fpadovani/nld-latn-100mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
nld-latn-100mb-100mb_seed3407/checkpoint-12000/trainer_state.json
ModelHub XC 0e904dde1b 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/nld-latn-100mb-100mb_seed3407
Source: Original Platform
2026-07-17 10:12:13 +08:00

24079 lines
658 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0885341074020318,
"eval_steps": 3000,
"global_step": 12000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.69201955795288,
"epoch": 0.00045355587808418,
"grad_norm": 14.125,
"learning_rate": 2e-06,
"loss": 10.8421,
"mean_token_accuracy": 0.0,
"num_tokens": 10827.0,
"step": 5
},
{
"entropy": 10.691895008087158,
"epoch": 0.00090711175616836,
"grad_norm": 13.0,
"learning_rate": 4.5e-06,
"loss": 10.7626,
"mean_token_accuracy": 0.0,
"num_tokens": 21025.0,
"step": 10
},
{
"entropy": 10.690971279144287,
"epoch": 0.00136066763425254,
"grad_norm": 9.875,
"learning_rate": 7e-06,
"loss": 10.5141,
"mean_token_accuracy": 0.02440520448144525,
"num_tokens": 31678.0,
"step": 15
},
{
"entropy": 10.678609657287598,
"epoch": 0.00181422351233672,
"grad_norm": 6.3125,
"learning_rate": 9.5e-06,
"loss": 10.1495,
"mean_token_accuracy": 0.04602932706475258,
"num_tokens": 41025.0,
"step": 20
},
{
"entropy": 10.615153217315674,
"epoch": 0.0022677793904208998,
"grad_norm": 4.25,
"learning_rate": 1.2e-05,
"loss": 9.9137,
"mean_token_accuracy": 0.043584797717630866,
"num_tokens": 51270.0,
"step": 25
},
{
"entropy": 10.595486927032471,
"epoch": 0.00272133526850508,
"grad_norm": 3.3125,
"learning_rate": 1.4500000000000002e-05,
"loss": 9.7488,
"mean_token_accuracy": 0.04369584359228611,
"num_tokens": 62142.0,
"step": 30
},
{
"entropy": 10.605365085601807,
"epoch": 0.00317489114658926,
"grad_norm": 2.890625,
"learning_rate": 1.7000000000000003e-05,
"loss": 9.6708,
"mean_token_accuracy": 0.0410692248493433,
"num_tokens": 72709.0,
"step": 35
},
{
"entropy": 10.59789743423462,
"epoch": 0.00362844702467344,
"grad_norm": 2.609375,
"learning_rate": 1.95e-05,
"loss": 9.6037,
"mean_token_accuracy": 0.042245643213391304,
"num_tokens": 82313.0,
"step": 40
},
{
"entropy": 10.562069129943847,
"epoch": 0.00408200290275762,
"grad_norm": 2.859375,
"learning_rate": 2.2e-05,
"loss": 9.5375,
"mean_token_accuracy": 0.047857620567083356,
"num_tokens": 92450.0,
"step": 45
},
{
"entropy": 10.572824478149414,
"epoch": 0.0045355587808417995,
"grad_norm": 2.5,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.5533,
"mean_token_accuracy": 0.04238316640257835,
"num_tokens": 103978.0,
"step": 50
},
{
"entropy": 10.59686508178711,
"epoch": 0.00498911465892598,
"grad_norm": 2.28125,
"learning_rate": 2.7e-05,
"loss": 9.451,
"mean_token_accuracy": 0.04611069448292256,
"num_tokens": 114134.0,
"step": 55
},
{
"entropy": 10.546150016784669,
"epoch": 0.00544267053701016,
"grad_norm": 2.40625,
"learning_rate": 2.95e-05,
"loss": 9.4197,
"mean_token_accuracy": 0.049464210495352746,
"num_tokens": 125037.0,
"step": 60
},
{
"entropy": 10.460656833648681,
"epoch": 0.00589622641509434,
"grad_norm": 2.40625,
"learning_rate": 3.2e-05,
"loss": 9.2345,
"mean_token_accuracy": 0.06379429697990417,
"num_tokens": 134721.0,
"step": 65
},
{
"entropy": 10.382928371429443,
"epoch": 0.00634978229317852,
"grad_norm": 2.546875,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.143,
"mean_token_accuracy": 0.06433153599500656,
"num_tokens": 144448.0,
"step": 70
},
{
"entropy": 10.468272972106934,
"epoch": 0.006803338171262699,
"grad_norm": 2.40625,
"learning_rate": 3.7e-05,
"loss": 9.0501,
"mean_token_accuracy": 0.0702673528343439,
"num_tokens": 154343.0,
"step": 75
},
{
"entropy": 10.42321710586548,
"epoch": 0.00725689404934688,
"grad_norm": 2.359375,
"learning_rate": 3.95e-05,
"loss": 9.0122,
"mean_token_accuracy": 0.07085253149271012,
"num_tokens": 165650.0,
"step": 80
},
{
"entropy": 10.375396060943604,
"epoch": 0.007710449927431059,
"grad_norm": 2.21875,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.0095,
"mean_token_accuracy": 0.06769064106047154,
"num_tokens": 177098.0,
"step": 85
},
{
"entropy": 10.209734153747558,
"epoch": 0.00816400580551524,
"grad_norm": 2.46875,
"learning_rate": 4.45e-05,
"loss": 8.7346,
"mean_token_accuracy": 0.08102216497063637,
"num_tokens": 187903.0,
"step": 90
},
{
"entropy": 10.184309768676759,
"epoch": 0.00861756168359942,
"grad_norm": 2.9375,
"learning_rate": 4.7000000000000004e-05,
"loss": 8.6625,
"mean_token_accuracy": 0.07625751569867134,
"num_tokens": 198077.0,
"step": 95
},
{
"entropy": 10.164143085479736,
"epoch": 0.009071117561683599,
"grad_norm": 2.0625,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.6488,
"mean_token_accuracy": 0.0765529602766037,
"num_tokens": 208641.0,
"step": 100
},
{
"entropy": 10.11404504776001,
"epoch": 0.009524673439767779,
"grad_norm": 2.21875,
"learning_rate": 5.2e-05,
"loss": 8.4459,
"mean_token_accuracy": 0.07904929965734482,
"num_tokens": 218398.0,
"step": 105
},
{
"entropy": 10.020095443725586,
"epoch": 0.00997822931785196,
"grad_norm": 1.796875,
"learning_rate": 5.45e-05,
"loss": 8.3096,
"mean_token_accuracy": 0.08072945550084114,
"num_tokens": 229059.0,
"step": 110
},
{
"entropy": 9.92689781188965,
"epoch": 0.01043178519593614,
"grad_norm": 1.6796875,
"learning_rate": 5.7e-05,
"loss": 8.2562,
"mean_token_accuracy": 0.07845577970147133,
"num_tokens": 239775.0,
"step": 115
},
{
"entropy": 9.88659610748291,
"epoch": 0.01088534107402032,
"grad_norm": 1.90625,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.0783,
"mean_token_accuracy": 0.08167890310287476,
"num_tokens": 249954.0,
"step": 120
},
{
"entropy": 9.705071830749512,
"epoch": 0.011338896952104499,
"grad_norm": 1.6875,
"learning_rate": 6.2e-05,
"loss": 7.9982,
"mean_token_accuracy": 0.08171927258372307,
"num_tokens": 260056.0,
"step": 125
},
{
"entropy": 9.503845977783204,
"epoch": 0.01179245283018868,
"grad_norm": 1.7421875,
"learning_rate": 6.450000000000001e-05,
"loss": 7.8979,
"mean_token_accuracy": 0.08274013251066208,
"num_tokens": 269904.0,
"step": 130
},
{
"entropy": 9.399583435058593,
"epoch": 0.01224600870827286,
"grad_norm": 1.640625,
"learning_rate": 6.7e-05,
"loss": 7.8002,
"mean_token_accuracy": 0.08209368959069252,
"num_tokens": 280900.0,
"step": 135
},
{
"entropy": 9.210491371154784,
"epoch": 0.01269956458635704,
"grad_norm": 1.84375,
"learning_rate": 6.950000000000001e-05,
"loss": 7.7037,
"mean_token_accuracy": 0.08104285597801208,
"num_tokens": 290844.0,
"step": 140
},
{
"entropy": 8.886974906921386,
"epoch": 0.013153120464441219,
"grad_norm": 1.203125,
"learning_rate": 7.2e-05,
"loss": 7.6102,
"mean_token_accuracy": 0.08451902270317077,
"num_tokens": 301345.0,
"step": 145
},
{
"entropy": 8.810495853424072,
"epoch": 0.013606676342525399,
"grad_norm": 1.2109375,
"learning_rate": 7.45e-05,
"loss": 7.5207,
"mean_token_accuracy": 0.08492689803242684,
"num_tokens": 312681.0,
"step": 150
},
{
"entropy": 8.537842845916748,
"epoch": 0.01406023222060958,
"grad_norm": 1.0546875,
"learning_rate": 7.7e-05,
"loss": 7.4392,
"mean_token_accuracy": 0.08759672790765763,
"num_tokens": 323492.0,
"step": 155
},
{
"entropy": 8.368613147735596,
"epoch": 0.01451378809869376,
"grad_norm": 1.4296875,
"learning_rate": 7.950000000000001e-05,
"loss": 7.4411,
"mean_token_accuracy": 0.08510910198092461,
"num_tokens": 334521.0,
"step": 160
},
{
"entropy": 8.248584079742432,
"epoch": 0.014967343976777939,
"grad_norm": 1.1953125,
"learning_rate": 8.2e-05,
"loss": 7.4246,
"mean_token_accuracy": 0.08887577280402184,
"num_tokens": 345468.0,
"step": 165
},
{
"entropy": 8.11111536026001,
"epoch": 0.015420899854862119,
"grad_norm": 1.4375,
"learning_rate": 8.450000000000001e-05,
"loss": 7.3297,
"mean_token_accuracy": 0.0877467267215252,
"num_tokens": 355911.0,
"step": 170
},
{
"entropy": 8.027539014816284,
"epoch": 0.0158744557329463,
"grad_norm": 1.25,
"learning_rate": 8.7e-05,
"loss": 7.3504,
"mean_token_accuracy": 0.08467918261885643,
"num_tokens": 367537.0,
"step": 175
},
{
"entropy": 8.06540780067444,
"epoch": 0.01632801161103048,
"grad_norm": 1.3984375,
"learning_rate": 8.95e-05,
"loss": 7.4771,
"mean_token_accuracy": 0.0847482830286026,
"num_tokens": 377918.0,
"step": 180
},
{
"entropy": 7.908505392074585,
"epoch": 0.01678156748911466,
"grad_norm": 1.3125,
"learning_rate": 9.2e-05,
"loss": 7.3165,
"mean_token_accuracy": 0.08902316763997078,
"num_tokens": 388461.0,
"step": 185
},
{
"entropy": 8.000589084625243,
"epoch": 0.01723512336719884,
"grad_norm": 1.3125,
"learning_rate": 9.45e-05,
"loss": 7.2895,
"mean_token_accuracy": 0.08787789791822434,
"num_tokens": 399771.0,
"step": 190
},
{
"entropy": 7.731726121902466,
"epoch": 0.01768867924528302,
"grad_norm": 1.234375,
"learning_rate": 9.7e-05,
"loss": 7.1673,
"mean_token_accuracy": 0.08963026329874993,
"num_tokens": 410254.0,
"step": 195
},
{
"entropy": 7.811504650115967,
"epoch": 0.018142235123367198,
"grad_norm": 1.390625,
"learning_rate": 9.95e-05,
"loss": 7.1894,
"mean_token_accuracy": 0.08560735136270523,
"num_tokens": 419792.0,
"step": 200
},
{
"entropy": 7.708149337768555,
"epoch": 0.018595791001451378,
"grad_norm": 1.46875,
"learning_rate": 0.000102,
"loss": 7.1962,
"mean_token_accuracy": 0.09234726577997207,
"num_tokens": 429264.0,
"step": 205
},
{
"entropy": 7.65235538482666,
"epoch": 0.019049346879535557,
"grad_norm": 1.4375,
"learning_rate": 0.00010449999999999999,
"loss": 7.1416,
"mean_token_accuracy": 0.09288287535309792,
"num_tokens": 439624.0,
"step": 210
},
{
"entropy": 7.664721393585205,
"epoch": 0.01950290275761974,
"grad_norm": 1.2578125,
"learning_rate": 0.000107,
"loss": 7.1224,
"mean_token_accuracy": 0.0913820892572403,
"num_tokens": 449556.0,
"step": 215
},
{
"entropy": 7.666817283630371,
"epoch": 0.01995645863570392,
"grad_norm": 1.53125,
"learning_rate": 0.0001095,
"loss": 7.1527,
"mean_token_accuracy": 0.09391891658306122,
"num_tokens": 460243.0,
"step": 220
},
{
"entropy": 7.6871044635772705,
"epoch": 0.0204100145137881,
"grad_norm": 1.3046875,
"learning_rate": 0.000112,
"loss": 7.193,
"mean_token_accuracy": 0.08998195976018905,
"num_tokens": 469767.0,
"step": 225
},
{
"entropy": 7.621262836456299,
"epoch": 0.02086357039187228,
"grad_norm": 1.3125,
"learning_rate": 0.0001145,
"loss": 7.1106,
"mean_token_accuracy": 0.09542692303657532,
"num_tokens": 480329.0,
"step": 230
},
{
"entropy": 7.601617050170899,
"epoch": 0.02131712626995646,
"grad_norm": 1.28125,
"learning_rate": 0.00011700000000000001,
"loss": 7.0818,
"mean_token_accuracy": 0.08891230374574662,
"num_tokens": 490096.0,
"step": 235
},
{
"entropy": 7.60865626335144,
"epoch": 0.02177068214804064,
"grad_norm": 1.3359375,
"learning_rate": 0.00011949999999999999,
"loss": 7.1143,
"mean_token_accuracy": 0.08756190165877342,
"num_tokens": 501231.0,
"step": 240
},
{
"entropy": 7.543486261367798,
"epoch": 0.022224238026124818,
"grad_norm": 1.6640625,
"learning_rate": 0.000122,
"loss": 7.0258,
"mean_token_accuracy": 0.09758798629045487,
"num_tokens": 510722.0,
"step": 245
},
{
"entropy": 7.489023685455322,
"epoch": 0.022677793904208998,
"grad_norm": 1.1796875,
"learning_rate": 0.0001245,
"loss": 7.0792,
"mean_token_accuracy": 0.09397116005420685,
"num_tokens": 521452.0,
"step": 250
},
{
"entropy": 7.515778923034668,
"epoch": 0.023131349782293177,
"grad_norm": 1.34375,
"learning_rate": 0.000127,
"loss": 7.0911,
"mean_token_accuracy": 0.09408443942666053,
"num_tokens": 532397.0,
"step": 255
},
{
"entropy": 7.523461151123047,
"epoch": 0.02358490566037736,
"grad_norm": 1.1328125,
"learning_rate": 0.0001295,
"loss": 7.0524,
"mean_token_accuracy": 0.09337728172540664,
"num_tokens": 542271.0,
"step": 260
},
{
"entropy": 7.498798656463623,
"epoch": 0.02403846153846154,
"grad_norm": 1.296875,
"learning_rate": 0.000132,
"loss": 7.0403,
"mean_token_accuracy": 0.09828322902321815,
"num_tokens": 552411.0,
"step": 265
},
{
"entropy": 7.433657121658325,
"epoch": 0.02449201741654572,
"grad_norm": 1.1171875,
"learning_rate": 0.00013450000000000002,
"loss": 7.0246,
"mean_token_accuracy": 0.09725753068923951,
"num_tokens": 561518.0,
"step": 270
},
{
"entropy": 7.546285247802734,
"epoch": 0.0249455732946299,
"grad_norm": 1.265625,
"learning_rate": 0.00013700000000000002,
"loss": 7.0652,
"mean_token_accuracy": 0.09419438466429711,
"num_tokens": 571361.0,
"step": 275
},
{
"entropy": 7.3379639148712155,
"epoch": 0.02539912917271408,
"grad_norm": 1.3515625,
"learning_rate": 0.0001395,
"loss": 6.9636,
"mean_token_accuracy": 0.10313116908073425,
"num_tokens": 580926.0,
"step": 280
},
{
"entropy": 7.439048814773559,
"epoch": 0.025852685050798258,
"grad_norm": 1.1875,
"learning_rate": 0.00014199999999999998,
"loss": 7.0476,
"mean_token_accuracy": 0.09056098833680153,
"num_tokens": 592282.0,
"step": 285
},
{
"entropy": 7.535349416732788,
"epoch": 0.026306240928882438,
"grad_norm": 1.2578125,
"learning_rate": 0.0001445,
"loss": 7.0665,
"mean_token_accuracy": 0.09440166875720024,
"num_tokens": 602767.0,
"step": 290
},
{
"entropy": 7.4240789890289305,
"epoch": 0.026759796806966617,
"grad_norm": 1.2109375,
"learning_rate": 0.000147,
"loss": 6.9896,
"mean_token_accuracy": 0.0924597330391407,
"num_tokens": 612763.0,
"step": 295
},
{
"entropy": 7.324904584884644,
"epoch": 0.027213352685050797,
"grad_norm": 1.046875,
"learning_rate": 0.0001495,
"loss": 6.9363,
"mean_token_accuracy": 0.09375968724489211,
"num_tokens": 622796.0,
"step": 300
},
{
"entropy": 7.385919713973999,
"epoch": 0.027666908563134977,
"grad_norm": 1.3203125,
"learning_rate": 0.000152,
"loss": 6.9192,
"mean_token_accuracy": 0.09953035265207291,
"num_tokens": 633136.0,
"step": 305
},
{
"entropy": 7.27361741065979,
"epoch": 0.02812046444121916,
"grad_norm": 1.1875,
"learning_rate": 0.00015450000000000001,
"loss": 6.9231,
"mean_token_accuracy": 0.10048090890049935,
"num_tokens": 643347.0,
"step": 310
},
{
"entropy": 7.3609613418579105,
"epoch": 0.02857402031930334,
"grad_norm": 1.28125,
"learning_rate": 0.000157,
"loss": 6.9037,
"mean_token_accuracy": 0.10072512328624725,
"num_tokens": 653096.0,
"step": 315
},
{
"entropy": 7.219988059997559,
"epoch": 0.02902757619738752,
"grad_norm": 1.2421875,
"learning_rate": 0.0001595,
"loss": 6.9318,
"mean_token_accuracy": 0.09513700902462005,
"num_tokens": 663557.0,
"step": 320
},
{
"entropy": 7.375028610229492,
"epoch": 0.0294811320754717,
"grad_norm": 1.4609375,
"learning_rate": 0.000162,
"loss": 7.0052,
"mean_token_accuracy": 0.0983717530965805,
"num_tokens": 674395.0,
"step": 325
},
{
"entropy": 7.270230674743653,
"epoch": 0.029934687953555878,
"grad_norm": 1.265625,
"learning_rate": 0.00016450000000000001,
"loss": 6.9129,
"mean_token_accuracy": 0.10161509737372398,
"num_tokens": 684776.0,
"step": 330
},
{
"entropy": 7.269373607635498,
"epoch": 0.030388243831640058,
"grad_norm": 1.1171875,
"learning_rate": 0.00016700000000000002,
"loss": 6.8875,
"mean_token_accuracy": 0.0997242234647274,
"num_tokens": 694886.0,
"step": 335
},
{
"entropy": 7.338870811462402,
"epoch": 0.030841799709724237,
"grad_norm": 1.1875,
"learning_rate": 0.00016950000000000003,
"loss": 6.9575,
"mean_token_accuracy": 0.09762076139450074,
"num_tokens": 705426.0,
"step": 340
},
{
"entropy": 7.253270244598388,
"epoch": 0.03129535558780842,
"grad_norm": 1.3203125,
"learning_rate": 0.00017199999999999998,
"loss": 6.9498,
"mean_token_accuracy": 0.09457632303237914,
"num_tokens": 716271.0,
"step": 345
},
{
"entropy": 7.2550302028656,
"epoch": 0.0317489114658926,
"grad_norm": 1.1171875,
"learning_rate": 0.00017449999999999999,
"loss": 6.9409,
"mean_token_accuracy": 0.09868695959448814,
"num_tokens": 726875.0,
"step": 350
},
{
"entropy": 7.2600914478302006,
"epoch": 0.032202467343976776,
"grad_norm": 1.0859375,
"learning_rate": 0.000177,
"loss": 6.8738,
"mean_token_accuracy": 0.09764649495482444,
"num_tokens": 737825.0,
"step": 355
},
{
"entropy": 7.240714120864868,
"epoch": 0.03265602322206096,
"grad_norm": 1.0703125,
"learning_rate": 0.0001795,
"loss": 6.8413,
"mean_token_accuracy": 0.10239819139242172,
"num_tokens": 747424.0,
"step": 360
},
{
"entropy": 7.195548105239868,
"epoch": 0.033109579100145135,
"grad_norm": 1.1640625,
"learning_rate": 0.000182,
"loss": 6.854,
"mean_token_accuracy": 0.10043720304965972,
"num_tokens": 757649.0,
"step": 365
},
{
"entropy": 7.267161417007446,
"epoch": 0.03356313497822932,
"grad_norm": 1.1953125,
"learning_rate": 0.0001845,
"loss": 6.8436,
"mean_token_accuracy": 0.09980886504054069,
"num_tokens": 768556.0,
"step": 370
},
{
"entropy": 7.0987019538879395,
"epoch": 0.034016690856313495,
"grad_norm": 1.234375,
"learning_rate": 0.000187,
"loss": 6.8373,
"mean_token_accuracy": 0.10219831764698029,
"num_tokens": 778331.0,
"step": 375
},
{
"entropy": 7.19783296585083,
"epoch": 0.03447024673439768,
"grad_norm": 1.265625,
"learning_rate": 0.0001895,
"loss": 6.8352,
"mean_token_accuracy": 0.09996062368154526,
"num_tokens": 788065.0,
"step": 380
},
{
"entropy": 7.185236835479737,
"epoch": 0.03492380261248186,
"grad_norm": 1.2265625,
"learning_rate": 0.000192,
"loss": 6.8695,
"mean_token_accuracy": 0.10359352231025695,
"num_tokens": 798912.0,
"step": 385
},
{
"entropy": 7.1850183486938475,
"epoch": 0.03537735849056604,
"grad_norm": 1.15625,
"learning_rate": 0.0001945,
"loss": 6.8211,
"mean_token_accuracy": 0.10563454255461693,
"num_tokens": 809580.0,
"step": 390
},
{
"entropy": 7.208040618896485,
"epoch": 0.03583091436865022,
"grad_norm": 1.15625,
"learning_rate": 0.00019700000000000002,
"loss": 6.8958,
"mean_token_accuracy": 0.10004983395338059,
"num_tokens": 820207.0,
"step": 395
},
{
"entropy": 7.000587606430054,
"epoch": 0.036284470246734396,
"grad_norm": 1.2421875,
"learning_rate": 0.00019950000000000002,
"loss": 6.7196,
"mean_token_accuracy": 0.09804629981517791,
"num_tokens": 831038.0,
"step": 400
},
{
"entropy": 7.166787433624267,
"epoch": 0.03673802612481858,
"grad_norm": 1.03125,
"learning_rate": 0.000202,
"loss": 6.8141,
"mean_token_accuracy": 0.10281971544027328,
"num_tokens": 842144.0,
"step": 405
},
{
"entropy": 7.101940298080445,
"epoch": 0.037191582002902755,
"grad_norm": 1.1328125,
"learning_rate": 0.00020449999999999998,
"loss": 6.8197,
"mean_token_accuracy": 0.10505499988794327,
"num_tokens": 851315.0,
"step": 410
},
{
"entropy": 7.090035247802734,
"epoch": 0.03764513788098694,
"grad_norm": 1.265625,
"learning_rate": 0.000207,
"loss": 6.8306,
"mean_token_accuracy": 0.10175477415323257,
"num_tokens": 861233.0,
"step": 415
},
{
"entropy": 7.051157379150391,
"epoch": 0.038098693759071114,
"grad_norm": 1.1484375,
"learning_rate": 0.0002095,
"loss": 6.8096,
"mean_token_accuracy": 0.0993909016251564,
"num_tokens": 871131.0,
"step": 420
},
{
"entropy": 7.143013191223145,
"epoch": 0.0385522496371553,
"grad_norm": 1.1015625,
"learning_rate": 0.000212,
"loss": 6.7764,
"mean_token_accuracy": 0.09792252779006957,
"num_tokens": 882195.0,
"step": 425
},
{
"entropy": 7.111673784255982,
"epoch": 0.03900580551523948,
"grad_norm": 1.203125,
"learning_rate": 0.0002145,
"loss": 6.761,
"mean_token_accuracy": 0.10355701595544815,
"num_tokens": 891806.0,
"step": 430
},
{
"entropy": 7.103079462051392,
"epoch": 0.03945936139332366,
"grad_norm": 1.1640625,
"learning_rate": 0.00021700000000000002,
"loss": 6.8234,
"mean_token_accuracy": 0.09959669709205628,
"num_tokens": 901350.0,
"step": 435
},
{
"entropy": 7.085621690750122,
"epoch": 0.03991291727140784,
"grad_norm": 1.4296875,
"learning_rate": 0.0002195,
"loss": 6.7362,
"mean_token_accuracy": 0.09833786115050316,
"num_tokens": 912061.0,
"step": 440
},
{
"entropy": 6.99485411643982,
"epoch": 0.040366473149492016,
"grad_norm": 1.1640625,
"learning_rate": 0.000222,
"loss": 6.6882,
"mean_token_accuracy": 0.1011364847421646,
"num_tokens": 922220.0,
"step": 445
},
{
"entropy": 7.05831937789917,
"epoch": 0.0408200290275762,
"grad_norm": 1.1875,
"learning_rate": 0.0002245,
"loss": 6.7282,
"mean_token_accuracy": 0.10195261687040329,
"num_tokens": 932319.0,
"step": 450
},
{
"entropy": 6.9827405452728275,
"epoch": 0.041273584905660375,
"grad_norm": 1.2421875,
"learning_rate": 0.00022700000000000002,
"loss": 6.7504,
"mean_token_accuracy": 0.10525996536016465,
"num_tokens": 943254.0,
"step": 455
},
{
"entropy": 7.01410927772522,
"epoch": 0.04172714078374456,
"grad_norm": 1.03125,
"learning_rate": 0.00022950000000000002,
"loss": 6.7792,
"mean_token_accuracy": 0.0980103112757206,
"num_tokens": 953921.0,
"step": 460
},
{
"entropy": 6.935913562774658,
"epoch": 0.042180696661828734,
"grad_norm": 1.1328125,
"learning_rate": 0.00023200000000000003,
"loss": 6.6898,
"mean_token_accuracy": 0.10519141629338265,
"num_tokens": 963832.0,
"step": 465
},
{
"entropy": 6.9887090682983395,
"epoch": 0.04263425253991292,
"grad_norm": 1.3671875,
"learning_rate": 0.00023449999999999998,
"loss": 6.7863,
"mean_token_accuracy": 0.0979742519557476,
"num_tokens": 974066.0,
"step": 470
},
{
"entropy": 7.043638467788696,
"epoch": 0.0430878084179971,
"grad_norm": 1.15625,
"learning_rate": 0.000237,
"loss": 6.7626,
"mean_token_accuracy": 0.10633939728140832,
"num_tokens": 984402.0,
"step": 475
},
{
"entropy": 6.972577047348023,
"epoch": 0.04354136429608128,
"grad_norm": 1.125,
"learning_rate": 0.0002395,
"loss": 6.6771,
"mean_token_accuracy": 0.10603873878717422,
"num_tokens": 993555.0,
"step": 480
},
{
"entropy": 6.969974994659424,
"epoch": 0.04399492017416546,
"grad_norm": 1.234375,
"learning_rate": 0.000242,
"loss": 6.6629,
"mean_token_accuracy": 0.11064062714576721,
"num_tokens": 1002920.0,
"step": 485
},
{
"entropy": 6.993032169342041,
"epoch": 0.044448476052249636,
"grad_norm": 1.1328125,
"learning_rate": 0.0002445,
"loss": 6.7527,
"mean_token_accuracy": 0.09811311662197113,
"num_tokens": 1013611.0,
"step": 490
},
{
"entropy": 6.870787668228149,
"epoch": 0.04490203193033382,
"grad_norm": 1.078125,
"learning_rate": 0.000247,
"loss": 6.6381,
"mean_token_accuracy": 0.102886413782835,
"num_tokens": 1024121.0,
"step": 495
},
{
"entropy": 7.048051881790161,
"epoch": 0.045355587808417995,
"grad_norm": 1.125,
"learning_rate": 0.0002495,
"loss": 6.8934,
"mean_token_accuracy": 0.1061844177544117,
"num_tokens": 1033845.0,
"step": 500
},
{
"entropy": 6.991887521743775,
"epoch": 0.04580914368650218,
"grad_norm": 1.0390625,
"learning_rate": 0.000252,
"loss": 6.6783,
"mean_token_accuracy": 0.1054874338209629,
"num_tokens": 1044642.0,
"step": 505
},
{
"entropy": 6.883953094482422,
"epoch": 0.046262699564586354,
"grad_norm": 1.2890625,
"learning_rate": 0.0002545,
"loss": 6.7359,
"mean_token_accuracy": 0.10576739087700844,
"num_tokens": 1054728.0,
"step": 510
},
{
"entropy": 6.952514743804931,
"epoch": 0.04671625544267054,
"grad_norm": 1.1015625,
"learning_rate": 0.000257,
"loss": 6.6919,
"mean_token_accuracy": 0.09888017326593398,
"num_tokens": 1065743.0,
"step": 515
},
{
"entropy": 6.847628164291382,
"epoch": 0.04716981132075472,
"grad_norm": 1.1796875,
"learning_rate": 0.0002595,
"loss": 6.6143,
"mean_token_accuracy": 0.10563460439443588,
"num_tokens": 1076798.0,
"step": 520
},
{
"entropy": 6.913150882720947,
"epoch": 0.047623367198838897,
"grad_norm": 1.15625,
"learning_rate": 0.000262,
"loss": 6.7491,
"mean_token_accuracy": 0.09963728263974189,
"num_tokens": 1087445.0,
"step": 525
},
{
"entropy": 6.850391578674317,
"epoch": 0.04807692307692308,
"grad_norm": 1.0703125,
"learning_rate": 0.00026450000000000003,
"loss": 6.6457,
"mean_token_accuracy": 0.10645358338952064,
"num_tokens": 1098379.0,
"step": 530
},
{
"entropy": 6.9510291576385494,
"epoch": 0.048530478955007256,
"grad_norm": 1.21875,
"learning_rate": 0.00026700000000000004,
"loss": 6.7504,
"mean_token_accuracy": 0.10485646203160286,
"num_tokens": 1108016.0,
"step": 535
},
{
"entropy": 6.889407110214234,
"epoch": 0.04898403483309144,
"grad_norm": 1.3046875,
"learning_rate": 0.00026950000000000005,
"loss": 6.6183,
"mean_token_accuracy": 0.10373573526740074,
"num_tokens": 1117887.0,
"step": 540
},
{
"entropy": 6.825150203704834,
"epoch": 0.049437590711175615,
"grad_norm": 1.171875,
"learning_rate": 0.00027200000000000005,
"loss": 6.6096,
"mean_token_accuracy": 0.10104240402579308,
"num_tokens": 1128497.0,
"step": 545
},
{
"entropy": 6.80326771736145,
"epoch": 0.0498911465892598,
"grad_norm": 1.1953125,
"learning_rate": 0.0002745,
"loss": 6.6592,
"mean_token_accuracy": 0.10856908708810806,
"num_tokens": 1138168.0,
"step": 550
},
{
"entropy": 6.879168224334717,
"epoch": 0.050344702467343974,
"grad_norm": 1.2578125,
"learning_rate": 0.000277,
"loss": 6.6407,
"mean_token_accuracy": 0.10094325393438339,
"num_tokens": 1148357.0,
"step": 555
},
{
"entropy": 6.832116413116455,
"epoch": 0.05079825834542816,
"grad_norm": 1.1640625,
"learning_rate": 0.0002795,
"loss": 6.6591,
"mean_token_accuracy": 0.10421488359570504,
"num_tokens": 1159964.0,
"step": 560
},
{
"entropy": 6.893824195861816,
"epoch": 0.05125181422351233,
"grad_norm": 1.1875,
"learning_rate": 0.00028199999999999997,
"loss": 6.7137,
"mean_token_accuracy": 0.1055032454431057,
"num_tokens": 1170683.0,
"step": 565
},
{
"entropy": 6.914015007019043,
"epoch": 0.051705370101596516,
"grad_norm": 1.2734375,
"learning_rate": 0.0002845,
"loss": 6.6792,
"mean_token_accuracy": 0.10148620307445526,
"num_tokens": 1181680.0,
"step": 570
},
{
"entropy": 6.74883451461792,
"epoch": 0.0521589259796807,
"grad_norm": 1.1328125,
"learning_rate": 0.000287,
"loss": 6.5654,
"mean_token_accuracy": 0.10732218623161316,
"num_tokens": 1191619.0,
"step": 575
},
{
"entropy": 6.685351276397705,
"epoch": 0.052612481857764876,
"grad_norm": 1.203125,
"learning_rate": 0.0002895,
"loss": 6.5629,
"mean_token_accuracy": 0.10640141144394874,
"num_tokens": 1202264.0,
"step": 580
},
{
"entropy": 6.805472993850708,
"epoch": 0.05306603773584906,
"grad_norm": 1.1875,
"learning_rate": 0.000292,
"loss": 6.5815,
"mean_token_accuracy": 0.1027822844684124,
"num_tokens": 1212119.0,
"step": 585
},
{
"entropy": 6.801087665557861,
"epoch": 0.053519593613933235,
"grad_norm": 1.171875,
"learning_rate": 0.0002945,
"loss": 6.7049,
"mean_token_accuracy": 0.10856684222817421,
"num_tokens": 1222317.0,
"step": 590
},
{
"entropy": 6.853984594345093,
"epoch": 0.05397314949201742,
"grad_norm": 1.21875,
"learning_rate": 0.000297,
"loss": 6.6443,
"mean_token_accuracy": 0.1087227538228035,
"num_tokens": 1232914.0,
"step": 595
},
{
"entropy": 6.680654525756836,
"epoch": 0.054426705370101594,
"grad_norm": 0.98046875,
"learning_rate": 0.0002995,
"loss": 6.487,
"mean_token_accuracy": 0.11319171860814095,
"num_tokens": 1244214.0,
"step": 600
},
{
"entropy": 6.830823040008545,
"epoch": 0.05488026124818578,
"grad_norm": 1.265625,
"learning_rate": 0.000302,
"loss": 6.5837,
"mean_token_accuracy": 0.10821337848901749,
"num_tokens": 1254626.0,
"step": 605
},
{
"entropy": 6.7139935970306395,
"epoch": 0.05533381712626995,
"grad_norm": 1.265625,
"learning_rate": 0.0003045,
"loss": 6.6937,
"mean_token_accuracy": 0.10711480379104614,
"num_tokens": 1265461.0,
"step": 610
},
{
"entropy": 6.8157871723175045,
"epoch": 0.055787373004354136,
"grad_norm": 1.21875,
"learning_rate": 0.000307,
"loss": 6.5595,
"mean_token_accuracy": 0.1111778050661087,
"num_tokens": 1275810.0,
"step": 615
},
{
"entropy": 6.6582403659820555,
"epoch": 0.05624092888243832,
"grad_norm": 1.359375,
"learning_rate": 0.0003095,
"loss": 6.5723,
"mean_token_accuracy": 0.11056382060050965,
"num_tokens": 1285170.0,
"step": 620
},
{
"entropy": 6.752780246734619,
"epoch": 0.056694484760522496,
"grad_norm": 1.03125,
"learning_rate": 0.000312,
"loss": 6.5104,
"mean_token_accuracy": 0.11242734342813492,
"num_tokens": 1295460.0,
"step": 625
},
{
"entropy": 6.783557653427124,
"epoch": 0.05714804063860668,
"grad_norm": 1.375,
"learning_rate": 0.0003145,
"loss": 6.647,
"mean_token_accuracy": 0.10393900722265244,
"num_tokens": 1306656.0,
"step": 630
},
{
"entropy": 6.771673727035522,
"epoch": 0.057601596516690855,
"grad_norm": 1.1015625,
"learning_rate": 0.000317,
"loss": 6.6099,
"mean_token_accuracy": 0.10909517779946327,
"num_tokens": 1318019.0,
"step": 635
},
{
"entropy": 6.627294826507568,
"epoch": 0.05805515239477504,
"grad_norm": 1.140625,
"learning_rate": 0.0003195,
"loss": 6.5314,
"mean_token_accuracy": 0.114180638641119,
"num_tokens": 1328120.0,
"step": 640
},
{
"entropy": 6.828495216369629,
"epoch": 0.058508708272859214,
"grad_norm": 1.15625,
"learning_rate": 0.000322,
"loss": 6.6283,
"mean_token_accuracy": 0.1077908605337143,
"num_tokens": 1338354.0,
"step": 645
},
{
"entropy": 6.780737257003784,
"epoch": 0.0589622641509434,
"grad_norm": 1.3515625,
"learning_rate": 0.00032450000000000003,
"loss": 6.6005,
"mean_token_accuracy": 0.10598717033863067,
"num_tokens": 1348666.0,
"step": 650
},
{
"entropy": 6.688767719268799,
"epoch": 0.05941582002902757,
"grad_norm": 1.09375,
"learning_rate": 0.00032700000000000003,
"loss": 6.6481,
"mean_token_accuracy": 0.11127779558300972,
"num_tokens": 1359713.0,
"step": 655
},
{
"entropy": 6.6816778659820555,
"epoch": 0.059869375907111756,
"grad_norm": 1.125,
"learning_rate": 0.00032950000000000004,
"loss": 6.4972,
"mean_token_accuracy": 0.11233635172247887,
"num_tokens": 1369507.0,
"step": 660
},
{
"entropy": 6.753318500518799,
"epoch": 0.06032293178519594,
"grad_norm": 1.15625,
"learning_rate": 0.00033200000000000005,
"loss": 6.6613,
"mean_token_accuracy": 0.10458081513643265,
"num_tokens": 1380726.0,
"step": 665
},
{
"entropy": 6.791505956649781,
"epoch": 0.060776487663280115,
"grad_norm": 1.0703125,
"learning_rate": 0.00033450000000000005,
"loss": 6.5884,
"mean_token_accuracy": 0.1085951142013073,
"num_tokens": 1391423.0,
"step": 670
},
{
"entropy": 6.758508014678955,
"epoch": 0.0612300435413643,
"grad_norm": 1.2734375,
"learning_rate": 0.000337,
"loss": 6.5566,
"mean_token_accuracy": 0.1129811353981495,
"num_tokens": 1402492.0,
"step": 675
},
{
"entropy": 6.64742956161499,
"epoch": 0.061683599419448475,
"grad_norm": 1.09375,
"learning_rate": 0.0003395,
"loss": 6.6151,
"mean_token_accuracy": 0.11265368685126305,
"num_tokens": 1412497.0,
"step": 680
},
{
"entropy": 6.80935697555542,
"epoch": 0.06213715529753266,
"grad_norm": 1.203125,
"learning_rate": 0.000342,
"loss": 6.5751,
"mean_token_accuracy": 0.11512991040945053,
"num_tokens": 1423214.0,
"step": 685
},
{
"entropy": 6.651473712921143,
"epoch": 0.06259071117561683,
"grad_norm": 1.4140625,
"learning_rate": 0.00034449999999999997,
"loss": 6.5673,
"mean_token_accuracy": 0.1103017933666706,
"num_tokens": 1434044.0,
"step": 690
},
{
"entropy": 6.6841058254241945,
"epoch": 0.06304426705370102,
"grad_norm": 1.1015625,
"learning_rate": 0.000347,
"loss": 6.5303,
"mean_token_accuracy": 0.1138860672712326,
"num_tokens": 1444387.0,
"step": 695
},
{
"entropy": 6.592344141006469,
"epoch": 0.0634978229317852,
"grad_norm": 1.1640625,
"learning_rate": 0.0003495,
"loss": 6.4809,
"mean_token_accuracy": 0.10669072344899178,
"num_tokens": 1455157.0,
"step": 700
},
{
"entropy": 6.729357099533081,
"epoch": 0.06395137880986937,
"grad_norm": 1.1328125,
"learning_rate": 0.000352,
"loss": 6.595,
"mean_token_accuracy": 0.11270184963941574,
"num_tokens": 1466145.0,
"step": 705
},
{
"entropy": 6.65928430557251,
"epoch": 0.06440493468795355,
"grad_norm": 1.03125,
"learning_rate": 0.0003545,
"loss": 6.5818,
"mean_token_accuracy": 0.11218819990754128,
"num_tokens": 1476713.0,
"step": 710
},
{
"entropy": 6.686679744720459,
"epoch": 0.06485849056603774,
"grad_norm": 1.09375,
"learning_rate": 0.000357,
"loss": 6.581,
"mean_token_accuracy": 0.11432325467467308,
"num_tokens": 1486908.0,
"step": 715
},
{
"entropy": 6.704012012481689,
"epoch": 0.06531204644412192,
"grad_norm": 0.96875,
"learning_rate": 0.0003595,
"loss": 6.5049,
"mean_token_accuracy": 0.1139966145157814,
"num_tokens": 1498185.0,
"step": 720
},
{
"entropy": 6.605288028717041,
"epoch": 0.0657656023222061,
"grad_norm": 1.1171875,
"learning_rate": 0.000362,
"loss": 6.5477,
"mean_token_accuracy": 0.11151361390948296,
"num_tokens": 1508376.0,
"step": 725
},
{
"entropy": 6.708197355270386,
"epoch": 0.06621915820029027,
"grad_norm": 1.0625,
"learning_rate": 0.0003645,
"loss": 6.6343,
"mean_token_accuracy": 0.1107229582965374,
"num_tokens": 1519953.0,
"step": 730
},
{
"entropy": 6.5456921577453615,
"epoch": 0.06667271407837445,
"grad_norm": 1.046875,
"learning_rate": 0.000367,
"loss": 6.4996,
"mean_token_accuracy": 0.11313785165548325,
"num_tokens": 1529793.0,
"step": 735
},
{
"entropy": 6.6532182693481445,
"epoch": 0.06712626995645864,
"grad_norm": 1.015625,
"learning_rate": 0.0003695,
"loss": 6.4993,
"mean_token_accuracy": 0.11373248919844628,
"num_tokens": 1540859.0,
"step": 740
},
{
"entropy": 6.61553168296814,
"epoch": 0.06757982583454282,
"grad_norm": 1.0859375,
"learning_rate": 0.000372,
"loss": 6.4324,
"mean_token_accuracy": 0.1168427474796772,
"num_tokens": 1549783.0,
"step": 745
},
{
"entropy": 6.514708757400513,
"epoch": 0.06803338171262699,
"grad_norm": 1.1015625,
"learning_rate": 0.0003745,
"loss": 6.4901,
"mean_token_accuracy": 0.11619689613580704,
"num_tokens": 1560522.0,
"step": 750
},
{
"entropy": 6.587539339065552,
"epoch": 0.06848693759071117,
"grad_norm": 1.125,
"learning_rate": 0.000377,
"loss": 6.4759,
"mean_token_accuracy": 0.11305172964930535,
"num_tokens": 1570668.0,
"step": 755
},
{
"entropy": 6.66793270111084,
"epoch": 0.06894049346879536,
"grad_norm": 1.09375,
"learning_rate": 0.0003795,
"loss": 6.5919,
"mean_token_accuracy": 0.11405429989099503,
"num_tokens": 1581523.0,
"step": 760
},
{
"entropy": 6.56304898262024,
"epoch": 0.06939404934687954,
"grad_norm": 1.171875,
"learning_rate": 0.000382,
"loss": 6.4933,
"mean_token_accuracy": 0.11518137082457543,
"num_tokens": 1591356.0,
"step": 765
},
{
"entropy": 6.5586266040802,
"epoch": 0.06984760522496372,
"grad_norm": 1.1171875,
"learning_rate": 0.0003845,
"loss": 6.4862,
"mean_token_accuracy": 0.11520140692591667,
"num_tokens": 1601779.0,
"step": 770
},
{
"entropy": 6.645415878295898,
"epoch": 0.07030116110304789,
"grad_norm": 1.125,
"learning_rate": 0.00038700000000000003,
"loss": 6.4799,
"mean_token_accuracy": 0.11675522029399872,
"num_tokens": 1611995.0,
"step": 775
},
{
"entropy": 6.643508338928223,
"epoch": 0.07075471698113207,
"grad_norm": 1.0546875,
"learning_rate": 0.00038950000000000003,
"loss": 6.51,
"mean_token_accuracy": 0.11030993685126304,
"num_tokens": 1621834.0,
"step": 780
},
{
"entropy": 6.5690505027771,
"epoch": 0.07120827285921626,
"grad_norm": 1.28125,
"learning_rate": 0.00039200000000000004,
"loss": 6.5502,
"mean_token_accuracy": 0.10731691345572472,
"num_tokens": 1632881.0,
"step": 785
},
{
"entropy": 6.497133684158325,
"epoch": 0.07166182873730044,
"grad_norm": 1.1875,
"learning_rate": 0.00039450000000000005,
"loss": 6.5383,
"mean_token_accuracy": 0.11029602736234664,
"num_tokens": 1642870.0,
"step": 790
},
{
"entropy": 6.5692647933959964,
"epoch": 0.07211538461538461,
"grad_norm": 1.21875,
"learning_rate": 0.00039700000000000005,
"loss": 6.4224,
"mean_token_accuracy": 0.12195004150271416,
"num_tokens": 1653390.0,
"step": 795
},
{
"entropy": 6.5903912544250485,
"epoch": 0.07256894049346879,
"grad_norm": 1.109375,
"learning_rate": 0.0003995,
"loss": 6.5359,
"mean_token_accuracy": 0.11937625855207443,
"num_tokens": 1663382.0,
"step": 800
},
{
"entropy": 6.648660278320312,
"epoch": 0.07302249637155298,
"grad_norm": 1.0625,
"learning_rate": 0.000402,
"loss": 6.5463,
"mean_token_accuracy": 0.11221321672201157,
"num_tokens": 1673601.0,
"step": 805
},
{
"entropy": 6.414686155319214,
"epoch": 0.07347605224963716,
"grad_norm": 1.328125,
"learning_rate": 0.0004045,
"loss": 6.4486,
"mean_token_accuracy": 0.11948997676372528,
"num_tokens": 1683565.0,
"step": 810
},
{
"entropy": 6.576791286468506,
"epoch": 0.07392960812772134,
"grad_norm": 1.9609375,
"learning_rate": 0.00040699999999999997,
"loss": 6.4393,
"mean_token_accuracy": 0.11694891527295112,
"num_tokens": 1693636.0,
"step": 815
},
{
"entropy": 6.726799058914184,
"epoch": 0.07438316400580551,
"grad_norm": 1.1953125,
"learning_rate": 0.0004095,
"loss": 6.543,
"mean_token_accuracy": 0.1100502498447895,
"num_tokens": 1703415.0,
"step": 820
},
{
"entropy": 6.53825044631958,
"epoch": 0.0748367198838897,
"grad_norm": 1.265625,
"learning_rate": 0.000412,
"loss": 6.5077,
"mean_token_accuracy": 0.1171564668416977,
"num_tokens": 1713579.0,
"step": 825
},
{
"entropy": 6.537848711013794,
"epoch": 0.07529027576197388,
"grad_norm": 1.09375,
"learning_rate": 0.0004145,
"loss": 6.5234,
"mean_token_accuracy": 0.11068456545472145,
"num_tokens": 1724583.0,
"step": 830
},
{
"entropy": 6.56220064163208,
"epoch": 0.07574383164005806,
"grad_norm": 1.1796875,
"learning_rate": 0.000417,
"loss": 6.4309,
"mean_token_accuracy": 0.12289080545306205,
"num_tokens": 1735293.0,
"step": 835
},
{
"entropy": 6.558898782730102,
"epoch": 0.07619738751814223,
"grad_norm": 1.03125,
"learning_rate": 0.0004195,
"loss": 6.4543,
"mean_token_accuracy": 0.11800618395209313,
"num_tokens": 1745623.0,
"step": 840
},
{
"entropy": 6.5176109790802,
"epoch": 0.07665094339622641,
"grad_norm": 1.03125,
"learning_rate": 0.000422,
"loss": 6.4841,
"mean_token_accuracy": 0.11565901786088943,
"num_tokens": 1755713.0,
"step": 845
},
{
"entropy": 6.523326635360718,
"epoch": 0.0771044992743106,
"grad_norm": 1.1796875,
"learning_rate": 0.0004245,
"loss": 6.4853,
"mean_token_accuracy": 0.11440500319004059,
"num_tokens": 1765264.0,
"step": 850
},
{
"entropy": 6.563967895507813,
"epoch": 0.07755805515239478,
"grad_norm": 1.21875,
"learning_rate": 0.000427,
"loss": 6.5002,
"mean_token_accuracy": 0.11388131231069565,
"num_tokens": 1775769.0,
"step": 855
},
{
"entropy": 6.483576726913452,
"epoch": 0.07801161103047896,
"grad_norm": 1.109375,
"learning_rate": 0.0004295,
"loss": 6.4249,
"mean_token_accuracy": 0.11765370443463326,
"num_tokens": 1786245.0,
"step": 860
},
{
"entropy": 6.497152996063233,
"epoch": 0.07846516690856313,
"grad_norm": 1.1796875,
"learning_rate": 0.000432,
"loss": 6.4549,
"mean_token_accuracy": 0.12084143534302712,
"num_tokens": 1796040.0,
"step": 865
},
{
"entropy": 6.451007413864136,
"epoch": 0.07891872278664731,
"grad_norm": 1.0390625,
"learning_rate": 0.0004345,
"loss": 6.4846,
"mean_token_accuracy": 0.11668049693107604,
"num_tokens": 1806007.0,
"step": 870
},
{
"entropy": 6.6562886238098145,
"epoch": 0.0793722786647315,
"grad_norm": 1.1953125,
"learning_rate": 0.000437,
"loss": 6.5641,
"mean_token_accuracy": 0.10741591677069665,
"num_tokens": 1816935.0,
"step": 875
},
{
"entropy": 6.411975765228272,
"epoch": 0.07982583454281568,
"grad_norm": 1.171875,
"learning_rate": 0.0004395,
"loss": 6.4055,
"mean_token_accuracy": 0.11737755388021469,
"num_tokens": 1827419.0,
"step": 880
},
{
"entropy": 6.580957078933716,
"epoch": 0.08027939042089985,
"grad_norm": 1.234375,
"learning_rate": 0.000442,
"loss": 6.5167,
"mean_token_accuracy": 0.11306339800357819,
"num_tokens": 1837933.0,
"step": 885
},
{
"entropy": 6.480378532409668,
"epoch": 0.08073294629898403,
"grad_norm": 1.09375,
"learning_rate": 0.0004445,
"loss": 6.5671,
"mean_token_accuracy": 0.11678754836320877,
"num_tokens": 1848551.0,
"step": 890
},
{
"entropy": 6.546552562713623,
"epoch": 0.08118650217706821,
"grad_norm": 1.21875,
"learning_rate": 0.000447,
"loss": 6.4439,
"mean_token_accuracy": 0.11490961909294128,
"num_tokens": 1858380.0,
"step": 895
},
{
"entropy": 6.448764801025391,
"epoch": 0.0816400580551524,
"grad_norm": 1.1015625,
"learning_rate": 0.00044950000000000003,
"loss": 6.3688,
"mean_token_accuracy": 0.11925693526864052,
"num_tokens": 1868291.0,
"step": 900
},
{
"entropy": 6.491116046905518,
"epoch": 0.08209361393323658,
"grad_norm": 1.03125,
"learning_rate": 0.00045200000000000004,
"loss": 6.4651,
"mean_token_accuracy": 0.11819546893239022,
"num_tokens": 1879219.0,
"step": 905
},
{
"entropy": 6.526374053955078,
"epoch": 0.08254716981132075,
"grad_norm": 1.0078125,
"learning_rate": 0.00045450000000000004,
"loss": 6.4579,
"mean_token_accuracy": 0.1181636244058609,
"num_tokens": 1889809.0,
"step": 910
},
{
"entropy": 6.414980411529541,
"epoch": 0.08300072568940493,
"grad_norm": 1.15625,
"learning_rate": 0.00045700000000000005,
"loss": 6.431,
"mean_token_accuracy": 0.11433882117271424,
"num_tokens": 1899986.0,
"step": 915
},
{
"entropy": 6.483574247360229,
"epoch": 0.08345428156748912,
"grad_norm": 1.09375,
"learning_rate": 0.00045950000000000006,
"loss": 6.4028,
"mean_token_accuracy": 0.11887709498405456,
"num_tokens": 1909314.0,
"step": 920
},
{
"entropy": 6.4847499370574955,
"epoch": 0.0839078374455733,
"grad_norm": 1.078125,
"learning_rate": 0.000462,
"loss": 6.4628,
"mean_token_accuracy": 0.11896910145878792,
"num_tokens": 1918443.0,
"step": 925
},
{
"entropy": 6.4445396900177006,
"epoch": 0.08436139332365747,
"grad_norm": 1.28125,
"learning_rate": 0.0004645,
"loss": 6.4045,
"mean_token_accuracy": 0.12024879306554795,
"num_tokens": 1928683.0,
"step": 930
},
{
"entropy": 6.317193126678466,
"epoch": 0.08481494920174165,
"grad_norm": 1.140625,
"learning_rate": 0.000467,
"loss": 6.3318,
"mean_token_accuracy": 0.1153000108897686,
"num_tokens": 1938293.0,
"step": 935
},
{
"entropy": 6.431926250457764,
"epoch": 0.08526850507982583,
"grad_norm": 1.0546875,
"learning_rate": 0.0004695,
"loss": 6.4353,
"mean_token_accuracy": 0.11641831621527672,
"num_tokens": 1948233.0,
"step": 940
},
{
"entropy": 6.480042028427124,
"epoch": 0.08572206095791002,
"grad_norm": 1.109375,
"learning_rate": 0.000472,
"loss": 6.3302,
"mean_token_accuracy": 0.12561678364872933,
"num_tokens": 1958611.0,
"step": 945
},
{
"entropy": 6.443802785873413,
"epoch": 0.0861756168359942,
"grad_norm": 1.015625,
"learning_rate": 0.0004745,
"loss": 6.4923,
"mean_token_accuracy": 0.12038066908717156,
"num_tokens": 1968957.0,
"step": 950
},
{
"entropy": 6.41972770690918,
"epoch": 0.08662917271407837,
"grad_norm": 1.2265625,
"learning_rate": 0.000477,
"loss": 6.3452,
"mean_token_accuracy": 0.12305001318454742,
"num_tokens": 1978503.0,
"step": 955
},
{
"entropy": 6.331937170028686,
"epoch": 0.08708272859216255,
"grad_norm": 1.1328125,
"learning_rate": 0.0004795,
"loss": 6.3407,
"mean_token_accuracy": 0.12022828236222267,
"num_tokens": 1988458.0,
"step": 960
},
{
"entropy": 6.463285970687866,
"epoch": 0.08753628447024674,
"grad_norm": 1.046875,
"learning_rate": 0.000482,
"loss": 6.4719,
"mean_token_accuracy": 0.11548013314604759,
"num_tokens": 1999984.0,
"step": 965
},
{
"entropy": 6.286208868026733,
"epoch": 0.08798984034833092,
"grad_norm": 1.0625,
"learning_rate": 0.0004845,
"loss": 6.3288,
"mean_token_accuracy": 0.12638000547885894,
"num_tokens": 2010645.0,
"step": 970
},
{
"entropy": 6.497215414047242,
"epoch": 0.08844339622641509,
"grad_norm": 1.109375,
"learning_rate": 0.000487,
"loss": 6.4547,
"mean_token_accuracy": 0.11661298498511315,
"num_tokens": 2020196.0,
"step": 975
},
{
"entropy": 6.386039924621582,
"epoch": 0.08889695210449927,
"grad_norm": 1.109375,
"learning_rate": 0.0004895,
"loss": 6.347,
"mean_token_accuracy": 0.11794270873069763,
"num_tokens": 2030101.0,
"step": 980
},
{
"entropy": 6.388373708724975,
"epoch": 0.08935050798258345,
"grad_norm": 1.1328125,
"learning_rate": 0.000492,
"loss": 6.4225,
"mean_token_accuracy": 0.11813583448529244,
"num_tokens": 2040454.0,
"step": 985
},
{
"entropy": 6.357129955291748,
"epoch": 0.08980406386066764,
"grad_norm": 1.03125,
"learning_rate": 0.0004945,
"loss": 6.3631,
"mean_token_accuracy": 0.11890257149934769,
"num_tokens": 2050378.0,
"step": 990
},
{
"entropy": 6.4591563701629635,
"epoch": 0.09025761973875182,
"grad_norm": 1.0625,
"learning_rate": 0.000497,
"loss": 6.4206,
"mean_token_accuracy": 0.11803730875253678,
"num_tokens": 2060282.0,
"step": 995
},
{
"entropy": 6.331292963027954,
"epoch": 0.09071117561683599,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995,
"loss": 6.4023,
"mean_token_accuracy": 0.11980631649494171,
"num_tokens": 2070563.0,
"step": 1000
},
{
"entropy": 6.491535806655884,
"epoch": 0.09116473149492017,
"grad_norm": 1.078125,
"learning_rate": 0.0004999999985110217,
"loss": 6.4912,
"mean_token_accuracy": 0.11830355301499366,
"num_tokens": 2080153.0,
"step": 1005
},
{
"entropy": 6.275622034072876,
"epoch": 0.09161828737300436,
"grad_norm": 1.0625,
"learning_rate": 0.0004999999924620471,
"loss": 6.3284,
"mean_token_accuracy": 0.1242703415453434,
"num_tokens": 2090590.0,
"step": 1010
},
{
"entropy": 6.441268587112427,
"epoch": 0.09207184325108854,
"grad_norm": 0.96875,
"learning_rate": 0.0004999999817600155,
"loss": 6.4116,
"mean_token_accuracy": 0.11299227103590966,
"num_tokens": 2101945.0,
"step": 1015
},
{
"entropy": 6.446867227554321,
"epoch": 0.09252539912917271,
"grad_norm": 0.9375,
"learning_rate": 0.0004999999664049268,
"loss": 6.394,
"mean_token_accuracy": 0.12263479754328728,
"num_tokens": 2112514.0,
"step": 1020
},
{
"entropy": 6.262259101867675,
"epoch": 0.09297895500725689,
"grad_norm": 1.09375,
"learning_rate": 0.0004999999463967816,
"loss": 6.3454,
"mean_token_accuracy": 0.12117168605327606,
"num_tokens": 2122296.0,
"step": 1025
},
{
"entropy": 6.523186826705933,
"epoch": 0.09343251088534107,
"grad_norm": 0.96484375,
"learning_rate": 0.00049999992173558,
"loss": 6.4913,
"mean_token_accuracy": 0.10856073573231698,
"num_tokens": 2134512.0,
"step": 1030
},
{
"entropy": 6.369787693023682,
"epoch": 0.09388606676342526,
"grad_norm": 1.15625,
"learning_rate": 0.0004999998924213228,
"loss": 6.421,
"mean_token_accuracy": 0.1210160993039608,
"num_tokens": 2143934.0,
"step": 1035
},
{
"entropy": 6.265434217453003,
"epoch": 0.09433962264150944,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999998584540105,
"loss": 6.298,
"mean_token_accuracy": 0.1182013526558876,
"num_tokens": 2155730.0,
"step": 1040
},
{
"entropy": 6.436200380325317,
"epoch": 0.09479317851959361,
"grad_norm": 1.046875,
"learning_rate": 0.0004999998198336437,
"loss": 6.323,
"mean_token_accuracy": 0.1117444783449173,
"num_tokens": 2166422.0,
"step": 1045
},
{
"entropy": 6.347907590866089,
"epoch": 0.09524673439767779,
"grad_norm": 1.03125,
"learning_rate": 0.0004999997765602233,
"loss": 6.4449,
"mean_token_accuracy": 0.1139302246272564,
"num_tokens": 2177842.0,
"step": 1050
},
{
"entropy": 6.457770681381225,
"epoch": 0.09570029027576198,
"grad_norm": 1.015625,
"learning_rate": 0.0004999997286337502,
"loss": 6.504,
"mean_token_accuracy": 0.11967741474509239,
"num_tokens": 2188614.0,
"step": 1055
},
{
"entropy": 6.377467250823974,
"epoch": 0.09615384615384616,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999996760542254,
"loss": 6.2878,
"mean_token_accuracy": 0.12623362839221955,
"num_tokens": 2197886.0,
"step": 1060
},
{
"entropy": 6.229464912414551,
"epoch": 0.09660740203193033,
"grad_norm": 1.0546875,
"learning_rate": 0.00049999961882165,
"loss": 6.3599,
"mean_token_accuracy": 0.12299013957381248,
"num_tokens": 2209930.0,
"step": 1065
},
{
"entropy": 6.440615034103393,
"epoch": 0.09706095791001451,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999995569360251,
"loss": 6.3759,
"mean_token_accuracy": 0.11413345336914063,
"num_tokens": 2219767.0,
"step": 1070
},
{
"entropy": 6.328685569763183,
"epoch": 0.0975145137880987,
"grad_norm": 1.046875,
"learning_rate": 0.000499999490397352,
"loss": 6.3049,
"mean_token_accuracy": 0.1205204777419567,
"num_tokens": 2230082.0,
"step": 1075
},
{
"entropy": 6.243273115158081,
"epoch": 0.09796806966618288,
"grad_norm": 1.046875,
"learning_rate": 0.0004999994192056321,
"loss": 6.3255,
"mean_token_accuracy": 0.12633560076355935,
"num_tokens": 2240066.0,
"step": 1080
},
{
"entropy": 6.374665594100952,
"epoch": 0.09842162554426705,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999993433608669,
"loss": 6.3561,
"mean_token_accuracy": 0.1187953770160675,
"num_tokens": 2250732.0,
"step": 1085
},
{
"entropy": 6.321237421035766,
"epoch": 0.09887518142235123,
"grad_norm": 1.0625,
"learning_rate": 0.0004999992628630579,
"loss": 6.2454,
"mean_token_accuracy": 0.11912132352590561,
"num_tokens": 2261452.0,
"step": 1090
},
{
"entropy": 6.250462293624878,
"epoch": 0.09932873730043541,
"grad_norm": 1.03125,
"learning_rate": 0.0004999991777122069,
"loss": 6.3636,
"mean_token_accuracy": 0.12190868705511093,
"num_tokens": 2271929.0,
"step": 1095
},
{
"entropy": 6.321765327453614,
"epoch": 0.0997822931785196,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999990879083156,
"loss": 6.3264,
"mean_token_accuracy": 0.12127640098333359,
"num_tokens": 2282987.0,
"step": 1100
},
{
"entropy": 6.319233846664429,
"epoch": 0.10023584905660378,
"grad_norm": 1.15625,
"learning_rate": 0.0004999989934513857,
"loss": 6.3136,
"mean_token_accuracy": 0.1254054971039295,
"num_tokens": 2292796.0,
"step": 1105
},
{
"entropy": 6.332448720932007,
"epoch": 0.10068940493468795,
"grad_norm": 1.078125,
"learning_rate": 0.0004999988943414193,
"loss": 6.3567,
"mean_token_accuracy": 0.12186149284243583,
"num_tokens": 2303057.0,
"step": 1110
},
{
"entropy": 6.252414035797119,
"epoch": 0.10114296081277213,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999987905784184,
"loss": 6.3422,
"mean_token_accuracy": 0.12239331230521203,
"num_tokens": 2314168.0,
"step": 1115
},
{
"entropy": 6.341477918624878,
"epoch": 0.10159651669085631,
"grad_norm": 1.015625,
"learning_rate": 0.0004999986821623853,
"loss": 6.3088,
"mean_token_accuracy": 0.1192399226129055,
"num_tokens": 2325241.0,
"step": 1120
},
{
"entropy": 6.351327323913575,
"epoch": 0.1020500725689405,
"grad_norm": 1.1875,
"learning_rate": 0.0004999985690933219,
"loss": 6.3276,
"mean_token_accuracy": 0.12044254839420318,
"num_tokens": 2334866.0,
"step": 1125
},
{
"entropy": 6.211726427078247,
"epoch": 0.10250362844702467,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999984513712311,
"loss": 6.3425,
"mean_token_accuracy": 0.11830557957291603,
"num_tokens": 2345422.0,
"step": 1130
},
{
"entropy": 6.417554903030395,
"epoch": 0.10295718432510885,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999983289961146,
"loss": 6.349,
"mean_token_accuracy": 0.11906943917274475,
"num_tokens": 2356285.0,
"step": 1135
},
{
"entropy": 6.214526510238647,
"epoch": 0.10341074020319303,
"grad_norm": 1.140625,
"learning_rate": 0.0004999982019679755,
"loss": 6.2793,
"mean_token_accuracy": 0.12734896019101144,
"num_tokens": 2366861.0,
"step": 1140
},
{
"entropy": 6.289835214614868,
"epoch": 0.10386429608127722,
"grad_norm": 1.03125,
"learning_rate": 0.0004999980702868164,
"loss": 6.2976,
"mean_token_accuracy": 0.12919359803199768,
"num_tokens": 2376851.0,
"step": 1145
},
{
"entropy": 6.3659947395324705,
"epoch": 0.1043178519593614,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999979339526396,
"loss": 6.3591,
"mean_token_accuracy": 0.12571399062871932,
"num_tokens": 2386857.0,
"step": 1150
},
{
"entropy": 6.303523445129395,
"epoch": 0.10477140783744557,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999977929654484,
"loss": 6.268,
"mean_token_accuracy": 0.12874414026737213,
"num_tokens": 2397035.0,
"step": 1155
},
{
"entropy": 6.3552806854248045,
"epoch": 0.10522496371552975,
"grad_norm": 1.21875,
"learning_rate": 0.0004999976473252453,
"loss": 6.3744,
"mean_token_accuracy": 0.12097596004605293,
"num_tokens": 2408177.0,
"step": 1160
},
{
"entropy": 6.359548139572143,
"epoch": 0.10567851959361393,
"grad_norm": 1.015625,
"learning_rate": 0.0004999974970320338,
"loss": 6.3076,
"mean_token_accuracy": 0.127739118039608,
"num_tokens": 2418445.0,
"step": 1165
},
{
"entropy": 6.25285906791687,
"epoch": 0.10613207547169812,
"grad_norm": 1.109375,
"learning_rate": 0.0004999973420858165,
"loss": 6.3116,
"mean_token_accuracy": 0.12409163489937783,
"num_tokens": 2428990.0,
"step": 1170
},
{
"entropy": 6.301597690582275,
"epoch": 0.10658563134978229,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999971824865968,
"loss": 6.3509,
"mean_token_accuracy": 0.11999165937304497,
"num_tokens": 2439643.0,
"step": 1175
},
{
"entropy": 6.453162050247192,
"epoch": 0.10703918722786647,
"grad_norm": 1.015625,
"learning_rate": 0.0004999970182343782,
"loss": 6.3972,
"mean_token_accuracy": 0.12346342280507087,
"num_tokens": 2450691.0,
"step": 1180
},
{
"entropy": 6.286817169189453,
"epoch": 0.10749274310595065,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999968493291638,
"loss": 6.3281,
"mean_token_accuracy": 0.12339968159794808,
"num_tokens": 2460706.0,
"step": 1185
},
{
"entropy": 6.101443719863892,
"epoch": 0.10794629898403484,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999966757709573,
"loss": 6.223,
"mean_token_accuracy": 0.12936578840017318,
"num_tokens": 2470947.0,
"step": 1190
},
{
"entropy": 6.308624172210694,
"epoch": 0.10839985486211902,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999964975597622,
"loss": 6.1877,
"mean_token_accuracy": 0.12678698301315308,
"num_tokens": 2481132.0,
"step": 1195
},
{
"entropy": 6.23999056816101,
"epoch": 0.10885341074020319,
"grad_norm": 1.078125,
"learning_rate": 0.0004999963146955821,
"loss": 6.2221,
"mean_token_accuracy": 0.13168013989925384,
"num_tokens": 2491701.0,
"step": 1200
},
{
"entropy": 6.209512996673584,
"epoch": 0.10930696661828737,
"grad_norm": 0.98046875,
"learning_rate": 0.000499996127178421,
"loss": 6.3154,
"mean_token_accuracy": 0.12373346984386444,
"num_tokens": 2502360.0,
"step": 1205
},
{
"entropy": 6.259164810180664,
"epoch": 0.10976052249637155,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999959350082825,
"loss": 6.1911,
"mean_token_accuracy": 0.12604912146925926,
"num_tokens": 2512140.0,
"step": 1210
},
{
"entropy": 6.277319002151489,
"epoch": 0.11021407837445574,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999957381851709,
"loss": 6.3226,
"mean_token_accuracy": 0.11895870566368102,
"num_tokens": 2522721.0,
"step": 1215
},
{
"entropy": 6.195718097686767,
"epoch": 0.1106676342525399,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999955367090901,
"loss": 6.1883,
"mean_token_accuracy": 0.12740988954901694,
"num_tokens": 2531786.0,
"step": 1220
},
{
"entropy": 6.27605619430542,
"epoch": 0.11112119013062409,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999953305800441,
"loss": 6.3046,
"mean_token_accuracy": 0.1145807795226574,
"num_tokens": 2542895.0,
"step": 1225
},
{
"entropy": 6.222184085845948,
"epoch": 0.11157474600870827,
"grad_norm": 1.046875,
"learning_rate": 0.0004999951197980374,
"loss": 6.2913,
"mean_token_accuracy": 0.12083523273468018,
"num_tokens": 2554120.0,
"step": 1230
},
{
"entropy": 6.31572117805481,
"epoch": 0.11202830188679246,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999949043630744,
"loss": 6.3231,
"mean_token_accuracy": 0.12528388127684592,
"num_tokens": 2563935.0,
"step": 1235
},
{
"entropy": 6.163136672973633,
"epoch": 0.11248185776487664,
"grad_norm": 1.140625,
"learning_rate": 0.0004999946842751593,
"loss": 6.2399,
"mean_token_accuracy": 0.12615554332733153,
"num_tokens": 2574465.0,
"step": 1240
},
{
"entropy": 6.261920547485351,
"epoch": 0.11293541364296081,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999944595342968,
"loss": 6.1907,
"mean_token_accuracy": 0.12804502844810486,
"num_tokens": 2584694.0,
"step": 1245
},
{
"entropy": 6.215864896774292,
"epoch": 0.11338896952104499,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999942301404916,
"loss": 6.165,
"mean_token_accuracy": 0.1279282294213772,
"num_tokens": 2594540.0,
"step": 1250
},
{
"entropy": 6.151381015777588,
"epoch": 0.11384252539912917,
"grad_norm": 1.078125,
"learning_rate": 0.0004999939960937484,
"loss": 6.0988,
"mean_token_accuracy": 0.1311326391994953,
"num_tokens": 2604896.0,
"step": 1255
},
{
"entropy": 6.2653666019439695,
"epoch": 0.11429608127721336,
"grad_norm": 1.09375,
"learning_rate": 0.0004999937573940721,
"loss": 6.2806,
"mean_token_accuracy": 0.13130278512835503,
"num_tokens": 2614143.0,
"step": 1260
},
{
"entropy": 6.085949420928955,
"epoch": 0.11474963715529753,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999935140414674,
"loss": 6.1738,
"mean_token_accuracy": 0.12467294484376908,
"num_tokens": 2623999.0,
"step": 1265
},
{
"entropy": 6.378130960464477,
"epoch": 0.11520319303338171,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999932660359395,
"loss": 6.3881,
"mean_token_accuracy": 0.12474165633320808,
"num_tokens": 2636208.0,
"step": 1270
},
{
"entropy": 6.177397966384888,
"epoch": 0.11565674891146589,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999930133774936,
"loss": 6.2635,
"mean_token_accuracy": 0.1310955010354519,
"num_tokens": 2645854.0,
"step": 1275
},
{
"entropy": 6.334166765213013,
"epoch": 0.11611030478955008,
"grad_norm": 1.0625,
"learning_rate": 0.0004999927560661348,
"loss": 6.2513,
"mean_token_accuracy": 0.13245694637298583,
"num_tokens": 2655877.0,
"step": 1280
},
{
"entropy": 6.226203727722168,
"epoch": 0.11656386066763426,
"grad_norm": 1.046875,
"learning_rate": 0.0004999924941018685,
"loss": 6.2578,
"mean_token_accuracy": 0.12858212292194365,
"num_tokens": 2666157.0,
"step": 1285
},
{
"entropy": 6.180522871017456,
"epoch": 0.11701741654571843,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999922274847,
"loss": 6.1825,
"mean_token_accuracy": 0.1282689355313778,
"num_tokens": 2676647.0,
"step": 1290
},
{
"entropy": 6.1607826232910154,
"epoch": 0.11747097242380261,
"grad_norm": 1.0703125,
"learning_rate": 0.000499991956214635,
"loss": 6.2123,
"mean_token_accuracy": 0.14043712615966797,
"num_tokens": 2686417.0,
"step": 1295
},
{
"entropy": 6.220397090911865,
"epoch": 0.1179245283018868,
"grad_norm": 0.9921875,
"learning_rate": 0.000499991680291679,
"loss": 6.2217,
"mean_token_accuracy": 0.1216107651591301,
"num_tokens": 2697540.0,
"step": 1300
},
{
"entropy": 6.173789548873901,
"epoch": 0.11837808417997098,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999913997158378,
"loss": 6.2622,
"mean_token_accuracy": 0.11780738160014152,
"num_tokens": 2709397.0,
"step": 1305
},
{
