688158809995,
"grad_norm": 1.4140625,
"learning_rate": 0.0004971914667275302,
"loss": 6.5538,
"mean_token_accuracy": 0.12072688415646553,
"num_tokens": 12243327.0,
"step": 5655
},
{
"entropy": 6.6140237808227536,
"epoch": 0.6057038899887635,
"grad_norm": 1.21875,
"learning_rate": 0.0004971854414402377,
"loss": 6.4752,
"mean_token_accuracy": 0.12842355743050576,
"num_tokens": 12254249.0,
"step": 5660
},
{
"entropy": 6.662600469589234,
"epoch": 0.6062389640965273,
"grad_norm": 1.296875,
"learning_rate": 0.0004971794097373716,
"loss": 6.5208,
"mean_token_accuracy": 0.12328208535909653,
"num_tokens": 12264487.0,
"step": 5665
},
{
"entropy": 6.569520473480225,
"epoch": 0.6067740382042913,
"grad_norm": 1.375,
"learning_rate": 0.0004971733716191063,
"loss": 6.5189,
"mean_token_accuracy": 0.1292000360786915,
"num_tokens": 12274801.0,
"step": 5670
},
{
"entropy": 6.636984205245971,
"epoch": 0.6073091123120552,
"grad_norm": 1.4375,
"learning_rate": 0.0004971673270856162,
"loss": 6.4777,
"mean_token_accuracy": 0.11988460049033164,
"num_tokens": 12284647.0,
"step": 5675
},
{
"entropy": 6.658068609237671,
"epoch": 0.6078441864198192,
"grad_norm": 1.2578125,
"learning_rate": 0.0004971612761370756,
"loss": 6.6033,
"mean_token_accuracy": 0.11951280906796455,
"num_tokens": 12295215.0,
"step": 5680
},
{
"entropy": 6.673429822921753,
"epoch": 0.6083792605275831,
"grad_norm": 1.4609375,
"learning_rate": 0.0004971552187736596,
"loss": 6.516,
"mean_token_accuracy": 0.12064156159758568,
"num_tokens": 12305324.0,
"step": 5685
},
{
"entropy": 6.668921661376953,
"epoch": 0.6089143346353469,
"grad_norm": 1.4375,
"learning_rate": 0.0004971491549955427,
"loss": 6.5372,
"mean_token_accuracy": 0.12246538251638413,
"num_tokens": 12316399.0,
"step": 5690
},
{
"entropy": 6.6165331363677975,
"epoch": 0.6094494087431109,
"grad_norm": 1.2265625,
"learning_rate": 0.0004971430848029002,
"loss": 6.5219,
"mean_token_accuracy": 0.12773501947522165,
"num_tokens": 12328033.0,
"step": 5695
},
{
"entropy": 6.681158256530762,
"epoch": 0.6099844828508748,
"grad_norm": 1.3984375,
"learning_rate": 0.0004971370081959073,
"loss": 6.5622,
"mean_token_accuracy": 0.12176148667931556,
"num_tokens": 12338962.0,
"step": 5700
},
{
"entropy": 6.697946166992187,
"epoch": 0.6105195569586388,
"grad_norm": 1.6328125,
"learning_rate": 0.0004971309251747396,
"loss": 6.5526,
"mean_token_accuracy": 0.11940340176224709,
"num_tokens": 12350216.0,
"step": 5705
},
{
"entropy": 6.745215272903442,
"epoch": 0.6110546310664027,
"grad_norm": 1.171875,
"learning_rate": 0.0004971248357395726,
"loss": 6.5806,
"mean_token_accuracy": 0.11893382146954537,
"num_tokens": 12361277.0,
"step": 5710
},
{
"entropy": 6.580714225769043,
"epoch": 0.6115897051741667,
"grad_norm": 1.4765625,
"learning_rate": 0.0004971187398905821,
"loss": 6.4858,
"mean_token_accuracy": 0.12100339159369469,
"num_tokens": 12373308.0,
"step": 5715
},
{
"entropy": 6.731028461456299,
"epoch": 0.6121247792819305,
"grad_norm": 1.3203125,
"learning_rate": 0.0004971126376279443,
"loss": 6.7259,
"mean_token_accuracy": 0.11148046106100082,
"num_tokens": 12383943.0,
"step": 5720
},
{
"entropy": 6.721910047531128,
"epoch": 0.6126598533896944,
"grad_norm": 1.3125,
"learning_rate": 0.0004971065289518354,
"loss": 6.482,
"mean_token_accuracy": 0.12279203087091446,
"num_tokens": 12394575.0,
"step": 5725
},
{
"entropy": 6.715807628631592,
"epoch": 0.6131949274974584,
"grad_norm": 1.265625,
"learning_rate": 0.0004971004138624315,
"loss": 6.5895,
"mean_token_accuracy": 0.11779340729117393,
"num_tokens": 12405536.0,
"step": 5730
},
{
"entropy": 6.606629800796509,
"epoch": 0.6137300016052223,
"grad_norm": 1.7109375,
"learning_rate": 0.0004970942923599094,
"loss": 6.4743,
"mean_token_accuracy": 0.12641754895448684,
"num_tokens": 12416148.0,
"step": 5735
},
{
"entropy": 6.612427473068237,
"epoch": 0.6142650757129863,
"grad_norm": 1.3828125,
"learning_rate": 0.0004970881644444459,
"loss": 6.48,
"mean_token_accuracy": 0.12701191157102584,
"num_tokens": 12426182.0,
"step": 5740
},
{
"entropy": 6.647663402557373,
"epoch": 0.6148001498207502,
"grad_norm": 1.1796875,
"learning_rate": 0.0004970820301162178,
"loss": 6.6625,
"mean_token_accuracy": 0.11574866473674775,
"num_tokens": 12437548.0,
"step": 5745
},
{
"entropy": 6.737516117095947,
"epoch": 0.6153352239285141,
"grad_norm": 1.7890625,
"learning_rate": 0.0004970758893754022,
"loss": 6.5979,
"mean_token_accuracy": 0.11561758294701577,
"num_tokens": 12449720.0,
"step": 5750
},
{
"entropy": 6.817887496948242,
"epoch": 0.615870298036278,
"grad_norm": 1.140625,
"learning_rate": 0.0004970697422221767,
"loss": 6.574,
"mean_token_accuracy": 0.1215938724577427,
"num_tokens": 12460666.0,
"step": 5755
},
{
"entropy": 6.7063861846923825,
"epoch": 0.6164053721440419,
"grad_norm": 1.3125,
"learning_rate": 0.0004970635886567185,
"loss": 6.5911,
"mean_token_accuracy": 0.12281558737158775,
"num_tokens": 12471472.0,
"step": 5760
},
{
"entropy": 6.611774969100952,
"epoch": 0.6169404462518059,
"grad_norm": 2.953125,
"learning_rate": 0.0004970574286792054,
"loss": 6.5681,
"mean_token_accuracy": 0.11770694702863693,
"num_tokens": 12482127.0,
"step": 5765
},
{
"entropy": 6.611656904220581,
"epoch": 0.6174755203595698,
"grad_norm": 1.34375,
"learning_rate": 0.0004970512622898152,
"loss": 6.5767,
"mean_token_accuracy": 0.12225481644272804,
"num_tokens": 12493680.0,
"step": 5770
},
{
"entropy": 6.65736231803894,
"epoch": 0.6180105944673338,
"grad_norm": 1.2734375,
"learning_rate": 0.000497045089488726,
"loss": 6.523,
"mean_token_accuracy": 0.12167870178818703,
"num_tokens": 12505838.0,
"step": 5775
},
{
"entropy": 6.704753160476685,
"epoch": 0.6185456685750976,
"grad_norm": 1.5078125,
"learning_rate": 0.0004970389102761161,
"loss": 6.5102,
"mean_token_accuracy": 0.12942416369915008,
"num_tokens": 12516334.0,
"step": 5780
},
{
"entropy": 6.661266231536866,
"epoch": 0.6190807426828616,
"grad_norm": 1.3984375,
"learning_rate": 0.0004970327246521638,
"loss": 6.5879,
"mean_token_accuracy": 0.12000072821974754,
"num_tokens": 12526440.0,
"step": 5785
},
{
"entropy": 6.680879926681518,
"epoch": 0.6196158167906255,
"grad_norm": 1.359375,
"learning_rate": 0.0004970265326170478,
"loss": 6.5266,
"mean_token_accuracy": 0.11807752102613449,
"num_tokens": 12538127.0,
"step": 5790
},
{
"entropy": 6.722283792495728,
"epoch": 0.6201508908983894,
"grad_norm": 1.34375,
"learning_rate": 0.0004970203341709469,
"loss": 6.5914,
"mean_token_accuracy": 0.12216417640447616,
"num_tokens": 12548459.0,
"step": 5795
},
{
"entropy": 6.58387131690979,
"epoch": 0.6206859650061534,
"grad_norm": 1.28125,
"learning_rate": 0.00049701412931404,
"loss": 6.5481,
"mean_token_accuracy": 0.12659149914979934,
"num_tokens": 12558989.0,
"step": 5800
},
{
"entropy": 6.62721471786499,
"epoch": 0.6212210391139172,
"grad_norm": 1.2109375,
"learning_rate": 0.0004970079180465064,
"loss": 6.4916,
"mean_token_accuracy": 0.1283690959215164,
"num_tokens": 12569449.0,
"step": 5805
},
{
"entropy": 6.644355916976929,
"epoch": 0.6217561132216812,
"grad_norm": 1.296875,
"learning_rate": 0.0004970017003685253,
"loss": 6.5297,
"mean_token_accuracy": 0.12029706984758377,
"num_tokens": 12580733.0,
"step": 5810
},
{
"entropy": 6.65497670173645,
"epoch": 0.6222911873294451,
"grad_norm": 1.3203125,
"learning_rate": 0.0004969954762802763,
"loss": 6.5326,
"mean_token_accuracy": 0.11786738261580468,
"num_tokens": 12590869.0,
"step": 5815
},
{
"entropy": 6.706523609161377,
"epoch": 0.6228262614372091,
"grad_norm": 1.40625,
"learning_rate": 0.000496989245781939,
"loss": 6.6305,
"mean_token_accuracy": 0.11878098174929619,
"num_tokens": 12602109.0,
"step": 5820
},
{
"entropy": 6.690067100524902,
"epoch": 0.623361335544973,
"grad_norm": 1.2578125,
"learning_rate": 0.0004969830088736937,
"loss": 6.5463,
"mean_token_accuracy": 0.11708159074187278,
"num_tokens": 12612234.0,
"step": 5825
},
{
"entropy": 6.761075687408447,
"epoch": 0.6238964096527368,
"grad_norm": 1.3125,
"learning_rate": 0.00049697676555572,
"loss": 6.6967,
"mean_token_accuracy": 0.11268940791487694,
"num_tokens": 12623252.0,
"step": 5830
},
{
"entropy": 6.709875154495239,
"epoch": 0.6244314837605008,
"grad_norm": 1.25,
"learning_rate": 0.0004969705158281985,
"loss": 6.4508,
"mean_token_accuracy": 0.12307255640625954,
"num_tokens": 12632796.0,
"step": 5835
},
{
"entropy": 6.652636289596558,
"epoch": 0.6249665578682647,
"grad_norm": 1.65625,
"learning_rate": 0.0004969642596913095,
"loss": 6.5512,
"mean_token_accuracy": 0.11847778558731079,
"num_tokens": 12643278.0,
"step": 5840
},
{
"entropy": 6.670278692245484,
"epoch": 0.6255016319760287,
"grad_norm": 5.25,
"learning_rate": 0.0004969579971452337,
"loss": 6.6218,
"mean_token_accuracy": 0.12104167938232421,
"num_tokens": 12654446.0,
"step": 5845
},
{
"entropy": 6.704672002792359,
"epoch": 0.6260367060837926,
"grad_norm": 1.4296875,
"learning_rate": 0.0004969517281901519,
"loss": 6.5988,
"mean_token_accuracy": 0.12173526436090469,
"num_tokens": 12664281.0,
"step": 5850
},
{
"entropy": 6.726960182189941,
"epoch": 0.6265717801915566,
"grad_norm": 1.671875,
"learning_rate": 0.0004969454528262451,
"loss": 6.5829,
"mean_token_accuracy": 0.11787922754883766,
"num_tokens": 12675252.0,
"step": 5855
},
{
"entropy": 6.665972471237183,
"epoch": 0.6271068542993204,
"grad_norm": 1.9765625,
"learning_rate": 0.0004969391710536946,
"loss": 6.5794,
"mean_token_accuracy": 0.1184937097132206,
"num_tokens": 12685764.0,
"step": 5860
},
{
"entropy": 6.591383934020996,
"epoch": 0.6276419284070843,
"grad_norm": 1.6328125,
"learning_rate": 0.0004969328828726817,
"loss": 6.4757,
"mean_token_accuracy": 0.12730447500944136,
"num_tokens": 12696974.0,
"step": 5865
},
{
"entropy": 6.757170295715332,
"epoch": 0.6281770025148483,
"grad_norm": 1.453125,
"learning_rate": 0.0004969265882833879,
"loss": 6.6372,
"mean_token_accuracy": 0.11200533658266068,
"num_tokens": 12708217.0,
"step": 5870
},
{
"entropy": 6.707298660278321,
"epoch": 0.6287120766226122,
"grad_norm": 1.6953125,
"learning_rate": 0.0004969202872859952,
"loss": 6.516,
"mean_token_accuracy": 0.12099341303110123,
"num_tokens": 12717840.0,
"step": 5875
},
{
"entropy": 6.622361516952514,
"epoch": 0.6292471507303762,
"grad_norm": 1.3828125,
"learning_rate": 0.0004969139798806854,
"loss": 6.6088,
"mean_token_accuracy": 0.11309082061052322,
"num_tokens": 12728390.0,
"step": 5880
},
{
"entropy": 6.653683233261108,
"epoch": 0.6297822248381401,
"grad_norm": 1.2734375,
"learning_rate": 0.0004969076660676405,
"loss": 6.6113,
"mean_token_accuracy": 0.11578750982880592,
"num_tokens": 12738650.0,
"step": 5885
},
{
"entropy": 6.735027837753296,
"epoch": 0.630317298945904,
"grad_norm": 1.484375,
"learning_rate": 0.000496901345847043,
"loss": 6.5966,
"mean_token_accuracy": 0.11822382658720017,
"num_tokens": 12748989.0,
"step": 5890
},
{
"entropy": 6.689414024353027,
"epoch": 0.6308523730536679,
"grad_norm": 1.2578125,
"learning_rate": 0.0004968950192190752,
"loss": 6.559,
"mean_token_accuracy": 0.12090246379375458,
"num_tokens": 12761132.0,
"step": 5895
},
{
"entropy": 6.6849761486053465,
"epoch": 0.6313874471614318,
"grad_norm": 1.25,
"learning_rate": 0.00049688868618392,
"loss": 6.5651,
"mean_token_accuracy": 0.1228780284523964,
"num_tokens": 12771478.0,
"step": 5900
},
{
"entropy": 6.716721773147583,
"epoch": 0.6319225212691958,
"grad_norm": 1.2109375,
"learning_rate": 0.0004968823467417602,
"loss": 6.6244,
"mean_token_accuracy": 0.1174764983355999,
"num_tokens": 12782096.0,
"step": 5905
},
{
"entropy": 6.716968536376953,
"epoch": 0.6324575953769597,
"grad_norm": 1.171875,
"learning_rate": 0.0004968760008927788,
"loss": 6.5262,
"mean_token_accuracy": 0.12529518380761145,
"num_tokens": 12792467.0,
"step": 5910
},
{
"entropy": 6.6767195701599125,
"epoch": 0.6329926694847237,
"grad_norm": 1.25,
"learning_rate": 0.000496869648637159,
"loss": 6.5373,
"mean_token_accuracy": 0.12640023306012155,
"num_tokens": 12803248.0,
"step": 5915
},
{
"entropy": 6.697324562072754,
"epoch": 0.6335277435924875,
"grad_norm": 1.3125,
"learning_rate": 0.0004968632899750844,
"loss": 6.5591,
"mean_token_accuracy": 0.12147547379136085,
"num_tokens": 12813701.0,
"step": 5920
},
{
"entropy": 6.800759792327881,
"epoch": 0.6340628177002515,
"grad_norm": 1.4453125,
"learning_rate": 0.0004968569249067383,
"loss": 6.6272,
"mean_token_accuracy": 0.11679861545562745,
"num_tokens": 12824945.0,
"step": 5925
},
{
"entropy": 6.671930313110352,
"epoch": 0.6345978918080154,
"grad_norm": 1.09375,
"learning_rate": 0.0004968505534323047,
"loss": 6.5786,
"mean_token_accuracy": 0.1162982702255249,
"num_tokens": 12836699.0,
"step": 5930
},
{
"entropy": 6.6336596488952635,
"epoch": 0.6351329659157794,
"grad_norm": 1.28125,
"learning_rate": 0.0004968441755519677,
"loss": 6.5291,
"mean_token_accuracy": 0.11951676458120346,
"num_tokens": 12846905.0,
"step": 5935
},
{
"entropy": 6.7043437480926515,
"epoch": 0.6356680400235433,
"grad_norm": 1.375,
"learning_rate": 0.0004968377912659112,
"loss": 6.5001,
"mean_token_accuracy": 0.12340181395411491,
"num_tokens": 12856919.0,
"step": 5940
},
{
"entropy": 6.60842661857605,
"epoch": 0.6362031141313071,
"grad_norm": 1.7109375,
"learning_rate": 0.0004968314005743196,
"loss": 6.4565,
"mean_token_accuracy": 0.12368626967072487,
"num_tokens": 12867863.0,
"step": 5945
},
{
"entropy": 6.611450290679931,
"epoch": 0.6367381882390711,
"grad_norm": 1.1796875,
"learning_rate": 0.0004968250034773776,
"loss": 6.5429,
"mean_token_accuracy": 0.12223007977008819,
"num_tokens": 12877877.0,
"step": 5950
},
{
"entropy": 6.737860918045044,
"epoch": 0.637273262346835,
"grad_norm": 1.59375,
"learning_rate": 0.0004968185999752697,
"loss": 6.6383,
"mean_token_accuracy": 0.11434445828199387,
"num_tokens": 12887954.0,
"step": 5955
},
{
"entropy": 6.707271003723145,
"epoch": 0.637808336454599,
"grad_norm": 1.1875,
"learning_rate": 0.0004968121900681809,
"loss": 6.5511,
"mean_token_accuracy": 0.12070679739117622,
"num_tokens": 12897744.0,
"step": 5960
},
{
"entropy": 6.68266224861145,
"epoch": 0.6383434105623629,
"grad_norm": 1.640625,
"learning_rate": 0.0004968057737562964,
"loss": 6.617,
"mean_token_accuracy": 0.12236501500010491,
"num_tokens": 12909193.0,
"step": 5965
},
{
"entropy": 6.663351440429688,
"epoch": 0.6388784846701269,
"grad_norm": 1.3125,
"learning_rate": 0.0004967993510398012,
"loss": 6.4973,
"mean_token_accuracy": 0.12318923175334931,
"num_tokens": 12919879.0,
"step": 5970
},
{
"entropy": 6.623578882217407,
"epoch": 0.6394135587778907,
"grad_norm": 1.2734375,
"learning_rate": 0.0004967929219188809,
"loss": 6.5387,
"mean_token_accuracy": 0.12540650963783265,
"num_tokens": 12930297.0,
"step": 5975
},
{
"entropy": 6.649103832244873,
"epoch": 0.6399486328856546,
"grad_norm": 1.375,
"learning_rate": 0.0004967864863937212,
"loss": 6.5329,
"mean_token_accuracy": 0.12860941588878633,
"num_tokens": 12940467.0,
"step": 5980
},
{
"entropy": 6.6491728782653805,
"epoch": 0.6404837069934186,
"grad_norm": 1.4296875,
"learning_rate": 0.0004967800444645079,
"loss": 6.5204,
"mean_token_accuracy": 0.11868370845913886,
"num_tokens": 12950417.0,
"step": 5985
},
{
"entropy": 6.636889123916626,
"epoch": 0.6410187811011825,
"grad_norm": 1.4765625,
"learning_rate": 0.0004967735961314269,
"loss": 6.4761,
"mean_token_accuracy": 0.122308399528265,
"num_tokens": 12960957.0,
"step": 5990
},
{
"entropy": 6.666227865219116,
"epoch": 0.6415538552089465,
"grad_norm": 1.125,
"learning_rate": 0.0004967671413946646,
"loss": 6.575,
"mean_token_accuracy": 0.12026591822504998,
"num_tokens": 12971194.0,
"step": 5995
},
{
"entropy": 6.6303346157073975,
"epoch": 0.6420889293167104,
"grad_norm": 1.5390625,
"learning_rate": 0.0004967606802544071,
"loss": 6.5736,
"mean_token_accuracy": 0.11933250203728676,
"num_tokens": 12982479.0,
"step": 6000
},
{
"epoch": 0.6420889293167104,
"eval_entropy": 6.587495564643263,
"eval_loss": 6.633632183074951,
"eval_mean_token_accuracy": 0.123240454485414,
"eval_num_tokens": 12982479.0,
"eval_runtime": 22.6134,
"eval_samples_per_second": 1312.45,
"eval_steps_per_second": 164.062,
"step": 6000
},
{
"entropy": 6.708343410491944,
"epoch": 0.6426240034244743,
"grad_norm": 1.2578125,
"learning_rate": 0.0004967542127108412,
"loss": 6.6257,
"mean_token_accuracy": 0.1161632239818573,
"num_tokens": 12994119.0,
"step": 6005
},
{
"entropy": 6.693649530410767,
"epoch": 0.6431590775322382,
"grad_norm": 1.140625,
"learning_rate": 0.0004967477387641537,
"loss": 6.5309,
"mean_token_accuracy": 0.1253574460744858,
"num_tokens": 13005437.0,
"step": 6010
},
{
"entropy": 6.6332213401794435,
"epoch": 0.6436941516400021,
"grad_norm": 1.65625,
"learning_rate": 0.0004967412584145312,
"loss": 6.555,
"mean_token_accuracy": 0.12132607325911522,
"num_tokens": 13017278.0,
"step": 6015
},
{
"entropy": 6.6542360305786135,
"epoch": 0.6442292257477661,
"grad_norm": 1.21875,
"learning_rate": 0.0004967347716621611,
"loss": 6.5282,
"mean_token_accuracy": 0.12165799587965012,
"num_tokens": 13028010.0,
"step": 6020
},
{
"entropy": 6.629656457901001,
"epoch": 0.64476429985553,
"grad_norm": 1.1484375,
"learning_rate": 0.0004967282785072306,
"loss": 6.5233,
"mean_token_accuracy": 0.12609869465231896,
"num_tokens": 13038740.0,
"step": 6025
},
{
"entropy": 6.6070208072662355,
"epoch": 0.645299373963294,
"grad_norm": 1.203125,
"learning_rate": 0.0004967217789499272,
"loss": 6.543,
"mean_token_accuracy": 0.12261505052447319,
"num_tokens": 13049921.0,
"step": 6030
},
{
"entropy": 6.6102134704589846,
"epoch": 0.6458344480710578,
"grad_norm": 1.5,
"learning_rate": 0.0004967152729904386,
"loss": 6.4664,
"mean_token_accuracy": 0.12932628467679025,
"num_tokens": 13060907.0,
"step": 6035
},
{
"entropy": 6.6852294445037845,
"epoch": 0.6463695221788218,
"grad_norm": 1.1875,
"learning_rate": 0.0004967087606289527,
"loss": 6.501,
"mean_token_accuracy": 0.12574908435344695,
"num_tokens": 13071720.0,
"step": 6040
},
{
"entropy": 6.666558074951172,
"epoch": 0.6469045962865857,
"grad_norm": 1.203125,
"learning_rate": 0.0004967022418656575,
"loss": 6.5475,
"mean_token_accuracy": 0.12468422725796699,
"num_tokens": 13082474.0,
"step": 6045
},
{
"entropy": 6.596642351150512,
"epoch": 0.6474396703943496,
"grad_norm": 1.359375,
"learning_rate": 0.0004966957167007412,
"loss": 6.468,
"mean_token_accuracy": 0.12356942817568779,
"num_tokens": 13092777.0,
"step": 6050
},
{
"entropy": 6.594368410110474,
"epoch": 0.6479747445021136,
"grad_norm": 1.328125,
"learning_rate": 0.0004966891851343922,
"loss": 6.4676,
"mean_token_accuracy": 0.12478138655424117,
"num_tokens": 13104019.0,
"step": 6055
},
{
"entropy": 6.70648775100708,
"epoch": 0.6485098186098774,
"grad_norm": 1.15625,
"learning_rate": 0.0004966826471667992,
"loss": 6.4973,
"mean_token_accuracy": 0.12690474689006806,
"num_tokens": 13115741.0,
"step": 6060
},
{
"entropy": 6.651901435852051,
"epoch": 0.6490448927176414,
"grad_norm": 1.21875,
"learning_rate": 0.0004966761027981509,
"loss": 6.5075,
"mean_token_accuracy": 0.12609103322029114,
"num_tokens": 13126838.0,
"step": 6065
},
{
"entropy": 6.642081689834595,