"entropy": 6.271051549911499,
"epoch": 0.11883164005805515,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999911144871169,
"loss": 6.2158,
"mean_token_accuracy": 0.1249967023730278,
"num_tokens": 2720430.0,
"step": 1310
},
{
"entropy": 6.155363607406616,
"epoch": 0.11928519593613933,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999908246055226,
"loss": 6.1669,
"mean_token_accuracy": 0.12677768394351005,
"num_tokens": 2730648.0,
"step": 1315
},
{
"entropy": 6.153974342346191,
"epoch": 0.11973875181422351,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999905300710607,
"loss": 6.1472,
"mean_token_accuracy": 0.1284649021923542,
"num_tokens": 2739979.0,
"step": 1320
},
{
"entropy": 6.25038104057312,
"epoch": 0.1201923076923077,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999902308837373,
"loss": 6.3028,
"mean_token_accuracy": 0.12176627144217492,
"num_tokens": 2750329.0,
"step": 1325
},
{
"entropy": 6.140483713150024,
"epoch": 0.12064586357039188,
"grad_norm": 1.046875,
"learning_rate": 0.0004999899270435584,
"loss": 6.1728,
"mean_token_accuracy": 0.1314334973692894,
"num_tokens": 2760535.0,
"step": 1330
},
{
"entropy": 6.230471897125244,
"epoch": 0.12109941944847605,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999896185505307,
"loss": 6.2044,
"mean_token_accuracy": 0.12265649288892747,
"num_tokens": 2770470.0,
"step": 1335
},
{
"entropy": 6.159397172927856,
"epoch": 0.12155297532656023,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999893054046603,
"loss": 6.2145,
"mean_token_accuracy": 0.12754668965935706,
"num_tokens": 2781506.0,
"step": 1340
},
{
"entropy": 6.125198554992676,
"epoch": 0.12200653120464441,
"grad_norm": 1.046875,
"learning_rate": 0.0004999889876059537,
"loss": 6.1021,
"mean_token_accuracy": 0.13616543412208557,
"num_tokens": 2791245.0,
"step": 1345
},
{
"entropy": 6.152018737792969,
"epoch": 0.1224600870827286,
"grad_norm": 1.09375,
"learning_rate": 0.0004999886651544176,
"loss": 6.1054,
"mean_token_accuracy": 0.13110152259469032,
"num_tokens": 2801680.0,
"step": 1350
},
{
"entropy": 6.157893323898316,
"epoch": 0.12291364296081277,
"grad_norm": 1.03125,
"learning_rate": 0.0004999883380500584,
"loss": 6.1862,
"mean_token_accuracy": 0.13278760090470315,
"num_tokens": 2812805.0,
"step": 1355
},
{
"entropy": 6.238238143920898,
"epoch": 0.12336719883889695,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999880062928831,
"loss": 6.2099,
"mean_token_accuracy": 0.12580350860953332,
"num_tokens": 2822834.0,
"step": 1360
},
{
"entropy": 6.129199361801147,
"epoch": 0.12382075471698113,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999876698828985,
"loss": 6.1785,
"mean_token_accuracy": 0.1322852909564972,
"num_tokens": 2832312.0,
"step": 1365
},
{
"entropy": 6.211675024032592,
"epoch": 0.12427431059506532,
"grad_norm": 1.0,
"learning_rate": 0.0004999873288201116,
"loss": 6.1859,
"mean_token_accuracy": 0.13278617039322854,
"num_tokens": 2842998.0,
"step": 1370
},
{
"entropy": 6.222190952301025,
"epoch": 0.1247278664731495,
"grad_norm": 1.4609375,
"learning_rate": 0.0004999869831045294,
"loss": 6.22,
"mean_token_accuracy": 0.12911526262760162,
"num_tokens": 2854218.0,
"step": 1375
},
{
"entropy": 6.1174554347991945,
"epoch": 0.12518142235123367,
"grad_norm": 1.03125,
"learning_rate": 0.0004999866327361589,
"loss": 6.0953,
"mean_token_accuracy": 0.13779579773545264,
"num_tokens": 2864484.0,
"step": 1380
},
{
"entropy": 6.212736892700195,
"epoch": 0.12563497822931785,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999862777150077,
"loss": 6.2057,
"mean_token_accuracy": 0.12786780819296836,
"num_tokens": 2873800.0,
"step": 1385
},
{
"entropy": 6.171827030181885,
"epoch": 0.12608853410740203,
"grad_norm": 0.984375,
"learning_rate": 0.0004999859180410829,
"loss": 6.2171,
"mean_token_accuracy": 0.1288387209177017,
"num_tokens": 2884247.0,
"step": 1390
},
{
"entropy": 6.141082382202148,
"epoch": 0.12654208998548622,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999855537143919,
"loss": 6.1257,
"mean_token_accuracy": 0.12706563696265222,
"num_tokens": 2894739.0,
"step": 1395
},
{
"entropy": 6.073680734634399,
"epoch": 0.1269956458635704,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999851847349425,
"loss": 6.1807,
"mean_token_accuracy": 0.13040847554802895,
"num_tokens": 2904376.0,
"step": 1400
},
{
"entropy": 6.245872640609742,
"epoch": 0.12744920174165458,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999848111027419,
"loss": 6.2325,
"mean_token_accuracy": 0.127915970236063,
"num_tokens": 2914304.0,
"step": 1405
},
{
"entropy": 6.183934211730957,
"epoch": 0.12790275761973874,
"grad_norm": 1.09375,
"learning_rate": 0.0004999844328177983,
"loss": 6.1235,
"mean_token_accuracy": 0.12716101333498955,
"num_tokens": 2924400.0,
"step": 1410
},
{
"entropy": 6.2829203605651855,
"epoch": 0.12835631349782292,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999840498801191,
"loss": 6.347,
"mean_token_accuracy": 0.13103133216500282,
"num_tokens": 2935040.0,
"step": 1415
},
{
"entropy": 6.135620164871216,
"epoch": 0.1288098693759071,
"grad_norm": 1.0625,
"learning_rate": 0.0004999836622897126,
"loss": 6.1366,
"mean_token_accuracy": 0.13396168127655983,
"num_tokens": 2945536.0,
"step": 1420
},
{
"entropy": 6.0857262134552,
"epoch": 0.1292634252539913,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999832700465865,
"loss": 6.0541,
"mean_token_accuracy": 0.13348359540104865,
"num_tokens": 2954969.0,
"step": 1425
},
{
"entropy": 6.16528549194336,
"epoch": 0.12971698113207547,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999828731507491,
"loss": 6.1774,
"mean_token_accuracy": 0.12718980759382248,
"num_tokens": 2966218.0,
"step": 1430
},
{
"entropy": 6.185882234573365,
"epoch": 0.13017053701015965,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999824716022085,
"loss": 6.1999,
"mean_token_accuracy": 0.12183153331279754,
"num_tokens": 2976640.0,
"step": 1435
},
{
"entropy": 6.093891048431397,
"epoch": 0.13062409288824384,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999820654009731,
"loss": 6.1687,
"mean_token_accuracy": 0.131185482442379,
"num_tokens": 2985883.0,
"step": 1440
},
{
"entropy": 6.16903977394104,
"epoch": 0.13107764876632802,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999816545470513,
"loss": 6.1739,
"mean_token_accuracy": 0.12456371188163758,
"num_tokens": 2996695.0,
"step": 1445
},
{
"entropy": 6.17958173751831,
"epoch": 0.1315312046444122,
"grad_norm": 1.09375,
"learning_rate": 0.0004999812390404514,
"loss": 6.1945,
"mean_token_accuracy": 0.128679159283638,
"num_tokens": 3007500.0,
"step": 1450
},
{
"entropy": 5.967514085769653,
"epoch": 0.13198476052249636,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999808188811823,
"loss": 6.0384,
"mean_token_accuracy": 0.13110998794436454,
"num_tokens": 3017299.0,
"step": 1455
},
{
"entropy": 6.2758769512176515,
"epoch": 0.13243831640058054,
"grad_norm": 1.015625,
"learning_rate": 0.0004999803940692524,
"loss": 6.3073,
"mean_token_accuracy": 0.1230485461652279,
"num_tokens": 3028389.0,
"step": 1460
},
{
"entropy": 6.220829582214355,
"epoch": 0.13289187227866472,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999799646046707,
"loss": 6.1307,
"mean_token_accuracy": 0.13422513082623483,
"num_tokens": 3038883.0,
"step": 1465
},
{
"entropy": 6.163057661056518,
"epoch": 0.1333454281567489,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999795304874461,
"loss": 6.2457,
"mean_token_accuracy": 0.13295281678438187,
"num_tokens": 3049904.0,
"step": 1470
},
{
"entropy": 6.084758853912353,
"epoch": 0.1337989840348331,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999790917175873,
"loss": 6.0809,
"mean_token_accuracy": 0.13047859445214272,
"num_tokens": 3059788.0,
"step": 1475
},
{
"entropy": 6.1702046394348145,
"epoch": 0.13425253991291727,
"grad_norm": 1.078125,
"learning_rate": 0.0004999786482951036,
"loss": 6.0729,
"mean_token_accuracy": 0.1356496147811413,
"num_tokens": 3070413.0,
"step": 1480
},
{
"entropy": 6.121113586425781,
"epoch": 0.13470609579100146,
"grad_norm": 0.890625,
"learning_rate": 0.0004999782002200041,
"loss": 6.2024,
"mean_token_accuracy": 0.12601577937602998,
"num_tokens": 3081093.0,
"step": 1485
},
{
"entropy": 6.136486053466797,
"epoch": 0.13515965166908564,
"grad_norm": 1.09375,
"learning_rate": 0.0004999777474922982,
"loss": 6.1304,
"mean_token_accuracy": 0.12720816284418107,
"num_tokens": 3090594.0,
"step": 1490
},
{
"entropy": 6.193129396438598,
"epoch": 0.13561320754716982,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999772901119951,
"loss": 6.2432,
"mean_token_accuracy": 0.12697395980358123,
"num_tokens": 3101061.0,
"step": 1495
},
{
"entropy": 6.138486433029175,
"epoch": 0.13606676342525398,
"grad_norm": 0.984375,
"learning_rate": 0.0004999768280791043,
"loss": 6.1538,
"mean_token_accuracy": 0.13143179193139076,
"num_tokens": 3111715.0,
"step": 1500
},
{
"entropy": 6.113825941085816,
"epoch": 0.13652031930333816,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999763613936352,
"loss": 6.0641,
"mean_token_accuracy": 0.13530101105570794,
"num_tokens": 3121378.0,
"step": 1505
},
{
"entropy": 6.079781913757325,
"epoch": 0.13697387518142234,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999758900555979,
"loss": 6.013,
"mean_token_accuracy": 0.13859829008579255,
"num_tokens": 3131229.0,
"step": 1510
},
{
"entropy": 6.068106603622437,
"epoch": 0.13742743105950653,
"grad_norm": 1.1875,
"learning_rate": 0.0004999754140650017,
"loss": 6.1133,
"mean_token_accuracy": 0.12841149419546127,
"num_tokens": 3140859.0,
"step": 1515
},
{
"entropy": 6.186432981491089,
"epoch": 0.1378809869375907,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999749334218567,
"loss": 6.1079,
"mean_token_accuracy": 0.13175800815224648,
"num_tokens": 3151692.0,
"step": 1520
},
{
"entropy": 6.034256410598755,
"epoch": 0.1383345428156749,
"grad_norm": 1.09375,
"learning_rate": 0.0004999744481261726,
"loss": 6.1773,
"mean_token_accuracy": 0.12917248532176018,
"num_tokens": 3161879.0,
"step": 1525
},
{
"entropy": 6.104421138763428,
"epoch": 0.13878809869375908,
"grad_norm": 1.171875,
"learning_rate": 0.0004999739581779597,
"loss": 6.0365,
"mean_token_accuracy": 0.1392836257815361,
"num_tokens": 3171282.0,
"step": 1530
},
{
"entropy": 6.098322439193725,
"epoch": 0.13924165457184326,
"grad_norm": 1.03125,
"learning_rate": 0.000499973463577228,
"loss": 6.0919,
"mean_token_accuracy": 0.13791340962052345,
"num_tokens": 3181398.0,
"step": 1535
},
{
"entropy": 6.116310262680054,
"epoch": 0.13969521044992744,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999729643239877,
"loss": 6.221,
"mean_token_accuracy": 0.13014425411820413,
"num_tokens": 3191431.0,
"step": 1540
},
{
"entropy": 6.162513589859008,
"epoch": 0.1401487663280116,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999724604182492,
"loss": 6.2272,
"mean_token_accuracy": 0.12255278900265694,
"num_tokens": 3202147.0,
"step": 1545
},
{
"entropy": 6.2134605884552006,
"epoch": 0.14060232220609578,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999719518600229,
"loss": 6.0894,
"mean_token_accuracy": 0.13580598905682564,
"num_tokens": 3212428.0,
"step": 1550
},
{
"entropy": 6.07160701751709,
"epoch": 0.14105587808417996,
"grad_norm": 1.09375,
"learning_rate": 0.0004999714386493192,
"loss": 6.1527,
"mean_token_accuracy": 0.13032715693116187,
"num_tokens": 3222099.0,
"step": 1555
},
{
"entropy": 6.135047817230225,
"epoch": 0.14150943396226415,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999709207861489,
"loss": 6.0821,
"mean_token_accuracy": 0.13343799188733102,
"num_tokens": 3232079.0,
"step": 1560
},
{
"entropy": 6.043800163269043,
"epoch": 0.14196298984034833,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999703982705227,
"loss": 6.0044,
"mean_token_accuracy": 0.14089397639036177,
"num_tokens": 3241762.0,
"step": 1565
},
{
"entropy": 6.056787490844727,
"epoch": 0.1424165457184325,
"grad_norm": 1.109375,
"learning_rate": 0.0004999698711024513,
"loss": 6.0942,
"mean_token_accuracy": 0.13566121682524682,
"num_tokens": 3252571.0,
"step": 1570
},
{
"entropy": 6.119260835647583,
"epoch": 0.1428701015965167,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999693392819456,
"loss": 6.1276,
"mean_token_accuracy": 0.12969395369291306,
"num_tokens": 3263542.0,
"step": 1575
},
{
"entropy": 6.129227113723755,
"epoch": 0.14332365747460088,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999688028090166,
"loss": 6.2,
"mean_token_accuracy": 0.12882068231701851,
"num_tokens": 3273295.0,
"step": 1580
},
{
"entropy": 6.217692804336548,
"epoch": 0.14377721335268506,
"grad_norm": 1.046875,
"learning_rate": 0.0004999682616836755,
"loss": 6.23,
"mean_token_accuracy": 0.13299182802438736,
"num_tokens": 3283605.0,
"step": 1585
},
{
"entropy": 6.094928216934204,
"epoch": 0.14423076923076922,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999677159059334,
"loss": 6.0657,
"mean_token_accuracy": 0.1365733303129673,
"num_tokens": 3293837.0,
"step": 1590
},
{
"entropy": 5.993037939071655,
"epoch": 0.1446843251088534,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999671654758016,
"loss": 6.0685,
"mean_token_accuracy": 0.14040320217609406,
"num_tokens": 3304131.0,
"step": 1595
},
{
"entropy": 6.073614025115967,
"epoch": 0.14513788098693758,
"grad_norm": 1.359375,
"learning_rate": 0.0004999666103932915,
"loss": 6.0992,
"mean_token_accuracy": 0.13104889690876007,
"num_tokens": 3313543.0,
"step": 1600
},
{
"entropy": 6.087307071685791,
"epoch": 0.14559143686502177,
"grad_norm": 1.0,
"learning_rate": 0.0004999660506584145,
"loss": 6.0907,
"mean_token_accuracy": 0.13283502534031869,
"num_tokens": 3324487.0,
"step": 1605
},
{
"entropy": 6.13335018157959,
"epoch": 0.14604499274310595,
"grad_norm": 0.96875,
"learning_rate": 0.0004999654862711823,
"loss": 6.0993,
"mean_token_accuracy": 0.13403137400746346,
"num_tokens": 3335094.0,
"step": 1610
},
{
"entropy": 6.077793455123901,
"epoch": 0.14649854862119013,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999649172316065,
"loss": 6.033,
"mean_token_accuracy": 0.13351959139108657,
"num_tokens": 3344252.0,
"step": 1615
},
{
"entropy": 6.051349306106568,
"epoch": 0.14695210449927432,
"grad_norm": 1.0546875,
"learning_rate": 0.000499964343539699,
"loss": 6.1092,
"mean_token_accuracy": 0.12948338314890862,
"num_tokens": 3354265.0,
"step": 1620
},
{
"entropy": 6.167207384109497,
"epoch": 0.1474056603773585,
"grad_norm": 1.0,
"learning_rate": 0.0004999637651954714,
"loss": 6.2254,
"mean_token_accuracy": 0.1327558323740959,
"num_tokens": 3364616.0,
"step": 1625
},
{
"entropy": 6.135294818878174,
"epoch": 0.14785921625544268,
"grad_norm": 1.078125,
"learning_rate": 0.0004999631821989358,
"loss": 6.1636,
"mean_token_accuracy": 0.1339091271162033,
"num_tokens": 3373650.0,
"step": 1630
},
{
"entropy": 5.991321992874146,
"epoch": 0.14831277213352684,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999625945501043,
"loss": 6.0624,
"mean_token_accuracy": 0.13642672672867776,
"num_tokens": 3384590.0,
"step": 1635
},
{
"entropy": 6.221540021896362,
"epoch": 0.14876632801161102,
"grad_norm": 1.1171875,
"learning_rate": 0.000499962002248989,
"loss": 6.1045,
"mean_token_accuracy": 0.13118307739496232,
"num_tokens": 3395616.0,
"step": 1640
},
{
"entropy": 6.03116774559021,
"epoch": 0.1492198838896952,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999614052956024,
"loss": 6.092,
"mean_token_accuracy": 0.13425416350364686,
"num_tokens": 3405810.0,
"step": 1645
},
{
"entropy": 6.0968505859375,
"epoch": 0.1496734397677794,
"grad_norm": 0.96875,
"learning_rate": 0.0004999608036899563,
"loss": 6.1287,
"mean_token_accuracy": 0.13202429711818695,
"num_tokens": 3415932.0,
"step": 1650
},
{
"entropy": 6.095308542251587,
"epoch": 0.15012699564586357,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999601974320636,
"loss": 6.0899,
"mean_token_accuracy": 0.12914085909724235,
"num_tokens": 3425626.0,
"step": 1655
},
{
"entropy": 6.025156164169312,
"epoch": 0.15058055152394775,
"grad_norm": 1.109375,
"learning_rate": 0.0004999595865219367,
"loss": 6.0764,
"mean_token_accuracy": 0.12880702540278435,
"num_tokens": 3434837.0,
"step": 1660
},
{
"entropy": 6.111604022979736,
"epoch": 0.15103410740203194,
"grad_norm": 1.09375,
"learning_rate": 0.0004999589709595881,
"loss": 6.1073,
"mean_token_accuracy": 0.13633522242307664,
"num_tokens": 3445238.0,
"step": 1665
},
{
"entropy": 6.0251708984375,
"epoch": 0.15148766328011612,
"grad_norm": 1.125,
"learning_rate": 0.0004999583507450308,
"loss": 6.0934,
"mean_token_accuracy": 0.1300239272415638,
"num_tokens": 3456738.0,
"step": 1670
},
{
"entropy": 5.999286699295044,
"epoch": 0.1519412191582003,
"grad_norm": 1.03125,
"learning_rate": 0.0004999577258782774,
"loss": 5.957,
"mean_token_accuracy": 0.13702991008758544,
"num_tokens": 3466296.0,
"step": 1675
},
{
"entropy": 6.128127813339233,
"epoch": 0.15239477503628446,
"grad_norm": 0.98828125,
"learning_rate": 0.000499957096359341,
"loss": 6.1536,
"mean_token_accuracy": 0.12765773385763168,
"num_tokens": 3476989.0,
"step": 1680
},
{
"entropy": 6.0421911716461185,
"epoch": 0.15284833091436864,
"grad_norm": 1.21875,
"learning_rate": 0.0004999564621882343,
"loss": 6.0026,
"mean_token_accuracy": 0.1342094786465168,
"num_tokens": 3487050.0,
"step": 1685
},
{
"entropy": 6.144343376159668,
"epoch": 0.15330188679245282,
"grad_norm": 1.046875,
"learning_rate": 0.0004999558233649708,
"loss": 6.2183,
"mean_token_accuracy": 0.12711812555789948,
"num_tokens": 3497777.0,
"step": 1690
},
{
"entropy": 6.131715631484985,
"epoch": 0.153755442670537,
"grad_norm": 1.015625,
"learning_rate": 0.0004999551798895636,
"loss": 6.1137,
"mean_token_accuracy": 0.12770443707704543,
"num_tokens": 3509066.0,
"step": 1695
},
{
"entropy": 6.020261526107788,
"epoch": 0.1542089985486212,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999545317620258,
"loss": 6.0385,
"mean_token_accuracy": 0.1325708270072937,
"num_tokens": 3519567.0,
"step": 1700
},
{
"entropy": 6.05970869064331,
"epoch": 0.15466255442670537,
"grad_norm": 0.984375,
"learning_rate": 0.0004999538789823711,
"loss": 6.0921,
"mean_token_accuracy": 0.13301847204566003,
"num_tokens": 3529402.0,
"step": 1705
},
{
"entropy": 6.176769733428955,
"epoch": 0.15511611030478956,
"grad_norm": 1.03125,
"learning_rate": 0.0004999532215506127,
"loss": 6.0909,
"mean_token_accuracy": 0.14401522800326347,
"num_tokens": 3539805.0,
"step": 1710
},
{
"entropy": 5.969664430618286,
"epoch": 0.15556966618287374,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999525594667645,
"loss": 5.9514,
"mean_token_accuracy": 0.13585785627365113,
"num_tokens": 3550477.0,
"step": 1715
},
{
"entropy": 5.984322881698608,
"epoch": 0.15602322206095792,
"grad_norm": 1.1171875,
"learning_rate": 0.00049995189273084,
"loss": 6.0522,
"mean_token_accuracy": 0.13302028104662894,
"num_tokens": 3560729.0,
"step": 1720
},
{
"entropy": 6.111390686035156,
"epoch": 0.15647677793904208,
"grad_norm": 1.046875,
"learning_rate": 0.000499951221342853,
"loss": 6.1395,
"mean_token_accuracy": 0.12391486763954163,
"num_tokens": 3571538.0,
"step": 1725
},
{
"entropy": 6.122786998748779,
"epoch": 0.15693033381712626,
"grad_norm": 1.015625,
"learning_rate": 0.0004999505453028174,
"loss": 6.1449,
"mean_token_accuracy": 0.1329157382249832,
"num_tokens": 3583128.0,
"step": 1730
},
{
"entropy": 6.065582609176635,
"epoch": 0.15738388969521044,
"grad_norm": 0.96875,
"learning_rate": 0.0004999498646107472,
"loss": 6.0667,
"mean_token_accuracy": 0.13652407601475716,
"num_tokens": 3594358.0,
"step": 1735
},
{
"entropy": 6.036913299560547,
"epoch": 0.15783744557329463,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999491792666566,
"loss": 6.0088,
"mean_token_accuracy": 0.1374991677701473,
"num_tokens": 3604345.0,
"step": 1740
},
{
"entropy": 6.039101409912109,
"epoch": 0.1582910014513788,
"grad_norm": 1.015625,
"learning_rate": 0.0004999484892705597,
"loss": 6.055,
"mean_token_accuracy": 0.13434040173888206,
"num_tokens": 3616129.0,
"step": 1745
},
{
"entropy": 6.09887547492981,
"epoch": 0.158744557329463,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999477946224705,
"loss": 6.1052,
"mean_token_accuracy": 0.13473358303308486,
"num_tokens": 3627458.0,
"step": 1750
},
{
"entropy": 6.10147271156311,
"epoch": 0.15919811320754718,
"grad_norm": 1.109375,
"learning_rate": 0.0004999470953224037,
"loss": 5.9868,
"mean_token_accuracy": 0.1373836562037468,
"num_tokens": 3636047.0,
"step": 1755
},
{
"entropy": 5.975709295272827,
"epoch": 0.15965166908563136,
"grad_norm": 1.0,
"learning_rate": 0.0004999463913703735,
"loss": 5.9717,
"mean_token_accuracy": 0.13532424792647363,
"num_tokens": 3646930.0,
"step": 1760
},
{
"entropy": 6.0145656108856205,
"epoch": 0.16010522496371554,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999456827663948,
"loss": 6.0672,
"mean_token_accuracy": 0.13636473268270494,
"num_tokens": 3657279.0,
"step": 1765
},
{
"entropy": 6.016043090820313,
"epoch": 0.1605587808417997,
"grad_norm": 1.0546875,
"learning_rate": 0.000499944969510482,
"loss": 5.9569,
"mean_token_accuracy": 0.13713241890072822,
"num_tokens": 3667070.0,
"step": 1770
},
{
"entropy": 6.058058595657348,
"epoch": 0.16101233671988388,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999442516026498,
"loss": 5.9892,
"mean_token_accuracy": 0.13559969365596772,
"num_tokens": 3677560.0,
"step": 1775
},
{
"entropy": 5.985709571838379,
"epoch": 0.16146589259796806,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999435290429133,
"loss": 5.946,
"mean_token_accuracy": 0.14185371547937392,
"num_tokens": 3686701.0,
"step": 1780
},
{
"entropy": 5.999032068252563,
"epoch": 0.16191944847605225,
"grad_norm": 1.046875,
"learning_rate": 0.0004999428018312872,
"loss": 6.0462,
"mean_token_accuracy": 0.14263416901230813,
"num_tokens": 3696338.0,
"step": 1785
},
{
"entropy": 6.082455444335937,
"epoch": 0.16237300435413643,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999420699677867,
"loss": 6.0388,
"mean_token_accuracy": 0.1367315612733364,
"num_tokens": 3705752.0,
"step": 1790
},
{
"entropy": 6.089331150054932,
"epoch": 0.1628265602322206,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999413334524269,
"loss": 6.1127,
"mean_token_accuracy": 0.13379622250795364,
"num_tokens": 3716225.0,
"step": 1795
},
{
"entropy": 6.001873111724853,
"epoch": 0.1632801161103048,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999405922852229,
"loss": 5.9867,
"mean_token_accuracy": 0.13304651603102685,
"num_tokens": 3726593.0,
"step": 1800
},
{
"entropy": 5.997346496582031,
"epoch": 0.16373367198838898,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999398464661901,
"loss": 6.0555,
"mean_token_accuracy": 0.13038523942232133,
"num_tokens": 3737563.0,
"step": 1805
},
{
"entropy": 6.145504331588745,
"epoch": 0.16418722786647316,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999390959953441,
"loss": 6.1043,
"mean_token_accuracy": 0.13330675959587096,
"num_tokens": 3747752.0,
"step": 1810
},
{
"entropy": 5.9703622341156,
"epoch": 0.16464078374455732,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999383408727002,
"loss": 6.0364,
"mean_token_accuracy": 0.13549003675580024,
"num_tokens": 3758795.0,
"step": 1815
},
{
"entropy": 6.059307003021241,
"epoch": 0.1650943396226415,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999375810982741,
"loss": 6.0177,
"mean_token_accuracy": 0.1396925114095211,
"num_tokens": 3768690.0,
"step": 1820
},
{
"entropy": 6.01865873336792,
"epoch": 0.16554789550072568,
"grad_norm": 1.046875,
"learning_rate": 0.0004999368166720815,
"loss": 6.0841,
"mean_token_accuracy": 0.13427145406603813,
"num_tokens": 3779018.0,
"step": 1825
},
{
"entropy": 6.140135812759399,
"epoch": 0.16600145137880987,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999360475941382,
"loss": 5.9842,
"mean_token_accuracy": 0.14041623398661612,
"num_tokens": 3788985.0,
"step": 1830
},
{
"entropy": 6.026188707351684,
"epoch": 0.16645500725689405,
"grad_norm": 1.03125,
"learning_rate": 0.00049993527386446,
"loss": 6.1309,
"mean_token_accuracy": 0.12635482400655745,
"num_tokens": 3800787.0,
"step": 1835
},
{
"entropy": 6.024060583114624,
"epoch": 0.16690856313497823,
"grad_norm": 1.125,
"learning_rate": 0.0004999344954830633,
"loss": 6.0076,
"mean_token_accuracy": 0.13396736234426498,
"num_tokens": 3811381.0,
"step": 1840
},
{
"entropy": 6.089121675491333,
"epoch": 0.16736211901306242,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999337124499638,
"loss": 6.1121,
"mean_token_accuracy": 0.13973291739821433,
"num_tokens": 3822055.0,
"step": 1845
},
{
"entropy": 5.998512506484985,
"epoch": 0.1678156748911466,
"grad_norm": 1.109375,
"learning_rate": 0.0004999329247651777,
"loss": 6.0637,
"mean_token_accuracy": 0.1306314118206501,
"num_tokens": 3831900.0,
"step": 1850
},
{
"entropy": 6.045335531234741,
"epoch": 0.16826923076923078,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999321324287213,
"loss": 6.0357,
"mean_token_accuracy": 0.13573757112026213,
"num_tokens": 3842702.0,
"step": 1855
},
{
"entropy": 6.1244423389434814,
"epoch": 0.16872278664731494,
"grad_norm": 1.09375,
"learning_rate": 0.0004999313354406114,
"loss": 5.9929,
"mean_token_accuracy": 0.13379636257886887,
"num_tokens": 3852690.0,
"step": 1860
},
{
"entropy": 5.9076738357543945,
"epoch": 0.16917634252539912,
"grad_norm": 1.046875,
"learning_rate": 0.000499930533800864,
"loss": 5.993,
"mean_token_accuracy": 0.14352267235517502,
"num_tokens": 3863040.0,
"step": 1865
},
{
"entropy": 6.074823522567749,
"epoch": 0.1696298984034833,
"grad_norm": 1.046875,
"learning_rate": 0.0004999297275094958,
"loss": 6.0803,
"mean_token_accuracy": 0.13035911098122596,
"num_tokens": 3873613.0,
"step": 1870
},
{
"entropy": 5.932005929946899,
"epoch": 0.1700834542815675,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999289165665235,
"loss": 5.8793,
"mean_token_accuracy": 0.14385106191039085,
"num_tokens": 3883936.0,
"step": 1875
},
{
"entropy": 5.965173816680908,
"epoch": 0.17053701015965167,
"grad_norm": 1.3984375,
"learning_rate": 0.0004999281009719639,
"loss": 5.8597,
"mean_token_accuracy": 0.14646447375416755,
"num_tokens": 3893788.0,
"step": 1880
},
{
"entropy": 5.964988851547242,
"epoch": 0.17099056603773585,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999272807258339,
"loss": 5.9165,
"mean_token_accuracy": 0.14181002825498581,
"num_tokens": 3903434.0,
"step": 1885
},
{
"entropy": 5.97712721824646,
"epoch": 0.17144412191582004,
"grad_norm": 1.125,
"learning_rate": 0.0004999264558281504,
"loss": 5.9377,
"mean_token_accuracy": 0.14012128487229347,
"num_tokens": 3913476.0,
"step": 1890
},
{
"entropy": 5.872003698348999,
"epoch": 0.17189767779390422,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999256262789305,
"loss": 5.8922,
"mean_token_accuracy": 0.14096907377243043,
"num_tokens": 3923301.0,
"step": 1895
},
{
"entropy": 5.992045307159424,
"epoch": 0.1723512336719884,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999247920781913,
"loss": 6.0344,
"mean_token_accuracy": 0.13852274641394616,
"num_tokens": 3933335.0,
"step": 1900
},
{
"entropy": 6.088766717910767,
"epoch": 0.17280478955007256,
"grad_norm": 1.015625,
"learning_rate": 0.0004999239532259502,
"loss": 6.0646,
"mean_token_accuracy": 0.1291509412229061,
"num_tokens": 3944171.0,
"step": 1905
},
{
"entropy": 6.080254316329956,
"epoch": 0.17325834542815674,
"grad_norm": 1.078125,
"learning_rate": 0.0004999231097222244,
"loss": 6.1305,
"mean_token_accuracy": 0.13909375816583633,
"num_tokens": 3954160.0,
"step": 1910
},
{
"entropy": 6.020297527313232,
"epoch": 0.17371190130624092,
"grad_norm": 1.125,
"learning_rate": 0.0004999222615670312,
"loss": 5.9253,
"mean_token_accuracy": 0.14530864879488944,
"num_tokens": 3964110.0,
"step": 1915
},
{
"entropy": 6.07570161819458,
"epoch": 0.1741654571843251,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999214087603885,
"loss": 6.0159,
"mean_token_accuracy": 0.13602590560913086,
"num_tokens": 3974293.0,
"step": 1920
},
{
"entropy": 5.922182607650757,
"epoch": 0.1746190130624093,
"grad_norm": 1.078125,
"learning_rate": 0.0004999205513023135,
"loss": 5.991,
"mean_token_accuracy": 0.1404339738190174,
"num_tokens": 3984346.0,
"step": 1925
},
{
"entropy": 5.97127251625061,
"epoch": 0.17507256894049347,
"grad_norm": 1.046875,
"learning_rate": 0.0004999196891928245,
"loss": 5.9094,
"mean_token_accuracy": 0.1484488442540169,
"num_tokens": 3992897.0,
"step": 1930
},
{
"entropy": 5.9334413528442385,
"epoch": 0.17552612481857766,
"grad_norm": 1.078125,
"learning_rate": 0.0004999188224319388,
"loss": 5.9525,
"mean_token_accuracy": 0.1443849354982376,
"num_tokens": 4002575.0,
"step": 1935
},
{
"entropy": 6.02183780670166,
"epoch": 0.17597968069666184,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999179510196746,
"loss": 6.027,
"mean_token_accuracy": 0.13418469578027725,
"num_tokens": 4013752.0,
"step": 1940
},
{
"entropy": 6.056840991973877,
"epoch": 0.17643323657474602,
"grad_norm": 1.125,
"learning_rate": 0.0004999170749560498,
"loss": 5.9132,
"mean_token_accuracy": 0.14471341520547867,
"num_tokens": 4023718.0,
"step": 1945
},
{
"entropy": 5.9298981666564945,
"epoch": 0.17688679245283018,
"grad_norm": 1.203125,
"learning_rate": 0.0004999161942410827,
"loss": 5.9974,
"mean_token_accuracy": 0.13362503573298454,
"num_tokens": 4034210.0,
"step": 1950
},
{
"entropy": 6.088183450698852,
"epoch": 0.17734034833091436,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999153088747913,
"loss": 6.0029,
"mean_token_accuracy": 0.1354498915374279,
"num_tokens": 4044868.0,
"step": 1955
},
{
"entropy": 5.887863874435425,
"epoch": 0.17779390420899854,
"grad_norm": 1.109375,
"learning_rate": 0.000499914418857194,
"loss": 6.0004,
"mean_token_accuracy": 0.13930529579520226,
"num_tokens": 4055049.0,
"step": 1960
},
{
"entropy": 6.098280572891236,
"epoch": 0.17824746008708273,
"grad_norm": 0.96875,
"learning_rate": 0.0004999135241883092,
"loss": 6.0537,
"mean_token_accuracy": 0.13831883370876313,
"num_tokens": 4065889.0,
"step": 1965
},
{
"entropy": 6.039399814605713,
"epoch": 0.1787010159651669,
"grad_norm": 1.15625,
"learning_rate": 0.0004999126248681555,
"loss": 6.0688,
"mean_token_accuracy": 0.13603949919342995,
"num_tokens": 4077288.0,
"step": 1970
},
{
"entropy": 6.006522798538208,
"epoch": 0.1791545718432511,
"grad_norm": 1.171875,
"learning_rate": 0.0004999117208967513,
"loss": 6.0125,
"mean_token_accuracy": 0.13738583624362946,
"num_tokens": 4087402.0,
"step": 1975
},
{
"entropy": 6.079785299301148,
"epoch": 0.17960812772133528,
"grad_norm": 0.984375,
"learning_rate": 0.0004999108122741154,
"loss": 6.0676,
"mean_token_accuracy": 0.13442711606621743,
"num_tokens": 4098150.0,
"step": 1980
},
{
"entropy": 5.926851367950439,
"epoch": 0.18006168359941946,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999098990002666,
"loss": 5.9194,
"mean_token_accuracy": 0.14138512015342714,
"num_tokens": 4108134.0,
"step": 1985
},
{
"entropy": 6.07075514793396,
"epoch": 0.18051523947750364,
"grad_norm": 1.078125,
"learning_rate": 0.0004999089810752237,
"loss": 6.0739,
"mean_token_accuracy": 0.12807230651378632,
"num_tokens": 4118251.0,
"step": 1990
},
{
"entropy": 5.945065689086914,
"epoch": 0.1809687953555878,
"grad_norm": 1.09375,
"learning_rate": 0.000499908058499006,
"loss": 5.9098,
"mean_token_accuracy": 0.14311220571398736,
"num_tokens": 4129228.0,
"step": 1995
},
{
"entropy": 5.959850263595581,
"epoch": 0.18142235123367198,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999071312716321,
"loss": 5.9727,
"mean_token_accuracy": 0.14519958719611167,
"num_tokens": 4139391.0,
"step": 2000
},
{
"entropy": 6.003384590148926,
"epoch": 0.18187590711175616,
"grad_norm": 1.1875,
"learning_rate": 0.0004999061993931215,
"loss": 5.9024,
"mean_token_accuracy": 0.1432528540492058,
"num_tokens": 4148891.0,
"step": 2005
},
{
"entropy": 6.182823896408081,
"epoch": 0.18232946298984035,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999052628634934,
"loss": 6.2626,
"mean_token_accuracy": 0.12997814267873764,
"num_tokens": 4159551.0,
"step": 2010
},
{
"entropy": 6.064494609832764,
"epoch": 0.18278301886792453,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999043216827671,
"loss": 6.0504,
"mean_token_accuracy": 0.13733192086219786,
"num_tokens": 4170055.0,
"step": 2015
},
{
"entropy": 5.916315412521362,
"epoch": 0.1832365747460087,
"grad_norm": 1.125,
"learning_rate": 0.0004999033758509621,
"loss": 5.9017,
"mean_token_accuracy": 0.1404242791235447,
"num_tokens": 4180767.0,
"step": 2020
},
{
"entropy": 5.9719846725463865,
"epoch": 0.1836901306240929,
"grad_norm": 1.03125,
"learning_rate": 0.000499902425368098,
"loss": 6.0294,
"mean_token_accuracy": 0.1393634244799614,
"num_tokens": 4190654.0,
"step": 2025
},
{
"entropy": 5.984382581710816,
"epoch": 0.18414368650217708,
"grad_norm": 1.125,
"learning_rate": 0.0004999014702341944,
"loss": 5.9548,
"mean_token_accuracy": 0.13900689631700516,
"num_tokens": 4200753.0,
"step": 2030
},
{
"entropy": 6.028017377853393,
"epoch": 0.18459724238026126,
"grad_norm": 1.09375,
"learning_rate": 0.0004999005104492712,
"loss": 6.0411,
"mean_token_accuracy": 0.1343386575579643,
"num_tokens": 4210994.0,
"step": 2035
},
{
"entropy": 5.9925353050231935,
"epoch": 0.18505079825834542,
"grad_norm": 1.125,
"learning_rate": 0.0004998995460133481,
"loss": 5.9756,
"mean_token_accuracy": 0.1370024561882019,
"num_tokens": 4220883.0,
"step": 2040
},
{
"entropy": 6.067998695373535,
"epoch": 0.1855043541364296,
"grad_norm": 1.015625,
"learning_rate": 0.0004998985769264451,
"loss": 6.0217,
"mean_token_accuracy": 0.13243423029780388,
"num_tokens": 4231269.0,
"step": 2045
},
{
"entropy": 5.975101900100708,
"epoch": 0.18595791001451378,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998976031885821,
"loss": 6.0222,
"mean_token_accuracy": 0.1368092842400074,
"num_tokens": 4241559.0,
"step": 2050
},
{
"entropy": 5.964649868011475,
"epoch": 0.18641146589259797,
"grad_norm": 1.171875,
"learning_rate": 0.0004998966247997794,
"loss": 5.8999,
"mean_token_accuracy": 0.14066432118415834,
"num_tokens": 4250870.0,
"step": 2055
},
{
"entropy": 5.934762763977051,
"epoch": 0.18686502177068215,
"grad_norm": 1.171875,
"learning_rate": 0.0004998956417600572,
"loss": 5.8707,
"mean_token_accuracy": 0.14538644403219222,
"num_tokens": 4260848.0,
"step": 2060
},
{
"entropy": 5.93128457069397,
"epoch": 0.18731857764876633,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998946540694358,
"loss": 5.9914,
"mean_token_accuracy": 0.14326113611459732,
"num_tokens": 4270973.0,
"step": 2065
},
{
"entropy": 6.073555898666382,
"epoch": 0.18777213352685052,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998936617279357,
"loss": 6.0634,
"mean_token_accuracy": 0.13676538169384003,
"num_tokens": 4281211.0,
"step": 2070
},
{
"entropy": 5.9082581996917725,
"epoch": 0.1882256894049347,
"grad_norm": 1.078125,
"learning_rate": 0.0004998926647355773,
"loss": 5.9195,
"mean_token_accuracy": 0.14342071637511253,
"num_tokens": 4291425.0,
"step": 2075
},
{
"entropy": 6.059259080886841,
"epoch": 0.18867924528301888,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998916630923813,
"loss": 6.0163,
"mean_token_accuracy": 0.13589237108826638,
"num_tokens": 4301079.0,
"step": 2080
},
{
"entropy": 5.99994306564331,
"epoch": 0.18913280116110304,
"grad_norm": 1.1875,
"learning_rate": 0.0004998906567983685,
"loss": 6.0042,
"mean_token_accuracy": 0.13732029348611832,
"num_tokens": 4312374.0,
"step": 2085
},
{
"entropy": 5.953618669509888,
"epoch": 0.18958635703918722,
"grad_norm": 1.09375,
"learning_rate": 0.0004998896458535595,
"loss": 5.9273,
"mean_token_accuracy": 0.1498640865087509,
"num_tokens": 4322810.0,
"step": 2090
},
{
"entropy": 5.99668779373169,
"epoch": 0.1900399129172714,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998886302579752,
"loss": 5.9219,
"mean_token_accuracy": 0.14000264406204224,
"num_tokens": 4332082.0,
"step": 2095
},
{
"entropy": 6.013230466842652,
"epoch": 0.19049346879535559,
"grad_norm": 1.1328125,
"learning_rate": 0.000499887610011637,
"loss": 6.0153,
"mean_token_accuracy": 0.13956746608018875,
"num_tokens": 4342632.0,
"step": 2100
},
{
"entropy": 5.974089479446411,
"epoch": 0.19094702467343977,
"grad_norm": 1.25,
"learning_rate": 0.0004998865851145656,
"loss": 5.9374,
"mean_token_accuracy": 0.13679615035653114,
"num_tokens": 4353544.0,
"step": 2105
},
{
"entropy": 5.911929941177368,
"epoch": 0.19140058055152395,
"grad_norm": 1.25,
"learning_rate": 0.0004998855555667822,
"loss": 6.0663,
"mean_token_accuracy": 0.13838788717985154,
"num_tokens": 4363636.0,
"step": 2110
},
{
"entropy": 6.095745468139649,
"epoch": 0.19185413642960814,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998845213683084,
"loss": 5.9418,
"mean_token_accuracy": 0.13569523617625237,
"num_tokens": 4373672.0,
"step": 2115
},
{
"entropy": 5.816546630859375,
"epoch": 0.19230769230769232,
"grad_norm": 1.1953125,
"learning_rate": 0.0004998834825191652,
"loss": 5.9133,
"mean_token_accuracy": 0.14208602756261826,
"num_tokens": 4383199.0,
"step": 2120
},
{
"entropy": 6.015786647796631,
"epoch": 0.1927612481857765,
"grad_norm": 1.09375,
"learning_rate": 0.0004998824390193743,
"loss": 5.983,
"mean_token_accuracy": 0.13925845474004744,
"num_tokens": 4394446.0,
"step": 2125
},
{
"entropy": 6.1271813869476315,
"epoch": 0.19321480406386066,
"grad_norm": 1.234375,
"learning_rate": 0.0004998813908689573,
"loss": 6.0262,
"mean_token_accuracy": 0.13644398227334023,
"num_tokens": 4405882.0,
"step": 2130
},
{
"entropy": 5.8932281017303465,
"epoch": 0.19366835994194484,
"grad_norm": 1.6171875,
"learning_rate": 0.0004998803380679359,
"loss": 5.8922,
"mean_token_accuracy": 0.14466665238142012,
"num_tokens": 4415280.0,
"step": 2135
},
{
"entropy": 5.9923975467681885,
"epoch": 0.19412191582002902,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998792806163317,
"loss": 6.0927,
"mean_token_accuracy": 0.13269788771867752,
"num_tokens": 4425755.0,
"step": 2140
},
{
"entropy": 6.003790664672851,
"epoch": 0.1945754716981132,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998782185141667,
"loss": 5.961,
"mean_token_accuracy": 0.13688114881515503,
"num_tokens": 4436259.0,
"step": 2145
},
{
"entropy": 6.004357957839966,
"epoch": 0.1950290275761974,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998771517614629,
"loss": 5.9919,
"mean_token_accuracy": 0.13306296169757842,
"num_tokens": 4446423.0,
"step": 2150
},
{
"entropy": 6.005547332763672,
"epoch": 0.19548258345428157,
"grad_norm": 1.171875,
"learning_rate": 0.0004998760803582422,
"loss": 5.9289,
"mean_token_accuracy": 0.14951801002025605,
"num_tokens": 4456678.0,
"step": 2155
},
{
"entropy": 5.9408186912536625,
"epoch": 0.19593613933236576,
"grad_norm": 1.21875,
"learning_rate": 0.0004998750043045269,
"loss": 5.924,
"mean_token_accuracy": 0.14363801032304763,
"num_tokens": 4467031.0,
"step": 2160
},
{
"entropy": 6.0042197704315186,
"epoch": 0.19638969521044994,
"grad_norm": 1.03125,
"learning_rate": 0.0004998739236003392,
"loss": 5.9952,
"mean_token_accuracy": 0.13384947553277016,
"num_tokens": 4477771.0,
"step": 2165
},
{
"entropy": 5.994749736785889,
"epoch": 0.1968432510885341,
"grad_norm": 1.171875,
"learning_rate": 0.0004998728382457014,
"loss": 6.0132,
"mean_token_accuracy": 0.13731033354997635,
"num_tokens": 4488426.0,
"step": 2170
},
{
"entropy": 6.060128211975098,
"epoch": 0.19729680696661828,
"grad_norm": 1.09375,
"learning_rate": 0.0004998717482406362,
"loss": 5.9842,
"mean_token_accuracy": 0.13427486047148704,
"num_tokens": 4499735.0,
"step": 2175
},
{
"entropy": 5.945842647552491,
"epoch": 0.19775036284470246,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998706535851657,
"loss": 5.9344,
"mean_token_accuracy": 0.13740400075912476,
"num_tokens": 4509500.0,
"step": 2180
},
{
"entropy": 5.9569385051727295,
"epoch": 0.19820391872278664,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998695542793129,
"loss": 5.9392,
"mean_token_accuracy": 0.13562830910086632,
"num_tokens": 4520431.0,
"step": 2185
},
{
"entropy": 5.98747034072876,
"epoch": 0.19865747460087083,
"grad_norm": 1.140625,
"learning_rate": 0.0004998684503231003,
"loss": 6.0309,
"mean_token_accuracy": 0.13986979350447654,
"num_tokens": 4532294.0,
"step": 2190
},
{
"entropy": 5.934312200546264,
"epoch": 0.199111030478955,
"grad_norm": 1.1484375,
"learning_rate": 0.000499867341716551,
"loss": 5.948,
"mean_token_accuracy": 0.1377948857843876,
"num_tokens": 4542488.0,
"step": 2195
},
{
"entropy": 5.966859626770019,
"epoch": 0.1995645863570392,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998662284596878,
"loss": 5.8828,
"mean_token_accuracy": 0.13661633133888246,
"num_tokens": 4552933.0,
"step": 2200
},
{
"entropy": 5.952669095993042,
"epoch": 0.20001814223512338,
"grad_norm": 1.09375,
"learning_rate": 0.0004998651105525335,
"loss": 5.8786,
"mean_token_accuracy": 0.1433972269296646,
"num_tokens": 4562373.0,
"step": 2205
},
{
"entropy": 5.883717727661133,
"epoch": 0.20047169811320756,
"grad_norm": 1.09375,
"learning_rate": 0.0004998639879951115,
"loss": 6.0106,
"mean_token_accuracy": 0.13773834556341172,
"num_tokens": 4572715.0,
"step": 2210
},
{
"entropy": 6.062313556671143,
"epoch": 0.2009252539912917,
"grad_norm": 1.125,
"learning_rate": 0.000499862860787445,
"loss": 5.8879,
"mean_token_accuracy": 0.13721229061484336,
"num_tokens": 4582782.0,
"step": 2215
},
{
"entropy": 5.980919361114502,
"epoch": 0.2013788098693759,
"grad_norm": 1.171875,
"learning_rate": 0.0004998617289295571,
"loss": 5.9939,
"mean_token_accuracy": 0.13345705717802048,
"num_tokens": 4594108.0,
"step": 2220
},
{
"entropy": 6.007939147949219,
"epoch": 0.20183236574746008,
"grad_norm": 1.3671875,
"learning_rate": 0.0004998605924214714,
"loss": 5.9503,
"mean_token_accuracy": 0.13709166496992112,
"num_tokens": 4604732.0,
"step": 2225
},
{
"entropy": 5.984397983551025,
"epoch": 0.20228592162554426,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998594512632114,
"loss": 6.0782,
"mean_token_accuracy": 0.12725423350930215,
"num_tokens": 4614998.0,
"step": 2230
},
{
"entropy": 5.951976251602173,
"epoch": 0.20273947750362845,
"grad_norm": 1.421875,
"learning_rate": 0.0004998583054548006,
"loss": 5.9087,
"mean_token_accuracy": 0.13761360049247742,
"num_tokens": 4624665.0,
"step": 2235
},
{
"entropy": 5.996311187744141,
"epoch": 0.20319303338171263,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998571549962628,
"loss": 5.9052,
"mean_token_accuracy": 0.14659312665462493,
"num_tokens": 4633731.0,
"step": 2240
},
{
"entropy": 5.99422254562378,
"epoch": 0.2036465892597968,
"grad_norm": 1.171875,
"learning_rate": 0.0004998559998876216,
"loss": 5.9297,
"mean_token_accuracy": 0.13845302835106849,
"num_tokens": 4643421.0,
"step": 2245
},
{
"entropy": 5.973086881637573,
"epoch": 0.204100145137881,
"grad_norm": 1.2890625,
"learning_rate": 0.0004998548401289012,
"loss": 6.0172,
"mean_token_accuracy": 0.13639747053384782,
"num_tokens": 4653795.0,
"step": 2250
},
{
"entropy": 6.01025357246399,
"epoch": 0.20455370101596518,
"grad_norm": 1.09375,
"learning_rate": 0.0004998536757201253,
"loss": 5.9909,
"mean_token_accuracy": 0.13689780980348587,
"num_tokens": 4664840.0,
"step": 2255
},
{
"entropy": 6.023299217224121,
"epoch": 0.20500725689404933,
"grad_norm": 1.3203125,
"learning_rate": 0.0004998525066613181,
"loss": 6.0004,
"mean_token_accuracy": 0.1407467246055603,
"num_tokens": 4675508.0,
"step": 2260
},
{
"entropy": 5.931134462356567,
"epoch": 0.20546081277213352,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998513329525038,
"loss": 5.9157,
"mean_token_accuracy": 0.14364449679851532,
"num_tokens": 4685995.0,
"step": 2265
},
{
"entropy": 6.031408452987671,
"epoch": 0.2059143686502177,
"grad_norm": 1.140625,
"learning_rate": 0.0004998501545937066,
"loss": 5.9397,
"mean_token_accuracy": 0.13526451364159583,
"num_tokens": 4696659.0,
"step": 2270
},
{
"entropy": 5.900203275680542,
"epoch": 0.20636792452830188,
"grad_norm": 1.1640625,
"learning_rate": 0.000499848971584951,
"loss": 5.8941,
"mean_token_accuracy": 0.14944595396518706,
"num_tokens": 4706438.0,
"step": 2275
},
{
"entropy": 5.911033678054809,
"epoch": 0.20682148040638607,
"grad_norm": 1.265625,
"learning_rate": 0.0004998477839262611,
"loss": 5.9105,
"mean_token_accuracy": 0.14088572859764098,
"num_tokens": 4716402.0,
"step": 2280
},
{
"entropy": 5.99247899055481,
"epoch": 0.20727503628447025,
"grad_norm": 1.109375,
"learning_rate": 0.000499846591617662,
"loss": 6.001,
"mean_token_accuracy": 0.1439184844493866,
"num_tokens": 4728488.0,
"step": 2285
},
{
"entropy": 5.915400600433349,
"epoch": 0.20772859216255443,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998453946591779,
"loss": 6.0241,
"mean_token_accuracy": 0.1419094666838646,
"num_tokens": 4738693.0,
"step": 2290
},
{
"entropy": 5.9869701862335205,
"epoch": 0.20818214804063861,
"grad_norm": 1.125,
"learning_rate": 0.0004998441930508339,
"loss": 5.9132,
"mean_token_accuracy": 0.1442976713180542,
"num_tokens": 4748895.0,
"step": 2295
},
{
"entropy": 6.020207309722901,
"epoch": 0.2086357039187228,
"grad_norm": 1.171875,
"learning_rate": 0.0004998429867926547,
"loss": 6.0203,
"mean_token_accuracy": 0.13464172631502153,
"num_tokens": 4760135.0,
"step": 2300
},
{
"entropy": 6.043022537231446,
"epoch": 0.20908925979680695,
"grad_norm": 1.171875,
"learning_rate": 0.0004998417758846651,
"loss": 6.0728,
"mean_token_accuracy": 0.14229417890310286,
"num_tokens": 4769935.0,
"step": 2305
},
{
"entropy": 6.088847494125366,
"epoch": 0.20954281567489114,
"grad_norm": 1.1953125,
"learning_rate": 0.0004998405603268903,
"loss": 5.9793,
"mean_token_accuracy": 0.14080933183431626,
"num_tokens": 4779520.0,
"step": 2310
},
{
"entropy": 5.910649347305298,
"epoch": 0.20999637155297532,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998393401193554,
"loss": 5.9004,
"mean_token_accuracy": 0.14728002548217772,
"num_tokens": 4791256.0,
"step": 2315
},
{
"entropy": 5.814291381835938,
"epoch": 0.2104499274310595,
"grad_norm": 1.171875,
"learning_rate": 0.0004998381152620857,
"loss": 5.8551,
"mean_token_accuracy": 0.14871017038822174,
"num_tokens": 4801334.0,
"step": 2320
},
{
"entropy": 5.910569524765014,
"epoch": 0.21090348330914369,
"grad_norm": 1.28125,
"learning_rate": 0.0004998368857551067,
"loss": 5.8394,
"mean_token_accuracy": 0.145636598020792,
"num_tokens": 4811361.0,
"step": 2325
},
{
"entropy": 5.91206111907959,
"epoch": 0.21135703918722787,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998356515984433,
"loss": 5.9207,
"mean_token_accuracy": 0.14299503937363625,
"num_tokens": 4821190.0,
"step": 2330
},
{
"entropy": 5.959600400924683,
"epoch": 0.21181059506531205,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998344127921213,
"loss": 5.9476,
"mean_token_accuracy": 0.14232189506292342,
"num_tokens": 4830694.0,
"step": 2335
},
{
"entropy": 5.944343614578247,
"epoch": 0.21226415094339623,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998331693361666,
"loss": 5.9274,
"mean_token_accuracy": 0.137859558314085,
"num_tokens": 4841344.0,
"step": 2340
},
{
"entropy": 5.9577583312988285,
"epoch": 0.21271770682148042,
"grad_norm": 1.265625,
"learning_rate": 0.0004998319212306046,
"loss": 5.8928,
"mean_token_accuracy": 0.14234141260385513,
"num_tokens": 4851161.0,
"step": 2345
},
{
"entropy": 5.885667610168457,
"epoch": 0.21317126269956457,
"grad_norm": 1.3984375,
"learning_rate": 0.000499830668475461,
"loss": 5.8742,
"mean_token_accuracy": 0.14236741438508033,
"num_tokens": 4860740.0,
"step": 2350
},
{
"entropy": 5.944697475433349,
"epoch": 0.21362481857764876,
"grad_norm": 1.2578125,
"learning_rate": 0.000499829411070762,
"loss": 5.9118,
"mean_token_accuracy": 0.14243801087141036,
"num_tokens": 4871733.0,
"step": 2355
},
{
"entropy": 5.945801448822022,
"epoch": 0.21407837445573294,
"grad_norm": 1.1875,
"learning_rate": 0.0004998281490165335,
"loss": 5.9669,
"mean_token_accuracy": 0.13472701087594033,
"num_tokens": 4883617.0,
"step": 2360
},
{
"entropy": 5.939307308197021,
"epoch": 0.21453193033381712,
"grad_norm": 1.203125,
"learning_rate": 0.0004998268823128016,
"loss": 5.9302,
"mean_token_accuracy": 0.1350124605000019,
"num_tokens": 4894519.0,
"step": 2365
},
{
"entropy": 5.977178955078125,
"epoch": 0.2149854862119013,
"grad_norm": 1.328125,
"learning_rate": 0.0004998256109595923,
"loss": 5.8817,
"mean_token_accuracy": 0.1414977490901947,
"num_tokens": 4904965.0,
"step": 2370
},
{
"entropy": 5.86356201171875,
"epoch": 0.2154390420899855,
"grad_norm": 1.3359375,
"learning_rate": 0.0004998243349569323,
"loss": 5.7913,
"mean_token_accuracy": 0.1485743671655655,
"num_tokens": 4914618.0,
"step": 2375
},
{
"entropy": 5.848833799362183,
"epoch": 0.21589259796806967,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998230543048477,
"loss": 5.8575,
"mean_token_accuracy": 0.14849701523780823,
"num_tokens": 4924341.0,
"step": 2380
},
{
"entropy": 5.913916826248169,
"epoch": 0.21634615384615385,
"grad_norm": 1.125,
"learning_rate": 0.000499821769003365,
"loss": 5.8344,
"mean_token_accuracy": 0.14421416819095612,
"num_tokens": 4934231.0,
"step": 2385
},
{
"entropy": 5.868072605133056,
"epoch": 0.21679970972423804,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998204790525109,
"loss": 5.8489,
"mean_token_accuracy": 0.1429114505648613,
"num_tokens": 4944303.0,
"step": 2390
},
{
"entropy": 5.908360385894776,
"epoch": 0.2172532656023222,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998191844523119,
"loss": 5.9351,
"mean_token_accuracy": 0.14515570178627968,
"num_tokens": 4954571.0,
"step": 2395
},
{
"entropy": 5.965419578552246,
"epoch": 0.21770682148040638,
"grad_norm": 1.203125,
"learning_rate": 0.000499817885202795,
"loss": 5.8968,
"mean_token_accuracy": 0.14096556901931762,
"num_tokens": 4964418.0,
"step": 2400
},
{
"entropy": 5.979696035385132,
"epoch": 0.21816037735849056,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998165813039869,
"loss": 5.8835,
"mean_token_accuracy": 0.14295963644981385,
"num_tokens": 4975036.0,
"step": 2405
},
{
"entropy": 5.873979759216309,
"epoch": 0.21861393323657474,
"grad_norm": 1.3359375,
"learning_rate": 0.0004998152727559145,
"loss": 5.9914,
"mean_token_accuracy": 0.13900846913456916,
"num_tokens": 4985272.0,
"step": 2410
},
{
"entropy": 6.005450963973999,
"epoch": 0.21906748911465893,
"grad_norm": 1.0859375,
"learning_rate": 0.000499813959558605,
"loss": 5.9213,
"mean_token_accuracy": 0.14803757071495055,
"num_tokens": 4995187.0,
"step": 2415
},
{
"entropy": 5.996357822418213,
"epoch": 0.2195210449927431,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998126417120856,
"loss": 5.9681,
"mean_token_accuracy": 0.14800111576914787,
"num_tokens": 5005013.0,
"step": 2420
},
{
"entropy": 6.017085218429566,
"epoch": 0.2199746008708273,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998113192163835,
"loss": 5.9755,
"mean_token_accuracy": 0.13796148598194122,
"num_tokens": 5016789.0,
"step": 2425
},
{
"entropy": 5.947689533233643,
"epoch": 0.22042815674891147,
"grad_norm": 1.0546875,
"learning_rate": 0.000499809992071526,
"loss": 5.9331,
"mean_token_accuracy": 0.14225904271006584,
"num_tokens": 5026913.0,
"step": 2430
},
{
"entropy": 5.922537040710449,
"epoch": 0.22088171262699566,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998086602775406,
"loss": 6.0365,
"mean_token_accuracy": 0.13978781476616858,
"num_tokens": 5037501.0,
"step": 2435
},
{
"entropy": 5.976426696777343,
"epoch": 0.2213352685050798,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998073238344547,
"loss": 5.8812,
"mean_token_accuracy": 0.13661929443478585,
"num_tokens": 5048478.0,
"step": 2440
},
{
"entropy": 5.931379604339599,
"epoch": 0.221788824383164,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998059827422962,
"loss": 5.9411,
"mean_token_accuracy": 0.1370401792228222,
"num_tokens": 5059118.0,
"step": 2445
},
{
"entropy": 5.984925699234009,
"epoch": 0.22224238026124818,
"grad_norm": 1.171875,
"learning_rate": 0.0004998046370010926,
"loss": 5.8956,
"mean_token_accuracy": 0.14508832097053528,
"num_tokens": 5070180.0,
"step": 2450
},
{
"entropy": 5.882190847396851,
"epoch": 0.22269593613933236,
"grad_norm": 1.2578125,
"learning_rate": 0.0004998032866108718,
"loss": 5.8958,
"mean_token_accuracy": 0.14208046942949296,
"num_tokens": 5079912.0,
"step": 2455
},
{
"entropy": 5.956144332885742,
"epoch": 0.22314949201741655,
"grad_norm": 1.140625,
"learning_rate": 0.0004998019315716618,
"loss": 5.9767,
"mean_token_accuracy": 0.14252351969480515,
"num_tokens": 5090549.0,
"step": 2460
},
{
"entropy": 5.902889060974121,
"epoch": 0.22360304789550073,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998005718834905,
"loss": 5.8293,
"mean_token_accuracy": 0.15000268071889877,
"num_tokens": 5100880.0,
"step": 2465
},
{
"entropy": 5.912509965896606,
"epoch": 0.2240566037735849,
"grad_norm": 1.140625,
"learning_rate": 0.0004997992075463862,
"loss": 5.9424,
"mean_token_accuracy": 0.1433528684079647,
"num_tokens": 5111796.0,
"step": 2470
},
{
"entropy": 5.958014440536499,
"epoch": 0.2245101596516691,
"grad_norm": 1.234375,
"learning_rate": 0.0004997978385603769,
"loss": 5.935,
"mean_token_accuracy": 0.14096981883049012,
"num_tokens": 5123468.0,
"step": 2475
},
{
"entropy": 5.9337944984436035,
"epoch": 0.22496371552975328,
"grad_norm": 1.171875,
"learning_rate": 0.0004997964649254911,
"loss": 5.8641,
"mean_token_accuracy": 0.1491854891180992,
"num_tokens": 5132925.0,
"step": 2480
},
{
"entropy": 6.011470174789428,
"epoch": 0.22541727140783743,
"grad_norm": 1.28125,
"learning_rate": 0.0004997950866417572,
"loss": 5.9448,
"mean_token_accuracy": 0.1389354422688484,
"num_tokens": 5143445.0,
"step": 2485
},
{
"entropy": 5.92656307220459,
"epoch": 0.22587082728592162,
"grad_norm": 1.3359375,
"learning_rate": 0.0004997937037092034,
"loss": 5.8551,
"mean_token_accuracy": 0.13968046829104425,
"num_tokens": 5153176.0,
"step": 2490
},
{
"entropy": 5.953498363494873,
"epoch": 0.2263243831640058,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997923161278587,
"loss": 5.9041,
"mean_token_accuracy": 0.1428646430373192,
"num_tokens": 5162765.0,
"step": 2495
},
{
"entropy": 5.978867244720459,
"epoch": 0.22677793904208998,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997909238977514,
"loss": 5.9606,
"mean_token_accuracy": 0.14364035278558732,
"num_tokens": 5173390.0,
"step": 2500
},
{
"entropy": 5.906553840637207,
"epoch": 0.22723149492017417,
"grad_norm": 1.1328125,
"learning_rate": 0.0004997895270189108,
"loss": 5.8628,
"mean_token_accuracy": 0.1373254232108593,
"num_tokens": 5184212.0,
"step": 2505
},
{
"entropy": 5.853435754776001,
"epoch": 0.22768505079825835,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997881254913653,
"loss": 5.8602,
"mean_token_accuracy": 0.13991171047091483,
"num_tokens": 5194311.0,
"step": 2510
},
{
"entropy": 5.888242435455322,
"epoch": 0.22813860667634253,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997867193151441,
"loss": 5.8408,
"mean_token_accuracy": 0.1456891916692257,
"num_tokens": 5204708.0,
"step": 2515
},
{
"entropy": 5.830229997634888,
"epoch": 0.22859216255442671,
"grad_norm": 1.21875,
"learning_rate": 0.0004997853084902763,
"loss": 5.843,
"mean_token_accuracy": 0.14071134552359582,
"num_tokens": 5214566.0,
"step": 2520
},
{
"entropy": 5.903006267547608,
"epoch": 0.2290457184325109,
"grad_norm": 1.1953125,
"learning_rate": 0.0004997838930167911,
"loss": 5.8779,
"mean_token_accuracy": 0.14581192210316657,
"num_tokens": 5224762.0,
"step": 2525
},
{
"entropy": 5.952599906921387,
"epoch": 0.22949927431059505,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997824728947176,
"loss": 5.9372,
"mean_token_accuracy": 0.13909635022282602,
"num_tokens": 5235146.0,
"step": 2530
},
{
"entropy": 5.936500406265258,
"epoch": 0.22995283018867924,
"grad_norm": 1.1875,
"learning_rate": 0.0004997810481240853,
"loss": 5.9115,
"mean_token_accuracy": 0.1376205012202263,
"num_tokens": 5243867.0,
"step": 2535
},
{
"entropy": 5.910394525527954,
"epoch": 0.23040638606676342,
"grad_norm": 1.21875,
"learning_rate": 0.0004997796187049238,
"loss": 5.831,
"mean_token_accuracy": 0.14185473769903184,
"num_tokens": 5254279.0,
"step": 2540
},
{
"entropy": 6.032854747772217,
"epoch": 0.2308599419448476,
"grad_norm": 1.2421875,
"learning_rate": 0.0004997781846372623,
"loss": 5.9444,
"mean_token_accuracy": 0.1413105957210064,
"num_tokens": 5264769.0,
"step": 2545
},
{
"entropy": 5.892349815368652,
"epoch": 0.23131349782293179,
"grad_norm": 1.2734375,
"learning_rate": 0.0004997767459211307,
"loss": 5.8757,
"mean_token_accuracy": 0.1384491316974163,
"num_tokens": 5274291.0,
"step": 2550
},
{
"entropy": 5.9222578525543215,
"epoch": 0.23176705370101597,
"grad_norm": 1.4453125,
"learning_rate": 0.0004997753025565587,
"loss": 5.8226,
"mean_token_accuracy": 0.14413735643029213,
"num_tokens": 5284366.0,
"step": 2555
},
{
"entropy": 5.923757791519165,
"epoch": 0.23222060957910015,
"grad_norm": 1.3515625,
"learning_rate": 0.0004997738545435763,
"loss": 5.9218,
"mean_token_accuracy": 0.14310500398278236,
"num_tokens": 5294494.0,
"step": 2560
},
{
"entropy": 5.835726594924926,
"epoch": 0.23267416545718433,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997724018822133,
"loss": 5.9402,
"mean_token_accuracy": 0.14283815771341324,
"num_tokens": 5305606.0,
"step": 2565
},
{
"entropy": 5.938771772384643,
"epoch": 0.23312772133526852,
"grad_norm": 1.4296875,
"learning_rate": 0.0004997709445724996,
"loss": 5.9326,
"mean_token_accuracy": 0.13846345022320747,
"num_tokens": 5315677.0,
"step": 2570
},
{
"entropy": 6.001786565780639,
"epoch": 0.23358127721335267,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997694826144656,
"loss": 5.987,
"mean_token_accuracy": 0.13663317039608955,
"num_tokens": 5326453.0,
"step": 2575
},
{
"entropy": 5.945255422592163,
"epoch": 0.23403483309143686,
"grad_norm": 1.015625,
"learning_rate": 0.0004997680160081414,
"loss": 5.8746,
"mean_token_accuracy": 0.14533428698778153,
"num_tokens": 5337591.0,
"step": 2580
},
{
"entropy": 5.880674076080322,
"epoch": 0.23448838896952104,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997665447535573,
"loss": 5.8792,
"mean_token_accuracy": 0.14644460529088973,
"num_tokens": 5349400.0,
"step": 2585
},
{
"entropy": 5.976725244522095,
"epoch": 0.23494194484760522,
"grad_norm": 1.296875,
"learning_rate": 0.0004997650688507438,
"loss": 5.8647,
"mean_token_accuracy": 0.14361676946282387,
"num_tokens": 5359984.0,
"step": 2590
},
{
"entropy": 5.860870790481568,
"epoch": 0.2353955007256894,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997635882997314,
"loss": 5.9341,
"mean_token_accuracy": 0.14296606034040452,
"num_tokens": 5369254.0,
"step": 2595
},
{
"entropy": 5.9094048023223875,
"epoch": 0.2358490566037736,
"grad_norm": 1.3984375,
"learning_rate": 0.0004997621031005507,
"loss": 5.8609,
"mean_token_accuracy": 0.14291658252477646,
"num_tokens": 5379648.0,
"step": 2600
},
{
"entropy": 5.866974496841431,
"epoch": 0.23630261248185777,
"grad_norm": 1.3046875,
"learning_rate": 0.0004997606132532325,
"loss": 5.8309,
"mean_token_accuracy": 0.14787738099694253,
"num_tokens": 5390321.0,
"step": 2605
},
{
"entropy": 5.867215776443482,
"epoch": 0.23675616835994195,
"grad_norm": 1.3359375,
"learning_rate": 0.0004997591187578075,
"loss": 5.8596,
"mean_token_accuracy": 0.14872283935546876,
"num_tokens": 5401078.0,
"step": 2610
},
{
"entropy": 5.8993391513824465,
"epoch": 0.23720972423802614,
"grad_norm": 1.390625,
"learning_rate": 0.0004997576196143067,
"loss": 5.8513,
"mean_token_accuracy": 0.1461376279592514,
"num_tokens": 5411977.0,
"step": 2615
},
{
"entropy": 5.88139500617981,
"epoch": 0.2376632801161103,
"grad_norm": 1.3203125,
"learning_rate": 0.000499756115822761,
"loss": 5.841,
"mean_token_accuracy": 0.13903868198394775,
"num_tokens": 5422967.0,
"step": 2620
},
{
"entropy": 5.967788743972778,
"epoch": 0.23811683599419448,
"grad_norm": 1.265625,
"learning_rate": 0.0004997546073832015,
"loss": 5.8848,
"mean_token_accuracy": 0.15244564265012742,
"num_tokens": 5432260.0,
"step": 2625
},
{
"entropy": 5.894135141372681,
"epoch": 0.23857039187227866,
"grad_norm": 1.53125,
"learning_rate": 0.0004997530942956595,
"loss": 5.8465,
"mean_token_accuracy": 0.14635367840528488,
"num_tokens": 5441706.0,
"step": 2630
},
{
"entropy": 5.897868919372558,
"epoch": 0.23902394775036284,
"grad_norm": 1.234375,
"learning_rate": 0.0004997515765601662,
"loss": 5.8558,
"mean_token_accuracy": 0.15220582336187363,
"num_tokens": 5452397.0,
"step": 2635
},
{
"entropy": 5.880718660354614,
"epoch": 0.23947750362844702,
"grad_norm": 1.109375,
"learning_rate": 0.0004997500541767531,
"loss": 5.8202,
"mean_token_accuracy": 0.14415100812911988,
"num_tokens": 5463762.0,
"step": 2640
},
{
"entropy": 5.870383977890015,
"epoch": 0.2399310595065312,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997485271454516,
"loss": 5.8788,
"mean_token_accuracy": 0.13774784579873084,
"num_tokens": 5474622.0,
"step": 2645
},
{
"entropy": 5.96224799156189,
"epoch": 0.2403846153846154,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997469954662933,
"loss": 5.8695,
"mean_token_accuracy": 0.14564383402466774,
"num_tokens": 5484697.0,
"step": 2650
},
{
"entropy": 5.851238155364991,
"epoch": 0.24083817126269957,
"grad_norm": 1.2734375,
"learning_rate": 0.0004997454591393097,
"loss": 5.8667,
"mean_token_accuracy": 0.1427861526608467,
"num_tokens": 5496433.0,
"step": 2655
},
{
"entropy": 5.973064804077149,
"epoch": 0.24129172714078376,
"grad_norm": 1.09375,
"learning_rate": 0.0004997439181645329,
"loss": 5.97,
"mean_token_accuracy": 0.14119771718978882,
"num_tokens": 5507917.0,
"step": 2660
},
{
"entropy": 6.047526073455811,
"epoch": 0.2417452830188679,
"grad_norm": 1.21875,
"learning_rate": 0.0004997423725419945,
"loss": 5.9544,
"mean_token_accuracy": 0.1415584534406662,
"num_tokens": 5519159.0,
"step": 2665
},
{
"entropy": 5.842634201049805,
"epoch": 0.2421988388969521,
"grad_norm": 1.3671875,
"learning_rate": 0.0004997408222717267,
"loss": 5.9242,
"mean_token_accuracy": 0.14665919542312622,
"num_tokens": 5529180.0,
"step": 2670
},
{
"entropy": 5.897845602035522,
"epoch": 0.24265239477503628,
"grad_norm": 1.1328125,
"learning_rate": 0.0004997392673537612,
"loss": 5.8381,
"mean_token_accuracy": 0.15111086070537566,
"num_tokens": 5539363.0,
"step": 2675
},
{
"entropy": 5.8370285987854,
"epoch": 0.24310595065312046,
"grad_norm": 1.234375,
"learning_rate": 0.0004997377077881306,
"loss": 5.946,
"mean_token_accuracy": 0.14330666437745093,
"num_tokens": 5549841.0,
"step": 2680
},
{
"entropy": 5.904944038391113,
"epoch": 0.24355950653120464,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997361435748668,
"loss": 5.8349,
"mean_token_accuracy": 0.14796256124973298,
"num_tokens": 5560299.0,
"step": 2685
},
{
"entropy": 5.9934006690979,
"epoch": 0.24401306240928883,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997345747140024,
"loss": 5.8923,
"mean_token_accuracy": 0.14425792098045348,
"num_tokens": 5571081.0,
"step": 2690
},
{
"entropy": 5.856541490554809,
"epoch": 0.244466618287373,
"grad_norm": 1.171875,
"learning_rate": 0.0004997330012055695,
"loss": 5.8447,
"mean_token_accuracy": 0.149576073884964,
"num_tokens": 5581389.0,
"step": 2695
},
{
"entropy": 5.9023360252380375,
"epoch": 0.2449201741654572,
"grad_norm": 1.265625,
"learning_rate": 0.0004997314230496011,
"loss": 5.8683,
"mean_token_accuracy": 0.14250419288873672,
"num_tokens": 5592347.0,
"step": 2700
},
{
"entropy": 5.878292608261108,
"epoch": 0.24537373004354138,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997298402461296,
"loss": 5.8384,
"mean_token_accuracy": 0.14773763567209244,
"num_tokens": 5603411.0,
"step": 2705
},
{
"entropy": 5.920032787322998,
"epoch": 0.24582728592162553,
"grad_norm": 1.15625,
"learning_rate": 0.0004997282527951876,
"loss": 5.844,
"mean_token_accuracy": 0.14941513240337373,
"num_tokens": 5614615.0,
"step": 2710
},
{
"entropy": 5.869028520584107,
"epoch": 0.24628084179970972,
"grad_norm": 1.390625,
"learning_rate": 0.0004997266606968081,
"loss": 5.8137,
"mean_token_accuracy": 0.1474539414048195,
"num_tokens": 5624582.0,
"step": 2715
},
{
"entropy": 5.876409101486206,
"epoch": 0.2467343976777939,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997250639510241,
"loss": 5.8213,
"mean_token_accuracy": 0.14212239235639573,
"num_tokens": 5634310.0,
"step": 2720
},
{
"entropy": 5.889693927764893,
"epoch": 0.24718795355587808,
"grad_norm": 1.328125,
"learning_rate": 0.0004997234625578684,
"loss": 5.9165,
"mean_token_accuracy": 0.14168269485235213,
"num_tokens": 5644018.0,
"step": 2725
},
{
"entropy": 5.872516393661499,
"epoch": 0.24764150943396226,
"grad_norm": 1.3125,
"learning_rate": 0.0004997218565173742,
"loss": 5.8501,
"mean_token_accuracy": 0.1477284774184227,
"num_tokens": 5653592.0,
"step": 2730
},
{
"entropy": 5.970296859741211,
"epoch": 0.24809506531204645,
"grad_norm": 1.3671875,
"learning_rate": 0.0004997202458295748,
"loss": 5.8983,
"mean_token_accuracy": 0.14910974949598313,
"num_tokens": 5662812.0,
"step": 2735
},
{
"entropy": 5.877613735198975,
"epoch": 0.24854862119013063,
"grad_norm": 1.140625,
"learning_rate": 0.0004997186304945034,
"loss": 5.8589,
"mean_token_accuracy": 0.14394548833370208,
"num_tokens": 5672939.0,
"step": 2740
},
{
"entropy": 5.869716453552246,
"epoch": 0.24900217706821481,
"grad_norm": 1.171875,
"learning_rate": 0.0004997170105121935,
"loss": 5.8577,
"mean_token_accuracy": 0.14621407687664031,
"num_tokens": 5683350.0,
"step": 2745
},
{
"entropy": 6.0468605041503904,
"epoch": 0.249455732946299,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997153858826786,
"loss": 5.9484,
"mean_token_accuracy": 0.13753940612077714,
"num_tokens": 5694516.0,
"step": 2750
},
{
"entropy": 5.836617755889892,
"epoch": 0.24990928882438315,
"grad_norm": 1.203125,
"learning_rate": 0.0004997137566059922,
"loss": 5.7658,
"mean_token_accuracy": 0.1503409966826439,
"num_tokens": 5705443.0,
"step": 2755
},
{
"entropy": 5.8198834419250485,
"epoch": 0.25036284470246734,
"grad_norm": 1.234375,
"learning_rate": 0.0004997121226821681,
"loss": 5.8859,
"mean_token_accuracy": 0.13909180611371993,
"num_tokens": 5715933.0,
"step": 2760
},
{
"entropy": 5.972828674316406,
"epoch": 0.25081640058055155,
"grad_norm": 1.2578125,
"learning_rate": 0.00049971048411124,
"loss": 5.9315,
"mean_token_accuracy": 0.1365823417901993,
"num_tokens": 5726630.0,
"step": 2765
},
{
"entropy": 6.025601530075074,
"epoch": 0.2512699564586357,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997088408932418,
"loss": 5.8606,
"mean_token_accuracy": 0.14626222401857375,
"num_tokens": 5736890.0,
"step": 2770
},
{
"entropy": 5.7838927745819095,
"epoch": 0.25172351233671986,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997071930282076,
"loss": 5.7814,
"mean_token_accuracy": 0.15587268024683,
"num_tokens": 5747042.0,
"step": 2775
},
{
"entropy": 5.836866760253907,
"epoch": 0.25217706821480407,
"grad_norm": 1.3359375,
"learning_rate": 0.0004997055405161712,
"loss": 5.904,
"mean_token_accuracy": 0.14357635378837585,
"num_tokens": 5757367.0,
"step": 2780
},
{
"entropy": 5.944021129608155,
"epoch": 0.2526306240928882,
"grad_norm": 1.1953125,
"learning_rate": 0.0004997038833571671,
"loss": 5.9615,
"mean_token_accuracy": 0.1422647625207901,
"num_tokens": 5769557.0,
"step": 2785
},
{
"entropy": 5.940666151046753,
"epoch": 0.25308417997097243,
"grad_norm": 1.34375,
"learning_rate": 0.0004997022215512295,
"loss": 5.8503,
"mean_token_accuracy": 0.15143156200647354,
"num_tokens": 5780615.0,
"step": 2790
},
{
"entropy": 5.937691497802734,
"epoch": 0.2535377358490566,
"grad_norm": 1.15625,
"learning_rate": 0.0004997005550983925,
"loss": 5.827,
"mean_token_accuracy": 0.14720471799373627,
"num_tokens": 5791271.0,
"step": 2795
},
{
"entropy": 5.830975341796875,
"epoch": 0.2539912917271408,
"grad_norm": 1.3046875,
"learning_rate": 0.000499698883998691,
"loss": 5.7719,
"mean_token_accuracy": 0.14439669102430344,
"num_tokens": 5801117.0,
"step": 2800
},
{
"entropy": 5.828149271011353,
"epoch": 0.25444484760522496,
"grad_norm": 1.2734375,
"learning_rate": 0.0004996972082521591,
"loss": 5.7958,
"mean_token_accuracy": 0.1420363336801529,
"num_tokens": 5811696.0,
"step": 2805
},
{
"entropy": 5.868321561813355,
"epoch": 0.25489840348330917,
"grad_norm": 1.25,
"learning_rate": 0.0004996955278588319,
"loss": 5.8318,
"mean_token_accuracy": 0.15537209510803224,
"num_tokens": 5822207.0,
"step": 2810
},
{
"entropy": 5.808701229095459,
"epoch": 0.2553519593613933,
"grad_norm": 1.4921875,
"learning_rate": 0.0004996938428187437,
"loss": 5.7714,
"mean_token_accuracy": 0.15279726982116698,
"num_tokens": 5831927.0,
"step": 2815
},
{
"entropy": 5.886981678009033,
"epoch": 0.2558055152394775,
"grad_norm": 1.1875,
"learning_rate": 0.0004996921531319297,
"loss": 5.8415,
"mean_token_accuracy": 0.14455378502607347,
"num_tokens": 5842093.0,
"step": 2820
},
{
"entropy": 5.86587119102478,
"epoch": 0.2562590711175617,
"grad_norm": 1.34375,
"learning_rate": 0.0004996904587984246,
"loss": 5.7803,
"mean_token_accuracy": 0.15339562743902208,
"num_tokens": 5850551.0,
"step": 2825
},
{
"entropy": 5.764548873901367,
"epoch": 0.25671262699564584,
"grad_norm": 1.3046875,
"learning_rate": 0.0004996887598182636,
"loss": 5.7714,
"mean_token_accuracy": 0.1484563708305359,
"num_tokens": 5860651.0,
"step": 2830
},
{
"entropy": 5.840695333480835,
"epoch": 0.25716618287373005,
"grad_norm": 1.2734375,
"learning_rate": 0.0004996870561914819,
"loss": 5.7878,
"mean_token_accuracy": 0.14319953173398972,
"num_tokens": 5871500.0,
"step": 2835
},
{
"entropy": 5.857464742660523,
"epoch": 0.2576197387518142,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996853479181145,
"loss": 5.7638,
"mean_token_accuracy": 0.1445399969816208,
"num_tokens": 5881523.0,
"step": 2840
},
{
"entropy": 5.928007984161377,
"epoch": 0.2580732946298984,
"grad_norm": 1.2578125,
"learning_rate": 0.0004996836349981969,
"loss": 5.9088,
"mean_token_accuracy": 0.14473725184798242,
"num_tokens": 5891162.0,
"step": 2845
},
{
"entropy": 5.846505928039551,
"epoch": 0.2585268505079826,
"grad_norm": 1.3828125,
"learning_rate": 0.0004996819174317645,
"loss": 5.7977,
"mean_token_accuracy": 0.14666571021080016,
"num_tokens": 5901818.0,
"step": 2850
},
{
"entropy": 5.829140758514404,
"epoch": 0.2589804063860668,
"grad_norm": 1.296875,
"learning_rate": 0.0004996801952188528,
"loss": 5.895,
"mean_token_accuracy": 0.14497522860765458,
"num_tokens": 5913889.0,
"step": 2855
},
{
"entropy": 5.888280963897705,
"epoch": 0.25943396226415094,
"grad_norm": 1.328125,
"learning_rate": 0.0004996784683594973,
"loss": 5.8485,
"mean_token_accuracy": 0.14567365497350693,
"num_tokens": 5923429.0,
"step": 2860
},
{
"entropy": 5.998221731185913,
"epoch": 0.2598875181422351,
"grad_norm": 1.3125,
"learning_rate": 0.0004996767368537338,
"loss": 5.8844,
"mean_token_accuracy": 0.14624838829040526,
"num_tokens": 5933847.0,
"step": 2865
},
{
"entropy": 5.8100090503692625,
"epoch": 0.2603410740203193,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996750007015983,
"loss": 5.7683,
"mean_token_accuracy": 0.14520363360643387,
"num_tokens": 5943874.0,
"step": 2870
},
{
"entropy": 5.848591136932373,
"epoch": 0.26079462989840346,
"grad_norm": 1.1875,
"learning_rate": 0.0004996732599031264,
"loss": 5.8297,
"mean_token_accuracy": 0.14542661756277084,
"num_tokens": 5953653.0,
"step": 2875
},
{
"entropy": 5.917650747299194,
"epoch": 0.2612481857764877,
"grad_norm": 1.265625,
"learning_rate": 0.0004996715144583543,
"loss": 5.7734,
"mean_token_accuracy": 0.14755018949508666,
"num_tokens": 5964039.0,
"step": 2880
},
{
"entropy": 5.900532245635986,
"epoch": 0.26170174165457183,
"grad_norm": 1.2265625,
"learning_rate": 0.000499669764367318,
"loss": 5.9012,
"mean_token_accuracy": 0.14746888726949692,
"num_tokens": 5974083.0,
"step": 2885
},
{
"entropy": 5.8273969173431395,
"epoch": 0.26215529753265604,
"grad_norm": 1.25,
"learning_rate": 0.0004996680096300537,
"loss": 5.8462,
"mean_token_accuracy": 0.14346672594547272,
"num_tokens": 5984629.0,
"step": 2890
},
{
"entropy": 5.839718961715699,
"epoch": 0.2626088534107402,
"grad_norm": 1.5,
"learning_rate": 0.0004996662502465978,
"loss": 5.7524,
"mean_token_accuracy": 0.16117057502269744,
"num_tokens": 5994295.0,
"step": 2895
},
{
"entropy": 5.838645267486572,
"epoch": 0.2630624092888244,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996644862169864,
"loss": 5.7866,
"mean_token_accuracy": 0.14446460008621215,
"num_tokens": 6004156.0,
"step": 2900
},
{
"entropy": 5.901477813720703,
"epoch": 0.26351596516690856,
"grad_norm": 1.28125,
"learning_rate": 0.0004996627175412563,
"loss": 5.8837,
"mean_token_accuracy": 0.14486180245876312,
"num_tokens": 6014188.0,
"step": 2905
},
{
"entropy": 5.949041032791138,
"epoch": 0.2639695210449927,
"grad_norm": 1.21875,
"learning_rate": 0.000499660944219444,
"loss": 5.9405,
"mean_token_accuracy": 0.1410454109311104,
"num_tokens": 6024879.0,
"step": 2910
},
{
"entropy": 5.967418909072876,
"epoch": 0.2644230769230769,
"grad_norm": 1.125,
"learning_rate": 0.0004996591662515861,
"loss": 5.9944,
"mean_token_accuracy": 0.13865989670157433,
"num_tokens": 6036374.0,
"step": 2915
},
{
"entropy": 5.901403713226318,
"epoch": 0.2648766328011611,
"grad_norm": 1.3671875,
"learning_rate": 0.0004996573836377193,
"loss": 5.8373,
"mean_token_accuracy": 0.14499613642692566,
"num_tokens": 6046432.0,
"step": 2920
},
{
"entropy": 5.943468523025513,
"epoch": 0.2653301886792453,
"grad_norm": 1.2890625,
"learning_rate": 0.0004996555963778806,
"loss": 5.8592,
"mean_token_accuracy": 0.14138642624020575,
"num_tokens": 6056864.0,
"step": 2925
},
{
"entropy": 5.995022583007812,
"epoch": 0.26578374455732945,
"grad_norm": 1.3671875,
"learning_rate": 0.000499653804472107,
"loss": 5.8844,
"mean_token_accuracy": 0.15080390572547914,
"num_tokens": 6067294.0,
"step": 2930
},
{
"entropy": 5.905005168914795,
"epoch": 0.26623730043541366,
"grad_norm": 1.3671875,
"learning_rate": 0.0004996520079204353,
"loss": 5.8871,
"mean_token_accuracy": 0.14075151458382607,
"num_tokens": 6077587.0,
"step": 2935
},
{
"entropy": 5.918993663787842,
"epoch": 0.2666908563134978,
"grad_norm": 1.28125,
"learning_rate": 0.000499650206722903,
"loss": 5.8385,
"mean_token_accuracy": 0.14215887933969498,
"num_tokens": 6089306.0,
"step": 2940
},
{
"entropy": 5.827713394165039,
"epoch": 0.267144412191582,
"grad_norm": 1.578125,
"learning_rate": 0.0004996484008795471,
"loss": 5.7875,
"mean_token_accuracy": 0.1460879400372505,
"num_tokens": 6099651.0,
"step": 2945
},
{
"entropy": 5.838403844833374,
"epoch": 0.2675979680696662,
"grad_norm": 1.34375,
"learning_rate": 0.000499646590390405,
"loss": 5.7932,
"mean_token_accuracy": 0.14292585104703903,
"num_tokens": 6110063.0,
"step": 2950
},
{
"entropy": 5.874997091293335,
"epoch": 0.26805152394775034,
"grad_norm": 1.234375,
"learning_rate": 0.0004996447752555142,
"loss": 5.8194,
"mean_token_accuracy": 0.1502259336411953,
"num_tokens": 6120031.0,
"step": 2955
},
{
"entropy": 5.802283811569214,
"epoch": 0.26850507982583455,
"grad_norm": 1.171875,
"learning_rate": 0.0004996429554749122,
"loss": 5.8655,
"mean_token_accuracy": 0.15023885816335678,
"num_tokens": 6131274.0,
"step": 2960
},
{
"entropy": 5.964536857604981,
"epoch": 0.2689586357039187,
"grad_norm": 1.2109375,
"learning_rate": 0.0004996411310486367,
"loss": 5.8932,
"mean_token_accuracy": 0.1471738301217556,
"num_tokens": 6140878.0,
"step": 2965
},
{
"entropy": 5.937205791473389,
"epoch": 0.2694121915820029,
"grad_norm": 1.1875,
"learning_rate": 0.0004996393019767253,
"loss": 5.8875,
"mean_token_accuracy": 0.14035830050706863,
"num_tokens": 6152229.0,
"step": 2970
},
{
"entropy": 5.76689043045044,
"epoch": 0.26986574746008707,
"grad_norm": 1.5703125,
"learning_rate": 0.0004996374682592158,
"loss": 5.7879,
"mean_token_accuracy": 0.14796984195709229,
"num_tokens": 6162311.0,
"step": 2975
},
{
"entropy": 5.934810733795166,
"epoch": 0.2703193033381713,
"grad_norm": 1.2890625,
"learning_rate": 0.0004996356298961463,
"loss": 5.9202,
"mean_token_accuracy": 0.14373642206192017,
"num_tokens": 6172806.0,
"step": 2980
},
{
"entropy": 5.891864681243897,
"epoch": 0.27077285921625543,
"grad_norm": 1.359375,
"learning_rate": 0.0004996337868875549,
"loss": 5.879,
"mean_token_accuracy": 0.14369452446699144,
"num_tokens": 6183479.0,
"step": 2985
},
{
"entropy": 5.89184536933899,
"epoch": 0.27122641509433965,
"grad_norm": 1.2421875,
"learning_rate": 0.0004996319392334792,
"loss": 5.8281,
"mean_token_accuracy": 0.14598268270492554,
"num_tokens": 6194201.0,
"step": 2990
},
{
"entropy": 5.824180364608765,
"epoch": 0.2716799709724238,
"grad_norm": 1.2734375,
"learning_rate": 0.000499630086933958,
"loss": 5.7868,
"mean_token_accuracy": 0.14525872468948364,
"num_tokens": 6204774.0,
"step": 2995
},
{
"entropy": 5.86897759437561,
"epoch": 0.27213352685050796,
"grad_norm": 1.34375,
"learning_rate": 0.0004996282299890292,
"loss": 5.7866,
"mean_token_accuracy": 0.1425388991832733,
"num_tokens": 6215056.0,
"step": 3000
},
{
"epoch": 0.27213352685050796,
"eval_entropy": 5.760963649555569,
"eval_loss": 5.879345417022705,
"eval_mean_token_accuracy": 0.15095411169330053,
"eval_num_tokens": 6215056.0,
"eval_runtime": 26.6633,
"eval_samples_per_second": 1326.167,
"eval_steps_per_second": 165.771,
"step": 3000
},
{
"entropy": 5.954884576797485,
"epoch": 0.27258708272859217,
"grad_norm": 1.28125,
"learning_rate": 0.0004996263683987316,
"loss": 5.8977,
"mean_token_accuracy": 0.14913439005613327,
"num_tokens": 6226264.0,
"step": 3005
},
{
"entropy": 5.886638212203979,
"epoch": 0.2730406386066763,
"grad_norm": 1.21875,
"learning_rate": 0.0004996245021631033,
"loss": 5.8156,
"mean_token_accuracy": 0.14778477251529692,
"num_tokens": 6235137.0,
"step": 3010
},
{
"entropy": 5.848877334594727,
"epoch": 0.27349419448476053,
"grad_norm": 1.3828125,
"learning_rate": 0.000499622631282183,
"loss": 5.8335,
"mean_token_accuracy": 0.1477284774184227,
"num_tokens": 6245411.0,
"step": 3015
},
{
"entropy": 5.880431079864502,
"epoch": 0.2739477503628447,
"grad_norm": 1.375,
"learning_rate": 0.0004996207557560095,
"loss": 5.9158,
"mean_token_accuracy": 0.14132191836833954,
"num_tokens": 6255633.0,
"step": 3020
},
{
"entropy": 5.930035352706909,
"epoch": 0.2744013062409289,
"grad_norm": 1.9453125,
"learning_rate": 0.0004996188755846216,
"loss": 5.8541,
"mean_token_accuracy": 0.14648665934801103,
"num_tokens": 6265109.0,
"step": 3025
},
{
"entropy": 5.880398845672607,
"epoch": 0.27485486211901305,
"grad_norm": 1.359375,
"learning_rate": 0.000499616990768058,
"loss": 5.9021,
"mean_token_accuracy": 0.1377116046845913,
"num_tokens": 6275241.0,
"step": 3030
},
{
"entropy": 5.896439170837402,
"epoch": 0.27530841799709727,
"grad_norm": 1.4921875,
"learning_rate": 0.0004996151013063579,
"loss": 5.9281,
"mean_token_accuracy": 0.13803014755249024,
"num_tokens": 6286603.0,
"step": 3035
},
{
"entropy": 5.930904483795166,
"epoch": 0.2757619738751814,
"grad_norm": 1.5625,
"learning_rate": 0.0004996132071995601,
"loss": 5.7778,
"mean_token_accuracy": 0.15983342826366426,
"num_tokens": 6297219.0,
"step": 3040
},
{
"entropy": 5.833100652694702,
"epoch": 0.2762155297532656,
"grad_norm": 1.390625,
"learning_rate": 0.0004996113084477042,
"loss": 5.8305,
"mean_token_accuracy": 0.15047609359025954,
"num_tokens": 6307694.0,
"step": 3045
},
{
"entropy": 5.810063695907592,
"epoch": 0.2766690856313498,
"grad_norm": 1.328125,
"learning_rate": 0.0004996094050508291,
"loss": 5.7577,
"mean_token_accuracy": 0.15339901745319368,
"num_tokens": 6317384.0,
"step": 3050
},
{
"entropy": 5.917490768432617,
"epoch": 0.27712264150943394,
"grad_norm": 1.4609375,
"learning_rate": 0.0004996074970089741,
"loss": 5.9031,
"mean_token_accuracy": 0.1421275481581688,
"num_tokens": 6327558.0,
"step": 3055
},
{
"entropy": 5.8848024845123295,
"epoch": 0.27757619738751815,
"grad_norm": 1.421875,
"learning_rate": 0.0004996055843221789,
"loss": 5.7545,
"mean_token_accuracy": 0.15326765477657317,
"num_tokens": 6337828.0,
"step": 3060
},
{
"entropy": 5.889102983474731,
"epoch": 0.2780297532656023,
"grad_norm": 1.4140625,
"learning_rate": 0.0004996036669904831,
"loss": 5.8111,
"mean_token_accuracy": 0.14792668223381042,
"num_tokens": 6348678.0,
"step": 3065
},
{
"entropy": 5.846058511734009,
"epoch": 0.2784833091436865,
"grad_norm": 1.359375,
"learning_rate": 0.0004996017450139261,
"loss": 5.9027,
"mean_token_accuracy": 0.13910972326993942,
"num_tokens": 6359670.0,
"step": 3070
},
{
"entropy": 5.913860940933228,
"epoch": 0.2789368650217707,
"grad_norm": 1.359375,
"learning_rate": 0.0004995998183925478,
"loss": 5.854,
"mean_token_accuracy": 0.14639595597982408,
"num_tokens": 6371034.0,
"step": 3075
},
{
"entropy": 5.873360013961792,
"epoch": 0.2793904208998549,
"grad_norm": 1.8671875,
"learning_rate": 0.0004995978871263881,
"loss": 5.8086,
"mean_token_accuracy": 0.14766118600964545,
"num_tokens": 6380672.0,
"step": 3080
},
{
"entropy": 5.863108396530151,
"epoch": 0.27984397677793904,
"grad_norm": 1.7109375,
"learning_rate": 0.0004995959512154868,
"loss": 5.9087,
"mean_token_accuracy": 0.1358805164694786,
"num_tokens": 6390799.0,
"step": 3085
},
{
"entropy": 5.892030858993531,
"epoch": 0.2802975326560232,
"grad_norm": 1.3515625,
"learning_rate": 0.000499594010659884,
"loss": 5.7886,
"mean_token_accuracy": 0.14606204628944397,
"num_tokens": 6400965.0,
"step": 3090
},
{
"entropy": 5.858356714248657,
"epoch": 0.2807510885341074,
"grad_norm": 1.40625,
"learning_rate": 0.0004995920654596199,
"loss": 5.8181,
"mean_token_accuracy": 0.14298465102910995,
"num_tokens": 6411567.0,
"step": 3095
},
{
"entropy": 5.93129563331604,
"epoch": 0.28120464441219156,
"grad_norm": 1.3671875,
"learning_rate": 0.0004995901156147345,
"loss": 5.8549,
"mean_token_accuracy": 0.14818204939365387,
"num_tokens": 6421193.0,
"step": 3100
},
{
"entropy": 5.919429969787598,
"epoch": 0.2816582002902758,
"grad_norm": 1.5,
"learning_rate": 0.0004995881611252684,
"loss": 5.8732,
"mean_token_accuracy": 0.15032653510570526,
"num_tokens": 6430281.0,
"step": 3105
},
{
"entropy": 5.843878746032715,
"epoch": 0.28211175616835993,
"grad_norm": 1.3359375,
"learning_rate": 0.0004995862019912619,
"loss": 5.8664,
"mean_token_accuracy": 0.13923195675015448,
"num_tokens": 6441055.0,
"step": 3110
},
{
"entropy": 5.923148584365845,
"epoch": 0.28256531204644414,
"grad_norm": 1.4140625,
"learning_rate": 0.0004995842382127555,
"loss": 5.7781,
"mean_token_accuracy": 0.14486080035567284,
"num_tokens": 6451082.0,
"step": 3115
},
{
"entropy": 5.824050951004028,
"epoch": 0.2830188679245283,
"grad_norm": 1.328125,
"learning_rate": 0.0004995822697897897,
"loss": 5.8814,
"mean_token_accuracy": 0.14269932955503464,
"num_tokens": 6461617.0,
"step": 3120
},
{
"entropy": 5.99531660079956,
"epoch": 0.2834724238026125,
"grad_norm": 1.390625,
"learning_rate": 0.0004995802967224056,
"loss": 5.8715,
"mean_token_accuracy": 0.14604993909597397,
"num_tokens": 6472780.0,
"step": 3125
},
{
"entropy": 5.821425580978394,
"epoch": 0.28392597968069666,
"grad_norm": 1.6015625,
"learning_rate": 0.0004995783190106434,
"loss": 5.7767,
"mean_token_accuracy": 0.1456163577735424,
"num_tokens": 6484078.0,
"step": 3130
},
{
"entropy": 5.887919139862061,
"epoch": 0.2843795355587808,
"grad_norm": 1.3046875,
"learning_rate": 0.0004995763366545445,
"loss": 5.8759,
"mean_token_accuracy": 0.14020358249545098,
"num_tokens": 6494366.0,
"step": 3135
},
{
"entropy": 5.793185806274414,
"epoch": 0.284833091436865,
"grad_norm": 1.1640625,
"learning_rate": 0.0004995743496541497,
"loss": 5.8095,
"mean_token_accuracy": 0.14859849438071251,
"num_tokens": 6505461.0,
"step": 3140
},
{
"entropy": 5.851748323440551,
"epoch": 0.2852866473149492,
"grad_norm": 1.359375,
"learning_rate": 0.0004995723580095001,
"loss": 5.8046,
"mean_token_accuracy": 0.14792132675647734,
"num_tokens": 6516141.0,
"step": 3145
},
{
"entropy": 5.804424190521241,
"epoch": 0.2857402031930334,
"grad_norm": 1.34375,
"learning_rate": 0.000499570361720637,
"loss": 5.8024,
"mean_token_accuracy": 0.1473324827849865,
"num_tokens": 6525536.0,
"step": 3150
},
{
"entropy": 5.917052936553955,
"epoch": 0.28619375907111755,
"grad_norm": 1.59375,
"learning_rate": 0.0004995683607876015,
"loss": 5.8618,
"mean_token_accuracy": 0.14245882481336594,
"num_tokens": 6535852.0,
"step": 3155
},
{
"entropy": 5.9194934368133545,
"epoch": 0.28664731494920176,
"grad_norm": 1.453125,
"learning_rate": 0.0004995663552104352,
"loss": 5.8379,
"mean_token_accuracy": 0.15086517333984376,
"num_tokens": 6546353.0,
"step": 3160
},
{
"entropy": 5.818840503692627,
"epoch": 0.2871008708272859,
"grad_norm": 1.453125,
"learning_rate": 0.0004995643449891793,
"loss": 5.7942,
"mean_token_accuracy": 0.1443500891327858,
"num_tokens": 6557782.0,
"step": 3165
},
{
"entropy": 5.908010578155517,
"epoch": 0.2875544267053701,
"grad_norm": 1.4609375,
"learning_rate": 0.0004995623301238757,
"loss": 5.8142,
"mean_token_accuracy": 0.150578635931015,
"num_tokens": 6568100.0,
"step": 3170
},
{
"entropy": 5.893179893493652,
"epoch": 0.2880079825834543,
"grad_norm": 1.3359375,
"learning_rate": 0.0004995603106145657,
"loss": 5.8712,
"mean_token_accuracy": 0.1439751148223877,
"num_tokens": 6578698.0,
"step": 3175
},
{
"entropy": 5.9084978103637695,
"epoch": 0.28846153846153844,
"grad_norm": 1.6328125,
"learning_rate": 0.0004995582864612913,
"loss": 5.8446,
"mean_token_accuracy": 0.145198056101799,
"num_tokens": 6588908.0,
"step": 3180
},
{
"entropy": 5.821733474731445,
"epoch": 0.28891509433962265,
"grad_norm": 1.3125,
"learning_rate": 0.0004995562576640943,
"loss": 5.7096,
"mean_token_accuracy": 0.14953436851501464,
"num_tokens": 6599445.0,
"step": 3185
},
{
"entropy": 5.891123914718628,
"epoch": 0.2893686502177068,
"grad_norm": 1.515625,
"learning_rate": 0.0004995542242230167,
"loss": 5.8133,
"mean_token_accuracy": 0.14369880110025407,
"num_tokens": 6610246.0,
"step": 3190
},
{
"entropy": 5.856010913848877,
"epoch": 0.289822206095791,
"grad_norm": 1.265625,
"learning_rate": 0.0004995521861381006,
"loss": 5.8187,
"mean_token_accuracy": 0.1429808184504509,
"num_tokens": 6620501.0,
"step": 3195
},
{
"entropy": 6.004367733001709,
"epoch": 0.29027576197387517,
"grad_norm": 1.2265625,
"learning_rate": 0.000499550143409388,
"loss": 5.8964,
"mean_token_accuracy": 0.1352742463350296,
"num_tokens": 6631206.0,
"step": 3200
},
{
"entropy": 5.821225309371949,
"epoch": 0.2907293178519594,
"grad_norm": 1.5546875,
"learning_rate": 0.0004995480960369211,
"loss": 5.8262,
"mean_token_accuracy": 0.14419071823358537,
"num_tokens": 6641600.0,
"step": 3205
},
{
"entropy": 5.829140472412109,
"epoch": 0.29118287373004353,
"grad_norm": 1.3359375,
"learning_rate": 0.0004995460440207425,
"loss": 5.8931,
"mean_token_accuracy": 0.14225514978170395,
"num_tokens": 6652319.0,
"step": 3210
},
{
"entropy": 5.9609822750091555,
"epoch": 0.29163642960812775,
"grad_norm": 1.34375,
"learning_rate": 0.0004995439873608945,
"loss": 5.8314,
"mean_token_accuracy": 0.1385738007724285,
"num_tokens": 6663702.0,
"step": 3215
},
{
"entropy": 5.875910472869873,
"epoch": 0.2920899854862119,
"grad_norm": 1.5859375,
"learning_rate": 0.0004995419260574196,
"loss": 5.8177,
"mean_token_accuracy": 0.1408117711544037,
"num_tokens": 6673908.0,
"step": 3220
},
{
"entropy": 5.849833536148071,
"epoch": 0.29254354136429606,
"grad_norm": 1.3671875,
"learning_rate": 0.0004995398601103603,
"loss": 5.8474,
"mean_token_accuracy": 0.1477855384349823,
"num_tokens": 6683237.0,
"step": 3225
},
{
"entropy": 5.877266454696655,
"epoch": 0.29299709724238027,
"grad_norm": 1.6015625,
"learning_rate": 0.0004995377895197597,
"loss": 5.7741,
"mean_token_accuracy": 0.1441804811358452,
"num_tokens": 6693451.0,
"step": 3230
},
{
"entropy": 5.826403617858887,
"epoch": 0.2934506531204644,
"grad_norm": 1.3046875,
"learning_rate": 0.0004995357142856602,
"loss": 5.7418,
"mean_token_accuracy": 0.1471038743853569,
"num_tokens": 6703479.0,
"step": 3235
},
{
"entropy": 5.786555290222168,
"epoch": 0.29390420899854863,
"grad_norm": 1.3125,
"learning_rate": 0.000499533634408105,
"loss": 5.7384,
"mean_token_accuracy": 0.1510653540492058,
"num_tokens": 6715321.0,
"step": 3240
},
{
"entropy": 5.851634502410889,
"epoch": 0.2943577648766328,
"grad_norm": 1.546875,
"learning_rate": 0.000499531549887137,
"loss": 5.7926,
"mean_token_accuracy": 0.14810622185468675,
"num_tokens": 6725541.0,
"step": 3245
},
{
"entropy": 5.840380907058716,
"epoch": 0.294811320754717,
"grad_norm": 1.578125,
"learning_rate": 0.0004995294607227993,
"loss": 5.8448,
"mean_token_accuracy": 0.1491126760840416,
"num_tokens": 6735165.0,
"step": 3250
},
{
"entropy": 5.938561058044433,
"epoch": 0.29526487663280115,
"grad_norm": 1.5546875,
"learning_rate": 0.0004995273669151351,
"loss": 5.8398,
"mean_token_accuracy": 0.1489756792783737,
"num_tokens": 6744245.0,
"step": 3255
},
{
"entropy": 5.934394836425781,
"epoch": 0.29571843251088537,
"grad_norm": 1.4375,
"learning_rate": 0.000499525268464188,
"loss": 5.7883,
"mean_token_accuracy": 0.1483618885278702,
"num_tokens": 6754894.0,
"step": 3260
},
{
"entropy": 5.832844591140747,
"epoch": 0.2961719883889695,
"grad_norm": 1.4765625,
"learning_rate": 0.0004995231653700009,
"loss": 5.8538,
"mean_token_accuracy": 0.14105366319417953,
"num_tokens": 6764442.0,
"step": 3265
},
{
"entropy": 5.8802886486053465,
"epoch": 0.2966255442670537,
"grad_norm": 1.375,
"learning_rate": 0.0004995210576326176,
"loss": 5.8285,
"mean_token_accuracy": 0.14174019694328308,
"num_tokens": 6775054.0,
"step": 3270
},
{
"entropy": 5.894062757492065,
"epoch": 0.2970791001451379,
"grad_norm": 1.3515625,
"learning_rate": 0.0004995189452520816,
"loss": 5.7856,
"mean_token_accuracy": 0.160012948513031,
"num_tokens": 6784825.0,
"step": 3275
},
{
"entropy": 5.887639808654785,
"epoch": 0.29753265602322204,
"grad_norm": 1.421875,
"learning_rate": 0.0004995168282284366,
"loss": 5.8532,
"mean_token_accuracy": 0.14631997644901276,
"num_tokens": 6795928.0,
"step": 3280
},
{
"entropy": 5.8099130153656,
"epoch": 0.29798621190130625,
"grad_norm": 1.2890625,
"learning_rate": 0.0004995147065617263,
"loss": 5.7761,
"mean_token_accuracy": 0.14812222272157669,
"num_tokens": 6805456.0,
"step": 3285
},
{
"entropy": 5.922118043899536,
"epoch": 0.2984397677793904,
"grad_norm": 1.6875,
"learning_rate": 0.0004995125802519946,
"loss": 5.8439,
"mean_token_accuracy": 0.14502991884946823,
"num_tokens": 6816215.0,
"step": 3290
},
{
"entropy": 5.972725343704224,
"epoch": 0.2988933236574746,
"grad_norm": 1.3984375,
"learning_rate": 0.0004995104492992857,
"loss": 5.8794,
"mean_token_accuracy": 0.14430792406201362,
"num_tokens": 6826545.0,
"step": 3295
},
{
"entropy": 5.7858147621154785,
"epoch": 0.2993468795355588,
"grad_norm": 1.546875,
"learning_rate": 0.0004995083137036434,
"loss": 5.7952,
"mean_token_accuracy": 0.1501883916556835,
"num_tokens": 6836184.0,
"step": 3300
},
{
"entropy": 5.940231418609619,
"epoch": 0.299800435413643,
"grad_norm": 2.484375,
"learning_rate": 0.0004995061734651121,
"loss": 5.8701,
"mean_token_accuracy": 0.14531150236725807,
"num_tokens": 6846646.0,
"step": 3305
},
{
"entropy": 5.887357616424561,
"epoch": 0.30025399129172714,
"grad_norm": 1.2109375,
"learning_rate": 0.0004995040285837358,
"loss": 5.7889,
"mean_token_accuracy": 0.1509875774383545,
"num_tokens": 6857597.0,
"step": 3310
},
{
"entropy": 5.866164779663086,
"epoch": 0.3007075471698113,
"grad_norm": 1.625,
"learning_rate": 0.0004995018790595589,
"loss": 5.8584,
"mean_token_accuracy": 0.14564741998910904,
"num_tokens": 6867523.0,
"step": 3315
},
{
"entropy": 5.948118543624878,
"epoch": 0.3011611030478955,
"grad_norm": 1.5546875,
"learning_rate": 0.0004994997248926261,
"loss": 5.9656,
"mean_token_accuracy": 0.14200707226991655,
"num_tokens": 6878190.0,
"step": 3320
},
{
"entropy": 5.900257682800293,
"epoch": 0.30161465892597966,
"grad_norm": 1.46875,
"learning_rate": 0.0004994975660829817,
"loss": 5.76,
"mean_token_accuracy": 0.15460998713970184,
"num_tokens": 6888483.0,
"step": 3325
},
{
"entropy": 5.831659650802612,
"epoch": 0.3020682148040639,
"grad_norm": 1.3671875,
"learning_rate": 0.0004994954026306705,
"loss": 5.7383,
"mean_token_accuracy": 0.14502749666571618,
"num_tokens": 6898337.0,
"step": 3330
},
{
"entropy": 5.909942626953125,
"epoch": 0.30252177068214803,
"grad_norm": 2.0625,
"learning_rate": 0.0004994932345357371,
"loss": 5.8354,
"mean_token_accuracy": 0.14530461728572847,
"num_tokens": 6909131.0,
"step": 3335
},
{
"entropy": 5.851958847045898,
"epoch": 0.30297532656023224,
"grad_norm": 1.4375,
"learning_rate": 0.0004994910617982264,
"loss": 5.7986,
"mean_token_accuracy": 0.14323278665542602,
"num_tokens": 6919790.0,
"step": 3340
},
{
"entropy": 5.855467844009399,
"epoch": 0.3034288824383164,
"grad_norm": 1.9765625,
"learning_rate": 0.0004994888844181833,
"loss": 5.8352,
"mean_token_accuracy": 0.1461854785680771,
"num_tokens": 6930072.0,
"step": 3345
},
{
"entropy": 5.8171202659606935,
"epoch": 0.3038824383164006,
"grad_norm": 1.3984375,
"learning_rate": 0.0004994867023956529,
"loss": 5.708,
"mean_token_accuracy": 0.1426757961511612,
"num_tokens": 6940646.0,
"step": 3350
},
{
"entropy": 5.799750900268554,
"epoch": 0.30433599419448476,
"grad_norm": 1.5859375,
"learning_rate": 0.0004994845157306803,
"loss": 5.7194,
"mean_token_accuracy": 0.1513740189373493,
"num_tokens": 6951645.0,
"step": 3355
},
{
"entropy": 5.950794792175293,
"epoch": 0.3047895500725689,
"grad_norm": 1.390625,
"learning_rate": 0.0004994823244233107,
"loss": 5.8849,
"mean_token_accuracy": 0.14364167153835297,
"num_tokens": 6963020.0,
"step": 3360
},
{
"entropy": 5.939484977722168,
"epoch": 0.3052431059506531,
"grad_norm": 1.6953125,
"learning_rate": 0.0004994801284735895,
"loss": 5.8,
"mean_token_accuracy": 0.15134814828634263,
"num_tokens": 6972895.0,
"step": 3365
},
{
"entropy": 5.890461444854736,
"epoch": 0.3056966618287373,
"grad_norm": 1.3359375,
"learning_rate": 0.0004994779278815619,
"loss": 5.9493,
"mean_token_accuracy": 0.13638330921530722,
"num_tokens": 6985363.0,
"step": 3370
},
{
"entropy": 5.943120670318604,
"epoch": 0.3061502177068215,
"grad_norm": 2.125,
"learning_rate": 0.0004994757226472736,
"loss": 5.8885,
"mean_token_accuracy": 0.14796184450387956,
"num_tokens": 6997040.0,
"step": 3375
},
{
"entropy": 5.883255338668823,
"epoch": 0.30660377358490565,
"grad_norm": 1.5078125,
"learning_rate": 0.0004994735127707702,
"loss": 5.8132,
"mean_token_accuracy": 0.14786105006933212,
"num_tokens": 7007591.0,
"step": 3380
},
{
"entropy": 5.877407550811768,
"epoch": 0.30705732946298986,
"grad_norm": 1.484375,
"learning_rate": 0.0004994712982520972,
"loss": 5.8239,
"mean_token_accuracy": 0.1532495453953743,
"num_tokens": 7017858.0,
"step": 3385
},
{
"entropy": 5.839944124221802,
"epoch": 0.307510885341074,
"grad_norm": 1.453125,
"learning_rate": 0.0004994690790913008,
"loss": 5.782,
"mean_token_accuracy": 0.145537555962801,
"num_tokens": 7027365.0,
"step": 3390
},
{
"entropy": 5.779080820083618,
"epoch": 0.3079644412191582,
"grad_norm": 1.6171875,
"learning_rate": 0.0004994668552884265,
"loss": 5.7916,
"mean_token_accuracy": 0.14543195366859435,
"num_tokens": 7038728.0,
"step": 3395
},
{
"entropy": 5.920030498504639,
"epoch": 0.3084179970972424,
"grad_norm": 1.8984375,
"learning_rate": 0.0004994646268435204,
"loss": 5.7284,
"mean_token_accuracy": 0.15286441892385483,
"num_tokens": 7048736.0,
"step": 3400
},
{
"entropy": 5.903434181213379,
"epoch": 0.30887155297532654,
"grad_norm": 1.5234375,
"learning_rate": 0.0004994623937566286,
"loss": 5.8435,
"mean_token_accuracy": 0.14856164008378983,
"num_tokens": 7059205.0,
"step": 3405
},
{
"entropy": 5.834399271011352,
"epoch": 0.30932510885341075,
"grad_norm": 1.578125,
"learning_rate": 0.0004994601560277974,
"loss": 5.8519,
"mean_token_accuracy": 0.13926964849233628,
"num_tokens": 7069513.0,
"step": 3410
},
{
"entropy": 5.890263462066651,
"epoch": 0.3097786647314949,
"grad_norm": 1.453125,
"learning_rate": 0.000499457913657073,
"loss": 5.7705,
"mean_token_accuracy": 0.147540445625782,
"num_tokens": 7079940.0,
"step": 3415
},
{
"entropy": 5.894211339950561,
"epoch": 0.3102322206095791,
"grad_norm": 1.4609375,
"learning_rate": 0.0004994556666445016,
"loss": 5.7803,
"mean_token_accuracy": 0.1465594932436943,
"num_tokens": 7090792.0,
"step": 3420
},
{
"entropy": 5.789063596725464,
"epoch": 0.31068577648766327,
"grad_norm": 1.3359375,
"learning_rate": 0.0004994534149901299,
"loss": 5.7383,
"mean_token_accuracy": 0.14637766480445863,
"num_tokens": 7101964.0,
"step": 3425
},
{
"entropy": 5.8579018115997314,
"epoch": 0.3111393323657475,
"grad_norm": 1.6328125,
"learning_rate": 0.0004994511586940044,
"loss": 5.7893,
"mean_token_accuracy": 0.14257095232605935,
"num_tokens": 7113332.0,
"step": 3430
},
{
"entropy": 5.97416934967041,
"epoch": 0.31159288824383163,
"grad_norm": 1.390625,
"learning_rate": 0.0004994488977561717,
"loss": 5.9521,
"mean_token_accuracy": 0.14183745980262757,
"num_tokens": 7123481.0,
"step": 3435
},
{
"entropy": 5.814324283599854,
"epoch": 0.31204644412191584,
"grad_norm": 1.78125,
"learning_rate": 0.0004994466321766786,
"loss": 5.7119,
"mean_token_accuracy": 0.15085338801145554,
"num_tokens": 7133694.0,
"step": 3440
},
{
"entropy": 5.849011659622192,
"epoch": 0.3125,
"grad_norm": 1.6171875,
"learning_rate": 0.000499444361955572,
"loss": 5.8351,
"mean_token_accuracy": 0.15081941783428193,
"num_tokens": 7144491.0,
"step": 3445
},
{
"entropy": 6.005057287216187,
"epoch": 0.31295355587808416,
"grad_norm": 1.6875,
"learning_rate": 0.0004994420870928988,
"loss": 5.9086,
"mean_token_accuracy": 0.14320788010954857,
"num_tokens": 7154716.0,
"step": 3450
},
{
"entropy": 5.896556758880616,
"epoch": 0.31340711175616837,
"grad_norm": 1.7265625,
"learning_rate": 0.0004994398075887059,
"loss": 5.9023,
"mean_token_accuracy": 0.14134786948561667,
"num_tokens": 7164258.0,
"step": 3455
},
{
"entropy": 5.83188419342041,
"epoch": 0.3138606676342525,
"grad_norm": 1.390625,
"learning_rate": 0.0004994375234430407,
"loss": 5.7675,
"mean_token_accuracy": 0.1461493268609047,
"num_tokens": 7175098.0,
"step": 3460
},
{
"entropy": 5.892137575149536,
"epoch": 0.31431422351233673,
"grad_norm": 1.5625,
"learning_rate": 0.0004994352346559504,
"loss": 5.8343,
"mean_token_accuracy": 0.14542945474386215,
"num_tokens": 7186480.0,
"step": 3465
},
{
"entropy": 5.799042797088623,
"epoch": 0.3147677793904209,
"grad_norm": 2.390625,
"learning_rate": 0.0004994329412274821,
"loss": 5.8072,
"mean_token_accuracy": 0.14921786934137343,
"num_tokens": 7197056.0,
"step": 3470
},
{
"entropy": 5.921955919265747,
"epoch": 0.3152213352685051,
"grad_norm": 1.8203125,
"learning_rate": 0.0004994306431576835,
"loss": 5.908,
"mean_token_accuracy": 0.14707638397812844,
"num_tokens": 7206040.0,
"step": 3475
},
{
"entropy": 5.987424421310425,
"epoch": 0.31567489114658925,
"grad_norm": 1.3203125,
"learning_rate": 0.000499428340446602,
"loss": 5.773,
"mean_token_accuracy": 0.15154775828123093,
"num_tokens": 7216344.0,
"step": 3480
},
{
"entropy": 5.893925428390503,
"epoch": 0.31612844702467346,
"grad_norm": 1.9609375,
"learning_rate": 0.0004994260330942851,
"loss": 5.8535,
"mean_token_accuracy": 0.14545960426330568,
"num_tokens": 7226729.0,
"step": 3485
},
{
"entropy": 5.822306489944458,
"epoch": 0.3165820029027576,
"grad_norm": 1.5703125,
"learning_rate": 0.0004994237211007808,
"loss": 5.8019,
"mean_token_accuracy": 0.14516061842441558,
"num_tokens": 7237616.0,
"step": 3490
},
{
"entropy": 5.896799755096436,
"epoch": 0.3170355587808418,
"grad_norm": 1.765625,
"learning_rate": 0.0004994214044661367,
"loss": 5.7364,
"mean_token_accuracy": 0.14803769513964654,
"num_tokens": 7247857.0,
"step": 3495
},
{
"entropy": 5.767730283737182,
"epoch": 0.317489114658926,
"grad_norm": 1.7578125,
"learning_rate": 0.0004994190831904007,
"loss": 5.711,
"mean_token_accuracy": 0.15267411172389983,
"num_tokens": 7257858.0,
"step": 3500
},
{
"entropy": 5.71488733291626,
"epoch": 0.31794267053701014,
"grad_norm": 1.6171875,
"learning_rate": 0.000499416757273621,
"loss": 5.7186,
"mean_token_accuracy": 0.1502937451004982,
"num_tokens": 7268440.0,
"step": 3505
},
{
"entropy": 5.9244773387908936,
"epoch": 0.31839622641509435,
"grad_norm": 1.390625,
"learning_rate": 0.0004994144267158454,
"loss": 5.7995,
"mean_token_accuracy": 0.15224562138319014,
"num_tokens": 7278560.0,
"step": 3510
},
{
"entropy": 5.970515632629395,
"epoch": 0.3188497822931785,
"grad_norm": 1.609375,
"learning_rate": 0.0004994120915171224,
"loss": 5.7927,
"mean_token_accuracy": 0.14984412491321564,
"num_tokens": 7288422.0,
"step": 3515
},
{
"entropy": 5.850483465194702,
"epoch": 0.3193033381712627,
"grad_norm": 2.34375,
"learning_rate": 0.0004994097516775002,
"loss": 5.8555,
"mean_token_accuracy": 0.14620980769395828,
"num_tokens": 7298613.0,
"step": 3520
},
{
"entropy": 5.9037727355957035,
"epoch": 0.3197568940493469,
"grad_norm": 1.625,
"learning_rate": 0.0004994074071970271,
"loss": 5.8943,
"mean_token_accuracy": 0.1411413997411728,
"num_tokens": 7309509.0,
"step": 3525
},
{
"entropy": 5.948377180099487,
"epoch": 0.3202104499274311,
"grad_norm": 1.7890625,
"learning_rate": 0.0004994050580757515,
"loss": 5.8095,
"mean_token_accuracy": 0.14360484182834626,
"num_tokens": 7319015.0,
"step": 3530
},
{
"entropy": 5.825267028808594,
"epoch": 0.32066400580551524,
"grad_norm": 1.984375,
"learning_rate": 0.0004994027043137223,
"loss": 5.7781,
"mean_token_accuracy": 0.15057334899902344,
"num_tokens": 7329026.0,
"step": 3535
},
{
"entropy": 5.940208768844604,
"epoch": 0.3211175616835994,
"grad_norm": 1.6015625,
"learning_rate": 0.0004994003459109879,
"loss": 5.8815,
"mean_token_accuracy": 0.14752597510814666,
"num_tokens": 7339638.0,
"step": 3540
},
{
"entropy": 5.925622272491455,
"epoch": 0.3215711175616836,
"grad_norm": 1.6328125,
"learning_rate": 0.0004993979828675972,
"loss": 5.7797,
"mean_token_accuracy": 0.1494549334049225,
"num_tokens": 7349979.0,
"step": 3545
},
{
"entropy": 5.7519083499908445,
"epoch": 0.32202467343976776,
"grad_norm": 2.53125,
"learning_rate": 0.000499395615183599,
"loss": 5.7062,
"mean_token_accuracy": 0.15184318274259567,
"num_tokens": 7359398.0,
"step": 3550
},
{
"entropy": 5.7866541862487795,
"epoch": 0.32247822931785197,
"grad_norm": 1.7265625,
"learning_rate": 0.0004993932428590422,
"loss": 5.7652,
"mean_token_accuracy": 0.1473848804831505,
"num_tokens": 7370418.0,
"step": 3555
},
{
"entropy": 5.946097373962402,
"epoch": 0.3229317851959361,
"grad_norm": 1.4609375,
"learning_rate": 0.0004993908658939762,
"loss": 5.8028,
"mean_token_accuracy": 0.1365408256649971,
"num_tokens": 7380461.0,
"step": 3560
},
{
"entropy": 5.865476608276367,
"epoch": 0.32338534107402034,
"grad_norm": 1.40625,
"learning_rate": 0.0004993884842884496,
"loss": 5.7909,
"mean_token_accuracy": 0.14746063798666,
"num_tokens": 7390646.0,
"step": 3565
},
{
"entropy": 5.833335304260254,
"epoch": 0.3238388969521045,
"grad_norm": 1.8046875,
"learning_rate": 0.0004993860980425121,
"loss": 5.9112,
"mean_token_accuracy": 0.14362555146217346,
"num_tokens": 7402106.0,
"step": 3570
},
{
"entropy": 5.881076526641846,
"epoch": 0.3242924528301887,
"grad_norm": 1.78125,
"learning_rate": 0.0004993837071562131,
"loss": 5.8834,
"mean_token_accuracy": 0.14547116905450821,
"num_tokens": 7412936.0,
"step": 3575
},
{
"entropy": 5.909448480606079,
"epoch": 0.32474600870827286,
"grad_norm": 1.71875,
"learning_rate": 0.0004993813116296017,
"loss": 5.7451,
"mean_token_accuracy": 0.15334491580724716,
"num_tokens": 7422303.0,
"step": 3580
},
{
"entropy": 5.858001613616944,
"epoch": 0.325199564586357,
"grad_norm": 2.3125,
"learning_rate": 0.0004993789114627275,
"loss": 5.8084,
"mean_token_accuracy": 0.1455884724855423,
"num_tokens": 7432829.0,
"step": 3585
},
{
"entropy": 5.897142839431763,
"epoch": 0.3256531204644412,
"grad_norm": 1.7421875,
"learning_rate": 0.0004993765066556403,
"loss": 5.8427,
"mean_token_accuracy": 0.14214815869927405,
"num_tokens": 7443218.0,
"step": 3590
},
{
"entropy": 5.87728099822998,
"epoch": 0.3261066763425254,
"grad_norm": 1.78125,
"learning_rate": 0.0004993740972083899,
"loss": 5.8019,
"mean_token_accuracy": 0.13930851370096206,
"num_tokens": 7454531.0,
"step": 3595
},
{
"entropy": 5.862864971160889,
"epoch": 0.3265602322206096,
"grad_norm": 1.7265625,
"learning_rate": 0.0004993716831210259,
"loss": 5.8417,
"mean_token_accuracy": 0.1489178165793419,
"num_tokens": 7464993.0,
"step": 3600
},
{
"entropy": 5.848693084716797,
"epoch": 0.32701378809869375,
"grad_norm": 1.4375,
"learning_rate": 0.0004993692643935983,
"loss": 5.8819,
"mean_token_accuracy": 0.14340077340602875,
"num_tokens": 7475755.0,
"step": 3605
},
{
"entropy": 5.898052215576172,
"epoch": 0.32746734397677796,
"grad_norm": 1.40625,
"learning_rate": 0.0004993668410261571,
"loss": 5.8648,
"mean_token_accuracy": 0.14696988239884376,
"num_tokens": 7486778.0,
"step": 3610
},
{
"entropy": 5.989887142181397,
"epoch": 0.3279208998548621,
"grad_norm": 1.7734375,
"learning_rate": 0.0004993644130187525,
"loss": 5.826,
"mean_token_accuracy": 0.14816126823425294,
"num_tokens": 7497439.0,
"step": 3615
},
{
"entropy": 5.916240262985229,
"epoch": 0.3283744557329463,
"grad_norm": 1.875,
"learning_rate": 0.0004993619803714347,
"loss": 5.8852,
"mean_token_accuracy": 0.14901129752397538,
"num_tokens": 7508274.0,
"step": 3620
},
{
"entropy": 5.900164651870727,
"epoch": 0.3288280116110305,
"grad_norm": 1.7109375,
"learning_rate": 0.0004993595430842539,
"loss": 5.8396,
"mean_token_accuracy": 0.14458931982517242,
"num_tokens": 7519315.0,
"step": 3625
},
{
"entropy": 5.966023588180542,
"epoch": 0.32928156748911463,
"grad_norm": 1.5859375,
"learning_rate": 0.0004993571011572606,
"loss": 5.8417,
"mean_token_accuracy": 0.14624714255332946,
"num_tokens": 7529910.0,
"step": 3630
},
{
"entropy": 5.87901463508606,
"epoch": 0.32973512336719885,
"grad_norm": 1.5703125,
"learning_rate": 0.0004993546545905053,
"loss": 5.7665,
"mean_token_accuracy": 0.15389569848775864,
"num_tokens": 7538966.0,
"step": 3635
},
{
"entropy": 5.901384925842285,
"epoch": 0.330188679245283,
"grad_norm": 1.53125,
"learning_rate": 0.0004993522033840386,
"loss": 5.8063,
"mean_token_accuracy": 0.1490241676568985,
"num_tokens": 7549683.0,
"step": 3640
},
{
"entropy": 5.766738080978394,
"epoch": 0.3306422351233672,
"grad_norm": 1.40625,
"learning_rate": 0.000499349747537911,
"loss": 5.788,
"mean_token_accuracy": 0.14788973480463027,
"num_tokens": 7560629.0,
"step": 3645
},
{
"entropy": 5.873026323318482,
"epoch": 0.33109579100145137,
"grad_norm": 1.8359375,
"learning_rate": 0.0004993472870521736,
"loss": 5.7384,
"mean_token_accuracy": 0.15600567162036896,
"num_tokens": 7570613.0,
"step": 3650
},
{
"entropy": 5.851700639724731,
"epoch": 0.3315493468795356,
"grad_norm": 1.5546875,
"learning_rate": 0.0004993448219268771,
"loss": 5.7104,
"mean_token_accuracy": 0.1479036659002304,
"num_tokens": 7580640.0,
"step": 3655
},
{
"entropy": 5.855727005004883,
"epoch": 0.33200290275761973,
"grad_norm": 1.7578125,
"learning_rate": 0.0004993423521620726,
"loss": 5.7926,
"mean_token_accuracy": 0.14974100142717361,
"num_tokens": 7590164.0,
"step": 3660
},
{
"entropy": 5.854600191116333,
"epoch": 0.33245645863570394,
"grad_norm": 1.4609375,
"learning_rate": 0.0004993398777578109,
"loss": 5.8563,
"mean_token_accuracy": 0.1418274387717247,
"num_tokens": 7600173.0,
"step": 3665
},
{
"entropy": 5.840301847457885,
"epoch": 0.3329100145137881,
"grad_norm": 1.484375,
"learning_rate": 0.0004993373987141435,
"loss": 5.6667,
"mean_token_accuracy": 0.15604306310415267,
"num_tokens": 7609958.0,
"step": 3670
},
{
"entropy": 5.945655298233032,
"epoch": 0.33336357039187225,
"grad_norm": 1.875,
"learning_rate": 0.0004993349150311214,
"loss": 5.7995,
"mean_token_accuracy": 0.14565619081258774,
"num_tokens": 7620237.0,
"step": 3675
},
{
"entropy": 5.800672483444214,
"epoch": 0.33381712626995647,
"grad_norm": 1.765625,
"learning_rate": 0.0004993324267087962,
"loss": 5.6914,
"mean_token_accuracy": 0.15652529150247574,
"num_tokens": 7629886.0,
"step": 3680
},
{
"entropy": 5.822660684585571,
"epoch": 0.3342706821480406,
"grad_norm": 1.875,
"learning_rate": 0.0004993299337472192,
"loss": 5.77,
"mean_token_accuracy": 0.15085675567388535,
"num_tokens": 7640057.0,
"step": 3685
},
{
"entropy": 5.817735576629639,
"epoch": 0.33472423802612483,
"grad_norm": 1.6015625,
"learning_rate": 0.000499327436146442,
"loss": 5.7257,
"mean_token_accuracy": 0.1475510910153389,
"num_tokens": 7649766.0,
"step": 3690
},
{
"entropy": 5.8362211227417,
"epoch": 0.335177793904209,
"grad_norm": 1.734375,
"learning_rate": 0.0004993249339065162,
"loss": 5.7994,
"mean_token_accuracy": 0.14809372201561927,
"num_tokens": 7660890.0,
"step": 3695
},
{
"entropy": 5.8948447704315186,
"epoch": 0.3356313497822932,
"grad_norm": 1.8828125,
"learning_rate": 0.0004993224270274937,
"loss": 5.8063,
"mean_token_accuracy": 0.14590197429060936,
"num_tokens": 7670737.0,
"step": 3700
},
{
"entropy": 5.797480678558349,
"epoch": 0.33608490566037735,
"grad_norm": 1.9609375,
"learning_rate": 0.0004993199155094263,
"loss": 5.7728,
"mean_token_accuracy": 0.14692779183387755,
"num_tokens": 7681407.0,
"step": 3705
},
{
"entropy": 5.8480466365814205,
"epoch": 0.33653846153846156,
"grad_norm": 1.5234375,
"learning_rate": 0.0004993173993523658,
"loss": 5.7183,
"mean_token_accuracy": 0.16032444536685944,
"num_tokens": 7691758.0,
"step": 3710
},
{
"entropy": 5.812997913360595,
"epoch": 0.3369920174165457,
"grad_norm": 1.59375,
"learning_rate": 0.0004993148785563643,
"loss": 5.7426,
"mean_token_accuracy": 0.14660506919026375,
"num_tokens": 7702934.0,
"step": 3715
},
{
"entropy": 5.723813915252686,
"epoch": 0.3374455732946299,
"grad_norm": 1.3671875,
"learning_rate": 0.000499312353121474,
"loss": 5.7393,
"mean_token_accuracy": 0.15231507793068885,
"num_tokens": 7713001.0,
"step": 3720
},
{
"entropy": 5.890734529495239,
"epoch": 0.3378991291727141,
"grad_norm": 1.4921875,
"learning_rate": 0.0004993098230477471,
"loss": 5.794,
"mean_token_accuracy": 0.14045751169323922,
"num_tokens": 7724005.0,
"step": 3725
},
{
"entropy": 5.900547075271606,
"epoch": 0.33835268505079824,
"grad_norm": 1.3125,
"learning_rate": 0.0004993072883352358,
"loss": 5.7467,
"mean_token_accuracy": 0.1473555825650692,
"num_tokens": 7733743.0,
"step": 3730
},
{
"entropy": 5.855947017669678,
"epoch": 0.33880624092888245,
"grad_norm": 1.640625,
"learning_rate": 0.0004993047489839927,
"loss": 5.9383,
"mean_token_accuracy": 0.13958059400320053,
"num_tokens": 7743762.0,
"step": 3735
},
{
"entropy": 5.930490493774414,
"epoch": 0.3392597968069666,
"grad_norm": 1.484375,
"learning_rate": 0.0004993022049940702,
"loss": 5.817,
"mean_token_accuracy": 0.14824567139148712,
"num_tokens": 7753758.0,
"step": 3740
},
{
"entropy": 5.871481132507324,
"epoch": 0.3397133526850508,
"grad_norm": 1.5234375,
"learning_rate": 0.0004992996563655209,
"loss": 5.7719,
"mean_token_accuracy": 0.1495535232126713,
"num_tokens": 7764692.0,
"step": 3745
},
{
"entropy": 5.848981046676636,
"epoch": 0.340166908563135,
"grad_norm": 1.6171875,
"learning_rate": 0.0004992971030983975,
"loss": 5.8128,
"mean_token_accuracy": 0.14353037178516387,
"num_tokens": 7776892.0,
"step": 3750
},
{
"entropy": 5.902813720703125,
"epoch": 0.3406204644412192,
"grad_norm": 1.5703125,
"learning_rate": 0.0004992945451927531,
"loss": 5.8642,
"mean_token_accuracy": 0.1436462000012398,
"num_tokens": 7787143.0,
"step": 3755
},
{
"entropy": 5.806349992752075,
"epoch": 0.34107402031930334,
"grad_norm": 1.6640625,
"learning_rate": 0.0004992919826486402,
"loss": 5.7675,
"mean_token_accuracy": 0.1482495039701462,
"num_tokens": 7798537.0,
"step": 3760
},
{
"entropy": 5.8226241111755375,
"epoch": 0.3415275761973875,
"grad_norm": 1.5234375,
"learning_rate": 0.0004992894154661118,
"loss": 5.7344,
"mean_token_accuracy": 0.15073239281773568,
"num_tokens": 7808434.0,
"step": 3765
},
{
"entropy": 5.856382083892822,
"epoch": 0.3419811320754717,
"grad_norm": 1.6484375,
"learning_rate": 0.0004992868436452213,
"loss": 5.851,
"mean_token_accuracy": 0.1481686621904373,
"num_tokens": 7819592.0,
"step": 3770
},
{
"entropy": 5.890800809860229,
"epoch": 0.34243468795355586,
"grad_norm": 1.78125,
"learning_rate": 0.0004992842671860217,
"loss": 5.7417,
"mean_token_accuracy": 0.15032289028167725,
"num_tokens": 7830602.0,
"step": 3775
},
{
"entropy": 5.879576730728149,
"epoch": 0.34288824383164007,
"grad_norm": 2.03125,
"learning_rate": 0.0004992816860885662,
"loss": 5.8953,
"mean_token_accuracy": 0.1439703606069088,
"num_tokens": 7840850.0,
"step": 3780
},
{
"entropy": 6.028309392929077,
"epoch": 0.3433417997097242,
"grad_norm": 2.46875,
"learning_rate": 0.0004992791003529084,
"loss": 5.8544,
"mean_token_accuracy": 0.14057560488581658,
"num_tokens": 7852371.0,
"step": 3785
},
{
"entropy": 5.9221659183502195,
"epoch": 0.34379535558780844,
"grad_norm": 1.640625,
"learning_rate": 0.0004992765099791017,
"loss": 5.8467,
"mean_token_accuracy": 0.14126079380512238,
"num_tokens": 7863290.0,
"step": 3790
},
{
"entropy": 5.783722972869873,
"epoch": 0.3442489114658926,
"grad_norm": 1.5546875,
"learning_rate": 0.0004992739149671995,
"loss": 5.687,
"mean_token_accuracy": 0.1527349516749382,
"num_tokens": 7873927.0,
"step": 3795
},
{
"entropy": 5.791727685928345,
"epoch": 0.3447024673439768,
"grad_norm": 1.4296875,
"learning_rate": 0.0004992713153172556,
"loss": 5.7173,
"mean_token_accuracy": 0.14965997114777566,
"num_tokens": 7884977.0,
"step": 3800
},
{
"entropy": 5.827057123184204,
"epoch": 0.34515602322206096,
"grad_norm": 1.546875,
"learning_rate": 0.0004992687110293238,
"loss": 5.7985,
"mean_token_accuracy": 0.15286190062761307,
"num_tokens": 7894620.0,
"step": 3805
},
{
"entropy": 5.84814682006836,
"epoch": 0.3456095791001451,
"grad_norm": 1.5703125,
"learning_rate": 0.0004992661021034579,
"loss": 5.7404,
"mean_token_accuracy": 0.15445290505886078,
"num_tokens": 7904541.0,
"step": 3810
},
{
"entropy": 5.818321228027344,
"epoch": 0.3460631349782293,
"grad_norm": 1.2578125,
"learning_rate": 0.0004992634885397118,
"loss": 5.7626,
"mean_token_accuracy": 0.14903516918420792,
"num_tokens": 7915435.0,
"step": 3815
},
{
"entropy": 5.83319125175476,
"epoch": 0.3465166908563135,
"grad_norm": 1.3359375,
"learning_rate": 0.0004992608703381396,
"loss": 5.742,
"mean_token_accuracy": 0.14871969372034072,
"num_tokens": 7925870.0,
"step": 3820
},
{
"entropy": 5.8578869819641115,
"epoch": 0.3469702467343977,
"grad_norm": 1.9140625,
"learning_rate": 0.0004992582474987955,
"loss": 5.7545,
"mean_token_accuracy": 0.14518966972827912,
"num_tokens": 7937046.0,
"step": 3825
},
{
"entropy": 5.77307505607605,
"epoch": 0.34742380261248185,
"grad_norm": 1.671875,
"learning_rate": 0.0004992556200217337,
"loss": 5.7064,
"mean_token_accuracy": 0.14739801585674286,
"num_tokens": 7947150.0,
"step": 3830
},
{
"entropy": 5.862250852584839,
"epoch": 0.34787735849056606,
"grad_norm": 1.640625,
"learning_rate": 0.0004992529879070085,
"loss": 5.8049,
"mean_token_accuracy": 0.1511588916182518,