"epoch": 0.6495799668254053,
"grad_norm": 1.1875,
"learning_rate": 0.0004966695520286362,
"loss": 6.6086,
"mean_token_accuracy": 0.11905003264546395,
"num_tokens": 13137485.0,
"step": 6070
},
{
"entropy": 6.6793499946594235,
"epoch": 0.6501150409331693,
"grad_norm": 1.2265625,
"learning_rate": 0.0004966629948584444,
"loss": 6.4684,
"mean_token_accuracy": 0.1274704895913601,
"num_tokens": 13147631.0,
"step": 6075
},
{
"entropy": 6.597648620605469,
"epoch": 0.6506501150409332,
"grad_norm": 1.46875,
"learning_rate": 0.0004966564312877647,
"loss": 6.5373,
"mean_token_accuracy": 0.1260016866028309,
"num_tokens": 13158468.0,
"step": 6080
},
{
"entropy": 6.723154020309448,
"epoch": 0.651185189148697,
"grad_norm": 3.109375,
"learning_rate": 0.0004966498613167868,
"loss": 6.5746,
"mean_token_accuracy": 0.11794209703803063,
"num_tokens": 13169263.0,
"step": 6085
},
{
"entropy": 6.676345491409302,
"epoch": 0.651720263256461,
"grad_norm": 1.375,
"learning_rate": 0.0004966432849457002,
"loss": 6.4874,
"mean_token_accuracy": 0.13016888946294786,
"num_tokens": 13178647.0,
"step": 6090
},
{
"entropy": 6.582516241073608,
"epoch": 0.6522553373642249,
"grad_norm": 1.1796875,
"learning_rate": 0.000496636702174695,
"loss": 6.4788,
"mean_token_accuracy": 0.12643413320183755,
"num_tokens": 13189883.0,
"step": 6095
},
{
"entropy": 6.648423719406128,
"epoch": 0.6527904114719889,
"grad_norm": 1.25,
"learning_rate": 0.0004966301130039611,
"loss": 6.6022,
"mean_token_accuracy": 0.12256921008229256,
"num_tokens": 13200920.0,
"step": 6100
},
{
"entropy": 6.701182746887207,
"epoch": 0.6533254855797528,
"grad_norm": 2.796875,
"learning_rate": 0.0004966235174336889,
"loss": 6.5918,
"mean_token_accuracy": 0.11332411095499992,
"num_tokens": 13211765.0,
"step": 6105
},
{
"entropy": 6.659097290039062,
"epoch": 0.6538605596875168,
"grad_norm": 1.390625,
"learning_rate": 0.0004966169154640687,
"loss": 6.5191,
"mean_token_accuracy": 0.1247679390013218,
"num_tokens": 13221768.0,
"step": 6110
},
{
"entropy": 6.609342384338379,
"epoch": 0.6543956337952807,
"grad_norm": 1.21875,
"learning_rate": 0.0004966103070952912,
"loss": 6.5209,
"mean_token_accuracy": 0.12406276315450668,
"num_tokens": 13233545.0,
"step": 6115
},
{
"entropy": 6.693496465682983,
"epoch": 0.6549307079030445,
"grad_norm": 1.2109375,
"learning_rate": 0.0004966036923275472,
"loss": 6.5591,
"mean_token_accuracy": 0.13287978991866112,
"num_tokens": 13245018.0,
"step": 6120
},
{
"entropy": 6.604655790328979,
"epoch": 0.6554657820108085,
"grad_norm": 1.7578125,
"learning_rate": 0.0004965970711610278,
"loss": 6.4761,
"mean_token_accuracy": 0.12369921505451202,
"num_tokens": 13255772.0,
"step": 6125
},
{
"entropy": 6.611993074417114,
"epoch": 0.6560008561185724,
"grad_norm": 1.6796875,
"learning_rate": 0.0004965904435959241,
"loss": 6.5591,
"mean_token_accuracy": 0.11679132431745529,
"num_tokens": 13267271.0,
"step": 6130
},
{
"entropy": 6.685591697692871,
"epoch": 0.6565359302263364,
"grad_norm": 1.5625,
"learning_rate": 0.0004965838096324274,
"loss": 6.558,
"mean_token_accuracy": 0.12077526301145554,
"num_tokens": 13278467.0,
"step": 6135
},
{
"entropy": 6.647955656051636,
"epoch": 0.6570710043341003,
"grad_norm": 1.296875,
"learning_rate": 0.0004965771692707294,
"loss": 6.625,
"mean_token_accuracy": 0.11517797484993934,
"num_tokens": 13288522.0,
"step": 6140
},
{
"entropy": 6.735994911193847,
"epoch": 0.6576060784418643,
"grad_norm": 1.65625,
"learning_rate": 0.0004965705225110218,
"loss": 6.5958,
"mean_token_accuracy": 0.1102604016661644,
"num_tokens": 13299355.0,
"step": 6145
},
{
"entropy": 6.69739933013916,
"epoch": 0.6581411525496281,
"grad_norm": 1.2578125,
"learning_rate": 0.0004965638693534964,
"loss": 6.5269,
"mean_token_accuracy": 0.12430242151021957,
"num_tokens": 13310334.0,
"step": 6150
},
{
"entropy": 6.717443895339966,
"epoch": 0.658676226657392,
"grad_norm": 1.59375,
"learning_rate": 0.0004965572097983455,
"loss": 6.5918,
"mean_token_accuracy": 0.11856616660952568,
"num_tokens": 13320917.0,
"step": 6155
},
{
"entropy": 6.674542045593261,
"epoch": 0.659211300765156,
"grad_norm": 1.078125,
"learning_rate": 0.0004965505438457613,
"loss": 6.4852,
"mean_token_accuracy": 0.12417368888854981,
"num_tokens": 13331707.0,
"step": 6160
},
{
"entropy": 6.640851926803589,
"epoch": 0.6597463748729199,
"grad_norm": 1.2109375,
"learning_rate": 0.0004965438714959363,
"loss": 6.629,
"mean_token_accuracy": 0.11581350415945053,
"num_tokens": 13341784.0,
"step": 6165
},
{
"entropy": 6.735370779037476,
"epoch": 0.6602814489806839,
"grad_norm": 1.4921875,
"learning_rate": 0.000496537192749063,
"loss": 6.5809,
"mean_token_accuracy": 0.11734402552247047,
"num_tokens": 13351633.0,
"step": 6170
},
{
"entropy": 6.681562995910644,
"epoch": 0.6608165230884477,
"grad_norm": 1.2421875,
"learning_rate": 0.0004965305076053345,
"loss": 6.4794,
"mean_token_accuracy": 0.12656131833791734,
"num_tokens": 13362444.0,
"step": 6175
},
{
"entropy": 6.592486047744751,
"epoch": 0.6613515971962117,
"grad_norm": 1.3984375,
"learning_rate": 0.0004965238160649437,
"loss": 6.6139,
"mean_token_accuracy": 0.11791914477944374,
"num_tokens": 13374322.0,
"step": 6180
},
{
"entropy": 6.651375102996826,
"epoch": 0.6618866713039756,
"grad_norm": 1.390625,
"learning_rate": 0.0004965171181280838,
"loss": 6.5627,
"mean_token_accuracy": 0.11961539909243583,
"num_tokens": 13384188.0,
"step": 6185
},
{
"entropy": 6.709027862548828,
"epoch": 0.6624217454117395,
"grad_norm": 1.2265625,
"learning_rate": 0.0004965104137949483,
"loss": 6.5697,
"mean_token_accuracy": 0.11759102568030358,
"num_tokens": 13395648.0,
"step": 6190
},
{
"entropy": 6.718053483963013,
"epoch": 0.6629568195195035,
"grad_norm": 1.28125,
"learning_rate": 0.0004965037030657308,
"loss": 6.594,
"mean_token_accuracy": 0.12067084982991219,
"num_tokens": 13406769.0,
"step": 6195
},
{
"entropy": 6.728432607650757,
"epoch": 0.6634918936272673,
"grad_norm": 1.5078125,
"learning_rate": 0.0004964969859406248,
"loss": 6.563,
"mean_token_accuracy": 0.11147924661636352,
"num_tokens": 13417509.0,
"step": 6200
},
{
"entropy": 6.6613959789276125,
"epoch": 0.6640269677350313,
"grad_norm": 2.703125,
"learning_rate": 0.0004964902624198246,
"loss": 6.4816,
"mean_token_accuracy": 0.11563964560627937,
"num_tokens": 13429441.0,
"step": 6205
},
{
"entropy": 6.6188640117645265,
"epoch": 0.6645620418427952,
"grad_norm": 1.375,
"learning_rate": 0.0004964835325035243,
"loss": 6.511,
"mean_token_accuracy": 0.12196975350379943,
"num_tokens": 13439332.0,
"step": 6210
},
{
"entropy": 6.617918682098389,
"epoch": 0.6650971159505592,
"grad_norm": 1.3046875,
"learning_rate": 0.0004964767961919179,
"loss": 6.4908,
"mean_token_accuracy": 0.12061900198459626,
"num_tokens": 13449400.0,
"step": 6215
},
{
"entropy": 6.680626916885376,
"epoch": 0.6656321900583231,
"grad_norm": 1.2734375,
"learning_rate": 0.0004964700534852003,
"loss": 6.5983,
"mean_token_accuracy": 0.11767825335264206,
"num_tokens": 13460134.0,
"step": 6220
},
{
"entropy": 6.686565780639649,
"epoch": 0.666167264166087,
"grad_norm": 1.390625,
"learning_rate": 0.0004964633043835658,
"loss": 6.5323,
"mean_token_accuracy": 0.12875395640730858,
"num_tokens": 13469954.0,
"step": 6225
},
{
"entropy": 6.679265308380127,
"epoch": 0.666702338273851,
"grad_norm": 1.5703125,
"learning_rate": 0.0004964565488872097,
"loss": 6.5229,
"mean_token_accuracy": 0.12348946034908295,
"num_tokens": 13479866.0,
"step": 6230
},
{
"entropy": 6.563807821273803,
"epoch": 0.6672374123816148,
"grad_norm": 1.3125,
"learning_rate": 0.0004964497869963268,
"loss": 6.478,
"mean_token_accuracy": 0.12527237683534623,
"num_tokens": 13490088.0,
"step": 6235
},
{
"entropy": 6.656466770172119,
"epoch": 0.6677724864893788,
"grad_norm": 1.3125,
"learning_rate": 0.0004964430187111125,
"loss": 6.5728,
"mean_token_accuracy": 0.12173057049512863,
"num_tokens": 13500898.0,
"step": 6240
},
{
"entropy": 6.687621164321899,
"epoch": 0.6683075605971427,
"grad_norm": 1.109375,
"learning_rate": 0.0004964362440317621,
"loss": 6.5927,
"mean_token_accuracy": 0.1209525316953659,
"num_tokens": 13511537.0,
"step": 6245
},
{
"entropy": 6.633641624450684,
"epoch": 0.6688426347049067,
"grad_norm": 1.2109375,
"learning_rate": 0.0004964294629584712,
"loss": 6.5303,
"mean_token_accuracy": 0.12212182357907295,
"num_tokens": 13522916.0,
"step": 6250
},
{
"entropy": 6.683465242385864,
"epoch": 0.6693777088126706,
"grad_norm": 1.2890625,
"learning_rate": 0.0004964226754914357,
"loss": 6.4758,
"mean_token_accuracy": 0.12490794360637665,
"num_tokens": 13533817.0,
"step": 6255
},
{
"entropy": 6.711468744277954,
"epoch": 0.6699127829204344,
"grad_norm": 1.625,
"learning_rate": 0.0004964158816308517,
"loss": 6.5852,
"mean_token_accuracy": 0.11734790429472923,
"num_tokens": 13543723.0,
"step": 6260
},
{
"entropy": 6.586713552474976,
"epoch": 0.6704478570281984,
"grad_norm": 1.4453125,
"learning_rate": 0.0004964090813769151,
"loss": 6.434,
"mean_token_accuracy": 0.1227193221449852,
"num_tokens": 13555269.0,
"step": 6265
},
{
"entropy": 6.64307050704956,
"epoch": 0.6709829311359623,
"grad_norm": 1.328125,
"learning_rate": 0.0004964022747298224,
"loss": 6.6139,
"mean_token_accuracy": 0.11729198768734932,
"num_tokens": 13567465.0,
"step": 6270
},
{
"entropy": 6.752734518051147,
"epoch": 0.6715180052437263,
"grad_norm": 1.4296875,
"learning_rate": 0.0004963954616897701,
"loss": 6.5446,
"mean_token_accuracy": 0.11841646209359169,
"num_tokens": 13578637.0,
"step": 6275
},
{
"entropy": 6.68422417640686,
"epoch": 0.6720530793514902,
"grad_norm": 1.5078125,
"learning_rate": 0.000496388642256955,
"loss": 6.5505,
"mean_token_accuracy": 0.12172395139932632,
"num_tokens": 13589074.0,
"step": 6280
},
{
"entropy": 6.646722793579102,
"epoch": 0.6725881534592542,
"grad_norm": 1.1640625,
"learning_rate": 0.000496381816431574,
"loss": 6.514,
"mean_token_accuracy": 0.12342069149017335,
"num_tokens": 13599624.0,
"step": 6285
},
{
"entropy": 6.719354677200317,
"epoch": 0.673123227567018,
"grad_norm": 1.2734375,
"learning_rate": 0.0004963749842138241,
"loss": 6.5363,
"mean_token_accuracy": 0.12487530261278153,
"num_tokens": 13610293.0,
"step": 6290
},
{
"entropy": 6.675761699676514,
"epoch": 0.6736583016747819,
"grad_norm": 1.9453125,
"learning_rate": 0.0004963681456039027,
"loss": 6.5611,
"mean_token_accuracy": 0.11623863652348518,
"num_tokens": 13621641.0,
"step": 6295
},
{
"entropy": 6.602420663833618,
"epoch": 0.6741933757825459,
"grad_norm": 1.2890625,
"learning_rate": 0.0004963613006020072,
"loss": 6.5252,
"mean_token_accuracy": 0.1208052784204483,
"num_tokens": 13632083.0,
"step": 6300
},
{
"entropy": 6.604093360900879,
"epoch": 0.6747284498903098,
"grad_norm": 1.5546875,
"learning_rate": 0.0004963544492083352,
"loss": 6.41,
"mean_token_accuracy": 0.13683966994285585,
"num_tokens": 13644084.0,
"step": 6305
},
{
"entropy": 6.6350510597229,
"epoch": 0.6752635239980738,
"grad_norm": 1.6328125,
"learning_rate": 0.0004963475914230845,
"loss": 6.5449,
"mean_token_accuracy": 0.11910239085555077,
"num_tokens": 13656231.0,
"step": 6310
},
{
"entropy": 6.699683904647827,
"epoch": 0.6757985981058376,
"grad_norm": 2.140625,
"learning_rate": 0.0004963407272464533,
"loss": 6.5952,
"mean_token_accuracy": 0.11810294091701508,
"num_tokens": 13667334.0,
"step": 6315
},
{
"entropy": 6.724069118499756,
"epoch": 0.6763336722136016,
"grad_norm": 1.40625,
"learning_rate": 0.0004963338566786398,
"loss": 6.5828,
"mean_token_accuracy": 0.12107535973191261,
"num_tokens": 13678615.0,
"step": 6320
},
{
"entropy": 6.693145990371704,
"epoch": 0.6768687463213655,
"grad_norm": 1.578125,
"learning_rate": 0.0004963269797198422,
"loss": 6.5874,
"mean_token_accuracy": 0.12346574515104294,
"num_tokens": 13689890.0,
"step": 6325
},
{
"entropy": 6.558933973312378,
"epoch": 0.6774038204291294,
"grad_norm": 1.6328125,
"learning_rate": 0.0004963200963702592,
"loss": 6.4168,
"mean_token_accuracy": 0.13032166957855223,
"num_tokens": 13699120.0,
"step": 6330
},
{
"entropy": 6.581582736968994,
"epoch": 0.6779388945368934,
"grad_norm": 1.25,
"learning_rate": 0.0004963132066300894,
"loss": 6.5395,
"mean_token_accuracy": 0.11724406257271766,
"num_tokens": 13710016.0,
"step": 6335
},
{
"entropy": 6.6938841342926025,
"epoch": 0.6784739686446573,
"grad_norm": 1.5546875,
"learning_rate": 0.0004963063104995319,
"loss": 6.6391,
"mean_token_accuracy": 0.11238991096615791,
"num_tokens": 13721576.0,
"step": 6340
},
{
"entropy": 6.745585584640503,
"epoch": 0.6790090427524212,
"grad_norm": 1.453125,
"learning_rate": 0.0004962994079787859,
"loss": 6.591,
"mean_token_accuracy": 0.12299842238426209,
"num_tokens": 13732568.0,
"step": 6345
},
{
"entropy": 6.657323408126831,
"epoch": 0.6795441168601851,
"grad_norm": 1.4296875,
"learning_rate": 0.0004962924990680504,
"loss": 6.4794,
"mean_token_accuracy": 0.12324386537075042,
"num_tokens": 13742569.0,
"step": 6350
},
{
"entropy": 6.627435684204102,
"epoch": 0.6800791909679491,
"grad_norm": 1.453125,
"learning_rate": 0.0004962855837675252,
"loss": 6.6034,
"mean_token_accuracy": 0.11716282293200493,
"num_tokens": 13752689.0,
"step": 6355
},
{
"entropy": 6.6892821311950685,
"epoch": 0.680614265075713,
"grad_norm": 1.453125,
"learning_rate": 0.0004962786620774099,
"loss": 6.5905,
"mean_token_accuracy": 0.11979246288537979,
"num_tokens": 13763709.0,
"step": 6360
},
{
"entropy": 6.705752182006836,
"epoch": 0.6811493391834769,
"grad_norm": 1.6796875,
"learning_rate": 0.0004962717339979043,
"loss": 6.4891,
"mean_token_accuracy": 0.1252245396375656,
"num_tokens": 13773726.0,
"step": 6365
},
{
"entropy": 6.608482742309571,
"epoch": 0.6816844132912409,
"grad_norm": 1.328125,
"learning_rate": 0.0004962647995292084,
"loss": 6.5296,
"mean_token_accuracy": 0.12159904614090919,
"num_tokens": 13785435.0,
"step": 6370
},
{
"entropy": 6.640556859970093,
"epoch": 0.6822194873990047,
"grad_norm": 1.71875,
"learning_rate": 0.0004962578586715226,
"loss": 6.52,
"mean_token_accuracy": 0.11594315767288207,
"num_tokens": 13797104.0,
"step": 6375
},
{
"entropy": 6.7496202945709225,
"epoch": 0.6827545615067687,
"grad_norm": 1.3515625,
"learning_rate": 0.0004962509114250471,
"loss": 6.6121,
"mean_token_accuracy": 0.11787962764501572,
"num_tokens": 13807756.0,
"step": 6380
},
{
"entropy": 6.491614484786988,
"epoch": 0.6832896356145326,
"grad_norm": 1.453125,
"learning_rate": 0.0004962439577899828,
"loss": 6.314,
"mean_token_accuracy": 0.13807514384388925,
"num_tokens": 13818490.0,
"step": 6385
},
{
"entropy": 6.646625661849976,
"epoch": 0.6838247097222966,
"grad_norm": 1.28125,
"learning_rate": 0.0004962369977665301,
"loss": 6.5814,
"mean_token_accuracy": 0.11426923871040344,
"num_tokens": 13829465.0,
"step": 6390
},
{
"entropy": 6.783056354522705,
"epoch": 0.6843597838300605,
"grad_norm": 1.171875,
"learning_rate": 0.0004962300313548903,
"loss": 6.5539,
"mean_token_accuracy": 0.12248866409063339,
"num_tokens": 13840863.0,
"step": 6395
},
{
"entropy": 6.695218420028686,
"epoch": 0.6848948579378243,
"grad_norm": 1.2265625,
"learning_rate": 0.0004962230585552645,
"loss": 6.549,
"mean_token_accuracy": 0.12046386897563935,
"num_tokens": 13851711.0,
"step": 6400
},
{
"entropy": 6.581887054443359,
"epoch": 0.6854299320455883,
"grad_norm": 1.6953125,
"learning_rate": 0.0004962160793678539,
"loss": 6.5789,
"mean_token_accuracy": 0.11757319420576096,
"num_tokens": 13864576.0,
"step": 6405
},
{
"entropy": 6.6630912780761715,
"epoch": 0.6859650061533522,
"grad_norm": 1.3359375,
"learning_rate": 0.00049620909379286,
"loss": 6.4883,
"mean_token_accuracy": 0.12629830315709115,
"num_tokens": 13875150.0,
"step": 6410
},
{
"entropy": 6.683273983001709,
"epoch": 0.6865000802611162,
"grad_norm": 2.546875,
"learning_rate": 0.0004962021018304847,
"loss": 6.5422,
"mean_token_accuracy": 0.12137580290436745,
"num_tokens": 13886070.0,
"step": 6415
},
{
"entropy": 6.600133419036865,
"epoch": 0.6870351543688801,
"grad_norm": 1.078125,
"learning_rate": 0.0004961951034809298,
"loss": 6.4596,
"mean_token_accuracy": 0.11954843401908874,
"num_tokens": 13897898.0,
"step": 6420
},
{
"entropy": 6.743383502960205,
"epoch": 0.6875702284766441,
"grad_norm": 1.2578125,
"learning_rate": 0.0004961880987443974,
"loss": 6.5996,
"mean_token_accuracy": 0.1191033512353897,
"num_tokens": 13909578.0,
"step": 6425
},
{
"entropy": 6.5636146068573,
"epoch": 0.6881053025844079,
"grad_norm": 1.46875,
"learning_rate": 0.0004961810876210897,
"loss": 6.4561,
"mean_token_accuracy": 0.12540539875626563,
"num_tokens": 13920537.0,
"step": 6430
},
{
"entropy": 6.548890447616577,
"epoch": 0.6886403766921718,
"grad_norm": 1.2734375,
"learning_rate": 0.0004961740701112091,
"loss": 6.5134,
"mean_token_accuracy": 0.12369008883833885,
"num_tokens": 13931015.0,
"step": 6435
},
{
"entropy": 6.668211126327515,
"epoch": 0.6891754507999358,
"grad_norm": 1.2109375,
"learning_rate": 0.0004961670462149583,
"loss": 6.5521,
"mean_token_accuracy": 0.12055359557271003,
"num_tokens": 13942691.0,
"step": 6440
},
{
"entropy": 6.767236375808716,
"epoch": 0.6897105249076997,
"grad_norm": 1.21875,
"learning_rate": 0.0004961600159325403,
"loss": 6.637,
"mean_token_accuracy": 0.10908434465527535,
"num_tokens": 13953945.0,
"step": 6445
},
{
"entropy": 6.751330995559693,
"epoch": 0.6902455990154637,
"grad_norm": 1.328125,
"learning_rate": 0.0004961529792641577,
"loss": 6.6236,
"mean_token_accuracy": 0.1173236459493637,
"num_tokens": 13965658.0,
"step": 6450
},
{
"entropy": 6.731899356842041,
"epoch": 0.6907806731232276,
"grad_norm": 1.2578125,
"learning_rate": 0.0004961459362100141,
"loss": 6.5992,
"mean_token_accuracy": 0.12082590013742447,
"num_tokens": 13976184.0,
"step": 6455
},
{
"entropy": 6.6650745391845705,
"epoch": 0.6913157472309915,
"grad_norm": 1.1953125,
"learning_rate": 0.0004961388867703125,
"loss": 6.5114,
"mean_token_accuracy": 0.11692757606506347,
"num_tokens": 13986243.0,
"step": 6460
},
{
"entropy": 6.65087571144104,
"epoch": 0.6918508213387554,
"grad_norm": 1.109375,
"learning_rate": 0.0004961318309452566,
"loss": 6.507,
"mean_token_accuracy": 0.12220369726419449,
"num_tokens": 13995904.0,
"step": 6465
},
{
"entropy": 6.659942245483398,
"epoch": 0.6923858954465193,
"grad_norm": 1.3359375,
"learning_rate": 0.0004961247687350504,
"loss": 6.5498,
"mean_token_accuracy": 0.12576377019286156,
"num_tokens": 14006711.0,
"step": 6470
},
{
"entropy": 6.628136539459229,
"epoch": 0.6929209695542833,
"grad_norm": 1.2890625,
"learning_rate": 0.0004961177001398974,
"loss": 6.527,
"mean_token_accuracy": 0.12062407061457633,
"num_tokens": 14018177.0,
"step": 6475
},
{
"entropy": 6.665041160583496,
"epoch": 0.6934560436620472,
"grad_norm": 1.40625,
"learning_rate": 0.0004961106251600018,
"loss": 6.4884,
"mean_token_accuracy": 0.11637992933392524,
"num_tokens": 14028725.0,
"step": 6480
},
{
"entropy": 6.6288951396942135,
"epoch": 0.6939911177698111,
"grad_norm": 1.2578125,
"learning_rate": 0.0004961035437955681,
"loss": 6.5271,
"mean_token_accuracy": 0.13461530953645706,
"num_tokens": 14040448.0,
"step": 6485
},
{
"entropy": 6.62433614730835,
"epoch": 0.694526191877575,
"grad_norm": 1.2109375,
"learning_rate": 0.0004960964560468005,
"loss": 6.5734,
"mean_token_accuracy": 0.11507597416639329,
"num_tokens": 14051731.0,
"step": 6490
},
{
"entropy": 6.7240547180175785,
"epoch": 0.695061265985339,
"grad_norm": 1.1953125,
"learning_rate": 0.0004960893619139039,
"loss": 6.5658,
"mean_token_accuracy": 0.11801211908459663,
"num_tokens": 14062330.0,
"step": 6495
},
{
"entropy": 6.683112001419067,
"epoch": 0.6955963400931029,
"grad_norm": 1.1640625,
"learning_rate": 0.0004960822613970831,
"loss": 6.4486,
"mean_token_accuracy": 0.12386781573295594,
"num_tokens": 14073991.0,
"step": 6500
},
{
"entropy": 6.640247869491577,
"epoch": 0.6961314142008668,
"grad_norm": 1.421875,
"learning_rate": 0.0004960751544965431,
"loss": 6.5076,
"mean_token_accuracy": 0.12210935950279236,
"num_tokens": 14085132.0,
"step": 6505
},
{
"entropy": 6.592376041412353,
"epoch": 0.6966664883086308,
"grad_norm": 1.3515625,
"learning_rate": 0.0004960680412124889,
"loss": 6.4499,
"mean_token_accuracy": 0.12720930054783822,
"num_tokens": 14095371.0,
"step": 6510
},
{
"entropy": 6.5528076171875,
"epoch": 0.6972015624163946,
"grad_norm": 1.7421875,
"learning_rate": 0.0004960609215451261,
"loss": 6.5341,
"mean_token_accuracy": 0.12374990880489349,
"num_tokens": 14106117.0,
"step": 6515
},
{
"entropy": 6.649059104919433,
"epoch": 0.6977366365241586,
"grad_norm": 1.2734375,
"learning_rate": 0.0004960537954946604,
"loss": 6.5267,
"mean_token_accuracy": 0.12003685981035232,
"num_tokens": 14117112.0,
"step": 6520
},
{
"entropy": 6.596162796020508,
"epoch": 0.6982717106319225,
"grad_norm": 1.3046875,
"learning_rate": 0.0004960466630612974,
"loss": 6.4541,
"mean_token_accuracy": 0.13205637782812119,
"num_tokens": 14128037.0,
"step": 6525
},
{
"entropy": 6.556289529800415,
"epoch": 0.6988067847396865,
"grad_norm": 1.2734375,
"learning_rate": 0.0004960395242452429,
"loss": 6.4584,
"mean_token_accuracy": 0.12506138905882835,
"num_tokens": 14140183.0,
"step": 6530
},
{
"entropy": 6.6649885177612305,
"epoch": 0.6993418588474504,
"grad_norm": 1.4921875,
"learning_rate": 0.0004960323790467033,
"loss": 6.5995,
"mean_token_accuracy": 0.12367913722991944,
"num_tokens": 14151635.0,
"step": 6535
},
{
"entropy": 6.737906551361084,
"epoch": 0.6998769329552142,
"grad_norm": 1.28125,
"learning_rate": 0.0004960252274658849,
"loss": 6.5577,
"mean_token_accuracy": 0.12105544358491897,
"num_tokens": 14162660.0,
"step": 6540
},
{
"entropy": 6.671280765533448,
"epoch": 0.7004120070629782,
"grad_norm": 1.3828125,
"learning_rate": 0.0004960180695029939,
"loss": 6.5407,
"mean_token_accuracy": 0.12037990167737007,
"num_tokens": 14173496.0,
"step": 6545
},
{
"entropy": 6.617800331115722,
"epoch": 0.7009470811707421,
"grad_norm": 1.34375,
"learning_rate": 0.0004960109051582374,
"loss": 6.4934,
"mean_token_accuracy": 0.11847614273428916,
"num_tokens": 14185479.0,
"step": 6550
},
{
"entropy": 6.60119400024414,
"epoch": 0.7014821552785061,
"grad_norm": 1.3046875,
"learning_rate": 0.0004960037344318221,
"loss": 6.4507,
"mean_token_accuracy": 0.12120893523097039,
"num_tokens": 14197441.0,
"step": 6555
},
{
"entropy": 6.566002035140992,
"epoch": 0.70201722938627,
"grad_norm": 1.2890625,
"learning_rate": 0.0004959965573239548,
"loss": 6.4247,
"mean_token_accuracy": 0.12884053736925125,
"num_tokens": 14207832.0,
"step": 6560
},
{
"entropy": 6.55321888923645,
"epoch": 0.702552303494034,
"grad_norm": 1.3203125,
"learning_rate": 0.0004959893738348432,
"loss": 6.4945,
"mean_token_accuracy": 0.12767787277698517,
"num_tokens": 14217976.0,
"step": 6565
},
{
"entropy": 6.674604034423828,
"epoch": 0.7030873776017978,
"grad_norm": 1.421875,
"learning_rate": 0.0004959821839646943,
"loss": 6.4431,
"mean_token_accuracy": 0.12329324334859848,
"num_tokens": 14227557.0,
"step": 6570
},
{
"entropy": 6.662657594680786,
"epoch": 0.7036224517095617,
"grad_norm": 1.28125,
"learning_rate": 0.000495974987713716,
"loss": 6.571,
"mean_token_accuracy": 0.11384310796856881,
"num_tokens": 14238912.0,
"step": 6575
},
{
"entropy": 6.638991737365723,
"epoch": 0.7041575258173257,
"grad_norm": 1.25,
"learning_rate": 0.0004959677850821159,
"loss": 6.5012,
"mean_token_accuracy": 0.12319194078445435,
"num_tokens": 14249968.0,
"step": 6580
},
{
"entropy": 6.620342445373535,
"epoch": 0.7046925999250896,
"grad_norm": 1.265625,
"learning_rate": 0.0004959605760701021,
"loss": 6.5569,
"mean_token_accuracy": 0.12182011604309081,
"num_tokens": 14260809.0,
"step": 6585
},
{
"entropy": 6.663771057128907,
"epoch": 0.7052276740328536,
"grad_norm": 1.5,
"learning_rate": 0.0004959533606778827,
"loss": 6.527,
"mean_token_accuracy": 0.1212913878262043,
"num_tokens": 14271971.0,
"step": 6590
},
{
"entropy": 6.70226559638977,
"epoch": 0.7057627481406175,
"grad_norm": 1.265625,
"learning_rate": 0.0004959461389056661,
"loss": 6.5913,
"mean_token_accuracy": 0.12781344801187516,
"num_tokens": 14283550.0,
"step": 6595
},
{
"entropy": 6.566189193725586,
"epoch": 0.7062978222483814,
"grad_norm": 1.25,
"learning_rate": 0.0004959389107536608,
"loss": 6.4158,
"mean_token_accuracy": 0.12276504635810852,
"num_tokens": 14293769.0,
"step": 6600
},
{
"entropy": 6.642466735839844,
"epoch": 0.7068328963561453,
"grad_norm": 1.3203125,
"learning_rate": 0.0004959316762220754,
"loss": 6.5368,
"mean_token_accuracy": 0.11977979466319084,
"num_tokens": 14304668.0,
"step": 6605
},
{
"entropy": 6.650493144989014,
"epoch": 0.7073679704639092,
"grad_norm": 1.234375,
"learning_rate": 0.000495924435311119,
"loss": 6.5975,
"mean_token_accuracy": 0.11864884719252586,
"num_tokens": 14315912.0,
"step": 6610
},
{
"entropy": 6.61599588394165,
"epoch": 0.7079030445716732,
"grad_norm": 1.1875,
"learning_rate": 0.0004959171880210005,
"loss": 6.5415,
"mean_token_accuracy": 0.1240161381661892,
"num_tokens": 14327386.0,
"step": 6615
},
{
"entropy": 6.61832332611084,
"epoch": 0.7084381186794371,
"grad_norm": 1.1640625,
"learning_rate": 0.0004959099343519294,
"loss": 6.5297,
"mean_token_accuracy": 0.11902955025434495,
"num_tokens": 14338629.0,
"step": 6620
},
{
"entropy": 6.745673322677613,
"epoch": 0.7089731927872011,
"grad_norm": 1.171875,
"learning_rate": 0.0004959026743041148,
"loss": 6.5523,
"mean_token_accuracy": 0.12611126601696016,
"num_tokens": 14349195.0,
"step": 6625
},
{
"entropy": 6.707384347915649,
"epoch": 0.7095082668949649,
"grad_norm": 1.4375,
"learning_rate": 0.0004958954078777665,
"loss": 6.5804,
"mean_token_accuracy": 0.11705741733312607,
"num_tokens": 14359449.0,
"step": 6630
},
{
"entropy": 6.607598972320557,
"epoch": 0.7100433410027289,
"grad_norm": 1.3828125,
"learning_rate": 0.0004958881350730944,
"loss": 6.4914,
"mean_token_accuracy": 0.12530679404735565,
"num_tokens": 14369904.0,
"step": 6635
},
{
"entropy": 6.640906763076782,
"epoch": 0.7105784151104928,
"grad_norm": 1.4453125,
"learning_rate": 0.0004958808558903085,
"loss": 6.502,
"mean_token_accuracy": 0.12301137670874596,
"num_tokens": 14382635.0,
"step": 6640
},
{
"entropy": 6.597404193878174,
"epoch": 0.7111134892182567,
"grad_norm": 1.4140625,
"learning_rate": 0.0004958735703296189,
"loss": 6.4504,
"mean_token_accuracy": 0.12635595500469207,
"num_tokens": 14393238.0,
"step": 6645
},
{
"entropy": 6.605778455734253,
"epoch": 0.7116485633260207,
"grad_norm": 1.25,
"learning_rate": 0.000495866278391236,
"loss": 6.4988,
"mean_token_accuracy": 0.12151379361748696,
"num_tokens": 14403734.0,
"step": 6650
},
{
"entropy": 6.662947177886963,
"epoch": 0.7121836374337845,
"grad_norm": 1.6640625,
"learning_rate": 0.0004958589800753703,
"loss": 6.493,
"mean_token_accuracy": 0.12501125037670135,
"num_tokens": 14414201.0,
"step": 6655
},
{
"entropy": 6.639647436141968,
"epoch": 0.7127187115415485,
"grad_norm": 1.21875,
"learning_rate": 0.0004958516753822326,
"loss": 6.5273,
"mean_token_accuracy": 0.12744878232479095,
"num_tokens": 14425153.0,
"step": 6660
},
{
"entropy": 6.599532556533814,
"epoch": 0.7132537856493124,
"grad_norm": 1.453125,
"learning_rate": 0.0004958443643120338,
"loss": 6.5016,
"mean_token_accuracy": 0.12863869071006775,
"num_tokens": 14435942.0,
"step": 6665
},
{
"entropy": 6.614558696746826,
"epoch": 0.7137888597570764,
"grad_norm": 1.4453125,
"learning_rate": 0.000495837046864985,
"loss": 6.4763,
"mean_token_accuracy": 0.12335674315690995,
"num_tokens": 14446371.0,
"step": 6670
},
{
"entropy": 6.617805099487304,
"epoch": 0.7143239338648403,
"grad_norm": 1.421875,
"learning_rate": 0.0004958297230412976,
"loss": 6.4252,
"mean_token_accuracy": 0.127166984975338,
"num_tokens": 14456598.0,
"step": 6675
},
{
"entropy": 6.555561685562134,
"epoch": 0.7148590079726042,
"grad_norm": 2.703125,
"learning_rate": 0.0004958223928411829,
"loss": 6.5309,
"mean_token_accuracy": 0.1143219605088234,
"num_tokens": 14466775.0,
"step": 6680
},
{
"entropy": 6.574054431915283,
"epoch": 0.7153940820803681,
"grad_norm": 1.375,
"learning_rate": 0.0004958150562648526,
"loss": 6.4665,
"mean_token_accuracy": 0.12366539910435677,
"num_tokens": 14476876.0,
"step": 6685
},
{
"entropy": 6.590529203414917,
"epoch": 0.715929156188132,
"grad_norm": 1.3984375,
"learning_rate": 0.0004958077133125187,
"loss": 6.4797,
"mean_token_accuracy": 0.12587342411279678,
"num_tokens": 14487698.0,
"step": 6690
},
{
"entropy": 6.692271709442139,
"epoch": 0.716464230295896,
"grad_norm": 1.640625,
"learning_rate": 0.000495800363984393,
"loss": 6.4987,
"mean_token_accuracy": 0.12271760776638985,
"num_tokens": 14498572.0,
"step": 6695
},
{
"entropy": 6.623218107223511,
"epoch": 0.7169993044036599,
"grad_norm": 1.3515625,
"learning_rate": 0.0004957930082806878,
"loss": 6.5438,
"mean_token_accuracy": 0.12247662395238876,
"num_tokens": 14509673.0,
"step": 6700
},
{
"entropy": 6.70994553565979,
"epoch": 0.7175343785114239,
"grad_norm": 1.515625,
"learning_rate": 0.0004957856462016155,
"loss": 6.6279,
"mean_token_accuracy": 0.1105528324842453,
"num_tokens": 14519974.0,
"step": 6705
},
{
"entropy": 6.691906595230103,
"epoch": 0.7180694526191878,
"grad_norm": 2.21875,
"learning_rate": 0.0004957782777473888,
"loss": 6.5373,
"mean_token_accuracy": 0.11652741655707359,
"num_tokens": 14531194.0,
"step": 6710
},
{
"entropy": 6.606967735290527,
"epoch": 0.7186045267269516,
"grad_norm": 1.75,
"learning_rate": 0.0004957709029182203,
"loss": 6.4862,
"mean_token_accuracy": 0.12221086099743843,
"num_tokens": 14542393.0,
"step": 6715
},
{
"entropy": 6.582061862945556,
"epoch": 0.7191396008347156,
"grad_norm": 1.375,
"learning_rate": 0.0004957635217143231,
"loss": 6.4944,
"mean_token_accuracy": 0.13097272515296937,
"num_tokens": 14552934.0,
"step": 6720
},
{
"entropy": 6.62083740234375,
"epoch": 0.7196746749424795,
"grad_norm": 1.3984375,
"learning_rate": 0.00049575613413591,
"loss": 6.4978,
"mean_token_accuracy": 0.120374695956707,
"num_tokens": 14563643.0,
"step": 6725
},
{
"entropy": 6.601111316680909,
"epoch": 0.7202097490502435,
"grad_norm": 1.296875,
"learning_rate": 0.0004957487401831947,
"loss": 6.445,
"mean_token_accuracy": 0.12362169623374938,
"num_tokens": 14574422.0,
"step": 6730
},
{
"entropy": 6.584991502761841,
"epoch": 0.7207448231580074,
"grad_norm": 1.140625,
"learning_rate": 0.0004957413398563906,
"loss": 6.4292,
"mean_token_accuracy": 0.11930982172489166,
"num_tokens": 14585002.0,
"step": 6735
},
{
"entropy": 6.625512647628784,
"epoch": 0.7212798972657714,
"grad_norm": 1.6875,
"learning_rate": 0.0004957339331557113,
"loss": 6.4612,
"mean_token_accuracy": 0.1259589172899723,
"num_tokens": 14595172.0,
"step": 6740
},
{
"entropy": 6.599657821655273,
"epoch": 0.7218149713735352,
"grad_norm": 1.2421875,
"learning_rate": 0.0004957265200813706,
"loss": 6.5742,
"mean_token_accuracy": 0.11676609218120575,
"num_tokens": 14606316.0,
"step": 6745
},
{
"entropy": 6.696068477630615,
"epoch": 0.7223500454812991,
"grad_norm": 1.609375,
"learning_rate": 0.0004957191006335827,
"loss": 6.4903,
"mean_token_accuracy": 0.12656715735793114,
"num_tokens": 14616394.0,
"step": 6750
},
{
"entropy": 6.555004119873047,
"epoch": 0.7228851195890631,
"grad_norm": 1.5,
"learning_rate": 0.0004957116748125618,
"loss": 6.4925,
"mean_token_accuracy": 0.11967325806617737,
"num_tokens": 14627474.0,
"step": 6755
},
{
"entropy": 6.575228595733643,
"epoch": 0.723420193696827,
"grad_norm": 1.28125,
"learning_rate": 0.0004957042426185223,
"loss": 6.4314,
"mean_token_accuracy": 0.13004304990172386,
"num_tokens": 14638343.0,
"step": 6760
},
{
"entropy": 6.614687061309814,
"epoch": 0.723955267804591,
"grad_norm": 1.328125,
"learning_rate": 0.0004956968040516789,
"loss": 6.4941,
"mean_token_accuracy": 0.12631681188941002,
"num_tokens": 14649115.0,
"step": 6765
},
{
"entropy": 6.607396507263184,
"epoch": 0.7244903419123548,
"grad_norm": 1.1328125,
"learning_rate": 0.0004956893591122461,
"loss": 6.5614,
"mean_token_accuracy": 0.1266578659415245,
"num_tokens": 14659714.0,
"step": 6770
},
{
"entropy": 6.730130004882812,
"epoch": 0.7250254160201188,
"grad_norm": 1.2578125,
"learning_rate": 0.0004956819078004392,
"loss": 6.5943,
"mean_token_accuracy": 0.1168831467628479,
"num_tokens": 14670481.0,
"step": 6775
},
{
"entropy": 6.612733459472656,
"epoch": 0.7255604901278827,
"grad_norm": 1.515625,
"learning_rate": 0.0004956744501164732,
"loss": 6.4208,
"mean_token_accuracy": 0.12815224677324294,
"num_tokens": 14680816.0,
"step": 6780
},
{
"entropy": 6.573790550231934,
"epoch": 0.7260955642356466,
"grad_norm": 1.1796875,
"learning_rate": 0.0004956669860605634,
"loss": 6.5177,
"mean_token_accuracy": 0.12638939991593362,
"num_tokens": 14692723.0,
"step": 6785
},
{
"entropy": 6.630371952056885,
"epoch": 0.7266306383434106,
"grad_norm": 1.3125,
"learning_rate": 0.0004956595156329254,
"loss": 6.5369,
"mean_token_accuracy": 0.12008581012487411,
"num_tokens": 14703824.0,
"step": 6790
},
{
"entropy": 6.706773328781128,
"epoch": 0.7271657124511745,
"grad_norm": 1.28125,
"learning_rate": 0.000495652038833775,
"loss": 6.5182,
"mean_token_accuracy": 0.12735376432538031,
"num_tokens": 14715934.0,
"step": 6795
},
{
"entropy": 6.598440217971802,
"epoch": 0.7277007865589384,
"grad_norm": 1.328125,
"learning_rate": 0.0004956445556633279,
"loss": 6.5177,
"mean_token_accuracy": 0.12028545215725898,
"num_tokens": 14725617.0,
"step": 6800
},
{
"entropy": 6.634641218185425,
"epoch": 0.7282358606667023,
"grad_norm": 1.328125,
"learning_rate": 0.0004956370661218003,
"loss": 6.4967,
"mean_token_accuracy": 0.12309338971972465,
"num_tokens": 14736430.0,
"step": 6805
},
{
"entropy": 6.666610288619995,
"epoch": 0.7287709347744663,
"grad_norm": 1.4375,
"learning_rate": 0.0004956295702094084,
"loss": 6.5171,
"mean_token_accuracy": 0.12309236079454422,
"num_tokens": 14746740.0,
"step": 6810
},
{
"entropy": 6.582339763641357,
"epoch": 0.7293060088822302,
"grad_norm": 1.5625,
"learning_rate": 0.0004956220679263687,
"loss": 6.4535,
"mean_token_accuracy": 0.12928269281983376,
"num_tokens": 14756884.0,
"step": 6815
},
{
"entropy": 6.589403057098389,
"epoch": 0.7298410829899941,
"grad_norm": 1.2734375,
"learning_rate": 0.0004956145592728976,
"loss": 6.4581,
"mean_token_accuracy": 0.1312728136777878,
"num_tokens": 14766974.0,
"step": 6820
},
{
"entropy": 6.623006343841553,
"epoch": 0.730376157097758,
"grad_norm": 1.359375,
"learning_rate": 0.0004956070442492123,
"loss": 6.6176,
"mean_token_accuracy": 0.11667682155966759,
"num_tokens": 14778449.0,
"step": 6825
},
{
"entropy": 6.7070536613464355,
"epoch": 0.7309112312055219,
"grad_norm": 1.3671875,
"learning_rate": 0.0004955995228555296,
"loss": 6.5422,
"mean_token_accuracy": 0.1225484624505043,
"num_tokens": 14788989.0,
"step": 6830
},
{
"entropy": 6.628703689575195,
"epoch": 0.7314463053132859,
"grad_norm": 1.2890625,
"learning_rate": 0.0004955919950920666,
"loss": 6.5023,
"mean_token_accuracy": 0.12728202119469642,
"num_tokens": 14798454.0,
"step": 6835
},
{
"entropy": 6.660853433609009,
"epoch": 0.7319813794210498,
"grad_norm": 1.515625,
"learning_rate": 0.0004955844609590408,
"loss": 6.4467,
"mean_token_accuracy": 0.12790426537394523,
"num_tokens": 14808920.0,
"step": 6840
},
{
"entropy": 6.635347509384156,
"epoch": 0.7325164535288138,
"grad_norm": 1.1875,
"learning_rate": 0.0004955769204566696,
"loss": 6.4426,
"mean_token_accuracy": 0.1228242613375187,
"num_tokens": 14819250.0,
"step": 6845
},
{
"entropy": 6.662122344970703,
"epoch": 0.7330515276365777,
"grad_norm": 1.5078125,
"learning_rate": 0.0004955693735851709,
"loss": 6.5277,
"mean_token_accuracy": 0.12443855032324791,
"num_tokens": 14829789.0,
"step": 6850
},
{
"entropy": 6.562832450866699,
"epoch": 0.7335866017443415,
"grad_norm": 1.5625,
"learning_rate": 0.0004955618203447625,
"loss": 6.5037,
"mean_token_accuracy": 0.1241986483335495,
"num_tokens": 14840527.0,
"step": 6855
},
{
"entropy": 6.608736944198609,
"epoch": 0.7341216758521055,
"grad_norm": 1.46875,
"learning_rate": 0.0004955542607356625,
"loss": 6.5588,
"mean_token_accuracy": 0.12313508540391922,
"num_tokens": 14850922.0,
"step": 6860
},
{
"entropy": 6.655763769149781,
"epoch": 0.7346567499598694,
"grad_norm": 1.9609375,
"learning_rate": 0.0004955466947580893,
"loss": 6.5061,
"mean_token_accuracy": 0.12578540593385695,
"num_tokens": 14862226.0,
"step": 6865
},
{
"entropy": 6.597611951828003,
"epoch": 0.7351918240676334,
"grad_norm": 1.5859375,
"learning_rate": 0.0004955391224122611,
"loss": 6.4097,
"mean_token_accuracy": 0.1363422118127346,
"num_tokens": 14872461.0,
"step": 6870
},
{
"entropy": 6.608497571945191,
"epoch": 0.7357268981753973,
"grad_norm": 1.4765625,
"learning_rate": 0.000495531543698397,
"loss": 6.5258,
"mean_token_accuracy": 0.12411358803510666,
"num_tokens": 14883088.0,
"step": 6875
},
{
"entropy": 6.693577766418457,
"epoch": 0.7362619722831613,
"grad_norm": 1.234375,
"learning_rate": 0.0004955239586167153,
"loss": 6.5405,
"mean_token_accuracy": 0.12336590811610222,
"num_tokens": 14894055.0,
"step": 6880
},
{
"entropy": 6.6278046607971195,
"epoch": 0.7367970463909251,
"grad_norm": 1.5703125,
"learning_rate": 0.0004955163671674354,
"loss": 6.5092,
"mean_token_accuracy": 0.1199507437646389,
"num_tokens": 14904760.0,
"step": 6885
},
{
"entropy": 6.727400636672973,
"epoch": 0.737332120498689,
"grad_norm": 2.65625,
"learning_rate": 0.0004955087693507764,
"loss": 6.6256,
"mean_token_accuracy": 0.11664849147200584,
"num_tokens": 14915983.0,
"step": 6890
},
{
"entropy": 6.65953311920166,
"epoch": 0.737867194606453,
"grad_norm": 1.5078125,
"learning_rate": 0.0004955011651669577,
"loss": 6.5148,
"mean_token_accuracy": 0.11811737567186356,
"num_tokens": 14925959.0,
"step": 6895
},
{
"entropy": 6.5658063888549805,
"epoch": 0.7384022687142169,
"grad_norm": 1.21875,
"learning_rate": 0.0004954935546161988,
"loss": 6.4756,
"mean_token_accuracy": 0.12480072304606438,
"num_tokens": 14936138.0,
"step": 6900
},
{
"entropy": 6.680220413208008,
"epoch": 0.7389373428219809,
"grad_norm": 1.234375,
"learning_rate": 0.0004954859376987195,
"loss": 6.5023,
"mean_token_accuracy": 0.1277581550180912,
"num_tokens": 14946979.0,
"step": 6905
},
{
"entropy": 6.584284877777099,
"epoch": 0.7394724169297447,
"grad_norm": 1.3359375,