"num_tokens": 7956815.0,
"step": 3835
},
{
"entropy": 5.875207948684692,
"epoch": 0.3483309143686502,
"grad_norm": 1.515625,
"learning_rate": 0.0004992503511546745,
"loss": 5.8322,
"mean_token_accuracy": 0.14654328525066376,
"num_tokens": 7968114.0,
"step": 3840
},
{
"entropy": 5.881295871734619,
"epoch": 0.3487844702467344,
"grad_norm": 1.671875,
"learning_rate": 0.0004992477097647859,
"loss": 5.7647,
"mean_token_accuracy": 0.14300089776515962,
"num_tokens": 7977840.0,
"step": 3845
},
{
"entropy": 5.8811851978302006,
"epoch": 0.3492380261248186,
"grad_norm": 1.4765625,
"learning_rate": 0.0004992450637373976,
"loss": 5.8822,
"mean_token_accuracy": 0.14847241342067719,
"num_tokens": 7988663.0,
"step": 3850
},
{
"entropy": 5.808912992477417,
"epoch": 0.34969158200290273,
"grad_norm": 1.3671875,
"learning_rate": 0.0004992424130725642,
"loss": 5.6955,
"mean_token_accuracy": 0.14893131777644159,
"num_tokens": 7999693.0,
"step": 3855
},
{
"entropy": 5.872442388534546,
"epoch": 0.35014513788098695,
"grad_norm": 1.53125,
"learning_rate": 0.0004992397577703406,
"loss": 5.8621,
"mean_token_accuracy": 0.14381872415542601,
"num_tokens": 8010154.0,
"step": 3860
},
{
"entropy": 5.946900177001953,
"epoch": 0.3505986937590711,
"grad_norm": 1.8515625,
"learning_rate": 0.0004992370978307816,
"loss": 5.8415,
"mean_token_accuracy": 0.14593325406312943,
"num_tokens": 8020921.0,
"step": 3865
},
{
"entropy": 5.8881330490112305,
"epoch": 0.3510522496371553,
"grad_norm": 2.5,
"learning_rate": 0.0004992344332539424,
"loss": 5.7708,
"mean_token_accuracy": 0.15435807704925536,
"num_tokens": 8031404.0,
"step": 3870
},
{
"entropy": 5.730764579772949,
"epoch": 0.35150580551523947,
"grad_norm": 1.4453125,
"learning_rate": 0.0004992317640398779,
"loss": 5.7562,
"mean_token_accuracy": 0.1508772239089012,
"num_tokens": 8041310.0,
"step": 3875
},
{
"entropy": 5.828841018676758,
"epoch": 0.3519593613933237,
"grad_norm": 1.6796875,
"learning_rate": 0.0004992290901886434,
"loss": 5.7351,
"mean_token_accuracy": 0.14970532357692717,
"num_tokens": 8051054.0,
"step": 3880
},
{
"entropy": 5.846589374542236,
"epoch": 0.35241291727140783,
"grad_norm": 1.859375,
"learning_rate": 0.000499226411700294,
"loss": 5.7467,
"mean_token_accuracy": 0.14929102063179017,
"num_tokens": 8061063.0,
"step": 3885
},
{
"entropy": 5.819320011138916,
"epoch": 0.35286647314949204,
"grad_norm": 1.7265625,
"learning_rate": 0.0004992237285748855,
"loss": 5.749,
"mean_token_accuracy": 0.15423073768615722,
"num_tokens": 8071715.0,
"step": 3890
},
{
"entropy": 5.911148452758789,
"epoch": 0.3533200290275762,
"grad_norm": 1.6953125,
"learning_rate": 0.0004992210408124731,
"loss": 5.8243,
"mean_token_accuracy": 0.14293495416641236,
"num_tokens": 8082679.0,
"step": 3895
},
{
"entropy": 5.898702573776245,
"epoch": 0.35377358490566035,
"grad_norm": 1.4453125,
"learning_rate": 0.0004992183484131123,
"loss": 5.8528,
"mean_token_accuracy": 0.14598763659596442,
"num_tokens": 8093785.0,
"step": 3900
},
{
"entropy": 5.834564685821533,
"epoch": 0.35422714078374457,
"grad_norm": 1.6796875,
"learning_rate": 0.0004992156513768591,
"loss": 5.7622,
"mean_token_accuracy": 0.1508698433637619,
"num_tokens": 8105822.0,
"step": 3905
},
{
"entropy": 5.7624317646026615,
"epoch": 0.3546806966618287,
"grad_norm": 1.7109375,
"learning_rate": 0.000499212949703769,
"loss": 5.682,
"mean_token_accuracy": 0.15023178458213807,
"num_tokens": 8115521.0,
"step": 3910
},
{
"entropy": 5.811681795120239,
"epoch": 0.35513425253991293,
"grad_norm": 1.6640625,
"learning_rate": 0.0004992102433938981,
"loss": 5.7786,
"mean_token_accuracy": 0.14909463599324227,
"num_tokens": 8125524.0,
"step": 3915
},
{
"entropy": 5.856146621704101,
"epoch": 0.3555878084179971,
"grad_norm": 1.6171875,
"learning_rate": 0.0004992075324473021,
"loss": 5.6655,
"mean_token_accuracy": 0.16061010211706161,
"num_tokens": 8135552.0,
"step": 3920
},
{
"entropy": 5.878396081924438,
"epoch": 0.3560413642960813,
"grad_norm": 1.9921875,
"learning_rate": 0.0004992048168640371,
"loss": 5.8021,
"mean_token_accuracy": 0.14835941046476364,
"num_tokens": 8145617.0,
"step": 3925
},
{
"entropy": 5.8598052024841305,
"epoch": 0.35649492017416545,
"grad_norm": 1.4765625,
"learning_rate": 0.0004992020966441595,
"loss": 5.8254,
"mean_token_accuracy": 0.14229780659079552,
"num_tokens": 8156729.0,
"step": 3930
},
{
"entropy": 5.834389066696167,
"epoch": 0.35694847605224966,
"grad_norm": 1.9609375,
"learning_rate": 0.0004991993717877254,
"loss": 5.8103,
"mean_token_accuracy": 0.1433609440922737,
"num_tokens": 8167637.0,
"step": 3935
},
{
"entropy": 5.883994483947754,
"epoch": 0.3574020319303338,
"grad_norm": 2.21875,
"learning_rate": 0.0004991966422947911,
"loss": 5.7989,
"mean_token_accuracy": 0.14962732493877412,
"num_tokens": 8177354.0,
"step": 3940
},
{
"entropy": 5.8065185546875,
"epoch": 0.357855587808418,
"grad_norm": 2.0,
"learning_rate": 0.0004991939081654132,
"loss": 5.7199,
"mean_token_accuracy": 0.16128290593624114,
"num_tokens": 8187580.0,
"step": 3945
},
{
"entropy": 5.752160549163818,
"epoch": 0.3583091436865022,
"grad_norm": 1.796875,
"learning_rate": 0.000499191169399648,
"loss": 5.8216,
"mean_token_accuracy": 0.14404607713222503,
"num_tokens": 8199382.0,
"step": 3950
},
{
"entropy": 5.846578645706177,
"epoch": 0.35876269956458634,
"grad_norm": 1.53125,
"learning_rate": 0.0004991884259975525,
"loss": 5.7256,
"mean_token_accuracy": 0.14961068257689475,
"num_tokens": 8208484.0,
"step": 3955
},
{
"entropy": 5.8810042381286625,
"epoch": 0.35921625544267055,
"grad_norm": 1.7421875,
"learning_rate": 0.000499185677959183,
"loss": 5.7582,
"mean_token_accuracy": 0.14644067436456681,
"num_tokens": 8219056.0,
"step": 3960
},
{
"entropy": 5.849036693572998,
"epoch": 0.3596698113207547,
"grad_norm": 1.4296875,
"learning_rate": 0.0004991829252845966,
"loss": 5.7966,
"mean_token_accuracy": 0.14414289593696594,
"num_tokens": 8230149.0,
"step": 3965
},
{
"entropy": 5.865396165847779,
"epoch": 0.3601233671988389,
"grad_norm": 2.234375,
"learning_rate": 0.0004991801679738502,
"loss": 5.8497,
"mean_token_accuracy": 0.1402447283267975,
"num_tokens": 8240803.0,
"step": 3970
},
{
"entropy": 5.896198892593384,
"epoch": 0.3605769230769231,
"grad_norm": 1.8046875,
"learning_rate": 0.0004991774060270008,
"loss": 5.8232,
"mean_token_accuracy": 0.14320749416947365,
"num_tokens": 8251563.0,
"step": 3975
},
{
"entropy": 5.7957603454589846,
"epoch": 0.3610304789550073,
"grad_norm": 1.6953125,
"learning_rate": 0.0004991746394441055,
"loss": 5.7534,
"mean_token_accuracy": 0.14512395709753037,
"num_tokens": 8261342.0,
"step": 3980
},
{
"entropy": 5.8211053848266605,
"epoch": 0.36148403483309144,
"grad_norm": 1.6875,
"learning_rate": 0.0004991718682252216,
"loss": 5.8592,
"mean_token_accuracy": 0.14282869845628737,
"num_tokens": 8271540.0,
"step": 3985
},
{
"entropy": 5.969235181808472,
"epoch": 0.3619375907111756,
"grad_norm": 2.078125,
"learning_rate": 0.0004991690923704063,
"loss": 5.774,
"mean_token_accuracy": 0.1459478497505188,
"num_tokens": 8280593.0,
"step": 3990
},
{
"entropy": 5.8458741188049315,
"epoch": 0.3623911465892598,
"grad_norm": 1.5703125,
"learning_rate": 0.0004991663118797171,
"loss": 5.7937,
"mean_token_accuracy": 0.1484941601753235,
"num_tokens": 8290126.0,
"step": 3995
},
{
"entropy": 5.810283374786377,
"epoch": 0.36284470246734396,
"grad_norm": 2.03125,
"learning_rate": 0.0004991635267532113,
"loss": 5.7276,
"mean_token_accuracy": 0.1489349991083145,
"num_tokens": 8301416.0,
"step": 4000
},
{
"entropy": 5.92182149887085,
"epoch": 0.36329825834542817,
"grad_norm": 1.75,
"learning_rate": 0.0004991607369909468,
"loss": 5.8051,
"mean_token_accuracy": 0.14902321994304657,
"num_tokens": 8311034.0,
"step": 4005
},
{
"entropy": 5.9140537738800045,
"epoch": 0.3637518142235123,
"grad_norm": 1.484375,
"learning_rate": 0.0004991579425929811,
"loss": 5.8382,
"mean_token_accuracy": 0.14275244921445845,
"num_tokens": 8322203.0,
"step": 4010
},
{
"entropy": 5.910188245773315,
"epoch": 0.36420537010159654,
"grad_norm": 1.5546875,
"learning_rate": 0.000499155143559372,
"loss": 5.703,
"mean_token_accuracy": 0.15037306100130082,
"num_tokens": 8330981.0,
"step": 4015
},
{
"entropy": 5.727109909057617,
"epoch": 0.3646589259796807,
"grad_norm": 1.9765625,
"learning_rate": 0.0004991523398901773,
"loss": 5.7422,
"mean_token_accuracy": 0.14837262779474258,
"num_tokens": 8341294.0,
"step": 4020
},
{
"entropy": 5.937906169891358,
"epoch": 0.3651124818577649,
"grad_norm": 1.6875,
"learning_rate": 0.0004991495315854553,
"loss": 5.8772,
"mean_token_accuracy": 0.14473702684044837,
"num_tokens": 8351824.0,
"step": 4025
},
{
"entropy": 5.887541055679321,
"epoch": 0.36556603773584906,
"grad_norm": 1.8828125,
"learning_rate": 0.0004991467186452637,
"loss": 5.7527,
"mean_token_accuracy": 0.14745676666498184,
"num_tokens": 8361975.0,
"step": 4030
},
{
"entropy": 5.788140058517456,
"epoch": 0.3660195936139332,
"grad_norm": 1.671875,
"learning_rate": 0.0004991439010696609,
"loss": 5.7179,
"mean_token_accuracy": 0.15196914672851564,
"num_tokens": 8372145.0,
"step": 4035
},
{
"entropy": 5.883623647689819,
"epoch": 0.3664731494920174,
"grad_norm": 1.5546875,
"learning_rate": 0.0004991410788587051,
"loss": 5.7945,
"mean_token_accuracy": 0.1431012563407421,
"num_tokens": 8382542.0,
"step": 4040
},
{
"entropy": 5.880132675170898,
"epoch": 0.3669267053701016,
"grad_norm": 1.8828125,
"learning_rate": 0.0004991382520124547,
"loss": 5.8053,
"mean_token_accuracy": 0.14454633221030236,
"num_tokens": 8393450.0,
"step": 4045
},
{
"entropy": 5.843823099136353,
"epoch": 0.3673802612481858,
"grad_norm": 1.578125,
"learning_rate": 0.0004991354205309681,
"loss": 5.7325,
"mean_token_accuracy": 0.1445598617196083,
"num_tokens": 8404929.0,
"step": 4050
},
{
"entropy": 5.8328451156616214,
"epoch": 0.36783381712626995,
"grad_norm": 2.015625,
"learning_rate": 0.000499132584414304,
"loss": 5.763,
"mean_token_accuracy": 0.14964553862810134,
"num_tokens": 8415703.0,
"step": 4055
},
{
"entropy": 5.886098432540893,
"epoch": 0.36828737300435416,
"grad_norm": 1.5859375,
"learning_rate": 0.0004991297436625207,
"loss": 5.8256,
"mean_token_accuracy": 0.14880608171224594,
"num_tokens": 8426220.0,
"step": 4060
},
{
"entropy": 5.749038791656494,
"epoch": 0.3687409288824383,
"grad_norm": 1.7265625,
"learning_rate": 0.0004991268982756774,
"loss": 5.7145,
"mean_token_accuracy": 0.15660099536180497,
"num_tokens": 8436220.0,
"step": 4065
},
{
"entropy": 5.905976867675781,
"epoch": 0.3691944847605225,
"grad_norm": 1.6484375,
"learning_rate": 0.0004991240482538327,
"loss": 5.8618,
"mean_token_accuracy": 0.140651698410511,
"num_tokens": 8446724.0,
"step": 4070
},
{
"entropy": 5.880495882034301,
"epoch": 0.3696480406386067,
"grad_norm": 1.625,
"learning_rate": 0.0004991211935970455,
"loss": 5.7217,
"mean_token_accuracy": 0.1466690830886364,
"num_tokens": 8457270.0,
"step": 4075
},
{
"entropy": 5.808180665969848,
"epoch": 0.37010159651669083,
"grad_norm": 1.6015625,
"learning_rate": 0.0004991183343053748,
"loss": 5.7933,
"mean_token_accuracy": 0.1437254838645458,
"num_tokens": 8468137.0,
"step": 4080
},
{
"entropy": 5.832044219970703,
"epoch": 0.37055515239477504,
"grad_norm": 1.59375,
"learning_rate": 0.00049911547037888,
"loss": 5.7281,
"mean_token_accuracy": 0.1513017401099205,
"num_tokens": 8477574.0,
"step": 4085
},
{
"entropy": 5.84261417388916,
"epoch": 0.3710087082728592,
"grad_norm": 1.8984375,
"learning_rate": 0.0004991126018176202,
"loss": 5.8255,
"mean_token_accuracy": 0.14621062725782394,
"num_tokens": 8487268.0,
"step": 4090
},
{
"entropy": 5.94973406791687,
"epoch": 0.3714622641509434,
"grad_norm": 1.75,
"learning_rate": 0.0004991097286216545,
"loss": 5.8993,
"mean_token_accuracy": 0.14729729145765305,
"num_tokens": 8497785.0,
"step": 4095
},
{
"entropy": 5.92133355140686,
"epoch": 0.37191582002902757,
"grad_norm": 1.734375,
"learning_rate": 0.0004991068507910427,
"loss": 5.7538,
"mean_token_accuracy": 0.14549212902784348,
"num_tokens": 8508169.0,
"step": 4100
},
{
"entropy": 5.796505546569824,
"epoch": 0.3723693759071118,
"grad_norm": 1.515625,
"learning_rate": 0.000499103968325844,
"loss": 5.7271,
"mean_token_accuracy": 0.14605697244405746,
"num_tokens": 8519201.0,
"step": 4105
},
{
"entropy": 5.881065845489502,
"epoch": 0.37282293178519593,
"grad_norm": 1.5,
"learning_rate": 0.0004991010812261181,
"loss": 5.8454,
"mean_token_accuracy": 0.13941508531570435,
"num_tokens": 8531093.0,
"step": 4110
},
{
"entropy": 5.850164127349854,
"epoch": 0.37327648766328014,
"grad_norm": 1.6328125,
"learning_rate": 0.0004990981894919248,
"loss": 5.6946,
"mean_token_accuracy": 0.14870947301387788,
"num_tokens": 8541730.0,
"step": 4115
},
{
"entropy": 5.792994403839112,
"epoch": 0.3737300435413643,
"grad_norm": 1.921875,
"learning_rate": 0.0004990952931233238,
"loss": 5.7656,
"mean_token_accuracy": 0.150176040828228,
"num_tokens": 8552570.0,
"step": 4120
},
{
"entropy": 5.824308729171753,
"epoch": 0.37418359941944845,
"grad_norm": 2.015625,
"learning_rate": 0.000499092392120375,
"loss": 5.6597,
"mean_token_accuracy": 0.1537941262125969,
"num_tokens": 8562340.0,
"step": 4125
},
{
"entropy": 5.876683855056763,
"epoch": 0.37463715529753266,
"grad_norm": 1.5234375,
"learning_rate": 0.0004990894864831385,
"loss": 5.8779,
"mean_token_accuracy": 0.14065658450126647,
"num_tokens": 8573240.0,
"step": 4130
},
{
"entropy": 5.889408159255981,
"epoch": 0.3750907111756168,
"grad_norm": 1.640625,
"learning_rate": 0.0004990865762116743,
"loss": 5.7275,
"mean_token_accuracy": 0.15267123952507972,
"num_tokens": 8583194.0,
"step": 4135
},
{
"entropy": 5.83022403717041,
"epoch": 0.37554426705370103,
"grad_norm": 2.03125,
"learning_rate": 0.0004990836613060426,
"loss": 5.7907,
"mean_token_accuracy": 0.1415669023990631,
"num_tokens": 8593603.0,
"step": 4140
},
{
"entropy": 5.852129411697388,
"epoch": 0.3759978229317852,
"grad_norm": 1.6953125,
"learning_rate": 0.0004990807417663036,
"loss": 5.8044,
"mean_token_accuracy": 0.14670791774988173,
"num_tokens": 8603176.0,
"step": 4145
},
{
"entropy": 5.894939708709717,
"epoch": 0.3764513788098694,
"grad_norm": 1.2734375,
"learning_rate": 0.0004990778175925179,
"loss": 5.8056,
"mean_token_accuracy": 0.15206588059663773,
"num_tokens": 8614801.0,
"step": 4150
},
{
"entropy": 5.8759352684021,
"epoch": 0.37690493468795355,
"grad_norm": 1.96875,
"learning_rate": 0.0004990748887847457,
"loss": 5.8379,
"mean_token_accuracy": 0.1481950357556343,
"num_tokens": 8624999.0,
"step": 4155
},
{
"entropy": 5.830811405181885,
"epoch": 0.37735849056603776,
"grad_norm": 1.6875,
"learning_rate": 0.0004990719553430477,
"loss": 5.7012,
"mean_token_accuracy": 0.1591061070561409,
"num_tokens": 8635358.0,
"step": 4160
},
{
"entropy": 5.838520669937134,
"epoch": 0.3778120464441219,
"grad_norm": 1.625,
"learning_rate": 0.0004990690172674847,
"loss": 5.7631,
"mean_token_accuracy": 0.1457097850739956,
"num_tokens": 8645430.0,
"step": 4165
},
{
"entropy": 5.834043788909912,
"epoch": 0.3782656023222061,
"grad_norm": 1.4453125,
"learning_rate": 0.0004990660745581172,
"loss": 5.7158,
"mean_token_accuracy": 0.15678425431251525,
"num_tokens": 8655115.0,
"step": 4170
},
{
"entropy": 5.725192070007324,
"epoch": 0.3787191582002903,
"grad_norm": 1.703125,
"learning_rate": 0.0004990631272150062,
"loss": 5.7122,
"mean_token_accuracy": 0.14877525568008423,
"num_tokens": 8665391.0,
"step": 4175
},
{
"entropy": 5.905309581756592,
"epoch": 0.37917271407837444,
"grad_norm": 1.28125,
"learning_rate": 0.0004990601752382127,
"loss": 5.8272,
"mean_token_accuracy": 0.14464897736907006,
"num_tokens": 8676819.0,
"step": 4180
},
{
"entropy": 5.917379760742188,
"epoch": 0.37962626995645865,
"grad_norm": 1.6796875,
"learning_rate": 0.0004990572186277975,
"loss": 5.7598,
"mean_token_accuracy": 0.15172497928142548,
"num_tokens": 8687052.0,
"step": 4185
},
{
"entropy": 5.85373740196228,
"epoch": 0.3800798258345428,
"grad_norm": 1.609375,
"learning_rate": 0.0004990542573838221,
"loss": 5.8318,
"mean_token_accuracy": 0.14298003166913986,
"num_tokens": 8697756.0,
"step": 4190
},
{
"entropy": 5.784401750564575,
"epoch": 0.380533381712627,
"grad_norm": 2.203125,
"learning_rate": 0.0004990512915063475,
"loss": 5.7493,
"mean_token_accuracy": 0.15153614282608033,
"num_tokens": 8708388.0,
"step": 4195
},
{
"entropy": 5.872270154953003,
"epoch": 0.38098693759071117,
"grad_norm": 1.609375,
"learning_rate": 0.0004990483209954352,
"loss": 5.8258,
"mean_token_accuracy": 0.1415306583046913,
"num_tokens": 8719475.0,
"step": 4200
},
{
"entropy": 5.8587422370910645,
"epoch": 0.3814404934687954,
"grad_norm": 1.7421875,
"learning_rate": 0.0004990453458511464,
"loss": 5.6861,
"mean_token_accuracy": 0.15934394747018815,
"num_tokens": 8729395.0,
"step": 4205
},
{
"entropy": 5.873230600357056,
"epoch": 0.38189404934687954,
"grad_norm": 1.4453125,
"learning_rate": 0.0004990423660735428,
"loss": 5.7549,
"mean_token_accuracy": 0.1521786004304886,
"num_tokens": 8740351.0,
"step": 4210
},
{
"entropy": 5.826301527023316,
"epoch": 0.3823476052249637,
"grad_norm": 1.78125,
"learning_rate": 0.000499039381662686,
"loss": 5.7769,
"mean_token_accuracy": 0.14992594420909883,
"num_tokens": 8749260.0,
"step": 4215
},
{
"entropy": 5.838196468353272,
"epoch": 0.3828011611030479,
"grad_norm": 1.4296875,
"learning_rate": 0.0004990363926186377,
"loss": 5.8601,
"mean_token_accuracy": 0.13967606276273728,
"num_tokens": 8759604.0,
"step": 4220
},
{
"entropy": 5.865482902526855,
"epoch": 0.38325471698113206,
"grad_norm": 1.671875,
"learning_rate": 0.0004990333989414597,
"loss": 5.8563,
"mean_token_accuracy": 0.14229664355516433,
"num_tokens": 8769815.0,
"step": 4225
},
{
"entropy": 5.842839145660401,
"epoch": 0.38370827285921627,
"grad_norm": 1.953125,
"learning_rate": 0.0004990304006312138,
"loss": 5.8042,
"mean_token_accuracy": 0.144217911362648,
"num_tokens": 8780557.0,
"step": 4230
},
{
"entropy": 5.893247365951538,
"epoch": 0.3841618287373004,
"grad_norm": 1.4296875,
"learning_rate": 0.0004990273976879622,
"loss": 5.746,
"mean_token_accuracy": 0.15741802453994752,
"num_tokens": 8790993.0,
"step": 4235
},
{
"entropy": 5.908751630783081,
"epoch": 0.38461538461538464,
"grad_norm": 1.6796875,
"learning_rate": 0.0004990243901117669,
"loss": 5.7993,
"mean_token_accuracy": 0.14164488315582274,
"num_tokens": 8800657.0,
"step": 4240
},
{
"entropy": 5.749859476089478,
"epoch": 0.3850689404934688,
"grad_norm": 1.703125,
"learning_rate": 0.0004990213779026903,
"loss": 5.8134,
"mean_token_accuracy": 0.14273017644882202,
"num_tokens": 8812983.0,
"step": 4245
},
{
"entropy": 5.878562164306641,
"epoch": 0.385522496371553,
"grad_norm": 1.4609375,
"learning_rate": 0.0004990183610607944,
"loss": 5.7723,
"mean_token_accuracy": 0.14839280098676683,
"num_tokens": 8822980.0,
"step": 4250
},
{
"entropy": 5.868063688278198,
"epoch": 0.38597605224963716,
"grad_norm": 1.6484375,
"learning_rate": 0.0004990153395861415,
"loss": 5.7693,
"mean_token_accuracy": 0.15079350471496583,
"num_tokens": 8833821.0,
"step": 4255
},
{
"entropy": 5.78454327583313,
"epoch": 0.3864296081277213,
"grad_norm": 1.4453125,
"learning_rate": 0.0004990123134787946,
"loss": 5.7049,
"mean_token_accuracy": 0.15205924212932587,
"num_tokens": 8843929.0,
"step": 4260
},
{
"entropy": 5.8183831691741945,
"epoch": 0.3868831640058055,
"grad_norm": 1.5390625,
"learning_rate": 0.0004990092827388159,
"loss": 5.6926,
"mean_token_accuracy": 0.14838198870420455,
"num_tokens": 8854180.0,
"step": 4265
},
{
"entropy": 5.767665433883667,
"epoch": 0.3873367198838897,
"grad_norm": 1.8828125,
"learning_rate": 0.0004990062473662681,
"loss": 5.708,
"mean_token_accuracy": 0.15309692025184632,
"num_tokens": 8865018.0,
"step": 4270
},
{
"entropy": 5.879072809219361,
"epoch": 0.3877902757619739,
"grad_norm": 1.765625,
"learning_rate": 0.000499003207361214,
"loss": 5.8918,
"mean_token_accuracy": 0.14059318900108336,
"num_tokens": 8875423.0,
"step": 4275
},
{
"entropy": 5.988429594039917,
"epoch": 0.38824383164005805,
"grad_norm": 1.765625,
"learning_rate": 0.0004990001627237165,
"loss": 5.8433,
"mean_token_accuracy": 0.14742626398801803,
"num_tokens": 8887850.0,
"step": 4280
},
{
"entropy": 5.77903642654419,
"epoch": 0.38869738751814226,
"grad_norm": 2.125,
"learning_rate": 0.0004989971134538386,
"loss": 5.7332,
"mean_token_accuracy": 0.152449893951416,
"num_tokens": 8897429.0,
"step": 4285
},
{
"entropy": 5.795289611816406,
"epoch": 0.3891509433962264,
"grad_norm": 1.859375,
"learning_rate": 0.0004989940595516432,
"loss": 5.7322,
"mean_token_accuracy": 0.15252401381731034,
"num_tokens": 8907048.0,
"step": 4290
},
{
"entropy": 5.867383098602295,
"epoch": 0.3896044992743106,
"grad_norm": 2.09375,
"learning_rate": 0.0004989910010171938,
"loss": 5.7955,
"mean_token_accuracy": 0.14241870269179344,
"num_tokens": 8917987.0,
"step": 4295
},
{
"entropy": 5.902405786514282,
"epoch": 0.3900580551523948,
"grad_norm": 1.8203125,
"learning_rate": 0.0004989879378505532,
"loss": 5.7042,
"mean_token_accuracy": 0.145450921356678,
"num_tokens": 8929339.0,
"step": 4300
},
{
"entropy": 5.787789440155029,
"epoch": 0.39051161103047893,
"grad_norm": 1.59375,
"learning_rate": 0.0004989848700517849,
"loss": 5.7059,
"mean_token_accuracy": 0.15002443864941598,
"num_tokens": 8939473.0,
"step": 4305
},
{
"entropy": 5.778521919250489,
"epoch": 0.39096516690856314,
"grad_norm": 3.09375,
"learning_rate": 0.0004989817976209526,
"loss": 5.7477,
"mean_token_accuracy": 0.14690720587968825,
"num_tokens": 8949173.0,
"step": 4310
},
{
"entropy": 5.890463495254517,
"epoch": 0.3914187227866473,
"grad_norm": 1.6015625,
"learning_rate": 0.0004989787205581196,
"loss": 5.7321,
"mean_token_accuracy": 0.15126073211431504,
"num_tokens": 8959879.0,
"step": 4315
},
{
"entropy": 5.9013659954071045,
"epoch": 0.3918722786647315,
"grad_norm": 1.7109375,
"learning_rate": 0.0004989756388633496,
"loss": 5.7962,
"mean_token_accuracy": 0.1466445192694664,
"num_tokens": 8970643.0,
"step": 4320
},
{
"entropy": 5.776039171218872,
"epoch": 0.39232583454281567,
"grad_norm": 2.421875,
"learning_rate": 0.0004989725525367063,
"loss": 5.662,
"mean_token_accuracy": 0.15230050086975097,
"num_tokens": 8979882.0,
"step": 4325
},
{
"entropy": 5.822905206680298,
"epoch": 0.3927793904208999,
"grad_norm": 1.8125,
"learning_rate": 0.0004989694615782534,
"loss": 5.8141,
"mean_token_accuracy": 0.14447067975997924,
"num_tokens": 8990925.0,
"step": 4330
},
{
"entropy": 5.899914932250977,
"epoch": 0.39323294629898403,
"grad_norm": 1.734375,
"learning_rate": 0.0004989663659880552,
"loss": 5.84,
"mean_token_accuracy": 0.1423853024840355,
"num_tokens": 9001365.0,
"step": 4335
},
{
"entropy": 5.923264074325561,
"epoch": 0.3936865021770682,
"grad_norm": 3.0,
"learning_rate": 0.0004989632657661754,
"loss": 5.8286,
"mean_token_accuracy": 0.14973220527172088,
"num_tokens": 9011942.0,
"step": 4340
},
{
"entropy": 5.935524415969849,
"epoch": 0.3941400580551524,
"grad_norm": 1.8828125,
"learning_rate": 0.000498960160912678,
"loss": 5.8534,
"mean_token_accuracy": 0.14469540640711784,
"num_tokens": 9022610.0,
"step": 4345
},
{
"entropy": 5.852841472625732,
"epoch": 0.39459361393323655,
"grad_norm": 1.7265625,
"learning_rate": 0.0004989570514276276,
"loss": 5.7577,
"mean_token_accuracy": 0.14448110163211822,
"num_tokens": 9033034.0,
"step": 4350
},
{
"entropy": 5.914149904251099,
"epoch": 0.39504716981132076,
"grad_norm": 1.8203125,
"learning_rate": 0.0004989539373110883,
"loss": 5.7943,
"mean_token_accuracy": 0.1416556268930435,
"num_tokens": 9043418.0,
"step": 4355
},
{
"entropy": 5.945955562591553,
"epoch": 0.3955007256894049,
"grad_norm": 1.75,
"learning_rate": 0.0004989508185631245,
"loss": 5.8783,
"mean_token_accuracy": 0.14210682958364487,
"num_tokens": 9052708.0,
"step": 4360
},
{
"entropy": 5.764258909225464,
"epoch": 0.39595428156748913,
"grad_norm": 1.6328125,
"learning_rate": 0.0004989476951838006,
"loss": 5.741,
"mean_token_accuracy": 0.14578800797462463,
"num_tokens": 9062883.0,
"step": 4365
},
{
"entropy": 5.867785787582397,
"epoch": 0.3964078374455733,
"grad_norm": 2.078125,
"learning_rate": 0.0004989445671731814,
"loss": 5.7054,
"mean_token_accuracy": 0.15434126257896424,
"num_tokens": 9072830.0,
"step": 4370
},
{
"entropy": 5.7734722137451175,
"epoch": 0.3968613933236575,
"grad_norm": 1.609375,
"learning_rate": 0.0004989414345313315,
"loss": 5.6661,
"mean_token_accuracy": 0.1422044441103935,
"num_tokens": 9082749.0,
"step": 4375
},
{
"entropy": 5.874911880493164,
"epoch": 0.39731494920174165,
"grad_norm": 1.5703125,
"learning_rate": 0.0004989382972583156,
"loss": 5.7632,
"mean_token_accuracy": 0.14635880067944526,
"num_tokens": 9094188.0,
"step": 4380
},
{
"entropy": 5.875842332839966,
"epoch": 0.3977685050798258,
"grad_norm": 1.78125,
"learning_rate": 0.0004989351553541986,
"loss": 5.7743,
"mean_token_accuracy": 0.1443245954811573,
"num_tokens": 9105023.0,
"step": 4385
},
{
"entropy": 5.7874414920806885,
"epoch": 0.39822206095791,
"grad_norm": 1.8515625,
"learning_rate": 0.0004989320088190456,
"loss": 5.725,
"mean_token_accuracy": 0.1519322693347931,
"num_tokens": 9115563.0,
"step": 4390
},
{
"entropy": 5.8170891284942625,
"epoch": 0.3986756168359942,
"grad_norm": 1.90625,
"learning_rate": 0.0004989288576529216,
"loss": 5.7447,
"mean_token_accuracy": 0.14550501853227615,
"num_tokens": 9125394.0,
"step": 4395
},
{
"entropy": 5.876050138473511,
"epoch": 0.3991291727140784,
"grad_norm": 1.7734375,
"learning_rate": 0.0004989257018558917,
"loss": 5.7457,
"mean_token_accuracy": 0.14662075340747832,
"num_tokens": 9137061.0,
"step": 4400
},
{
"entropy": 5.858101940155029,
"epoch": 0.39958272859216254,
"grad_norm": 1.6640625,
"learning_rate": 0.0004989225414280213,
"loss": 5.7085,
"mean_token_accuracy": 0.14827828407287597,
"num_tokens": 9148196.0,
"step": 4405
},
{
"entropy": 5.830370330810547,
"epoch": 0.40003628447024675,
"grad_norm": 1.9140625,
"learning_rate": 0.0004989193763693756,
"loss": 5.7425,
"mean_token_accuracy": 0.15035545378923415,
"num_tokens": 9158013.0,
"step": 4410
},
{
"entropy": 5.81728343963623,
"epoch": 0.4004898403483309,
"grad_norm": 1.828125,
"learning_rate": 0.0004989162066800202,
"loss": 5.6504,
"mean_token_accuracy": 0.1513180986046791,
"num_tokens": 9167813.0,
"step": 4415
},
{
"entropy": 5.791998386383057,
"epoch": 0.4009433962264151,
"grad_norm": 2.203125,
"learning_rate": 0.0004989130323600205,
"loss": 5.7369,
"mean_token_accuracy": 0.15377046391367913,
"num_tokens": 9177783.0,
"step": 4420
},
{
"entropy": 5.829497003555298,
"epoch": 0.40139695210449927,
"grad_norm": 1.9140625,
"learning_rate": 0.0004989098534094424,
"loss": 5.7658,
"mean_token_accuracy": 0.15063516944646835,
"num_tokens": 9187556.0,
"step": 4425
},
{
"entropy": 5.810343599319458,
"epoch": 0.4018505079825834,
"grad_norm": 1.609375,
"learning_rate": 0.0004989066698283512,
"loss": 5.8178,
"mean_token_accuracy": 0.14447112008929253,
"num_tokens": 9198659.0,
"step": 4430
},
{
"entropy": 5.862035846710205,
"epoch": 0.40230406386066764,
"grad_norm": 1.9921875,
"learning_rate": 0.0004989034816168132,
"loss": 5.7796,
"mean_token_accuracy": 0.14605519771575928,
"num_tokens": 9209546.0,
"step": 4435
},
{
"entropy": 5.83269739151001,
"epoch": 0.4027576197387518,
"grad_norm": 2.234375,
"learning_rate": 0.0004989002887748941,
"loss": 5.7002,
"mean_token_accuracy": 0.1521790012717247,
"num_tokens": 9219339.0,
"step": 4440
},
{
"entropy": 5.877965784072876,
"epoch": 0.403211175616836,
"grad_norm": 1.71875,
"learning_rate": 0.0004988970913026601,
"loss": 5.8796,
"mean_token_accuracy": 0.14370514899492265,
"num_tokens": 9230364.0,
"step": 4445
},
{
"entropy": 5.9046632766723635,
"epoch": 0.40366473149492016,
"grad_norm": 2.484375,
"learning_rate": 0.0004988938892001771,
"loss": 5.7459,
"mean_token_accuracy": 0.14931874126195907,
"num_tokens": 9240732.0,
"step": 4450
},
{
"entropy": 5.796473550796509,
"epoch": 0.40411828737300437,
"grad_norm": 1.9296875,
"learning_rate": 0.0004988906824675113,
"loss": 5.6842,
"mean_token_accuracy": 0.15042189061641692,
"num_tokens": 9251049.0,
"step": 4455
},
{
"entropy": 5.852061557769775,
"epoch": 0.4045718432510885,
"grad_norm": 2.125,
"learning_rate": 0.0004988874711047293,
"loss": 5.727,
"mean_token_accuracy": 0.15044807642698288,
"num_tokens": 9261660.0,
"step": 4460
},
{
"entropy": 5.830964851379394,
"epoch": 0.40502539912917274,
"grad_norm": 1.625,
"learning_rate": 0.0004988842551118972,
"loss": 5.798,
"mean_token_accuracy": 0.14962558969855308,
"num_tokens": 9272114.0,
"step": 4465
},
{
"entropy": 5.867624425888062,
"epoch": 0.4054789550072569,
"grad_norm": 2.125,
"learning_rate": 0.0004988810344890818,
"loss": 5.767,
"mean_token_accuracy": 0.15117275416851045,
"num_tokens": 9282125.0,
"step": 4470
},
{
"entropy": 5.913155126571655,
"epoch": 0.40593251088534105,
"grad_norm": 1.765625,
"learning_rate": 0.0004988778092363494,
"loss": 5.8271,
"mean_token_accuracy": 0.14333474934101104,
"num_tokens": 9294078.0,
"step": 4475
},
{
"entropy": 5.834015846252441,
"epoch": 0.40638606676342526,
"grad_norm": 1.7421875,
"learning_rate": 0.0004988745793537671,
"loss": 5.7602,
"mean_token_accuracy": 0.14512751549482344,
"num_tokens": 9304026.0,
"step": 4480
},
{
"entropy": 5.8868653774261475,
"epoch": 0.4068396226415094,
"grad_norm": 1.78125,
"learning_rate": 0.0004988713448414011,
"loss": 5.7957,
"mean_token_accuracy": 0.15431096106767656,
"num_tokens": 9313933.0,
"step": 4485
},
{
"entropy": 5.919203186035157,
"epoch": 0.4072931785195936,
"grad_norm": 2.390625,
"learning_rate": 0.000498868105699319,
"loss": 5.8216,
"mean_token_accuracy": 0.14666167348623277,
"num_tokens": 9323932.0,
"step": 4490
},
{
"entropy": 5.755291604995728,
"epoch": 0.4077467343976778,
"grad_norm": 1.8828125,
"learning_rate": 0.0004988648619275872,
"loss": 5.7265,
"mean_token_accuracy": 0.15795115977525712,
"num_tokens": 9333467.0,
"step": 4495
},
{
"entropy": 5.849260568618774,
"epoch": 0.408200290275762,
"grad_norm": 1.609375,
"learning_rate": 0.0004988616135262729,
"loss": 5.7941,
"mean_token_accuracy": 0.1402094468474388,
"num_tokens": 9344909.0,
"step": 4500
},
{
"entropy": 5.984080076217651,
"epoch": 0.40865384615384615,
"grad_norm": 1.8671875,
"learning_rate": 0.0004988583604954436,
"loss": 5.8291,
"mean_token_accuracy": 0.1422007627785206,
"num_tokens": 9355630.0,
"step": 4505
},
{
"entropy": 5.847146606445312,
"epoch": 0.40910740203193036,
"grad_norm": 1.9609375,
"learning_rate": 0.0004988551028351663,
"loss": 5.7674,
"mean_token_accuracy": 0.14505108445882797,
"num_tokens": 9365700.0,
"step": 4510
},
{
"entropy": 5.8058990955352785,
"epoch": 0.4095609579100145,
"grad_norm": 1.96875,
"learning_rate": 0.0004988518405455084,
"loss": 5.8489,
"mean_token_accuracy": 0.14073796719312667,
"num_tokens": 9377088.0,
"step": 4515
},
{
"entropy": 5.843289089202881,
"epoch": 0.41001451378809867,
"grad_norm": 2.171875,
"learning_rate": 0.0004988485736265374,
"loss": 5.6492,
"mean_token_accuracy": 0.1532072402536869,
"num_tokens": 9386855.0,
"step": 4520
},
{
"entropy": 5.841462135314941,
"epoch": 0.4104680696661829,
"grad_norm": 1.6875,
"learning_rate": 0.000498845302078321,
"loss": 5.772,
"mean_token_accuracy": 0.14480848461389542,
"num_tokens": 9398661.0,
"step": 4525
},
{
"entropy": 5.877380418777466,
"epoch": 0.41092162554426703,
"grad_norm": 2.40625,
"learning_rate": 0.0004988420259009266,
"loss": 5.7597,
"mean_token_accuracy": 0.15487963408231736,
"num_tokens": 9408670.0,
"step": 4530
},
{
"entropy": 5.917161273956299,
"epoch": 0.41137518142235124,
"grad_norm": 3.078125,
"learning_rate": 0.000498838745094422,
"loss": 5.7748,
"mean_token_accuracy": 0.14555237144231797,
"num_tokens": 9418740.0,
"step": 4535
},
{
"entropy": 5.8036597728729244,
"epoch": 0.4118287373004354,
"grad_norm": 3.25,
"learning_rate": 0.0004988354596588751,
"loss": 5.7659,
"mean_token_accuracy": 0.15228321254253388,
"num_tokens": 9429339.0,
"step": 4540
},
{
"entropy": 5.7854917526245115,
"epoch": 0.4122822931785196,
"grad_norm": 1.6953125,
"learning_rate": 0.0004988321695943539,
"loss": 5.7236,
"mean_token_accuracy": 0.15588060170412063,
"num_tokens": 9438996.0,
"step": 4545
},
{
"entropy": 5.850178670883179,
"epoch": 0.41273584905660377,
"grad_norm": 1.640625,
"learning_rate": 0.0004988288749009264,
"loss": 5.7185,
"mean_token_accuracy": 0.15071613490581512,
"num_tokens": 9448669.0,
"step": 4550
},
{
"entropy": 5.810451078414917,
"epoch": 0.413189404934688,
"grad_norm": 2.671875,
"learning_rate": 0.0004988255755786608,
"loss": 5.7114,
"mean_token_accuracy": 0.14924689680337905,
"num_tokens": 9458857.0,
"step": 4555
},
{
"entropy": 5.801498889923096,
"epoch": 0.41364296081277213,
"grad_norm": 2.796875,
"learning_rate": 0.0004988222716276251,
"loss": 5.7071,
"mean_token_accuracy": 0.15637744963169098,
"num_tokens": 9470241.0,
"step": 4560
},
{
"entropy": 5.892170333862305,
"epoch": 0.4140965166908563,
"grad_norm": 2.5625,
"learning_rate": 0.000498818963047888,
"loss": 5.8641,
"mean_token_accuracy": 0.14676743596792222,
"num_tokens": 9480528.0,
"step": 4565
},
{
"entropy": 5.931382036209106,
"epoch": 0.4145500725689405,
"grad_norm": 2.0625,
"learning_rate": 0.0004988156498395176,
"loss": 5.786,
"mean_token_accuracy": 0.14832354038953782,
"num_tokens": 9491000.0,
"step": 4570
},
{
"entropy": 5.862825870513916,
"epoch": 0.41500362844702465,
"grad_norm": 1.921875,
"learning_rate": 0.0004988123320025825,
"loss": 5.7668,
"mean_token_accuracy": 0.1513970524072647,
"num_tokens": 9501612.0,
"step": 4575
},
{
"entropy": 5.8863836288452145,
"epoch": 0.41545718432510886,
"grad_norm": 1.734375,
"learning_rate": 0.0004988090095371514,
"loss": 5.8554,
"mean_token_accuracy": 0.14679653570055962,
"num_tokens": 9513999.0,
"step": 4580
},
{
"entropy": 5.8441126346588135,
"epoch": 0.415910740203193,
"grad_norm": 1.78125,
"learning_rate": 0.0004988056824432928,
"loss": 5.7543,
"mean_token_accuracy": 0.14250365793704986,
"num_tokens": 9524392.0,
"step": 4585
},
{
"entropy": 5.861783695220947,
"epoch": 0.41636429608127723,
"grad_norm": 1.8828125,
"learning_rate": 0.0004988023507210758,
"loss": 5.7453,
"mean_token_accuracy": 0.14709829092025756,
"num_tokens": 9534227.0,
"step": 4590
},
{
"entropy": 5.811742496490479,
"epoch": 0.4168178519593614,
"grad_norm": 1.8515625,
"learning_rate": 0.0004987990143705692,
"loss": 5.7413,
"mean_token_accuracy": 0.15399961918592453,
"num_tokens": 9544242.0,
"step": 4595
},
{
"entropy": 5.838329458236695,
"epoch": 0.4172714078374456,
"grad_norm": 2.03125,
"learning_rate": 0.0004987956733918418,
"loss": 5.6717,
"mean_token_accuracy": 0.1506224110722542,
"num_tokens": 9553921.0,
"step": 4600
},
{
"entropy": 5.757907295227051,
"epoch": 0.41772496371552975,
"grad_norm": 1.953125,
"learning_rate": 0.0004987923277849629,
"loss": 5.7169,
"mean_token_accuracy": 0.14515915364027024,
"num_tokens": 9564768.0,
"step": 4605
},
{
"entropy": 5.806964540481568,
"epoch": 0.4181785195936139,
"grad_norm": 1.609375,
"learning_rate": 0.0004987889775500017,
"loss": 5.7797,
"mean_token_accuracy": 0.14279748350381852,
"num_tokens": 9576360.0,
"step": 4610
},
{
"entropy": 5.898333644866943,
"epoch": 0.4186320754716981,
"grad_norm": 2.03125,
"learning_rate": 0.0004987856226870272,
"loss": 5.7643,
"mean_token_accuracy": 0.14810260534286498,
"num_tokens": 9587623.0,
"step": 4615
},
{
"entropy": 5.835229158401489,
"epoch": 0.4190856313497823,
"grad_norm": 2.109375,
"learning_rate": 0.0004987822631961092,
"loss": 5.7959,
"mean_token_accuracy": 0.14757825285196305,
"num_tokens": 9597837.0,
"step": 4620
},
{
"entropy": 5.865486764907837,
"epoch": 0.4195391872278665,
"grad_norm": 5.03125,
"learning_rate": 0.000498778899077317,
"loss": 5.7521,
"mean_token_accuracy": 0.1463379107415676,
"num_tokens": 9609509.0,
"step": 4625
},
{
"entropy": 5.824074077606201,
"epoch": 0.41999274310595064,
"grad_norm": 1.9765625,
"learning_rate": 0.00049877553033072,
"loss": 5.7348,
"mean_token_accuracy": 0.15134868174791336,
"num_tokens": 9619954.0,
"step": 4630
},
{
"entropy": 5.8004528999328615,
"epoch": 0.42044629898403485,
"grad_norm": 1.875,
"learning_rate": 0.0004987721569563882,
"loss": 5.7403,
"mean_token_accuracy": 0.14655848294496537,
"num_tokens": 9629742.0,
"step": 4635
},
{
"entropy": 5.837172031402588,
"epoch": 0.420899854862119,
"grad_norm": 1.6484375,
"learning_rate": 0.000498768778954391,
"loss": 5.7608,
"mean_token_accuracy": 0.14900261014699936,
"num_tokens": 9641378.0,
"step": 4640
},
{
"entropy": 5.948794746398926,
"epoch": 0.4213534107402032,
"grad_norm": 1.7109375,
"learning_rate": 0.0004987653963247985,
"loss": 5.8219,
"mean_token_accuracy": 0.14978400021791458,
"num_tokens": 9652279.0,
"step": 4645
},
{
"entropy": 5.867143392562866,
"epoch": 0.42180696661828737,
"grad_norm": 1.71875,
"learning_rate": 0.0004987620090676805,
"loss": 5.838,
"mean_token_accuracy": 0.14259155243635177,
"num_tokens": 9663416.0,
"step": 4650
},
{
"entropy": 5.858006381988526,
"epoch": 0.4222605224963715,
"grad_norm": 1.5703125,
"learning_rate": 0.0004987586171831072,
"loss": 5.7857,
"mean_token_accuracy": 0.1458583153784275,
"num_tokens": 9674647.0,
"step": 4655
},
{
"entropy": 5.827457094192505,
"epoch": 0.42271407837445574,
"grad_norm": 1.5,
"learning_rate": 0.0004987552206711487,
"loss": 5.625,
"mean_token_accuracy": 0.1527285732328892,
"num_tokens": 9685084.0,
"step": 4660
},
{
"entropy": 5.700690031051636,
"epoch": 0.4231676342525399,
"grad_norm": 1.75,
"learning_rate": 0.0004987518195318752,
"loss": 5.6882,
"mean_token_accuracy": 0.1495150238275528,
"num_tokens": 9695661.0,
"step": 4665
},
{
"entropy": 5.858991527557373,
"epoch": 0.4236211901306241,
"grad_norm": 1.671875,
"learning_rate": 0.000498748413765357,
"loss": 5.7407,
"mean_token_accuracy": 0.14949771612882615,
"num_tokens": 9705773.0,
"step": 4670
},
{
"entropy": 5.8706464767456055,
"epoch": 0.42407474600870826,
"grad_norm": 1.4453125,
"learning_rate": 0.0004987450033716646,
"loss": 5.8052,
"mean_token_accuracy": 0.14631288051605223,
"num_tokens": 9715799.0,
"step": 4675
},
{
"entropy": 5.8481348037719725,
"epoch": 0.42452830188679247,
"grad_norm": 1.96875,
"learning_rate": 0.0004987415883508686,
"loss": 5.7683,
"mean_token_accuracy": 0.15039038732647897,
"num_tokens": 9727012.0,
"step": 4680
},
{
"entropy": 5.842692613601685,
"epoch": 0.4249818577648766,
"grad_norm": 1.4296875,
"learning_rate": 0.0004987381687030394,
"loss": 5.7363,
"mean_token_accuracy": 0.148531074821949,
"num_tokens": 9737044.0,
"step": 4685
},
{
"entropy": 5.8995438575744625,
"epoch": 0.42543541364296084,
"grad_norm": 2.171875,
"learning_rate": 0.0004987347444282479,
"loss": 5.8423,
"mean_token_accuracy": 0.1406104564666748,
"num_tokens": 9747648.0,
"step": 4690
},
{
"entropy": 5.810405111312866,
"epoch": 0.425888969521045,
"grad_norm": 2.734375,
"learning_rate": 0.0004987313155265649,
"loss": 5.7585,
"mean_token_accuracy": 0.14476184248924256,
"num_tokens": 9757812.0,
"step": 4695
},
{
"entropy": 5.901001262664795,
"epoch": 0.42634252539912915,
"grad_norm": 1.7890625,
"learning_rate": 0.0004987278819980613,
"loss": 5.7523,
"mean_token_accuracy": 0.15064995884895324,
"num_tokens": 9768337.0,
"step": 4700
},
{
"entropy": 5.868827390670776,
"epoch": 0.42679608127721336,
"grad_norm": 1.5390625,
"learning_rate": 0.0004987244438428081,
"loss": 5.7938,
"mean_token_accuracy": 0.14378955513238906,
"num_tokens": 9779433.0,
"step": 4705
},
{
"entropy": 5.774157524108887,
"epoch": 0.4272496371552975,
"grad_norm": 2.28125,
"learning_rate": 0.0004987210010608764,
"loss": 5.6983,
"mean_token_accuracy": 0.14819056391716004,
"num_tokens": 9790069.0,
"step": 4710
},
{
"entropy": 5.828324174880981,
"epoch": 0.4277031930333817,
"grad_norm": 1.9140625,
"learning_rate": 0.0004987175536523373,
"loss": 5.6902,
"mean_token_accuracy": 0.15661128610372543,
"num_tokens": 9799669.0,
"step": 4715
},
{
"entropy": 5.810670280456543,
"epoch": 0.4281567489114659,
"grad_norm": 1.546875,
"learning_rate": 0.0004987141016172622,
"loss": 5.7101,
"mean_token_accuracy": 0.15046556890010834,
"num_tokens": 9809615.0,
"step": 4720
},
{
"entropy": 5.774604225158692,
"epoch": 0.4286103047895501,
"grad_norm": 3.5625,
"learning_rate": 0.0004987106449557225,
"loss": 5.6934,
"mean_token_accuracy": 0.14824203699827193,
"num_tokens": 9819666.0,
"step": 4725
},
{
"entropy": 5.898928689956665,
"epoch": 0.42906386066763424,
"grad_norm": 1.8828125,
"learning_rate": 0.0004987071836677896,
"loss": 5.8448,
"mean_token_accuracy": 0.14774569869041443,
"num_tokens": 9831015.0,
"step": 4730
},
{
"entropy": 5.842265319824219,
"epoch": 0.42951741654571846,
"grad_norm": 1.421875,
"learning_rate": 0.0004987037177535353,
"loss": 5.7116,
"mean_token_accuracy": 0.1472062274813652,
"num_tokens": 9841354.0,
"step": 4735
},
{
"entropy": 5.716348314285279,
"epoch": 0.4299709724238026,
"grad_norm": 3.171875,
"learning_rate": 0.0004987002472130309,
"loss": 5.7335,
"mean_token_accuracy": 0.14573415368795395,
"num_tokens": 9850467.0,
"step": 4740
},
{
"entropy": 5.853896808624268,
"epoch": 0.43042452830188677,
"grad_norm": 1.5625,
"learning_rate": 0.0004986967720463485,
"loss": 5.7836,
"mean_token_accuracy": 0.14540919959545134,
"num_tokens": 9861115.0,
"step": 4745
},
{
"entropy": 5.881145763397217,
"epoch": 0.430878084179971,
"grad_norm": 1.6640625,
"learning_rate": 0.0004986932922535598,
"loss": 5.7928,
"mean_token_accuracy": 0.1494162768125534,
"num_tokens": 9871172.0,
"step": 4750
},
{
"entropy": 5.8132791996002195,
"epoch": 0.43133164005805513,
"grad_norm": 1.71875,
"learning_rate": 0.0004986898078347368,
"loss": 5.6845,
"mean_token_accuracy": 0.15211455523967743,
"num_tokens": 9881997.0,
"step": 4755
},
{
"entropy": 5.795827627182007,
"epoch": 0.43178519593613934,
"grad_norm": 1.609375,
"learning_rate": 0.0004986863187899516,
"loss": 5.6945,
"mean_token_accuracy": 0.15255386531352996,
"num_tokens": 9892310.0,
"step": 4760
},
{
"entropy": 5.739163732528686,
"epoch": 0.4322387518142235,
"grad_norm": 1.65625,
"learning_rate": 0.0004986828251192764,
"loss": 5.7317,
"mean_token_accuracy": 0.15006684958934785,
"num_tokens": 9902664.0,
"step": 4765
},
{
"entropy": 5.765441131591797,
"epoch": 0.4326923076923077,
"grad_norm": 2.265625,
"learning_rate": 0.0004986793268227833,
"loss": 5.7226,
"mean_token_accuracy": 0.14934629648923875,
"num_tokens": 9912491.0,
"step": 4770
},
{
"entropy": 5.810837459564209,
"epoch": 0.43314586357039186,
"grad_norm": 1.734375,
"learning_rate": 0.0004986758239005448,
"loss": 5.6559,
"mean_token_accuracy": 0.15213295966386794,
"num_tokens": 9922057.0,
"step": 4775
},
{
"entropy": 5.772084760665893,
"epoch": 0.4335994194484761,
"grad_norm": 1.5625,
"learning_rate": 0.0004986723163526331,
"loss": 5.7265,
"mean_token_accuracy": 0.1473941296339035,
"num_tokens": 9933661.0,
"step": 4780
},
{
"entropy": 5.798309230804444,
"epoch": 0.43405297532656023,
"grad_norm": 1.5,
"learning_rate": 0.0004986688041791209,
"loss": 5.7003,
"mean_token_accuracy": 0.14621387347579,
"num_tokens": 9945177.0,
"step": 4785
},
{
"entropy": 5.851141834259034,
"epoch": 0.4345065312046444,
"grad_norm": 1.78125,
"learning_rate": 0.000498665287380081,
"loss": 5.7308,
"mean_token_accuracy": 0.14761214107275009,
"num_tokens": 9956226.0,
"step": 4790
},
{
"entropy": 5.874855804443359,
"epoch": 0.4349600870827286,
"grad_norm": 1.9296875,
"learning_rate": 0.0004986617659555859,
"loss": 5.7916,
"mean_token_accuracy": 0.14956312030553817,
"num_tokens": 9965877.0,
"step": 4795
},
{
"entropy": 5.805899953842163,
"epoch": 0.43541364296081275,
"grad_norm": 1.84375,
"learning_rate": 0.0004986582399057085,
"loss": 5.7519,
"mean_token_accuracy": 0.160671429336071,
"num_tokens": 9975592.0,
"step": 4800
},
{
"entropy": 5.817458915710449,
"epoch": 0.43586719883889696,
"grad_norm": 1.78125,
"learning_rate": 0.0004986547092305216,
"loss": 5.7795,
"mean_token_accuracy": 0.14750458002090455,
"num_tokens": 9986094.0,
"step": 4805
},
{
"entropy": 5.865862941741943,
"epoch": 0.4363207547169811,
"grad_norm": 1.96875,
"learning_rate": 0.0004986511739300984,
"loss": 5.8163,
"mean_token_accuracy": 0.14368128031492233,
"num_tokens": 9997241.0,
"step": 4810
},
{
"entropy": 5.855591487884522,
"epoch": 0.43677431059506533,
"grad_norm": 1.734375,
"learning_rate": 0.0004986476340045119,
"loss": 5.7254,
"mean_token_accuracy": 0.1495523065328598,
"num_tokens": 10008091.0,
"step": 4815
},
{
"entropy": 5.795777940750122,
"epoch": 0.4372278664731495,
"grad_norm": 1.9921875,
"learning_rate": 0.0004986440894538354,
"loss": 5.7912,
"mean_token_accuracy": 0.14798561334609986,
"num_tokens": 10018794.0,
"step": 4820
},
{
"entropy": 5.776800107955933,
"epoch": 0.4376814223512337,
"grad_norm": 2.359375,
"learning_rate": 0.0004986405402781421,
"loss": 5.6904,
"mean_token_accuracy": 0.14998709559440612,
"num_tokens": 10028414.0,
"step": 4825
},
{
"entropy": 5.859041595458985,
"epoch": 0.43813497822931785,
"grad_norm": 1.7734375,
"learning_rate": 0.0004986369864775054,
"loss": 5.7709,
"mean_token_accuracy": 0.14849811941385269,
"num_tokens": 10038749.0,
"step": 4830
},
{
"entropy": 5.881982755661011,
"epoch": 0.438588534107402,
"grad_norm": 1.8828125,
"learning_rate": 0.0004986334280519989,
"loss": 5.7227,
"mean_token_accuracy": 0.14857089072465895,
"num_tokens": 10049866.0,
"step": 4835
},
{
"entropy": 5.816167640686035,
"epoch": 0.4390420899854862,
"grad_norm": 2.53125,
"learning_rate": 0.000498629865001696,
"loss": 5.7514,
"mean_token_accuracy": 0.1548391565680504,
"num_tokens": 10060680.0,
"step": 4840
},
{
"entropy": 5.862650632858276,
"epoch": 0.43949564586357037,
"grad_norm": 1.5625,
"learning_rate": 0.0004986262973266706,
"loss": 5.8481,
"mean_token_accuracy": 0.14525855779647828,
"num_tokens": 10070931.0,
"step": 4845
},
{
"entropy": 5.7950972557067875,
"epoch": 0.4399492017416546,
"grad_norm": 2.015625,
"learning_rate": 0.0004986227250269964,
"loss": 5.7682,
"mean_token_accuracy": 0.1551634520292282,
"num_tokens": 10080189.0,
"step": 4850
},
{
"entropy": 5.814848613739014,
"epoch": 0.44040275761973874,
"grad_norm": 2.46875,
"learning_rate": 0.0004986191481027472,
"loss": 5.6731,
"mean_token_accuracy": 0.15222601294517518,
"num_tokens": 10088813.0,
"step": 4855
},
{
"entropy": 5.785250854492188,
"epoch": 0.44085631349782295,
"grad_norm": 3.125,
"learning_rate": 0.0004986155665539971,
"loss": 5.7113,
"mean_token_accuracy": 0.15193801522254943,
"num_tokens": 10098125.0,
"step": 4860
},
{
"entropy": 5.8375824928283695,
"epoch": 0.4413098693759071,
"grad_norm": 1.71875,
"learning_rate": 0.00049861198038082,
"loss": 5.7335,
"mean_token_accuracy": 0.1524299070239067,
"num_tokens": 10109125.0,
"step": 4865
},
{
"entropy": 5.789612865447998,
"epoch": 0.4417634252539913,
"grad_norm": 1.78125,
"learning_rate": 0.0004986083895832903,
"loss": 5.7042,
"mean_token_accuracy": 0.15111606419086457,
"num_tokens": 10120182.0,
"step": 4870
},
{
"entropy": 5.803175401687622,
"epoch": 0.44221698113207547,
"grad_norm": 1.3984375,
"learning_rate": 0.0004986047941614821,
"loss": 5.7837,
"mean_token_accuracy": 0.14436635076999665,
"num_tokens": 10130941.0,
"step": 4875
},
{
"entropy": 5.944940805435181,
"epoch": 0.4426705370101596,
"grad_norm": 1.796875,
"learning_rate": 0.0004986011941154696,
"loss": 5.7612,
"mean_token_accuracy": 0.14681328535079957,
"num_tokens": 10141708.0,
"step": 4880
},
{
"entropy": 5.836460781097412,
"epoch": 0.44312409288824384,
"grad_norm": 1.4921875,
"learning_rate": 0.0004985975894453275,
"loss": 5.6717,
"mean_token_accuracy": 0.1519172064960003,
"num_tokens": 10151816.0,
"step": 4885
},
{
"entropy": 5.8214634418487545,
"epoch": 0.443577648766328,
"grad_norm": 1.609375,
"learning_rate": 0.0004985939801511304,
"loss": 5.7987,
"mean_token_accuracy": 0.15131675601005554,
"num_tokens": 10162470.0,
"step": 4890
},
{
"entropy": 5.886533737182617,
"epoch": 0.4440312046444122,
"grad_norm": 1.59375,
"learning_rate": 0.0004985903662329527,
"loss": 5.6955,
"mean_token_accuracy": 0.1570177286863327,
"num_tokens": 10173171.0,
"step": 4895
},
{
"entropy": 5.8201783180236815,
"epoch": 0.44448476052249636,
"grad_norm": 1.96875,
"learning_rate": 0.0004985867476908693,
"loss": 5.7265,
"mean_token_accuracy": 0.1524544969201088,
"num_tokens": 10184185.0,
"step": 4900
},
{
"entropy": 5.7874305725097654,
"epoch": 0.44493831640058057,
"grad_norm": 1.453125,
"learning_rate": 0.0004985831245249549,
"loss": 5.7848,
"mean_token_accuracy": 0.14795219525694847,
"num_tokens": 10195516.0,
"step": 4905
},
{
"entropy": 5.796139764785766,
"epoch": 0.4453918722786647,
"grad_norm": 3.34375,
"learning_rate": 0.0004985794967352846,
"loss": 5.7202,
"mean_token_accuracy": 0.15106817930936814,
"num_tokens": 10204936.0,
"step": 4910
},
{
"entropy": 5.855431652069091,
"epoch": 0.44584542815674894,
"grad_norm": 1.4375,
"learning_rate": 0.0004985758643219334,
"loss": 5.7589,
"mean_token_accuracy": 0.14915177449584008,
"num_tokens": 10216218.0,
"step": 4915
},
{
"entropy": 5.896589231491089,
"epoch": 0.4462989840348331,
"grad_norm": 1.71875,
"learning_rate": 0.0004985722272849762,
"loss": 5.7595,
"mean_token_accuracy": 0.1486516922712326,
"num_tokens": 10226370.0,
"step": 4920
},
{
"entropy": 5.761085844039917,
"epoch": 0.44675253991291725,
"grad_norm": 1.609375,
"learning_rate": 0.0004985685856244884,
"loss": 5.6748,
"mean_token_accuracy": 0.15937331169843674,
"num_tokens": 10236170.0,
"step": 4925
},
{
"entropy": 5.8097490787506105,
"epoch": 0.44720609579100146,
"grad_norm": 3.734375,
"learning_rate": 0.0004985649393405453,
"loss": 5.7641,
"mean_token_accuracy": 0.1467866986989975,
"num_tokens": 10246119.0,
"step": 4930
},
{
"entropy": 5.838061428070068,
"epoch": 0.4476596516690856,
"grad_norm": 2.28125,
"learning_rate": 0.0004985612884332223,
"loss": 5.7597,
"mean_token_accuracy": 0.14624503403902053,
"num_tokens": 10256762.0,
"step": 4935
},
{
"entropy": 5.759787511825562,
"epoch": 0.4481132075471698,
"grad_norm": 1.6015625,
"learning_rate": 0.0004985576329025949,
"loss": 5.7058,
"mean_token_accuracy": 0.1520256593823433,
"num_tokens": 10267317.0,
"step": 4940
},
{
"entropy": 5.869536733627319,
"epoch": 0.448566763425254,
"grad_norm": 2.484375,
"learning_rate": 0.0004985539727487385,
"loss": 5.8374,
"mean_token_accuracy": 0.13999374434351922,
"num_tokens": 10279111.0,
"step": 4945
},
{
"entropy": 5.854530906677246,
"epoch": 0.4490203193033382,
"grad_norm": 3.28125,
"learning_rate": 0.0004985503079717292,
"loss": 5.718,
"mean_token_accuracy": 0.15307011604309081,
"num_tokens": 10289383.0,
"step": 4950
},
{
"entropy": 5.875187730789184,
"epoch": 0.44947387518142234,
"grad_norm": 2.359375,
"learning_rate": 0.0004985466385716425,
"loss": 5.7275,
"mean_token_accuracy": 0.1502274066209793,
"num_tokens": 10299806.0,
"step": 4955
},
{
"entropy": 5.806089878082275,
"epoch": 0.44992743105950656,
"grad_norm": 2.890625,
"learning_rate": 0.0004985429645485543,
"loss": 5.7382,
"mean_token_accuracy": 0.1461896352469921,
"num_tokens": 10310348.0,
"step": 4960
},
{
"entropy": 5.827132749557495,
"epoch": 0.4503809869375907,
"grad_norm": 2.265625,
"learning_rate": 0.0004985392859025407,
"loss": 5.7644,
"mean_token_accuracy": 0.1516891375184059,
"num_tokens": 10320354.0,
"step": 4965
},
{
"entropy": 5.757310485839843,
"epoch": 0.45083454281567487,
"grad_norm": 1.7734375,
"learning_rate": 0.0004985356026336776,
"loss": 5.6703,
"mean_token_accuracy": 0.1507007971405983,
"num_tokens": 10330292.0,
"step": 4970
},
{
"entropy": 5.858229494094848,
"epoch": 0.4512880986937591,
"grad_norm": 1.8984375,
"learning_rate": 0.0004985319147420414,
"loss": 5.6474,
"mean_token_accuracy": 0.15889344215393067,
"num_tokens": 10340449.0,
"step": 4975
},
{
"entropy": 5.822214269638062,
"epoch": 0.45174165457184323,
"grad_norm": 2.28125,
"learning_rate": 0.0004985282222277082,
"loss": 5.6964,
"mean_token_accuracy": 0.14781344383955003,
"num_tokens": 10350637.0,
"step": 4980
},
{
"entropy": 5.680160665512085,
"epoch": 0.45219521044992744,
"grad_norm": 2.0,
"learning_rate": 0.0004985245250907544,
"loss": 5.6621,
"mean_token_accuracy": 0.15216653496026994,
"num_tokens": 10361548.0,
"step": 4985
},
{
"entropy": 5.7882819175720215,
"epoch": 0.4526487663280116,
"grad_norm": 1.984375,
"learning_rate": 0.0004985208233312564,
"loss": 5.6475,
"mean_token_accuracy": 0.14174009189009668,
"num_tokens": 10371765.0,
"step": 4990
},
{
"entropy": 5.8632818222045895,
"epoch": 0.4531023222060958,
"grad_norm": 1.828125,
"learning_rate": 0.0004985171169492908,
"loss": 5.7738,
"mean_token_accuracy": 0.15304608941078185,
"num_tokens": 10382108.0,
"step": 4995
},
{
"entropy": 5.799547719955444,
"epoch": 0.45355587808417996,
"grad_norm": 1.59375,
"learning_rate": 0.0004985134059449344,
"loss": 5.7167,
"mean_token_accuracy": 0.15100618749856948,
"num_tokens": 10392390.0,
"step": 5000
},
{
"entropy": 5.829653596878051,
"epoch": 0.4540094339622642,
"grad_norm": 1.6484375,
"learning_rate": 0.0004985096903182637,
"loss": 5.7569,
"mean_token_accuracy": 0.15035886466503143,
"num_tokens": 10402094.0,
"step": 5005
},
{
"entropy": 5.823336267471314,
"epoch": 0.45446298984034833,
"grad_norm": 1.5859375,
"learning_rate": 0.0004985059700693557,
"loss": 5.7002,
"mean_token_accuracy": 0.15384241491556166,
"num_tokens": 10413350.0,
"step": 5010
},
{
"entropy": 5.836109685897827,
"epoch": 0.4549165457184325,
"grad_norm": 1.65625,
"learning_rate": 0.0004985022451982872,
"loss": 5.8458,
"mean_token_accuracy": 0.14465131014585494,
"num_tokens": 10424292.0,
"step": 5015
},
{
"entropy": 5.888298034667969,
"epoch": 0.4553701015965167,
"grad_norm": 1.578125,
"learning_rate": 0.0004984985157051354,
"loss": 5.6868,
"mean_token_accuracy": 0.15214828550815582,
"num_tokens": 10433724.0,
"step": 5020
},
{
"entropy": 5.819750642776489,
"epoch": 0.45582365747460085,
"grad_norm": 2.953125,
"learning_rate": 0.0004984947815899774,
"loss": 5.779,
"mean_token_accuracy": 0.14743213057518006,
"num_tokens": 10444724.0,
"step": 5025
},
{
"entropy": 5.838175058364868,
"epoch": 0.45627721335268506,
"grad_norm": 1.859375,
"learning_rate": 0.0004984910428528902,
"loss": 5.7053,
"mean_token_accuracy": 0.15772704035043716,
"num_tokens": 10455086.0,
"step": 5030
},
{
"entropy": 5.7590735912322994,
"epoch": 0.4567307692307692,
"grad_norm": 2.375,
"learning_rate": 0.0004984872994939514,
"loss": 5.6661,
"mean_token_accuracy": 0.1486493766307831,
"num_tokens": 10465076.0,
"step": 5035
},
{
"entropy": 5.837244844436645,
"epoch": 0.45718432510885343,
"grad_norm": 1.609375,
"learning_rate": 0.0004984835515132382,
"loss": 5.7028,
"mean_token_accuracy": 0.1532436266541481,
"num_tokens": 10474817.0,
"step": 5040
},
{
"entropy": 5.835756492614746,
"epoch": 0.4576378809869376,
"grad_norm": 1.7421875,
"learning_rate": 0.0004984797989108282,
"loss": 5.7118,
"mean_token_accuracy": 0.14745451956987382,
"num_tokens": 10485735.0,
"step": 5045
},
{
"entropy": 5.790848636627198,
"epoch": 0.4580914368650218,
"grad_norm": 1.5546875,
"learning_rate": 0.0004984760416867991,
"loss": 5.7919,
"mean_token_accuracy": 0.14835846349596976,
"num_tokens": 10496731.0,
"step": 5050
},
{
"entropy": 5.760481691360473,
"epoch": 0.45854499274310595,
"grad_norm": 3.8125,
"learning_rate": 0.0004984722798412286,
"loss": 5.6739,
"mean_token_accuracy": 0.14979240894317628,
"num_tokens": 10506787.0,
"step": 5055
},
{
"entropy": 5.826909589767456,
"epoch": 0.4589985486211901,
"grad_norm": 1.859375,
"learning_rate": 0.0004984685133741941,
"loss": 5.7387,
"mean_token_accuracy": 0.1452489860355854,
"num_tokens": 10517170.0,
"step": 5060
},
{
"entropy": 5.772784662246704,
"epoch": 0.4594521044992743,
"grad_norm": 1.875,
"learning_rate": 0.000498464742285774,
"loss": 5.7163,
"mean_token_accuracy": 0.15115534663200378,
"num_tokens": 10527342.0,
"step": 5065
},
{
"entropy": 5.81494026184082,
"epoch": 0.45990566037735847,
"grad_norm": 1.5,
"learning_rate": 0.000498460966576046,
"loss": 5.7022,
"mean_token_accuracy": 0.15441648811101913,
"num_tokens": 10537253.0,
"step": 5070
},
{
"entropy": 5.858026695251465,
"epoch": 0.4603592162554427,
"grad_norm": 1.59375,
"learning_rate": 0.0004984571862450884,
"loss": 5.7309,
"mean_token_accuracy": 0.15261692702770233,
"num_tokens": 10547249.0,
"step": 5075
},
{
"entropy": 5.862608909606934,
"epoch": 0.46081277213352684,
"grad_norm": 1.390625,
"learning_rate": 0.0004984534012929791,
"loss": 5.7598,
"mean_token_accuracy": 0.14884943068027495,
"num_tokens": 10558174.0,
"step": 5080
},
{
"entropy": 5.871862173080444,
"epoch": 0.46126632801161105,
"grad_norm": 2.265625,
"learning_rate": 0.0004984496117197967,
"loss": 5.721,
"mean_token_accuracy": 0.15395016670227052,
"num_tokens": 10567610.0,
"step": 5085
},
{
"entropy": 5.7651379108428955,
"epoch": 0.4617198838896952,
"grad_norm": 1.8515625,
"learning_rate": 0.0004984458175256193,
"loss": 5.7356,
"mean_token_accuracy": 0.14666065573692322,
"num_tokens": 10578494.0,
"step": 5090
},
{
"entropy": 5.876489353179932,
"epoch": 0.4621734397677794,
"grad_norm": 2.171875,
"learning_rate": 0.0004984420187105253,
"loss": 5.8277,
"mean_token_accuracy": 0.14741355702280998,
"num_tokens": 10589482.0,
"step": 5095
},
{
"entropy": 5.92623553276062,
"epoch": 0.46262699564586357,
"grad_norm": 1.84375,
"learning_rate": 0.0004984382152745936,
"loss": 5.7812,
"mean_token_accuracy": 0.14485349804162978,
"num_tokens": 10599965.0,
"step": 5100
},
{
"entropy": 5.797193765640259,
"epoch": 0.4630805515239477,
"grad_norm": 1.796875,
"learning_rate": 0.0004984344072179026,
"loss": 5.8158,
"mean_token_accuracy": 0.15118358284235,
"num_tokens": 10610500.0,
"step": 5105
},
{
"entropy": 5.832291603088379,
"epoch": 0.46353410740203194,
"grad_norm": 1.703125,
"learning_rate": 0.0004984305945405312,
"loss": 5.7608,
"mean_token_accuracy": 0.15085830986499787,
"num_tokens": 10619843.0,