"learning_rate": 0.0004954783144147397,
"loss": 6.5679,
"mean_token_accuracy": 0.12042608857154846,
"num_tokens": 14958324.0,
"step": 6910
},
{
"entropy": 6.737890386581421,
"epoch": 0.7400074910375087,
"grad_norm": 1.234375,
"learning_rate": 0.0004954706847644796,
"loss": 6.5574,
"mean_token_accuracy": 0.11832675114274024,
"num_tokens": 14969048.0,
"step": 6915
},
{
"entropy": 6.710620450973511,
"epoch": 0.7405425651452726,
"grad_norm": 1.265625,
"learning_rate": 0.0004954630487481594,
"loss": 6.5496,
"mean_token_accuracy": 0.12008848264813424,
"num_tokens": 14979825.0,
"step": 6920
},
{
"entropy": 6.595880651473999,
"epoch": 0.7410776392530365,
"grad_norm": 1.3671875,
"learning_rate": 0.0004954554063659998,
"loss": 6.481,
"mean_token_accuracy": 0.12603807896375657,
"num_tokens": 14990138.0,
"step": 6925
},
{
"entropy": 6.596027612686157,
"epoch": 0.7416127133608005,
"grad_norm": 1.4453125,
"learning_rate": 0.0004954477576182212,
"loss": 6.4498,
"mean_token_accuracy": 0.1261916309595108,
"num_tokens": 15002276.0,
"step": 6930
},
{
"entropy": 6.646682214736939,
"epoch": 0.7421477874685644,
"grad_norm": 1.3828125,
"learning_rate": 0.0004954401025050445,
"loss": 6.5709,
"mean_token_accuracy": 0.12357406765222549,
"num_tokens": 15013622.0,
"step": 6935
},
{
"entropy": 6.676917219161988,
"epoch": 0.7426828615763283,
"grad_norm": 1.3203125,
"learning_rate": 0.0004954324410266909,
"loss": 6.531,
"mean_token_accuracy": 0.12281017452478409,
"num_tokens": 15024067.0,
"step": 6940
},
{
"entropy": 6.628924036026001,
"epoch": 0.7432179356840922,
"grad_norm": 1.359375,
"learning_rate": 0.0004954247731833815,
"loss": 6.4168,
"mean_token_accuracy": 0.13402525782585145,
"num_tokens": 15033646.0,
"step": 6945
},
{
"entropy": 6.559259271621704,
"epoch": 0.7437530097918562,
"grad_norm": 1.2734375,
"learning_rate": 0.0004954170989753377,
"loss": 6.4686,
"mean_token_accuracy": 0.12331727594137191,
"num_tokens": 15044655.0,
"step": 6950
},
{
"entropy": 6.714093255996704,
"epoch": 0.7442880838996201,
"grad_norm": 1.6328125,
"learning_rate": 0.0004954094184027813,
"loss": 6.5556,
"mean_token_accuracy": 0.11801010146737098,
"num_tokens": 15055492.0,
"step": 6955
},
{
"entropy": 6.666389083862304,
"epoch": 0.744823158007384,
"grad_norm": 1.3046875,
"learning_rate": 0.0004954017314659339,
"loss": 6.53,
"mean_token_accuracy": 0.1155826836824417,
"num_tokens": 15066472.0,
"step": 6960
},
{
"entropy": 6.671958112716675,
"epoch": 0.745358232115148,
"grad_norm": 1.328125,
"learning_rate": 0.0004953940381650174,
"loss": 6.5027,
"mean_token_accuracy": 0.12497956231236458,
"num_tokens": 15076851.0,
"step": 6965
},
{
"entropy": 6.544204425811768,
"epoch": 0.7458933062229118,
"grad_norm": 1.4453125,
"learning_rate": 0.0004953863385002541,
"loss": 6.4337,
"mean_token_accuracy": 0.12553054392337798,
"num_tokens": 15086997.0,
"step": 6970
},
{
"entropy": 6.669872713088989,
"epoch": 0.7464283803306758,
"grad_norm": 1.3359375,
"learning_rate": 0.0004953786324718661,
"loss": 6.5781,
"mean_token_accuracy": 0.1174256332218647,
"num_tokens": 15098625.0,
"step": 6975
},
{
"entropy": 6.628860807418823,
"epoch": 0.7469634544384397,
"grad_norm": 1.2890625,
"learning_rate": 0.0004953709200800761,
"loss": 6.3992,
"mean_token_accuracy": 0.1296637736260891,
"num_tokens": 15108265.0,
"step": 6980
},
{
"entropy": 6.58307580947876,
"epoch": 0.7474985285462037,
"grad_norm": 1.4453125,
"learning_rate": 0.0004953632013251068,
"loss": 6.4934,
"mean_token_accuracy": 0.12321807146072387,
"num_tokens": 15118794.0,
"step": 6985
},
{
"entropy": 6.648694610595703,
"epoch": 0.7480336026539676,
"grad_norm": 1.5859375,
"learning_rate": 0.0004953554762071811,
"loss": 6.5249,
"mean_token_accuracy": 0.12125022262334824,
"num_tokens": 15130084.0,
"step": 6990
},
{
"entropy": 6.664281702041626,
"epoch": 0.7485686767617316,
"grad_norm": 1.515625,
"learning_rate": 0.0004953477447265218,
"loss": 6.5118,
"mean_token_accuracy": 0.11916638761758805,
"num_tokens": 15142193.0,
"step": 6995
},
{
"entropy": 6.637905645370483,
"epoch": 0.7491037508694954,
"grad_norm": 1.546875,
"learning_rate": 0.0004953400068833524,
"loss": 6.6153,
"mean_token_accuracy": 0.12101433128118515,
"num_tokens": 15152817.0,
"step": 7000
},
{
"entropy": 6.557371330261231,
"epoch": 0.7496388249772593,
"grad_norm": 1.3203125,
"learning_rate": 0.0004953322626778964,
"loss": 6.317,
"mean_token_accuracy": 0.13028131946921348,
"num_tokens": 15162352.0,
"step": 7005
},
{
"entropy": 6.673658227920532,
"epoch": 0.7501738990850233,
"grad_norm": 1.421875,
"learning_rate": 0.0004953245121103771,
"loss": 6.4892,
"mean_token_accuracy": 0.12943382561206818,
"num_tokens": 15173091.0,
"step": 7010
},
{
"entropy": 6.693431568145752,
"epoch": 0.7507089731927872,
"grad_norm": 1.4765625,
"learning_rate": 0.0004953167551810185,
"loss": 6.6457,
"mean_token_accuracy": 0.11348064690828323,
"num_tokens": 15184422.0,
"step": 7015
},
{
"entropy": 6.567291927337647,
"epoch": 0.7512440473005512,
"grad_norm": 1.4609375,
"learning_rate": 0.0004953089918900447,
"loss": 6.3901,
"mean_token_accuracy": 0.13483602181077003,
"num_tokens": 15194765.0,
"step": 7020
},
{
"entropy": 6.645245218276978,
"epoch": 0.751779121408315,
"grad_norm": 1.5546875,
"learning_rate": 0.0004953012222376795,
"loss": 6.5368,
"mean_token_accuracy": 0.11706858575344085,
"num_tokens": 15206007.0,
"step": 7025
},
{
"entropy": 6.627328014373779,
"epoch": 0.752314195516079,
"grad_norm": 1.4609375,
"learning_rate": 0.0004952934462241476,
"loss": 6.489,
"mean_token_accuracy": 0.12645872682332993,
"num_tokens": 15216634.0,
"step": 7030
},
{
"entropy": 6.62941837310791,
"epoch": 0.7528492696238429,
"grad_norm": 1.109375,
"learning_rate": 0.0004952856638496734,
"loss": 6.5516,
"mean_token_accuracy": 0.1180113211274147,
"num_tokens": 15226931.0,
"step": 7035
},
{
"entropy": 6.64382848739624,
"epoch": 0.7533843437316068,
"grad_norm": 1.5,
"learning_rate": 0.0004952778751144817,
"loss": 6.5267,
"mean_token_accuracy": 0.12257011234760284,
"num_tokens": 15239034.0,
"step": 7040
},
{
"entropy": 6.664216995239258,
"epoch": 0.7539194178393708,
"grad_norm": 1.390625,
"learning_rate": 0.0004952700800187971,
"loss": 6.5176,
"mean_token_accuracy": 0.12180257961153984,
"num_tokens": 15250088.0,
"step": 7045
},
{
"entropy": 6.677762746810913,
"epoch": 0.7544544919471347,
"grad_norm": 1.5,
"learning_rate": 0.000495262278562845,
"loss": 6.4659,
"mean_token_accuracy": 0.12146370336413384,
"num_tokens": 15261150.0,
"step": 7050
},
{
"entropy": 6.664985132217407,
"epoch": 0.7549895660548986,
"grad_norm": 1.28125,
"learning_rate": 0.0004952544707468506,
"loss": 6.5414,
"mean_token_accuracy": 0.12395042702555656,
"num_tokens": 15270865.0,
"step": 7055
},
{
"entropy": 6.677373027801513,
"epoch": 0.7555246401626625,
"grad_norm": 1.40625,
"learning_rate": 0.0004952466565710391,
"loss": 6.5552,
"mean_token_accuracy": 0.118691186606884,
"num_tokens": 15282809.0,
"step": 7060
},
{
"entropy": 6.694124698638916,
"epoch": 0.7560597142704265,
"grad_norm": 1.3046875,
"learning_rate": 0.0004952388360356366,
"loss": 6.5083,
"mean_token_accuracy": 0.12859989702701569,
"num_tokens": 15293166.0,
"step": 7065
},
{
"entropy": 6.616050100326538,
"epoch": 0.7565947883781904,
"grad_norm": 1.5,
"learning_rate": 0.0004952310091408685,
"loss": 6.4966,
"mean_token_accuracy": 0.12518092021346092,
"num_tokens": 15302985.0,
"step": 7070
},
{
"entropy": 6.651115751266479,
"epoch": 0.7571298624859543,
"grad_norm": 1.59375,
"learning_rate": 0.0004952231758869611,
"loss": 6.5415,
"mean_token_accuracy": 0.1267145797610283,
"num_tokens": 15314737.0,
"step": 7075
},
{
"entropy": 6.614376068115234,
"epoch": 0.7576649365937183,
"grad_norm": 1.2421875,
"learning_rate": 0.0004952153362741403,
"loss": 6.464,
"mean_token_accuracy": 0.13092187345027922,
"num_tokens": 15325441.0,
"step": 7080
},
{
"entropy": 6.599036788940429,
"epoch": 0.7582000107014821,
"grad_norm": 1.2265625,
"learning_rate": 0.0004952074903026327,
"loss": 6.3991,
"mean_token_accuracy": 0.12750762775540353,
"num_tokens": 15334939.0,
"step": 7085
},
{
"entropy": 6.595050287246704,
"epoch": 0.7587350848092461,
"grad_norm": 1.21875,
"learning_rate": 0.0004951996379726647,
"loss": 6.5734,
"mean_token_accuracy": 0.12048307284712792,
"num_tokens": 15345655.0,
"step": 7090
},
{
"entropy": 6.538129758834839,
"epoch": 0.75927015891701,
"grad_norm": 1.4375,
"learning_rate": 0.0004951917792844631,
"loss": 6.3968,
"mean_token_accuracy": 0.13058631792664527,
"num_tokens": 15356035.0,
"step": 7095
},
{
"entropy": 6.659978437423706,
"epoch": 0.759805233024774,
"grad_norm": 1.1875,
"learning_rate": 0.0004951839142382548,
"loss": 6.5094,
"mean_token_accuracy": 0.12489819005131722,
"num_tokens": 15368509.0,
"step": 7100
},
{
"entropy": 6.615253734588623,
"epoch": 0.7603403071325379,
"grad_norm": 1.171875,
"learning_rate": 0.000495176042834267,
"loss": 6.4777,
"mean_token_accuracy": 0.1264044873416424,
"num_tokens": 15379662.0,
"step": 7105
},
{
"entropy": 6.601353359222412,
"epoch": 0.7608753812403017,
"grad_norm": 1.4609375,
"learning_rate": 0.0004951681650727268,
"loss": 6.4844,
"mean_token_accuracy": 0.11898941099643708,
"num_tokens": 15389698.0,
"step": 7110
},
{
"entropy": 6.746938896179199,
"epoch": 0.7614104553480657,
"grad_norm": 1.203125,
"learning_rate": 0.0004951602809538619,
"loss": 6.5772,
"mean_token_accuracy": 0.11864528879523277,
"num_tokens": 15400724.0,
"step": 7115
},
{
"entropy": 6.733620023727417,
"epoch": 0.7619455294558296,
"grad_norm": 1.25,
"learning_rate": 0.0004951523904778997,
"loss": 6.5659,
"mean_token_accuracy": 0.11942593604326249,
"num_tokens": 15410732.0,
"step": 7120
},
{
"entropy": 6.618440341949463,
"epoch": 0.7624806035635936,
"grad_norm": 1.421875,
"learning_rate": 0.0004951444936450682,
"loss": 6.4671,
"mean_token_accuracy": 0.13049914836883544,
"num_tokens": 15420185.0,
"step": 7125
},
{
"entropy": 6.591977787017822,
"epoch": 0.7630156776713575,
"grad_norm": 1.515625,
"learning_rate": 0.0004951365904555954,
"loss": 6.4038,
"mean_token_accuracy": 0.1246684692800045,
"num_tokens": 15431904.0,
"step": 7130
},
{
"entropy": 6.575604772567749,
"epoch": 0.7635507517791215,
"grad_norm": 1.2890625,
"learning_rate": 0.0004951286809097094,
"loss": 6.5303,
"mean_token_accuracy": 0.11952223181724549,
"num_tokens": 15442650.0,
"step": 7135
},
{
"entropy": 6.564087724685669,
"epoch": 0.7640858258868853,
"grad_norm": 1.484375,
"learning_rate": 0.0004951207650076388,
"loss": 6.5026,
"mean_token_accuracy": 0.12019338682293892,
"num_tokens": 15454273.0,
"step": 7140
},
{
"entropy": 6.638538455963134,
"epoch": 0.7646208999946492,
"grad_norm": 1.2890625,
"learning_rate": 0.000495112842749612,
"loss": 6.4855,
"mean_token_accuracy": 0.1241270050406456,
"num_tokens": 15465751.0,
"step": 7145
},
{
"entropy": 6.656053924560547,
"epoch": 0.7651559741024132,
"grad_norm": 1.359375,
"learning_rate": 0.0004951049141358578,
"loss": 6.5927,
"mean_token_accuracy": 0.11400035619735718,
"num_tokens": 15475340.0,
"step": 7150
},
{
"entropy": 6.713410806655884,
"epoch": 0.7656910482101771,
"grad_norm": 1.265625,
"learning_rate": 0.0004950969791666051,
"loss": 6.5507,
"mean_token_accuracy": 0.1195062555372715,
"num_tokens": 15486246.0,
"step": 7155
},
{
"entropy": 6.696958923339844,
"epoch": 0.7662261223179411,
"grad_norm": 1.390625,
"learning_rate": 0.0004950890378420831,
"loss": 6.5556,
"mean_token_accuracy": 0.12101633250713348,
"num_tokens": 15497331.0,
"step": 7160
},
{
"entropy": 6.624008226394653,
"epoch": 0.766761196425705,
"grad_norm": 1.1328125,
"learning_rate": 0.000495081090162521,
"loss": 6.496,
"mean_token_accuracy": 0.1252144269645214,
"num_tokens": 15508036.0,
"step": 7165
},
{
"entropy": 6.651372194290161,
"epoch": 0.7672962705334689,
"grad_norm": 1.3984375,
"learning_rate": 0.0004950731361281483,
"loss": 6.4892,
"mean_token_accuracy": 0.12409828379750251,
"num_tokens": 15518381.0,
"step": 7170
},
{
"entropy": 6.648929023742676,
"epoch": 0.7678313446412328,
"grad_norm": 1.296875,
"learning_rate": 0.0004950651757391948,
"loss": 6.4927,
"mean_token_accuracy": 0.12453809827566147,
"num_tokens": 15528939.0,
"step": 7175
},
{
"entropy": 6.691262435913086,
"epoch": 0.7683664187489967,
"grad_norm": 1.25,
"learning_rate": 0.0004950572089958904,
"loss": 6.4844,
"mean_token_accuracy": 0.11920621544122696,
"num_tokens": 15539718.0,
"step": 7180
},
{
"entropy": 6.671784210205078,
"epoch": 0.7689014928567607,
"grad_norm": 1.375,
"learning_rate": 0.0004950492358984648,
"loss": 6.6442,
"mean_token_accuracy": 0.11607598587870598,
"num_tokens": 15550970.0,
"step": 7185
},
{
"entropy": 6.673526191711426,
"epoch": 0.7694365669645246,
"grad_norm": 1.296875,
"learning_rate": 0.0004950412564471486,
"loss": 6.4146,
"mean_token_accuracy": 0.1274817906320095,
"num_tokens": 15560490.0,
"step": 7190
},
{
"entropy": 6.616913270950318,
"epoch": 0.7699716410722885,
"grad_norm": 1.390625,
"learning_rate": 0.000495033270642172,
"loss": 6.4266,
"mean_token_accuracy": 0.1294262781739235,
"num_tokens": 15570188.0,
"step": 7195
},
{
"entropy": 6.589165592193604,
"epoch": 0.7705067151800524,
"grad_norm": 1.296875,
"learning_rate": 0.0004950252784837655,
"loss": 6.4459,
"mean_token_accuracy": 0.12019804939627647,
"num_tokens": 15581446.0,
"step": 7200
},
{
"entropy": 6.63636999130249,
"epoch": 0.7710417892878164,
"grad_norm": 1.8125,
"learning_rate": 0.0004950172799721601,
"loss": 6.5597,
"mean_token_accuracy": 0.12001867443323136,
"num_tokens": 15594362.0,
"step": 7205
},
{
"entropy": 6.646881103515625,
"epoch": 0.7715768633955803,
"grad_norm": 1.5390625,
"learning_rate": 0.0004950092751075866,
"loss": 6.4336,
"mean_token_accuracy": 0.13300900235772134,
"num_tokens": 15605531.0,
"step": 7210
},
{
"entropy": 6.632643985748291,
"epoch": 0.7721119375033442,
"grad_norm": 1.65625,
"learning_rate": 0.0004950012638902763,
"loss": 6.4747,
"mean_token_accuracy": 0.12664539813995362,
"num_tokens": 15614962.0,
"step": 7215
},
{
"entropy": 6.527265357971191,
"epoch": 0.7726470116111082,
"grad_norm": 2.078125,
"learning_rate": 0.0004949932463204603,
"loss": 6.5196,
"mean_token_accuracy": 0.12367821410298348,
"num_tokens": 15625475.0,
"step": 7220
},
{
"entropy": 6.600604248046875,
"epoch": 0.773182085718872,
"grad_norm": 1.1953125,
"learning_rate": 0.0004949852223983703,
"loss": 6.5051,
"mean_token_accuracy": 0.11932418122887611,
"num_tokens": 15637359.0,
"step": 7225
},
{
"entropy": 6.792949867248535,
"epoch": 0.773717159826636,
"grad_norm": 1.5078125,
"learning_rate": 0.0004949771921242379,
"loss": 6.5462,
"mean_token_accuracy": 0.117109165340662,
"num_tokens": 15648368.0,
"step": 7230
},
{
"entropy": 6.6926452159881595,
"epoch": 0.7742522339343999,
"grad_norm": 1.34375,
"learning_rate": 0.0004949691554982951,
"loss": 6.4385,
"mean_token_accuracy": 0.13027268201112746,
"num_tokens": 15659295.0,
"step": 7235
},
{
"entropy": 6.543884658813477,
"epoch": 0.7747873080421639,
"grad_norm": 1.5859375,
"learning_rate": 0.0004949611125207737,
"loss": 6.4051,
"mean_token_accuracy": 0.12192230448126792,
"num_tokens": 15669975.0,
"step": 7240
},
{
"entropy": 6.570065546035766,
"epoch": 0.7753223821499278,
"grad_norm": 1.5390625,
"learning_rate": 0.000494953063191906,
"loss": 6.4946,
"mean_token_accuracy": 0.11872415691614151,
"num_tokens": 15680343.0,
"step": 7245
},
{
"entropy": 6.683220624923706,
"epoch": 0.7758574562576916,
"grad_norm": 1.34375,
"learning_rate": 0.0004949450075119247,
"loss": 6.5525,
"mean_token_accuracy": 0.12026704177260399,
"num_tokens": 15692246.0,
"step": 7250
},
{
"entropy": 6.6822953701019285,
"epoch": 0.7763925303654556,
"grad_norm": 1.140625,
"learning_rate": 0.0004949369454810621,
"loss": 6.5062,
"mean_token_accuracy": 0.12393845468759537,
"num_tokens": 15704262.0,
"step": 7255
},
{
"entropy": 6.547556638717651,
"epoch": 0.7769276044732195,
"grad_norm": 1.5234375,
"learning_rate": 0.0004949288770995512,
"loss": 6.4991,
"mean_token_accuracy": 0.12381231859326362,
"num_tokens": 15715078.0,
"step": 7260
},
{
"entropy": 6.660402727127075,
"epoch": 0.7774626785809835,
"grad_norm": 1.3125,
"learning_rate": 0.0004949208023676249,
"loss": 6.5459,
"mean_token_accuracy": 0.1232883907854557,
"num_tokens": 15726503.0,
"step": 7265
},
{
"entropy": 6.587308025360107,
"epoch": 0.7779977526887474,
"grad_norm": 2.140625,
"learning_rate": 0.0004949127212855161,
"loss": 6.4327,
"mean_token_accuracy": 0.12670731246471406,
"num_tokens": 15738854.0,
"step": 7270
},
{
"entropy": 6.611951589584351,
"epoch": 0.7785328267965114,
"grad_norm": 1.5859375,
"learning_rate": 0.0004949046338534587,
"loss": 6.5893,
"mean_token_accuracy": 0.12085817605257035,
"num_tokens": 15750836.0,
"step": 7275
},
{
"entropy": 6.783895874023438,
"epoch": 0.7790679009042752,
"grad_norm": 1.4765625,
"learning_rate": 0.0004948965400716857,
"loss": 6.5934,
"mean_token_accuracy": 0.12155032604932785,
"num_tokens": 15762355.0,
"step": 7280
},
{
"entropy": 6.6959075927734375,
"epoch": 0.7796029750120391,
"grad_norm": 1.3828125,
"learning_rate": 0.000494888439940431,
"loss": 6.5948,
"mean_token_accuracy": 0.12169316858053207,
"num_tokens": 15772952.0,
"step": 7285
},
{
"entropy": 6.562328147888183,
"epoch": 0.7801380491198031,
"grad_norm": 1.5703125,
"learning_rate": 0.0004948803334599286,
"loss": 6.4051,
"mean_token_accuracy": 0.13314662128686905,
"num_tokens": 15782997.0,
"step": 7290
},
{
"entropy": 6.598123836517334,
"epoch": 0.780673123227567,
"grad_norm": 1.6875,
"learning_rate": 0.0004948722206304123,
"loss": 6.5065,
"mean_token_accuracy": 0.12343615815043449,
"num_tokens": 15794060.0,
"step": 7295
},
{
"entropy": 6.707820177078247,
"epoch": 0.781208197335331,
"grad_norm": 1.1953125,
"learning_rate": 0.0004948641014521167,
"loss": 6.5704,
"mean_token_accuracy": 0.12031665593385696,
"num_tokens": 15804149.0,
"step": 7300
},
{
"entropy": 6.668608617782593,
"epoch": 0.7817432714430949,
"grad_norm": 1.40625,
"learning_rate": 0.000494855975925276,
"loss": 6.496,
"mean_token_accuracy": 0.12902145832777023,
"num_tokens": 15814643.0,
"step": 7305
},
{
"entropy": 6.645229959487915,
"epoch": 0.7822783455508588,
"grad_norm": 1.4921875,
"learning_rate": 0.0004948478440501249,
"loss": 6.4438,
"mean_token_accuracy": 0.1254532441496849,
"num_tokens": 15824640.0,
"step": 7310
},
{
"entropy": 6.553537178039551,
"epoch": 0.7828134196586227,
"grad_norm": 1.40625,
"learning_rate": 0.0004948397058268982,
"loss": 6.4806,
"mean_token_accuracy": 0.12377227991819381,
"num_tokens": 15836160.0,
"step": 7315
},
{
"entropy": 6.6396448612213135,
"epoch": 0.7833484937663866,
"grad_norm": 1.453125,
"learning_rate": 0.0004948315612558308,
"loss": 6.5149,
"mean_token_accuracy": 0.12054053619503975,
"num_tokens": 15846886.0,
"step": 7320
},
{
"entropy": 6.6480477809906,
"epoch": 0.7838835678741506,
"grad_norm": 1.5859375,
"learning_rate": 0.000494823410337158,
"loss": 6.4443,
"mean_token_accuracy": 0.13052470311522485,
"num_tokens": 15857309.0,
"step": 7325
},
{
"entropy": 6.601436614990234,
"epoch": 0.7844186419819145,
"grad_norm": 1.5546875,
"learning_rate": 0.0004948152530711153,
"loss": 6.5795,
"mean_token_accuracy": 0.12191757187247276,
"num_tokens": 15869620.0,
"step": 7330
},
{
"entropy": 6.614558506011963,
"epoch": 0.7849537160896785,
"grad_norm": 1.359375,
"learning_rate": 0.0004948070894579378,
"loss": 6.4211,
"mean_token_accuracy": 0.129998816549778,
"num_tokens": 15880915.0,
"step": 7335
},
{
"entropy": 6.605517482757568,
"epoch": 0.7854887901974423,
"grad_norm": 1.296875,
"learning_rate": 0.0004947989194978616,
"loss": 6.495,
"mean_token_accuracy": 0.1268579512834549,
"num_tokens": 15891150.0,
"step": 7340
},
{
"entropy": 6.656474876403808,
"epoch": 0.7860238643052063,
"grad_norm": 1.2265625,
"learning_rate": 0.0004947907431911225,
"loss": 6.4923,
"mean_token_accuracy": 0.12626128271222115,
"num_tokens": 15902526.0,
"step": 7345
},
{
"entropy": 6.591264724731445,
"epoch": 0.7865589384129702,
"grad_norm": 1.53125,
"learning_rate": 0.0004947825605379566,
"loss": 6.4372,
"mean_token_accuracy": 0.12627596408128738,
"num_tokens": 15913425.0,
"step": 7350
},
{
"entropy": 6.613666772842407,
"epoch": 0.7870940125207341,
"grad_norm": 1.3125,
"learning_rate": 0.0004947743715386,
"loss": 6.4193,
"mean_token_accuracy": 0.12343008667230607,
"num_tokens": 15924324.0,
"step": 7355
},
{
"entropy": 6.57948784828186,
"epoch": 0.7876290866284981,
"grad_norm": 1.6640625,
"learning_rate": 0.0004947661761932894,
"loss": 6.4727,
"mean_token_accuracy": 0.12277128249406814,
"num_tokens": 15935533.0,
"step": 7360
},
{
"entropy": 6.5657641887664795,
"epoch": 0.7881641607362619,
"grad_norm": 1.3203125,
"learning_rate": 0.0004947579745022613,
"loss": 6.4295,
"mean_token_accuracy": 0.12148091346025466,
"num_tokens": 15947007.0,
"step": 7365
},
{
"entropy": 6.581766366958618,
"epoch": 0.7886992348440259,
"grad_norm": 1.421875,
"learning_rate": 0.0004947497664657527,
"loss": 6.4862,
"mean_token_accuracy": 0.12112942487001419,
"num_tokens": 15957405.0,
"step": 7370
},
{
"entropy": 6.61507477760315,
"epoch": 0.7892343089517898,
"grad_norm": 1.296875,
"learning_rate": 0.0004947415520840003,
"loss": 6.4761,
"mean_token_accuracy": 0.12532250955700874,
"num_tokens": 15969895.0,
"step": 7375
},
{
"entropy": 6.637605905532837,
"epoch": 0.7897693830595538,
"grad_norm": 1.4921875,
"learning_rate": 0.0004947333313572416,
"loss": 6.3931,
"mean_token_accuracy": 0.13630961254239082,
"num_tokens": 15979854.0,
"step": 7380
},
{
"entropy": 6.567370319366455,
"epoch": 0.7903044571673177,
"grad_norm": 1.390625,
"learning_rate": 0.0004947251042857137,
"loss": 6.3963,
"mean_token_accuracy": 0.13095270618796348,
"num_tokens": 15991341.0,
"step": 7385
},
{
"entropy": 6.531961727142334,
"epoch": 0.7908395312750816,
"grad_norm": 1.640625,
"learning_rate": 0.0004947168708696544,
"loss": 6.4554,
"mean_token_accuracy": 0.12993446215987206,
"num_tokens": 16001944.0,
"step": 7390
},
{
"entropy": 6.560093545913697,
"epoch": 0.7913746053828455,
"grad_norm": 1.2734375,
"learning_rate": 0.0004947086311093013,
"loss": 6.4423,
"mean_token_accuracy": 0.1281399607658386,
"num_tokens": 16011771.0,
"step": 7395
},
{
"entropy": 6.687586164474487,
"epoch": 0.7919096794906094,
"grad_norm": 1.3046875,
"learning_rate": 0.0004947003850048924,
"loss": 6.6005,
"mean_token_accuracy": 0.11408599764108658,
"num_tokens": 16023901.0,
"step": 7400
},
{
"entropy": 6.740469646453858,
"epoch": 0.7924447535983734,
"grad_norm": 1.4296875,
"learning_rate": 0.0004946921325566656,
"loss": 6.5191,
"mean_token_accuracy": 0.1196585789322853,
"num_tokens": 16035014.0,
"step": 7405
},
{
"entropy": 6.650988531112671,
"epoch": 0.7929798277061373,
"grad_norm": 1.2109375,
"learning_rate": 0.0004946838737648595,
"loss": 6.5891,
"mean_token_accuracy": 0.11678230240941048,
"num_tokens": 16046927.0,
"step": 7410
},
{
"entropy": 6.613309526443482,
"epoch": 0.7935149018139013,
"grad_norm": 1.296875,
"learning_rate": 0.0004946756086297124,
"loss": 6.4909,
"mean_token_accuracy": 0.12795686945319176,
"num_tokens": 16057341.0,
"step": 7415
},
{
"entropy": 6.654987621307373,
"epoch": 0.7940499759216652,
"grad_norm": 1.5078125,
"learning_rate": 0.0004946673371514628,
"loss": 6.4782,
"mean_token_accuracy": 0.12236208170652389,
"num_tokens": 16068408.0,
"step": 7420
},
{
"entropy": 6.5727095127105715,
"epoch": 0.794585050029429,
"grad_norm": 1.6875,
"learning_rate": 0.0004946590593303499,
"loss": 6.4494,
"mean_token_accuracy": 0.1296915277838707,
"num_tokens": 16078693.0,
"step": 7425
},
{
"entropy": 6.60039119720459,
"epoch": 0.795120124137193,
"grad_norm": 1.3203125,
"learning_rate": 0.0004946507751666124,
"loss": 6.4975,
"mean_token_accuracy": 0.1274369865655899,
"num_tokens": 16089127.0,
"step": 7430
},
{
"entropy": 6.586338949203491,
"epoch": 0.7956551982449569,
"grad_norm": 1.5078125,
"learning_rate": 0.0004946424846604897,
"loss": 6.4071,
"mean_token_accuracy": 0.13582724407315255,
"num_tokens": 16099726.0,
"step": 7435
},
{
"entropy": 6.615121269226075,
"epoch": 0.7961902723527209,
"grad_norm": 1.234375,
"learning_rate": 0.0004946341878122212,
"loss": 6.4573,
"mean_token_accuracy": 0.12698574587702752,
"num_tokens": 16111017.0,
"step": 7440
},
{
"entropy": 6.624665117263794,
"epoch": 0.7967253464604848,
"grad_norm": 1.484375,
"learning_rate": 0.0004946258846220462,
"loss": 6.5288,
"mean_token_accuracy": 0.12629680335521698,
"num_tokens": 16121406.0,
"step": 7445
},
{
"entropy": 6.636161994934082,
"epoch": 0.7972604205682488,
"grad_norm": 1.875,
"learning_rate": 0.0004946175750902049,
"loss": 6.4516,
"mean_token_accuracy": 0.12538782581686975,
"num_tokens": 16131587.0,
"step": 7450
},
{
"entropy": 6.5716126441955565,
"epoch": 0.7977954946760126,
"grad_norm": 2.0625,
"learning_rate": 0.0004946092592169368,
"loss": 6.4116,
"mean_token_accuracy": 0.12958415150642394,
"num_tokens": 16142578.0,
"step": 7455
},
{
"entropy": 6.619257640838623,
"epoch": 0.7983305687837765,
"grad_norm": 1.4921875,
"learning_rate": 0.0004946009370024822,
"loss": 6.426,
"mean_token_accuracy": 0.1310078866779804,
"num_tokens": 16153501.0,
"step": 7460
},
{
"entropy": 6.572661924362182,
"epoch": 0.7988656428915405,
"grad_norm": 1.890625,
"learning_rate": 0.0004945926084470814,
"loss": 6.5052,
"mean_token_accuracy": 0.12284256890416145,
"num_tokens": 16164273.0,
"step": 7465
},
{
"entropy": 6.610775899887085,
"epoch": 0.7994007169993044,
"grad_norm": 1.3046875,
"learning_rate": 0.0004945842735509749,
"loss": 6.4856,
"mean_token_accuracy": 0.12552264481782913,
"num_tokens": 16175071.0,
"step": 7470
},
{
"entropy": 6.680575323104859,
"epoch": 0.7999357911070684,
"grad_norm": 1.2421875,
"learning_rate": 0.0004945759323144034,
"loss": 6.5288,
"mean_token_accuracy": 0.11573776230216026,
"num_tokens": 16185404.0,
"step": 7475
},
{
"entropy": 6.629665231704712,
"epoch": 0.8004708652148322,
"grad_norm": 1.4609375,
"learning_rate": 0.0004945675847376077,
"loss": 6.5139,
"mean_token_accuracy": 0.12014818266034126,
"num_tokens": 16196881.0,
"step": 7480
},
{
"entropy": 6.639012718200684,
"epoch": 0.8010059393225962,
"grad_norm": 1.2734375,
"learning_rate": 0.0004945592308208288,
"loss": 6.5439,
"mean_token_accuracy": 0.12018415406346321,
"num_tokens": 16207534.0,
"step": 7485
},
{
"entropy": 6.646149730682373,
"epoch": 0.8015410134303601,
"grad_norm": 1.1875,
"learning_rate": 0.000494550870564308,
"loss": 6.4188,
"mean_token_accuracy": 0.13105716928839684,
"num_tokens": 16217371.0,
"step": 7490
},
{
"entropy": 6.6225439548492435,
"epoch": 0.802076087538124,
"grad_norm": 1.484375,
"learning_rate": 0.0004945425039682866,
"loss": 6.467,
"mean_token_accuracy": 0.1279752753674984,
"num_tokens": 16227898.0,
"step": 7495
},
{
"entropy": 6.564583015441895,
"epoch": 0.802611161645888,
"grad_norm": 1.4921875,
"learning_rate": 0.0004945341310330064,
"loss": 6.4619,
"mean_token_accuracy": 0.1270508736371994,
"num_tokens": 16238966.0,
"step": 7500
},
{
"entropy": 6.590152359008789,
"epoch": 0.8031462357536518,
"grad_norm": 1.7890625,
"learning_rate": 0.0004945257517587089,
"loss": 6.4931,
"mean_token_accuracy": 0.12701256275177003,
"num_tokens": 16250129.0,
"step": 7505
},
{
"entropy": 6.5930369853973385,
"epoch": 0.8036813098614158,
"grad_norm": 1.2421875,
"learning_rate": 0.0004945173661456361,
"loss": 6.5149,
"mean_token_accuracy": 0.12291709631681443,
"num_tokens": 16260877.0,
"step": 7510
},
{
"entropy": 6.699715852737427,
"epoch": 0.8042163839691797,
"grad_norm": 1.1640625,
"learning_rate": 0.0004945089741940303,
"loss": 6.472,
"mean_token_accuracy": 0.12913578376173973,
"num_tokens": 16270548.0,
"step": 7515
},
{
"entropy": 6.676292562484742,
"epoch": 0.8047514580769437,
"grad_norm": 1.2734375,
"learning_rate": 0.0004945005759041338,
"loss": 6.497,
"mean_token_accuracy": 0.12249005734920501,
"num_tokens": 16281014.0,
"step": 7520
},
{
"entropy": 6.624485683441162,
"epoch": 0.8052865321847076,
"grad_norm": 1.2421875,
"learning_rate": 0.0004944921712761889,
"loss": 6.5405,
"mean_token_accuracy": 0.12573966085910798,
"num_tokens": 16291653.0,
"step": 7525
},
{
"entropy": 6.5947753429412845,
"epoch": 0.8058216062924715,
"grad_norm": 1.28125,
"learning_rate": 0.0004944837603104383,
"loss": 6.3863,
"mean_token_accuracy": 0.13147774934768677,
"num_tokens": 16301703.0,
"step": 7530
},
{
"entropy": 6.560491371154785,
"epoch": 0.8063566804002354,
"grad_norm": 1.4453125,
"learning_rate": 0.0004944753430071252,
"loss": 6.4576,
"mean_token_accuracy": 0.12715979367494584,
"num_tokens": 16312252.0,
"step": 7535
},
{
"entropy": 6.584918880462647,
"epoch": 0.8068917545079993,
"grad_norm": 1.375,
"learning_rate": 0.0004944669193664923,
"loss": 6.4331,
"mean_token_accuracy": 0.12487595155835152,
"num_tokens": 16324073.0,
"step": 7540
},
{
"entropy": 6.6717156887054445,
"epoch": 0.8074268286157633,
"grad_norm": 1.25,
"learning_rate": 0.0004944584893887829,
"loss": 6.4445,
"mean_token_accuracy": 0.12484904229640961,
"num_tokens": 16334637.0,
"step": 7545
},
{
"entropy": 6.605461120605469,
"epoch": 0.8079619027235272,
"grad_norm": 1.34375,
"learning_rate": 0.0004944500530742404,
"loss": 6.5743,
"mean_token_accuracy": 0.11794036030769348,
"num_tokens": 16345581.0,
"step": 7550
},
{
"entropy": 6.6252374172210695,
"epoch": 0.8084969768312912,
"grad_norm": 1.453125,
"learning_rate": 0.0004944416104231086,
"loss": 6.5663,
"mean_token_accuracy": 0.1216730572283268,
"num_tokens": 16356995.0,
"step": 7555
},
{
"entropy": 6.688077592849732,
"epoch": 0.8090320509390551,
"grad_norm": 1.3046875,
"learning_rate": 0.0004944331614356311,
"loss": 6.5093,
"mean_token_accuracy": 0.11890427842736244,
"num_tokens": 16368841.0,
"step": 7560
},
{
"entropy": 6.692157506942749,
"epoch": 0.8095671250468189,
"grad_norm": 1.2578125,
"learning_rate": 0.0004944247061120519,
"loss": 6.4279,
"mean_token_accuracy": 0.12462588772177696,
"num_tokens": 16379295.0,
"step": 7565
},
{
"entropy": 6.576985025405884,
"epoch": 0.8101021991545829,
"grad_norm": 1.6953125,
"learning_rate": 0.0004944162444526151,
"loss": 6.4721,
"mean_token_accuracy": 0.13099455311894417,
"num_tokens": 16389512.0,
"step": 7570
},
{
"entropy": 6.476492547988892,
"epoch": 0.8106372732623468,
"grad_norm": 1.1796875,
"learning_rate": 0.0004944077764575651,
"loss": 6.3894,
"mean_token_accuracy": 0.13501969650387763,
"num_tokens": 16400720.0,
"step": 7575
},
{
"entropy": 6.580502033233643,
"epoch": 0.8111723473701108,
"grad_norm": 1.7890625,
"learning_rate": 0.0004943993021271463,
"loss": 6.4658,
"mean_token_accuracy": 0.12537092193961144,
"num_tokens": 16411278.0,
"step": 7580
},
{
"entropy": 6.67013144493103,
"epoch": 0.8117074214778747,
"grad_norm": 1.28125,
"learning_rate": 0.0004943908214616035,
"loss": 6.5088,
"mean_token_accuracy": 0.12090714648365974,
"num_tokens": 16421665.0,
"step": 7585
},
{
"entropy": 6.671138763427734,
"epoch": 0.8122424955856387,
"grad_norm": 1.25,
"learning_rate": 0.0004943823344611818,
"loss": 6.5485,
"mean_token_accuracy": 0.11637551337480545,
"num_tokens": 16432943.0,
"step": 7590
},
{
"entropy": 6.666892385482788,
"epoch": 0.8127775696934025,
"grad_norm": 1.421875,
"learning_rate": 0.0004943738411261258,
"loss": 6.451,
"mean_token_accuracy": 0.12569426372647285,
"num_tokens": 16444506.0,
"step": 7595
},
{
"entropy": 6.5838854789733885,
"epoch": 0.8133126438011664,
"grad_norm": 2.578125,
"learning_rate": 0.0004943653414566809,
"loss": 6.4841,
"mean_token_accuracy": 0.12255563437938691,
"num_tokens": 16455403.0,
"step": 7600
},
{
"entropy": 6.685585021972656,
"epoch": 0.8138477179089304,
"grad_norm": 1.7421875,
"learning_rate": 0.0004943568354530927,
"loss": 6.5825,
"mean_token_accuracy": 0.11601466611027718,
"num_tokens": 16465616.0,
"step": 7605
},
{
"entropy": 6.6593701362609865,
"epoch": 0.8143827920166943,
"grad_norm": 1.2890625,
"learning_rate": 0.0004943483231156068,
"loss": 6.5438,
"mean_token_accuracy": 0.12109178379178047,
"num_tokens": 16476850.0,
"step": 7610
},
{
"entropy": 6.668148422241211,
"epoch": 0.8149178661244583,
"grad_norm": 1.21875,
"learning_rate": 0.0004943398044444687,
"loss": 6.4975,
"mean_token_accuracy": 0.1246594287455082,
"num_tokens": 16489118.0,
"step": 7615
},
{
"entropy": 6.676653909683227,
"epoch": 0.8154529402322221,
"grad_norm": 1.3984375,
"learning_rate": 0.0004943312794399246,
"loss": 6.4363,
"mean_token_accuracy": 0.12908004522323607,
"num_tokens": 16499336.0,
"step": 7620
},
{
"entropy": 6.601690435409546,
"epoch": 0.8159880143399861,
"grad_norm": 1.875,
"learning_rate": 0.0004943227481022207,
"loss": 6.5219,
"mean_token_accuracy": 0.11842733025550842,
"num_tokens": 16510158.0,
"step": 7625
},
{
"entropy": 6.564496660232544,
"epoch": 0.81652308844775,
"grad_norm": 1.453125,
"learning_rate": 0.0004943142104316033,
"loss": 6.4621,
"mean_token_accuracy": 0.12559000104665757,
"num_tokens": 16521080.0,
"step": 7630
},
{
"entropy": 6.656149530410767,
"epoch": 0.8170581625555139,
"grad_norm": 1.296875,
"learning_rate": 0.0004943056664283189,
"loss": 6.4704,
"mean_token_accuracy": 0.12910144105553628,
"num_tokens": 16531072.0,
"step": 7635
},
{
"entropy": 6.6499816417694095,
"epoch": 0.8175932366632779,
"grad_norm": 1.359375,
"learning_rate": 0.000494297116092614,
"loss": 6.5586,
"mean_token_accuracy": 0.11900619938969612,
"num_tokens": 16542588.0,
"step": 7640
},
{
"entropy": 6.663675832748413,
"epoch": 0.8181283107710418,
"grad_norm": 1.609375,
"learning_rate": 0.0004942885594247359,
"loss": 6.4237,
"mean_token_accuracy": 0.12086946368217469,
"num_tokens": 16553466.0,
"step": 7645
},
{
"entropy": 6.618226099014282,
"epoch": 0.8186633848788057,
"grad_norm": 1.40625,
"learning_rate": 0.0004942799964249314,
"loss": 6.4553,
"mean_token_accuracy": 0.1253214955329895,
"num_tokens": 16564084.0,
"step": 7650
},
{
"entropy": 6.56568751335144,
"epoch": 0.8191984589865696,
"grad_norm": 1.3828125,
"learning_rate": 0.0004942714270934479,
"loss": 6.4666,
"mean_token_accuracy": 0.13621146976947784,
"num_tokens": 16575665.0,
"step": 7655
},
{
"entropy": 6.610156917572022,
"epoch": 0.8197335330943336,
"grad_norm": 1.2734375,
"learning_rate": 0.0004942628514305326,
"loss": 6.4752,
"mean_token_accuracy": 0.12168620154261589,
"num_tokens": 16586518.0,
"step": 7660
},
{
"entropy": 6.63839054107666,
"epoch": 0.8202686072020975,
"grad_norm": 1.2734375,
"learning_rate": 0.0004942542694364333,
"loss": 6.4753,
"mean_token_accuracy": 0.12710672691464425,
"num_tokens": 16597345.0,
"step": 7665
},
{
"entropy": 6.6390495777130125,
"epoch": 0.8208036813098614,
"grad_norm": 1.9609375,
"learning_rate": 0.0004942456811113978,
"loss": 6.4888,
"mean_token_accuracy": 0.12848000451922417,
"num_tokens": 16609586.0,
"step": 7670
},
{
"entropy": 6.635290050506592,
"epoch": 0.8213387554176254,
"grad_norm": 1.3125,
"learning_rate": 0.0004942370864556741,
"loss": 6.3679,
"mean_token_accuracy": 0.13500959426164627,
"num_tokens": 16620165.0,
"step": 7675
},
{
"entropy": 6.571346569061279,
"epoch": 0.8218738295253892,
"grad_norm": 1.640625,
"learning_rate": 0.0004942284854695104,
"loss": 6.4753,
"mean_token_accuracy": 0.12547776997089385,
"num_tokens": 16631200.0,
"step": 7680
},
{
"entropy": 6.609015321731567,
"epoch": 0.8224089036331532,
"grad_norm": 1.3125,
"learning_rate": 0.0004942198781531549,
"loss": 6.5048,
"mean_token_accuracy": 0.1238760806620121,
"num_tokens": 16642553.0,
"step": 7685
},
{
"entropy": 6.665633964538574,
"epoch": 0.8229439777409171,
"grad_norm": 1.2265625,
"learning_rate": 0.0004942112645068561,
"loss": 6.5156,
"mean_token_accuracy": 0.12217526063323021,
"num_tokens": 16653431.0,
"step": 7690
},
{
"entropy": 6.571784734725952,
"epoch": 0.8234790518486811,
"grad_norm": 1.2265625,
"learning_rate": 0.000494202644530863,
"loss": 6.4593,
"mean_token_accuracy": 0.12343985810875893,
"num_tokens": 16664880.0,
"step": 7695
},
{
"entropy": 6.5380603790283205,
"epoch": 0.824014125956445,
"grad_norm": 1.4921875,
"learning_rate": 0.0004941940182254244,
"loss": 6.4767,
"mean_token_accuracy": 0.12872528284788132,
"num_tokens": 16676224.0,
"step": 7700
},
{
"entropy": 6.582237911224365,
"epoch": 0.8245492000642088,
"grad_norm": 1.4296875,
"learning_rate": 0.0004941853855907892,
"loss": 6.5396,
"mean_token_accuracy": 0.12095732614398003,
"num_tokens": 16687732.0,
"step": 7705
},
{
"entropy": 6.682282161712647,
"epoch": 0.8250842741719728,
"grad_norm": 1.3671875,
"learning_rate": 0.0004941767466272068,
"loss": 6.4688,
"mean_token_accuracy": 0.129934361577034,
"num_tokens": 16698257.0,
"step": 7710
},
{
"entropy": 6.666169214248657,
"epoch": 0.8256193482797367,
"grad_norm": 1.1796875,
"learning_rate": 0.0004941681013349267,
"loss": 6.4383,
"mean_token_accuracy": 0.12359966635704041,
"num_tokens": 16708860.0,
"step": 7715
},
{
"entropy": 6.602855110168457,
"epoch": 0.8261544223875007,
"grad_norm": 1.34375,
"learning_rate": 0.0004941594497141985,
"loss": 6.438,
"mean_token_accuracy": 0.12608251199126244,
"num_tokens": 16719987.0,
"step": 7720
},
{
"entropy": 6.643445825576782,
"epoch": 0.8266894964952646,
"grad_norm": 1.4609375,
"learning_rate": 0.0004941507917652718,
"loss": 6.5341,
"mean_token_accuracy": 0.11999792009592056,
"num_tokens": 16730861.0,
"step": 7725
},
{
"entropy": 6.553995418548584,
"epoch": 0.8272245706030286,
"grad_norm": 1.5546875,
"learning_rate": 0.0004941421274883969,
"loss": 6.383,
"mean_token_accuracy": 0.13347533941268921,
"num_tokens": 16740968.0,
"step": 7730
},
{
"entropy": 6.578843641281128,
"epoch": 0.8277596447107924,
"grad_norm": 1.234375,
"learning_rate": 0.000494133456883824,
"loss": 6.4878,
"mean_token_accuracy": 0.12768243104219437,
"num_tokens": 16750700.0,
"step": 7735
},
{
"entropy": 6.617146635055542,
"epoch": 0.8282947188185563,
"grad_norm": 1.4609375,
"learning_rate": 0.0004941247799518031,
"loss": 6.476,
"mean_token_accuracy": 0.12326866835355758,
"num_tokens": 16760719.0,
"step": 7740
},
{
"entropy": 6.594090604782105,
"epoch": 0.8288297929263203,
"grad_norm": 1.234375,
"learning_rate": 0.0004941160966925851,
"loss": 6.3544,
"mean_token_accuracy": 0.1272714115679264,
"num_tokens": 16771331.0,
"step": 7745
},
{
"entropy": 6.494041490554809,
"epoch": 0.8293648670340842,
"grad_norm": 1.2734375,
"learning_rate": 0.0004941074071064205,
"loss": 6.4433,