"step": 5110
},
{
"entropy": 5.8759160995483395,
"epoch": 0.4639876632801161,
"grad_norm": 2.546875,
"learning_rate": 0.000498426777242558,
"loss": 5.6799,
"mean_token_accuracy": 0.1561095081269741,
"num_tokens": 10631026.0,
"step": 5115
},
{
"entropy": 5.648857736587525,
"epoch": 0.4644412191582003,
"grad_norm": 2.078125,
"learning_rate": 0.0004984229553240623,
"loss": 5.6994,
"mean_token_accuracy": 0.15481915920972825,
"num_tokens": 10641130.0,
"step": 5120
},
{
"entropy": 5.7813159942626955,
"epoch": 0.46489477503628446,
"grad_norm": 2.09375,
"learning_rate": 0.0004984191287851228,
"loss": 5.7541,
"mean_token_accuracy": 0.1515883758664131,
"num_tokens": 10652305.0,
"step": 5125
},
{
"entropy": 5.945868587493896,
"epoch": 0.46534833091436867,
"grad_norm": 1.7421875,
"learning_rate": 0.0004984152976258188,
"loss": 5.7772,
"mean_token_accuracy": 0.15462889671325683,
"num_tokens": 10663083.0,
"step": 5130
},
{
"entropy": 5.808717489242554,
"epoch": 0.4658018867924528,
"grad_norm": 1.9375,
"learning_rate": 0.0004984114618462296,
"loss": 5.8091,
"mean_token_accuracy": 0.14427706003189086,
"num_tokens": 10673602.0,
"step": 5135
},
{
"entropy": 5.731534767150879,
"epoch": 0.46625544267053703,
"grad_norm": 1.5234375,
"learning_rate": 0.0004984076214464343,
"loss": 5.6296,
"mean_token_accuracy": 0.15683090388774873,
"num_tokens": 10684622.0,
"step": 5140
},
{
"entropy": 5.81270661354065,
"epoch": 0.4667089985486212,
"grad_norm": 1.7890625,
"learning_rate": 0.0004984037764265126,
"loss": 5.7507,
"mean_token_accuracy": 0.14948814660310744,
"num_tokens": 10695129.0,
"step": 5145
},
{
"entropy": 5.793086814880371,
"epoch": 0.46716255442670535,
"grad_norm": 1.90625,
"learning_rate": 0.0004983999267865438,
"loss": 5.6087,
"mean_token_accuracy": 0.15965820848941803,
"num_tokens": 10705236.0,
"step": 5150
},
{
"entropy": 5.784671878814697,
"epoch": 0.46761611030478956,
"grad_norm": 2.140625,
"learning_rate": 0.0004983960725266078,
"loss": 5.7217,
"mean_token_accuracy": 0.14745316356420518,
"num_tokens": 10715748.0,
"step": 5155
},
{
"entropy": 5.831699419021606,
"epoch": 0.4680696661828737,
"grad_norm": 1.5625,
"learning_rate": 0.0004983922136467838,
"loss": 5.6555,
"mean_token_accuracy": 0.15694286674261093,
"num_tokens": 10725799.0,
"step": 5160
},
{
"entropy": 5.813610506057739,
"epoch": 0.4685232220609579,
"grad_norm": 1.8046875,
"learning_rate": 0.000498388350147152,
"loss": 5.751,
"mean_token_accuracy": 0.14641758650541306,
"num_tokens": 10735668.0,
"step": 5165
},
{
"entropy": 5.861870384216308,
"epoch": 0.4689767779390421,
"grad_norm": 2.171875,
"learning_rate": 0.0004983844820277922,
"loss": 5.87,
"mean_token_accuracy": 0.1426786296069622,
"num_tokens": 10746466.0,
"step": 5170
},
{
"entropy": 5.888263511657715,
"epoch": 0.4694303338171263,
"grad_norm": 3.140625,
"learning_rate": 0.0004983806092887843,
"loss": 5.7561,
"mean_token_accuracy": 0.15089037865400315,
"num_tokens": 10756724.0,
"step": 5175
},
{
"entropy": 5.855864477157593,
"epoch": 0.46988388969521044,
"grad_norm": 2.546875,
"learning_rate": 0.0004983767319302086,
"loss": 5.8112,
"mean_token_accuracy": 0.1499728061258793,
"num_tokens": 10766919.0,
"step": 5180
},
{
"entropy": 5.766339826583862,
"epoch": 0.47033744557329465,
"grad_norm": 2.203125,
"learning_rate": 0.0004983728499521451,
"loss": 5.6087,
"mean_token_accuracy": 0.1593857705593109,
"num_tokens": 10777550.0,
"step": 5185
},
{
"entropy": 5.767188262939453,
"epoch": 0.4707910014513788,
"grad_norm": 1.625,
"learning_rate": 0.0004983689633546741,
"loss": 5.6586,
"mean_token_accuracy": 0.14953138381242753,
"num_tokens": 10788456.0,
"step": 5190
},
{
"entropy": 5.820877647399902,
"epoch": 0.47124455732946297,
"grad_norm": 1.6328125,
"learning_rate": 0.0004983650721378761,
"loss": 5.7126,
"mean_token_accuracy": 0.1494821533560753,
"num_tokens": 10800049.0,
"step": 5195
},
{
"entropy": 5.8708991527557375,
"epoch": 0.4716981132075472,
"grad_norm": 1.609375,
"learning_rate": 0.0004983611763018314,
"loss": 5.7514,
"mean_token_accuracy": 0.1483040362596512,
"num_tokens": 10810743.0,
"step": 5200
},
{
"entropy": 5.935766887664795,
"epoch": 0.47215166908563133,
"grad_norm": 1.5078125,
"learning_rate": 0.0004983572758466208,
"loss": 5.9259,
"mean_token_accuracy": 0.14382343962788582,
"num_tokens": 10821897.0,
"step": 5205
},
{
"entropy": 5.911200714111328,
"epoch": 0.47260522496371554,
"grad_norm": 2.203125,
"learning_rate": 0.0004983533707723247,
"loss": 5.78,
"mean_token_accuracy": 0.15174445733428002,
"num_tokens": 10831589.0,
"step": 5210
},
{
"entropy": 5.881593132019043,
"epoch": 0.4730587808417997,
"grad_norm": 2.125,
"learning_rate": 0.0004983494610790241,
"loss": 5.6938,
"mean_token_accuracy": 0.14457672983407974,
"num_tokens": 10841404.0,
"step": 5215
},
{
"entropy": 5.787980270385742,
"epoch": 0.4735123367198839,
"grad_norm": 2.515625,
"learning_rate": 0.0004983455467667996,
"loss": 5.7393,
"mean_token_accuracy": 0.1467074014246464,
"num_tokens": 10851577.0,
"step": 5220
},
{
"entropy": 5.788097476959228,
"epoch": 0.47396589259796806,
"grad_norm": 2.234375,
"learning_rate": 0.0004983416278357322,
"loss": 5.6485,
"mean_token_accuracy": 0.1567993551492691,
"num_tokens": 10861403.0,
"step": 5225
},
{
"entropy": 5.844185018539429,
"epoch": 0.4744194484760523,
"grad_norm": 2.28125,
"learning_rate": 0.0004983377042859032,
"loss": 5.702,
"mean_token_accuracy": 0.14790162742137908,
"num_tokens": 10871933.0,
"step": 5230
},
{
"entropy": 5.761803388595581,
"epoch": 0.47487300435413643,
"grad_norm": 2.984375,
"learning_rate": 0.0004983337761173936,
"loss": 5.773,
"mean_token_accuracy": 0.14696283340454103,
"num_tokens": 10881800.0,
"step": 5235
},
{
"entropy": 5.915017938613891,
"epoch": 0.4753265602322206,
"grad_norm": 2.203125,
"learning_rate": 0.0004983298433302843,
"loss": 5.74,
"mean_token_accuracy": 0.14846391826868058,
"num_tokens": 10891830.0,
"step": 5240
},
{
"entropy": 5.812942409515381,
"epoch": 0.4757801161103048,
"grad_norm": 2.0,
"learning_rate": 0.0004983259059246572,
"loss": 5.655,
"mean_token_accuracy": 0.15466391891241074,
"num_tokens": 10901557.0,
"step": 5245
},
{
"entropy": 5.7293657779693605,
"epoch": 0.47623367198838895,
"grad_norm": 4.0625,
"learning_rate": 0.0004983219639005934,
"loss": 5.764,
"mean_token_accuracy": 0.14264445900917053,
"num_tokens": 10912155.0,
"step": 5250
},
{
"entropy": 5.743430709838867,
"epoch": 0.47668722786647316,
"grad_norm": 2.1875,
"learning_rate": 0.0004983180172581745,
"loss": 5.6787,
"mean_token_accuracy": 0.15394465178251265,
"num_tokens": 10922291.0,
"step": 5255
},
{
"entropy": 5.898777866363526,
"epoch": 0.4771407837445573,
"grad_norm": 1.78125,
"learning_rate": 0.0004983140659974819,
"loss": 5.7493,
"mean_token_accuracy": 0.1463671863079071,
"num_tokens": 10933176.0,
"step": 5260
},
{
"entropy": 5.883290433883667,
"epoch": 0.47759433962264153,
"grad_norm": 2.96875,
"learning_rate": 0.0004983101101185977,
"loss": 5.8327,
"mean_token_accuracy": 0.145734940469265,
"num_tokens": 10944049.0,
"step": 5265
},
{
"entropy": 5.773307657241821,
"epoch": 0.4780478955007257,
"grad_norm": 2.703125,
"learning_rate": 0.0004983061496216035,
"loss": 5.6794,
"mean_token_accuracy": 0.1519426852464676,
"num_tokens": 10954347.0,
"step": 5270
},
{
"entropy": 5.770019006729126,
"epoch": 0.4785014513788099,
"grad_norm": 2.375,
"learning_rate": 0.0004983021845065812,
"loss": 5.7304,
"mean_token_accuracy": 0.1466204196214676,
"num_tokens": 10965123.0,
"step": 5275
},
{
"entropy": 5.798799276351929,
"epoch": 0.47895500725689405,
"grad_norm": 3.578125,
"learning_rate": 0.0004982982147736128,
"loss": 5.7335,
"mean_token_accuracy": 0.1530810609459877,
"num_tokens": 10976395.0,
"step": 5280
},
{
"entropy": 5.84113039970398,
"epoch": 0.4794085631349782,
"grad_norm": 2.515625,
"learning_rate": 0.0004982942404227805,
"loss": 5.6978,
"mean_token_accuracy": 0.1490973189473152,
"num_tokens": 10986797.0,
"step": 5285
},
{
"entropy": 5.87009449005127,
"epoch": 0.4798621190130624,
"grad_norm": 1.984375,
"learning_rate": 0.0004982902614541662,
"loss": 5.7569,
"mean_token_accuracy": 0.14732319861650467,
"num_tokens": 10997229.0,
"step": 5290
},
{
"entropy": 5.840237808227539,
"epoch": 0.48031567489114657,
"grad_norm": 2.328125,
"learning_rate": 0.0004982862778678526,
"loss": 5.6982,
"mean_token_accuracy": 0.15742301493883132,
"num_tokens": 11008180.0,
"step": 5295
},
{
"entropy": 5.926225042343139,
"epoch": 0.4807692307692308,
"grad_norm": 1.6484375,
"learning_rate": 0.0004982822896639218,
"loss": 5.7236,
"mean_token_accuracy": 0.1533257931470871,
"num_tokens": 11019248.0,
"step": 5300
},
{
"entropy": 5.791766881942749,
"epoch": 0.48122278664731494,
"grad_norm": 2.15625,
"learning_rate": 0.0004982782968424563,
"loss": 5.649,
"mean_token_accuracy": 0.15525579005479812,
"num_tokens": 11029760.0,
"step": 5305
},
{
"entropy": 5.7674986839294435,
"epoch": 0.48167634252539915,
"grad_norm": 3.171875,
"learning_rate": 0.0004982742994035388,
"loss": 5.7206,
"mean_token_accuracy": 0.14568620324134826,
"num_tokens": 11039211.0,
"step": 5310
},
{
"entropy": 5.839351844787598,
"epoch": 0.4821298984034833,
"grad_norm": 2.0,
"learning_rate": 0.0004982702973472518,
"loss": 5.6961,
"mean_token_accuracy": 0.1473838321864605,
"num_tokens": 11049892.0,
"step": 5315
},
{
"entropy": 5.943693685531616,
"epoch": 0.4825834542815675,
"grad_norm": 2.15625,
"learning_rate": 0.0004982662906736783,
"loss": 5.7621,
"mean_token_accuracy": 0.14789720624685287,
"num_tokens": 11060823.0,
"step": 5320
},
{
"entropy": 5.757358884811401,
"epoch": 0.48303701015965167,
"grad_norm": 2.1875,
"learning_rate": 0.0004982622793829009,
"loss": 5.7463,
"mean_token_accuracy": 0.15420066863298415,
"num_tokens": 11072353.0,
"step": 5325
},
{
"entropy": 5.798610353469849,
"epoch": 0.4834905660377358,
"grad_norm": 1.796875,
"learning_rate": 0.0004982582634750027,
"loss": 5.7571,
"mean_token_accuracy": 0.14944387078285218,
"num_tokens": 11082655.0,
"step": 5330
},
{
"entropy": 5.828657102584839,
"epoch": 0.48394412191582004,
"grad_norm": 1.7578125,
"learning_rate": 0.0004982542429500667,
"loss": 5.7616,
"mean_token_accuracy": 0.15139087289571762,
"num_tokens": 11093315.0,
"step": 5335
},
{
"entropy": 5.714878511428833,
"epoch": 0.4843976777939042,
"grad_norm": 1.8046875,
"learning_rate": 0.0004982502178081761,
"loss": 5.6677,
"mean_token_accuracy": 0.15089037865400315,
"num_tokens": 11103074.0,
"step": 5340
},
{
"entropy": 5.84651894569397,
"epoch": 0.4848512336719884,
"grad_norm": 2.421875,
"learning_rate": 0.0004982461880494141,
"loss": 5.7078,
"mean_token_accuracy": 0.15072353780269623,
"num_tokens": 11113636.0,
"step": 5345
},
{
"entropy": 5.837501668930054,
"epoch": 0.48530478955007256,
"grad_norm": 1.9765625,
"learning_rate": 0.000498242153673864,
"loss": 5.6617,
"mean_token_accuracy": 0.15472064316272735,
"num_tokens": 11123509.0,
"step": 5350
},
{
"entropy": 5.811651086807251,
"epoch": 0.48575834542815677,
"grad_norm": 2.71875,
"learning_rate": 0.0004982381146816094,
"loss": 5.732,
"mean_token_accuracy": 0.1466941475868225,
"num_tokens": 11133405.0,
"step": 5355
},
{
"entropy": 5.874093151092529,
"epoch": 0.4862119013062409,
"grad_norm": 2.140625,
"learning_rate": 0.0004982340710727336,
"loss": 5.784,
"mean_token_accuracy": 0.1411792032420635,
"num_tokens": 11144725.0,
"step": 5360
},
{
"entropy": 5.835286045074463,
"epoch": 0.48666545718432513,
"grad_norm": 1.9453125,
"learning_rate": 0.0004982300228473203,
"loss": 5.6809,
"mean_token_accuracy": 0.14511124938726425,
"num_tokens": 11155446.0,
"step": 5365
},
{
"entropy": 5.843464946746826,
"epoch": 0.4871190130624093,
"grad_norm": 1.8046875,
"learning_rate": 0.0004982259700054533,
"loss": 5.7373,
"mean_token_accuracy": 0.15083874613046647,
"num_tokens": 11165380.0,
"step": 5370
},
{
"entropy": 5.833011054992676,
"epoch": 0.48757256894049344,
"grad_norm": 1.796875,
"learning_rate": 0.0004982219125472163,
"loss": 5.7805,
"mean_token_accuracy": 0.15250826701521875,
"num_tokens": 11175641.0,
"step": 5375
},
{
"entropy": 5.952740001678467,
"epoch": 0.48802612481857766,
"grad_norm": 1.953125,
"learning_rate": 0.0004982178504726933,
"loss": 5.9333,
"mean_token_accuracy": 0.1492827832698822,
"num_tokens": 11186257.0,
"step": 5380
},
{
"entropy": 5.917461729049682,
"epoch": 0.4884796806966618,
"grad_norm": 2.125,
"learning_rate": 0.0004982137837819682,
"loss": 5.7581,
"mean_token_accuracy": 0.14257122650742532,
"num_tokens": 11197284.0,
"step": 5385
},
{
"entropy": 5.890596342086792,
"epoch": 0.488933236574746,
"grad_norm": 1.7890625,
"learning_rate": 0.0004982097124751252,
"loss": 5.7259,
"mean_token_accuracy": 0.15704453438520433,
"num_tokens": 11208413.0,
"step": 5390
},
{
"entropy": 5.756617689132691,
"epoch": 0.4893867924528302,
"grad_norm": 2.0625,
"learning_rate": 0.0004982056365522486,
"loss": 5.7385,
"mean_token_accuracy": 0.15123408585786818,
"num_tokens": 11219016.0,
"step": 5395
},
{
"entropy": 5.743880319595337,
"epoch": 0.4898403483309144,
"grad_norm": 2.046875,
"learning_rate": 0.0004982015560134225,
"loss": 5.7147,
"mean_token_accuracy": 0.1525794416666031,
"num_tokens": 11228934.0,
"step": 5400
},
{
"entropy": 5.801280117034912,
"epoch": 0.49029390420899854,
"grad_norm": 2.0625,
"learning_rate": 0.0004981974708587312,
"loss": 5.6416,
"mean_token_accuracy": 0.15183479487895965,
"num_tokens": 11239596.0,
"step": 5405
},
{
"entropy": 5.771134996414185,
"epoch": 0.49074746008708275,
"grad_norm": 2.078125,
"learning_rate": 0.0004981933810882594,
"loss": 5.6322,
"mean_token_accuracy": 0.14867930263280868,
"num_tokens": 11249560.0,
"step": 5410
},
{
"entropy": 5.860190916061401,
"epoch": 0.4912010159651669,
"grad_norm": 2.0625,
"learning_rate": 0.0004981892867020917,
"loss": 5.8181,
"mean_token_accuracy": 0.14608463272452354,
"num_tokens": 11259787.0,
"step": 5415
},
{
"entropy": 5.8484776496887205,
"epoch": 0.49165457184325106,
"grad_norm": 2.078125,
"learning_rate": 0.0004981851877003125,
"loss": 5.7599,
"mean_token_accuracy": 0.14876429736614227,
"num_tokens": 11269717.0,
"step": 5420
},
{
"entropy": 5.834319734573365,
"epoch": 0.4921081277213353,
"grad_norm": 2.4375,
"learning_rate": 0.0004981810840830068,
"loss": 5.7344,
"mean_token_accuracy": 0.15793785825371742,
"num_tokens": 11279863.0,
"step": 5425
},
{
"entropy": 5.850916337966919,
"epoch": 0.49256168359941943,
"grad_norm": 2.90625,
"learning_rate": 0.0004981769758502596,
"loss": 5.8704,
"mean_token_accuracy": 0.1406077429652214,
"num_tokens": 11291161.0,
"step": 5430
},
{
"entropy": 5.7810022830963135,
"epoch": 0.49301523947750364,
"grad_norm": 2.34375,
"learning_rate": 0.0004981728630021555,
"loss": 5.6669,
"mean_token_accuracy": 0.15320780575275422,
"num_tokens": 11301151.0,
"step": 5435
},
{
"entropy": 5.785669231414795,
"epoch": 0.4934687953555878,
"grad_norm": 1.6484375,
"learning_rate": 0.0004981687455387797,
"loss": 5.6748,
"mean_token_accuracy": 0.15113574340939523,
"num_tokens": 11311466.0,
"step": 5440
},
{
"entropy": 5.812684631347656,
"epoch": 0.493922351233672,
"grad_norm": 3.765625,
"learning_rate": 0.0004981646234602173,
"loss": 5.6467,
"mean_token_accuracy": 0.15649000257253648,
"num_tokens": 11321522.0,
"step": 5445
},
{
"entropy": 5.847482442855835,
"epoch": 0.49437590711175616,
"grad_norm": 2.703125,
"learning_rate": 0.0004981604967665539,
"loss": 5.761,
"mean_token_accuracy": 0.14855778515338897,
"num_tokens": 11332317.0,
"step": 5450
},
{
"entropy": 5.819078731536865,
"epoch": 0.4948294629898404,
"grad_norm": 2.203125,
"learning_rate": 0.0004981563654578743,
"loss": 5.6897,
"mean_token_accuracy": 0.14819339513778687,
"num_tokens": 11343024.0,
"step": 5455
},
{
"entropy": 5.848114490509033,
"epoch": 0.49528301886792453,
"grad_norm": 2.546875,
"learning_rate": 0.0004981522295342643,
"loss": 5.6589,
"mean_token_accuracy": 0.15384979099035262,
"num_tokens": 11353114.0,
"step": 5460
},
{
"entropy": 5.850389432907105,
"epoch": 0.4957365747460087,
"grad_norm": 2.109375,
"learning_rate": 0.0004981480889958092,
"loss": 5.7088,
"mean_token_accuracy": 0.15356869697570802,
"num_tokens": 11363446.0,
"step": 5465
},
{
"entropy": 5.791082286834717,
"epoch": 0.4961901306240929,
"grad_norm": 2.5625,
"learning_rate": 0.0004981439438425949,
"loss": 5.7092,
"mean_token_accuracy": 0.15395713597536087,
"num_tokens": 11373413.0,
"step": 5470
},
{
"entropy": 5.702733945846558,
"epoch": 0.49664368650217705,
"grad_norm": 3.265625,
"learning_rate": 0.000498139794074707,
"loss": 5.5953,
"mean_token_accuracy": 0.15706127732992173,
"num_tokens": 11383142.0,
"step": 5475
},
{
"entropy": 5.815199375152588,
"epoch": 0.49709724238026126,
"grad_norm": 2.328125,
"learning_rate": 0.0004981356396922313,
"loss": 5.7352,
"mean_token_accuracy": 0.14936144053936004,
"num_tokens": 11394635.0,
"step": 5480
},
{
"entropy": 5.790682744979859,
"epoch": 0.4975507982583454,
"grad_norm": 3.125,
"learning_rate": 0.0004981314806952535,
"loss": 5.641,
"mean_token_accuracy": 0.15651369094848633,
"num_tokens": 11404442.0,
"step": 5485
},
{
"entropy": 5.753213262557983,
"epoch": 0.49800435413642963,
"grad_norm": 2.703125,
"learning_rate": 0.00049812731708386,
"loss": 5.6012,
"mean_token_accuracy": 0.151226407289505,
"num_tokens": 11414217.0,
"step": 5490
},
{
"entropy": 5.8672788619995115,
"epoch": 0.4984579100145138,
"grad_norm": 2.046875,
"learning_rate": 0.0004981231488581367,
"loss": 5.7436,
"mean_token_accuracy": 0.1472836948931217,
"num_tokens": 11423834.0,
"step": 5495
},
{
"entropy": 5.754918384552002,
"epoch": 0.498911465892598,
"grad_norm": 4.875,
"learning_rate": 0.0004981189760181699,
"loss": 5.6162,
"mean_token_accuracy": 0.1506452038884163,
"num_tokens": 11433613.0,
"step": 5500
},
{
"entropy": 5.744574022293091,
"epoch": 0.49936502177068215,
"grad_norm": 2.484375,
"learning_rate": 0.0004981147985640456,
"loss": 5.6953,
"mean_token_accuracy": 0.15378040224313735,
"num_tokens": 11443915.0,
"step": 5505
},
{
"entropy": 5.902297592163086,
"epoch": 0.4998185776487663,
"grad_norm": 2.40625,
"learning_rate": 0.0004981106164958506,
"loss": 5.7512,
"mean_token_accuracy": 0.14607876390218735,
"num_tokens": 11454588.0,
"step": 5510
},
{
"entropy": 5.8737144947052,
"epoch": 0.5002721335268505,
"grad_norm": 2.9375,
"learning_rate": 0.0004981064298136712,
"loss": 5.7896,
"mean_token_accuracy": 0.14923810511827468,
"num_tokens": 11465533.0,
"step": 5515
},
{
"entropy": 5.816037034988403,
"epoch": 0.5007256894049347,
"grad_norm": 2.515625,
"learning_rate": 0.0004981022385175939,
"loss": 5.7349,
"mean_token_accuracy": 0.14456416741013528,
"num_tokens": 11475776.0,
"step": 5520
},
{
"entropy": 5.810181140899658,
"epoch": 0.5011792452830188,
"grad_norm": 1.8828125,
"learning_rate": 0.0004980980426077054,
"loss": 5.7047,
"mean_token_accuracy": 0.15837073922157288,
"num_tokens": 11486200.0,
"step": 5525
},
{
"entropy": 5.8160165309906,
"epoch": 0.5016328011611031,
"grad_norm": 3.59375,
"learning_rate": 0.0004980938420840926,
"loss": 5.739,
"mean_token_accuracy": 0.14561933577060698,
"num_tokens": 11496768.0,
"step": 5530
},
{
"entropy": 5.837372159957885,
"epoch": 0.5020863570391872,
"grad_norm": 2.09375,
"learning_rate": 0.0004980896369468423,
"loss": 5.7015,
"mean_token_accuracy": 0.15523829013109208,
"num_tokens": 11507096.0,
"step": 5535
},
{
"entropy": 5.847644376754761,
"epoch": 0.5025399129172714,
"grad_norm": 4.46875,
"learning_rate": 0.0004980854271960415,
"loss": 5.7618,
"mean_token_accuracy": 0.15393387004733086,
"num_tokens": 11517301.0,
"step": 5540
},
{
"entropy": 5.862361669540405,
"epoch": 0.5029934687953556,
"grad_norm": 2.28125,
"learning_rate": 0.0004980812128317772,
"loss": 5.8145,
"mean_token_accuracy": 0.14930694699287414,
"num_tokens": 11528616.0,
"step": 5545
},
{
"entropy": 5.91984920501709,
"epoch": 0.5034470246734397,
"grad_norm": 3.046875,
"learning_rate": 0.0004980769938541364,
"loss": 5.7421,
"mean_token_accuracy": 0.15533946603536605,
"num_tokens": 11539344.0,
"step": 5550
},
{
"entropy": 5.893903303146362,
"epoch": 0.503900580551524,
"grad_norm": 1.953125,
"learning_rate": 0.0004980727702632066,
"loss": 5.8022,
"mean_token_accuracy": 0.14859232157468796,
"num_tokens": 11549511.0,
"step": 5555
},
{
"entropy": 5.845568323135376,
"epoch": 0.5043541364296081,
"grad_norm": 3.671875,
"learning_rate": 0.0004980685420590752,
"loss": 5.7162,
"mean_token_accuracy": 0.14990185871720313,
"num_tokens": 11559425.0,
"step": 5560
},
{
"entropy": 5.871610403060913,
"epoch": 0.5048076923076923,
"grad_norm": 2.171875,
"learning_rate": 0.0004980643092418294,
"loss": 5.8337,
"mean_token_accuracy": 0.1418799228966236,
"num_tokens": 11570406.0,
"step": 5565
},
{
"entropy": 5.810862731933594,
"epoch": 0.5052612481857764,
"grad_norm": 1.7421875,
"learning_rate": 0.0004980600718115568,
"loss": 5.7154,
"mean_token_accuracy": 0.15226266756653786,
"num_tokens": 11580509.0,
"step": 5570
},
{
"entropy": 5.8717328071594235,
"epoch": 0.5057148040638607,
"grad_norm": 2.03125,
"learning_rate": 0.0004980558297683452,
"loss": 5.7229,
"mean_token_accuracy": 0.15168080031871795,
"num_tokens": 11590860.0,
"step": 5575
},
{
"entropy": 5.922792911529541,
"epoch": 0.5061683599419449,
"grad_norm": 3.828125,
"learning_rate": 0.0004980515831122822,
"loss": 5.7336,
"mean_token_accuracy": 0.15418553948402405,
"num_tokens": 11601954.0,
"step": 5580
},
{
"entropy": 5.866706800460816,
"epoch": 0.506621915820029,
"grad_norm": 3.34375,
"learning_rate": 0.0004980473318434555,
"loss": 5.7499,
"mean_token_accuracy": 0.1491977170109749,
"num_tokens": 11612581.0,
"step": 5585
},
{
"entropy": 5.79789137840271,
"epoch": 0.5070754716981132,
"grad_norm": 2.421875,
"learning_rate": 0.0004980430759619532,
"loss": 5.6749,
"mean_token_accuracy": 0.15472817420959473,
"num_tokens": 11622599.0,
"step": 5590
},
{
"entropy": 5.773746681213379,
"epoch": 0.5075290275761973,
"grad_norm": 5.09375,
"learning_rate": 0.0004980388154678633,
"loss": 5.682,
"mean_token_accuracy": 0.15070171654224396,
"num_tokens": 11633611.0,
"step": 5595
},
{
"entropy": 5.802550792694092,
"epoch": 0.5079825834542816,
"grad_norm": 2.0625,
"learning_rate": 0.0004980345503612738,
"loss": 5.6761,
"mean_token_accuracy": 0.14901446551084518,
"num_tokens": 11643159.0,
"step": 5600
},
{
"entropy": 5.781796932220459,
"epoch": 0.5084361393323658,
"grad_norm": 2.484375,
"learning_rate": 0.0004980302806422729,
"loss": 5.6478,
"mean_token_accuracy": 0.15606949031352996,
"num_tokens": 11653174.0,
"step": 5605
},
{
"entropy": 5.798026847839355,
"epoch": 0.5088896952104499,
"grad_norm": 3.375,
"learning_rate": 0.0004980260063109491,
"loss": 5.694,
"mean_token_accuracy": 0.15325014442205429,
"num_tokens": 11662621.0,
"step": 5610
},
{
"entropy": 5.7645632266998295,
"epoch": 0.5093432510885341,
"grad_norm": 2.3125,
"learning_rate": 0.0004980217273673906,
"loss": 5.6488,
"mean_token_accuracy": 0.1541965365409851,
"num_tokens": 11673538.0,
"step": 5615
},
{
"entropy": 5.807270860671997,
"epoch": 0.5097968069666183,
"grad_norm": 2.5,
"learning_rate": 0.0004980174438116859,
"loss": 5.6645,
"mean_token_accuracy": 0.15759781897068023,
"num_tokens": 11683203.0,
"step": 5620
},
{
"entropy": 5.808385896682739,
"epoch": 0.5102503628447025,
"grad_norm": 5.0,
"learning_rate": 0.0004980131556439236,
"loss": 5.6727,
"mean_token_accuracy": 0.15713814795017242,
"num_tokens": 11693313.0,
"step": 5625
},
{
"entropy": 5.834480953216553,
"epoch": 0.5107039187227866,
"grad_norm": 2.328125,
"learning_rate": 0.0004980088628641925,
"loss": 5.7533,
"mean_token_accuracy": 0.15072424262762069,
"num_tokens": 11704194.0,
"step": 5630
},
{
"entropy": 5.80663833618164,
"epoch": 0.5111574746008708,
"grad_norm": 3.515625,
"learning_rate": 0.0004980045654725812,
"loss": 5.6112,
"mean_token_accuracy": 0.15680344700813292,
"num_tokens": 11715109.0,
"step": 5635
},
{
"entropy": 5.749876594543457,
"epoch": 0.511611030478955,
"grad_norm": 3.84375,
"learning_rate": 0.0004980002634691787,
"loss": 5.6527,
"mean_token_accuracy": 0.15471058934926987,
"num_tokens": 11724961.0,
"step": 5640
},
{
"entropy": 5.743565797805786,
"epoch": 0.5120645863570392,
"grad_norm": 2.515625,
"learning_rate": 0.000497995956854074,
"loss": 5.6811,
"mean_token_accuracy": 0.15321707502007484,
"num_tokens": 11735334.0,
"step": 5645
},
{
"entropy": 5.80997257232666,
"epoch": 0.5125181422351234,
"grad_norm": 2.265625,
"learning_rate": 0.0004979916456273561,
"loss": 5.6801,
"mean_token_accuracy": 0.1457176223397255,
"num_tokens": 11746858.0,
"step": 5650
},
{
"entropy": 5.898235130310058,
"epoch": 0.5129716981132075,
"grad_norm": 3.765625,
"learning_rate": 0.0004979873297891141,
"loss": 5.7874,
"mean_token_accuracy": 0.14425734728574752,
"num_tokens": 11757819.0,
"step": 5655
},
{
"entropy": 5.879952812194825,
"epoch": 0.5134252539912917,
"grad_norm": 2.21875,
"learning_rate": 0.0004979830093394372,
"loss": 5.7022,
"mean_token_accuracy": 0.1498760312795639,
"num_tokens": 11767470.0,
"step": 5660
},
{
"entropy": 5.842562484741211,
"epoch": 0.513878809869376,
"grad_norm": 2.6875,
"learning_rate": 0.0004979786842784149,
"loss": 5.6223,
"mean_token_accuracy": 0.15927857905626297,
"num_tokens": 11777060.0,
"step": 5665
},
{
"entropy": 5.728598117828369,
"epoch": 0.5143323657474601,
"grad_norm": 2.15625,
"learning_rate": 0.0004979743546061366,
"loss": 5.6863,
"mean_token_accuracy": 0.15647987723350526,
"num_tokens": 11787176.0,
"step": 5670
},
{
"entropy": 5.639022636413574,
"epoch": 0.5147859216255443,
"grad_norm": 2.21875,
"learning_rate": 0.0004979700203226918,
"loss": 5.5698,
"mean_token_accuracy": 0.15824757367372513,
"num_tokens": 11796419.0,
"step": 5675
},
{
"entropy": 5.869472932815552,
"epoch": 0.5152394775036284,
"grad_norm": 1.8359375,
"learning_rate": 0.0004979656814281702,
"loss": 5.8015,
"mean_token_accuracy": 0.14808933883905412,
"num_tokens": 11807279.0,
"step": 5680
},
{
"entropy": 5.864350700378418,
"epoch": 0.5156930333817126,
"grad_norm": 4.15625,
"learning_rate": 0.0004979613379226615,
"loss": 5.6757,
"mean_token_accuracy": 0.14973624050617218,
"num_tokens": 11817669.0,
"step": 5685
},
{
"entropy": 5.7543477535247805,
"epoch": 0.5161465892597968,
"grad_norm": 2.03125,
"learning_rate": 0.0004979569898062554,
"loss": 5.6793,
"mean_token_accuracy": 0.15661610066890716,
"num_tokens": 11827247.0,
"step": 5690
},
{
"entropy": 5.805063152313233,
"epoch": 0.516600145137881,
"grad_norm": 2.0,
"learning_rate": 0.000497952637079042,
"loss": 5.7034,
"mean_token_accuracy": 0.147939595580101,
"num_tokens": 11838201.0,
"step": 5695
},
{
"entropy": 5.80039496421814,
"epoch": 0.5170537010159652,
"grad_norm": 2.515625,
"learning_rate": 0.0004979482797411114,
"loss": 5.6264,
"mean_token_accuracy": 0.15654842257499696,
"num_tokens": 11847561.0,
"step": 5700
},
{
"entropy": 5.81939845085144,
"epoch": 0.5175072568940493,
"grad_norm": 2.25,
"learning_rate": 0.0004979439177925534,
"loss": 5.7384,
"mean_token_accuracy": 0.1583888441324234,
"num_tokens": 11858272.0,
"step": 5705
},
{
"entropy": 5.923577070236206,
"epoch": 0.5179608127721336,
"grad_norm": 1.890625,
"learning_rate": 0.0004979395512334584,
"loss": 5.7881,
"mean_token_accuracy": 0.148820598423481,
"num_tokens": 11869660.0,
"step": 5710
},
{
"entropy": 5.893476390838623,
"epoch": 0.5184143686502177,
"grad_norm": 2.171875,
"learning_rate": 0.0004979351800639166,
"loss": 5.8122,
"mean_token_accuracy": 0.14301129132509233,
"num_tokens": 11880239.0,
"step": 5715
},
{
"entropy": 5.912051820755005,
"epoch": 0.5188679245283019,
"grad_norm": 2.125,
"learning_rate": 0.0004979308042840186,
"loss": 5.7862,
"mean_token_accuracy": 0.14405866116285324,
"num_tokens": 11890630.0,
"step": 5720
},
{
"entropy": 5.940375518798828,
"epoch": 0.519321480406386,
"grad_norm": 2.0625,
"learning_rate": 0.0004979264238938547,
"loss": 5.8105,
"mean_token_accuracy": 0.14483141005039216,
"num_tokens": 11900948.0,
"step": 5725
},
{
"entropy": 5.788791179656982,
"epoch": 0.5197750362844702,
"grad_norm": 2.390625,
"learning_rate": 0.0004979220388935155,
"loss": 5.6941,
"mean_token_accuracy": 0.15050781145691872,
"num_tokens": 11912528.0,
"step": 5730
},
{
"entropy": 5.782585954666137,
"epoch": 0.5202285921625545,
"grad_norm": 2.46875,
"learning_rate": 0.0004979176492830918,
"loss": 5.6521,
"mean_token_accuracy": 0.15226173400878906,
"num_tokens": 11922211.0,
"step": 5735
},
{
"entropy": 5.8104184627532955,
"epoch": 0.5206821480406386,
"grad_norm": 2.421875,
"learning_rate": 0.0004979132550626742,
"loss": 5.6858,
"mean_token_accuracy": 0.14871056675910949,
"num_tokens": 11932520.0,
"step": 5740
},
{
"entropy": 5.8572922706604,
"epoch": 0.5211357039187228,
"grad_norm": 2.25,
"learning_rate": 0.0004979088562323538,
"loss": 5.6209,
"mean_token_accuracy": 0.15842307060956956,
"num_tokens": 11943312.0,
"step": 5745
},
{
"entropy": 5.783070468902588,
"epoch": 0.5215892597968069,
"grad_norm": 2.453125,
"learning_rate": 0.0004979044527922214,
"loss": 5.7697,
"mean_token_accuracy": 0.1494937114417553,
"num_tokens": 11954455.0,
"step": 5750
},
{
"entropy": 5.808009147644043,
"epoch": 0.5220428156748912,
"grad_norm": 2.8125,
"learning_rate": 0.0004979000447423681,
"loss": 5.6679,
"mean_token_accuracy": 0.1576451078057289,
"num_tokens": 11964650.0,
"step": 5755
},
{
"entropy": 5.751487350463867,
"epoch": 0.5224963715529753,
"grad_norm": 2.359375,
"learning_rate": 0.0004978956320828851,
"loss": 5.7023,
"mean_token_accuracy": 0.15352913439273835,
"num_tokens": 11975134.0,
"step": 5760
},
{
"entropy": 5.815647315979004,
"epoch": 0.5229499274310595,
"grad_norm": 2.734375,
"learning_rate": 0.0004978912148138637,
"loss": 5.7442,
"mean_token_accuracy": 0.14895366728305817,
"num_tokens": 11986486.0,
"step": 5765
},
{
"entropy": 5.89086766242981,
"epoch": 0.5234034833091437,
"grad_norm": 2.484375,
"learning_rate": 0.000497886792935395,
"loss": 5.7652,
"mean_token_accuracy": 0.14870162308216095,
"num_tokens": 11998045.0,
"step": 5770
},
{
"entropy": 5.9239579200744625,
"epoch": 0.5238570391872278,
"grad_norm": 2.03125,
"learning_rate": 0.0004978823664475707,
"loss": 5.7574,
"mean_token_accuracy": 0.15128326863050462,
"num_tokens": 12008130.0,
"step": 5775
},
{
"entropy": 5.880002403259278,
"epoch": 0.5243105950653121,
"grad_norm": 2.578125,
"learning_rate": 0.0004978779353504824,
"loss": 5.8101,
"mean_token_accuracy": 0.14375324547290802,
"num_tokens": 12018753.0,
"step": 5780
},
{
"entropy": 5.836540651321411,
"epoch": 0.5247641509433962,
"grad_norm": 1.90625,
"learning_rate": 0.0004978734996442214,
"loss": 5.6768,
"mean_token_accuracy": 0.1561191499233246,
"num_tokens": 12028202.0,
"step": 5785
},
{
"entropy": 5.83511757850647,
"epoch": 0.5252177068214804,
"grad_norm": 3.0,
"learning_rate": 0.0004978690593288798,
"loss": 5.7631,
"mean_token_accuracy": 0.15131065398454666,
"num_tokens": 12039652.0,
"step": 5790
},
{
"entropy": 5.919553232192993,
"epoch": 0.5256712626995645,
"grad_norm": 2.984375,
"learning_rate": 0.0004978646144045491,
"loss": 5.7,
"mean_token_accuracy": 0.14130512177944182,
"num_tokens": 12051269.0,
"step": 5795
},
{
"entropy": 5.771428203582763,
"epoch": 0.5261248185776488,
"grad_norm": 2.078125,
"learning_rate": 0.0004978601648713215,
"loss": 5.5623,
"mean_token_accuracy": 0.15090945810079576,
"num_tokens": 12060647.0,
"step": 5800
},
{
"entropy": 5.792234802246094,
"epoch": 0.526578374455733,
"grad_norm": 2.03125,
"learning_rate": 0.0004978557107292889,
"loss": 5.7274,
"mean_token_accuracy": 0.1574611634016037,
"num_tokens": 12071669.0,
"step": 5805
},
{
"entropy": 5.736810398101807,
"epoch": 0.5270319303338171,
"grad_norm": 2.25,
"learning_rate": 0.0004978512519785433,
"loss": 5.6046,
"mean_token_accuracy": 0.15474170595407485,
"num_tokens": 12081971.0,
"step": 5810
},
{
"entropy": 5.7692629337310795,
"epoch": 0.5274854862119013,
"grad_norm": 2.203125,
"learning_rate": 0.000497846788619177,
"loss": 5.7381,
"mean_token_accuracy": 0.15169834345579147,
"num_tokens": 12092465.0,
"step": 5815
},
{
"entropy": 5.852247858047486,
"epoch": 0.5279390420899854,
"grad_norm": 2.40625,
"learning_rate": 0.0004978423206512825,
"loss": 5.6706,
"mean_token_accuracy": 0.15314571261405946,
"num_tokens": 12101752.0,
"step": 5820
},
{
"entropy": 5.919044303894043,
"epoch": 0.5283925979680697,
"grad_norm": 2.296875,
"learning_rate": 0.000497837848074952,
"loss": 5.7533,
"mean_token_accuracy": 0.1472005233168602,
"num_tokens": 12111805.0,
"step": 5825
},
{
"entropy": 5.848569869995117,
"epoch": 0.5288461538461539,
"grad_norm": 2.171875,
"learning_rate": 0.000497833370890278,
"loss": 5.7834,
"mean_token_accuracy": 0.14824461191892624,
"num_tokens": 12122511.0,
"step": 5830
},
{
"entropy": 5.827112245559692,
"epoch": 0.529299709724238,
"grad_norm": 1.9765625,
"learning_rate": 0.0004978288890973529,
"loss": 5.6648,
"mean_token_accuracy": 0.15470681414008142,
"num_tokens": 12132510.0,
"step": 5835
},
{
"entropy": 5.7759528160095215,
"epoch": 0.5297532656023222,
"grad_norm": 2.34375,
"learning_rate": 0.0004978244026962698,
"loss": 5.6782,
"mean_token_accuracy": 0.15179212763905525,
"num_tokens": 12142419.0,
"step": 5840
},
{
"entropy": 5.80072717666626,
"epoch": 0.5302068214804064,
"grad_norm": 2.109375,
"learning_rate": 0.0004978199116871212,
"loss": 5.7047,
"mean_token_accuracy": 0.14955106675624846,
"num_tokens": 12152759.0,
"step": 5845
},
{
"entropy": 5.8397301197052,
"epoch": 0.5306603773584906,
"grad_norm": 1.8125,
"learning_rate": 0.00049781541607,
"loss": 5.6693,
"mean_token_accuracy": 0.15128355473279953,
"num_tokens": 12162758.0,
"step": 5850
},
{
"entropy": 5.860444831848144,
"epoch": 0.5311139332365747,
"grad_norm": 2.328125,
"learning_rate": 0.0004978109158449993,
"loss": 5.7164,
"mean_token_accuracy": 0.1495297521352768,
"num_tokens": 12172664.0,
"step": 5855
},
{
"entropy": 5.820270347595215,
"epoch": 0.5315674891146589,
"grad_norm": 1.9765625,
"learning_rate": 0.0004978064110122121,
"loss": 5.6847,
"mean_token_accuracy": 0.15420603305101394,
"num_tokens": 12182028.0,
"step": 5860
},
{
"entropy": 5.807129764556885,
"epoch": 0.532021044992743,
"grad_norm": 3.328125,
"learning_rate": 0.0004978019015717314,
"loss": 5.7133,
"mean_token_accuracy": 0.15545213222503662,
"num_tokens": 12192168.0,
"step": 5865
},
{
"entropy": 5.763472938537598,
"epoch": 0.5324746008708273,
"grad_norm": 1.9921875,
"learning_rate": 0.0004977973875236508,
"loss": 5.6219,
"mean_token_accuracy": 0.15059156119823455,
"num_tokens": 12202889.0,
"step": 5870
},
{
"entropy": 5.8198224067687985,
"epoch": 0.5329281567489115,
"grad_norm": 1.890625,
"learning_rate": 0.0004977928688680634,
"loss": 5.7018,
"mean_token_accuracy": 0.15930801928043364,
"num_tokens": 12212514.0,
"step": 5875
},
{
"entropy": 5.7587038516998295,
"epoch": 0.5333817126269956,
"grad_norm": 2.46875,
"learning_rate": 0.0004977883456050626,
"loss": 5.5441,
"mean_token_accuracy": 0.158404940366745,
"num_tokens": 12222353.0,
"step": 5880
},
{
"entropy": 5.8088079452514645,
"epoch": 0.5338352685050798,
"grad_norm": 1.8515625,
"learning_rate": 0.000497783817734742,
"loss": 5.7913,
"mean_token_accuracy": 0.15034977793693544,
"num_tokens": 12232845.0,
"step": 5885
},
{
"entropy": 5.813466548919678,
"epoch": 0.534288824383164,
"grad_norm": 1.671875,
"learning_rate": 0.0004977792852571954,
"loss": 5.7517,
"mean_token_accuracy": 0.1444981053471565,
"num_tokens": 12242950.0,
"step": 5890
},
{
"entropy": 5.866303157806397,
"epoch": 0.5347423802612482,
"grad_norm": 2.890625,
"learning_rate": 0.0004977747481725164,
"loss": 5.7329,
"mean_token_accuracy": 0.14827523976564408,
"num_tokens": 12253132.0,
"step": 5895
},
{
"entropy": 5.847100877761841,
"epoch": 0.5351959361393324,
"grad_norm": 2.71875,
"learning_rate": 0.0004977702064807987,
"loss": 5.6634,
"mean_token_accuracy": 0.15220373421907424,
"num_tokens": 12263214.0,
"step": 5900
},
{
"entropy": 5.721499347686768,
"epoch": 0.5356494920174165,
"grad_norm": 1.7890625,
"learning_rate": 0.0004977656601821366,
"loss": 5.6111,
"mean_token_accuracy": 0.16137520372867584,
"num_tokens": 12273250.0,
"step": 5905
},
{
"entropy": 5.920711851119995,
"epoch": 0.5361030478955007,
"grad_norm": 2.15625,
"learning_rate": 0.0004977611092766238,
"loss": 5.7896,
"mean_token_accuracy": 0.15604286193847655,
"num_tokens": 12283767.0,
"step": 5910
},
{
"entropy": 5.821226930618286,
"epoch": 0.5365566037735849,
"grad_norm": 4.5625,
"learning_rate": 0.0004977565537643544,
"loss": 5.6478,
"mean_token_accuracy": 0.14899064600467682,
"num_tokens": 12294612.0,
"step": 5915
},
{
"entropy": 5.75635838508606,
"epoch": 0.5370101596516691,
"grad_norm": 2.3125,
"learning_rate": 0.0004977519936454227,
"loss": 5.7324,
"mean_token_accuracy": 0.15379261076450348,
"num_tokens": 12305614.0,
"step": 5920
},
{
"entropy": 5.7148298740386965,
"epoch": 0.5374637155297532,
"grad_norm": 2.03125,
"learning_rate": 0.0004977474289199231,
"loss": 5.6016,
"mean_token_accuracy": 0.14994118511676788,
"num_tokens": 12316152.0,
"step": 5925
},
{
"entropy": 5.771187782287598,
"epoch": 0.5379172714078374,
"grad_norm": 2.03125,
"learning_rate": 0.0004977428595879499,
"loss": 5.6922,
"mean_token_accuracy": 0.156583671271801,
"num_tokens": 12325940.0,
"step": 5930
},
{
"entropy": 5.850227928161621,
"epoch": 0.5383708272859217,
"grad_norm": 1.8984375,
"learning_rate": 0.0004977382856495975,
"loss": 5.6883,
"mean_token_accuracy": 0.15705590099096298,
"num_tokens": 12335482.0,
"step": 5935
},
{
"entropy": 5.897109460830689,
"epoch": 0.5388243831640058,
"grad_norm": 2.328125,
"learning_rate": 0.0004977337071049606,
"loss": 5.7659,
"mean_token_accuracy": 0.14804613441228867,
"num_tokens": 12346386.0,
"step": 5940
},
{
"entropy": 5.881980228424072,
"epoch": 0.53927793904209,
"grad_norm": 1.859375,
"learning_rate": 0.000497729123954134,
"loss": 5.7127,
"mean_token_accuracy": 0.15534048080444335,
"num_tokens": 12356282.0,
"step": 5945
},
{
"entropy": 5.8060948848724365,
"epoch": 0.5397314949201741,
"grad_norm": 1.828125,
"learning_rate": 0.0004977245361972123,
"loss": 5.7073,
"mean_token_accuracy": 0.15651538074016572,
"num_tokens": 12366994.0,
"step": 5950
},
{
"entropy": 5.762063932418823,
"epoch": 0.5401850507982583,
"grad_norm": 3.984375,
"learning_rate": 0.0004977199438342904,
"loss": 5.5904,
"mean_token_accuracy": 0.15490693598985672,
"num_tokens": 12377524.0,
"step": 5955
},
{
"entropy": 5.752496385574341,
"epoch": 0.5406386066763426,
"grad_norm": 2.65625,
"learning_rate": 0.0004977153468654634,
"loss": 5.7448,
"mean_token_accuracy": 0.15410600006580352,
"num_tokens": 12387593.0,
"step": 5960
},
{
"entropy": 5.861761140823364,
"epoch": 0.5410921625544267,
"grad_norm": 2.875,
"learning_rate": 0.0004977107452908262,
"loss": 5.7276,
"mean_token_accuracy": 0.14881233870983124,
"num_tokens": 12398732.0,
"step": 5965
},
{
"entropy": 5.880192470550537,
"epoch": 0.5415457184325109,
"grad_norm": 2.578125,
"learning_rate": 0.000497706139110474,
"loss": 5.7008,
"mean_token_accuracy": 0.15882584676146508,
"num_tokens": 12409821.0,
"step": 5970
},
{
"entropy": 5.842268705368042,
"epoch": 0.541999274310595,
"grad_norm": 2.265625,
"learning_rate": 0.0004977015283245021,
"loss": 5.7328,
"mean_token_accuracy": 0.14897113144397736,
"num_tokens": 12420044.0,
"step": 5975
},
{
"entropy": 5.760256767272949,
"epoch": 0.5424528301886793,
"grad_norm": 3.03125,
"learning_rate": 0.0004976969129330059,
"loss": 5.6512,
"mean_token_accuracy": 0.15109718292951585,
"num_tokens": 12430980.0,
"step": 5980
},
{
"entropy": 5.786817407608032,
"epoch": 0.5429063860667634,
"grad_norm": 1.9375,
"learning_rate": 0.0004976922929360808,
"loss": 5.7144,
"mean_token_accuracy": 0.15620118379592896,
"num_tokens": 12441999.0,
"step": 5985
},
{
"entropy": 5.855512714385986,
"epoch": 0.5433599419448476,
"grad_norm": 3.65625,
"learning_rate": 0.0004976876683338222,
"loss": 5.7487,
"mean_token_accuracy": 0.14880464673042298,
"num_tokens": 12452837.0,
"step": 5990
},
{
"entropy": 5.810596656799317,
"epoch": 0.5438134978229318,
"grad_norm": 2.421875,
"learning_rate": 0.000497683039126326,
"loss": 5.666,
"mean_token_accuracy": 0.1583888977766037,
"num_tokens": 12462142.0,
"step": 5995
},
{
"entropy": 5.80487813949585,
"epoch": 0.5442670537010159,
"grad_norm": 2.375,
"learning_rate": 0.0004976784053136878,
"loss": 5.706,
"mean_token_accuracy": 0.15387148559093475,
"num_tokens": 12472756.0,
"step": 6000
},
{
"epoch": 0.5442670537010159,
"eval_entropy": 5.610305411999042,
"eval_loss": 5.737067699432373,
"eval_mean_token_accuracy": 0.15633446278547808,
"eval_num_tokens": 12472756.0,
"eval_runtime": 26.751,
"eval_samples_per_second": 1321.818,
"eval_steps_per_second": 165.227,
"step": 6000
},
{
"entropy": 5.83179087638855,
"epoch": 0.5447206095791002,
"grad_norm": 2.734375,
"learning_rate": 0.0004976737668960034,
"loss": 5.8127,
"mean_token_accuracy": 0.14738722145557404,
"num_tokens": 12483115.0,
"step": 6005
},
{
"entropy": 5.831946611404419,
"epoch": 0.5451741654571843,
"grad_norm": 2.0,
"learning_rate": 0.0004976691238733688,
"loss": 5.6753,
"mean_token_accuracy": 0.1467287912964821,
"num_tokens": 12493280.0,
"step": 6010
},
{
"entropy": 5.849226713180542,
"epoch": 0.5456277213352685,
"grad_norm": 2.65625,
"learning_rate": 0.00049766447624588,
"loss": 5.7941,
"mean_token_accuracy": 0.15283684581518173,
"num_tokens": 12503665.0,
"step": 6015
},
{
"entropy": 5.815784883499146,
"epoch": 0.5460812772133526,
"grad_norm": 4.0625,
"learning_rate": 0.000497659824013633,
"loss": 5.7144,
"mean_token_accuracy": 0.14608054608106613,
"num_tokens": 12513381.0,
"step": 6020
},
{
"entropy": 5.895172071456909,
"epoch": 0.5465348330914369,
"grad_norm": 2.8125,
"learning_rate": 0.0004976551671767241,
"loss": 5.754,
"mean_token_accuracy": 0.1483724221587181,
"num_tokens": 12524121.0,
"step": 6025
},
{
"entropy": 5.840110015869141,
"epoch": 0.5469883889695211,
"grad_norm": 3.328125,
"learning_rate": 0.0004976505057352496,
"loss": 5.7072,
"mean_token_accuracy": 0.15462427884340285,
"num_tokens": 12533972.0,
"step": 6030
},
{
"entropy": 5.871407890319825,
"epoch": 0.5474419448476052,
"grad_norm": 5.8125,
"learning_rate": 0.0004976458396893059,
"loss": 5.8162,
"mean_token_accuracy": 0.14339689910411835,
"num_tokens": 12544110.0,
"step": 6035
},
{
"entropy": 5.787636804580688,
"epoch": 0.5478955007256894,
"grad_norm": 2.46875,
"learning_rate": 0.0004976411690389896,
"loss": 5.6686,
"mean_token_accuracy": 0.14848399460315703,
"num_tokens": 12554273.0,
"step": 6040
},
{
"entropy": 5.903516960144043,
"epoch": 0.5483490566037735,
"grad_norm": 3.125,
"learning_rate": 0.000497636493784397,
"loss": 5.7501,
"mean_token_accuracy": 0.1464612677693367,
"num_tokens": 12564633.0,
"step": 6045
},
{
"entropy": 5.910893726348877,
"epoch": 0.5488026124818578,
"grad_norm": 3.984375,
"learning_rate": 0.0004976318139256251,
"loss": 5.6743,
"mean_token_accuracy": 0.1580335780978203,
"num_tokens": 12573790.0,
"step": 6050
},
{
"entropy": 5.878701543807983,
"epoch": 0.549256168359942,
"grad_norm": 2.96875,
"learning_rate": 0.0004976271294627706,
"loss": 5.747,
"mean_token_accuracy": 0.14373386204242705,
"num_tokens": 12584265.0,
"step": 6055
},
{
"entropy": 5.770555639266968,
"epoch": 0.5497097242380261,
"grad_norm": 7.21875,
"learning_rate": 0.0004976224403959303,
"loss": 5.6169,
"mean_token_accuracy": 0.15551466792821883,
"num_tokens": 12594413.0,
"step": 6060
},
{
"entropy": 5.840236854553223,
"epoch": 0.5501632801161103,
"grad_norm": 2.796875,
"learning_rate": 0.0004976177467252012,
"loss": 5.6874,
"mean_token_accuracy": 0.15790211111307145,
"num_tokens": 12604637.0,
"step": 6065
},
{
"entropy": 5.8618628025054935,
"epoch": 0.5506168359941945,
"grad_norm": 2.34375,
"learning_rate": 0.0004976130484506804,
"loss": 5.7417,
"mean_token_accuracy": 0.15392341166734697,
"num_tokens": 12614873.0,
"step": 6070
},
{
"entropy": 5.796398115158081,
"epoch": 0.5510703918722787,
"grad_norm": 2.25,
"learning_rate": 0.0004976083455724649,
"loss": 5.6818,
"mean_token_accuracy": 0.15129643827676773,
"num_tokens": 12626182.0,
"step": 6075
},
{
"entropy": 5.792143869400024,
"epoch": 0.5515239477503628,
"grad_norm": 2.390625,
"learning_rate": 0.0004976036380906523,
"loss": 5.6363,
"mean_token_accuracy": 0.15263219848275184,
"num_tokens": 12635418.0,
"step": 6080
},
{
"entropy": 5.859647369384765,
"epoch": 0.551977503628447,
"grad_norm": 3.046875,
"learning_rate": 0.0004975989260053395,
"loss": 5.7718,
"mean_token_accuracy": 0.1413577154278755,
"num_tokens": 12645942.0,
"step": 6085
},
{
"entropy": 5.834799003601074,
"epoch": 0.5524310595065312,
"grad_norm": 3.328125,
"learning_rate": 0.0004975942093166242,
"loss": 5.6965,
"mean_token_accuracy": 0.1482281967997551,
"num_tokens": 12656460.0,
"step": 6090
},
{
"entropy": 5.838398408889771,
"epoch": 0.5528846153846154,
"grad_norm": 2.453125,
"learning_rate": 0.0004975894880246041,
"loss": 5.6335,
"mean_token_accuracy": 0.15041477978229523,
"num_tokens": 12666916.0,
"step": 6095
},
{
"entropy": 5.82815203666687,
"epoch": 0.5533381712626996,
"grad_norm": 2.90625,
"learning_rate": 0.0004975847621293766,
"loss": 5.7116,
"mean_token_accuracy": 0.1567052870988846,
"num_tokens": 12677562.0,
"step": 6100
},
{
"entropy": 5.815052461624146,
"epoch": 0.5537917271407837,
"grad_norm": 2.78125,
"learning_rate": 0.0004975800316310394,
"loss": 5.6512,
"mean_token_accuracy": 0.15303489565849304,
"num_tokens": 12687349.0,
"step": 6105
},
{
"entropy": 5.88008337020874,
"epoch": 0.5542452830188679,
"grad_norm": 2.953125,
"learning_rate": 0.0004975752965296906,
"loss": 5.7239,
"mean_token_accuracy": 0.15200326591730118,
"num_tokens": 12697850.0,
"step": 6110
},
{
"entropy": 5.805438089370727,
"epoch": 0.5546988388969522,
"grad_norm": 2.46875,
"learning_rate": 0.0004975705568254278,
"loss": 5.5851,
"mean_token_accuracy": 0.15197097063064574,
"num_tokens": 12708283.0,
"step": 6115
},
{
"entropy": 5.785101699829101,
"epoch": 0.5551523947750363,
"grad_norm": 2.78125,
"learning_rate": 0.0004975658125183493,
"loss": 5.6675,
"mean_token_accuracy": 0.15603910833597184,
"num_tokens": 12718201.0,
"step": 6120
},
{
"entropy": 5.77977442741394,
"epoch": 0.5556059506531205,
"grad_norm": 3.28125,
"learning_rate": 0.000497561063608553,
"loss": 5.6441,
"mean_token_accuracy": 0.1591254323720932,
"num_tokens": 12728860.0,
"step": 6125
},
{
"entropy": 5.882054996490479,
"epoch": 0.5560595065312046,
"grad_norm": 2.3125,
"learning_rate": 0.0004975563100961372,
"loss": 5.7045,
"mean_token_accuracy": 0.14953402429819107,
"num_tokens": 12739007.0,
"step": 6130
},
{
"entropy": 5.851760959625244,
"epoch": 0.5565130624092888,
"grad_norm": 2.765625,
"learning_rate": 0.0004975515519812002,
"loss": 5.5603,
"mean_token_accuracy": 0.1571223884820938,
"num_tokens": 12749004.0,
"step": 6135
},
{
"entropy": 5.719464206695557,
"epoch": 0.556966618287373,
"grad_norm": 2.484375,
"learning_rate": 0.0004975467892638402,
"loss": 5.7312,
"mean_token_accuracy": 0.15219081193208694,
"num_tokens": 12760085.0,
"step": 6140
},
{
"entropy": 5.780846977233887,
"epoch": 0.5574201741654572,
"grad_norm": 2.125,
"learning_rate": 0.0004975420219441561,
"loss": 5.6557,
"mean_token_accuracy": 0.15899507701396942,
"num_tokens": 12770303.0,
"step": 6145
},
{
"entropy": 5.880466079711914,
"epoch": 0.5578737300435413,
"grad_norm": 2.1875,
"learning_rate": 0.0004975372500222462,
"loss": 5.6693,
"mean_token_accuracy": 0.15486729443073272,
"num_tokens": 12780334.0,
"step": 6150
},
{
"entropy": 5.8313295364379885,
"epoch": 0.5583272859216255,
"grad_norm": 2.5,
"learning_rate": 0.0004975324734982094,
"loss": 5.7253,
"mean_token_accuracy": 0.1489449195563793,
"num_tokens": 12790980.0,
"step": 6155
},
{
"entropy": 5.845129108428955,
"epoch": 0.5587808417997098,
"grad_norm": 1.6875,
"learning_rate": 0.0004975276923721442,
"loss": 5.7754,
"mean_token_accuracy": 0.14961862117052077,
"num_tokens": 12801713.0,
"step": 6160
},
{
"entropy": 5.870355415344238,
"epoch": 0.5592343976777939,
"grad_norm": 2.234375,
"learning_rate": 0.0004975229066441496,
"loss": 5.758,
"mean_token_accuracy": 0.14657938480377197,
"num_tokens": 12813040.0,
"step": 6165
},
{
"entropy": 5.833003377914428,
"epoch": 0.5596879535558781,
"grad_norm": 1.8828125,
"learning_rate": 0.0004975181163143247,
"loss": 5.713,
"mean_token_accuracy": 0.1453935071825981,
"num_tokens": 12824048.0,
"step": 6170
},
{
"entropy": 5.837733316421509,
"epoch": 0.5601415094339622,
"grad_norm": 1.890625,
"learning_rate": 0.0004975133213827684,
"loss": 5.6621,
"mean_token_accuracy": 0.14674486815929413,
"num_tokens": 12834297.0,
"step": 6175
},
{
"entropy": 5.896747970581055,
"epoch": 0.5605950653120464,
"grad_norm": 1.75,
"learning_rate": 0.0004975085218495798,
"loss": 5.7316,
"mean_token_accuracy": 0.1485936962068081,
"num_tokens": 12844574.0,
"step": 6180
},
{
"entropy": 5.765360546112061,
"epoch": 0.5610486211901307,
"grad_norm": 2.265625,
"learning_rate": 0.0004975037177148585,
"loss": 5.6848,
"mean_token_accuracy": 0.15071796774864196,
"num_tokens": 12854670.0,
"step": 6185
},
{
"entropy": 5.713732099533081,
"epoch": 0.5615021770682148,
"grad_norm": 2.203125,
"learning_rate": 0.0004974989089787035,
"loss": 5.7079,
"mean_token_accuracy": 0.15234477072954178,
"num_tokens": 12865281.0,
"step": 6190
},
{
"entropy": 5.8670814514160154,
"epoch": 0.561955732946299,
"grad_norm": 1.953125,
"learning_rate": 0.0004974940956412143,
"loss": 5.6978,
"mean_token_accuracy": 0.1518659010529518,
"num_tokens": 12875222.0,
"step": 6195
},
{
"entropy": 5.996776294708252,
"epoch": 0.5624092888243831,
"grad_norm": 1.7578125,
"learning_rate": 0.0004974892777024905,
"loss": 5.7535,
"mean_token_accuracy": 0.14664066731929778,
"num_tokens": 12885079.0,
"step": 6200
},
{
"entropy": 5.816704797744751,
"epoch": 0.5628628447024674,
"grad_norm": 2.328125,
"learning_rate": 0.0004974844551626318,
"loss": 5.7058,
"mean_token_accuracy": 0.15577784329652786,
"num_tokens": 12896402.0,
"step": 6205
},
{
"entropy": 5.7654026508331295,
"epoch": 0.5633164005805515,
"grad_norm": 2.03125,
"learning_rate": 0.0004974796280217379,
"loss": 5.6738,
"mean_token_accuracy": 0.16155672520399095,
"num_tokens": 12906403.0,
"step": 6210
},
{
"entropy": 5.881820344924927,
"epoch": 0.5637699564586357,
"grad_norm": 2.21875,
"learning_rate": 0.0004974747962799086,
"loss": 5.694,
"mean_token_accuracy": 0.14602868854999543,
"num_tokens": 12917061.0,
"step": 6215
},
{
"entropy": 5.8594695091247555,
"epoch": 0.5642235123367199,
"grad_norm": 7.03125,
"learning_rate": 0.0004974699599372437,
"loss": 5.7495,
"mean_token_accuracy": 0.15344917625188828,
"num_tokens": 12927182.0,
"step": 6220
},
{
"entropy": 5.867999505996704,
"epoch": 0.564677068214804,
"grad_norm": 1.9140625,
"learning_rate": 0.0004974651189938434,
"loss": 5.683,
"mean_token_accuracy": 0.15102926790714263,
"num_tokens": 12937371.0,
"step": 6225
},
{
"entropy": 5.83708872795105,
"epoch": 0.5651306240928883,
"grad_norm": 2.1875,
"learning_rate": 0.0004974602734498079,
"loss": 5.7392,
"mean_token_accuracy": 0.1440092071890831,
"num_tokens": 12948569.0,
"step": 6230
},
{
"entropy": 5.8283039093017575,
"epoch": 0.5655841799709724,
"grad_norm": 1.9140625,
"learning_rate": 0.000497455423305237,
"loss": 5.6769,
"mean_token_accuracy": 0.14810833483934402,
"num_tokens": 12958503.0,
"step": 6235
},
{
"entropy": 5.854265213012695,
"epoch": 0.5660377358490566,
"grad_norm": 2.265625,
"learning_rate": 0.0004974505685602314,
"loss": 5.6917,
"mean_token_accuracy": 0.15404328405857087,
"num_tokens": 12968714.0,
"step": 6240
},
{
"entropy": 5.803040409088135,
"epoch": 0.5664912917271407,
"grad_norm": 2.640625,
"learning_rate": 0.0004974457092148912,
"loss": 5.7329,
"mean_token_accuracy": 0.14936806112527848,
"num_tokens": 12978857.0,
"step": 6245
},
{
"entropy": 5.967836856842041,
"epoch": 0.566944847605225,
"grad_norm": 2.375,
"learning_rate": 0.0004974408452693172,
"loss": 5.8515,
"mean_token_accuracy": 0.14415958821773528,
"num_tokens": 12989863.0,
"step": 6250
},
{
"entropy": 5.897352266311645,
"epoch": 0.5673984034833092,
"grad_norm": 2.484375,
"learning_rate": 0.0004974359767236099,
"loss": 5.648,
"mean_token_accuracy": 0.15427448004484176,
"num_tokens": 13000307.0,
"step": 6255
},
{
"entropy": 5.835937643051148,
"epoch": 0.5678519593613933,
"grad_norm": 2.671875,
"learning_rate": 0.0004974311035778698,
"loss": 5.7076,
"mean_token_accuracy": 0.16386753916740418,
"num_tokens": 13009053.0,
"step": 6260
},
{
"entropy": 5.788226127624512,
"epoch": 0.5683055152394775,
"grad_norm": 1.859375,
"learning_rate": 0.0004974262258321979,
"loss": 5.7386,
"mean_token_accuracy": 0.14352438002824783,
"num_tokens": 13019136.0,
"step": 6265
},
{
"entropy": 5.879953765869141,
"epoch": 0.5687590711175616,
"grad_norm": 2.390625,
"learning_rate": 0.0004974213434866949,
"loss": 5.7667,
"mean_token_accuracy": 0.15057471692562102,
"num_tokens": 13030356.0,
"step": 6270
},
{
"entropy": 5.882538843154907,
"epoch": 0.5692126269956459,
"grad_norm": 2.953125,
"learning_rate": 0.0004974164565414617,
"loss": 5.7889,
"mean_token_accuracy": 0.1513417199254036,
"num_tokens": 13040477.0,
"step": 6275
},
{
"entropy": 5.837933874130249,
"epoch": 0.56966618287373,
"grad_norm": 2.046875,
"learning_rate": 0.0004974115649965998,
"loss": 5.6785,
"mean_token_accuracy": 0.1549770399928093,
"num_tokens": 13051755.0,
"step": 6280
},
{
"entropy": 5.841934442520142,
"epoch": 0.5701197387518142,
"grad_norm": 1.9375,
"learning_rate": 0.0004974066688522098,
"loss": 5.6869,
"mean_token_accuracy": 0.15811796933412553,
"num_tokens": 13061670.0,
"step": 6285
},
{
"entropy": 5.760716295242309,
"epoch": 0.5705732946298984,
"grad_norm": 2.34375,
"learning_rate": 0.0004974017681083934,
"loss": 5.6512,
"mean_token_accuracy": 0.15475476533174515,
"num_tokens": 13070917.0,
"step": 6290
},
{
"entropy": 5.800657749176025,
"epoch": 0.5710268505079826,
"grad_norm": 2.828125,
"learning_rate": 0.0004973968627652516,
"loss": 5.7069,
"mean_token_accuracy": 0.15006187707185745,
"num_tokens": 13080577.0,
"step": 6295
},
{
"entropy": 5.851555442810058,
"epoch": 0.5714804063860668,
"grad_norm": 2.359375,
"learning_rate": 0.000497391952822886,
"loss": 5.7457,
"mean_token_accuracy": 0.15279553458094597,
"num_tokens": 13091446.0,
"step": 6300
},
{
"entropy": 5.947834491729736,
"epoch": 0.5719339622641509,
"grad_norm": 2.375,
"learning_rate": 0.0004973870382813983,
"loss": 5.7611,
"mean_token_accuracy": 0.15199481695890427,
"num_tokens": 13102752.0,
"step": 6305
},
{
"entropy": 5.940124607086181,
"epoch": 0.5723875181422351,
"grad_norm": 2.34375,
"learning_rate": 0.0004973821191408899,
"loss": 5.7886,
"mean_token_accuracy": 0.1465797521173954,
"num_tokens": 13113944.0,
"step": 6310
},
{
"entropy": 5.7535113334655765,
"epoch": 0.5728410740203193,
"grad_norm": 1.953125,
"learning_rate": 0.0004973771954014626,
"loss": 5.7191,
"mean_token_accuracy": 0.14859745651483536,
"num_tokens": 13124316.0,
"step": 6315
},
{
"entropy": 5.874121236801147,
"epoch": 0.5732946298984035,
"grad_norm": 2.609375,
"learning_rate": 0.0004973722670632182,
"loss": 5.8198,
"mean_token_accuracy": 0.14575577676296234,
"num_tokens": 13133567.0,
"step": 6320
},
{
"entropy": 5.899265956878662,
"epoch": 0.5737481857764877,
"grad_norm": 4.46875,
"learning_rate": 0.0004973673341262587,
"loss": 5.7334,
"mean_token_accuracy": 0.14762221351265908,
"num_tokens": 13144279.0,
"step": 6325
},
{
"entropy": 5.887506055831909,
"epoch": 0.5742017416545718,
"grad_norm": 2.671875,
"learning_rate": 0.000497362396590686,
"loss": 5.7578,
"mean_token_accuracy": 0.1508809007704258,
"num_tokens": 13154463.0,
"step": 6330
},
{
"entropy": 5.8973757266998295,
"epoch": 0.574655297532656,
"grad_norm": 2.5,
"learning_rate": 0.0004973574544566024,
"loss": 5.7407,
"mean_token_accuracy": 0.14477827101945878,
"num_tokens": 13165565.0,
"step": 6335
},
{
"entropy": 5.804881763458252,
"epoch": 0.5751088534107403,
"grad_norm": 4.0,
"learning_rate": 0.0004973525077241098,
"loss": 5.6568,
"mean_token_accuracy": 0.15822704434394835,
"num_tokens": 13175839.0,
"step": 6340
},
{
"entropy": 5.780976009368897,
"epoch": 0.5755624092888244,
"grad_norm": 2.046875,
"learning_rate": 0.0004973475563933109,
"loss": 5.6726,
"mean_token_accuracy": 0.14872757494449615,
"num_tokens": 13186372.0,
"step": 6345
},
{
"entropy": 5.8148088455200195,
"epoch": 0.5760159651669086,
"grad_norm": 3.4375,
"learning_rate": 0.0004973426004643077,
"loss": 5.702,
"mean_token_accuracy": 0.15282864421606063,
"num_tokens": 13196259.0,
"step": 6350
},
{
"entropy": 5.822752189636231,
"epoch": 0.5764695210449927,
"grad_norm": 3.390625,
"learning_rate": 0.0004973376399372029,
"loss": 5.694,
"mean_token_accuracy": 0.15161322802305222,
"num_tokens": 13206497.0,
"step": 6355
},
{
"entropy": 5.866645383834839,
"epoch": 0.5769230769230769,
"grad_norm": 1.5625,
"learning_rate": 0.0004973326748120991,
"loss": 5.7283,
"mean_token_accuracy": 0.15126238614320756,
"num_tokens": 13217188.0,
"step": 6360
},
{
"entropy": 5.846437931060791,
"epoch": 0.5773766328011611,
"grad_norm": 2.078125,
"learning_rate": 0.0004973277050890988,
"loss": 5.7288,
"mean_token_accuracy": 0.1515586942434311,
"num_tokens": 13226544.0,
"step": 6365
},
{
"entropy": 5.839645147323608,
"epoch": 0.5778301886792453,
"grad_norm": 2.0625,
"learning_rate": 0.0004973227307683049,
"loss": 5.7764,
"mean_token_accuracy": 0.14674135893583298,
"num_tokens": 13236572.0,
"step": 6370
},
{
"entropy": 5.9877115249633786,
"epoch": 0.5782837445573294,
"grad_norm": 2.09375,
"learning_rate": 0.0004973177518498204,
"loss": 5.7779,
"mean_token_accuracy": 0.14902501702308654,
"num_tokens": 13247958.0,
"step": 6375
},
{
"entropy": 5.8630200862884525,
"epoch": 0.5787373004354136,
"grad_norm": 1.9609375,
"learning_rate": 0.0004973127683337482,
"loss": 5.6749,
"mean_token_accuracy": 0.14949809312820433,
"num_tokens": 13258255.0,