"mean_token_accuracy": 0.12712007239460946,
"num_tokens": 16781306.0,
"step": 7750
},
{
"entropy": 6.638606166839599,
"epoch": 0.8298999411418482,
"grad_norm": 1.4921875,
"learning_rate": 0.0004940987111935605,
"loss": 6.4686,
"mean_token_accuracy": 0.12214233577251435,
"num_tokens": 16793013.0,
"step": 7755
},
{
"entropy": 6.612168407440185,
"epoch": 0.830435015249612,
"grad_norm": 1.2109375,
"learning_rate": 0.0004940900089542558,
"loss": 6.4737,
"mean_token_accuracy": 0.12796134501695633,
"num_tokens": 16803987.0,
"step": 7760
},
{
"entropy": 6.5910180568695065,
"epoch": 0.830970089357376,
"grad_norm": 1.2734375,
"learning_rate": 0.000494081300388758,
"loss": 6.4697,
"mean_token_accuracy": 0.11712961420416831,
"num_tokens": 16815031.0,
"step": 7765
},
{
"entropy": 6.678561735153198,
"epoch": 0.8315051634651399,
"grad_norm": 1.265625,
"learning_rate": 0.0004940725854973184,
"loss": 6.5281,
"mean_token_accuracy": 0.12335440516471863,
"num_tokens": 16826313.0,
"step": 7770
},
{
"entropy": 6.646420955657959,
"epoch": 0.8320402375729038,
"grad_norm": 1.2109375,
"learning_rate": 0.0004940638642801886,
"loss": 6.5086,
"mean_token_accuracy": 0.12422936633229256,
"num_tokens": 16838470.0,
"step": 7775
},
{
"entropy": 6.546121120452881,
"epoch": 0.8325753116806678,
"grad_norm": 1.5078125,
"learning_rate": 0.0004940551367376205,
"loss": 6.4028,
"mean_token_accuracy": 0.12903384640812873,
"num_tokens": 16849266.0,
"step": 7780
},
{
"entropy": 6.566736698150635,
"epoch": 0.8331103857884317,
"grad_norm": 1.28125,
"learning_rate": 0.0004940464028698662,
"loss": 6.5309,
"mean_token_accuracy": 0.12425655499100685,
"num_tokens": 16860720.0,
"step": 7785
},
{
"entropy": 6.63223614692688,
"epoch": 0.8336454598961957,
"grad_norm": 1.34375,
"learning_rate": 0.0004940376626771779,
"loss": 6.4921,
"mean_token_accuracy": 0.12509471029043198,
"num_tokens": 16872004.0,
"step": 7790
},
{
"entropy": 6.648682880401611,
"epoch": 0.8341805340039595,
"grad_norm": 1.9453125,
"learning_rate": 0.0004940289161598076,
"loss": 6.5292,
"mean_token_accuracy": 0.1173703834414482,
"num_tokens": 16882898.0,
"step": 7795
},
{
"entropy": 6.628533267974854,
"epoch": 0.8347156081117235,
"grad_norm": 1.515625,
"learning_rate": 0.0004940201633180084,
"loss": 6.4511,
"mean_token_accuracy": 0.11566249057650566,
"num_tokens": 16894331.0,
"step": 7800
},
{
"entropy": 6.6906242847442625,
"epoch": 0.8352506822194874,
"grad_norm": 1.1953125,
"learning_rate": 0.0004940114041520326,
"loss": 6.5387,
"mean_token_accuracy": 0.1246345192193985,
"num_tokens": 16903656.0,
"step": 7805
},
{
"entropy": 6.581471395492554,
"epoch": 0.8357857563272513,
"grad_norm": 1.1953125,
"learning_rate": 0.0004940026386621333,
"loss": 6.5019,
"mean_token_accuracy": 0.12531133443117143,
"num_tokens": 16915878.0,
"step": 7810
},
{
"entropy": 6.6755578994750975,
"epoch": 0.8363208304350153,
"grad_norm": 1.4453125,
"learning_rate": 0.0004939938668485636,
"loss": 6.5432,
"mean_token_accuracy": 0.1215199887752533,
"num_tokens": 16926577.0,
"step": 7815
},
{
"entropy": 6.565039253234863,
"epoch": 0.8368559045427791,
"grad_norm": 1.3359375,
"learning_rate": 0.0004939850887115768,
"loss": 6.4234,
"mean_token_accuracy": 0.1319645993411541,
"num_tokens": 16938012.0,
"step": 7820
},
{
"entropy": 6.569745969772339,
"epoch": 0.8373909786505431,
"grad_norm": 1.3828125,
"learning_rate": 0.0004939763042514263,
"loss": 6.4132,
"mean_token_accuracy": 0.12320912629365921,
"num_tokens": 16949007.0,
"step": 7825
},
{
"entropy": 6.491839742660522,
"epoch": 0.837926052758307,
"grad_norm": 1.2578125,
"learning_rate": 0.0004939675134683658,
"loss": 6.3726,
"mean_token_accuracy": 0.13265551403164863,
"num_tokens": 16959425.0,
"step": 7830
},
{
"entropy": 6.643397045135498,
"epoch": 0.838461126866071,
"grad_norm": 1.4765625,
"learning_rate": 0.0004939587163626491,
"loss": 6.4972,
"mean_token_accuracy": 0.11856562122702599,
"num_tokens": 16970002.0,
"step": 7835
},
{
"entropy": 6.622504758834839,
"epoch": 0.8389962009738349,
"grad_norm": 1.8359375,
"learning_rate": 0.0004939499129345302,
"loss": 6.5261,
"mean_token_accuracy": 0.12169675678014755,
"num_tokens": 16980739.0,
"step": 7840
},
{
"entropy": 6.638087224960327,
"epoch": 0.8395312750815987,
"grad_norm": 1.78125,
"learning_rate": 0.0004939411031842633,
"loss": 6.4766,
"mean_token_accuracy": 0.1255985088646412,
"num_tokens": 16991643.0,
"step": 7845
},
{
"entropy": 6.454816770553589,
"epoch": 0.8400663491893627,
"grad_norm": 1.5390625,
"learning_rate": 0.000493932287112103,
"loss": 6.3395,
"mean_token_accuracy": 0.1409579671919346,
"num_tokens": 17001975.0,
"step": 7850
},
{
"entropy": 6.637519359588623,
"epoch": 0.8406014232971266,
"grad_norm": 1.34375,
"learning_rate": 0.0004939234647183035,
"loss": 6.4786,
"mean_token_accuracy": 0.12199027836322784,
"num_tokens": 17012658.0,
"step": 7855
},
{
"entropy": 6.653904962539673,
"epoch": 0.8411364974048906,
"grad_norm": 1.3359375,
"learning_rate": 0.0004939146360031197,
"loss": 6.4875,
"mean_token_accuracy": 0.12131756246089935,
"num_tokens": 17023196.0,
"step": 7860
},
{
"entropy": 6.606409931182862,
"epoch": 0.8416715715126545,
"grad_norm": 1.4140625,
"learning_rate": 0.0004939058009668067,
"loss": 6.5086,
"mean_token_accuracy": 0.12194685339927673,
"num_tokens": 17034589.0,
"step": 7865
},
{
"entropy": 6.642972040176391,
"epoch": 0.8422066456204185,
"grad_norm": 2.25,
"learning_rate": 0.0004938969596096194,
"loss": 6.4733,
"mean_token_accuracy": 0.12159970998764039,
"num_tokens": 17046139.0,
"step": 7870
},
{
"entropy": 6.589673328399658,
"epoch": 0.8427417197281823,
"grad_norm": 1.375,
"learning_rate": 0.000493888111931813,
"loss": 6.4945,
"mean_token_accuracy": 0.12014141380786895,
"num_tokens": 17056926.0,
"step": 7875
},
{
"entropy": 6.539459180831909,
"epoch": 0.8432767938359462,
"grad_norm": 3.125,
"learning_rate": 0.0004938792579336433,
"loss": 6.4375,
"mean_token_accuracy": 0.12531604021787643,
"num_tokens": 17067339.0,
"step": 7880
},
{
"entropy": 6.603718090057373,
"epoch": 0.8438118679437102,
"grad_norm": 2.140625,
"learning_rate": 0.0004938703976153657,
"loss": 6.4518,
"mean_token_accuracy": 0.1254698507487774,
"num_tokens": 17078037.0,
"step": 7885
},
{
"entropy": 6.632722282409668,
"epoch": 0.8443469420514741,
"grad_norm": 1.2890625,
"learning_rate": 0.0004938615309772362,
"loss": 6.4862,
"mean_token_accuracy": 0.11615241914987565,
"num_tokens": 17089069.0,
"step": 7890
},
{
"entropy": 6.616069984436035,
"epoch": 0.8448820161592381,
"grad_norm": 1.2578125,
"learning_rate": 0.0004938526580195107,
"loss": 6.4657,
"mean_token_accuracy": 0.12840015292167664,
"num_tokens": 17100006.0,
"step": 7895
},
{
"entropy": 6.608232641220093,
"epoch": 0.845417090267002,
"grad_norm": 1.34375,
"learning_rate": 0.0004938437787424454,
"loss": 6.501,
"mean_token_accuracy": 0.12470883950591087,
"num_tokens": 17111896.0,
"step": 7900
},
{
"entropy": 6.601596450805664,
"epoch": 0.845952164374766,
"grad_norm": 1.9609375,
"learning_rate": 0.0004938348931462969,
"loss": 6.5061,
"mean_token_accuracy": 0.1278610810637474,
"num_tokens": 17123202.0,
"step": 7905
},
{
"entropy": 6.600058317184448,
"epoch": 0.8464872384825298,
"grad_norm": 1.4375,
"learning_rate": 0.0004938260012313215,
"loss": 6.4451,
"mean_token_accuracy": 0.12930461540818214,
"num_tokens": 17134013.0,
"step": 7910
},
{
"entropy": 6.536851596832276,
"epoch": 0.8470223125902937,
"grad_norm": 1.890625,
"learning_rate": 0.0004938171029977761,
"loss": 6.4564,
"mean_token_accuracy": 0.12916264310479164,
"num_tokens": 17144041.0,
"step": 7915
},
{
"entropy": 6.553992843627929,
"epoch": 0.8475573866980577,
"grad_norm": 1.4375,
"learning_rate": 0.0004938081984459176,
"loss": 6.4397,
"mean_token_accuracy": 0.12721440196037292,
"num_tokens": 17155334.0,
"step": 7920
},
{
"entropy": 6.6116251945495605,
"epoch": 0.8480924608058216,
"grad_norm": 1.6171875,
"learning_rate": 0.000493799287576003,
"loss": 6.3996,
"mean_token_accuracy": 0.12881384789943695,
"num_tokens": 17166243.0,
"step": 7925
},
{
"entropy": 6.599978303909301,
"epoch": 0.8486275349135856,
"grad_norm": 1.421875,
"learning_rate": 0.0004937903703882898,
"loss": 6.4664,
"mean_token_accuracy": 0.12380218654870986,
"num_tokens": 17177709.0,
"step": 7930
},
{
"entropy": 6.550316333770752,
"epoch": 0.8491626090213494,
"grad_norm": 1.28125,
"learning_rate": 0.0004937814468830353,
"loss": 6.4306,
"mean_token_accuracy": 0.12615659311413766,
"num_tokens": 17188275.0,
"step": 7935
},
{
"entropy": 6.639209222793579,
"epoch": 0.8496976831291134,
"grad_norm": 1.2890625,
"learning_rate": 0.0004937725170604975,
"loss": 6.5549,
"mean_token_accuracy": 0.12421398386359214,
"num_tokens": 17199527.0,
"step": 7940
},
{
"entropy": 6.658242416381836,
"epoch": 0.8502327572368773,
"grad_norm": 1.4453125,
"learning_rate": 0.0004937635809209339,
"loss": 6.4738,
"mean_token_accuracy": 0.12579896971583365,
"num_tokens": 17209500.0,
"step": 7945
},
{
"entropy": 6.567161989212036,
"epoch": 0.8507678313446412,
"grad_norm": 1.828125,
"learning_rate": 0.0004937546384646027,
"loss": 6.4189,
"mean_token_accuracy": 0.1228828877210617,
"num_tokens": 17219589.0,
"step": 7950
},
{
"entropy": 6.668274545669556,
"epoch": 0.8513029054524052,
"grad_norm": 1.8125,
"learning_rate": 0.0004937456896917619,
"loss": 6.4642,
"mean_token_accuracy": 0.1268440864980221,
"num_tokens": 17230508.0,
"step": 7955
},
{
"entropy": 6.628099966049194,
"epoch": 0.851837979560169,
"grad_norm": 1.328125,
"learning_rate": 0.0004937367346026702,
"loss": 6.4648,
"mean_token_accuracy": 0.12629646956920623,
"num_tokens": 17240886.0,
"step": 7960
},
{
"entropy": 6.5766232967376705,
"epoch": 0.852373053667933,
"grad_norm": 1.5078125,
"learning_rate": 0.0004937277731975858,
"loss": 6.4924,
"mean_token_accuracy": 0.12928350567817687,
"num_tokens": 17252247.0,
"step": 7965
},
{
"entropy": 6.573000049591064,
"epoch": 0.8529081277756969,
"grad_norm": 1.71875,
"learning_rate": 0.0004937188054767679,
"loss": 6.4584,
"mean_token_accuracy": 0.12772766575217248,
"num_tokens": 17262845.0,
"step": 7970
},
{
"entropy": 6.6421280860900875,
"epoch": 0.8534432018834609,
"grad_norm": 1.578125,
"learning_rate": 0.0004937098314404751,
"loss": 6.4562,
"mean_token_accuracy": 0.12444490268826484,
"num_tokens": 17274247.0,
"step": 7975
},
{
"entropy": 6.644790935516357,
"epoch": 0.8539782759912248,
"grad_norm": 1.40625,
"learning_rate": 0.0004937008510889668,
"loss": 6.4871,
"mean_token_accuracy": 0.12767765745520593,
"num_tokens": 17285471.0,
"step": 7980
},
{
"entropy": 6.589143466949463,
"epoch": 0.8545133500989887,
"grad_norm": 1.4453125,
"learning_rate": 0.0004936918644225021,
"loss": 6.5025,
"mean_token_accuracy": 0.12948852255940438,
"num_tokens": 17296243.0,
"step": 7985
},
{
"entropy": 6.579883146286011,
"epoch": 0.8550484242067526,
"grad_norm": 1.3828125,
"learning_rate": 0.0004936828714413405,
"loss": 6.4633,
"mean_token_accuracy": 0.12740985825657844,
"num_tokens": 17307301.0,
"step": 7990
},
{
"entropy": 6.545937824249267,
"epoch": 0.8555834983145165,
"grad_norm": 1.5078125,
"learning_rate": 0.0004936738721457417,
"loss": 6.3969,
"mean_token_accuracy": 0.13528102710843087,
"num_tokens": 17318550.0,
"step": 7995
},
{
"entropy": 6.5888251781463625,
"epoch": 0.8561185724222805,
"grad_norm": 1.4609375,
"learning_rate": 0.0004936648665359657,
"loss": 6.4592,
"mean_token_accuracy": 0.12461457923054695,
"num_tokens": 17328577.0,
"step": 8000
},
{
"entropy": 6.586920166015625,
"epoch": 0.8566536465300444,
"grad_norm": 1.3984375,
"learning_rate": 0.0004936558546122723,
"loss": 6.5184,
"mean_token_accuracy": 0.12257596775889397,
"num_tokens": 17341013.0,
"step": 8005
},
{
"entropy": 6.589303112030029,
"epoch": 0.8571887206378084,
"grad_norm": 1.453125,
"learning_rate": 0.0004936468363749217,
"loss": 6.4655,
"mean_token_accuracy": 0.12056496143341064,
"num_tokens": 17351956.0,
"step": 8010
},
{
"entropy": 6.6829746723175045,
"epoch": 0.8577237947455723,
"grad_norm": 1.3515625,
"learning_rate": 0.0004936378118241746,
"loss": 6.4386,
"mean_token_accuracy": 0.117622260004282,
"num_tokens": 17363465.0,
"step": 8015
},
{
"entropy": 6.584981489181518,
"epoch": 0.8582588688533361,
"grad_norm": 1.28125,
"learning_rate": 0.0004936287809602913,
"loss": 6.4964,
"mean_token_accuracy": 0.1258731223642826,
"num_tokens": 17374458.0,
"step": 8020
},
{
"entropy": 6.558772945404053,
"epoch": 0.8587939429611001,
"grad_norm": 1.546875,
"learning_rate": 0.0004936197437835329,
"loss": 6.4979,
"mean_token_accuracy": 0.12399565950036048,
"num_tokens": 17385492.0,
"step": 8025
},
{
"entropy": 6.684000682830811,
"epoch": 0.859329017068864,
"grad_norm": 1.28125,
"learning_rate": 0.0004936107002941599,
"loss": 6.446,
"mean_token_accuracy": 0.1296046145260334,
"num_tokens": 17395640.0,
"step": 8030
},
{
"entropy": 6.621039772033692,
"epoch": 0.859864091176628,
"grad_norm": 1.296875,
"learning_rate": 0.0004936016504924337,
"loss": 6.4657,
"mean_token_accuracy": 0.12447260543704033,
"num_tokens": 17405211.0,
"step": 8035
},
{
"entropy": 6.572212600708008,
"epoch": 0.8603991652843919,
"grad_norm": 1.578125,
"learning_rate": 0.0004935925943786155,
"loss": 6.4385,
"mean_token_accuracy": 0.12494494318962097,
"num_tokens": 17415421.0,
"step": 8040
},
{
"entropy": 6.5851781368255615,
"epoch": 0.8609342393921559,
"grad_norm": 1.3203125,
"learning_rate": 0.0004935835319529669,
"loss": 6.4882,
"mean_token_accuracy": 0.12248977199196816,
"num_tokens": 17426040.0,
"step": 8045
},
{
"entropy": 6.55501275062561,
"epoch": 0.8614693134999197,
"grad_norm": 1.34375,
"learning_rate": 0.0004935744632157496,
"loss": 6.4693,
"mean_token_accuracy": 0.12371492311358452,
"num_tokens": 17436305.0,
"step": 8050
},
{
"entropy": 6.602463960647583,
"epoch": 0.8620043876076836,
"grad_norm": 1.3515625,
"learning_rate": 0.0004935653881672253,
"loss": 6.5244,
"mean_token_accuracy": 0.12265571355819702,
"num_tokens": 17446786.0,
"step": 8055
},
{
"entropy": 6.612880706787109,
"epoch": 0.8625394617154476,
"grad_norm": 1.578125,
"learning_rate": 0.0004935563068076561,
"loss": 6.422,
"mean_token_accuracy": 0.1300605744123459,
"num_tokens": 17456613.0,
"step": 8060
},
{
"entropy": 6.542842960357666,
"epoch": 0.8630745358232115,
"grad_norm": 1.46875,
"learning_rate": 0.0004935472191373044,
"loss": 6.3675,
"mean_token_accuracy": 0.12812848910689353,
"num_tokens": 17467281.0,
"step": 8065
},
{
"entropy": 6.56179370880127,
"epoch": 0.8636096099309755,
"grad_norm": 1.40625,
"learning_rate": 0.0004935381251564324,
"loss": 6.4896,
"mean_token_accuracy": 0.12458937540650368,
"num_tokens": 17478271.0,
"step": 8070
},
{
"entropy": 6.59501895904541,
"epoch": 0.8641446840387393,
"grad_norm": 1.8671875,
"learning_rate": 0.0004935290248653027,
"loss": 6.4665,
"mean_token_accuracy": 0.12613890841603279,
"num_tokens": 17489519.0,
"step": 8075
},
{
"entropy": 6.614124202728272,
"epoch": 0.8646797581465033,
"grad_norm": 1.25,
"learning_rate": 0.0004935199182641781,
"loss": 6.4417,
"mean_token_accuracy": 0.12557547613978387,
"num_tokens": 17500076.0,
"step": 8080
},
{
"entropy": 6.614273643493652,
"epoch": 0.8652148322542672,
"grad_norm": 1.53125,
"learning_rate": 0.0004935108053533216,
"loss": 6.4453,
"mean_token_accuracy": 0.12125986739993096,
"num_tokens": 17509898.0,
"step": 8085
},
{
"entropy": 6.550960397720337,
"epoch": 0.8657499063620312,
"grad_norm": 1.609375,
"learning_rate": 0.0004935016861329964,
"loss": 6.406,
"mean_token_accuracy": 0.1296715520322323,
"num_tokens": 17519890.0,
"step": 8090
},
{
"entropy": 6.530986309051514,
"epoch": 0.8662849804697951,
"grad_norm": 1.2734375,
"learning_rate": 0.0004934925606034656,
"loss": 6.4381,
"mean_token_accuracy": 0.12463523522019386,
"num_tokens": 17530927.0,
"step": 8095
},
{
"entropy": 6.6647419929504395,
"epoch": 0.866820054577559,
"grad_norm": 1.484375,
"learning_rate": 0.0004934834287649928,
"loss": 6.5129,
"mean_token_accuracy": 0.12179197296500206,
"num_tokens": 17541174.0,
"step": 8100
},
{
"entropy": 6.6586121082305905,
"epoch": 0.8673551286853229,
"grad_norm": 2.21875,
"learning_rate": 0.0004934742906178417,
"loss": 6.4765,
"mean_token_accuracy": 0.11956845298409462,
"num_tokens": 17552460.0,
"step": 8105
},
{
"entropy": 6.6295825958251955,
"epoch": 0.8678902027930868,
"grad_norm": 1.5,
"learning_rate": 0.0004934651461622762,
"loss": 6.483,
"mean_token_accuracy": 0.12741912975907327,
"num_tokens": 17563908.0,
"step": 8110
},
{
"entropy": 6.586279296875,
"epoch": 0.8684252769008508,
"grad_norm": 1.1953125,
"learning_rate": 0.0004934559953985603,
"loss": 6.4953,
"mean_token_accuracy": 0.1282585471868515,
"num_tokens": 17574314.0,
"step": 8115
},
{
"entropy": 6.640317058563232,
"epoch": 0.8689603510086147,
"grad_norm": 1.3671875,
"learning_rate": 0.0004934468383269582,
"loss": 6.4568,
"mean_token_accuracy": 0.12750404104590415,
"num_tokens": 17583719.0,
"step": 8120
},
{
"entropy": 6.520741510391235,
"epoch": 0.8694954251163787,
"grad_norm": 1.5390625,
"learning_rate": 0.0004934376749477344,
"loss": 6.4474,
"mean_token_accuracy": 0.1261451207101345,
"num_tokens": 17594522.0,
"step": 8125
},
{
"entropy": 6.550720930099487,
"epoch": 0.8700304992241426,
"grad_norm": 1.5703125,
"learning_rate": 0.0004934285052611533,
"loss": 6.384,
"mean_token_accuracy": 0.13392736613750458,
"num_tokens": 17604632.0,
"step": 8130
},
{
"entropy": 6.7096068382263185,
"epoch": 0.8705655733319064,
"grad_norm": 2.53125,
"learning_rate": 0.00049341932926748,
"loss": 6.5216,
"mean_token_accuracy": 0.12129078730940819,
"num_tokens": 17615459.0,
"step": 8135
},
{
"entropy": 6.650518894195557,
"epoch": 0.8711006474396704,
"grad_norm": 1.359375,
"learning_rate": 0.0004934101469669791,
"loss": 6.4636,
"mean_token_accuracy": 0.1257311299443245,
"num_tokens": 17625800.0,
"step": 8140
},
{
"entropy": 6.517386484146118,
"epoch": 0.8716357215474343,
"grad_norm": 1.4296875,
"learning_rate": 0.0004934009583599159,
"loss": 6.3112,
"mean_token_accuracy": 0.13819433227181435,
"num_tokens": 17637264.0,
"step": 8145
},
{
"entropy": 6.515216493606568,
"epoch": 0.8721707956551983,
"grad_norm": 1.421875,
"learning_rate": 0.0004933917634465558,
"loss": 6.3499,
"mean_token_accuracy": 0.12682703286409377,
"num_tokens": 17648256.0,
"step": 8150
},
{
"entropy": 6.564549684524536,
"epoch": 0.8727058697629622,
"grad_norm": 1.625,
"learning_rate": 0.0004933825622271642,
"loss": 6.4983,
"mean_token_accuracy": 0.12459777966141701,
"num_tokens": 17659096.0,
"step": 8155
},
{
"entropy": 6.629525566101075,
"epoch": 0.8732409438707261,
"grad_norm": 1.2578125,
"learning_rate": 0.0004933733547020069,
"loss": 6.4714,
"mean_token_accuracy": 0.12370945960283279,
"num_tokens": 17669808.0,
"step": 8160
},
{
"entropy": 6.615776252746582,
"epoch": 0.87377601797849,
"grad_norm": 1.4609375,
"learning_rate": 0.0004933641408713495,
"loss": 6.4171,
"mean_token_accuracy": 0.12331488877534866,
"num_tokens": 17680375.0,
"step": 8165
},
{
"entropy": 6.513298511505127,
"epoch": 0.8743110920862539,
"grad_norm": 1.4921875,
"learning_rate": 0.0004933549207354582,
"loss": 6.3829,
"mean_token_accuracy": 0.13013615384697913,
"num_tokens": 17690780.0,
"step": 8170
},
{
"entropy": 6.601052331924438,
"epoch": 0.8748461661940179,
"grad_norm": 2.125,
"learning_rate": 0.0004933456942945994,
"loss": 6.483,
"mean_token_accuracy": 0.12874177172780038,
"num_tokens": 17702170.0,
"step": 8175
},
{
"entropy": 6.5550049304962155,
"epoch": 0.8753812403017818,
"grad_norm": 1.9609375,
"learning_rate": 0.0004933364615490393,
"loss": 6.4206,
"mean_token_accuracy": 0.12599021792411805,
"num_tokens": 17712631.0,
"step": 8180
},
{
"entropy": 6.611193418502808,
"epoch": 0.8759163144095458,
"grad_norm": 1.4375,
"learning_rate": 0.0004933272224990445,
"loss": 6.5003,
"mean_token_accuracy": 0.12176733165979385,
"num_tokens": 17723692.0,
"step": 8185
},
{
"entropy": 6.6318916320800785,
"epoch": 0.8764513885173096,
"grad_norm": 1.5234375,
"learning_rate": 0.0004933179771448819,
"loss": 6.4129,
"mean_token_accuracy": 0.13463898748159409,
"num_tokens": 17735264.0,
"step": 8190
},
{
"entropy": 6.5342247009277346,
"epoch": 0.8769864626250736,
"grad_norm": 1.3984375,
"learning_rate": 0.0004933087254868183,
"loss": 6.3354,
"mean_token_accuracy": 0.1375872351229191,
"num_tokens": 17747060.0,
"step": 8195
},
{
"entropy": 6.510075902938842,
"epoch": 0.8775215367328375,
"grad_norm": 1.6484375,
"learning_rate": 0.000493299467525121,
"loss": 6.4422,
"mean_token_accuracy": 0.12407493069767953,
"num_tokens": 17758159.0,
"step": 8200
},
{
"entropy": 6.57683801651001,
"epoch": 0.8780566108406014,
"grad_norm": 1.265625,
"learning_rate": 0.0004932902032600573,
"loss": 6.4594,
"mean_token_accuracy": 0.12217129170894622,
"num_tokens": 17768444.0,
"step": 8205
},
{
"entropy": 6.646522092819214,
"epoch": 0.8785916849483654,
"grad_norm": 1.4296875,
"learning_rate": 0.0004932809326918945,
"loss": 6.3517,
"mean_token_accuracy": 0.1339459776878357,
"num_tokens": 17779303.0,
"step": 8210
},
{
"entropy": 6.622037935256958,
"epoch": 0.8791267590561292,
"grad_norm": 1.265625,
"learning_rate": 0.0004932716558209005,
"loss": 6.5405,
"mean_token_accuracy": 0.11944242864847184,
"num_tokens": 17789853.0,
"step": 8215
},
{
"entropy": 6.602828121185302,
"epoch": 0.8796618331638932,
"grad_norm": 1.6953125,
"learning_rate": 0.0004932623726473432,
"loss": 6.3894,
"mean_token_accuracy": 0.1371438518166542,
"num_tokens": 17799232.0,
"step": 8220
},
{
"entropy": 6.487914800643921,
"epoch": 0.8801969072716571,
"grad_norm": 1.640625,
"learning_rate": 0.0004932530831714904,
"loss": 6.4097,
"mean_token_accuracy": 0.12824408933520318,
"num_tokens": 17808870.0,
"step": 8225
},
{
"entropy": 6.5319726943969725,
"epoch": 0.8807319813794211,
"grad_norm": 1.3046875,
"learning_rate": 0.0004932437873936107,
"loss": 6.4906,
"mean_token_accuracy": 0.11963605657219886,
"num_tokens": 17819394.0,
"step": 8230
},
{
"entropy": 6.662888050079346,
"epoch": 0.881267055487185,
"grad_norm": 1.484375,
"learning_rate": 0.0004932344853139721,
"loss": 6.4679,
"mean_token_accuracy": 0.12477930113673211,
"num_tokens": 17830220.0,
"step": 8235
},
{
"entropy": 6.598001623153687,
"epoch": 0.8818021295949489,
"grad_norm": 1.5234375,
"learning_rate": 0.0004932251769328436,
"loss": 6.4034,
"mean_token_accuracy": 0.1247767224907875,
"num_tokens": 17841971.0,
"step": 8240
},
{
"entropy": 6.5990701675415036,
"epoch": 0.8823372037027128,
"grad_norm": 1.359375,
"learning_rate": 0.0004932158622504937,
"loss": 6.4173,
"mean_token_accuracy": 0.12920651137828826,
"num_tokens": 17851890.0,
"step": 8245
},
{
"entropy": 6.519874525070191,
"epoch": 0.8828722778104767,
"grad_norm": 1.421875,
"learning_rate": 0.0004932065412671915,
"loss": 6.4123,
"mean_token_accuracy": 0.13252334222197532,
"num_tokens": 17862855.0,
"step": 8250
},
{
"entropy": 6.629786729812622,
"epoch": 0.8834073519182407,
"grad_norm": 1.4921875,
"learning_rate": 0.0004931972139832058,
"loss": 6.4243,
"mean_token_accuracy": 0.12248511165380478,
"num_tokens": 17873550.0,
"step": 8255
},
{
"entropy": 6.638599252700805,
"epoch": 0.8839424260260046,
"grad_norm": 2.46875,
"learning_rate": 0.0004931878803988066,
"loss": 6.4432,
"mean_token_accuracy": 0.13054898232221604,
"num_tokens": 17884067.0,
"step": 8260
},
{
"entropy": 6.601341295242309,
"epoch": 0.8844775001337686,
"grad_norm": 1.5234375,
"learning_rate": 0.0004931785405142627,
"loss": 6.4715,
"mean_token_accuracy": 0.12773810103535652,
"num_tokens": 17894587.0,
"step": 8265
},
{
"entropy": 6.557901048660279,
"epoch": 0.8850125742415325,
"grad_norm": 1.375,
"learning_rate": 0.0004931691943298443,
"loss": 6.4536,
"mean_token_accuracy": 0.12345588281750679,
"num_tokens": 17905297.0,
"step": 8270
},
{
"entropy": 6.472777080535889,
"epoch": 0.8855476483492963,
"grad_norm": 1.3046875,
"learning_rate": 0.000493159841845821,
"loss": 6.3637,
"mean_token_accuracy": 0.12871376648545266,
"num_tokens": 17916225.0,
"step": 8275
},
{
"entropy": 6.641252183914185,
"epoch": 0.8860827224570603,
"grad_norm": 2.890625,
"learning_rate": 0.0004931504830624629,
"loss": 6.4114,
"mean_token_accuracy": 0.12872292771935462,
"num_tokens": 17925795.0,
"step": 8280
},
{
"entropy": 6.560930967330933,
"epoch": 0.8866177965648242,
"grad_norm": 1.2734375,
"learning_rate": 0.0004931411179800403,
"loss": 6.4212,
"mean_token_accuracy": 0.12332068607211114,
"num_tokens": 17936134.0,
"step": 8285
},
{
"entropy": 6.576021671295166,
"epoch": 0.8871528706725882,
"grad_norm": 1.7734375,
"learning_rate": 0.0004931317465988235,
"loss": 6.4877,
"mean_token_accuracy": 0.12288779988884926,
"num_tokens": 17946393.0,
"step": 8290
},
{
"entropy": 6.620288515090943,
"epoch": 0.8876879447803521,
"grad_norm": 1.5546875,
"learning_rate": 0.0004931223689190832,
"loss": 6.4979,
"mean_token_accuracy": 0.12131898999214172,
"num_tokens": 17957527.0,
"step": 8295
},
{
"entropy": 6.60401668548584,
"epoch": 0.8882230188881161,
"grad_norm": 1.390625,
"learning_rate": 0.0004931129849410903,
"loss": 6.5001,
"mean_token_accuracy": 0.12653226852416993,
"num_tokens": 17968390.0,
"step": 8300
},
{
"entropy": 6.624185705184937,
"epoch": 0.8887580929958799,
"grad_norm": 1.484375,
"learning_rate": 0.0004931035946651154,
"loss": 6.424,
"mean_token_accuracy": 0.13672848865389825,
"num_tokens": 17978667.0,
"step": 8305
},
{
"entropy": 6.649472332000732,
"epoch": 0.8892931671036438,
"grad_norm": 1.734375,
"learning_rate": 0.0004930941980914299,
"loss": 6.484,
"mean_token_accuracy": 0.12073868960142135,
"num_tokens": 17989008.0,
"step": 8310
},
{
"entropy": 6.55554428100586,
"epoch": 0.8898282412114078,
"grad_norm": 1.6796875,
"learning_rate": 0.0004930847952203052,
"loss": 6.4167,
"mean_token_accuracy": 0.11992595419287681,
"num_tokens": 18000199.0,
"step": 8315
},
{
"entropy": 6.6081115245819095,
"epoch": 0.8903633153191717,
"grad_norm": 2.21875,
"learning_rate": 0.0004930753860520126,
"loss": 6.3714,
"mean_token_accuracy": 0.12737504318356513,
"num_tokens": 18010941.0,
"step": 8320
},
{
"entropy": 6.636107873916626,
"epoch": 0.8908983894269357,
"grad_norm": 1.3828125,
"learning_rate": 0.0004930659705868237,
"loss": 6.4755,
"mean_token_accuracy": 0.12668566182255744,
"num_tokens": 18021519.0,
"step": 8325
},
{
"entropy": 6.576786231994629,
"epoch": 0.8914334635346995,
"grad_norm": 2.296875,
"learning_rate": 0.0004930565488250108,
"loss": 6.4282,
"mean_token_accuracy": 0.12401631250977516,
"num_tokens": 18032773.0,
"step": 8330
},
{
"entropy": 6.560948514938355,
"epoch": 0.8919685376424635,
"grad_norm": 1.625,
"learning_rate": 0.0004930471207668456,
"loss": 6.3931,
"mean_token_accuracy": 0.1288905918598175,
"num_tokens": 18043250.0,
"step": 8335
},
{
"entropy": 6.577342367172241,
"epoch": 0.8925036117502274,
"grad_norm": 1.4453125,
"learning_rate": 0.0004930376864126005,
"loss": 6.4045,
"mean_token_accuracy": 0.13135718554258347,
"num_tokens": 18055165.0,
"step": 8340
},
{
"entropy": 6.533294534683227,
"epoch": 0.8930386858579913,
"grad_norm": 1.375,
"learning_rate": 0.0004930282457625476,
"loss": 6.3655,
"mean_token_accuracy": 0.1319275178015232,
"num_tokens": 18065296.0,
"step": 8345
},
{
"entropy": 6.483318853378296,
"epoch": 0.8935737599657553,
"grad_norm": 1.734375,
"learning_rate": 0.00049301879881696,
"loss": 6.3713,
"mean_token_accuracy": 0.12717922553420066,
"num_tokens": 18075536.0,
"step": 8350
},
{
"entropy": 6.523620080947876,
"epoch": 0.8941088340735192,
"grad_norm": 1.4296875,
"learning_rate": 0.0004930093455761101,
"loss": 6.4792,
"mean_token_accuracy": 0.1299550086259842,
"num_tokens": 18086253.0,
"step": 8355
},
{
"entropy": 6.625021934509277,
"epoch": 0.8946439081812831,
"grad_norm": 2.15625,
"learning_rate": 0.000492999886040271,
"loss": 6.4867,
"mean_token_accuracy": 0.12763941064476966,
"num_tokens": 18097335.0,
"step": 8360
},
{
"entropy": 6.678513479232788,
"epoch": 0.895178982289047,
"grad_norm": 1.53125,
"learning_rate": 0.0004929904202097159,
"loss": 6.4165,
"mean_token_accuracy": 0.1255946137011051,
"num_tokens": 18107293.0,
"step": 8365
},
{
"entropy": 6.574617719650268,
"epoch": 0.895714056396811,
"grad_norm": 2.0,
"learning_rate": 0.0004929809480847179,
"loss": 6.529,
"mean_token_accuracy": 0.11832507178187371,
"num_tokens": 18117705.0,
"step": 8370
},
{
"entropy": 6.6313234806060795,
"epoch": 0.8962491305045749,
"grad_norm": 2.015625,
"learning_rate": 0.0004929714696655507,
"loss": 6.4331,
"mean_token_accuracy": 0.12140770703554153,
"num_tokens": 18129003.0,
"step": 8375
},
{
"entropy": 6.6004438400268555,
"epoch": 0.8967842046123388,
"grad_norm": 1.6796875,
"learning_rate": 0.000492961984952488,
"loss": 6.3848,
"mean_token_accuracy": 0.1367577515542507,
"num_tokens": 18140843.0,
"step": 8380
},
{
"entropy": 6.557983160018921,
"epoch": 0.8973192787201028,
"grad_norm": 1.375,
"learning_rate": 0.0004929524939458037,
"loss": 6.4735,
"mean_token_accuracy": 0.12216648012399674,
"num_tokens": 18151476.0,
"step": 8385
},
{
"entropy": 6.597401762008667,
"epoch": 0.8978543528278666,
"grad_norm": 1.75,
"learning_rate": 0.0004929429966457716,
"loss": 6.4974,
"mean_token_accuracy": 0.11995471492409707,
"num_tokens": 18161304.0,
"step": 8390
},
{
"entropy": 6.655696821212769,
"epoch": 0.8983894269356306,
"grad_norm": 1.3671875,
"learning_rate": 0.0004929334930526662,
"loss": 6.4287,
"mean_token_accuracy": 0.13061922267079354,
"num_tokens": 18173696.0,
"step": 8395
},
{
"entropy": 6.572501087188721,
"epoch": 0.8989245010433945,
"grad_norm": 1.9609375,
"learning_rate": 0.0004929239831667617,
"loss": 6.4387,
"mean_token_accuracy": 0.12057856544852256,
"num_tokens": 18186024.0,
"step": 8400
},
{
"entropy": 6.567807579040528,
"epoch": 0.8994595751511585,
"grad_norm": 1.328125,
"learning_rate": 0.0004929144669883329,
"loss": 6.4628,
"mean_token_accuracy": 0.13044245541095734,
"num_tokens": 18196976.0,
"step": 8405
},
{
"entropy": 6.587893533706665,
"epoch": 0.8999946492589224,
"grad_norm": 1.8984375,
"learning_rate": 0.0004929049445176545,
"loss": 6.4553,
"mean_token_accuracy": 0.12222403511404992,
"num_tokens": 18208086.0,
"step": 8410
},
{
"entropy": 6.598808670043946,
"epoch": 0.9005297233666862,
"grad_norm": 1.4609375,
"learning_rate": 0.0004928954157550013,
"loss": 6.4026,
"mean_token_accuracy": 0.1364985041320324,
"num_tokens": 18218468.0,
"step": 8415
},
{
"entropy": 6.641229772567749,
"epoch": 0.9010647974744502,
"grad_norm": 1.640625,
"learning_rate": 0.0004928858807006488,
"loss": 6.511,
"mean_token_accuracy": 0.12634748592972755,
"num_tokens": 18230489.0,
"step": 8420
},
{
"entropy": 6.616871500015259,
"epoch": 0.9015998715822141,
"grad_norm": 1.5625,
"learning_rate": 0.0004928763393548721,
"loss": 6.4344,
"mean_token_accuracy": 0.12947860807180406,
"num_tokens": 18239968.0,
"step": 8425
},
{
"entropy": 6.5853495597839355,
"epoch": 0.9021349456899781,
"grad_norm": 1.7734375,
"learning_rate": 0.0004928667917179465,
"loss": 6.4898,
"mean_token_accuracy": 0.12196068912744522,
"num_tokens": 18250902.0,
"step": 8430
},
{
"entropy": 6.663750267028808,
"epoch": 0.902670019797742,
"grad_norm": 1.453125,
"learning_rate": 0.0004928572377901481,
"loss": 6.4837,
"mean_token_accuracy": 0.12794666811823846,
"num_tokens": 18262094.0,
"step": 8435
},
{
"entropy": 6.584063005447388,
"epoch": 0.903205093905506,
"grad_norm": 1.4296875,
"learning_rate": 0.0004928476775717524,
"loss": 6.4512,
"mean_token_accuracy": 0.12302127629518508,
"num_tokens": 18272765.0,
"step": 8440
},
{
"entropy": 6.568285655975342,
"epoch": 0.9037401680132698,
"grad_norm": 1.6796875,
"learning_rate": 0.0004928381110630358,
"loss": 6.3773,
"mean_token_accuracy": 0.13741017803549765,
"num_tokens": 18284570.0,
"step": 8445
},
{
"entropy": 6.740869760513306,
"epoch": 0.9042752421210337,
"grad_norm": 1.65625,
"learning_rate": 0.0004928285382642742,
"loss": 6.4835,
"mean_token_accuracy": 0.1234114944934845,
"num_tokens": 18295170.0,
"step": 8450
},
{
"entropy": 6.594295454025269,
"epoch": 0.9048103162287977,
"grad_norm": 1.4140625,
"learning_rate": 0.0004928189591757443,
"loss": 6.4201,
"mean_token_accuracy": 0.122423455119133,
"num_tokens": 18306388.0,
"step": 8455
},
{
"entropy": 6.5661896705627445,
"epoch": 0.9053453903365616,
"grad_norm": 1.359375,
"learning_rate": 0.0004928093737977225,
"loss": 6.429,
"mean_token_accuracy": 0.13050043657422067,
"num_tokens": 18316363.0,
"step": 8460
},
{
"entropy": 6.4683513164520265,
"epoch": 0.9058804644443256,
"grad_norm": 8.5625,
"learning_rate": 0.0004927997821304856,
"loss": 6.3861,
"mean_token_accuracy": 0.12661736011505126,
"num_tokens": 18327397.0,
"step": 8465
},
{
"entropy": 6.5845592498779295,
"epoch": 0.9064155385520895,
"grad_norm": 1.890625,
"learning_rate": 0.0004927901841743107,
"loss": 6.4818,
"mean_token_accuracy": 0.12432613149285317,
"num_tokens": 18338909.0,
"step": 8470
},
{
"entropy": 6.639092874526978,
"epoch": 0.9069506126598534,
"grad_norm": 1.703125,
"learning_rate": 0.0004927805799294747,
"loss": 6.4568,
"mean_token_accuracy": 0.12108288407325744,
"num_tokens": 18350110.0,
"step": 8475
},
{
"entropy": 6.679194068908691,
"epoch": 0.9074856867676173,
"grad_norm": 2.515625,
"learning_rate": 0.0004927709693962551,
"loss": 6.5177,
"mean_token_accuracy": 0.11563151106238365,
"num_tokens": 18361590.0,
"step": 8480
},
{
"entropy": 6.59504246711731,
"epoch": 0.9080207608753812,
"grad_norm": 1.9453125,
"learning_rate": 0.0004927613525749295,
"loss": 6.4634,
"mean_token_accuracy": 0.12366845831274986,
"num_tokens": 18372602.0,
"step": 8485
},
{
"entropy": 6.630734634399414,
"epoch": 0.9085558349831452,
"grad_norm": 1.4609375,
"learning_rate": 0.0004927517294657753,
"loss": 6.499,
"mean_token_accuracy": 0.12221666947007179,
"num_tokens": 18384653.0,
"step": 8490
},
{
"entropy": 6.684356164932251,
"epoch": 0.9090909090909091,
"grad_norm": 1.8828125,
"learning_rate": 0.0004927421000690705,
"loss": 6.4559,
"mean_token_accuracy": 0.12245200499892235,
"num_tokens": 18395379.0,
"step": 8495
},
{
"entropy": 6.63131685256958,
"epoch": 0.909625983198673,
"grad_norm": 1.7890625,
"learning_rate": 0.0004927324643850933,
"loss": 6.4256,
"mean_token_accuracy": 0.12843620777130127,
"num_tokens": 18405915.0,
"step": 8500
},
{
"entropy": 6.438924121856689,
"epoch": 0.9101610573064369,
"grad_norm": 1.6484375,
"learning_rate": 0.0004927228224141217,
"loss": 6.3293,
"mean_token_accuracy": 0.14597983360290528,
"num_tokens": 18416656.0,
"step": 8505
},
{
"entropy": 6.542063999176025,
"epoch": 0.9106961314142009,
"grad_norm": 2.5,
"learning_rate": 0.0004927131741564342,
"loss": 6.4477,
"mean_token_accuracy": 0.12013882920145988,
"num_tokens": 18427141.0,
"step": 8510
},
{
"entropy": 6.607851362228393,
"epoch": 0.9112312055219648,
"grad_norm": 1.734375,
"learning_rate": 0.0004927035196123094,
"loss": 6.4915,
"mean_token_accuracy": 0.12988717779517173,
"num_tokens": 18437135.0,
"step": 8515
},
{
"entropy": 6.685029554367065,
"epoch": 0.9117662796297287,
"grad_norm": 1.4453125,
"learning_rate": 0.000492693858782026,
"loss": 6.5639,
"mean_token_accuracy": 0.11965355947613716,
"num_tokens": 18448567.0,
"step": 8520
},
{
"entropy": 6.679243803024292,
"epoch": 0.9123013537374927,
"grad_norm": 1.5546875,
"learning_rate": 0.000492684191665863,
"loss": 6.511,
"mean_token_accuracy": 0.11756190657615662,
"num_tokens": 18459385.0,
"step": 8525
},
{
"entropy": 6.561179542541504,
"epoch": 0.9128364278452565,
"grad_norm": 1.421875,
"learning_rate": 0.0004926745182640996,
"loss": 6.4677,
"mean_token_accuracy": 0.1285193495452404,
"num_tokens": 18471376.0,
"step": 8530
},
{
"entropy": 6.649504232406616,
"epoch": 0.9133715019530205,
"grad_norm": 1.7890625,
"learning_rate": 0.0004926648385770152,
"loss": 6.5335,
"mean_token_accuracy": 0.11931721493601799,
"num_tokens": 18481950.0,
"step": 8535
},
{
"entropy": 6.654511833190918,
"epoch": 0.9139065760607844,
"grad_norm": 1.4375,
"learning_rate": 0.0004926551526048891,
"loss": 6.4524,
"mean_token_accuracy": 0.12727693393826484,
"num_tokens": 18492436.0,
"step": 8540
},
{
"entropy": 6.575570774078369,
"epoch": 0.9144416501685484,
"grad_norm": 2.390625,
"learning_rate": 0.0004926454603480009,
"loss": 6.2964,
"mean_token_accuracy": 0.12638833299279212,
"num_tokens": 18504279.0,
"step": 8545
},
{
"entropy": 6.573821878433227,
"epoch": 0.9149767242763123,
"grad_norm": 3.546875,
"learning_rate": 0.0004926357618066307,
"loss": 6.4443,
"mean_token_accuracy": 0.12484021857380867,
"num_tokens": 18515262.0,
"step": 8550
},
{
"entropy": 6.63035249710083,
"epoch": 0.9155117983840761,
"grad_norm": 1.6484375,
"learning_rate": 0.0004926260569810586,
"loss": 6.4951,
"mean_token_accuracy": 0.12227512821555138,
"num_tokens": 18525176.0,
"step": 8555
},
{
"entropy": 6.5566153049469,
"epoch": 0.9160468724918401,
"grad_norm": 1.59375,
"learning_rate": 0.0004926163458715645,
"loss": 6.4617,
"mean_token_accuracy": 0.12596529051661493,
"num_tokens": 18536674.0,
"step": 8560
},
{
"entropy": 6.5605353832244875,
"epoch": 0.916581946599604,
"grad_norm": 4.03125,
"learning_rate": 0.0004926066284784292,
"loss": 6.4496,
"mean_token_accuracy": 0.12597779259085656,
"num_tokens": 18547398.0,
"step": 8565
},
{
"entropy": 6.563134813308716,
"epoch": 0.917117020707368,
"grad_norm": 1.7578125,
"learning_rate": 0.0004925969048019329,
"loss": 6.4943,
"mean_token_accuracy": 0.12652412205934524,
"num_tokens": 18558838.0,
"step": 8570
},
{
"entropy": 6.629328727722168,
"epoch": 0.9176520948151319,
"grad_norm": 2.265625,
"learning_rate": 0.0004925871748423566,
"loss": 6.4395,
"mean_token_accuracy": 0.1258179649710655,
"num_tokens": 18569881.0,
"step": 8575
},
{
"entropy": 6.6013552188873295,
"epoch": 0.9181871689228959,
"grad_norm": 1.6640625,
"learning_rate": 0.0004925774385999812,
"loss": 6.3501,
"mean_token_accuracy": 0.13269954398274422,
"num_tokens": 18579556.0,
"step": 8580
},
{
"entropy": 6.4663420677185055,
"epoch": 0.9187222430306597,
"grad_norm": 1.71875,
"learning_rate": 0.0004925676960750878,
"loss": 6.3516,
"mean_token_accuracy": 0.1361738361418247,
"num_tokens": 18590580.0,
"step": 8585
},
{
"entropy": 6.634561157226562,
"epoch": 0.9192573171384236,
"grad_norm": 1.71875,
"learning_rate": 0.0004925579472679578,
"loss": 6.4828,
"mean_token_accuracy": 0.1265283964574337,