"step": 6380
},
{
"entropy": 5.744321727752686,
"epoch": 0.5791908563134979,
"grad_norm": 2.25,
"learning_rate": 0.0004973077802201911,
"loss": 5.6909,
"mean_token_accuracy": 0.15299025177955627,
"num_tokens": 13268416.0,
"step": 6385
},
{
"entropy": 5.801497316360473,
"epoch": 0.579644412191582,
"grad_norm": 2.875,
"learning_rate": 0.0004973027875092526,
"loss": 5.6943,
"mean_token_accuracy": 0.15058967620134353,
"num_tokens": 13278087.0,
"step": 6390
},
{
"entropy": 5.81386513710022,
"epoch": 0.5800979680696662,
"grad_norm": 1.9765625,
"learning_rate": 0.0004972977902010357,
"loss": 5.592,
"mean_token_accuracy": 0.1598864108324051,
"num_tokens": 13288121.0,
"step": 6395
},
{
"entropy": 5.926956272125244,
"epoch": 0.5805515239477503,
"grad_norm": 2.25,
"learning_rate": 0.0004972927882956439,
"loss": 5.7942,
"mean_token_accuracy": 0.14235642403364182,
"num_tokens": 13298620.0,
"step": 6400
},
{
"entropy": 5.7990469455719,
"epoch": 0.5810050798258345,
"grad_norm": 2.1875,
"learning_rate": 0.0004972877817931807,
"loss": 5.6142,
"mean_token_accuracy": 0.14995205849409105,
"num_tokens": 13308413.0,
"step": 6405
},
{
"entropy": 5.788681268692017,
"epoch": 0.5814586357039188,
"grad_norm": 2.265625,
"learning_rate": 0.0004972827706937493,
"loss": 5.7003,
"mean_token_accuracy": 0.1515623390674591,
"num_tokens": 13319387.0,
"step": 6410
},
{
"entropy": 5.825255298614502,
"epoch": 0.5819121915820029,
"grad_norm": 2.390625,
"learning_rate": 0.0004972777549974537,
"loss": 5.6635,
"mean_token_accuracy": 0.1523427128791809,
"num_tokens": 13330494.0,
"step": 6415
},
{
"entropy": 5.863323593139649,
"epoch": 0.5823657474600871,
"grad_norm": 2.171875,
"learning_rate": 0.0004972727347043975,
"loss": 5.663,
"mean_token_accuracy": 0.15184202492237092,
"num_tokens": 13341009.0,
"step": 6420
},
{
"entropy": 5.795520925521851,
"epoch": 0.5828193033381712,
"grad_norm": 2.515625,
"learning_rate": 0.0004972677098146845,
"loss": 5.6168,
"mean_token_accuracy": 0.15109016746282578,
"num_tokens": 13351269.0,
"step": 6425
},
{
"entropy": 5.784488582611084,
"epoch": 0.5832728592162555,
"grad_norm": 2.625,
"learning_rate": 0.0004972626803284185,
"loss": 5.7503,
"mean_token_accuracy": 0.14944482147693633,
"num_tokens": 13362160.0,
"step": 6430
},
{
"entropy": 5.889904022216797,
"epoch": 0.5837264150943396,
"grad_norm": 3.375,
"learning_rate": 0.0004972576462457038,
"loss": 5.7866,
"mean_token_accuracy": 0.14930133521556854,
"num_tokens": 13372497.0,
"step": 6435
},
{
"entropy": 5.957094621658325,
"epoch": 0.5841799709724238,
"grad_norm": 2.125,
"learning_rate": 0.0004972526075666442,
"loss": 5.7257,
"mean_token_accuracy": 0.15872014164924622,
"num_tokens": 13381665.0,
"step": 6440
},
{
"entropy": 5.8321576595306395,
"epoch": 0.584633526850508,
"grad_norm": 2.546875,
"learning_rate": 0.0004972475642913442,
"loss": 5.6392,
"mean_token_accuracy": 0.1554955556988716,
"num_tokens": 13392507.0,
"step": 6445
},
{
"entropy": 5.76501989364624,
"epoch": 0.5850870827285921,
"grad_norm": 3.5,
"learning_rate": 0.0004972425164199078,
"loss": 5.6346,
"mean_token_accuracy": 0.1577868416905403,
"num_tokens": 13402171.0,
"step": 6450
},
{
"entropy": 5.8788012027740475,
"epoch": 0.5855406386066764,
"grad_norm": 1.90625,
"learning_rate": 0.0004972374639524396,
"loss": 5.8384,
"mean_token_accuracy": 0.14709555059671403,
"num_tokens": 13412933.0,
"step": 6455
},
{
"entropy": 5.856866788864136,
"epoch": 0.5859941944847605,
"grad_norm": 2.15625,
"learning_rate": 0.000497232406889044,
"loss": 5.6662,
"mean_token_accuracy": 0.1488827720284462,
"num_tokens": 13423563.0,
"step": 6460
},
{
"entropy": 5.896422910690307,
"epoch": 0.5864477503628447,
"grad_norm": 2.015625,
"learning_rate": 0.0004972273452298255,
"loss": 5.6954,
"mean_token_accuracy": 0.15475627183914184,
"num_tokens": 13433998.0,
"step": 6465
},
{
"entropy": 5.828073978424072,
"epoch": 0.5869013062409288,
"grad_norm": 2.375,
"learning_rate": 0.0004972222789748888,
"loss": 5.6352,
"mean_token_accuracy": 0.15622286945581437,
"num_tokens": 13443878.0,
"step": 6470
},
{
"entropy": 5.796691513061523,
"epoch": 0.5873548621190131,
"grad_norm": 1.7265625,
"learning_rate": 0.0004972172081243389,
"loss": 5.7249,
"mean_token_accuracy": 0.1440662868320942,
"num_tokens": 13456051.0,
"step": 6475
},
{
"entropy": 5.815340518951416,
"epoch": 0.5878084179970973,
"grad_norm": 4.375,
"learning_rate": 0.0004972121326782806,
"loss": 5.7373,
"mean_token_accuracy": 0.15519822090864183,
"num_tokens": 13467276.0,
"step": 6480
},
{
"entropy": 5.861797666549682,
"epoch": 0.5882619738751814,
"grad_norm": 2.421875,
"learning_rate": 0.0004972070526368186,
"loss": 5.7039,
"mean_token_accuracy": 0.14981682896614074,
"num_tokens": 13478331.0,
"step": 6485
},
{
"entropy": 5.927264070510864,
"epoch": 0.5887155297532656,
"grad_norm": 1.7734375,
"learning_rate": 0.0004972019680000582,
"loss": 5.6366,
"mean_token_accuracy": 0.1500559240579605,
"num_tokens": 13489879.0,
"step": 6490
},
{
"entropy": 5.851372814178466,
"epoch": 0.5891690856313497,
"grad_norm": 3.046875,
"learning_rate": 0.0004971968787681044,
"loss": 5.763,
"mean_token_accuracy": 0.1548596978187561,
"num_tokens": 13500879.0,
"step": 6495
},
{
"entropy": 5.862674856185913,
"epoch": 0.589622641509434,
"grad_norm": 3.15625,
"learning_rate": 0.0004971917849410627,
"loss": 5.7439,
"mean_token_accuracy": 0.14779864698648454,
"num_tokens": 13511719.0,
"step": 6500
},
{
"entropy": 5.834804105758667,
"epoch": 0.5900761973875182,
"grad_norm": 2.0625,
"learning_rate": 0.0004971866865190381,
"loss": 5.6717,
"mean_token_accuracy": 0.1496259018778801,
"num_tokens": 13521850.0,
"step": 6505
},
{
"entropy": 5.923253965377808,
"epoch": 0.5905297532656023,
"grad_norm": 2.828125,
"learning_rate": 0.0004971815835021362,
"loss": 5.6981,
"mean_token_accuracy": 0.15618499219417573,
"num_tokens": 13532095.0,
"step": 6510
},
{
"entropy": 5.815547132492066,
"epoch": 0.5909833091436865,
"grad_norm": 3.203125,
"learning_rate": 0.0004971764758904626,
"loss": 5.6828,
"mean_token_accuracy": 0.14781468808650972,
"num_tokens": 13543133.0,
"step": 6515
},
{
"entropy": 5.811095142364502,
"epoch": 0.5914368650217707,
"grad_norm": 2.9375,
"learning_rate": 0.0004971713636841229,
"loss": 5.7401,
"mean_token_accuracy": 0.14562103599309922,
"num_tokens": 13553771.0,
"step": 6520
},
{
"entropy": 5.810519790649414,
"epoch": 0.5918904208998549,
"grad_norm": 2.03125,
"learning_rate": 0.0004971662468832228,
"loss": 5.6905,
"mean_token_accuracy": 0.15237817242741586,
"num_tokens": 13564548.0,
"step": 6525
},
{
"entropy": 5.833587503433227,
"epoch": 0.592343976777939,
"grad_norm": 2.265625,
"learning_rate": 0.000497161125487868,
"loss": 5.6855,
"mean_token_accuracy": 0.15271230041980743,
"num_tokens": 13574567.0,
"step": 6530
},
{
"entropy": 5.78583288192749,
"epoch": 0.5927975326560232,
"grad_norm": 2.453125,
"learning_rate": 0.0004971559994981645,
"loss": 5.7089,
"mean_token_accuracy": 0.1522444397211075,
"num_tokens": 13585302.0,
"step": 6535
},
{
"entropy": 5.878629112243653,
"epoch": 0.5932510885341074,
"grad_norm": 2.8125,
"learning_rate": 0.0004971508689142184,
"loss": 5.7447,
"mean_token_accuracy": 0.14830908924341202,
"num_tokens": 13594754.0,
"step": 6540
},
{
"entropy": 5.83653244972229,
"epoch": 0.5937046444121916,
"grad_norm": 2.328125,
"learning_rate": 0.0004971457337361356,
"loss": 5.6665,
"mean_token_accuracy": 0.156255804002285,
"num_tokens": 13604691.0,
"step": 6545
},
{
"entropy": 5.710748481750488,
"epoch": 0.5941582002902758,
"grad_norm": 2.484375,
"learning_rate": 0.0004971405939640225,
"loss": 5.5835,
"mean_token_accuracy": 0.15645805299282073,
"num_tokens": 13613911.0,
"step": 6550
},
{
"entropy": 5.8865584373474125,
"epoch": 0.5946117561683599,
"grad_norm": 2.3125,
"learning_rate": 0.0004971354495979854,
"loss": 5.7821,
"mean_token_accuracy": 0.15241619274020196,
"num_tokens": 13624285.0,
"step": 6555
},
{
"entropy": 5.771980142593383,
"epoch": 0.5950653120464441,
"grad_norm": 3.34375,
"learning_rate": 0.0004971303006381305,
"loss": 5.6256,
"mean_token_accuracy": 0.1564767062664032,
"num_tokens": 13635434.0,
"step": 6560
},
{
"entropy": 5.796686601638794,
"epoch": 0.5955188679245284,
"grad_norm": 3.84375,
"learning_rate": 0.0004971251470845643,
"loss": 5.6973,
"mean_token_accuracy": 0.15278116166591643,
"num_tokens": 13645582.0,
"step": 6565
},
{
"entropy": 5.8198816776275635,
"epoch": 0.5959724238026125,
"grad_norm": 2.875,
"learning_rate": 0.0004971199889373935,
"loss": 5.6479,
"mean_token_accuracy": 0.15285747945308686,
"num_tokens": 13655029.0,
"step": 6570
},
{
"entropy": 5.809751796722412,
"epoch": 0.5964259796806967,
"grad_norm": 2.390625,
"learning_rate": 0.0004971148261967249,
"loss": 5.6286,
"mean_token_accuracy": 0.15282736122608184,
"num_tokens": 13665204.0,
"step": 6575
},
{
"entropy": 5.816667556762695,
"epoch": 0.5968795355587808,
"grad_norm": 2.78125,
"learning_rate": 0.0004971096588626649,
"loss": 5.7408,
"mean_token_accuracy": 0.15095006227493285,
"num_tokens": 13676281.0,
"step": 6580
},
{
"entropy": 5.780574703216553,
"epoch": 0.597333091436865,
"grad_norm": 2.25,
"learning_rate": 0.0004971044869353206,
"loss": 5.7281,
"mean_token_accuracy": 0.15921155661344527,
"num_tokens": 13687144.0,
"step": 6585
},
{
"entropy": 5.795083045959473,
"epoch": 0.5977866473149492,
"grad_norm": 1.953125,
"learning_rate": 0.000497099310414799,
"loss": 5.688,
"mean_token_accuracy": 0.15438786298036575,
"num_tokens": 13696371.0,
"step": 6590
},
{
"entropy": 5.84175672531128,
"epoch": 0.5982402031930334,
"grad_norm": 1.9296875,
"learning_rate": 0.0004970941293012069,
"loss": 5.6345,
"mean_token_accuracy": 0.15077201277017593,
"num_tokens": 13706900.0,
"step": 6595
},
{
"entropy": 5.836261129379272,
"epoch": 0.5986937590711175,
"grad_norm": 2.09375,
"learning_rate": 0.0004970889435946518,
"loss": 5.7144,
"mean_token_accuracy": 0.15443189740180968,
"num_tokens": 13717874.0,
"step": 6600
},
{
"entropy": 5.776962184906006,
"epoch": 0.5991473149492017,
"grad_norm": 2.75,
"learning_rate": 0.0004970837532952406,
"loss": 5.6737,
"mean_token_accuracy": 0.1497507020831108,
"num_tokens": 13728020.0,
"step": 6605
},
{
"entropy": 5.830104970932007,
"epoch": 0.599600870827286,
"grad_norm": 2.515625,
"learning_rate": 0.0004970785584030808,
"loss": 5.7138,
"mean_token_accuracy": 0.1525508925318718,
"num_tokens": 13738007.0,
"step": 6610
},
{
"entropy": 5.879718828201294,
"epoch": 0.6000544267053701,
"grad_norm": 2.515625,
"learning_rate": 0.0004970733589182799,
"loss": 5.736,
"mean_token_accuracy": 0.1473026007413864,
"num_tokens": 13749429.0,
"step": 6615
},
{
"entropy": 5.8097755908966064,
"epoch": 0.6005079825834543,
"grad_norm": 3.234375,
"learning_rate": 0.0004970681548409452,
"loss": 5.5973,
"mean_token_accuracy": 0.1569398522377014,
"num_tokens": 13760279.0,
"step": 6620
},
{
"entropy": 5.83681287765503,
"epoch": 0.6009615384615384,
"grad_norm": 2.265625,
"learning_rate": 0.0004970629461711846,
"loss": 5.8122,
"mean_token_accuracy": 0.14485863000154495,
"num_tokens": 13770575.0,
"step": 6625
},
{
"entropy": 5.8300028324127195,
"epoch": 0.6014150943396226,
"grad_norm": 2.4375,
"learning_rate": 0.0004970577329091057,
"loss": 5.6667,
"mean_token_accuracy": 0.15231173038482665,
"num_tokens": 13781068.0,
"step": 6630
},
{
"entropy": 5.8861555576324465,
"epoch": 0.6018686502177069,
"grad_norm": 2.96875,
"learning_rate": 0.0004970525150548162,
"loss": 5.7296,
"mean_token_accuracy": 0.15105716288089752,
"num_tokens": 13790745.0,
"step": 6635
},
{
"entropy": 5.805358266830444,
"epoch": 0.602322206095791,
"grad_norm": 1.7734375,
"learning_rate": 0.0004970472926084242,
"loss": 5.6751,
"mean_token_accuracy": 0.1507732406258583,
"num_tokens": 13800196.0,
"step": 6640
},
{
"entropy": 5.770063829421997,
"epoch": 0.6027757619738752,
"grad_norm": 16.875,
"learning_rate": 0.0004970420655700375,
"loss": 5.5856,
"mean_token_accuracy": 0.15864112824201584,
"num_tokens": 13809954.0,
"step": 6645
},
{
"entropy": 5.792227268218994,
"epoch": 0.6032293178519593,
"grad_norm": 1.9453125,
"learning_rate": 0.0004970368339397643,
"loss": 5.7318,
"mean_token_accuracy": 0.14497790783643721,
"num_tokens": 13820281.0,
"step": 6650
},
{
"entropy": 5.885860538482666,
"epoch": 0.6036828737300436,
"grad_norm": 2.046875,
"learning_rate": 0.0004970315977177129,
"loss": 5.7127,
"mean_token_accuracy": 0.15290192514657974,
"num_tokens": 13830614.0,
"step": 6655
},
{
"entropy": 5.7723931789398195,
"epoch": 0.6041364296081277,
"grad_norm": 8.3125,
"learning_rate": 0.0004970263569039914,
"loss": 5.6539,
"mean_token_accuracy": 0.15475403964519502,
"num_tokens": 13840551.0,
"step": 6660
},
{
"entropy": 5.899419689178467,
"epoch": 0.6045899854862119,
"grad_norm": 2.453125,
"learning_rate": 0.0004970211114987083,
"loss": 5.8363,
"mean_token_accuracy": 0.14742269665002822,
"num_tokens": 13851830.0,
"step": 6665
},
{
"entropy": 5.820240592956543,
"epoch": 0.6050435413642961,
"grad_norm": 4.5625,
"learning_rate": 0.0004970158615019719,
"loss": 5.6443,
"mean_token_accuracy": 0.15448153614997864,
"num_tokens": 13862299.0,
"step": 6670
},
{
"entropy": 5.92971625328064,
"epoch": 0.6054970972423802,
"grad_norm": 3.359375,
"learning_rate": 0.0004970106069138912,
"loss": 5.769,
"mean_token_accuracy": 0.14795244932174684,
"num_tokens": 13872852.0,
"step": 6675
},
{
"entropy": 5.852387952804565,
"epoch": 0.6059506531204645,
"grad_norm": 1.9765625,
"learning_rate": 0.0004970053477345743,
"loss": 5.7018,
"mean_token_accuracy": 0.16219263672828674,
"num_tokens": 13883601.0,
"step": 6680
},
{
"entropy": 5.804964447021485,
"epoch": 0.6064042089985486,
"grad_norm": 2.265625,
"learning_rate": 0.0004970000839641304,
"loss": 5.709,
"mean_token_accuracy": 0.15332725495100022,
"num_tokens": 13894246.0,
"step": 6685
},
{
"entropy": 5.776219987869263,
"epoch": 0.6068577648766328,
"grad_norm": 1.96875,
"learning_rate": 0.0004969948156026681,
"loss": 5.7196,
"mean_token_accuracy": 0.15061549395322799,
"num_tokens": 13904823.0,
"step": 6690
},
{
"entropy": 5.924747323989868,
"epoch": 0.6073113207547169,
"grad_norm": 2.65625,
"learning_rate": 0.0004969895426502965,
"loss": 5.7394,
"mean_token_accuracy": 0.147009214758873,
"num_tokens": 13915338.0,
"step": 6695
},
{
"entropy": 5.8624742984771725,
"epoch": 0.6077648766328012,
"grad_norm": 9.75,
"learning_rate": 0.0004969842651071246,
"loss": 5.6918,
"mean_token_accuracy": 0.15186342597007751,
"num_tokens": 13925838.0,
"step": 6700
},
{
"entropy": 5.8716295719146725,
"epoch": 0.6082184325108854,
"grad_norm": 3.71875,
"learning_rate": 0.0004969789829732615,
"loss": 5.6917,
"mean_token_accuracy": 0.14943995475769042,
"num_tokens": 13936341.0,
"step": 6705
},
{
"entropy": 5.775670528411865,
"epoch": 0.6086719883889695,
"grad_norm": 2.5,
"learning_rate": 0.0004969736962488164,
"loss": 5.7052,
"mean_token_accuracy": 0.15158903300762178,
"num_tokens": 13945438.0,
"step": 6710
},
{
"entropy": 5.7914879322052,
"epoch": 0.6091255442670537,
"grad_norm": 2.59375,
"learning_rate": 0.0004969684049338987,
"loss": 5.6138,
"mean_token_accuracy": 0.1479756087064743,
"num_tokens": 13956029.0,
"step": 6715
},
{
"entropy": 5.867767429351806,
"epoch": 0.6095791001451378,
"grad_norm": 2.21875,
"learning_rate": 0.000496963109028618,
"loss": 5.7231,
"mean_token_accuracy": 0.14818116053938865,
"num_tokens": 13966498.0,
"step": 6720
},
{
"entropy": 5.8269449234008786,
"epoch": 0.6100326560232221,
"grad_norm": 2.921875,
"learning_rate": 0.0004969578085330834,
"loss": 5.6769,
"mean_token_accuracy": 0.149531489610672,
"num_tokens": 13976997.0,
"step": 6725
},
{
"entropy": 5.888552141189575,
"epoch": 0.6104862119013063,
"grad_norm": 2.40625,
"learning_rate": 0.0004969525034474049,
"loss": 5.6642,
"mean_token_accuracy": 0.15069333463907242,
"num_tokens": 13986664.0,
"step": 6730
},
{
"entropy": 5.795527458190918,
"epoch": 0.6109397677793904,
"grad_norm": 2.984375,
"learning_rate": 0.000496947193771692,
"loss": 5.6811,
"mean_token_accuracy": 0.1521475911140442,
"num_tokens": 13997828.0,
"step": 6735
},
{
"entropy": 5.740749216079712,
"epoch": 0.6113933236574746,
"grad_norm": 2.6875,
"learning_rate": 0.0004969418795060548,
"loss": 5.6303,
"mean_token_accuracy": 0.15071731954813003,
"num_tokens": 14008293.0,
"step": 6740
},
{
"entropy": 5.788277959823608,
"epoch": 0.6118468795355588,
"grad_norm": 1.8203125,
"learning_rate": 0.0004969365606506027,
"loss": 5.6121,
"mean_token_accuracy": 0.15342239588499068,
"num_tokens": 14018031.0,
"step": 6745
},
{
"entropy": 5.80244402885437,
"epoch": 0.612300435413643,
"grad_norm": 2.9375,
"learning_rate": 0.0004969312372054461,
"loss": 5.5759,
"mean_token_accuracy": 0.1577866330742836,
"num_tokens": 14028013.0,
"step": 6750
},
{
"entropy": 5.764116668701172,
"epoch": 0.6127539912917271,
"grad_norm": 2.546875,
"learning_rate": 0.0004969259091706948,
"loss": 5.7075,
"mean_token_accuracy": 0.1519751712679863,
"num_tokens": 14039787.0,
"step": 6755
},
{
"entropy": 5.7666487216949465,
"epoch": 0.6132075471698113,
"grad_norm": 2.140625,
"learning_rate": 0.0004969205765464593,
"loss": 5.6029,
"mean_token_accuracy": 0.1544177196919918,
"num_tokens": 14049223.0,
"step": 6760
},
{
"entropy": 5.835097169876098,
"epoch": 0.6136611030478955,
"grad_norm": 4.03125,
"learning_rate": 0.0004969152393328497,
"loss": 5.6536,
"mean_token_accuracy": 0.15525499433279039,
"num_tokens": 14059004.0,
"step": 6765
},
{
"entropy": 5.888670158386231,
"epoch": 0.6141146589259797,
"grad_norm": 2.90625,
"learning_rate": 0.0004969098975299764,
"loss": 5.7625,
"mean_token_accuracy": 0.14184020161628724,
"num_tokens": 14069615.0,
"step": 6770
},
{
"entropy": 5.841464519500732,
"epoch": 0.6145682148040639,
"grad_norm": 3.328125,
"learning_rate": 0.0004969045511379499,
"loss": 5.6905,
"mean_token_accuracy": 0.15196973383426665,
"num_tokens": 14078886.0,
"step": 6775
},
{
"entropy": 5.859313821792602,
"epoch": 0.615021770682148,
"grad_norm": 3.328125,
"learning_rate": 0.0004968992001568806,
"loss": 5.7616,
"mean_token_accuracy": 0.14708426594734192,
"num_tokens": 14089021.0,
"step": 6780
},
{
"entropy": 5.780261850357055,
"epoch": 0.6154753265602322,
"grad_norm": 2.734375,
"learning_rate": 0.0004968938445868795,
"loss": 5.6091,
"mean_token_accuracy": 0.15852924063801765,
"num_tokens": 14098316.0,
"step": 6785
},
{
"entropy": 5.832832527160645,
"epoch": 0.6159288824383164,
"grad_norm": 3.234375,
"learning_rate": 0.0004968884844280569,
"loss": 5.6729,
"mean_token_accuracy": 0.15131434947252273,
"num_tokens": 14108874.0,
"step": 6790
},
{
"entropy": 5.833796405792237,
"epoch": 0.6163824383164006,
"grad_norm": 3.0,
"learning_rate": 0.0004968831196805242,
"loss": 5.7507,
"mean_token_accuracy": 0.15231581628322602,
"num_tokens": 14119613.0,
"step": 6795
},
{
"entropy": 5.878158283233643,
"epoch": 0.6168359941944848,
"grad_norm": 2.015625,
"learning_rate": 0.0004968777503443918,
"loss": 5.6809,
"mean_token_accuracy": 0.15027406364679335,
"num_tokens": 14129085.0,
"step": 6800
},
{
"entropy": 5.774784278869629,
"epoch": 0.6172895500725689,
"grad_norm": 2.609375,
"learning_rate": 0.0004968723764197711,
"loss": 5.5675,
"mean_token_accuracy": 0.16327838152647017,
"num_tokens": 14138461.0,
"step": 6805
},
{
"entropy": 5.768008708953857,
"epoch": 0.6177431059506531,
"grad_norm": 2.875,
"learning_rate": 0.000496866997906773,
"loss": 5.6372,
"mean_token_accuracy": 0.161294561624527,
"num_tokens": 14149126.0,
"step": 6810
},
{
"entropy": 5.774297380447388,
"epoch": 0.6181966618287373,
"grad_norm": 4.25,
"learning_rate": 0.0004968616148055089,
"loss": 5.6585,
"mean_token_accuracy": 0.15676031559705733,
"num_tokens": 14160068.0,
"step": 6815
},
{
"entropy": 5.860025787353516,
"epoch": 0.6186502177068215,
"grad_norm": 1.8984375,
"learning_rate": 0.0004968562271160901,
"loss": 5.7546,
"mean_token_accuracy": 0.1506920039653778,
"num_tokens": 14170288.0,
"step": 6820
},
{
"entropy": 5.866980457305909,
"epoch": 0.6191037735849056,
"grad_norm": 2.09375,
"learning_rate": 0.000496850834838628,
"loss": 5.7489,
"mean_token_accuracy": 0.15387054532766342,
"num_tokens": 14181231.0,
"step": 6825
},
{
"entropy": 5.823310661315918,
"epoch": 0.6195573294629898,
"grad_norm": 2.578125,
"learning_rate": 0.0004968454379732342,
"loss": 5.6475,
"mean_token_accuracy": 0.15675568729639053,
"num_tokens": 14191715.0,
"step": 6830
},
{
"entropy": 5.823959398269653,
"epoch": 0.6200108853410741,
"grad_norm": 2.046875,
"learning_rate": 0.00049684003652002,
"loss": 5.6493,
"mean_token_accuracy": 0.15095681548118592,
"num_tokens": 14203416.0,
"step": 6835
},
{
"entropy": 5.77118558883667,
"epoch": 0.6204644412191582,
"grad_norm": 2.703125,
"learning_rate": 0.0004968346304790973,
"loss": 5.6142,
"mean_token_accuracy": 0.15649014487862586,
"num_tokens": 14213658.0,
"step": 6840
},
{
"entropy": 5.8414740562438965,
"epoch": 0.6209179970972424,
"grad_norm": 2.5,
"learning_rate": 0.0004968292198505781,
"loss": 5.7781,
"mean_token_accuracy": 0.1473475828766823,
"num_tokens": 14223956.0,
"step": 6845
},
{
"entropy": 5.85752501487732,
"epoch": 0.6213715529753265,
"grad_norm": 3.140625,
"learning_rate": 0.000496823804634574,
"loss": 5.7716,
"mean_token_accuracy": 0.1460169270634651,
"num_tokens": 14236801.0,
"step": 6850
},
{
"entropy": 5.755843448638916,
"epoch": 0.6218251088534107,
"grad_norm": 2.875,
"learning_rate": 0.0004968183848311972,
"loss": 5.5906,
"mean_token_accuracy": 0.15509454011917115,
"num_tokens": 14247382.0,
"step": 6855
},
{
"entropy": 5.873498058319091,
"epoch": 0.622278664731495,
"grad_norm": 2.3125,
"learning_rate": 0.0004968129604405596,
"loss": 5.6613,
"mean_token_accuracy": 0.15263956636190415,
"num_tokens": 14258627.0,
"step": 6860
},
{
"entropy": 5.784880971908569,
"epoch": 0.6227322206095791,
"grad_norm": 2.5,
"learning_rate": 0.0004968075314627734,
"loss": 5.6584,
"mean_token_accuracy": 0.15659964829683304,
"num_tokens": 14269013.0,
"step": 6865
},
{
"entropy": 5.7966248989105225,
"epoch": 0.6231857764876633,
"grad_norm": 3.34375,
"learning_rate": 0.000496802097897951,
"loss": 5.6326,
"mean_token_accuracy": 0.15666757971048356,
"num_tokens": 14279620.0,
"step": 6870
},
{
"entropy": 5.860994577407837,
"epoch": 0.6236393323657474,
"grad_norm": 2.046875,
"learning_rate": 0.0004967966597462046,
"loss": 5.6651,
"mean_token_accuracy": 0.16124931573867798,
"num_tokens": 14289448.0,
"step": 6875
},
{
"entropy": 5.885032081604004,
"epoch": 0.6240928882438317,
"grad_norm": 2.21875,
"learning_rate": 0.0004967912170076469,
"loss": 5.6426,
"mean_token_accuracy": 0.15155680626630783,
"num_tokens": 14299999.0,
"step": 6880
},
{
"entropy": 5.80174822807312,
"epoch": 0.6245464441219158,
"grad_norm": 2.296875,
"learning_rate": 0.0004967857696823901,
"loss": 5.6443,
"mean_token_accuracy": 0.14934637099504472,
"num_tokens": 14310384.0,
"step": 6885
},
{
"entropy": 5.7569050788879395,
"epoch": 0.625,
"grad_norm": 4.21875,
"learning_rate": 0.0004967803177705472,
"loss": 5.7437,
"mean_token_accuracy": 0.14541705399751664,
"num_tokens": 14320644.0,
"step": 6890
},
{
"entropy": 5.833973693847656,
"epoch": 0.6254535558780842,
"grad_norm": 4.90625,
"learning_rate": 0.0004967748612722308,
"loss": 5.7003,
"mean_token_accuracy": 0.15331256091594697,
"num_tokens": 14330077.0,
"step": 6895
},
{
"entropy": 5.9259500980377195,
"epoch": 0.6259071117561683,
"grad_norm": 2.296875,
"learning_rate": 0.0004967694001875538,
"loss": 5.8098,
"mean_token_accuracy": 0.14541816860437393,
"num_tokens": 14341549.0,
"step": 6900
},
{
"entropy": 5.868324947357178,
"epoch": 0.6263606676342526,
"grad_norm": 2.984375,
"learning_rate": 0.000496763934516629,
"loss": 5.6829,
"mean_token_accuracy": 0.1533267766237259,
"num_tokens": 14352146.0,
"step": 6905
},
{
"entropy": 5.799675416946411,
"epoch": 0.6268142235123367,
"grad_norm": 3.578125,
"learning_rate": 0.0004967584642595696,
"loss": 5.6725,
"mean_token_accuracy": 0.14867498576641083,
"num_tokens": 14362521.0,
"step": 6910
},
{
"entropy": 5.820127725601196,
"epoch": 0.6272677793904209,
"grad_norm": 4.25,
"learning_rate": 0.0004967529894164887,
"loss": 5.7218,
"mean_token_accuracy": 0.15302577912807463,
"num_tokens": 14373216.0,
"step": 6915
},
{
"entropy": 5.8009356498718265,
"epoch": 0.627721335268505,
"grad_norm": 2.078125,
"learning_rate": 0.0004967475099874992,
"loss": 5.664,
"mean_token_accuracy": 0.1593104839324951,
"num_tokens": 14383104.0,
"step": 6920
},
{
"entropy": 5.81319580078125,
"epoch": 0.6281748911465893,
"grad_norm": 3.1875,
"learning_rate": 0.000496742025972715,
"loss": 5.6722,
"mean_token_accuracy": 0.16146908774971963,
"num_tokens": 14393854.0,
"step": 6925
},
{
"entropy": 5.859877109527588,
"epoch": 0.6286284470246735,
"grad_norm": 2.109375,
"learning_rate": 0.000496736537372249,
"loss": 5.6132,
"mean_token_accuracy": 0.16212746649980544,
"num_tokens": 14403213.0,
"step": 6930
},
{
"entropy": 5.876523447036743,
"epoch": 0.6290820029027576,
"grad_norm": 2.421875,
"learning_rate": 0.0004967310441862151,
"loss": 5.8071,
"mean_token_accuracy": 0.14938885420560838,
"num_tokens": 14412792.0,
"step": 6935
},
{
"entropy": 5.8900281429290775,
"epoch": 0.6295355587808418,
"grad_norm": 2.921875,
"learning_rate": 0.0004967255464147265,
"loss": 5.6745,
"mean_token_accuracy": 0.1568540945649147,
"num_tokens": 14423061.0,
"step": 6940
},
{
"entropy": 5.874559831619263,
"epoch": 0.6299891146589259,
"grad_norm": 2.078125,
"learning_rate": 0.0004967200440578972,
"loss": 5.6251,
"mean_token_accuracy": 0.16184545755386354,
"num_tokens": 14433338.0,
"step": 6945
},
{
"entropy": 5.755834102630615,
"epoch": 0.6304426705370102,
"grad_norm": 2.453125,
"learning_rate": 0.0004967145371158408,
"loss": 5.6117,
"mean_token_accuracy": 0.16235730051994324,
"num_tokens": 14444124.0,
"step": 6950
},
{
"entropy": 5.839078283309936,
"epoch": 0.6308962264150944,
"grad_norm": 2.359375,
"learning_rate": 0.0004967090255886714,
"loss": 5.7494,
"mean_token_accuracy": 0.15068836957216264,
"num_tokens": 14453868.0,
"step": 6955
},
{
"entropy": 5.887261533737183,
"epoch": 0.6313497822931785,
"grad_norm": 3.359375,
"learning_rate": 0.0004967035094765028,
"loss": 5.6452,
"mean_token_accuracy": 0.14719320237636566,
"num_tokens": 14465015.0,
"step": 6960
},
{
"entropy": 5.804389810562133,
"epoch": 0.6318033381712627,
"grad_norm": 2.734375,
"learning_rate": 0.0004966979887794491,
"loss": 5.6197,
"mean_token_accuracy": 0.16230510920286179,
"num_tokens": 14475523.0,
"step": 6965
},
{
"entropy": 5.753247451782227,
"epoch": 0.6322568940493469,
"grad_norm": 2.46875,
"learning_rate": 0.0004966924634976245,
"loss": 5.6644,
"mean_token_accuracy": 0.1508348599076271,
"num_tokens": 14485523.0,
"step": 6970
},
{
"entropy": 5.880261182785034,
"epoch": 0.6327104499274311,
"grad_norm": 2.421875,
"learning_rate": 0.0004966869336311433,
"loss": 5.7506,
"mean_token_accuracy": 0.15484288036823274,
"num_tokens": 14495237.0,
"step": 6975
},
{
"entropy": 5.907916021347046,
"epoch": 0.6331640058055152,
"grad_norm": 3.8125,
"learning_rate": 0.0004966813991801197,
"loss": 5.8437,
"mean_token_accuracy": 0.14812066629529,
"num_tokens": 14506431.0,
"step": 6980
},
{
"entropy": 5.925472545623779,
"epoch": 0.6336175616835994,
"grad_norm": 2.265625,
"learning_rate": 0.0004966758601446684,
"loss": 5.7761,
"mean_token_accuracy": 0.1462002858519554,
"num_tokens": 14517445.0,
"step": 6985
},
{
"entropy": 5.854861927032471,
"epoch": 0.6340711175616836,
"grad_norm": 4.75,
"learning_rate": 0.0004966703165249038,
"loss": 5.6887,
"mean_token_accuracy": 0.14496298283338546,
"num_tokens": 14527818.0,
"step": 6990
},
{
"entropy": 5.769768190383911,
"epoch": 0.6345246734397678,
"grad_norm": 2.390625,
"learning_rate": 0.0004966647683209406,
"loss": 5.7801,
"mean_token_accuracy": 0.14265136122703553,
"num_tokens": 14538235.0,
"step": 6995
},
{
"entropy": 5.903887510299683,
"epoch": 0.634978229317852,
"grad_norm": 2.28125,
"learning_rate": 0.0004966592155328935,
"loss": 5.7874,
"mean_token_accuracy": 0.1466798961162567,
"num_tokens": 14549784.0,
"step": 7000
},
{
"entropy": 5.886717748641968,
"epoch": 0.6354317851959361,
"grad_norm": 3.90625,
"learning_rate": 0.0004966536581608771,
"loss": 5.7825,
"mean_token_accuracy": 0.14561346769332886,
"num_tokens": 14561057.0,
"step": 7005
},
{
"entropy": 5.904143190383911,
"epoch": 0.6358853410740203,
"grad_norm": 2.125,
"learning_rate": 0.0004966480962050069,
"loss": 5.7119,
"mean_token_accuracy": 0.1525163695216179,
"num_tokens": 14570859.0,
"step": 7010
},
{
"entropy": 5.81612000465393,
"epoch": 0.6363388969521045,
"grad_norm": 2.921875,
"learning_rate": 0.0004966425296653974,
"loss": 5.6954,
"mean_token_accuracy": 0.15854416489601136,
"num_tokens": 14581066.0,
"step": 7015
},
{
"entropy": 5.780066299438476,
"epoch": 0.6367924528301887,
"grad_norm": 17.5,
"learning_rate": 0.000496636958542164,
"loss": 5.6691,
"mean_token_accuracy": 0.15080804228782654,
"num_tokens": 14590709.0,
"step": 7020
},
{
"entropy": 5.813197469711303,
"epoch": 0.6372460087082729,
"grad_norm": 2.765625,
"learning_rate": 0.0004966313828354218,
"loss": 5.666,
"mean_token_accuracy": 0.15459553748369217,
"num_tokens": 14600849.0,
"step": 7025
},
{
"entropy": 5.7914598941802975,
"epoch": 0.637699564586357,
"grad_norm": 5.125,
"learning_rate": 0.000496625802545286,
"loss": 5.5547,
"mean_token_accuracy": 0.15562158226966857,
"num_tokens": 14611427.0,
"step": 7030
},
{
"entropy": 5.883208084106445,
"epoch": 0.6381531204644412,
"grad_norm": 3.15625,
"learning_rate": 0.0004966202176718723,
"loss": 5.7263,
"mean_token_accuracy": 0.15225352942943574,
"num_tokens": 14623860.0,
"step": 7035
},
{
"entropy": 5.855080842971802,
"epoch": 0.6386066763425254,
"grad_norm": 4.6875,
"learning_rate": 0.0004966146282152959,
"loss": 5.6323,
"mean_token_accuracy": 0.1527144968509674,
"num_tokens": 14633481.0,
"step": 7040
},
{
"entropy": 5.809521245956421,
"epoch": 0.6390602322206096,
"grad_norm": 2.5,
"learning_rate": 0.0004966090341756726,
"loss": 5.6193,
"mean_token_accuracy": 0.16110151559114455,
"num_tokens": 14643519.0,
"step": 7045
},
{
"entropy": 5.722095584869384,
"epoch": 0.6395137880986937,
"grad_norm": 2.75,
"learning_rate": 0.000496603435553118,
"loss": 5.6306,
"mean_token_accuracy": 0.16035314351320268,
"num_tokens": 14654570.0,
"step": 7050
},
{
"entropy": 5.759281158447266,
"epoch": 0.6399673439767779,
"grad_norm": 3.796875,
"learning_rate": 0.0004965978323477478,
"loss": 5.6755,
"mean_token_accuracy": 0.14542552083730698,
"num_tokens": 14665596.0,
"step": 7055
},
{
"entropy": 5.8258264541625975,
"epoch": 0.6404208998548622,
"grad_norm": 2.78125,
"learning_rate": 0.000496592224559678,
"loss": 5.6142,
"mean_token_accuracy": 0.1545528292655945,
"num_tokens": 14675660.0,
"step": 7060
},
{
"entropy": 5.889820384979248,
"epoch": 0.6408744557329463,
"grad_norm": 3.25,
"learning_rate": 0.0004965866121890245,
"loss": 5.7523,
"mean_token_accuracy": 0.1493598237633705,
"num_tokens": 14685229.0,
"step": 7065
},
{
"entropy": 5.895591783523559,
"epoch": 0.6413280116110305,
"grad_norm": 2.484375,
"learning_rate": 0.0004965809952359034,
"loss": 5.6761,
"mean_token_accuracy": 0.1506911963224411,
"num_tokens": 14695672.0,
"step": 7070
},
{
"entropy": 5.819560384750366,
"epoch": 0.6417815674891146,
"grad_norm": 2.375,
"learning_rate": 0.0004965753737004309,
"loss": 5.6134,
"mean_token_accuracy": 0.15645401328802108,
"num_tokens": 14705160.0,
"step": 7075
},
{
"entropy": 5.791318702697754,
"epoch": 0.6422351233671988,
"grad_norm": 2.421875,
"learning_rate": 0.0004965697475827231,
"loss": 5.6414,
"mean_token_accuracy": 0.1544739320874214,
"num_tokens": 14715487.0,
"step": 7080
},
{
"entropy": 5.8290979862213135,
"epoch": 0.6426886792452831,
"grad_norm": 3.609375,
"learning_rate": 0.0004965641168828965,
"loss": 5.685,
"mean_token_accuracy": 0.15276286453008653,
"num_tokens": 14726987.0,
"step": 7085
},
{
"entropy": 5.785283803939819,
"epoch": 0.6431422351233672,
"grad_norm": 3.40625,
"learning_rate": 0.0004965584816010676,
"loss": 5.6615,
"mean_token_accuracy": 0.15806744545698165,
"num_tokens": 14737810.0,
"step": 7090
},
{
"entropy": 5.8629076957702635,
"epoch": 0.6435957910014514,
"grad_norm": 2.65625,
"learning_rate": 0.0004965528417373528,
"loss": 5.6796,
"mean_token_accuracy": 0.1506367415189743,
"num_tokens": 14748997.0,
"step": 7095
},
{
"entropy": 5.868148517608643,
"epoch": 0.6440493468795355,
"grad_norm": 2.203125,
"learning_rate": 0.0004965471972918688,
"loss": 5.6911,
"mean_token_accuracy": 0.15026735216379167,
"num_tokens": 14760439.0,
"step": 7100
},
{
"entropy": 5.735028457641602,
"epoch": 0.6445029027576198,
"grad_norm": 2.625,
"learning_rate": 0.0004965415482647322,
"loss": 5.6046,
"mean_token_accuracy": 0.15349232405424118,
"num_tokens": 14770828.0,
"step": 7105
},
{
"entropy": 5.755033874511719,
"epoch": 0.6449564586357039,
"grad_norm": 2.328125,
"learning_rate": 0.00049653589465606,
"loss": 5.7486,
"mean_token_accuracy": 0.15289024412631988,
"num_tokens": 14781753.0,
"step": 7110
},
{
"entropy": 5.863271188735962,
"epoch": 0.6454100145137881,
"grad_norm": 2.234375,
"learning_rate": 0.000496530236465969,
"loss": 5.7505,
"mean_token_accuracy": 0.14865993708372116,
"num_tokens": 14792241.0,
"step": 7115
},
{
"entropy": 5.7898180961608885,
"epoch": 0.6458635703918723,
"grad_norm": 2.8125,
"learning_rate": 0.0004965245736945763,
"loss": 5.5854,
"mean_token_accuracy": 0.1545030027627945,
"num_tokens": 14802580.0,
"step": 7120
},
{
"entropy": 5.820276069641113,
"epoch": 0.6463171262699564,
"grad_norm": 2.046875,
"learning_rate": 0.0004965189063419989,
"loss": 5.7744,
"mean_token_accuracy": 0.151085764169693,
"num_tokens": 14813103.0,
"step": 7125
},
{
"entropy": 5.845597743988037,
"epoch": 0.6467706821480407,
"grad_norm": 4.96875,
"learning_rate": 0.0004965132344083541,
"loss": 5.6569,
"mean_token_accuracy": 0.15386222600936889,
"num_tokens": 14823334.0,
"step": 7130
},
{
"entropy": 5.861242008209229,
"epoch": 0.6472242380261248,
"grad_norm": 2.109375,
"learning_rate": 0.0004965075578937593,
"loss": 5.6942,
"mean_token_accuracy": 0.14701148271560668,
"num_tokens": 14834850.0,
"step": 7135
},
{
"entropy": 5.782326364517212,
"epoch": 0.647677793904209,
"grad_norm": 3.015625,
"learning_rate": 0.0004965018767983316,
"loss": 5.7155,
"mean_token_accuracy": 0.1517026126384735,
"num_tokens": 14844276.0,
"step": 7140
},
{
"entropy": 5.726510858535766,
"epoch": 0.6481313497822931,
"grad_norm": 2.21875,
"learning_rate": 0.0004964961911221886,
"loss": 5.5679,
"mean_token_accuracy": 0.16671131700277328,
"num_tokens": 14854352.0,
"step": 7145
},
{
"entropy": 5.780787515640259,
"epoch": 0.6485849056603774,
"grad_norm": 2.609375,
"learning_rate": 0.000496490500865448,
"loss": 5.7241,
"mean_token_accuracy": 0.15255769789218904,
"num_tokens": 14864395.0,
"step": 7150
},
{
"entropy": 5.805046892166137,
"epoch": 0.6490384615384616,
"grad_norm": 3.375,
"learning_rate": 0.0004964848060282275,
"loss": 5.6223,
"mean_token_accuracy": 0.15498533099889755,
"num_tokens": 14874373.0,
"step": 7155
},
{
"entropy": 5.788227224349976,
"epoch": 0.6494920174165457,
"grad_norm": 2.046875,
"learning_rate": 0.0004964791066106447,
"loss": 5.6011,
"mean_token_accuracy": 0.1550254613161087,
"num_tokens": 14884932.0,
"step": 7160
},
{
"entropy": 5.826607799530029,
"epoch": 0.6499455732946299,
"grad_norm": 2.515625,
"learning_rate": 0.0004964734026128175,
"loss": 5.7555,
"mean_token_accuracy": 0.14865583181381226,
"num_tokens": 14896254.0,
"step": 7165
},
{
"entropy": 5.9093794345855715,
"epoch": 0.650399129172714,
"grad_norm": 2.265625,
"learning_rate": 0.0004964676940348638,
"loss": 5.6961,
"mean_token_accuracy": 0.1514290004968643,
"num_tokens": 14907392.0,
"step": 7170
},
{
"entropy": 5.843709802627563,
"epoch": 0.6508526850507983,
"grad_norm": 2.40625,
"learning_rate": 0.000496461980876902,
"loss": 5.8001,
"mean_token_accuracy": 0.14915761500597,
"num_tokens": 14917676.0,
"step": 7175
},
{
"entropy": 5.839835214614868,
"epoch": 0.6513062409288825,
"grad_norm": 3.140625,
"learning_rate": 0.0004964562631390498,
"loss": 5.6594,
"mean_token_accuracy": 0.1541726604104042,
"num_tokens": 14928500.0,
"step": 7180
},
{
"entropy": 5.849341201782226,
"epoch": 0.6517597968069666,
"grad_norm": 2.265625,
"learning_rate": 0.0004964505408214257,
"loss": 5.6167,
"mean_token_accuracy": 0.15815220177173614,
"num_tokens": 14938970.0,
"step": 7185
},
{
"entropy": 5.796083402633667,
"epoch": 0.6522133526850508,
"grad_norm": 2.515625,
"learning_rate": 0.000496444813924148,
"loss": 5.6739,
"mean_token_accuracy": 0.15429000556468964,
"num_tokens": 14949017.0,
"step": 7190
},
{
"entropy": 5.80718240737915,
"epoch": 0.652666908563135,
"grad_norm": 2.53125,
"learning_rate": 0.0004964390824473351,
"loss": 5.6101,
"mean_token_accuracy": 0.1598324030637741,
"num_tokens": 14958529.0,
"step": 7195
},
{
"entropy": 5.796051073074341,
"epoch": 0.6531204644412192,
"grad_norm": 2.734375,
"learning_rate": 0.0004964333463911056,
"loss": 5.6836,
"mean_token_accuracy": 0.15045272409915925,
"num_tokens": 14968044.0,
"step": 7200
},
{
"entropy": 5.83974404335022,
"epoch": 0.6535740203193033,
"grad_norm": 3.359375,
"learning_rate": 0.0004964276057555779,
"loss": 5.7166,
"mean_token_accuracy": 0.1514724761247635,
"num_tokens": 14977867.0,
"step": 7205
},
{
"entropy": 5.976543140411377,
"epoch": 0.6540275761973875,
"grad_norm": 2.109375,
"learning_rate": 0.000496421860540871,
"loss": 5.7153,
"mean_token_accuracy": 0.15487430840730668,
"num_tokens": 14986936.0,
"step": 7210
},
{
"entropy": 5.7960106372833256,
"epoch": 0.6544811320754716,
"grad_norm": 3.015625,
"learning_rate": 0.0004964161107471036,
"loss": 5.6711,
"mean_token_accuracy": 0.1544204205274582,
"num_tokens": 14996681.0,
"step": 7215
},
{
"entropy": 5.7676044464111325,
"epoch": 0.6549346879535559,
"grad_norm": 2.09375,
"learning_rate": 0.0004964103563743946,
"loss": 5.682,
"mean_token_accuracy": 0.15293529480695725,
"num_tokens": 15007100.0,
"step": 7220
},
{
"entropy": 5.930917358398437,
"epoch": 0.6553882438316401,
"grad_norm": 1.875,
"learning_rate": 0.000496404597422863,
"loss": 5.7601,
"mean_token_accuracy": 0.15238583236932754,
"num_tokens": 15017452.0,
"step": 7225
},
{
"entropy": 5.838077640533447,
"epoch": 0.6558417997097242,
"grad_norm": 2.640625,
"learning_rate": 0.000496398833892628,
"loss": 5.6059,
"mean_token_accuracy": 0.15244304686784743,
"num_tokens": 15028362.0,
"step": 7230
},
{
"entropy": 5.829388523101807,
"epoch": 0.6562953555878084,
"grad_norm": 2.234375,
"learning_rate": 0.0004963930657838086,
"loss": 5.7011,
"mean_token_accuracy": 0.14786461889743804,
"num_tokens": 15038691.0,
"step": 7235
},
{
"entropy": 5.791590785980224,
"epoch": 0.6567489114658926,
"grad_norm": 2.34375,
"learning_rate": 0.0004963872930965242,
"loss": 5.6779,
"mean_token_accuracy": 0.15659743323922157,
"num_tokens": 15050037.0,
"step": 7240
},
{
"entropy": 5.781030893325806,
"epoch": 0.6572024673439768,
"grad_norm": 3.15625,
"learning_rate": 0.0004963815158308941,
"loss": 5.63,
"mean_token_accuracy": 0.15438777953386307,
"num_tokens": 15059793.0,
"step": 7245
},
{
"entropy": 5.81955099105835,
"epoch": 0.657656023222061,
"grad_norm": 2.03125,
"learning_rate": 0.0004963757339870379,
"loss": 5.7275,
"mean_token_accuracy": 0.149324406683445,
"num_tokens": 15070500.0,
"step": 7250
},
{
"entropy": 5.926506376266479,
"epoch": 0.6581095791001451,
"grad_norm": 3.6875,
"learning_rate": 0.0004963699475650751,
"loss": 5.7868,
"mean_token_accuracy": 0.14556344151496886,
"num_tokens": 15081109.0,
"step": 7255
},
{
"entropy": 5.816445398330688,
"epoch": 0.6585631349782293,
"grad_norm": 2.40625,
"learning_rate": 0.0004963641565651255,
"loss": 5.6792,
"mean_token_accuracy": 0.15753332376480103,
"num_tokens": 15092315.0,
"step": 7260
},
{
"entropy": 5.796482992172241,
"epoch": 0.6590166908563135,
"grad_norm": 2.5,
"learning_rate": 0.0004963583609873084,
"loss": 5.7455,
"mean_token_accuracy": 0.14933345466852188,
"num_tokens": 15103957.0,
"step": 7265
},
{
"entropy": 5.874239158630371,
"epoch": 0.6594702467343977,
"grad_norm": 2.421875,
"learning_rate": 0.0004963525608317443,
"loss": 5.7189,
"mean_token_accuracy": 0.15017946064472198,
"num_tokens": 15114996.0,
"step": 7270
},
{
"entropy": 5.8762486457824705,
"epoch": 0.6599238026124818,
"grad_norm": 2.25,
"learning_rate": 0.0004963467560985527,
"loss": 5.6769,
"mean_token_accuracy": 0.14785903841257095,
"num_tokens": 15125098.0,
"step": 7275
},
{
"entropy": 5.787175512313842,
"epoch": 0.660377358490566,
"grad_norm": 2.421875,
"learning_rate": 0.0004963409467878538,
"loss": 5.6366,
"mean_token_accuracy": 0.1504684343934059,
"num_tokens": 15134746.0,
"step": 7280
},
{
"entropy": 5.808853626251221,
"epoch": 0.6608309143686503,
"grad_norm": 3.21875,
"learning_rate": 0.0004963351328997678,
"loss": 5.6695,
"mean_token_accuracy": 0.1455580174922943,
"num_tokens": 15145698.0,
"step": 7285
},
{
"entropy": 5.806615495681763,
"epoch": 0.6612844702467344,
"grad_norm": 6.875,
"learning_rate": 0.0004963293144344148,
"loss": 5.7427,
"mean_token_accuracy": 0.14489729702472687,
"num_tokens": 15156026.0,
"step": 7290
},
{
"entropy": 5.85854206085205,
"epoch": 0.6617380261248186,
"grad_norm": 2.90625,
"learning_rate": 0.0004963234913919151,
"loss": 5.6739,
"mean_token_accuracy": 0.1468849129974842,
"num_tokens": 15165900.0,
"step": 7295
},
{
"entropy": 5.836259555816651,
"epoch": 0.6621915820029027,
"grad_norm": 2.671875,
"learning_rate": 0.0004963176637723893,
"loss": 5.7369,
"mean_token_accuracy": 0.14796246439218522,
"num_tokens": 15176693.0,
"step": 7300
},
{
"entropy": 5.850326919555664,
"epoch": 0.6626451378809869,
"grad_norm": 2.890625,
"learning_rate": 0.0004963118315759578,
"loss": 5.7717,
"mean_token_accuracy": 0.14276785105466844,
"num_tokens": 15186535.0,
"step": 7305
},
{
"entropy": 5.898229312896729,
"epoch": 0.6630986937590712,
"grad_norm": 2.234375,
"learning_rate": 0.0004963059948027411,
"loss": 5.6684,
"mean_token_accuracy": 0.14483277201652528,
"num_tokens": 15196019.0,
"step": 7310
},
{
"entropy": 5.776585054397583,
"epoch": 0.6635522496371553,
"grad_norm": 3.703125,
"learning_rate": 0.0004963001534528601,
"loss": 5.5405,
"mean_token_accuracy": 0.1623222604393959,
"num_tokens": 15205273.0,
"step": 7315
},
{
"entropy": 5.7897053241729735,
"epoch": 0.6640058055152395,
"grad_norm": 2.578125,
"learning_rate": 0.0004962943075264355,
"loss": 5.7013,
"mean_token_accuracy": 0.1488634929060936,
"num_tokens": 15216716.0,
"step": 7320
},
{
"entropy": 5.807294750213623,
"epoch": 0.6644593613933236,
"grad_norm": 4.40625,
"learning_rate": 0.0004962884570235884,
"loss": 5.7236,
"mean_token_accuracy": 0.15068677812814713,
"num_tokens": 15227306.0,
"step": 7325
},
{
"entropy": 5.742990303039551,
"epoch": 0.6649129172714079,
"grad_norm": 3.671875,
"learning_rate": 0.0004962826019444394,
"loss": 5.599,
"mean_token_accuracy": 0.1591808944940567,
"num_tokens": 15237302.0,
"step": 7330
},
{
"entropy": 5.827797937393188,
"epoch": 0.665366473149492,
"grad_norm": 2.25,
"learning_rate": 0.0004962767422891099,
"loss": 5.6475,
"mean_token_accuracy": 0.15172240287065505,
"num_tokens": 15247064.0,
"step": 7335
},
{
"entropy": 5.857191896438598,
"epoch": 0.6658200290275762,
"grad_norm": 3.078125,
"learning_rate": 0.000496270878057721,
"loss": 5.7054,
"mean_token_accuracy": 0.15546690672636032,
"num_tokens": 15257006.0,
"step": 7340
},
{
"entropy": 5.8285126209259035,
"epoch": 0.6662735849056604,
"grad_norm": 3.515625,
"learning_rate": 0.0004962650092503939,
"loss": 5.683,
"mean_token_accuracy": 0.15548086911439896,
"num_tokens": 15266720.0,
"step": 7345
},
{
"entropy": 5.79559907913208,
"epoch": 0.6667271407837445,
"grad_norm": 2.3125,
"learning_rate": 0.0004962591358672502,
"loss": 5.6971,
"mean_token_accuracy": 0.150240645557642,
"num_tokens": 15277330.0,
"step": 7350
},
{
"entropy": 5.8351600646972654,
"epoch": 0.6671806966618288,
"grad_norm": 7.84375,
"learning_rate": 0.000496253257908411,
"loss": 5.6504,
"mean_token_accuracy": 0.15267718434333802,
"num_tokens": 15287401.0,
"step": 7355
},
{
"entropy": 5.7537617683410645,
"epoch": 0.6676342525399129,
"grad_norm": 3.328125,
"learning_rate": 0.0004962473753739981,
"loss": 5.5934,
"mean_token_accuracy": 0.1605018511414528,
"num_tokens": 15297426.0,
"step": 7360
},
{
"entropy": 5.8254279613494875,
"epoch": 0.6680878084179971,
"grad_norm": 2.578125,
"learning_rate": 0.000496241488264133,
"loss": 5.5919,
"mean_token_accuracy": 0.15262950733304023,
"num_tokens": 15308412.0,
"step": 7365
},
{
"entropy": 5.769833183288574,
"epoch": 0.6685413642960812,
"grad_norm": 3.3125,
"learning_rate": 0.0004962355965789377,
"loss": 5.6122,
"mean_token_accuracy": 0.14959425926208497,
"num_tokens": 15318821.0,
"step": 7370
},
{
"entropy": 5.800112342834472,
"epoch": 0.6689949201741655,
"grad_norm": 2.484375,
"learning_rate": 0.0004962297003185338,
"loss": 5.6561,
"mean_token_accuracy": 0.1527245745062828,
"num_tokens": 15328737.0,
"step": 7375
},
{
"entropy": 5.918911552429199,
"epoch": 0.6694484760522497,
"grad_norm": 5.28125,
"learning_rate": 0.0004962237994830433,
"loss": 5.7712,
"mean_token_accuracy": 0.14499044567346572,
"num_tokens": 15339695.0,
"step": 7380
},
{
"entropy": 5.899818372726441,
"epoch": 0.6699020319303338,
"grad_norm": 1.9609375,
"learning_rate": 0.0004962178940725883,
"loss": 5.7177,
"mean_token_accuracy": 0.14840016067028045,
"num_tokens": 15350270.0,
"step": 7385
},
{
"entropy": 5.827748537063599,
"epoch": 0.670355587808418,
"grad_norm": 2.796875,
"learning_rate": 0.0004962119840872908,
"loss": 5.7215,
"mean_token_accuracy": 0.15222820043563842,
"num_tokens": 15360986.0,
"step": 7390
},
{
"entropy": 5.822416067123413,
"epoch": 0.6708091436865021,
"grad_norm": 2.078125,
"learning_rate": 0.0004962060695272733,
"loss": 5.7284,
"mean_token_accuracy": 0.14872562736272812,
"num_tokens": 15371786.0,
"step": 7395
},
{
"entropy": 5.847713041305542,
"epoch": 0.6712626995645864,
"grad_norm": 2.671875,
"learning_rate": 0.0004962001503926577,
"loss": 5.664,
"mean_token_accuracy": 0.15866391956806183,
"num_tokens": 15382960.0,
"step": 7400
},
{
"entropy": 5.81244649887085,
"epoch": 0.6717162554426706,
"grad_norm": 2.765625,
"learning_rate": 0.0004961942266835667,
"loss": 5.6593,
"mean_token_accuracy": 0.15917277187108994,
"num_tokens": 15392807.0,
"step": 7405
},
{
"entropy": 5.838684129714966,
"epoch": 0.6721698113207547,
"grad_norm": 2.65625,
"learning_rate": 0.0004961882984001227,
"loss": 5.7278,
"mean_token_accuracy": 0.14638093262910842,
"num_tokens": 15403712.0,
"step": 7410
},
{
"entropy": 5.842763519287109,
"epoch": 0.6726233671988389,
"grad_norm": 2.59375,
"learning_rate": 0.0004961823655424484,
"loss": 5.7594,
"mean_token_accuracy": 0.14760402739048004,
"num_tokens": 15414839.0,
"step": 7415
},
{
"entropy": 5.85579662322998,
"epoch": 0.6730769230769231,
"grad_norm": 3.375,
"learning_rate": 0.0004961764281106664,
"loss": 5.6555,
"mean_token_accuracy": 0.15158698484301567,
"num_tokens": 15425320.0,
"step": 7420
},
{
"entropy": 5.880425500869751,
"epoch": 0.6735304789550073,
"grad_norm": 1.984375,
"learning_rate": 0.0004961704861048995,
"loss": 5.6446,
"mean_token_accuracy": 0.15110456049442292,
"num_tokens": 15435476.0,
"step": 7425
},
{
"entropy": 5.793002033233643,
"epoch": 0.6739840348330914,
"grad_norm": 5.34375,
"learning_rate": 0.0004961645395252705,
"loss": 5.7385,
"mean_token_accuracy": 0.14751248359680175,
"num_tokens": 15445786.0,
"step": 7430
},
{
"entropy": 5.762565755844117,
"epoch": 0.6744375907111756,
"grad_norm": 2.78125,
"learning_rate": 0.0004961585883719025,
"loss": 5.7483,
"mean_token_accuracy": 0.152704431861639,
"num_tokens": 15455915.0,
"step": 7435
},
{
"entropy": 5.846822690963745,
"epoch": 0.6748911465892597,
"grad_norm": 2.828125,
"learning_rate": 0.0004961526326449186,
"loss": 5.7176,
"mean_token_accuracy": 0.15189335495233536,
"num_tokens": 15466523.0,
"step": 7440
},
{
"entropy": 5.879119920730591,
"epoch": 0.675344702467344,
"grad_norm": 2.171875,
"learning_rate": 0.0004961466723444419,
"loss": 5.7177,
"mean_token_accuracy": 0.1526683509349823,
"num_tokens": 15476821.0,
"step": 7445
},
{
"entropy": 5.80903730392456,
"epoch": 0.6757982583454282,
"grad_norm": 2.296875,
"learning_rate": 0.0004961407074705955,
"loss": 5.6795,
"mean_token_accuracy": 0.15011930987238883,
"num_tokens": 15487182.0,
"step": 7450
},
{
"entropy": 5.812462520599365,
"epoch": 0.6762518142235123,
"grad_norm": 3.578125,
"learning_rate": 0.0004961347380235031,
"loss": 5.6465,
"mean_token_accuracy": 0.15409299433231355,
"num_tokens": 15497600.0,
"step": 7455
},
{
"entropy": 5.885445642471313,
"epoch": 0.6767053701015965,
"grad_norm": 2.796875,
"learning_rate": 0.0004961287640032879,
"loss": 5.7799,
"mean_token_accuracy": 0.14085702300071717,
"num_tokens": 15508278.0,
"step": 7460
},
{
"entropy": 5.931916284561157,
"epoch": 0.6771589259796807,
"grad_norm": 2.828125,
"learning_rate": 0.0004961227854100734,
"loss": 5.8092,
"mean_token_accuracy": 0.14859505146741867,
"num_tokens": 15518458.0,
"step": 7465
},
{
"entropy": 5.752091789245606,
"epoch": 0.6776124818577649,
"grad_norm": 3.03125,
"learning_rate": 0.0004961168022439835,
"loss": 5.6075,
"mean_token_accuracy": 0.15934648662805556,
"num_tokens": 15528738.0,
"step": 7470
},
{
"entropy": 5.784538269042969,
"epoch": 0.6780660377358491,
"grad_norm": 2.46875,
"learning_rate": 0.0004961108145051417,
"loss": 5.6542,
"mean_token_accuracy": 0.15490319281816484,
"num_tokens": 15538656.0,
"step": 7475
},
{
"entropy": 5.7982035160064695,
"epoch": 0.6785195936139332,
"grad_norm": 3.515625,
"learning_rate": 0.000496104822193672,
"loss": 5.6647,
"mean_token_accuracy": 0.1550088033080101,
"num_tokens": 15549285.0,
"step": 7480
},
{
"entropy": 5.904118776321411,
"epoch": 0.6789731494920174,
"grad_norm": 2.015625,
"learning_rate": 0.0004960988253096982,
"loss": 5.7249,
"mean_token_accuracy": 0.15838196128606796,
"num_tokens": 15559766.0,
"step": 7485
},
{
"entropy": 5.829934310913086,
"epoch": 0.6794267053701016,
"grad_norm": 1.6875,
"learning_rate": 0.0004960928238533443,
"loss": 5.6062,
"mean_token_accuracy": 0.15969916731119155,
"num_tokens": 15570873.0,
"step": 7490
},
{
"entropy": 5.828317451477051,
"epoch": 0.6798802612481858,
"grad_norm": 2.09375,
"learning_rate": 0.0004960868178247344,
"loss": 5.8395,
"mean_token_accuracy": 0.1505518451333046,
"num_tokens": 15581229.0,
"step": 7495
},
{
"entropy": 5.747046995162964,
"epoch": 0.68033381712627,
"grad_norm": 2.03125,
"learning_rate": 0.0004960808072239929,
"loss": 5.7085,
"mean_token_accuracy": 0.1462366871535778,
"num_tokens": 15591686.0,
"step": 7500
},
{
"entropy": 5.912446784973144,
"epoch": 0.6807873730043541,
"grad_norm": 2.3125,
"learning_rate": 0.0004960747920512439,
"loss": 5.7215,
"mean_token_accuracy": 0.1496862567961216,
"num_tokens": 15601869.0,
"step": 7505
},
{
"entropy": 5.829827451705933,
"epoch": 0.6812409288824384,
"grad_norm": 2.234375,
"learning_rate": 0.000496068772306612,
"loss": 5.7228,
"mean_token_accuracy": 0.14869892299175264,
"num_tokens": 15612946.0,
"step": 7510
},
{
"entropy": 5.813806962966919,
"epoch": 0.6816944847605225,
"grad_norm": 3.984375,
"learning_rate": 0.0004960627479902214,
"loss": 5.7616,
"mean_token_accuracy": 0.14657169133424758,
"num_tokens": 15623761.0,
"step": 7515
},
{
"entropy": 5.797790241241455,
"epoch": 0.6821480406386067,
"grad_norm": 3.453125,
"learning_rate": 0.000496056719102197,
"loss": 5.7404,
"mean_token_accuracy": 0.14438045471906663,
"num_tokens": 15634500.0,
"step": 7520
},
{
"entropy": 5.746096229553222,
"epoch": 0.6826015965166908,
"grad_norm": 2.515625,
"learning_rate": 0.0004960506856426631,
"loss": 5.6124,
"mean_token_accuracy": 0.149461767077446,
"num_tokens": 15644896.0,
"step": 7525
},
{
"entropy": 5.8980224609375,
"epoch": 0.683055152394775,
"grad_norm": 2.28125,
"learning_rate": 0.0004960446476117448,
"loss": 5.7132,
"mean_token_accuracy": 0.15735317766666412,
"num_tokens": 15655008.0,
"step": 7530
},
{
"entropy": 5.8324168682098385,
"epoch": 0.6835087082728593,
"grad_norm": 3.03125,
"learning_rate": 0.0004960386050095668,
"loss": 5.6322,
"mean_token_accuracy": 0.14784073680639268,
"num_tokens": 15665944.0,
"step": 7535
},
{
"entropy": 5.824177026748657,
"epoch": 0.6839622641509434,
"grad_norm": 2.4375,
"learning_rate": 0.0004960325578362543,
"loss": 5.7564,
"mean_token_accuracy": 0.15302710086107255,
"num_tokens": 15677099.0,
"step": 7540
},
{
"entropy": 5.838755798339844,
"epoch": 0.6844158200290276,
"grad_norm": 3.109375,
"learning_rate": 0.000496026506091932,
"loss": 5.6928,
"mean_token_accuracy": 0.1540417954325676,
"num_tokens": 15687531.0,
"step": 7545
},
{
"entropy": 5.907812118530273,
"epoch": 0.6848693759071117,
"grad_norm": 2.046875,
"learning_rate": 0.0004960204497767253,
"loss": 5.7004,
"mean_token_accuracy": 0.14849745184183122,
"num_tokens": 15697002.0,
"step": 7550
},
{
"entropy": 5.784596347808838,
"epoch": 0.685322931785196,
"grad_norm": 2.921875,
"learning_rate": 0.0004960143888907594,
"loss": 5.6055,
"mean_token_accuracy": 0.14923751652240752,
"num_tokens": 15707249.0,
"step": 7555
},
{
"entropy": 5.801516056060791,
"epoch": 0.6857764876632801,
"grad_norm": 2.734375,
"learning_rate": 0.0004960083234341595,
"loss": 5.6647,
"mean_token_accuracy": 0.15030739307403565,
"num_tokens": 15718764.0,
"step": 7560
},
{
"entropy": 5.818806838989258,
"epoch": 0.6862300435413643,
"grad_norm": 2.296875,
"learning_rate": 0.0004960022534070513,
"loss": 5.6246,
"mean_token_accuracy": 0.15101515799760817,
"num_tokens": 15728854.0,
"step": 7565
},
{
"entropy": 5.813277912139893,
"epoch": 0.6866835994194485,
"grad_norm": 2.125,
"learning_rate": 0.00049599617880956,
"loss": 5.6366,
"mean_token_accuracy": 0.15327175408601762,
"num_tokens": 15739682.0,
"step": 7570
},
{
"entropy": 5.815754461288452,
"epoch": 0.6871371552975326,
"grad_norm": 2.28125,
"learning_rate": 0.0004959900996418115,
"loss": 5.6278,
"mean_token_accuracy": 0.16303820610046388,
"num_tokens": 15749628.0,
"step": 7575
},
{
"entropy": 5.768430137634278,
"epoch": 0.6875907111756169,
"grad_norm": 2.140625,
"learning_rate": 0.0004959840159039315,
"loss": 5.7055,
"mean_token_accuracy": 0.15597921907901763,
"num_tokens": 15760137.0,
"step": 7580
},
{
"entropy": 5.7924089431762695,
"epoch": 0.688044267053701,
"grad_norm": 2.390625,
"learning_rate": 0.0004959779275960456,
"loss": 5.6656,
"mean_token_accuracy": 0.150012369453907,
"num_tokens": 15769889.0,
"step": 7585
},
{
"entropy": 5.846169757843017,
"epoch": 0.6884978229317852,
"grad_norm": 5.4375,
"learning_rate": 0.00049597183471828,
"loss": 5.7515,
"mean_token_accuracy": 0.15068214386701584,
"num_tokens": 15780912.0,
"step": 7590
},
{
"entropy": 5.89394326210022,
"epoch": 0.6889513788098693,
"grad_norm": 3.4375,
"learning_rate": 0.0004959657372707604,
"loss": 5.634,
"mean_token_accuracy": 0.15749605894088745,
"num_tokens": 15791719.0,
"step": 7595
},
{
"entropy": 5.874384260177612,
"epoch": 0.6894049346879536,
"grad_norm": 4.96875,
"learning_rate": 0.0004959596352536132,
"loss": 5.6705,
"mean_token_accuracy": 0.15589183270931245,
"num_tokens": 15802551.0,
"step": 7600
},
{
"entropy": 5.864084911346436,
"epoch": 0.6898584905660378,
"grad_norm": 3.53125,
"learning_rate": 0.0004959535286669643,
"loss": 5.7224,
"mean_token_accuracy": 0.15257749259471892,
"num_tokens": 15813551.0,
"step": 7605
},
{
"entropy": 5.830462169647217,
"epoch": 0.6903120464441219,
"grad_norm": 2.671875,
"learning_rate": 0.0004959474175109402,
"loss": 5.6989,
"mean_token_accuracy": 0.15375184565782546,
"num_tokens": 15824262.0,
"step": 7610
},
{
"entropy": 5.864915943145752,
"epoch": 0.6907656023222061,
"grad_norm": 2.65625,
"learning_rate": 0.0004959413017856672,
"loss": 5.7468,
"mean_token_accuracy": 0.1529516577720642,
"num_tokens": 15835493.0,
"step": 7615
},
{
"entropy": 5.750688648223877,
"epoch": 0.6912191582002902,
"grad_norm": 3.40625,
"learning_rate": 0.0004959351814912716,
"loss": 5.6216,
"mean_token_accuracy": 0.15190897583961488,
"num_tokens": 15846224.0,
"step": 7620
},
{
"entropy": 5.812427377700805,
"epoch": 0.6916727140783745,
"grad_norm": 2.1875,
"learning_rate": 0.0004959290566278803,
"loss": 5.6666,
"mean_token_accuracy": 0.15235227942466736,
"num_tokens": 15855622.0,
"step": 7625
},
{
"entropy": 5.867595767974853,
"epoch": 0.6921262699564587,
"grad_norm": 2.25,
"learning_rate": 0.0004959229271956199,
"loss": 5.7211,
"mean_token_accuracy": 0.15340991020202638,
"num_tokens": 15865835.0,
"step": 7630
},
{
"entropy": 5.828436231613159,
"epoch": 0.6925798258345428,
"grad_norm": 3.453125,
"learning_rate": 0.0004959167931946169,
"loss": 5.6652,
"mean_token_accuracy": 0.14981707483530043,
"num_tokens": 15875915.0,
"step": 7635
},
{
"entropy": 5.883977460861206,
"epoch": 0.693033381712627,
"grad_norm": 2.953125,
"learning_rate": 0.0004959106546249984,
"loss": 5.6936,
"mean_token_accuracy": 0.1512582406401634,
"num_tokens": 15887111.0,
"step": 7640
},
{
"entropy": 5.899054718017578,
"epoch": 0.6934869375907112,
"grad_norm": 2.40625,
"learning_rate": 0.0004959045114868913,
"loss": 5.8056,
"mean_token_accuracy": 0.13988058194518088,
"num_tokens": 15899528.0,
"step": 7645
},
{
"entropy": 5.864440107345581,
"epoch": 0.6939404934687954,
"grad_norm": 2.765625,
"learning_rate": 0.0004958983637804225,
"loss": 5.696,
"mean_token_accuracy": 0.15691021382808684,
"num_tokens": 15910881.0,
"step": 7650
},
{
"entropy": 5.844481897354126,
"epoch": 0.6943940493468795,
"grad_norm": 3.03125,
"learning_rate": 0.0004958922115057194,
"loss": 5.7601,
"mean_token_accuracy": 0.14802518784999846,
"num_tokens": 15921189.0,
"step": 7655
},
{
"entropy": 5.890579032897949,
"epoch": 0.6948476052249637,
"grad_norm": 2.015625,
"learning_rate": 0.0004958860546629089,
"loss": 5.7344,
"mean_token_accuracy": 0.15592280477285386,
"num_tokens": 15931276.0,
"step": 7660
},
{
"entropy": 5.924163198471069,
"epoch": 0.6953011611030478,
"grad_norm": 2.28125,
"learning_rate": 0.0004958798932521186,
"loss": 5.8199,
"mean_token_accuracy": 0.14040146842598916,
"num_tokens": 15941506.0,
"step": 7665
},
{
"entropy": 5.763565111160278,
"epoch": 0.6957547169811321,
"grad_norm": 2.875,
"learning_rate": 0.0004958737272734759,
"loss": 5.639,
"mean_token_accuracy": 0.15290124863386154,
"num_tokens": 15951158.0,
"step": 7670
},
{
"entropy": 5.818205165863037,
"epoch": 0.6962082728592163,
"grad_norm": 2.171875,
"learning_rate": 0.0004958675567271081,
"loss": 5.6771,
"mean_token_accuracy": 0.15895491987466812,
"num_tokens": 15960852.0,
"step": 7675
},
{
"entropy": 5.8290722370147705,
"epoch": 0.6966618287373004,
"grad_norm": 2.609375,
"learning_rate": 0.000495861381613143,
"loss": 5.6803,
"mean_token_accuracy": 0.15305335968732833,
"num_tokens": 15971291.0,
"step": 7680
},
{
"entropy": 5.846955585479736,
"epoch": 0.6971153846153846,
"grad_norm": 3.328125,
"learning_rate": 0.0004958552019317082,
"loss": 5.7539,
"mean_token_accuracy": 0.1479065515100956,
"num_tokens": 15982180.0,
"step": 7685
},
{
"entropy": 5.797716379165649,
"epoch": 0.6975689404934688,
"grad_norm": 3.265625,
"learning_rate": 0.0004958490176829316,
"loss": 5.6355,
"mean_token_accuracy": 0.15521086752414703,
"num_tokens": 15992363.0,
"step": 7690
},
{
"entropy": 5.861604118347168,
"epoch": 0.698022496371553,
"grad_norm": 3.265625,
"learning_rate": 0.000495842828866941,
"loss": 5.7186,
"mean_token_accuracy": 0.1435869313776493,
"num_tokens": 16003477.0,
"step": 7695
},
{
"entropy": 5.885314273834228,
"epoch": 0.6984760522496372,
"grad_norm": 2.28125,
"learning_rate": 0.0004958366354838645,
"loss": 5.6907,
"mean_token_accuracy": 0.15125567615032195,
"num_tokens": 16013747.0,
"step": 7700
},
{
"entropy": 5.860510921478271,
"epoch": 0.6989296081277213,
"grad_norm": 4.0,
"learning_rate": 0.00049583043753383,
"loss": 5.7163,
"mean_token_accuracy": 0.1556942120194435,
"num_tokens": 16025662.0,
"step": 7705
},
{
"entropy": 5.890257215499878,
"epoch": 0.6993831640058055,
"grad_norm": 2.640625,
"learning_rate": 0.0004958242350169658,
"loss": 5.7906,
"mean_token_accuracy": 0.15480666384100913,
"num_tokens": 16037315.0,
"step": 7710
},
{
"entropy": 5.741261672973633,
"epoch": 0.6998367198838897,
"grad_norm": 2.5,
"learning_rate": 0.0004958180279334001,
"loss": 5.5955,
"mean_token_accuracy": 0.1540132313966751,
"num_tokens": 16047297.0,
"step": 7715
},
{
"entropy": 5.783619117736817,
"epoch": 0.7002902757619739,
"grad_norm": 2.859375,
"learning_rate": 0.0004958118162832614,
"loss": 5.5993,
"mean_token_accuracy": 0.1512565240263939,
"num_tokens": 16057573.0,
"step": 7720
},
{
"entropy": 5.8567568302154545,
"epoch": 0.700743831640058,
"grad_norm": 3.0,
"learning_rate": 0.0004958056000666781,
"loss": 5.7383,
"mean_token_accuracy": 0.14643106907606124,
"num_tokens": 16067926.0,
"step": 7725
},
{
"entropy": 5.880869722366333,
"epoch": 0.7011973875181422,
"grad_norm": 3.796875,
"learning_rate": 0.0004957993792837786,
"loss": 5.6846,
"mean_token_accuracy": 0.14718203395605087,
"num_tokens": 16077938.0,
"step": 7730
},
{
"entropy": 5.8295722007751465,
"epoch": 0.7016509433962265,
"grad_norm": 3.484375,
"learning_rate": 0.0004957931539346917,
"loss": 5.5905,
"mean_token_accuracy": 0.15416589826345445,
"num_tokens": 16088163.0,
"step": 7735
},
{
"entropy": 5.909097576141358,
"epoch": 0.7021044992743106,
"grad_norm": 3.078125,
"learning_rate": 0.0004957869240195462,
"loss": 5.8017,
"mean_token_accuracy": 0.14014994204044343,
"num_tokens": 16100640.0,
"step": 7740
},
{
"entropy": 5.823917055130005,
"epoch": 0.7025580551523948,
"grad_norm": 2.046875,
"learning_rate": 0.0004957806895384707,
"loss": 5.6202,
"mean_token_accuracy": 0.15346488058567048,
"num_tokens": 16111018.0,
"step": 7745
},
{
"entropy": 5.7712815284729,
"epoch": 0.7030116110304789,
"grad_norm": 2.828125,
"learning_rate": 0.0004957744504915944,
"loss": 5.6513,
"mean_token_accuracy": 0.15386599451303482,
"num_tokens": 16120873.0,
"step": 7750
},
{
"entropy": 5.8102219581604,
"epoch": 0.7034651669085631,
"grad_norm": 3.09375,
"learning_rate": 0.0004957682068790463,
"loss": 5.666,
"mean_token_accuracy": 0.15427257865667343,
"num_tokens": 16131178.0,
"step": 7755
},
{
"entropy": 5.777589178085327,
"epoch": 0.7039187227866474,
"grad_norm": 2.90625,
"learning_rate": 0.0004957619587009553,
"loss": 5.662,
"mean_token_accuracy": 0.15195267498493195,
"num_tokens": 16141156.0,
"step": 7760
},
{
"entropy": 5.823400831222534,
"epoch": 0.7043722786647315,
"grad_norm": 2.28125,
"learning_rate": 0.0004957557059574507,
"loss": 5.7764,
"mean_token_accuracy": 0.14289068877696992,
"num_tokens": 16150077.0,
"step": 7765
},
{
"entropy": 5.917142295837403,
"epoch": 0.7048258345428157,
"grad_norm": 2.15625,
"learning_rate": 0.0004957494486486618,
"loss": 5.6877,
"mean_token_accuracy": 0.15153144299983978,
"num_tokens": 16160514.0,
"step": 7770
},
{
"entropy": 5.9026023864746096,
"epoch": 0.7052793904208998,
"grad_norm": 2.78125,
"learning_rate": 0.0004957431867747182,
"loss": 5.7275,
"mean_token_accuracy": 0.15452097058296205,
"num_tokens": 16170811.0,
"step": 7775
},
{
"entropy": 5.7668298244476315,
"epoch": 0.7057329462989841,
"grad_norm": 4.15625,
"learning_rate": 0.0004957369203357492,
"loss": 5.6154,
"mean_token_accuracy": 0.15205578207969667,
"num_tokens": 16181553.0,
"step": 7780
},
{
"entropy": 5.758222198486328,
"epoch": 0.7061865021770682,
"grad_norm": 2.25,
"learning_rate": 0.0004957306493318844,
"loss": 5.6736,
"mean_token_accuracy": 0.15890712887048722,
"num_tokens": 16191308.0,
"step": 7785
},
{
"entropy": 5.816359376907348,
"epoch": 0.7066400580551524,
"grad_norm": 2.546875,
"learning_rate": 0.0004957243737632535,
"loss": 5.6282,
"mean_token_accuracy": 0.15420712679624557,
"num_tokens": 16201311.0,
"step": 7790
},
{
"entropy": 5.890701484680176,
"epoch": 0.7070936139332366,
"grad_norm": 3.125,
"learning_rate": 0.0004957180936299863,
"loss": 5.6813,
"mean_token_accuracy": 0.1514902353286743,
"num_tokens": 16211105.0,
"step": 7795
},
{
"entropy": 5.8345139503479,
"epoch": 0.7075471698113207,
"grad_norm": 3.59375,
"learning_rate": 0.0004957118089322127,
"loss": 5.6679,
"mean_token_accuracy": 0.14869241565465927,
"num_tokens": 16220836.0,
"step": 7800
},
{
"entropy": 5.7877466678619385,
"epoch": 0.708000725689405,
"grad_norm": 3.21875,
"learning_rate": 0.0004957055196700628,
"loss": 5.6262,
"mean_token_accuracy": 0.1567522630095482,
"num_tokens": 16231630.0,
"step": 7805
},
{
"entropy": 5.844862079620361,
"epoch": 0.7084542815674891,
"grad_norm": 3.09375,
"learning_rate": 0.0004956992258436663,
"loss": 5.7023,
"mean_token_accuracy": 0.1523435965180397,
"num_tokens": 16242272.0,
"step": 7810
},
{
"entropy": 5.744817018508911,
"epoch": 0.7089078374455733,
"grad_norm": 5.3125,
"learning_rate": 0.0004956929274531536,
"loss": 5.5858,
"mean_token_accuracy": 0.1529250368475914,
"num_tokens": 16252555.0,
"step": 7815
},
{
"entropy": 5.78554220199585,
"epoch": 0.7093613933236574,
"grad_norm": 2.9375,
"learning_rate": 0.000495686624498655,
"loss": 5.72,
"mean_token_accuracy": 0.15313845574855806,
"num_tokens": 16261964.0,
"step": 7820
},
{
"entropy": 5.9008283615112305,
"epoch": 0.7098149492017417,
"grad_norm": 3.015625,
"learning_rate": 0.0004956803169803007,
"loss": 5.778,
"mean_token_accuracy": 0.14940793365240096,
"num_tokens": 16274543.0,
"step": 7825
},
{
"entropy": 5.838950538635254,
"epoch": 0.7102685050798259,
"grad_norm": 2.453125,
"learning_rate": 0.0004956740048982212,
"loss": 5.6871,
"mean_token_accuracy": 0.14389506876468658,
"num_tokens": 16285124.0,
"step": 7830
},
{
"entropy": 5.811396932601928,
"epoch": 0.71072206095791,
"grad_norm": 4.34375,
"learning_rate": 0.0004956676882525472,
"loss": 5.7509,
"mean_token_accuracy": 0.15540383011102676,
"num_tokens": 16295727.0,
"step": 7835
},
{
"entropy": 5.8709159851074215,
"epoch": 0.7111756168359942,
"grad_norm": 3.234375,
"learning_rate": 0.000495661367043409,
"loss": 5.6656,
"mean_token_accuracy": 0.15600730031728743,
"num_tokens": 16305496.0,
"step": 7840
},
{
"entropy": 5.87369556427002,
"epoch": 0.7116291727140783,
"grad_norm": 6.0,
"learning_rate": 0.0004956550412709376,
"loss": 5.6378,
"mean_token_accuracy": 0.15536927729845046,
"num_tokens": 16315112.0,
"step": 7845
},
{
"entropy": 5.886559772491455,
"epoch": 0.7120827285921626,
"grad_norm": 3.90625,
"learning_rate": 0.0004956487109352637,
"loss": 5.6906,
"mean_token_accuracy": 0.1533990427851677,
"num_tokens": 16325613.0,
"step": 7850
},
{
"entropy": 5.820042085647583,
"epoch": 0.7125362844702468,
"grad_norm": 2.328125,
"learning_rate": 0.0004956423760365181,
"loss": 5.6057,
"mean_token_accuracy": 0.15789132863283156,
"num_tokens": 16335317.0,
"step": 7855
},
{
"entropy": 5.794498109817505,
"epoch": 0.7129898403483309,
"grad_norm": 7.625,
"learning_rate": 0.0004956360365748321,
"loss": 5.7599,
"mean_token_accuracy": 0.14459077566862105,
"num_tokens": 16345003.0,
"step": 7860
},
{
"entropy": 5.861040449142456,
"epoch": 0.7134433962264151,
"grad_norm": 3.078125,
"learning_rate": 0.0004956296925503367,
"loss": 5.7328,
"mean_token_accuracy": 0.14798754081130028,
"num_tokens": 16356108.0,
"step": 7865
},
{
"entropy": 5.858279132843018,
"epoch": 0.7138969521044993,
"grad_norm": 5.15625,
"learning_rate": 0.0004956233439631628,
"loss": 5.562,
"mean_token_accuracy": 0.15614887624979018,
"num_tokens": 16364733.0,
"step": 7870
},
{
"entropy": 5.874885559082031,
"epoch": 0.7143505079825835,
"grad_norm": 3.265625,
"learning_rate": 0.0004956169908134421,
"loss": 5.6638,
"mean_token_accuracy": 0.15401551127433777,
"num_tokens": 16375229.0,
"step": 7875
},
{
"entropy": 5.813601684570313,
"epoch": 0.7148040638606676,
"grad_norm": 2.65625,
"learning_rate": 0.0004956106331013057,
"loss": 5.756,
"mean_token_accuracy": 0.15068020671606064,
"num_tokens": 16386366.0,
"step": 7880
},
{
"entropy": 5.802809762954712,
"epoch": 0.7152576197387518,
"grad_norm": 2.875,
"learning_rate": 0.0004956042708268853,
"loss": 5.6342,
"mean_token_accuracy": 0.15512944683432578,
"num_tokens": 16397067.0,
"step": 7885
},
{
"entropy": 5.80576000213623,
"epoch": 0.715711175616836,
"grad_norm": 3.546875,
"learning_rate": 0.0004955979039903123,
"loss": 5.6023,
"mean_token_accuracy": 0.15368096828460692,
"num_tokens": 16407256.0,
"step": 7890
},
{
"entropy": 5.754089975357056,
"epoch": 0.7161647314949202,
"grad_norm": 2.90625,
"learning_rate": 0.0004955915325917183,
"loss": 5.6564,
"mean_token_accuracy": 0.15151944160461425,
"num_tokens": 16418701.0,
"step": 7895
},
{
"entropy": 5.782627153396606,
"epoch": 0.7166182873730044,
"grad_norm": 3.109375,
"learning_rate": 0.0004955851566312354,
"loss": 5.5988,
"mean_token_accuracy": 0.1544649764895439,
"num_tokens": 16428019.0,
"step": 7900
},
{
"entropy": 5.766601467132569,
"epoch": 0.7170718432510885,
"grad_norm": 4.15625,
"learning_rate": 0.0004955787761089952,
"loss": 5.6641,
"mean_token_accuracy": 0.1493433177471161,
"num_tokens": 16440122.0,
"step": 7905
},
{
"entropy": 5.884334135055542,
"epoch": 0.7175253991291727,
"grad_norm": 3.078125,
"learning_rate": 0.0004955723910251296,
"loss": 5.7265,
"mean_token_accuracy": 0.1512959912419319,
"num_tokens": 16449801.0,
"step": 7910
},
{
"entropy": 5.7857404232025145,
"epoch": 0.717978955007257,
"grad_norm": 3.375,
"learning_rate": 0.0004955660013797709,
"loss": 5.5926,
"mean_token_accuracy": 0.15442865192890168,
"num_tokens": 16460155.0,
"step": 7915
},
{
"entropy": 5.845020294189453,
"epoch": 0.7184325108853411,
"grad_norm": 2.75,
"learning_rate": 0.000495559607173051,
"loss": 5.7866,
"mean_token_accuracy": 0.14254856258630752,
"num_tokens": 16470656.0,
"step": 7920
},
{
"entropy": 5.876619863510132,
"epoch": 0.7188860667634253,
"grad_norm": 2.3125,
"learning_rate": 0.0004955532084051022,
"loss": 5.6977,
"mean_token_accuracy": 0.15091280341148378,
"num_tokens": 16481449.0,
"step": 7925
},
{
"entropy": 5.9208259105682375,
"epoch": 0.7193396226415094,
"grad_norm": 3.3125,
"learning_rate": 0.0004955468050760568,
"loss": 5.7113,
"mean_token_accuracy": 0.15328221768140793,
"num_tokens": 16492032.0,
"step": 7930
},
{
"entropy": 5.832501029968261,
"epoch": 0.7197931785195936,
"grad_norm": 2.4375,
"learning_rate": 0.0004955403971860474,
"loss": 5.6446,
"mean_token_accuracy": 0.1522648885846138,
"num_tokens": 16502221.0,
"step": 7935
},
{
"entropy": 5.8008585453033445,
"epoch": 0.7202467343976778,
"grad_norm": 3.1875,
"learning_rate": 0.0004955339847352063,
"loss": 5.6355,
"mean_token_accuracy": 0.15295949429273606,
"num_tokens": 16512021.0,
"step": 7940
},
{
"entropy": 5.8326164245605465,
"epoch": 0.720700290275762,
"grad_norm": 4.875,
"learning_rate": 0.0004955275677236662,
"loss": 5.6849,
"mean_token_accuracy": 0.14992091357707976,
"num_tokens": 16522060.0,
"step": 7945
},
{
"entropy": 5.882563972473145,
"epoch": 0.7211538461538461,
"grad_norm": 3.296875,
"learning_rate": 0.0004955211461515599,
"loss": 5.6814,
"mean_token_accuracy": 0.155203914642334,
"num_tokens": 16532827.0,
"step": 7950
},
{
"entropy": 5.8585892677307125,
"epoch": 0.7216074020319303,
"grad_norm": 3.46875,
"learning_rate": 0.00049551472001902,
"loss": 5.7052,
"mean_token_accuracy": 0.15561985224485397,
"num_tokens": 16542838.0,
"step": 7955
},
{
"entropy": 5.779575490951538,
"epoch": 0.7220609579100146,
"grad_norm": 2.859375,
"learning_rate": 0.0004955082893261796,
"loss": 5.6582,
"mean_token_accuracy": 0.15778737068176268,
"num_tokens": 16552615.0,
"step": 7960
},
{
"entropy": 5.908274412155151,
"epoch": 0.7225145137880987,
"grad_norm": 2.515625,
"learning_rate": 0.0004955018540731714,
"loss": 5.726,
"mean_token_accuracy": 0.14718418419361115,
"num_tokens": 16561777.0,
"step": 7965
},
{
"entropy": 5.918739891052246,
"epoch": 0.7229680696661829,
"grad_norm": 5.125,
"learning_rate": 0.0004954954142601288,
"loss": 5.6905,
"mean_token_accuracy": 0.1526739552617073,
"num_tokens": 16572111.0,
"step": 7970
},
{
"entropy": 5.830246782302856,
"epoch": 0.723421625544267,
"grad_norm": 1.875,
"learning_rate": 0.0004954889698871849,
"loss": 5.7246,
"mean_token_accuracy": 0.15518662035465242,
"num_tokens": 16581674.0,
"step": 7975
},
{
"entropy": 5.847196817398071,
"epoch": 0.7238751814223512,
"grad_norm": 2.296875,
"learning_rate": 0.0004954825209544728,
"loss": 5.7107,
"mean_token_accuracy": 0.15622829049825668,
"num_tokens": 16591802.0,
"step": 7980
},
{
"entropy": 5.819644355773926,
"epoch": 0.7243287373004355,
"grad_norm": 3.90625,
"learning_rate": 0.000495476067462126,
"loss": 5.6259,
"mean_token_accuracy": 0.1592419072985649,
"num_tokens": 16602772.0,
"step": 7985
},
{
"entropy": 5.8849729061126705,
"epoch": 0.7247822931785196,
"grad_norm": 3.421875,
"learning_rate": 0.000495469609410278,
"loss": 5.669,
"mean_token_accuracy": 0.15404960215091706,
"num_tokens": 16613526.0,
"step": 7990
},
{
"entropy": 5.807653331756592,
"epoch": 0.7252358490566038,
"grad_norm": 2.890625,
"learning_rate": 0.0004954631467990622,
"loss": 5.6769,
"mean_token_accuracy": 0.1550093710422516,
"num_tokens": 16623583.0,
"step": 7995
},
{
"entropy": 5.852053880691528,
"epoch": 0.7256894049346879,
"grad_norm": 2.65625,
"learning_rate": 0.0004954566796286124,
"loss": 5.7295,
"mean_token_accuracy": 0.1478324055671692,
"num_tokens": 16635068.0,
"step": 8000
},
{
"entropy": 5.870395565032959,
"epoch": 0.7261429608127722,
"grad_norm": 2.890625,
"learning_rate": 0.0004954502078990623,
"loss": 5.6539,
"mean_token_accuracy": 0.15322620123624803,
"num_tokens": 16645021.0,
"step": 8005
},
{
"entropy": 5.84760684967041,
"epoch": 0.7265965166908563,
"grad_norm": 4.90625,
"learning_rate": 0.0004954437316105457,
"loss": 5.7219,
"mean_token_accuracy": 0.14631387591362,
"num_tokens": 16656021.0,
"step": 8010
},
{
"entropy": 5.818306398391724,
"epoch": 0.7270500725689405,
"grad_norm": 1.8671875,
"learning_rate": 0.0004954372507631965,
"loss": 5.6209,
"mean_token_accuracy": 0.15864094197750092,
"num_tokens": 16665236.0,
"step": 8015
},
{
"entropy": 5.840885925292969,
"epoch": 0.7275036284470247,
"grad_norm": 2.203125,
"learning_rate": 0.0004954307653571489,
"loss": 5.8054,
"mean_token_accuracy": 0.14249199628829956,
"num_tokens": 16676690.0,
"step": 8020
},
{
"entropy": 5.868116760253907,
"epoch": 0.7279571843251088,
"grad_norm": 4.5625,
"learning_rate": 0.0004954242753925367,
"loss": 5.6937,
"mean_token_accuracy": 0.14094067886471748,
"num_tokens": 16687288.0,
"step": 8025
},
{
"entropy": 5.850828790664673,
"epoch": 0.7284107402031931,
"grad_norm": 2.875,
"learning_rate": 0.0004954177808694946,
"loss": 5.7177,
"mean_token_accuracy": 0.14539122730493545,
"num_tokens": 16697873.0,
"step": 8030
},
{
"entropy": 5.858789253234863,
"epoch": 0.7288642960812772,
"grad_norm": 3.484375,
"learning_rate": 0.0004954112817881565,
"loss": 5.7026,
"mean_token_accuracy": 0.15312349796295166,
"num_tokens": 16709002.0,
"step": 8035
},
{
"entropy": 5.85047926902771,
"epoch": 0.7293178519593614,
"grad_norm": 3.15625,
"learning_rate": 0.0004954047781486568,
"loss": 5.6745,
"mean_token_accuracy": 0.1481028601527214,
"num_tokens": 16718989.0,
"step": 8040
},
{
"entropy": 5.945074272155762,
"epoch": 0.7297714078374455,
"grad_norm": 4.96875,
"learning_rate": 0.0004953982699511303,
"loss": 5.7757,
"mean_token_accuracy": 0.14789580926299095,
"num_tokens": 16730425.0,
"step": 8045
},
{
"entropy": 5.838289260864258,
"epoch": 0.7302249637155298,
"grad_norm": 2.4375,
"learning_rate": 0.0004953917571957114,
"loss": 5.6377,
"mean_token_accuracy": 0.15277301967144014,
"num_tokens": 16740216.0,
"step": 8050
},
{
"entropy": 5.768487119674683,
"epoch": 0.730678519593614,
"grad_norm": 5.03125,
"learning_rate": 0.0004953852398825346,
"loss": 5.6622,
"mean_token_accuracy": 0.15074789077043532,
"num_tokens": 16752436.0,
"step": 8055
},
{
"entropy": 5.748122119903565,
"epoch": 0.7311320754716981,
"grad_norm": 3.453125,
"learning_rate": 0.000495378718011735,
"loss": 5.6738,
"mean_token_accuracy": 0.15310948193073273,
"num_tokens": 16762631.0,
"step": 8060
},
{
"entropy": 5.8812836647033695,
"epoch": 0.7315856313497823,
"grad_norm": 3.1875,
"learning_rate": 0.0004953721915834473,
"loss": 5.6898,
"mean_token_accuracy": 0.1502485454082489,
"num_tokens": 16772613.0,
"step": 8065
},
{
"entropy": 5.8909612655639645,
"epoch": 0.7320391872278664,
"grad_norm": 2.984375,
"learning_rate": 0.0004953656605978067,
"loss": 5.7026,
"mean_token_accuracy": 0.14999075829982758,
"num_tokens": 16782963.0,
"step": 8070
},
{
"entropy": 5.819808101654052,
"epoch": 0.7324927431059507,
"grad_norm": 3.4375,
"learning_rate": 0.000495359125054948,
"loss": 5.6277,
"mean_token_accuracy": 0.15619223564863205,
"num_tokens": 16793892.0,
"step": 8075
},
{
"entropy": 5.7889190196990965,
"epoch": 0.7329462989840348,
"grad_norm": 2.75,
"learning_rate": 0.0004953525849550063,
"loss": 5.5992,
"mean_token_accuracy": 0.15081862434744836,
"num_tokens": 16803388.0,
"step": 8080
},
{
"entropy": 5.842611455917359,
"epoch": 0.733399854862119,
"grad_norm": 2.46875,
"learning_rate": 0.0004953460402981172,
"loss": 5.5759,
"mean_token_accuracy": 0.1550586625933647,
"num_tokens": 16813201.0,
"step": 8085
},
{
"entropy": 5.900736331939697,
"epoch": 0.7338534107402032,
"grad_norm": 2.125,
"learning_rate": 0.0004953394910844157,
"loss": 5.8161,
"mean_token_accuracy": 0.15295570343732834,
"num_tokens": 16823265.0,
"step": 8090
},
{
"entropy": 5.833998298645019,
"epoch": 0.7343069666182874,
"grad_norm": 1.921875,
"learning_rate": 0.0004953329373140374,
"loss": 5.5994,
"mean_token_accuracy": 0.15627037435770036,
"num_tokens": 16833057.0,
"step": 8095
},
{
"entropy": 5.8118898391723635,
"epoch": 0.7347605224963716,
"grad_norm": 3.4375,
"learning_rate": 0.0004953263789871179,
"loss": 5.5855,
"mean_token_accuracy": 0.15333315581083298,
"num_tokens": 16843488.0,
"step": 8100
},
{
"entropy": 5.778717279434204,
"epoch": 0.7352140783744557,
"grad_norm": 2.484375,
"learning_rate": 0.0004953198161037925,
"loss": 5.6183,
"mean_token_accuracy": 0.15874518752098082,
"num_tokens": 16852448.0,
"step": 8105
},
{
"entropy": 5.782966566085816,
"epoch": 0.7356676342525399,
"grad_norm": 9.5625,
"learning_rate": 0.0004953132486641975,
"loss": 5.6685,
"mean_token_accuracy": 0.15144542902708052,
"num_tokens": 16863944.0,
"step": 8110
},
{
"entropy": 5.844910621643066,
"epoch": 0.736121190130624,
"grad_norm": 2.515625,
"learning_rate": 0.000495306676668468,
"loss": 5.5967,
"mean_token_accuracy": 0.15626500099897384,
"num_tokens": 16873923.0,
"step": 8115
},
{
"entropy": 5.809791517257691,
"epoch": 0.7365747460087083,
"grad_norm": 2.09375,
"learning_rate": 0.0004953001001167406,
"loss": 5.6782,
"mean_token_accuracy": 0.15416514575481416,
"num_tokens": 16884373.0,
"step": 8120
},
{
"entropy": 5.8766303062438965,
"epoch": 0.7370283018867925,
"grad_norm": 2.703125,
"learning_rate": 0.0004952935190091508,
"loss": 5.7971,
"mean_token_accuracy": 0.14941829442977905,
"num_tokens": 16895869.0,
"step": 8125
},
{
"entropy": 5.778482818603516,
"epoch": 0.7374818577648766,
"grad_norm": 3.421875,
"learning_rate": 0.0004952869333458349,
"loss": 5.6802,
"mean_token_accuracy": 0.1520070217549801,
"num_tokens": 16906737.0,
"step": 8130
},
{
"entropy": 5.758775615692139,
"epoch": 0.7379354136429608,
"grad_norm": 5.1875,
"learning_rate": 0.0004952803431269292,
"loss": 5.5958,
"mean_token_accuracy": 0.15876534283161164,
"num_tokens": 16916083.0,
"step": 8135
},
{
"entropy": 5.893442249298095,
"epoch": 0.738388969521045,
"grad_norm": 3.546875,
"learning_rate": 0.0004952737483525699,
"loss": 5.7493,
"mean_token_accuracy": 0.14965495616197586,
"num_tokens": 16926419.0,
"step": 8140
},
{
"entropy": 5.902286958694458,
"epoch": 0.7388425253991292,
"grad_norm": 2.734375,
"learning_rate": 0.0004952671490228932,
"loss": 5.6993,
"mean_token_accuracy": 0.14944136291742324,
"num_tokens": 16936067.0,
"step": 8145
},
{
"entropy": 5.920256948471069,
"epoch": 0.7392960812772134,
"grad_norm": 3.1875,
"learning_rate": 0.0004952605451380357,
"loss": 5.6441,
"mean_token_accuracy": 0.1541039377450943,
"num_tokens": 16946571.0,
"step": 8150
},
{
"entropy": 5.820447397232056,
"epoch": 0.7397496371552975,
"grad_norm": 1.9375,
"learning_rate": 0.0004952539366981342,
"loss": 5.7088,
"mean_token_accuracy": 0.14866316616535186,
"num_tokens": 16956764.0,
"step": 8155
},
{
"entropy": 5.835457468032837,
"epoch": 0.7402031930333817,
"grad_norm": 2.375,
"learning_rate": 0.000495247323703325,
"loss": 5.6093,
"mean_token_accuracy": 0.1589105397462845,
"num_tokens": 16966002.0,
"step": 8160
},
{
"entropy": 5.838615369796753,
"epoch": 0.7406567489114659,
"grad_norm": 2.4375,
"learning_rate": 0.000495240706153745,
"loss": 5.7271,
"mean_token_accuracy": 0.1552692323923111,
"num_tokens": 16975705.0,
"step": 8165
},
{
"entropy": 5.8247520446777346,
"epoch": 0.7411103047895501,
"grad_norm": 2.390625,
"learning_rate": 0.0004952340840495312,
"loss": 5.6383,
"mean_token_accuracy": 0.15176112204790115,
"num_tokens": 16985945.0,
"step": 8170
},
{
"entropy": 5.8564215183258055,
"epoch": 0.7415638606676342,
"grad_norm": 2.1875,
"learning_rate": 0.0004952274573908203,
"loss": 5.6046,
"mean_token_accuracy": 0.15430220663547517,
"num_tokens": 16995475.0,
"step": 8175
},
{
"entropy": 5.7460408210754395,
"epoch": 0.7420174165457184,
"grad_norm": 2.5625,
"learning_rate": 0.0004952208261777496,
"loss": 5.6529,
"mean_token_accuracy": 0.15099769681692124,
"num_tokens": 17005028.0,
"step": 8180
},
{
"entropy": 5.812251472473145,
"epoch": 0.7424709724238027,
"grad_norm": 2.40625,
"learning_rate": 0.000495214190410456,
"loss": 5.6912,
"mean_token_accuracy": 0.15830041766166686,
"num_tokens": 17014786.0,
"step": 8185
},
{
"entropy": 5.9062474250793455,
"epoch": 0.7429245283018868,
"grad_norm": 2.546875,
"learning_rate": 0.0004952075500890768,
"loss": 5.6934,
"mean_token_accuracy": 0.1491142690181732,
"num_tokens": 17024431.0,
"step": 8190
},
{
"entropy": 5.860283613204956,
"epoch": 0.743378084179971,
"grad_norm": 3.25,
"learning_rate": 0.0004952009052137493,
"loss": 5.6797,
"mean_token_accuracy": 0.1560024857521057,
"num_tokens": 17034368.0,
"step": 8195
},
{
"entropy": 5.803822708129883,
"epoch": 0.7438316400580551,
"grad_norm": 2.859375,
"learning_rate": 0.0004951942557846112,
"loss": 5.7083,
"mean_token_accuracy": 0.15151755660772323,
"num_tokens": 17043818.0,
"step": 8200
},
{
"entropy": 5.9088828563690186,
"epoch": 0.7442851959361393,
"grad_norm": 3.0625,
"learning_rate": 0.0004951876018017997,
"loss": 5.7013,
"mean_token_accuracy": 0.15114538073539735,
"num_tokens": 17054843.0,
"step": 8205
},
{
"entropy": 5.918909502029419,
"epoch": 0.7447387518142236,
"grad_norm": 2.078125,
"learning_rate": 0.0004951809432654524,
"loss": 5.7167,
"mean_token_accuracy": 0.15287568420171738,
"num_tokens": 17065539.0,
"step": 8210
},
{
"entropy": 5.796352529525757,
"epoch": 0.7451923076923077,
"grad_norm": 2.84375,
"learning_rate": 0.000495174280175707,
"loss": 5.6057,
"mean_token_accuracy": 0.1571377620100975,
"num_tokens": 17075443.0,
"step": 8215
},
{
"entropy": 5.875583744049072,
"epoch": 0.7456458635703919,
"grad_norm": 3.078125,
"learning_rate": 0.0004951676125327015,
"loss": 5.7409,
"mean_token_accuracy": 0.15380513221025466,
"num_tokens": 17085490.0,
"step": 8220
},
{
"entropy": 5.88790078163147,
"epoch": 0.746099419448476,
"grad_norm": 2.265625,
"learning_rate": 0.0004951609403365737,
"loss": 5.7283,
"mean_token_accuracy": 0.15716300457715987,
"num_tokens": 17094803.0,
"step": 8225
},
{
"entropy": 5.826867723464966,
"epoch": 0.7465529753265603,
"grad_norm": 2.953125,
"learning_rate": 0.0004951542635874615,
"loss": 5.5683,
"mean_token_accuracy": 0.15698309987783432,
"num_tokens": 17104373.0,
"step": 8230
},
{
"entropy": 5.838105249404907,
"epoch": 0.7470065312046444,
"grad_norm": 3.546875,
"learning_rate": 0.0004951475822855029,
"loss": 5.6744,
"mean_token_accuracy": 0.1605461299419403,
"num_tokens": 17115469.0,
"step": 8235
},
{
"entropy": 5.7935174942016605,
"epoch": 0.7474600870827286,
"grad_norm": 2.65625,
"learning_rate": 0.0004951408964308364,
"loss": 5.6534,
"mean_token_accuracy": 0.15251931995153428,
"num_tokens": 17125443.0,
"step": 8240
},
{
"entropy": 5.811115503311157,
"epoch": 0.7479136429608128,
"grad_norm": 3.484375,
"learning_rate": 0.0004951342060236,
"loss": 5.6107,
"mean_token_accuracy": 0.15216806083917617,
"num_tokens": 17136253.0,
"step": 8245
},
{
"entropy": 5.925781488418579,
"epoch": 0.7483671988388969,
"grad_norm": 2.421875,
"learning_rate": 0.0004951275110639321,
"loss": 5.6833,
"mean_token_accuracy": 0.1573117569088936,
"num_tokens": 17145544.0,
"step": 8250
},
{
"entropy": 5.821874618530273,
"epoch": 0.7488207547169812,
"grad_norm": 2.203125,
"learning_rate": 0.0004951208115519712,
"loss": 5.6796,
"mean_token_accuracy": 0.15335864275693895,
"num_tokens": 17155043.0,
"step": 8255
},
{
"entropy": 5.746136140823364,
"epoch": 0.7492743105950653,
"grad_norm": 2.828125,
"learning_rate": 0.0004951141074878558,
"loss": 5.7308,
"mean_token_accuracy": 0.1515687733888626,
"num_tokens": 17164885.0,
"step": 8260
},
{
"entropy": 5.913036060333252,
"epoch": 0.7497278664731495,
"grad_norm": 2.328125,
"learning_rate": 0.0004951073988717246,
"loss": 5.7102,
"mean_token_accuracy": 0.15280442237854003,
"num_tokens": 17175681.0,
"step": 8265
},
{
"entropy": 5.837083959579468,
"epoch": 0.7501814223512336,
"grad_norm": 2.28125,
"learning_rate": 0.0004951006857037163,
"loss": 5.577,
"mean_token_accuracy": 0.15716822296380997,
"num_tokens": 17186002.0,
"step": 8270
},
{
"entropy": 5.79591007232666,
"epoch": 0.7506349782293179,
"grad_norm": 4.0625,
"learning_rate": 0.0004950939679839696,
"loss": 5.634,
"mean_token_accuracy": 0.1542076975107193,
"num_tokens": 17195499.0,
"step": 8275
},
{
"entropy": 5.812315368652344,
"epoch": 0.7510885341074021,
"grad_norm": 9.1875,
"learning_rate": 0.0004950872457126237,
"loss": 5.6553,
"mean_token_accuracy": 0.1538190796971321,
"num_tokens": 17204684.0,
"step": 8280
},
{
"entropy": 5.837529897689819,
"epoch": 0.7515420899854862,
"grad_norm": 2.4375,
"learning_rate": 0.0004950805188898175,
"loss": 5.6769,
"mean_token_accuracy": 0.1444774940609932,
"num_tokens": 17213987.0,
"step": 8285
},
{
"entropy": 5.825541210174561,
"epoch": 0.7519956458635704,
"grad_norm": 3.25,
"learning_rate": 0.00049507378751569,
"loss": 5.7259,
"mean_token_accuracy": 0.1459720402956009,
"num_tokens": 17224512.0,
"step": 8290
},
{
"entropy": 5.912886762619019,
"epoch": 0.7524492017416545,
"grad_norm": 3.265625,
"learning_rate": 0.0004950670515903805,
"loss": 5.7117,
"mean_token_accuracy": 0.1544044181704521,
"num_tokens": 17234362.0,
"step": 8295
},
{
"entropy": 5.879697990417481,
"epoch": 0.7529027576197388,
"grad_norm": 2.015625,
"learning_rate": 0.0004950603111140283,
"loss": 5.7077,
"mean_token_accuracy": 0.1570012852549553,
"num_tokens": 17244241.0,
"step": 8300
},
{
"entropy": 5.806952905654907,
"epoch": 0.753356313497823,
"grad_norm": 2.609375,
"learning_rate": 0.0004950535660867728,
"loss": 5.6538,
"mean_token_accuracy": 0.15973338037729262,
"num_tokens": 17254689.0,
"step": 8305
},
{
"entropy": 5.83306655883789,
"epoch": 0.7538098693759071,
"grad_norm": 2.8125,
"learning_rate": 0.0004950468165087535,
"loss": 5.7449,
"mean_token_accuracy": 0.14953402578830718,
"num_tokens": 17265506.0,
"step": 8310
},
{
"entropy": 5.895908164978027,
"epoch": 0.7542634252539913,
"grad_norm": 5.375,
"learning_rate": 0.0004950400623801098,
"loss": 5.7896,
"mean_token_accuracy": 0.1489114761352539,
"num_tokens": 17276376.0,
"step": 8315
},
{
"entropy": 5.948290395736694,
"epoch": 0.7547169811320755,
"grad_norm": 2.171875,
"learning_rate": 0.0004950333037009818,
"loss": 5.7097,
"mean_token_accuracy": 0.14384376108646393,
"num_tokens": 17287005.0,
"step": 8320
},
{
"entropy": 5.916612863540649,
"epoch": 0.7551705370101597,
"grad_norm": 2.71875,
"learning_rate": 0.0004950265404715089,
"loss": 5.713,
"mean_token_accuracy": 0.14719957262277603,
"num_tokens": 17297595.0,
"step": 8325
},
{
"entropy": 5.76622200012207,
"epoch": 0.7556240928882438,
"grad_norm": 2.359375,
"learning_rate": 0.000495019772691831,
"loss": 5.5757,
"mean_token_accuracy": 0.15127571076154708,
"num_tokens": 17307319.0,
"step": 8330
},
{
"entropy": 5.819165468215942,
"epoch": 0.756077648766328,
"grad_norm": 8.3125,
"learning_rate": 0.0004950130003620882,
"loss": 5.7815,
"mean_token_accuracy": 0.15138889104127884,
"num_tokens": 17317978.0,
"step": 8335
},
{
"entropy": 5.835421848297119,
"epoch": 0.7565312046444121,
"grad_norm": 3.078125,
"learning_rate": 0.0004950062234824204,
"loss": 5.6732,
"mean_token_accuracy": 0.14981579184532165,
"num_tokens": 17328339.0,
"step": 8340
},
{
"entropy": 5.840034532546997,
"epoch": 0.7569847605224964,
"grad_norm": 2.78125,
"learning_rate": 0.0004949994420529679,
"loss": 5.7422,
"mean_token_accuracy": 0.14759217202663422,
"num_tokens": 17338668.0,
"step": 8345
},
{
"entropy": 5.747548532485962,
"epoch": 0.7574383164005806,
"grad_norm": 2.453125,
"learning_rate": 0.0004949926560738709,
"loss": 5.6809,
"mean_token_accuracy": 0.1471247635781765,
"num_tokens": 17348609.0,
"step": 8350
},
{
"entropy": 5.8837117671966555,
"epoch": 0.7578918722786647,
"grad_norm": 2.640625,
"learning_rate": 0.0004949858655452697,
"loss": 5.6749,
"mean_token_accuracy": 0.1528981566429138,
"num_tokens": 17358014.0,
"step": 8355
},
{
"entropy": 5.897519922256469,
"epoch": 0.7583454281567489,
"grad_norm": 2.84375,
"learning_rate": 0.0004949790704673048,
"loss": 5.7057,
"mean_token_accuracy": 0.150955131649971,
"num_tokens": 17367975.0,
"step": 8360
},
{
"entropy": 5.8025383949279785,
"epoch": 0.7587989840348331,
"grad_norm": 2.8125,
"learning_rate": 0.0004949722708401166,
"loss": 5.6525,
"mean_token_accuracy": 0.15427399575710296,
"num_tokens": 17377960.0,
"step": 8365
},
{
"entropy": 5.839379596710205,
"epoch": 0.7592525399129173,
"grad_norm": 4.8125,
"learning_rate": 0.0004949654666638457,
"loss": 5.7023,
"mean_token_accuracy": 0.15956505686044692,
"num_tokens": 17387334.0,
"step": 8370
},
{
"entropy": 5.810500812530518,
"epoch": 0.7597060957910015,
"grad_norm": 3.359375,
"learning_rate": 0.000494958657938633,
"loss": 5.6878,
"mean_token_accuracy": 0.15003183782100676,
"num_tokens": 17396521.0,
"step": 8375
},
{
"entropy": 5.75392165184021,
"epoch": 0.7601596516690856,
"grad_norm": 2.1875,
"learning_rate": 0.0004949518446646192,
"loss": 5.6549,
"mean_token_accuracy": 0.15234262943267823,
"num_tokens": 17407217.0,
"step": 8380
},
{
"entropy": 5.897473382949829,
"epoch": 0.7606132075471698,
"grad_norm": 2.03125,
"learning_rate": 0.0004949450268419451,
"loss": 5.7301,
"mean_token_accuracy": 0.15099592208862306,
"num_tokens": 17416982.0,
"step": 8385
},
{
"entropy": 5.7785039901733395,
"epoch": 0.761066763425254,
"grad_norm": 2.84375,
"learning_rate": 0.000494938204470752,
"loss": 5.6164,
"mean_token_accuracy": 0.153751140832901,
"num_tokens": 17427636.0,
"step": 8390
},
{
"entropy": 5.792123603820801,
"epoch": 0.7615203193033382,
"grad_norm": 2.140625,
"learning_rate": 0.0004949313775511806,
"loss": 5.7164,
"mean_token_accuracy": 0.1500251740217209,
"num_tokens": 17437488.0,
"step": 8395
},
{
"entropy": 5.826402997970581,
"epoch": 0.7619738751814223,
"grad_norm": 2.390625,
"learning_rate": 0.0004949245460833723,
"loss": 5.5337,
"mean_token_accuracy": 0.15660372674465178,
"num_tokens": 17447838.0,
"step": 8400
},
{
"entropy": 5.9221526145935055,
"epoch": 0.7624274310595065,
"grad_norm": 2.546875,
"learning_rate": 0.0004949177100674684,
"loss": 5.7284,
"mean_token_accuracy": 0.15142788887023925,
"num_tokens": 17457257.0,
"step": 8405
},
{
"entropy": 5.868713808059693,
"epoch": 0.7628809869375908,
"grad_norm": 3.75,
"learning_rate": 0.0004949108695036102,
"loss": 5.6654,
"mean_token_accuracy": 0.1565401151776314,
"num_tokens": 17467729.0,
"step": 8410
},
{
"entropy": 5.8393584251403805,
"epoch": 0.7633345428156749,
"grad_norm": 3.328125,
"learning_rate": 0.0004949040243919392,
"loss": 5.6878,
"mean_token_accuracy": 0.1448308013379574,
"num_tokens": 17477692.0,
"step": 8415
},
{
"entropy": 5.799105072021485,
"epoch": 0.7637880986937591,
"grad_norm": 3.015625,
"learning_rate": 0.0004948971747325968,
"loss": 5.6738,
"mean_token_accuracy": 0.15476165413856507,
"num_tokens": 17488730.0,
"step": 8420
},
{
"entropy": 5.824799346923828,
"epoch": 0.7642416545718432,
"grad_norm": 3.640625,
"learning_rate": 0.000494890320525725,
"loss": 5.7578,
"mean_token_accuracy": 0.1491442933678627,
"num_tokens": 17498374.0,
"step": 8425
},
{
"entropy": 5.872576713562012,
"epoch": 0.7646952104499274,
"grad_norm": 2.953125,
"learning_rate": 0.0004948834617714652,
"loss": 5.6221,
"mean_token_accuracy": 0.16017669290304185,
"num_tokens": 17508601.0,
"step": 8430
},
{
"entropy": 5.882133340835571,
"epoch": 0.7651487663280117,
"grad_norm": 3.875,
"learning_rate": 0.0004948765984699595,
"loss": 5.6619,
"mean_token_accuracy": 0.15254097878932954,
"num_tokens": 17518164.0,
"step": 8435
},
{
"entropy": 5.8255561828613285,
"epoch": 0.7656023222060958,
"grad_norm": 8.625,
"learning_rate": 0.0004948697306213494,
"loss": 5.6263,
"mean_token_accuracy": 0.15994967967271806,
"num_tokens": 17527244.0,
"step": 8440
},
{
"entropy": 5.820489692687988,
"epoch": 0.76605587808418,
"grad_norm": 2.71875,
"learning_rate": 0.0004948628582257775,
"loss": 5.7001,
"mean_token_accuracy": 0.15338760912418364,
"num_tokens": 17538045.0,
"step": 8445
},
{
"entropy": 5.837648820877075,
"epoch": 0.7665094339622641,
"grad_norm": 2.578125,
"learning_rate": 0.0004948559812833857,
"loss": 5.6657,
"mean_token_accuracy": 0.1533903732895851,
"num_tokens": 17547584.0,
"step": 8450
},
{
"entropy": 5.848302984237671,
"epoch": 0.7669629898403484,
"grad_norm": 4.28125,
"learning_rate": 0.0004948490997943162,
"loss": 5.6926,
"mean_token_accuracy": 0.14964775890111923,
"num_tokens": 17558925.0,
"step": 8455
},
{
"entropy": 5.825404071807862,
"epoch": 0.7674165457184325,
"grad_norm": 2.28125,
"learning_rate": 0.0004948422137587111,
"loss": 5.7734,
"mean_token_accuracy": 0.1478324979543686,
"num_tokens": 17569355.0,
"step": 8460
},
{
"entropy": 5.838606214523315,
"epoch": 0.7678701015965167,
"grad_norm": 2.71875,
"learning_rate": 0.0004948353231767131,
"loss": 5.65,
"mean_token_accuracy": 0.15104758143424987,
"num_tokens": 17579782.0,
"step": 8465
},
{
"entropy": 5.851181888580323,
"epoch": 0.7683236574746009,
"grad_norm": 2.125,
"learning_rate": 0.0004948284280484645,
"loss": 5.6885,
"mean_token_accuracy": 0.15970910787582399,
"num_tokens": 17589772.0,
"step": 8470
},
{
"entropy": 5.735399580001831,
"epoch": 0.768777213352685,
"grad_norm": 2.765625,
"learning_rate": 0.0004948215283741081,
"loss": 5.6243,
"mean_token_accuracy": 0.15218766033649445,
"num_tokens": 17599613.0,
"step": 8475
},
{
"entropy": 5.824175262451172,
"epoch": 0.7692307692307693,
"grad_norm": 4.34375,
"learning_rate": 0.0004948146241537866,
"loss": 5.704,
"mean_token_accuracy": 0.15038759112358094,
"num_tokens": 17609056.0,
"step": 8480
},
{
"entropy": 5.834938859939575,
"epoch": 0.7696843251088534,
"grad_norm": 2.15625,
"learning_rate": 0.0004948077153876425,
"loss": 5.6183,
"mean_token_accuracy": 0.15458367764949799,
"num_tokens": 17618620.0,
"step": 8485
},
{
"entropy": 5.845889282226563,
"epoch": 0.7701378809869376,
"grad_norm": 3.953125,
"learning_rate": 0.0004948008020758188,
"loss": 5.5825,
"mean_token_accuracy": 0.1617045596241951,
"num_tokens": 17627630.0,
"step": 8490
},
{
"entropy": 5.747945404052734,
"epoch": 0.7705914368650217,
"grad_norm": 4.0,
"learning_rate": 0.0004947938842184585,
"loss": 5.615,
"mean_token_accuracy": 0.15477360188961028,
"num_tokens": 17638903.0,
"step": 8495
},
{
"entropy": 5.8060832023620605,
"epoch": 0.771044992743106,
"grad_norm": 3.515625,
"learning_rate": 0.0004947869618157048,
"loss": 5.6517,
"mean_token_accuracy": 0.15006260126829146,
"num_tokens": 17649257.0,
"step": 8500
},
{
"entropy": 5.867640686035156,
"epoch": 0.7714985486211902,
"grad_norm": 2.734375,
"learning_rate": 0.0004947800348677006,
"loss": 5.6267,
"mean_token_accuracy": 0.16068045347929,
"num_tokens": 17659495.0,
"step": 8505
},
{
"entropy": 5.778565263748169,
"epoch": 0.7719521044992743,
"grad_norm": 4.09375,
"learning_rate": 0.0004947731033745893,
"loss": 5.607,
"mean_token_accuracy": 0.15641812831163407,
"num_tokens": 17669671.0,
"step": 8510
},
{
"entropy": 5.852364349365234,
"epoch": 0.7724056603773585,
"grad_norm": 3.140625,
"learning_rate": 0.0004947661673365142,
"loss": 5.7728,
"mean_token_accuracy": 0.1515066832304001,
"num_tokens": 17680927.0,
"step": 8515
},
{
"entropy": 5.812708568572998,
"epoch": 0.7728592162554426,
"grad_norm": 3.28125,
"learning_rate": 0.0004947592267536188,
"loss": 5.6459,
"mean_token_accuracy": 0.1577283501625061,
"num_tokens": 17691378.0,
"step": 8520
},
{
"entropy": 5.806360816955566,
"epoch": 0.7733127721335269,
"grad_norm": 4.125,
"learning_rate": 0.0004947522816260466,
"loss": 5.7186,
"mean_token_accuracy": 0.14888853952288628,
"num_tokens": 17700942.0,
"step": 8525
},
{
"entropy": 5.797103643417358,
"epoch": 0.773766328011611,
"grad_norm": 3.140625,
"learning_rate": 0.0004947453319539411,
"loss": 5.6122,
"mean_token_accuracy": 0.15577647536993028,
"num_tokens": 17710076.0,
"step": 8530
},
{
"entropy": 5.8159160137176515,
"epoch": 0.7742198838896952,
"grad_norm": 2.640625,
"learning_rate": 0.0004947383777374462,
"loss": 5.6645,
"mean_token_accuracy": 0.15285882651805877,
"num_tokens": 17721009.0,
"step": 8535
},
{
"entropy": 5.78311185836792,
"epoch": 0.7746734397677794,
"grad_norm": 3.265625,
"learning_rate": 0.0004947314189767058,
"loss": 5.6808,
"mean_token_accuracy": 0.15498824268579484,
"num_tokens": 17731821.0,
"step": 8540
},
{
"entropy": 5.730218362808228,
"epoch": 0.7751269956458636,
"grad_norm": 3.15625,
"learning_rate": 0.0004947244556718635,
"loss": 5.56,
"mean_token_accuracy": 0.16283880770206452,
"num_tokens": 17741425.0,
"step": 8545
},
{
"entropy": 5.806784772872925,
"epoch": 0.7755805515239478,
"grad_norm": 2.390625,
"learning_rate": 0.0004947174878230635,
"loss": 5.6207,
"mean_token_accuracy": 0.15820681154727936,
"num_tokens": 17752381.0,
"step": 8550
},
{
"entropy": 5.77905125617981,
"epoch": 0.7760341074020319,
"grad_norm": 3.421875,
"learning_rate": 0.0004947105154304498,
"loss": 5.6129,
"mean_token_accuracy": 0.15068303048610687,
"num_tokens": 17762768.0,
"step": 8555
},
{
"entropy": 5.791327667236328,
"epoch": 0.7764876632801161,
"grad_norm": 3.125,
"learning_rate": 0.0004947035384941667,
"loss": 5.6467,
"mean_token_accuracy": 0.15347497612237931,
"num_tokens": 17774032.0,
"step": 8560
},
{
"entropy": 5.882103633880615,
"epoch": 0.7769412191582002,
"grad_norm": 3.84375,
"learning_rate": 0.0004946965570143584,
"loss": 5.7596,
"mean_token_accuracy": 0.14765684455633163,
"num_tokens": 17785098.0,
"step": 8565
},
{
"entropy": 5.897441005706787,
"epoch": 0.7773947750362845,
"grad_norm": 2.625,
"learning_rate": 0.0004946895709911695,
"loss": 5.6493,
"mean_token_accuracy": 0.15527293384075164,
"num_tokens": 17795487.0,
"step": 8570
},
{
"entropy": 5.820744800567627,
"epoch": 0.7778483309143687,
"grad_norm": 4.09375,
"learning_rate": 0.0004946825804247442,
"loss": 5.7713,
"mean_token_accuracy": 0.14488444924354554,
"num_tokens": 17806045.0,
"step": 8575
},
{
"entropy": 5.891991424560547,
"epoch": 0.7783018867924528,
"grad_norm": 3.296875,
"learning_rate": 0.000494675585315227,
"loss": 5.7503,
"mean_token_accuracy": 0.14853301644325256,
"num_tokens": 17816948.0,
"step": 8580
},
{
"entropy": 5.856547927856445,
"epoch": 0.778755442670537,
"grad_norm": 4.90625,
"learning_rate": 0.0004946685856627628,
"loss": 5.6418,
"mean_token_accuracy": 0.15747622698545455,
"num_tokens": 17826929.0,
"step": 8585
},
{
"entropy": 5.876050043106079,
"epoch": 0.7792089985486212,
"grad_norm": 3.0625,
"learning_rate": 0.0004946615814674963,
"loss": 5.7598,
"mean_token_accuracy": 0.15428425073623658,
"num_tokens": 17837780.0,
"step": 8590
},
{
"entropy": 5.781055736541748,
"epoch": 0.7796625544267054,
"grad_norm": 4.4375,
"learning_rate": 0.0004946545727295724,
"loss": 5.533,
"mean_token_accuracy": 0.15893158614635466,
"num_tokens": 17847835.0,
"step": 8595
},
{
"entropy": 5.804688358306885,
"epoch": 0.7801161103047896,
"grad_norm": 2.21875,
"learning_rate": 0.0004946475594491359,
"loss": 5.6283,
"mean_token_accuracy": 0.15911484211683274,
"num_tokens": 17858324.0,
"step": 8600
},
{
"entropy": 5.783213043212891,
"epoch": 0.7805696661828737,
"grad_norm": 2.78125,
"learning_rate": 0.0004946405416263319,
"loss": 5.6679,
"mean_token_accuracy": 0.1482814759016037,
"num_tokens": 17868455.0,
"step": 8605
},
{
"entropy": 5.856879234313965,
"epoch": 0.7810232220609579,
"grad_norm": 2.6875,
"learning_rate": 0.0004946335192613054,
"loss": 5.6486,
"mean_token_accuracy": 0.15848930925130844,
"num_tokens": 17879038.0,
"step": 8610
},
{
"entropy": 5.730532503128051,
"epoch": 0.7814767779390421,
"grad_norm": 2.65625,
"learning_rate": 0.0004946264923542019,
"loss": 5.5259,
"mean_token_accuracy": 0.15444619804620743,
"num_tokens": 17888153.0,
"step": 8615
},
{
"entropy": 5.750776576995849,
"epoch": 0.7819303338171263,
"grad_norm": 1.9921875,
"learning_rate": 0.0004946194609051664,
"loss": 5.6526,
"mean_token_accuracy": 0.15871244743466378,
"num_tokens": 17898289.0,
"step": 8620
},
{
"entropy": 5.8846351146698,
"epoch": 0.7823838896952104,
"grad_norm": 3.203125,
"learning_rate": 0.0004946124249143447,
"loss": 5.6435,
"mean_token_accuracy": 0.16097084879875184,
"num_tokens": 17909192.0,
"step": 8625
},
{
"entropy": 5.874616003036499,
"epoch": 0.7828374455732946,
"grad_norm": 3.40625,
"learning_rate": 0.000494605384381882,
"loss": 5.6155,
"mean_token_accuracy": 0.15133094936609268,
"num_tokens": 17918937.0,
"step": 8630
},
{
"entropy": 5.837487697601318,
"epoch": 0.7832910014513788,
"grad_norm": 3.90625,
"learning_rate": 0.0004945983393079239,
"loss": 5.6594,
"mean_token_accuracy": 0.16026734709739685,
"num_tokens": 17929346.0,
"step": 8635
},
{
"entropy": 5.757245635986328,
"epoch": 0.783744557329463,
"grad_norm": 3.828125,
"learning_rate": 0.0004945912896926163,
"loss": 5.6674,
"mean_token_accuracy": 0.15303874313831328,
"num_tokens": 17939479.0,
"step": 8640
},
{
"entropy": 5.875200843811035,
"epoch": 0.7841981132075472,
"grad_norm": 4.1875,
"learning_rate": 0.0004945842355361049,
"loss": 5.741,
"mean_token_accuracy": 0.14947696030139923,
"num_tokens": 17949841.0,
"step": 8645
},
{
"entropy": 5.829590225219727,
"epoch": 0.7846516690856313,
"grad_norm": 3.59375,
"learning_rate": 0.0004945771768385355,
"loss": 5.6726,
"mean_token_accuracy": 0.15517295747995377,
"num_tokens": 17960334.0,
"step": 8650
},
{
"entropy": 5.814268445968628,
"epoch": 0.7851052249637155,
"grad_norm": 4.59375,
"learning_rate": 0.000494570113600054,
"loss": 5.6043,
"mean_token_accuracy": 0.15772475600242614,
"num_tokens": 17970595.0,
"step": 8655
},
{
"entropy": 5.84750337600708,
"epoch": 0.7855587808417998,
"grad_norm": 6.25,
"learning_rate": 0.0004945630458208068,
"loss": 5.7539,
"mean_token_accuracy": 0.14532526433467866,
"num_tokens": 17981301.0,
"step": 8660
},
{
"entropy": 5.788166093826294,
"epoch": 0.7860123367198839,
"grad_norm": 3.203125,
"learning_rate": 0.0004945559735009397,
"loss": 5.6026,
"mean_token_accuracy": 0.1555872455239296,
"num_tokens": 17991262.0,
"step": 8665
},
{
"entropy": 5.872292470932007,
"epoch": 0.7864658925979681,
"grad_norm": 2.890625,
"learning_rate": 0.0004945488966405993,
"loss": 5.6259,
"mean_token_accuracy": 0.15453413426876067,
"num_tokens": 18001212.0,
"step": 8670
},
{
"entropy": 5.79869122505188,
"epoch": 0.7869194484760522,
"grad_norm": 7.03125,
"learning_rate": 0.0004945418152399317,
"loss": 5.6668,
"mean_token_accuracy": 0.15686399787664412,
"num_tokens": 18010432.0,
"step": 8675
},
{
"entropy": 5.775568866729737,
"epoch": 0.7873730043541364,
"grad_norm": 2.609375,
"learning_rate": 0.0004945347292990833,
"loss": 5.6717,
"mean_token_accuracy": 0.14896780550479888,
"num_tokens": 18020902.0,
"step": 8680
},
{
"entropy": 5.856520271301269,
"epoch": 0.7878265602322206,
"grad_norm": 3.625,
"learning_rate": 0.0004945276388182008,
"loss": 5.5571,
"mean_token_accuracy": 0.1509336993098259,
"num_tokens": 18031133.0,
"step": 8685
},
{
"entropy": 5.967869615554809,
"epoch": 0.7882801161103048,
"grad_norm": 3.421875,
"learning_rate": 0.0004945205437974309,
"loss": 5.6777,
"mean_token_accuracy": 0.15307042598724366,
"num_tokens": 18042050.0,
"step": 8690
},
{
"entropy": 5.793651819229126,
"epoch": 0.788733671988389,
"grad_norm": 3.75,
"learning_rate": 0.0004945134442369201,
"loss": 5.6197,
"mean_token_accuracy": 0.1614017441868782,
"num_tokens": 18051642.0,
"step": 8695
},
{
"entropy": 5.821029710769653,
"epoch": 0.7891872278664731,
"grad_norm": 4.28125,
"learning_rate": 0.0004945063401368154,
"loss": 5.6847,
"mean_token_accuracy": 0.14467733949422837,
"num_tokens": 18062180.0,
"step": 8700
},
{
"entropy": 5.86651382446289,
"epoch": 0.7896407837445574,
"grad_norm": 2.71875,
"learning_rate": 0.0004944992314972637,
"loss": 5.6323,
"mean_token_accuracy": 0.15366172045469284,
"num_tokens": 18072302.0,
"step": 8705
},
{
"entropy": 5.817298030853271,
"epoch": 0.7900943396226415,
"grad_norm": 2.5625,
"learning_rate": 0.0004944921183184118,
"loss": 5.6356,
"mean_token_accuracy": 0.15914198458194734,
"num_tokens": 18081824.0,
"step": 8710
},
{
"entropy": 5.774461650848389,
"epoch": 0.7905478955007257,
"grad_norm": 3.53125,
"learning_rate": 0.000494485000600407,
"loss": 5.6877,
"mean_token_accuracy": 0.1560596004128456,
"num_tokens": 18092696.0,
"step": 8715
},
{
"entropy": 5.762605333328247,
"epoch": 0.7910014513788098,
"grad_norm": 2.109375,
"learning_rate": 0.0004944778783433965,
"loss": 5.5936,
"mean_token_accuracy": 0.1632349118590355,
"num_tokens": 18101790.0,
"step": 8720
},
{
"entropy": 5.901923751831054,
"epoch": 0.791455007256894,
"grad_norm": 2.9375,
"learning_rate": 0.0004944707515475276,
"loss": 5.7263,
"mean_token_accuracy": 0.1511840596795082,
"num_tokens": 18111572.0,
"step": 8725
},
{
"entropy": 5.903787612915039,
"epoch": 0.7919085631349783,
"grad_norm": 3.125,
"learning_rate": 0.0004944636202129475,
"loss": 5.7068,
"mean_token_accuracy": 0.16270414739847183,
"num_tokens": 18120650.0,
"step": 8730
},
{
"entropy": 5.817595291137695,
"epoch": 0.7923621190130624,
"grad_norm": 2.59375,
"learning_rate": 0.0004944564843398039,
"loss": 5.5961,
"mean_token_accuracy": 0.16382812857627868,
"num_tokens": 18131055.0,
"step": 8735
},
{
"entropy": 5.860651016235352,
"epoch": 0.7928156748911466,
"grad_norm": 2.15625,
"learning_rate": 0.0004944493439282443,
"loss": 5.6884,
"mean_token_accuracy": 0.1546909973025322,
"num_tokens": 18141423.0,
"step": 8740
},
{
"entropy": 5.7539262771606445,
"epoch": 0.7932692307692307,
"grad_norm": 3.5625,
"learning_rate": 0.0004944421989784162,
"loss": 5.6177,
"mean_token_accuracy": 0.1582576185464859,
"num_tokens": 18152015.0,
"step": 8745
},
{
"entropy": 5.820552015304566,
"epoch": 0.793722786647315,
"grad_norm": 3.5,
"learning_rate": 0.0004944350494904676,
"loss": 5.6973,
"mean_token_accuracy": 0.15252969190478324,
"num_tokens": 18161934.0,
"step": 8750
},
{
"entropy": 5.889200210571289,
"epoch": 0.7941763425253991,
"grad_norm": 3.34375,
"learning_rate": 0.0004944278954645463,
"loss": 5.5921,
"mean_token_accuracy": 0.15543517842888832,
"num_tokens": 18172287.0,
"step": 8755
},
{
"entropy": 5.862867212295532,
"epoch": 0.7946298984034833,
"grad_norm": 3.9375,
"learning_rate": 0.0004944207369008002,
"loss": 5.6721,
"mean_token_accuracy": 0.1599373534321785,
"num_tokens": 18181015.0,
"step": 8760
},
{
"entropy": 5.802975034713745,
"epoch": 0.7950834542815675,
"grad_norm": 2.234375,
"learning_rate": 0.0004944135737993773,
"loss": 5.7508,
"mean_token_accuracy": 0.15021200925111772,
"num_tokens": 18192382.0,
"step": 8765
},
{
"entropy": 5.775304937362671,
"epoch": 0.7955370101596516,
"grad_norm": 3.625,
"learning_rate": 0.0004944064061604258,
"loss": 5.627,
"mean_token_accuracy": 0.16368629038333893,
"num_tokens": 18202597.0,
"step": 8770
},
{
"entropy": 5.934780597686768,
"epoch": 0.7959905660377359,
"grad_norm": 2.125,
"learning_rate": 0.0004943992339840939,
"loss": 5.6766,
"mean_token_accuracy": 0.1494644358754158,
"num_tokens": 18213756.0,
"step": 8775
},
{
"entropy": 5.783631753921509,
"epoch": 0.79644412191582,
"grad_norm": 2.96875,
"learning_rate": 0.00049439205727053,
"loss": 5.6229,
"mean_token_accuracy": 0.14898163676261902,
"num_tokens": 18223669.0,
"step": 8780
},
{
"entropy": 5.778256940841675,
"epoch": 0.7968976777939042,
"grad_norm": 2.109375,
"learning_rate": 0.0004943848760198823,
"loss": 5.6404,
"mean_token_accuracy": 0.15757868587970733,
"num_tokens": 18234884.0,
"step": 8785
},
{
"entropy": 5.814851236343384,
"epoch": 0.7973512336719883,
"grad_norm": 2.46875,
"learning_rate": 0.0004943776902322995,
"loss": 5.5969,
"mean_token_accuracy": 0.16033483147621155,
"num_tokens": 18245155.0,
"step": 8790
},
{
"entropy": 5.731310796737671,
"epoch": 0.7978047895500726,
"grad_norm": 2.421875,
"learning_rate": 0.0004943704999079302,
"loss": 5.5437,
"mean_token_accuracy": 0.15940094143152236,
"num_tokens": 18254686.0,
"step": 8795
},
{
"entropy": 5.79235520362854,
"epoch": 0.7982583454281568,
"grad_norm": 2.03125,
"learning_rate": 0.0004943633050469231,
"loss": 5.5433,
"mean_token_accuracy": 0.15957473516464232,
"num_tokens": 18265473.0,
"step": 8800
},
{
"entropy": 5.91292028427124,
"epoch": 0.7987119013062409,
"grad_norm": 3.28125,
"learning_rate": 0.0004943561056494269,
"loss": 5.7515,
"mean_token_accuracy": 0.15094480961561202,
"num_tokens": 18276024.0,
"step": 8805
},
{
"entropy": 5.838796234130859,
"epoch": 0.7991654571843251,
"grad_norm": 2.234375,
"learning_rate": 0.0004943489017155906,
"loss": 5.6977,
"mean_token_accuracy": 0.14398460388183593,
"num_tokens": 18287261.0,
"step": 8810
},
{
"entropy": 5.832255887985229,
"epoch": 0.7996190130624092,
"grad_norm": 1.90625,
"learning_rate": 0.0004943416932455631,
"loss": 5.7026,
"mean_token_accuracy": 0.15213215947151185,
"num_tokens": 18299374.0,
"step": 8815
},
{
"entropy": 5.7932600498199465,
"epoch": 0.8000725689404935,
"grad_norm": 2.25,
"learning_rate": 0.0004943344802394935,
"loss": 5.6894,
"mean_token_accuracy": 0.1507257863879204,
"num_tokens": 18309344.0,
"step": 8820
},
{
"entropy": 5.860401153564453,
"epoch": 0.8005261248185777,
"grad_norm": 3.828125,
"learning_rate": 0.0004943272626975309,
"loss": 5.6701,
"mean_token_accuracy": 0.15301405191421508,
"num_tokens": 18319778.0,
"step": 8825
},
{
"entropy": 5.827208948135376,
"epoch": 0.8009796806966618,
"grad_norm": 2.875,
"learning_rate": 0.0004943200406198246,
"loss": 5.6558,
"mean_token_accuracy": 0.1550200790166855,
"num_tokens": 18329787.0,
"step": 8830
},
{
"entropy": 5.767522525787354,
"epoch": 0.801433236574746,
"grad_norm": 2.609375,
"learning_rate": 0.000494312814006524,
"loss": 5.6147,
"mean_token_accuracy": 0.16383739709854125,
"num_tokens": 18340670.0,
"step": 8835
},
{
"entropy": 5.788567638397216,
"epoch": 0.8018867924528302,
"grad_norm": 2.546875,
"learning_rate": 0.0004943055828577784,
"loss": 5.6304,
"mean_token_accuracy": 0.14750366061925888,
"num_tokens": 18352484.0,
"step": 8840
},
{
"entropy": 5.787442207336426,
"epoch": 0.8023403483309144,
"grad_norm": 2.796875,
"learning_rate": 0.0004942983471737376,
"loss": 5.5716,
"mean_token_accuracy": 0.1558992326259613,
"num_tokens": 18363103.0,
"step": 8845
},
{
"entropy": 5.772452688217163,
"epoch": 0.8027939042089985,
"grad_norm": 2.40625,
"learning_rate": 0.0004942911069545511,
"loss": 5.6234,
"mean_token_accuracy": 0.1594387948513031,
"num_tokens": 18373925.0,
"step": 8850
},
{
"entropy": 5.807936382293701,
"epoch": 0.8032474600870827,
"grad_norm": 4.21875,
"learning_rate": 0.0004942838622003686,
"loss": 5.701,
"mean_token_accuracy": 0.1509519785642624,
"num_tokens": 18384081.0,
"step": 8855
},
{
"entropy": 5.827019786834716,
"epoch": 0.8037010159651669,
"grad_norm": 4.28125,
"learning_rate": 0.0004942766129113399,
"loss": 5.5911,
"mean_token_accuracy": 0.15893699526786803,
"num_tokens": 18392510.0,
"step": 8860
},
{
"entropy": 5.900896501541138,
"epoch": 0.8041545718432511,
"grad_norm": 2.359375,
"learning_rate": 0.0004942693590876151,
"loss": 5.7146,
"mean_token_accuracy": 0.1524460308253765,
"num_tokens": 18402809.0,
"step": 8865
},
{
"entropy": 5.886493778228759,
"epoch": 0.8046081277213353,
"grad_norm": 2.578125,
"learning_rate": 0.000494262100729344,
"loss": 5.7361,
"mean_token_accuracy": 0.1542189508676529,
"num_tokens": 18413342.0,
"step": 8870
},
{
"entropy": 5.874359846115112,
"epoch": 0.8050616835994194,
"grad_norm": 3.109375,
"learning_rate": 0.0004942548378366768,
"loss": 5.7501,
"mean_token_accuracy": 0.15199500173330308,
"num_tokens": 18424735.0,
"step": 8875
},
{
"entropy": 5.829610395431518,
"epoch": 0.8055152394775036,
"grad_norm": 2.015625,
"learning_rate": 0.0004942475704097635,
"loss": 5.6184,
"mean_token_accuracy": 0.15787999778985978,
"num_tokens": 18434797.0,
"step": 8880
},
{
"entropy": 5.796105766296387,
"epoch": 0.8059687953555879,
"grad_norm": 6.125,
"learning_rate": 0.0004942402984487548,
"loss": 5.6877,
"mean_token_accuracy": 0.14930816888809204,
"num_tokens": 18445456.0,
"step": 8885
},
{
"entropy": 5.800718307495117,
"epoch": 0.806422351233672,
"grad_norm": 3.703125,
"learning_rate": 0.0004942330219538008,
"loss": 5.6542,
"mean_token_accuracy": 0.14945662915706634,
"num_tokens": 18456728.0,
"step": 8890
},
{
"entropy": 6.033436584472656,
"epoch": 0.8068759071117562,
"grad_norm": 2.78125,
"learning_rate": 0.000494225740925052,
"loss": 5.7818,
"mean_token_accuracy": 0.14793658256530762,
"num_tokens": 18468813.0,
"step": 8895
},
{
"entropy": 5.932030200958252,
"epoch": 0.8073294629898403,
"grad_norm": 3.421875,
"learning_rate": 0.0004942184553626591,
"loss": 5.6778,
"mean_token_accuracy": 0.154929731041193,
"num_tokens": 18477912.0,
"step": 8900
},
{
"entropy": 5.841903495788574,
"epoch": 0.8077830188679245,
"grad_norm": 3.078125,
"learning_rate": 0.0004942111652667726,
"loss": 5.6177,
"mean_token_accuracy": 0.15868613123893738,
"num_tokens": 18488115.0,
"step": 8905
},
{
"entropy": 5.836884880065918,
"epoch": 0.8082365747460087,
"grad_norm": 4.3125,
"learning_rate": 0.0004942038706375433,
"loss": 5.6752,
"mean_token_accuracy": 0.15271092504262923,
"num_tokens": 18498853.0,
"step": 8910
},
{
"entropy": 5.817429161071777,
"epoch": 0.8086901306240929,
"grad_norm": 3.328125,
"learning_rate": 0.0004941965714751222,
"loss": 5.7144,
"mean_token_accuracy": 0.1488034538924694,
"num_tokens": 18510047.0,
"step": 8915
},
{
"entropy": 5.796209287643433,
"epoch": 0.809143686502177,
"grad_norm": 2.3125,
"learning_rate": 0.0004941892677796601,
"loss": 5.6739,
"mean_token_accuracy": 0.1562424249947071,
"num_tokens": 18519673.0,
"step": 8920
},
{
"entropy": 5.834105730056763,
"epoch": 0.8095972423802612,
"grad_norm": 4.8125,
"learning_rate": 0.0004941819595513081,
"loss": 5.7058,
"mean_token_accuracy": 0.15310655534267426,
"num_tokens": 18530090.0,
"step": 8925
},
{
"entropy": 5.897770929336548,
"epoch": 0.8100507982583455,
"grad_norm": 4.09375,
"learning_rate": 0.0004941746467902173,
"loss": 5.7393,
"mean_token_accuracy": 0.14204500168561934,
"num_tokens": 18541862.0,
"step": 8930
},
{
"entropy": 5.893235349655152,
"epoch": 0.8105043541364296,
"grad_norm": 2.53125,
"learning_rate": 0.000494167329496539,
"loss": 5.713,
"mean_token_accuracy": 0.14384791254997253,
"num_tokens": 18551624.0,
"step": 8935
},
{
"entropy": 5.793308925628662,
"epoch": 0.8109579100145138,
"grad_norm": 1.890625,
"learning_rate": 0.0004941600076704244,
"loss": 5.6614,
"mean_token_accuracy": 0.15391065627336503,
"num_tokens": 18561784.0,
"step": 8940
},
{
"entropy": 5.871464252471924,
"epoch": 0.8114114658925979,
"grad_norm": 3.53125,
"learning_rate": 0.0004941526813120251,
"loss": 5.7296,
"mean_token_accuracy": 0.15403200089931487,
"num_tokens": 18572110.0,
"step": 8945
},
{
"entropy": 5.898204517364502,
"epoch": 0.8118650217706821,
"grad_norm": 2.265625,
"learning_rate": 0.0004941453504214924,
"loss": 5.688,
"mean_token_accuracy": 0.14993584677577018,
"num_tokens": 18582012.0,
"step": 8950
},
{
"entropy": 5.844607734680176,
"epoch": 0.8123185776487664,
"grad_norm": 2.171875,
"learning_rate": 0.0004941380149989781,
"loss": 5.5412,
"mean_token_accuracy": 0.16359425187110901,
"num_tokens": 18592850.0,
"step": 8955
},
{
"entropy": 5.773560285568237,
"epoch": 0.8127721335268505,
"grad_norm": 2.5,
"learning_rate": 0.0004941306750446336,
"loss": 5.6954,
"mean_token_accuracy": 0.1542632609605789,
"num_tokens": 18602590.0,
"step": 8960
},
{
"entropy": 5.790944147109985,
"epoch": 0.8132256894049347,
"grad_norm": 3.3125,
"learning_rate": 0.000494123330558611,
"loss": 5.5988,
"mean_token_accuracy": 0.15769584104418755,
"num_tokens": 18612607.0,
"step": 8965
},
{
"entropy": 5.793989849090576,
"epoch": 0.8136792452830188,
"grad_norm": 4.3125,
"learning_rate": 0.0004941159815410622,
"loss": 5.5872,
"mean_token_accuracy": 0.15930003374814988,
"num_tokens": 18622936.0,
"step": 8970
},
{
"entropy": 5.842741632461548,
"epoch": 0.8141328011611031,
"grad_norm": 2.8125,
"learning_rate": 0.000494108627992139,
"loss": 5.7771,
"mean_token_accuracy": 0.1470210388302803,
"num_tokens": 18633997.0,
"step": 8975
},
{
"entropy": 5.867417573928833,
"epoch": 0.8145863570391872,
"grad_norm": 3.671875,
"learning_rate": 0.0004941012699119936,
"loss": 5.5763,
"mean_token_accuracy": 0.1638900578022003,
"num_tokens": 18644347.0,
"step": 8980
},
{
"entropy": 5.838050317764282,
"epoch": 0.8150399129172714,
"grad_norm": 3.0625,
"learning_rate": 0.0004940939073007779,
"loss": 5.6593,
"mean_token_accuracy": 0.15335242450237274,
"num_tokens": 18654989.0,
"step": 8985
},
{
"entropy": 5.7650409698486325,
"epoch": 0.8154934687953556,
"grad_norm": 2.34375,
"learning_rate": 0.0004940865401586446,
"loss": 5.6335,
"mean_token_accuracy": 0.15253210663795472,
"num_tokens": 18665526.0,
"step": 8990
},
{
"entropy": 5.834532165527344,
"epoch": 0.8159470246734397,
"grad_norm": 2.546875,
"learning_rate": 0.0004940791684857458,
"loss": 5.6554,
"mean_token_accuracy": 0.15907703042030336,
"num_tokens": 18676226.0,
"step": 8995
},
{
"entropy": 5.847899627685547,
"epoch": 0.816400580551524,
"grad_norm": 4.53125,
"learning_rate": 0.0004940717922822339,
"loss": 5.5808,
"mean_token_accuracy": 0.15463961511850358,
"num_tokens": 18686499.0,
"step": 9000
},
{
"epoch": 0.816400580551524,
"eval_entropy": 5.683191412606391,
"eval_loss": 5.684431552886963,
"eval_mean_token_accuracy": 0.15958393904131868,
"eval_num_tokens": 18686499.0,
"eval_runtime": 26.6876,
"eval_samples_per_second": 1324.96,
"eval_steps_per_second": 165.62,
"step": 9000
},
{
"entropy": 5.843375539779663,
"epoch": 0.8168541364296081,
"grad_norm": 3.78125,
"learning_rate": 0.0004940644115482615,
"loss": 5.6417,
"mean_token_accuracy": 0.15436051040887833,
"num_tokens": 18696543.0,
"step": 9005
},
{
"entropy": 5.793101692199707,
"epoch": 0.8173076923076923,
"grad_norm": 3.15625,
"learning_rate": 0.0004940570262839812,
"loss": 5.6896,
"mean_token_accuracy": 0.15068877935409547,
"num_tokens": 18707029.0,
"step": 9010
},
{
"entropy": 5.8139701843261715,
"epoch": 0.8177612481857764,
"grad_norm": 2.625,
"learning_rate": 0.0004940496364895457,
"loss": 5.6789,
"mean_token_accuracy": 0.15451287627220153,
"num_tokens": 18716937.0,
"step": 9015
},
{
"entropy": 5.897455024719238,
"epoch": 0.8182148040638607,
"grad_norm": 4.25,
"learning_rate": 0.0004940422421651082,
"loss": 5.6724,
"mean_token_accuracy": 0.15182604044675826,
"num_tokens": 18727478.0,
"step": 9020
},
{
"entropy": 5.864028549194336,
"epoch": 0.8186683599419449,
"grad_norm": 3.25,
"learning_rate": 0.0004940348433108211,
"loss": 5.5951,
"mean_token_accuracy": 0.14794762730598449,
"num_tokens": 18737339.0,
"step": 9025
},
{
"entropy": 5.851188659667969,
"epoch": 0.819121915820029,
"grad_norm": 2.96875,
"learning_rate": 0.0004940274399268377,
"loss": 5.7162,
"mean_token_accuracy": 0.15362059324979782,
"num_tokens": 18747810.0,
"step": 9030
},
{
"entropy": 5.7702428817749025,
"epoch": 0.8195754716981132,
"grad_norm": 2.171875,
"learning_rate": 0.0004940200320133109,
"loss": 5.6525,
"mean_token_accuracy": 0.15094198137521744,
"num_tokens": 18757564.0,
"step": 9035
},
{
"entropy": 5.884185075759888,
"epoch": 0.8200290275761973,
"grad_norm": 6.25,
"learning_rate": 0.0004940126195703941,
"loss": 5.7132,
"mean_token_accuracy": 0.1497935563325882,
"num_tokens": 18768196.0,
"step": 9040
},
{
"entropy": 5.898662471771241,
"epoch": 0.8204825834542816,
"grad_norm": 3.65625,
"learning_rate": 0.0004940052025982405,
"loss": 5.6839,
"mean_token_accuracy": 0.16041729897260665,
"num_tokens": 18778350.0,
"step": 9045
},
{
"entropy": 5.814507389068604,
"epoch": 0.8209361393323658,
"grad_norm": 2.609375,
"learning_rate": 0.0004939977810970035,
"loss": 5.6194,
"mean_token_accuracy": 0.14908331036567687,
"num_tokens": 18789825.0,
"step": 9050
},
{
"entropy": 5.8118775367736815,
"epoch": 0.8213896952104499,
"grad_norm": 3.390625,
"learning_rate": 0.0004939903550668366,
"loss": 5.6354,
"mean_token_accuracy": 0.1541058212518692,
"num_tokens": 18799793.0,
"step": 9055
},
{
"entropy": 5.806805038452149,
"epoch": 0.8218432510885341,
"grad_norm": 5.65625,
"learning_rate": 0.0004939829245078934,
"loss": 5.6747,
"mean_token_accuracy": 0.15679603368043898,
"num_tokens": 18809771.0,
"step": 9060
},
{
"entropy": 5.904952669143677,
"epoch": 0.8222968069666183,
"grad_norm": 2.6875,
"learning_rate": 0.0004939754894203275,
"loss": 5.6922,
"mean_token_accuracy": 0.1526167705655098,
"num_tokens": 18820529.0,
"step": 9065
},
{
"entropy": 5.800691223144531,
"epoch": 0.8227503628447025,
"grad_norm": 2.640625,
"learning_rate": 0.0004939680498042926,
"loss": 5.607,
"mean_token_accuracy": 0.15987065434455872,
"num_tokens": 18830797.0,
"step": 9070
},
{
"entropy": 5.785319471359253,
"epoch": 0.8232039187227866,
"grad_norm": 2.671875,
"learning_rate": 0.0004939606056599428,
"loss": 5.6072,
"mean_token_accuracy": 0.15448503717780113,
"num_tokens": 18840526.0,
"step": 9075
},
{
"entropy": 5.772441053390503,
"epoch": 0.8236574746008708,
"grad_norm": 2.9375,
"learning_rate": 0.0004939531569874316,
"loss": 5.6298,
"mean_token_accuracy": 0.15375131517648696,
"num_tokens": 18851201.0,
"step": 9080
},
{
"entropy": 5.755581855773926,
"epoch": 0.824111030478955,
"grad_norm": 2.21875,
"learning_rate": 0.0004939457037869135,
"loss": 5.469,
"mean_token_accuracy": 0.16889793276786805,
"num_tokens": 18861676.0,
"step": 9085
},
{
"entropy": 5.7853155612945555,
"epoch": 0.8245645863570392,
"grad_norm": 3.28125,
"learning_rate": 0.0004939382460585424,
"loss": 5.5963,
"mean_token_accuracy": 0.15494900494813918,
"num_tokens": 18872669.0,
"step": 9090
},
{
"entropy": 5.752126216888428,
"epoch": 0.8250181422351234,
"grad_norm": 2.296875,
"learning_rate": 0.0004939307838024726,
"loss": 5.6265,
"mean_token_accuracy": 0.15291281417012215,
"num_tokens": 18882130.0,
"step": 9095
},
{
"entropy": 5.73763689994812,
"epoch": 0.8254716981132075,
"grad_norm": 2.890625,
"learning_rate": 0.0004939233170188585,
"loss": 5.6226,
"mean_token_accuracy": 0.15411534905433655,
"num_tokens": 18892457.0,
"step": 9100
},
{
"entropy": 5.894270277023315,
"epoch": 0.8259252539912917,
"grad_norm": 2.5625,
"learning_rate": 0.0004939158457078543,
"loss": 5.6544,
"mean_token_accuracy": 0.15148564502596856,
"num_tokens": 18903152.0,
"step": 9105
},
{
"entropy": 5.817868185043335,
"epoch": 0.826378809869376,
"grad_norm": 3.671875,
"learning_rate": 0.0004939083698696146,
"loss": 5.6276,
"mean_token_accuracy": 0.15088266208767892,
"num_tokens": 18913740.0,
"step": 9110
},
{
"entropy": 5.8163670063018795,
"epoch": 0.8268323657474601,
"grad_norm": 3.9375,
"learning_rate": 0.000493900889504294,
"loss": 5.6636,
"mean_token_accuracy": 0.15680994540452958,
"num_tokens": 18923501.0,
"step": 9115
},
{
"entropy": 5.822394514083863,
"epoch": 0.8272859216255443,
"grad_norm": 2.96875,
"learning_rate": 0.0004938934046120473,
"loss": 5.6693,
"mean_token_accuracy": 0.15569230169057846,
"num_tokens": 18933255.0,
"step": 9120
},
{
"entropy": 5.795421028137207,
"epoch": 0.8277394775036284,
"grad_norm": 3.25,
"learning_rate": 0.0004938859151930293,
"loss": 5.6055,
"mean_token_accuracy": 0.15849126726388932,
"num_tokens": 18943649.0,
"step": 9125
},
{
"entropy": 5.788185548782349,
"epoch": 0.8281930333817126,
"grad_norm": 3.203125,
"learning_rate": 0.0004938784212473947,
"loss": 5.5362,
"mean_token_accuracy": 0.1615826740860939,
"num_tokens": 18953187.0,
"step": 9130
},
{
"entropy": 5.836130285263062,
"epoch": 0.8286465892597968,
"grad_norm": 15.75,
"learning_rate": 0.0004938709227752985,
"loss": 5.6835,
"mean_token_accuracy": 0.15411966443061828,
"num_tokens": 18963015.0,
"step": 9135
},
{
"entropy": 5.836290073394776,
"epoch": 0.829100145137881,
"grad_norm": 3.21875,
"learning_rate": 0.000493863419776896,
"loss": 5.685,
"mean_token_accuracy": 0.14958571940660476,
"num_tokens": 18974184.0,
"step": 9140
},
{
"entropy": 5.796621084213257,
"epoch": 0.8295537010159652,
"grad_norm": 2.21875,
"learning_rate": 0.0004938559122523421,
"loss": 5.6169,
"mean_token_accuracy": 0.1585509866476059,
"num_tokens": 18985036.0,
"step": 9145
},
{
"entropy": 5.808863782882691,
"epoch": 0.8300072568940493,
"grad_norm": 2.828125,
"learning_rate": 0.0004938484002017921,
"loss": 5.7302,
"mean_token_accuracy": 0.1534656725823879,
"num_tokens": 18995237.0,
"step": 9150
},
{
"entropy": 5.818014335632324,
"epoch": 0.8304608127721336,
"grad_norm": 2.390625,
"learning_rate": 0.0004938408836254015,
"loss": 5.73,
"mean_token_accuracy": 0.14893487244844436,
"num_tokens": 19006272.0,
"step": 9155
},
{
"entropy": 5.842798709869385,
"epoch": 0.8309143686502177,
"grad_norm": 3.15625,
"learning_rate": 0.0004938333625233258,
"loss": 5.6512,
"mean_token_accuracy": 0.1608937107026577,
"num_tokens": 19016728.0,
"step": 9160
},
{
"entropy": 5.886384630203247,
"epoch": 0.8313679245283019,
"grad_norm": 4.4375,
"learning_rate": 0.0004938258368957202,
"loss": 5.6925,
"mean_token_accuracy": 0.14620290994644164,
"num_tokens": 19028024.0,
"step": 9165
},
{
"entropy": 5.8492639541625975,
"epoch": 0.831821480406386,
"grad_norm": 3.640625,
"learning_rate": 0.0004938183067427406,
"loss": 5.628,
"mean_token_accuracy": 0.1614862248301506,
"num_tokens": 19038218.0,
"step": 9170
},
{
"entropy": 5.802521514892578,
"epoch": 0.8322750362844702,
"grad_norm": 5.125,
"learning_rate": 0.0004938107720645426,
"loss": 5.675,
"mean_token_accuracy": 0.15266878306865692,
"num_tokens": 19048574.0,
"step": 9175
},
{
"entropy": 5.856499767303466,
"epoch": 0.8327285921625545,
"grad_norm": 3.0625,
"learning_rate": 0.0004938032328612821,
"loss": 5.6296,
"mean_token_accuracy": 0.15636837631464004,
"num_tokens": 19057721.0,
"step": 9180
},
{
"entropy": 5.860321521759033,
"epoch": 0.8331821480406386,
"grad_norm": 3.59375,
"learning_rate": 0.000493795689133115,
"loss": 5.6015,
"mean_token_accuracy": 0.1484677277505398,
"num_tokens": 19067582.0,
"step": 9185
},
{
"entropy": 5.790124559402466,
"epoch": 0.8336357039187228,
"grad_norm": 4.4375,
"learning_rate": 0.0004937881408801974,
"loss": 5.5827,
"mean_token_accuracy": 0.15845344811677933,
"num_tokens": 19077896.0,
"step": 9190
},
{
"entropy": 5.808318042755127,
"epoch": 0.8340892597968069,
"grad_norm": 2.640625,
"learning_rate": 0.000493780588102685,
"loss": 5.6812,
"mean_token_accuracy": 0.15800550431013108,
"num_tokens": 19088082.0,
"step": 9195
},
{
"entropy": 5.863030862808228,
"epoch": 0.8345428156748912,
"grad_norm": 3.0,
"learning_rate": 0.0004937730308007346,
"loss": 5.7038,
"mean_token_accuracy": 0.151945798099041,
"num_tokens": 19097176.0,
"step": 9200
},
{
"entropy": 5.761180305480957,
"epoch": 0.8349963715529753,
"grad_norm": 2.640625,
"learning_rate": 0.0004937654689745021,
"loss": 5.6062,
"mean_token_accuracy": 0.15972303003072738,
"num_tokens": 19107835.0,
"step": 9205
},
{
"entropy": 5.709291219711304,
"epoch": 0.8354499274310595,
"grad_norm": 4.96875,
"learning_rate": 0.0004937579026241439,
"loss": 5.6228,
"mean_token_accuracy": 0.1632474735379219,
"num_tokens": 19118028.0,
"step": 9210
},
{
"entropy": 5.85373101234436,
"epoch": 0.8359034833091437,
"grad_norm": 13.3125,
"learning_rate": 0.0004937503317498165,
"loss": 5.6329,
"mean_token_accuracy": 0.15670134127140045,
"num_tokens": 19129223.0,
"step": 9215
},
{
"entropy": 5.874120140075684,
"epoch": 0.8363570391872278,
"grad_norm": 2.609375,
"learning_rate": 0.0004937427563516765,
"loss": 5.7367,
"mean_token_accuracy": 0.15082149356603622,
"num_tokens": 19140048.0,
"step": 9220
},
{
"entropy": 5.816678190231324,
"epoch": 0.8368105950653121,
"grad_norm": 4.53125,
"learning_rate": 0.0004937351764298805,
"loss": 5.7069,
"mean_token_accuracy": 0.14715564996004105,
"num_tokens": 19150853.0,
"step": 9225
},
{
"entropy": 5.82272515296936,
"epoch": 0.8372641509433962,
"grad_norm": 7.46875,
"learning_rate": 0.0004937275919845855,
"loss": 5.681,
"mean_token_accuracy": 0.15349098891019822,
"num_tokens": 19161645.0,
"step": 9230
},
{
"entropy": 5.871696996688843,
"epoch": 0.8377177068214804,
"grad_norm": 3.453125,
"learning_rate": 0.0004937200030159481,
"loss": 5.7667,
"mean_token_accuracy": 0.14946603924036025,
"num_tokens": 19173276.0,
"step": 9235
},
{
"entropy": 5.915431976318359,
"epoch": 0.8381712626995645,
"grad_norm": 3.203125,
"learning_rate": 0.0004937124095241252,
"loss": 5.6702,
"mean_token_accuracy": 0.1593783065676689,
"num_tokens": 19182949.0,
"step": 9240
},
{
"entropy": 5.884314680099488,
"epoch": 0.8386248185776488,
"grad_norm": 5.90625,
"learning_rate": 0.0004937048115092739,
"loss": 5.651,
"mean_token_accuracy": 0.15436612367630004,
"num_tokens": 19193504.0,
"step": 9245
},
{
"entropy": 5.835222673416138,
"epoch": 0.839078374455733,
"grad_norm": 2.6875,
"learning_rate": 0.0004936972089715515,
"loss": 5.7284,
"mean_token_accuracy": 0.14974120259284973,
"num_tokens": 19203406.0,
"step": 9250
},
{
"entropy": 5.815620374679566,
"epoch": 0.8395319303338171,
"grad_norm": 4.03125,
"learning_rate": 0.000493689601911115,
"loss": 5.6596,
"mean_token_accuracy": 0.15398818552494048,
"num_tokens": 19214325.0,
"step": 9255
},
{
"entropy": 5.845991325378418,
"epoch": 0.8399854862119013,
"grad_norm": 2.484375,
"learning_rate": 0.0004936819903281219,
"loss": 5.7333,
"mean_token_accuracy": 0.14285178184509278,
"num_tokens": 19226466.0,
"step": 9260
},
{
"entropy": 5.934034585952759,
"epoch": 0.8404390420899854,
"grad_norm": 2.359375,
"learning_rate": 0.0004936743742227295,
"loss": 5.7372,
"mean_token_accuracy": 0.15123442113399505,
"num_tokens": 19237032.0,
"step": 9265
},
{
"entropy": 5.916327333450317,
"epoch": 0.8408925979680697,
"grad_norm": 3.03125,
"learning_rate": 0.0004936667535950952,
"loss": 5.7308,
"mean_token_accuracy": 0.14760565385222435,
"num_tokens": 19248009.0,
"step": 9270
},
{
"entropy": 5.850819635391235,
"epoch": 0.8413461538461539,
"grad_norm": 4.1875,
"learning_rate": 0.0004936591284453768,
"loss": 5.7935,
"mean_token_accuracy": 0.14887545853853226,
"num_tokens": 19259363.0,
"step": 9275
},
{
"entropy": 5.817532968521118,
"epoch": 0.841799709724238,
"grad_norm": 3.0625,
"learning_rate": 0.0004936514987737319,
"loss": 5.6921,
"mean_token_accuracy": 0.15453299582004548,
"num_tokens": 19269191.0,
"step": 9280
},
{
"entropy": 5.868520593643188,
"epoch": 0.8422532656023222,
"grad_norm": 2.953125,
"learning_rate": 0.0004936438645803182,
"loss": 5.6331,
"mean_token_accuracy": 0.15244528353214265,
"num_tokens": 19280278.0,
"step": 9285
},
{
"entropy": 5.887366008758545,
"epoch": 0.8427068214804064,
"grad_norm": 2.84375,
"learning_rate": 0.0004936362258652938,
"loss": 5.6228,
"mean_token_accuracy": 0.15615448504686355,
"num_tokens": 19290130.0,
"step": 9290
},
{
"entropy": 5.975782442092895,
"epoch": 0.8431603773584906,
"grad_norm": 2.828125,
"learning_rate": 0.0004936285826288165,
"loss": 5.7635,
"mean_token_accuracy": 0.15192786306142808,
"num_tokens": 19301669.0,
"step": 9295
},
{
"entropy": 5.813149261474609,
"epoch": 0.8436139332365747,
"grad_norm": 3.0,
"learning_rate": 0.0004936209348710444,
"loss": 5.6288,
"mean_token_accuracy": 0.15411185771226882,
"num_tokens": 19312490.0,
"step": 9300
},
{
"entropy": 5.775185203552246,
"epoch": 0.8440674891146589,
"grad_norm": 3.171875,
"learning_rate": 0.0004936132825921355,
"loss": 5.6643,
"mean_token_accuracy": 0.1526573419570923,
"num_tokens": 19321743.0,
"step": 9305
},
{
"entropy": 5.795629644393921,
"epoch": 0.844521044992743,
"grad_norm": 2.375,
"learning_rate": 0.0004936056257922484,
"loss": 5.5721,
"mean_token_accuracy": 0.1604461818933487,
"num_tokens": 19332149.0,
"step": 9310
},
{
"entropy": 5.8589396476745605,
"epoch": 0.8449746008708273,
"grad_norm": 2.84375,
"learning_rate": 0.0004935979644715412,
"loss": 5.6545,
"mean_token_accuracy": 0.15000626295804978,
"num_tokens": 19341655.0,
"step": 9315
},
{
"entropy": 5.888137102127075,
"epoch": 0.8454281567489115,
"grad_norm": 3.0,
"learning_rate": 0.0004935902986301724,
"loss": 5.7102,
"mean_token_accuracy": 0.149715293943882,
"num_tokens": 19351404.0,
"step": 9320
},
{
"entropy": 5.82078709602356,
"epoch": 0.8458817126269956,
"grad_norm": 3.34375,
"learning_rate": 0.0004935826282683005,
"loss": 5.7112,
"mean_token_accuracy": 0.1484207406640053,
"num_tokens": 19361788.0,
"step": 9325
},
{
"entropy": 5.812763786315918,
"epoch": 0.8463352685050798,
"grad_norm": 2.375,
"learning_rate": 0.0004935749533860842,
"loss": 5.6817,
"mean_token_accuracy": 0.15633453130722047,
"num_tokens": 19372902.0,
"step": 9330
},
{
"entropy": 5.739981412887573,
"epoch": 0.846788824383164,
"grad_norm": 2.3125,
"learning_rate": 0.000493567273983682,
"loss": 5.5744,
"mean_token_accuracy": 0.1500100888311863,
"num_tokens": 19382181.0,
"step": 9335
},
{
"entropy": 5.848348140716553,
"epoch": 0.8472423802612482,
"grad_norm": 12.0,
"learning_rate": 0.000493559590061253,
"loss": 5.6457,
"mean_token_accuracy": 0.15094297528266906,
"num_tokens": 19393692.0,
"step": 9340
},
{
"entropy": 5.887370777130127,
"epoch": 0.8476959361393324,
"grad_norm": 6.15625,
"learning_rate": 0.0004935519016189559,
"loss": 5.7067,
"mean_token_accuracy": 0.15922312512993814,
"num_tokens": 19404587.0,
"step": 9345
},
{
"entropy": 5.892665004730224,
"epoch": 0.8481494920174165,
"grad_norm": 3.375,
"learning_rate": 0.0004935442086569498,
"loss": 5.6613,
"mean_token_accuracy": 0.15366343706846236,
"num_tokens": 19415296.0,
"step": 9350
},
{
"entropy": 5.887033700942993,
"epoch": 0.8486030478955007,
"grad_norm": 2.484375,
"learning_rate": 0.0004935365111753939,
"loss": 5.7199,
"mean_token_accuracy": 0.15258205384016038,
"num_tokens": 19426119.0,
"step": 9355
},
{
"entropy": 5.8707421779632565,
"epoch": 0.8490566037735849,
"grad_norm": 2.328125,
"learning_rate": 0.0004935288091744472,
"loss": 5.6975,
"mean_token_accuracy": 0.15224236696958543,
"num_tokens": 19436828.0,
"step": 9360
},
{
"entropy": 5.894240617752075,
"epoch": 0.8495101596516691,
"grad_norm": 2.25,
"learning_rate": 0.000493521102654269,
"loss": 5.6845,
"mean_token_accuracy": 0.14839867651462554,
"num_tokens": 19446583.0,
"step": 9365
},
{
"entropy": 5.792936849594116,
"epoch": 0.8499637155297532,
"grad_norm": 3.03125,
"learning_rate": 0.0004935133916150187,
"loss": 5.565,
"mean_token_accuracy": 0.15941736847162247,
"num_tokens": 19456757.0,
"step": 9370
},
{
"entropy": 5.754615211486817,
"epoch": 0.8504172714078374,
"grad_norm": 2.109375,
"learning_rate": 0.0004935056760568558,
"loss": 5.6343,
"mean_token_accuracy": 0.1472718082368374,
"num_tokens": 19467624.0,
"step": 9375
},
{
"entropy": 5.768962860107422,
"epoch": 0.8508708272859217,
"grad_norm": 3.09375,
"learning_rate": 0.0004934979559799399,
"loss": 5.6101,
"mean_token_accuracy": 0.15519556999206544,
"num_tokens": 19477759.0,
"step": 9380
},
{
"entropy": 5.826801013946533,
"epoch": 0.8513243831640058,
"grad_norm": 2.78125,
"learning_rate": 0.0004934902313844306,
"loss": 5.6141,
"mean_token_accuracy": 0.15790425688028337,
"num_tokens": 19488877.0,
"step": 9385
},
{
"entropy": 5.80237717628479,
"epoch": 0.85177793904209,
"grad_norm": 3.21875,
"learning_rate": 0.0004934825022704875,
"loss": 5.6248,
"mean_token_accuracy": 0.15270906090736389,
"num_tokens": 19499170.0,
"step": 9390
},
{
"entropy": 5.8249390602111815,
"epoch": 0.8522314949201741,
"grad_norm": 3.25,
"learning_rate": 0.0004934747686382707,
"loss": 5.6502,
"mean_token_accuracy": 0.14975784569978715,
"num_tokens": 19509944.0,
"step": 9395
},
{
"entropy": 5.788189125061035,
"epoch": 0.8526850507982583,
"grad_norm": 3.265625,
"learning_rate": 0.0004934670304879398,
"loss": 5.5969,
"mean_token_accuracy": 0.15632303059101105,
"num_tokens": 19519014.0,
"step": 9400
},
{
"entropy": 5.77811369895935,
"epoch": 0.8531386066763426,
"grad_norm": 1.7578125,
"learning_rate": 0.0004934592878196553,
"loss": 5.6028,
"mean_token_accuracy": 0.1545301377773285,
"num_tokens": 19528586.0,
"step": 9405
},
{
"entropy": 5.871918487548828,
"epoch": 0.8535921625544267,
"grad_norm": 2.796875,
"learning_rate": 0.000493451540633577,
"loss": 5.7379,
"mean_token_accuracy": 0.15096943005919455,
"num_tokens": 19539912.0,
"step": 9410
},
{
"entropy": 5.886108541488648,
"epoch": 0.8540457184325109,
"grad_norm": 2.875,
"learning_rate": 0.0004934437889298651,
"loss": 5.6926,
"mean_token_accuracy": 0.14735344499349595,
"num_tokens": 19549860.0,
"step": 9415
},
{
"entropy": 5.843727397918701,
"epoch": 0.854499274310595,
"grad_norm": 2.8125,
"learning_rate": 0.0004934360327086799,
"loss": 5.6329,
"mean_token_accuracy": 0.1558349311351776,
"num_tokens": 19559446.0,
"step": 9420
},
{
"entropy": 5.875371599197388,
"epoch": 0.8549528301886793,
"grad_norm": 3.578125,
"learning_rate": 0.000493428271970182,
"loss": 5.6847,
"mean_token_accuracy": 0.15351752042770386,
"num_tokens": 19570282.0,
"step": 9425
},
{
"entropy": 5.826085567474365,
"epoch": 0.8554063860667634,
"grad_norm": 3.0,
"learning_rate": 0.0004934205067145318,
"loss": 5.6183,
"mean_token_accuracy": 0.14981367737054824,
"num_tokens": 19581087.0,
"step": 9430
},
{
"entropy": 5.771932888031006,
"epoch": 0.8558599419448476,
"grad_norm": 2.4375,
"learning_rate": 0.0004934127369418898,
"loss": 5.7064,
"mean_token_accuracy": 0.15382656902074815,
"num_tokens": 19592100.0,
"step": 9435
},
{
"entropy": 5.816186094284058,
"epoch": 0.8563134978229318,
"grad_norm": 4.0,
"learning_rate": 0.0004934049626524168,
"loss": 5.6612,
"mean_token_accuracy": 0.15463659167289734,
"num_tokens": 19602557.0,
"step": 9440
},
{
"entropy": 5.917910575866699,
"epoch": 0.8567670537010159,
"grad_norm": 2.40625,
"learning_rate": 0.0004933971838462734,
"loss": 5.7029,
"mean_token_accuracy": 0.14646178781986235,
"num_tokens": 19613643.0,
"step": 9445
},
{
"entropy": 5.840442657470703,
"epoch": 0.8572206095791002,
"grad_norm": 3.890625,
"learning_rate": 0.0004933894005236207,
"loss": 5.62,
"mean_token_accuracy": 0.1540495663881302,
"num_tokens": 19624180.0,
"step": 9450
},
{
"entropy": 5.812321996688842,
"epoch": 0.8576741654571843,
"grad_norm": 5.71875,
"learning_rate": 0.0004933816126846195,
"loss": 5.6377,
"mean_token_accuracy": 0.1541966527700424,
"num_tokens": 19634799.0,
"step": 9455
},
{
"entropy": 5.77674446105957,
"epoch": 0.8581277213352685,
"grad_norm": 3.96875,
"learning_rate": 0.0004933738203294309,
"loss": 5.6135,
"mean_token_accuracy": 0.160749851167202,
"num_tokens": 19644618.0,
"step": 9460
},
{
"entropy": 5.760012865066528,
"epoch": 0.8585812772133526,
"grad_norm": 2.625,
"learning_rate": 0.000493366023458216,
"loss": 5.6447,
"mean_token_accuracy": 0.1502933233976364,
"num_tokens": 19655550.0,
"step": 9465
},
{
"entropy": 5.8344627857208256,
"epoch": 0.8590348330914369,
"grad_norm": 3.25,
"learning_rate": 0.0004933582220711362,
"loss": 5.6693,
"mean_token_accuracy": 0.15924992710351943,
"num_tokens": 19665734.0,
"step": 9470
},
{
"entropy": 5.90078501701355,
"epoch": 0.8594883889695211,
"grad_norm": 2.828125,
"learning_rate": 0.0004933504161683526,
"loss": 5.6556,
"mean_token_accuracy": 0.14944824129343032,
"num_tokens": 19677130.0,
"step": 9475
},
{
"entropy": 5.780390214920044,
"epoch": 0.8599419448476052,
"grad_norm": 2.78125,
"learning_rate": 0.0004933426057500269,
"loss": 5.5754,
"mean_token_accuracy": 0.15734326392412185,
"num_tokens": 19687374.0,
"step": 9480
},
{
"entropy": 5.87744369506836,
"epoch": 0.8603955007256894,
"grad_norm": 2.203125,
"learning_rate": 0.0004933347908163203,
"loss": 5.663,
"mean_token_accuracy": 0.15029311627149583,
"num_tokens": 19698182.0,
"step": 9485
},
{
"entropy": 5.847681093215942,
"epoch": 0.8608490566037735,
"grad_norm": 1.8671875,
"learning_rate": 0.0004933269713673947,
"loss": 5.6149,
"mean_token_accuracy": 0.15695204436779023,
"num_tokens": 19707664.0,
"step": 9490
},
{
"entropy": 5.727297925949097,
"epoch": 0.8613026124818578,
"grad_norm": 2.9375,
"learning_rate": 0.0004933191474034116,
"loss": 5.6623,
"mean_token_accuracy": 0.15112429186701776,
"num_tokens": 19717647.0,
"step": 9495
},
{
"entropy": 5.788974571228027,
"epoch": 0.861756168359942,
"grad_norm": 2.484375,
"learning_rate": 0.000493311318924533,
"loss": 5.6965,
"mean_token_accuracy": 0.15508863627910613,
"num_tokens": 19727444.0,
"step": 9500
},
{
"entropy": 5.8454258918762205,
"epoch": 0.8622097242380261,
"grad_norm": 4.21875,
"learning_rate": 0.0004933034859309208,
"loss": 5.7237,
"mean_token_accuracy": 0.14808909147977828,
"num_tokens": 19738719.0,
"step": 9505
},
{
"entropy": 5.907476949691772,
"epoch": 0.8626632801161103,
"grad_norm": 1.9453125,
"learning_rate": 0.0004932956484227367,
"loss": 5.6937,
"mean_token_accuracy": 0.15322445929050446,
"num_tokens": 19749786.0,
"step": 9510
},
{
"entropy": 5.898360776901245,
"epoch": 0.8631168359941945,
"grad_norm": 2.4375,
"learning_rate": 0.000493287806400143,
"loss": 5.677,
"mean_token_accuracy": 0.16384984850883483,
"num_tokens": 19759623.0,
"step": 9515
},
{
"entropy": 5.84280366897583,
"epoch": 0.8635703918722787,
"grad_norm": 2.75,
"learning_rate": 0.0004932799598633018,
"loss": 5.6466,
"mean_token_accuracy": 0.1526094436645508,
"num_tokens": 19770117.0,
"step": 9520
},
{
"entropy": 5.793286323547363,
"epoch": 0.8640239477503628,
"grad_norm": 3.125,
"learning_rate": 0.0004932721088123753,
"loss": 5.7419,
"mean_token_accuracy": 0.15547346323728561,
"num_tokens": 19780748.0,
"step": 9525
},
{
"entropy": 5.82646107673645,
"epoch": 0.864477503628447,
"grad_norm": 1.796875,
"learning_rate": 0.0004932642532475262,
"loss": 5.712,
"mean_token_accuracy": 0.1523199200630188,
"num_tokens": 19791816.0,
"step": 9530
},
{
"entropy": 5.908600378036499,
"epoch": 0.8649310595065312,
"grad_norm": 2.40625,
"learning_rate": 0.0004932563931689168,
"loss": 5.7278,
"mean_token_accuracy": 0.15349765717983246,
"num_tokens": 19801096.0,
"step": 9535
},
{
"entropy": 5.856309509277343,
"epoch": 0.8653846153846154,
"grad_norm": 3.640625,
"learning_rate": 0.0004932485285767093,
"loss": 5.5903,
"mean_token_accuracy": 0.1556379422545433,
"num_tokens": 19811258.0,
"step": 9540
},
{
"entropy": 5.814109754562378,
"epoch": 0.8658381712626996,
"grad_norm": 2.203125,
"learning_rate": 0.0004932406594710667,
"loss": 5.6841,
"mean_token_accuracy": 0.15989995896816253,
"num_tokens": 19821765.0,
"step": 9545
},
{
"entropy": 5.784682178497315,
"epoch": 0.8662917271407837,
"grad_norm": 2.359375,
"learning_rate": 0.0004932327858521517,
"loss": 5.5602,
"mean_token_accuracy": 0.16248162388801574,
"num_tokens": 19831537.0,
"step": 9550
},
{
"entropy": 5.778852939605713,
"epoch": 0.8667452830188679,
"grad_norm": 2.71875,
"learning_rate": 0.000493224907720127,
"loss": 5.5746,
"mean_token_accuracy": 0.15467148125171662,
"num_tokens": 19841830.0,
"step": 9555
},
{
"entropy": 5.863460540771484,
"epoch": 0.8671988388969522,
"grad_norm": 2.0625,
"learning_rate": 0.0004932170250751556,
"loss": 5.6807,
"mean_token_accuracy": 0.1475910425186157,
"num_tokens": 19851896.0,
"step": 9560
},
{
"entropy": 5.828677225112915,
"epoch": 0.8676523947750363,
"grad_norm": 3.265625,
"learning_rate": 0.0004932091379174006,
"loss": 5.7099,
"mean_token_accuracy": 0.14795950353145598,
"num_tokens": 19861735.0,
"step": 9565
},
{
"entropy": 5.848706912994385,
"epoch": 0.8681059506531205,
"grad_norm": 2.71875,
"learning_rate": 0.000493201246247025,
"loss": 5.6144,
"mean_token_accuracy": 0.1619333043694496,
"num_tokens": 19872333.0,
"step": 9570
},
{
"entropy": 5.8216742992401125,
"epoch": 0.8685595065312046,
"grad_norm": 3.578125,
"learning_rate": 0.000493193350064192,
"loss": 5.6359,
"mean_token_accuracy": 0.14962711334228515,
"num_tokens": 19882896.0,
"step": 9575
},
{
"entropy": 5.810192537307739,
"epoch": 0.8690130624092888,
"grad_norm": 2.78125,
"learning_rate": 0.0004931854493690649,
"loss": 5.5871,
"mean_token_accuracy": 0.1537919670343399,
"num_tokens": 19893767.0,
"step": 9580
},
{
"entropy": 5.7544848918914795,
"epoch": 0.869466618287373,
"grad_norm": 3.265625,
"learning_rate": 0.0004931775441618071,
"loss": 5.5776,
"mean_token_accuracy": 0.15696486085653305,
"num_tokens": 19904242.0,
"step": 9585
},
{
"entropy": 5.792415761947632,
"epoch": 0.8699201741654572,
"grad_norm": 3.203125,
"learning_rate": 0.0004931696344425821,
"loss": 5.6734,
"mean_token_accuracy": 0.15541588217020036,
"num_tokens": 19913898.0,
"step": 9590
},
{
"entropy": 5.925848960876465,
"epoch": 0.8703737300435413,
"grad_norm": 3.328125,
"learning_rate": 0.0004931617202115535,
"loss": 5.6865,
"mean_token_accuracy": 0.15690164417028427,
"num_tokens": 19923756.0,
"step": 9595
},
{
"entropy": 5.770646429061889,
"epoch": 0.8708272859216255,
"grad_norm": 4.53125,
"learning_rate": 0.0004931538014688849,
"loss": 5.6273,
"mean_token_accuracy": 0.15638579279184342,
"num_tokens": 19932720.0,
"step": 9600
},
{
"entropy": 5.847241497039795,
"epoch": 0.8712808417997098,
"grad_norm": 2.765625,
"learning_rate": 0.0004931458782147401,
"loss": 5.8113,
"mean_token_accuracy": 0.1438422456383705,
"num_tokens": 19942962.0,
"step": 9605
},
{
"entropy": 5.976376676559449,
"epoch": 0.8717343976777939,
"grad_norm": 2.484375,
"learning_rate": 0.000493137950449283,
"loss": 5.6977,
"mean_token_accuracy": 0.15650282204151153,
"num_tokens": 19953735.0,
"step": 9610
},
{
"entropy": 5.831784582138061,
"epoch": 0.8721879535558781,
"grad_norm": 2.8125,
"learning_rate": 0.0004931300181726775,
"loss": 5.6815,
"mean_token_accuracy": 0.15415865927934647,
"num_tokens": 19964416.0,
"step": 9615
},
{
"entropy": 5.827550315856934,
"epoch": 0.8726415094339622,
"grad_norm": 3.078125,
"learning_rate": 0.0004931220813850875,
"loss": 5.6933,
"mean_token_accuracy": 0.14974114447832107,
"num_tokens": 19974531.0,
"step": 9620
},
{
"entropy": 5.856517553329468,
"epoch": 0.8730950653120464,
"grad_norm": 2.28125,
"learning_rate": 0.0004931141400866773,
"loss": 5.6931,
"mean_token_accuracy": 0.15448979288339615,
"num_tokens": 19984964.0,
"step": 9625
},
{
"entropy": 5.804881381988525,
"epoch": 0.8735486211901307,
"grad_norm": 2.640625,
"learning_rate": 0.0004931061942776112,
"loss": 5.6258,
"mean_token_accuracy": 0.159735769033432,
"num_tokens": 19995568.0,
"step": 9630
},
{
"entropy": 5.927798700332642,
"epoch": 0.8740021770682148,
"grad_norm": 2.34375,
"learning_rate": 0.0004930982439580534,
"loss": 5.8032,
"mean_token_accuracy": 0.14587000608444214,
"num_tokens": 20005856.0,
"step": 9635
},
{
"entropy": 5.904760313034058,
"epoch": 0.874455732946299,
"grad_norm": 2.5625,
"learning_rate": 0.0004930902891281683,
"loss": 5.6923,
"mean_token_accuracy": 0.150040340423584,
"num_tokens": 20015683.0,
"step": 9640
},
{
"entropy": 5.7723430633544925,
"epoch": 0.8749092888243831,
"grad_norm": 4.34375,
"learning_rate": 0.0004930823297881205,
"loss": 5.534,
"mean_token_accuracy": 0.15704142898321152,
"num_tokens": 20025998.0,
"step": 9645
},
{
"entropy": 5.798841381072998,
"epoch": 0.8753628447024674,
"grad_norm": 9.5,
"learning_rate": 0.0004930743659380745,
"loss": 5.6927,
"mean_token_accuracy": 0.15218583345413209,
"num_tokens": 20036486.0,
"step": 9650
},
{
"entropy": 5.741021728515625,
"epoch": 0.8758164005805515,
"grad_norm": 2.859375,
"learning_rate": 0.000493066397578195,
"loss": 5.6093,
"mean_token_accuracy": 0.15795097947120668,
"num_tokens": 20046784.0,
"step": 9655
},
{
"entropy": 5.851765108108521,
"epoch": 0.8762699564586357,
"grad_norm": 2.53125,
"learning_rate": 0.0004930584247086468,
"loss": 5.6552,
"mean_token_accuracy": 0.15040991678833962,
"num_tokens": 20056999.0,
"step": 9660
},
{
"entropy": 5.8147345066070555,
"epoch": 0.8767235123367199,
"grad_norm": 2.546875,
"learning_rate": 0.000493050447329595,
"loss": 5.5419,
"mean_token_accuracy": 0.1588175803422928,
"num_tokens": 20067612.0,
"step": 9665
},
{
"entropy": 5.856226444244385,
"epoch": 0.877177068214804,
"grad_norm": 2.578125,
"learning_rate": 0.0004930424654412043,
"loss": 5.7058,
"mean_token_accuracy": 0.1530055969953537,
"num_tokens": 20077330.0,
"step": 9670
},
{
"entropy": 5.81877875328064,
"epoch": 0.8776306240928883,
"grad_norm": 3.9375,
"learning_rate": 0.0004930344790436398,
"loss": 5.6697,
"mean_token_accuracy": 0.1515201061964035,
"num_tokens": 20086801.0,
"step": 9675
},
{
"entropy": 5.8394787311553955,
"epoch": 0.8780841799709724,
"grad_norm": 2.296875,
"learning_rate": 0.0004930264881370666,
"loss": 5.6409,
"mean_token_accuracy": 0.15190551653504372,
"num_tokens": 20097047.0,
"step": 9680
},
{
"entropy": 5.822015190124512,
"epoch": 0.8785377358490566,
"grad_norm": 2.8125,
"learning_rate": 0.0004930184927216504,
"loss": 5.6708,
"mean_token_accuracy": 0.1601392611861229,
"num_tokens": 20107185.0,
"step": 9685
},
{
"entropy": 5.829152631759643,
"epoch": 0.8789912917271407,
"grad_norm": 2.375,
"learning_rate": 0.0004930104927975559,
"loss": 5.6213,
"mean_token_accuracy": 0.15499529242515564,
"num_tokens": 20118493.0,
"step": 9690
},
{
"entropy": 5.790694379806519,
"epoch": 0.879444847605225,
"grad_norm": 2.390625,
"learning_rate": 0.0004930024883649489,
"loss": 5.671,
"mean_token_accuracy": 0.15504609644412995,
"num_tokens": 20129294.0,
"step": 9695
},
{
"entropy": 5.7521278858184814,
"epoch": 0.8798984034833092,
"grad_norm": 2.421875,
"learning_rate": 0.000492994479423995,
"loss": 5.6159,
"mean_token_accuracy": 0.1620168849825859,
"num_tokens": 20139194.0,
"step": 9700
},
{
"entropy": 5.803879928588867,
"epoch": 0.8803519593613933,
"grad_norm": 2.234375,
"learning_rate": 0.0004929864659748597,
"loss": 5.6527,
"mean_token_accuracy": 0.1494958892464638,
"num_tokens": 20149681.0,
"step": 9705
},
{
"entropy": 5.84139928817749,
"epoch": 0.8808055152394775,
"grad_norm": 2.234375,
"learning_rate": 0.0004929784480177087,
"loss": 5.7388,
"mean_token_accuracy": 0.1462699994444847,
"num_tokens": 20160557.0,
"step": 9710
},
{
"entropy": 5.8120523452758786,
"epoch": 0.8812590711175616,
"grad_norm": 4.21875,
"learning_rate": 0.0004929704255527078,
"loss": 5.6344,
"mean_token_accuracy": 0.15634542852640151,
"num_tokens": 20171047.0,
"step": 9715
},
{
"entropy": 5.867328405380249,
"epoch": 0.8817126269956459,
"grad_norm": 3.28125,
"learning_rate": 0.000492962398580023,
"loss": 5.7423,
"mean_token_accuracy": 0.15117280706763267,
"num_tokens": 20182148.0,
"step": 9720
},
{
"entropy": 5.777922821044922,
"epoch": 0.88216618287373,
"grad_norm": 2.28125,
"learning_rate": 0.0004929543670998203,
"loss": 5.5569,
"mean_token_accuracy": 0.15868174731731416,
"num_tokens": 20192131.0,
"step": 9725
},
{
"entropy": 5.855689859390258,
"epoch": 0.8826197387518142,
"grad_norm": 2.484375,
"learning_rate": 0.0004929463311122657,
"loss": 5.6737,
"mean_token_accuracy": 0.15011551976203918,
"num_tokens": 20202805.0,
"step": 9730
},
{
"entropy": 5.817514896392822,
"epoch": 0.8830732946298984,
"grad_norm": 3.828125,
"learning_rate": 0.0004929382906175253,
"loss": 5.6053,
"mean_token_accuracy": 0.1565737932920456,
"num_tokens": 20212702.0,
"step": 9735
},
{
"entropy": 5.780954742431641,
"epoch": 0.8835268505079826,
"grad_norm": 2.40625,
"learning_rate": 0.0004929302456157657,
"loss": 5.6272,
"mean_token_accuracy": 0.1581820070743561,
"num_tokens": 20222661.0,
"step": 9740
},
{
"entropy": 5.840067434310913,
"epoch": 0.8839804063860668,
"grad_norm": 3.0,
"learning_rate": 0.0004929221961071531,
"loss": 5.6645,
"mean_token_accuracy": 0.1551720142364502,
"num_tokens": 20233012.0,
"step": 9745
},
{
"entropy": 5.858145475387573,
"epoch": 0.8844339622641509,
"grad_norm": 2.0625,
"learning_rate": 0.0004929141420918538,
"loss": 5.6646,
"mean_token_accuracy": 0.1597403571009636,
"num_tokens": 20242918.0,
"step": 9750
},
{
"entropy": 5.802211093902588,
"epoch": 0.8848875181422351,
"grad_norm": 3.953125,
"learning_rate": 0.0004929060835700347,
"loss": 5.6187,
"mean_token_accuracy": 0.15960576385259628,
"num_tokens": 20253254.0,
"step": 9755
},
{
"entropy": 5.8462591648101805,
"epoch": 0.8853410740203193,
"grad_norm": 2.546875,
"learning_rate": 0.000492898020541862,
"loss": 5.6733,
"mean_token_accuracy": 0.14833195358514786,
"num_tokens": 20263355.0,
"step": 9760
},
{
"entropy": 5.767037773132325,
"epoch": 0.8857946298984035,
"grad_norm": 2.03125,
"learning_rate": 0.0004928899530075029,
"loss": 5.6089,
"mean_token_accuracy": 0.1538158804178238,
"num_tokens": 20274716.0,
"step": 9765
},
{
"entropy": 5.755797386169434,
"epoch": 0.8862481857764877,
"grad_norm": 2.4375,
"learning_rate": 0.0004928818809671241,
"loss": 5.5942,
"mean_token_accuracy": 0.15015506744384766,
"num_tokens": 20285631.0,
"step": 9770
},
{
"entropy": 5.870469570159912,
"epoch": 0.8867017416545718,
"grad_norm": 2.4375,
"learning_rate": 0.0004928738044208923,
"loss": 5.6521,
"mean_token_accuracy": 0.15331775397062303,
"num_tokens": 20295717.0,
"step": 9775
},
{
"entropy": 5.748063564300537,
"epoch": 0.887155297532656,
"grad_norm": 3.3125,
"learning_rate": 0.0004928657233689748,
"loss": 5.6387,
"mean_token_accuracy": 0.15630463659763336,
"num_tokens": 20306637.0,
"step": 9780
},
{
"entropy": 5.818713998794555,
"epoch": 0.8876088534107403,
"grad_norm": 2.328125,
"learning_rate": 0.0004928576378115387,
"loss": 5.5891,
"mean_token_accuracy": 0.16262200921773912,
"num_tokens": 20316629.0,
"step": 9785
},
{
"entropy": 5.896687126159668,
"epoch": 0.8880624092888244,
"grad_norm": 4.5,
"learning_rate": 0.0004928495477487509,
"loss": 5.788,
"mean_token_accuracy": 0.14297403693199157,
"num_tokens": 20327935.0,
"step": 9790
},
{
"entropy": 5.855598497390747,
"epoch": 0.8885159651669086,
"grad_norm": 2.40625,
"learning_rate": 0.0004928414531807792,
"loss": 5.6003,
"mean_token_accuracy": 0.1532953843474388,
"num_tokens": 20339142.0,
"step": 9795
},
{
"entropy": 5.786568069458008,
"epoch": 0.8889695210449927,
"grad_norm": 2.265625,
"learning_rate": 0.0004928333541077906,
"loss": 5.6035,
"mean_token_accuracy": 0.16063882559537887,
"num_tokens": 20348474.0,
"step": 9800
},
{
"entropy": 5.816909885406494,
"epoch": 0.8894230769230769,
"grad_norm": 2.1875,
"learning_rate": 0.0004928252505299527,
"loss": 5.6577,
"mean_token_accuracy": 0.1501513257622719,
"num_tokens": 20359440.0,
"step": 9805
},
{
"entropy": 5.875770139694214,
"epoch": 0.8898766328011611,
"grad_norm": 2.1875,
"learning_rate": 0.0004928171424474332,
"loss": 5.7049,
"mean_token_accuracy": 0.15303952395915985,
"num_tokens": 20370355.0,
"step": 9810
},
{
"entropy": 5.778997087478638,
"epoch": 0.8903301886792453,
"grad_norm": 2.28125,
"learning_rate": 0.0004928090298603996,
"loss": 5.6012,
"mean_token_accuracy": 0.15646144449710847,
"num_tokens": 20380214.0,
"step": 9815
},
{
"entropy": 5.799949645996094,
"epoch": 0.8907837445573294,
"grad_norm": 1.90625,
"learning_rate": 0.0004928009127690199,
"loss": 5.6068,
"mean_token_accuracy": 0.15871121734380722,
"num_tokens": 20389190.0,
"step": 9820
},
{
"entropy": 5.775676584243774,
"epoch": 0.8912373004354136,
"grad_norm": 2.515625,
"learning_rate": 0.0004927927911734617,
"loss": 5.6645,
"mean_token_accuracy": 0.15490778982639314,
"num_tokens": 20399531.0,
"step": 9825
},
{
"entropy": 5.819215440750122,
"epoch": 0.8916908563134979,
"grad_norm": 2.015625,
"learning_rate": 0.0004927846650738931,
"loss": 5.6684,
"mean_token_accuracy": 0.15653486102819442,
"num_tokens": 20410201.0,
"step": 9830
},
{
"entropy": 5.820386266708374,
"epoch": 0.892144412191582,
"grad_norm": 2.296875,
"learning_rate": 0.0004927765344704821,
"loss": 5.6198,
"mean_token_accuracy": 0.15861736387014388,
"num_tokens": 20421579.0,
"step": 9835
},
{
"entropy": 5.789028739929199,
"epoch": 0.8925979680696662,
"grad_norm": 2.515625,
"learning_rate": 0.0004927683993633969,
"loss": 5.6932,
"mean_token_accuracy": 0.15348317474126816,
"num_tokens": 20432048.0,
"step": 9840
},
{
"entropy": 5.8670917510986325,
"epoch": 0.8930515239477503,
"grad_norm": 2.0625,
"learning_rate": 0.0004927602597528057,
"loss": 5.7223,
"mean_token_accuracy": 0.15003873780369759,
"num_tokens": 20441752.0,
"step": 9845
},
{
"entropy": 5.822627258300781,
"epoch": 0.8935050798258345,
"grad_norm": 2.734375,
"learning_rate": 0.0004927521156388769,
"loss": 5.5655,
"mean_token_accuracy": 0.15744317770004274,
"num_tokens": 20451693.0,
"step": 9850
},
{
"entropy": 5.830567121505737,
"epoch": 0.8939586357039188,
"grad_norm": 3.859375,
"learning_rate": 0.0004927439670217788,
"loss": 5.6907,
"mean_token_accuracy": 0.1509220041334629,
"num_tokens": 20462312.0,
"step": 9855
},
{
"entropy": 5.790587472915649,
"epoch": 0.8944121915820029,
"grad_norm": 2.625,
"learning_rate": 0.00049273581390168,
"loss": 5.6059,
"mean_token_accuracy": 0.1467101141810417,
"num_tokens": 20472546.0,
"step": 9860
},
{
"entropy": 5.781812763214111,
"epoch": 0.8948657474600871,
"grad_norm": 2.96875,
"learning_rate": 0.000492727656278749,
"loss": 5.6558,
"mean_token_accuracy": 0.14958946257829667,
"num_tokens": 20483266.0,
"step": 9865
},
{
"entropy": 5.750693082809448,
"epoch": 0.8953193033381712,
"grad_norm": 1.9609375,
"learning_rate": 0.0004927194941531547,
"loss": 5.5756,
"mean_token_accuracy": 0.15675682723522186,
"num_tokens": 20492490.0,
"step": 9870
},
{
"entropy": 5.833357667922973,
"epoch": 0.8957728592162555,
"grad_norm": 2.3125,
"learning_rate": 0.0004927113275250657,
"loss": 5.6116,
"mean_token_accuracy": 0.15225069373846054,
"num_tokens": 20503077.0,
"step": 9875
},
{
"entropy": 5.757365465164185,
"epoch": 0.8962264150943396,
"grad_norm": 2.578125,
"learning_rate": 0.0004927031563946511,
"loss": 5.6305,
"mean_token_accuracy": 0.15087197571992875,
"num_tokens": 20513771.0,
"step": 9880
},
{
"entropy": 5.80473780632019,
"epoch": 0.8966799709724238,
"grad_norm": 2.640625,
"learning_rate": 0.0004926949807620796,
"loss": 5.7378,
"mean_token_accuracy": 0.1536948338150978,
"num_tokens": 20523616.0,
"step": 9885
},
{
"entropy": 5.843550491333008,
"epoch": 0.897133526850508,
"grad_norm": 2.734375,
"learning_rate": 0.0004926868006275204,
"loss": 5.5529,
"mean_token_accuracy": 0.15692562013864517,
"num_tokens": 20533613.0,
"step": 9890
},
{
"entropy": 5.7657022953033445,
"epoch": 0.8975870827285921,
"grad_norm": 2.375,
"learning_rate": 0.0004926786159911427,
"loss": 5.6365,
"mean_token_accuracy": 0.1459011420607567,
"num_tokens": 20543812.0,
"step": 9895
},
{
"entropy": 5.727877378463745,
"epoch": 0.8980406386066764,
"grad_norm": 2.234375,
"learning_rate": 0.000492670426853116,
"loss": 5.5854,
"mean_token_accuracy": 0.15662455111742019,
"num_tokens": 20553785.0,
"step": 9900
},
{
"entropy": 5.887316608428955,
"epoch": 0.8984941944847605,
"grad_norm": 1.8359375,
"learning_rate": 0.0004926622332136092,
"loss": 5.6642,
"mean_token_accuracy": 0.15611073970794678,
"num_tokens": 20563326.0,
"step": 9905
},
{
"entropy": 5.789510202407837,
"epoch": 0.8989477503628447,
"grad_norm": 2.890625,
"learning_rate": 0.0004926540350727918,
"loss": 5.5444,
"mean_token_accuracy": 0.15752981156110762,
"num_tokens": 20573643.0,
"step": 9910
},
{
"entropy": 5.686382246017456,
"epoch": 0.8994013062409288,
"grad_norm": 2.265625,
"learning_rate": 0.0004926458324308337,
"loss": 5.5834,
"mean_token_accuracy": 0.15729566514492035,
"num_tokens": 20583916.0,
"step": 9915
},
{
"entropy": 5.836192798614502,
"epoch": 0.8998548621190131,
"grad_norm": 2.265625,
"learning_rate": 0.0004926376252879043,
"loss": 5.7371,
"mean_token_accuracy": 0.14930728375911712,
"num_tokens": 20593984.0,
"step": 9920
},
{
"entropy": 5.861156177520752,
"epoch": 0.9003084179970973,
"grad_norm": 3.125,
"learning_rate": 0.0004926294136441733,
"loss": 5.5535,
"mean_token_accuracy": 0.161505389213562,
"num_tokens": 20603927.0,
"step": 9925
},
{
"entropy": 5.867801284790039,
"epoch": 0.9007619738751814,
"grad_norm": 6.1875,
"learning_rate": 0.0004926211974998106,
"loss": 5.7063,
"mean_token_accuracy": 0.15205266922712327,
"num_tokens": 20614147.0,
"step": 9930
},
{
"entropy": 5.894407272338867,
"epoch": 0.9012155297532656,
"grad_norm": 2.09375,
"learning_rate": 0.000492612976854986,
"loss": 5.6583,
"mean_token_accuracy": 0.1546430841088295,
"num_tokens": 20623631.0,
"step": 9935
},
{
"entropy": 5.720052576065063,
"epoch": 0.9016690856313497,
"grad_norm": 2.53125,
"learning_rate": 0.0004926047517098697,
"loss": 5.5485,
"mean_token_accuracy": 0.15941712707281114,
"num_tokens": 20632566.0,
"step": 9940
},
{
"entropy": 5.809675502777099,
"epoch": 0.902122641509434,
"grad_norm": 2.734375,
"learning_rate": 0.0004925965220646315,
"loss": 5.6289,
"mean_token_accuracy": 0.15629922449588776,
"num_tokens": 20643246.0,
"step": 9945
},
{
"entropy": 5.817538738250732,
"epoch": 0.9025761973875182,
"grad_norm": 2.734375,
"learning_rate": 0.0004925882879194419,
"loss": 5.7432,
"mean_token_accuracy": 0.1489126816391945,
"num_tokens": 20653372.0,
"step": 9950
},
{
"entropy": 5.853244304656982,
"epoch": 0.9030297532656023,
"grad_norm": 2.015625,
"learning_rate": 0.0004925800492744712,
"loss": 5.6704,
"mean_token_accuracy": 0.15181304961442948,
"num_tokens": 20662687.0,
"step": 9955
},
{
"entropy": 5.815691566467285,
"epoch": 0.9034833091436865,
"grad_norm": 2.53125,
"learning_rate": 0.0004925718061298895,
"loss": 5.6074,
"mean_token_accuracy": 0.16012194007635117,
"num_tokens": 20673168.0,
"step": 9960
},
{
"entropy": 5.829687976837159,
"epoch": 0.9039368650217707,
"grad_norm": 3.0625,
"learning_rate": 0.0004925635584858675,
"loss": 5.709,
"mean_token_accuracy": 0.15368616580963135,
"num_tokens": 20683100.0,
"step": 9965
},
{
"entropy": 5.898300218582153,
"epoch": 0.9043904208998549,
"grad_norm": 2.0625,
"learning_rate": 0.0004925553063425755,
"loss": 5.7214,
"mean_token_accuracy": 0.1502762645483017,
"num_tokens": 20694097.0,
"step": 9970
},
{
"entropy": 5.8318281173706055,
"epoch": 0.904843976777939,
"grad_norm": 3.46875,
"learning_rate": 0.0004925470497001844,
"loss": 5.6888,
"mean_token_accuracy": 0.1505840465426445,
"num_tokens": 20704640.0,
"step": 9975
},
{
"entropy": 5.79219069480896,
"epoch": 0.9052975326560232,
"grad_norm": 2.1875,
"learning_rate": 0.0004925387885588648,
"loss": 5.6115,
"mean_token_accuracy": 0.15379339158535005,
"num_tokens": 20714419.0,
"step": 9980
},
{
"entropy": 5.930364370346069,
"epoch": 0.9057510885341074,
"grad_norm": 2.859375,
"learning_rate": 0.0004925305229187877,
"loss": 5.7051,
"mean_token_accuracy": 0.15184128955006598,
"num_tokens": 20724705.0,
"step": 9985
},
{
"entropy": 5.943155097961426,
"epoch": 0.9062046444121916,
"grad_norm": 2.3125,
"learning_rate": 0.000492522252780124,
"loss": 5.7953,
"mean_token_accuracy": 0.1515795536339283,
"num_tokens": 20734753.0,
"step": 9990
},
{
"entropy": 5.797655057907105,
"epoch": 0.9066582002902758,
"grad_norm": 2.28125,
"learning_rate": 0.0004925139781430446,
"loss": 5.5928,
"mean_token_accuracy": 0.156381855905056,
"num_tokens": 20745389.0,
"step": 9995
},
{
"entropy": 5.760052299499511,
"epoch": 0.9071117561683599,
"grad_norm": 1.8515625,
"learning_rate": 0.0004925056990077207,
"loss": 5.6167,
"mean_token_accuracy": 0.15757633447647096,
"num_tokens": 20755702.0,
"step": 10000
},
{
"entropy": 5.840163516998291,
"epoch": 0.9075653120464441,
"grad_norm": 2.828125,
"learning_rate": 0.0004924974153743235,
"loss": 5.5611,
"mean_token_accuracy": 0.15907627418637277,
"num_tokens": 20766046.0,
"step": 10005
},
{
"entropy": 5.841227865219116,
"epoch": 0.9080188679245284,
"grad_norm": 2.421875,
"learning_rate": 0.0004924891272430244,
"loss": 5.6409,
"mean_token_accuracy": 0.15797481536865235,
"num_tokens": 20776654.0,
"step": 10010
},
{
"entropy": 5.900266551971436,
"epoch": 0.9084724238026125,
"grad_norm": 4.375,
"learning_rate": 0.0004924808346139948,
"loss": 5.7158,
"mean_token_accuracy": 0.1568849690258503,
"num_tokens": 20787066.0,
"step": 10015
},
{
"entropy": 5.8268232345581055,
"epoch": 0.9089259796806967,
"grad_norm": 3.421875,
"learning_rate": 0.0004924725374874059,
"loss": 5.5981,
"mean_token_accuracy": 0.15794997215270995,
"num_tokens": 20796822.0,
"step": 10020
},
{
"entropy": 5.798414707183838,
"epoch": 0.9093795355587808,
"grad_norm": 3.265625,
"learning_rate": 0.0004924642358634297,
"loss": 5.6485,
"mean_token_accuracy": 0.15123971551656723,
"num_tokens": 20807602.0,
"step": 10025
},
{
"entropy": 5.838248157501221,
"epoch": 0.909833091436865,
"grad_norm": 4.375,
"learning_rate": 0.0004924559297422376,
"loss": 5.6415,
"mean_token_accuracy": 0.1583787500858307,
"num_tokens": 20819046.0,
"step": 10030
},
{
"entropy": 5.831664276123047,
"epoch": 0.9102866473149492,
"grad_norm": 2.296875,
"learning_rate": 0.0004924476191240014,
"loss": 5.6566,
"mean_token_accuracy": 0.15371111333370208,
"num_tokens": 20828989.0,
"step": 10035
},
{
"entropy": 5.811376571655273,
"epoch": 0.9107402031930334,
"grad_norm": 3.21875,
"learning_rate": 0.000492439304008893,
"loss": 5.6601,
"mean_token_accuracy": 0.1526098906993866,
"num_tokens": 20839366.0,
"step": 10040
},
{
"entropy": 5.830955839157104,
"epoch": 0.9111937590711175,
"grad_norm": 2.0625,
"learning_rate": 0.0004924309843970847,
"loss": 5.6463,
"mean_token_accuracy": 0.15849197879433632,
"num_tokens": 20849513.0,
"step": 10045
},
{
"entropy": 5.910466909408569,
"epoch": 0.9116473149492017,
"grad_norm": 2.234375,
"learning_rate": 0.0004924226602887479,
"loss": 5.7029,
"mean_token_accuracy": 0.14674522280693053,
"num_tokens": 20861406.0,
"step": 10050
},
{
"entropy": 5.799602031707764,
"epoch": 0.912100870827286,
"grad_norm": 3.171875,
"learning_rate": 0.0004924143316840552,
"loss": 5.6015,
"mean_token_accuracy": 0.16461957097053528,
"num_tokens": 20871190.0,
"step": 10055
},
{
"entropy": 5.843008947372437,
"epoch": 0.9125544267053701,
"grad_norm": 2.71875,
"learning_rate": 0.0004924059985831787,
"loss": 5.7387,
"mean_token_accuracy": 0.15071364492177963,
"num_tokens": 20881589.0,
"step": 10060
},
{
"entropy": 5.8244470119476315,
"epoch": 0.9130079825834543,
"grad_norm": 2.765625,
"learning_rate": 0.0004923976609862908,
"loss": 5.6717,
"mean_token_accuracy": 0.14722735732793807,
"num_tokens": 20892072.0,
"step": 10065
},
{
"entropy": 5.886385250091553,
"epoch": 0.9134615384615384,
"grad_norm": 2.125,
"learning_rate": 0.0004923893188935638,
"loss": 5.6589,
"mean_token_accuracy": 0.15132659822702407,
"num_tokens": 20902616.0,
"step": 10070
},
{
"entropy": 5.869722700119018,
"epoch": 0.9139150943396226,
"grad_norm": 2.015625,
"learning_rate": 0.0004923809723051702,
"loss": 5.6827,
"mean_token_accuracy": 0.1534601032733917,
"num_tokens": 20913832.0,
"step": 10075
},
{
"entropy": 5.739912796020508,
"epoch": 0.9143686502177069,
"grad_norm": 2.53125,
"learning_rate": 0.0004923726212212829,
"loss": 5.5464,
"mean_token_accuracy": 0.16056860238313675,
"num_tokens": 20923322.0,
"step": 10080
},
{
"entropy": 5.758353137969971,
"epoch": 0.914822206095791,
"grad_norm": 2.796875,
"learning_rate": 0.0004923642656420742,
"loss": 5.6584,
"mean_token_accuracy": 0.15374343991279601,
"num_tokens": 20933334.0,
"step": 10085
},
{
"entropy": 5.862091779708862,
"epoch": 0.9152757619738752,
"grad_norm": 2.484375,
"learning_rate": 0.0004923559055677173,
"loss": 5.6023,
"mean_token_accuracy": 0.1538617268204689,
"num_tokens": 20944171.0,
"step": 10090
},
{
"entropy": 5.853946924209595,
"epoch": 0.9157293178519593,
"grad_norm": 2.8125,
"learning_rate": 0.0004923475409983848,
"loss": 5.6513,
"mean_token_accuracy": 0.14951413869857788,
"num_tokens": 20954200.0,
"step": 10095
},
{
"entropy": 5.786560821533203,
"epoch": 0.9161828737300436,
"grad_norm": 2.015625,
"learning_rate": 0.0004923391719342497,
"loss": 5.6525,
"mean_token_accuracy": 0.1497160390019417,
"num_tokens": 20964935.0,
"step": 10100
},
{
"entropy": 5.829623651504517,
"epoch": 0.9166364296081277,
"grad_norm": 1.953125,
"learning_rate": 0.0004923307983754852,
"loss": 5.651,
"mean_token_accuracy": 0.15524252504110336,
"num_tokens": 20975499.0,
"step": 10105
},
{
"entropy": 5.718609046936035,
"epoch": 0.9170899854862119,
"grad_norm": 1.9765625,
"learning_rate": 0.0004923224203222645,
"loss": 5.5592,
"mean_token_accuracy": 0.1539004072546959,
"num_tokens": 20986199.0,
"step": 10110
},
{
"entropy": 5.755342245101929,
"epoch": 0.9175435413642961,
"grad_norm": 3.265625,
"learning_rate": 0.0004923140377747606,
"loss": 5.6396,
"mean_token_accuracy": 0.15343184471130372,
"num_tokens": 20996313.0,
"step": 10115
},
{
"entropy": 5.819121789932251,
"epoch": 0.9179970972423802,
"grad_norm": 3.203125,
"learning_rate": 0.0004923056507331471,
"loss": 5.6099,
"mean_token_accuracy": 0.1543293446302414,
"num_tokens": 21006133.0,
"step": 10120
},
{
"entropy": 5.8339330673217775,
"epoch": 0.9184506531204645,
"grad_norm": 3.234375,
"learning_rate": 0.0004922972591975974,
"loss": 5.5978,
"mean_token_accuracy": 0.15862650871276857,
"num_tokens": 21016448.0,
"step": 10125
},
{
"entropy": 5.770276260375977,
"epoch": 0.9189042089985486,
"grad_norm": 2.015625,
"learning_rate": 0.0004922888631682849,
"loss": 5.5403,
"mean_token_accuracy": 0.15439814180135727,
"num_tokens": 21026458.0,
"step": 10130
},
{
"entropy": 5.825389194488525,
"epoch": 0.9193577648766328,
"grad_norm": 2.375,
"learning_rate": 0.0004922804626453834,
"loss": 5.5758,
"mean_token_accuracy": 0.15657375603914261,
"num_tokens": 21037111.0,
"step": 10135
},
{
"entropy": 5.813408279418946,
"epoch": 0.9198113207547169,
"grad_norm": 2.203125,
"learning_rate": 0.0004922720576290665,
"loss": 5.6685,
"mean_token_accuracy": 0.15609802007675172,
"num_tokens": 21048041.0,
"step": 10140
},
{
"entropy": 5.73240475654602,
"epoch": 0.9202648766328012,
"grad_norm": 3.296875,
"learning_rate": 0.0004922636481195082,
"loss": 5.5345,
"mean_token_accuracy": 0.16573669612407685,
"num_tokens": 21058099.0,
"step": 10145
},
{
"entropy": 5.816942262649536,
"epoch": 0.9207184325108854,
"grad_norm": 2.078125,
"learning_rate": 0.0004922552341168823,
"loss": 5.6739,
"mean_token_accuracy": 0.14541862457990645,
"num_tokens": 21068137.0,
"step": 10150
},
{
"entropy": 5.747505617141724,
"epoch": 0.9211719883889695,
"grad_norm": 2.625,
"learning_rate": 0.0004922468156213627,
"loss": 5.5484,
"mean_token_accuracy": 0.1605380281805992,
"num_tokens": 21077556.0,
"step": 10155
},
{
"entropy": 5.804719352722168,
"epoch": 0.9216255442670537,
"grad_norm": 2.328125,
"learning_rate": 0.0004922383926331236,
"loss": 5.5838,
"mean_token_accuracy": 0.15588331669569017,
"num_tokens": 21087524.0,
"step": 10160
},
{
"entropy": 5.7460955619812015,
"epoch": 0.9220791001451378,
"grad_norm": 2.96875,
"learning_rate": 0.0004922299651523392,
"loss": 5.6607,
"mean_token_accuracy": 0.15176035612821578,
"num_tokens": 21098446.0,
"step": 10165
},
{
"entropy": 5.776041746139526,
"epoch": 0.9225326560232221,
"grad_norm": 2.03125,
"learning_rate": 0.0004922215331791837,
"loss": 5.6514,
"mean_token_accuracy": 0.1542374908924103,
"num_tokens": 21109291.0,
"step": 10170
},
{
"entropy": 5.798018217086792,
"epoch": 0.9229862119013063,
"grad_norm": 2.8125,
"learning_rate": 0.0004922130967138316,
"loss": 5.5424,
"mean_token_accuracy": 0.1597289949655533,
"num_tokens": 21120043.0,
"step": 10175
},
{
"entropy": 5.842937231063843,
"epoch": 0.9234397677793904,
"grad_norm": 2.4375,
"learning_rate": 0.0004922046557564574,
"loss": 5.6839,
"mean_token_accuracy": 0.1505768433213234,
"num_tokens": 21130339.0,
"step": 10180
},
{
"entropy": 5.839677381515503,
"epoch": 0.9238933236574746,
"grad_norm": 2.0625,
"learning_rate": 0.0004921962103072355,
"loss": 5.7313,
"mean_token_accuracy": 0.15170611292123795,
"num_tokens": 21142307.0,
"step": 10185
},
{
"entropy": 5.841834831237793,
"epoch": 0.9243468795355588,
"grad_norm": 2.265625,
"learning_rate": 0.0004921877603663406,
"loss": 5.6285,
"mean_token_accuracy": 0.15684072822332382,
"num_tokens": 21151843.0,
"step": 10190
},
{
"entropy": 5.84149398803711,
"epoch": 0.924800435413643,
"grad_norm": 2.671875,
"learning_rate": 0.0004921793059339474,
"loss": 5.6204,
"mean_token_accuracy": 0.1543716825544834,
"num_tokens": 21162574.0,
"step": 10195
},
{
"entropy": 5.825618267059326,
"epoch": 0.9252539912917271,
"grad_norm": 2.15625,
"learning_rate": 0.0004921708470102309,
"loss": 5.5822,
"mean_token_accuracy": 0.15569959282875062,
"num_tokens": 21172637.0,
"step": 10200
},
{
"entropy": 5.734502506256104,
"epoch": 0.9257075471698113,
"grad_norm": 1.734375,
"learning_rate": 0.000492162383595366,
"loss": 5.5819,
"mean_token_accuracy": 0.15843692421913147,
"num_tokens": 21182272.0,
"step": 10205
},
{
"entropy": 5.8643608570098875,
"epoch": 0.9261611030478955,
"grad_norm": 2.296875,
"learning_rate": 0.0004921539156895275,
"loss": 5.7286,
"mean_token_accuracy": 0.15053602