"num_tokens": 18600813.0,
"step": 8590
},
{
"entropy": 6.598533678054809,
"epoch": 0.9197923912461876,
"grad_norm": 1.484375,
"learning_rate": 0.0004925481921788726,
"loss": 6.391,
"mean_token_accuracy": 0.1304919548332691,
"num_tokens": 18610807.0,
"step": 8595
},
{
"entropy": 6.510357618331909,
"epoch": 0.9203274653539515,
"grad_norm": 1.515625,
"learning_rate": 0.000492538430808114,
"loss": 6.3577,
"mean_token_accuracy": 0.12540487572550774,
"num_tokens": 18622108.0,
"step": 8600
},
{
"entropy": 6.557230615615845,
"epoch": 0.9208625394617155,
"grad_norm": 1.4140625,
"learning_rate": 0.0004925286631559636,
"loss": 6.5401,
"mean_token_accuracy": 0.12565935403108597,
"num_tokens": 18633098.0,
"step": 8605
},
{
"entropy": 6.648445987701416,
"epoch": 0.9213976135694794,
"grad_norm": 1.5859375,
"learning_rate": 0.0004925188892227038,
"loss": 6.4225,
"mean_token_accuracy": 0.12731703594326974,
"num_tokens": 18643603.0,
"step": 8610
},
{
"entropy": 6.638711643218994,
"epoch": 0.9219326876772433,
"grad_norm": 2.15625,
"learning_rate": 0.0004925091090086166,
"loss": 6.4527,
"mean_token_accuracy": 0.1319430947303772,
"num_tokens": 18654354.0,
"step": 8615
},
{
"entropy": 6.621751213073731,
"epoch": 0.9224677617850072,
"grad_norm": 1.6171875,
"learning_rate": 0.0004924993225139845,
"loss": 6.5275,
"mean_token_accuracy": 0.12473874017596245,
"num_tokens": 18665349.0,
"step": 8620
},
{
"entropy": 6.635911178588867,
"epoch": 0.9230028358927711,
"grad_norm": 1.53125,
"learning_rate": 0.0004924895297390899,
"loss": 6.4661,
"mean_token_accuracy": 0.1247866302728653,
"num_tokens": 18676575.0,
"step": 8625
},
{
"entropy": 6.631935882568359,
"epoch": 0.9235379100005351,
"grad_norm": 1.4296875,
"learning_rate": 0.0004924797306842157,
"loss": 6.4019,
"mean_token_accuracy": 0.13183975219726562,
"num_tokens": 18686501.0,
"step": 8630
},
{
"entropy": 6.582294940948486,
"epoch": 0.924072984108299,
"grad_norm": 2.140625,
"learning_rate": 0.0004924699253496449,
"loss": 6.474,
"mean_token_accuracy": 0.1282108373939991,
"num_tokens": 18697246.0,
"step": 8635
},
{
"entropy": 6.533542108535767,
"epoch": 0.924608058216063,
"grad_norm": 1.921875,
"learning_rate": 0.0004924601137356605,
"loss": 6.3935,
"mean_token_accuracy": 0.1339188501238823,
"num_tokens": 18708177.0,
"step": 8640
},
{
"entropy": 6.60825548171997,
"epoch": 0.9251431323238268,
"grad_norm": 1.546875,
"learning_rate": 0.000492450295842546,
"loss": 6.458,
"mean_token_accuracy": 0.12657491490244865,
"num_tokens": 18719334.0,
"step": 8645
},
{
"entropy": 6.618236446380616,
"epoch": 0.9256782064315908,
"grad_norm": 1.6171875,
"learning_rate": 0.0004924404716705847,
"loss": 6.4091,
"mean_token_accuracy": 0.13286964371800422,
"num_tokens": 18729174.0,
"step": 8650
},
{
"entropy": 6.537341594696045,
"epoch": 0.9262132805393547,
"grad_norm": 1.703125,
"learning_rate": 0.0004924306412200603,
"loss": 6.517,
"mean_token_accuracy": 0.122911486774683,
"num_tokens": 18740277.0,
"step": 8655
},
{
"entropy": 6.627772474288941,
"epoch": 0.9267483546471186,
"grad_norm": 1.46875,
"learning_rate": 0.0004924208044912567,
"loss": 6.4767,
"mean_token_accuracy": 0.12486343458294868,
"num_tokens": 18751864.0,
"step": 8660
},
{
"entropy": 6.609443330764771,
"epoch": 0.9272834287548826,
"grad_norm": 2.828125,
"learning_rate": 0.0004924109614844579,
"loss": 6.3891,
"mean_token_accuracy": 0.1359869807958603,
"num_tokens": 18762786.0,
"step": 8665
},
{
"entropy": 6.555660629272461,
"epoch": 0.9278185028626464,
"grad_norm": 1.7890625,
"learning_rate": 0.0004924011121999482,
"loss": 6.4339,
"mean_token_accuracy": 0.13116455525159837,
"num_tokens": 18772913.0,
"step": 8670
},
{
"entropy": 6.5695160865783695,
"epoch": 0.9283535769704104,
"grad_norm": 1.5390625,
"learning_rate": 0.0004923912566380118,
"loss": 6.4058,
"mean_token_accuracy": 0.12949569448828696,
"num_tokens": 18784204.0,
"step": 8675
},
{
"entropy": 6.5018853664398195,
"epoch": 0.9288886510781743,
"grad_norm": 1.59375,
"learning_rate": 0.0004923813947989336,
"loss": 6.4945,
"mean_token_accuracy": 0.12146714031696319,
"num_tokens": 18795425.0,
"step": 8680
},
{
"entropy": 6.64973201751709,
"epoch": 0.9294237251859383,
"grad_norm": 1.796875,
"learning_rate": 0.0004923715266829981,
"loss": 6.4345,
"mean_token_accuracy": 0.1265689440071583,
"num_tokens": 18806850.0,
"step": 8685
},
{
"entropy": 6.641017436981201,
"epoch": 0.9299587992937022,
"grad_norm": 1.3671875,
"learning_rate": 0.0004923616522904903,
"loss": 6.3843,
"mean_token_accuracy": 0.1314132884144783,
"num_tokens": 18817272.0,
"step": 8690
},
{
"entropy": 6.597755098342896,
"epoch": 0.930493873401466,
"grad_norm": 1.7734375,
"learning_rate": 0.0004923517716216953,
"loss": 6.5085,
"mean_token_accuracy": 0.12215960323810578,
"num_tokens": 18827075.0,
"step": 8695
},
{
"entropy": 6.62264838218689,
"epoch": 0.93102894750923,
"grad_norm": 3.390625,
"learning_rate": 0.0004923418846768985,
"loss": 6.4421,
"mean_token_accuracy": 0.12476111203432083,
"num_tokens": 18837635.0,
"step": 8700
},
{
"entropy": 6.642798328399659,
"epoch": 0.9315640216169939,
"grad_norm": 1.5546875,
"learning_rate": 0.0004923319914563852,
"loss": 6.4275,
"mean_token_accuracy": 0.13177590519189836,
"num_tokens": 18848559.0,
"step": 8705
},
{
"entropy": 6.670168113708496,
"epoch": 0.9320990957247579,
"grad_norm": 1.8828125,
"learning_rate": 0.0004923220919604413,
"loss": 6.4681,
"mean_token_accuracy": 0.1240908034145832,
"num_tokens": 18859327.0,
"step": 8710
},
{
"entropy": 6.591074991226196,
"epoch": 0.9326341698325218,
"grad_norm": 1.5859375,
"learning_rate": 0.0004923121861893525,
"loss": 6.4075,
"mean_token_accuracy": 0.1274246610701084,
"num_tokens": 18869403.0,
"step": 8715
},
{
"entropy": 6.583342885971069,
"epoch": 0.9331692439402858,
"grad_norm": 1.7421875,
"learning_rate": 0.0004923022741434049,
"loss": 6.4379,
"mean_token_accuracy": 0.12099683731794357,
"num_tokens": 18879890.0,
"step": 8720
},
{
"entropy": 6.615302467346192,
"epoch": 0.9337043180480497,
"grad_norm": 1.9375,
"learning_rate": 0.0004922923558228846,
"loss": 6.4721,
"mean_token_accuracy": 0.12057366967201233,
"num_tokens": 18890361.0,
"step": 8725
},
{
"entropy": 6.579403305053711,
"epoch": 0.9342393921558135,
"grad_norm": 2.171875,
"learning_rate": 0.0004922824312280781,
"loss": 6.3868,
"mean_token_accuracy": 0.13047217801213265,
"num_tokens": 18902344.0,
"step": 8730
},
{
"entropy": 6.569160079956054,
"epoch": 0.9347744662635775,
"grad_norm": 1.890625,
"learning_rate": 0.000492272500359272,
"loss": 6.3847,
"mean_token_accuracy": 0.13155013620853423,
"num_tokens": 18911364.0,
"step": 8735
},
{
"entropy": 6.588317584991455,
"epoch": 0.9353095403713414,
"grad_norm": 1.7265625,
"learning_rate": 0.0004922625632167529,
"loss": 6.488,
"mean_token_accuracy": 0.11757436320185662,
"num_tokens": 18922254.0,
"step": 8740
},
{
"entropy": 6.664959383010864,
"epoch": 0.9358446144791054,
"grad_norm": 1.40625,
"learning_rate": 0.0004922526198008078,
"loss": 6.4452,
"mean_token_accuracy": 0.1260450467467308,
"num_tokens": 18932943.0,
"step": 8745
},
{
"entropy": 6.618773317337036,
"epoch": 0.9363796885868693,
"grad_norm": 1.578125,
"learning_rate": 0.0004922426701117239,
"loss": 6.5241,
"mean_token_accuracy": 0.12151310667395591,
"num_tokens": 18943523.0,
"step": 8750
},
{
"entropy": 6.609319257736206,
"epoch": 0.9369147626946333,
"grad_norm": 1.4296875,
"learning_rate": 0.0004922327141497884,
"loss": 6.508,
"mean_token_accuracy": 0.1161932684481144,
"num_tokens": 18955220.0,
"step": 8755
},
{
"entropy": 6.675847339630127,
"epoch": 0.9374498368023971,
"grad_norm": 1.5390625,
"learning_rate": 0.0004922227519152888,
"loss": 6.5421,
"mean_token_accuracy": 0.12342641204595566,
"num_tokens": 18966714.0,
"step": 8760
},
{
"entropy": 6.56199951171875,
"epoch": 0.937984910910161,
"grad_norm": 1.8203125,
"learning_rate": 0.0004922127834085128,
"loss": 6.3913,
"mean_token_accuracy": 0.12646044045686722,
"num_tokens": 18977998.0,
"step": 8765
},
{
"entropy": 6.567882633209228,
"epoch": 0.938519985017925,
"grad_norm": 1.9609375,
"learning_rate": 0.0004922028086297481,
"loss": 6.3818,
"mean_token_accuracy": 0.127099596709013,
"num_tokens": 18990210.0,
"step": 8770
},
{
"entropy": 6.6765382289886475,
"epoch": 0.9390550591256889,
"grad_norm": 1.5078125,
"learning_rate": 0.0004921928275792828,
"loss": 6.4995,
"mean_token_accuracy": 0.12616633251309395,
"num_tokens": 19000529.0,
"step": 8775
},
{
"entropy": 6.628943872451782,
"epoch": 0.9395901332334529,
"grad_norm": 2.6875,
"learning_rate": 0.0004921828402574052,
"loss": 6.4251,
"mean_token_accuracy": 0.13188672289252282,
"num_tokens": 19011386.0,
"step": 8780
},
{
"entropy": 6.550263404846191,
"epoch": 0.9401252073412167,
"grad_norm": 1.6640625,
"learning_rate": 0.0004921728466644037,
"loss": 6.4043,
"mean_token_accuracy": 0.12869123220443726,
"num_tokens": 19022196.0,
"step": 8785
},
{
"entropy": 6.533733606338501,
"epoch": 0.9406602814489807,
"grad_norm": 1.5625,
"learning_rate": 0.0004921628468005666,
"loss": 6.4358,
"mean_token_accuracy": 0.13189285323023797,
"num_tokens": 19033221.0,
"step": 8790
},
{
"entropy": 6.611112642288208,
"epoch": 0.9411953555567446,
"grad_norm": 1.4765625,
"learning_rate": 0.000492152840666183,
"loss": 6.4878,
"mean_token_accuracy": 0.12585045173764228,
"num_tokens": 19044953.0,
"step": 8795
},
{
"entropy": 6.711111831665039,
"epoch": 0.9417304296645085,
"grad_norm": 1.515625,
"learning_rate": 0.0004921428282615415,
"loss": 6.5018,
"mean_token_accuracy": 0.12096112072467805,
"num_tokens": 19055512.0,
"step": 8800
},
{
"entropy": 6.635407400131226,
"epoch": 0.9422655037722725,
"grad_norm": 1.5859375,
"learning_rate": 0.0004921328095869313,
"loss": 6.5027,
"mean_token_accuracy": 0.1203993707895279,
"num_tokens": 19065659.0,
"step": 8805
},
{
"entropy": 6.658339595794677,
"epoch": 0.9428005778800364,
"grad_norm": 1.8125,
"learning_rate": 0.0004921227846426418,
"loss": 6.433,
"mean_token_accuracy": 0.1371263563632965,
"num_tokens": 19076955.0,
"step": 8810
},
{
"entropy": 6.572042465209961,
"epoch": 0.9433356519878003,
"grad_norm": 1.625,
"learning_rate": 0.0004921127534289624,
"loss": 6.5106,
"mean_token_accuracy": 0.11860493347048759,
"num_tokens": 19088305.0,
"step": 8815
},
{
"entropy": 6.568140172958374,
"epoch": 0.9438707260955642,
"grad_norm": 1.65625,
"learning_rate": 0.0004921027159461829,
"loss": 6.5098,
"mean_token_accuracy": 0.12441008612513542,
"num_tokens": 19098046.0,
"step": 8820
},
{
"entropy": 6.6841308116912845,
"epoch": 0.9444058002033282,
"grad_norm": 1.75,
"learning_rate": 0.0004920926721945927,
"loss": 6.4226,
"mean_token_accuracy": 0.12864333316683768,
"num_tokens": 19109019.0,
"step": 8825
},
{
"entropy": 6.662001609802246,
"epoch": 0.9449408743110921,
"grad_norm": 1.484375,
"learning_rate": 0.0004920826221744823,
"loss": 6.4228,
"mean_token_accuracy": 0.12871572971343995,
"num_tokens": 19119624.0,
"step": 8830
},
{
"entropy": 6.636339902877808,
"epoch": 0.945475948418856,
"grad_norm": 1.4453125,
"learning_rate": 0.0004920725658861417,
"loss": 6.4741,
"mean_token_accuracy": 0.12303901016712189,
"num_tokens": 19130577.0,
"step": 8835
},
{
"entropy": 6.589366340637207,
"epoch": 0.94601102252662,
"grad_norm": 1.4140625,
"learning_rate": 0.0004920625033298611,
"loss": 6.3814,
"mean_token_accuracy": 0.12903112173080444,
"num_tokens": 19140254.0,
"step": 8840
},
{
"entropy": 6.569903993606568,
"epoch": 0.9465460966343838,
"grad_norm": 1.6015625,
"learning_rate": 0.0004920524345059314,
"loss": 6.4247,
"mean_token_accuracy": 0.12659077122807502,
"num_tokens": 19151357.0,
"step": 8845
},
{
"entropy": 6.6806480884552,
"epoch": 0.9470811707421478,
"grad_norm": 1.3515625,
"learning_rate": 0.0004920423594146431,
"loss": 6.4208,
"mean_token_accuracy": 0.13126432448625563,
"num_tokens": 19162069.0,
"step": 8850
},
{
"entropy": 6.546973991394043,
"epoch": 0.9476162448499117,
"grad_norm": 1.484375,
"learning_rate": 0.0004920322780562872,
"loss": 6.3129,
"mean_token_accuracy": 0.13278165608644485,
"num_tokens": 19172515.0,
"step": 8855
},
{
"entropy": 6.469416666030884,
"epoch": 0.9481513189576757,
"grad_norm": 1.3125,
"learning_rate": 0.0004920221904311546,
"loss": 6.3304,
"mean_token_accuracy": 0.1281422682106495,
"num_tokens": 19182450.0,
"step": 8860
},
{
"entropy": 6.6383439064025875,
"epoch": 0.9486863930654396,
"grad_norm": 1.53125,
"learning_rate": 0.0004920120965395369,
"loss": 6.5292,
"mean_token_accuracy": 0.12080127596855164,
"num_tokens": 19193833.0,
"step": 8865
},
{
"entropy": 6.678140926361084,
"epoch": 0.9492214671732034,
"grad_norm": 2.640625,
"learning_rate": 0.0004920019963817253,
"loss": 6.4762,
"mean_token_accuracy": 0.1225566104054451,
"num_tokens": 19204970.0,
"step": 8870
},
{
"entropy": 6.609140682220459,
"epoch": 0.9497565412809674,
"grad_norm": 1.7265625,
"learning_rate": 0.0004919918899580117,
"loss": 6.4331,
"mean_token_accuracy": 0.1269660323858261,
"num_tokens": 19215795.0,
"step": 8875
},
{
"entropy": 6.611617612838745,
"epoch": 0.9502916153887313,
"grad_norm": 1.8671875,
"learning_rate": 0.0004919817772686876,
"loss": 6.4649,
"mean_token_accuracy": 0.11650814116001129,
"num_tokens": 19226375.0,
"step": 8880
},
{
"entropy": 6.680630016326904,
"epoch": 0.9508266894964953,
"grad_norm": 1.8515625,
"learning_rate": 0.0004919716583140453,
"loss": 6.468,
"mean_token_accuracy": 0.13823570236563681,
"num_tokens": 19236066.0,
"step": 8885
},
{
"entropy": 6.529271268844605,
"epoch": 0.9513617636042592,
"grad_norm": 2.203125,
"learning_rate": 0.0004919615330943769,
"loss": 6.3353,
"mean_token_accuracy": 0.1377024307847023,
"num_tokens": 19246757.0,
"step": 8890
},
{
"entropy": 6.614864778518677,
"epoch": 0.9518968377120232,
"grad_norm": 1.4921875,
"learning_rate": 0.0004919514016099746,
"loss": 6.5,
"mean_token_accuracy": 0.12600353509187698,
"num_tokens": 19257875.0,
"step": 8895
},
{
"entropy": 6.6503785133361815,
"epoch": 0.952431911819787,
"grad_norm": 1.5,
"learning_rate": 0.000491941263861131,
"loss": 6.4781,
"mean_token_accuracy": 0.125810706615448,
"num_tokens": 19270236.0,
"step": 8900
},
{
"entropy": 6.588436317443848,
"epoch": 0.9529669859275509,
"grad_norm": 2.609375,
"learning_rate": 0.000491931119848139,
"loss": 6.4612,
"mean_token_accuracy": 0.12382268756628037,
"num_tokens": 19280877.0,
"step": 8905
},
{
"entropy": 6.525938892364502,
"epoch": 0.9535020600353149,
"grad_norm": 2.0,
"learning_rate": 0.0004919209695712915,
"loss": 6.3569,
"mean_token_accuracy": 0.13123888075351714,
"num_tokens": 19292852.0,
"step": 8910
},
{
"entropy": 6.636586141586304,
"epoch": 0.9540371341430788,
"grad_norm": 7.90625,
"learning_rate": 0.0004919108130308814,
"loss": 6.5481,
"mean_token_accuracy": 0.11724269837141037,
"num_tokens": 19305516.0,
"step": 8915
},
{
"entropy": 6.625235986709595,
"epoch": 0.9545722082508428,
"grad_norm": 1.4609375,
"learning_rate": 0.0004919006502272019,
"loss": 6.3795,
"mean_token_accuracy": 0.13500440642237663,
"num_tokens": 19315436.0,
"step": 8920
},
{
"entropy": 6.508727741241455,
"epoch": 0.9551072823586066,
"grad_norm": 1.453125,
"learning_rate": 0.0004918904811605468,
"loss": 6.4376,
"mean_token_accuracy": 0.1264245629310608,
"num_tokens": 19325393.0,
"step": 8925
},
{
"entropy": 6.5293982982635494,
"epoch": 0.9556423564663706,
"grad_norm": 1.71875,
"learning_rate": 0.0004918803058312094,
"loss": 6.4306,
"mean_token_accuracy": 0.12973214834928512,
"num_tokens": 19336722.0,
"step": 8930
},
{
"entropy": 6.7406854152679445,
"epoch": 0.9561774305741345,
"grad_norm": 1.796875,
"learning_rate": 0.0004918701242394837,
"loss": 6.5851,
"mean_token_accuracy": 0.11917414888739586,
"num_tokens": 19348406.0,
"step": 8935
},
{
"entropy": 6.599468946456909,
"epoch": 0.9567125046818984,
"grad_norm": 1.609375,
"learning_rate": 0.0004918599363856637,
"loss": 6.3923,
"mean_token_accuracy": 0.1359890453517437,
"num_tokens": 19359714.0,
"step": 8940
},
{
"entropy": 6.630413579940796,
"epoch": 0.9572475787896624,
"grad_norm": 1.765625,
"learning_rate": 0.0004918497422700434,
"loss": 6.4143,
"mean_token_accuracy": 0.12925480753183366,
"num_tokens": 19370833.0,
"step": 8945
},
{
"entropy": 6.610364055633545,
"epoch": 0.9577826528974263,
"grad_norm": 1.5078125,
"learning_rate": 0.0004918395418929174,
"loss": 6.4944,
"mean_token_accuracy": 0.12302698343992233,
"num_tokens": 19382098.0,
"step": 8950
},
{
"entropy": 6.624408388137818,
"epoch": 0.9583177270051902,
"grad_norm": 2.015625,
"learning_rate": 0.00049182933525458,
"loss": 6.4757,
"mean_token_accuracy": 0.12349115982651711,
"num_tokens": 19393595.0,
"step": 8955
},
{
"entropy": 6.640421295166016,
"epoch": 0.9588528011129541,
"grad_norm": 1.625,
"learning_rate": 0.000491819122355326,
"loss": 6.4125,
"mean_token_accuracy": 0.13156883418560028,
"num_tokens": 19405296.0,
"step": 8960
},
{
"entropy": 6.634973096847534,
"epoch": 0.9593878752207181,
"grad_norm": 1.4609375,
"learning_rate": 0.0004918089031954503,
"loss": 6.4472,
"mean_token_accuracy": 0.12182890325784683,
"num_tokens": 19415693.0,
"step": 8965
},
{
"entropy": 6.580959510803223,
"epoch": 0.959922949328482,
"grad_norm": 1.78125,
"learning_rate": 0.000491798677775248,
"loss": 6.4961,
"mean_token_accuracy": 0.12708681151270868,
"num_tokens": 19426140.0,
"step": 8970
},
{
"entropy": 6.693287134170532,
"epoch": 0.9604580234362459,
"grad_norm": 1.578125,
"learning_rate": 0.0004917884460950144,
"loss": 6.5297,
"mean_token_accuracy": 0.1222914569079876,
"num_tokens": 19437543.0,
"step": 8975
},
{
"entropy": 6.634545707702637,
"epoch": 0.9609930975440099,
"grad_norm": 1.390625,
"learning_rate": 0.000491778208155045,
"loss": 6.4384,
"mean_token_accuracy": 0.1252037465572357,
"num_tokens": 19447788.0,
"step": 8980
},
{
"entropy": 6.65233883857727,
"epoch": 0.9615281716517737,
"grad_norm": 1.796875,
"learning_rate": 0.0004917679639556351,
"loss": 6.4577,
"mean_token_accuracy": 0.12330029979348182,
"num_tokens": 19457587.0,
"step": 8985
},
{
"entropy": 6.6310333728790285,
"epoch": 0.9620632457595377,
"grad_norm": 1.8515625,
"learning_rate": 0.0004917577134970808,
"loss": 6.3826,
"mean_token_accuracy": 0.1255684621632099,
"num_tokens": 19467422.0,
"step": 8990
},
{
"entropy": 6.586491394042969,
"epoch": 0.9625983198673016,
"grad_norm": 1.609375,
"learning_rate": 0.0004917474567796778,
"loss": 6.4322,
"mean_token_accuracy": 0.12582321763038634,
"num_tokens": 19479037.0,
"step": 8995
},
{
"entropy": 6.5694173812866214,
"epoch": 0.9631333939750656,
"grad_norm": 1.5546875,
"learning_rate": 0.0004917371938037226,
"loss": 6.3603,
"mean_token_accuracy": 0.1418306939303875,
"num_tokens": 19489639.0,
"step": 9000
},
{
"epoch": 0.9631333939750656,
"eval_entropy": 6.45046752675203,
"eval_loss": 6.510419845581055,
"eval_mean_token_accuracy": 0.12865163465319177,
"eval_num_tokens": 19489639.0,
"eval_runtime": 22.7134,
"eval_samples_per_second": 1306.673,
"eval_steps_per_second": 163.34,
"step": 9000
}
],
"logging_steps": 5,
"max_steps": 93440,
"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": 2.8647633788928e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink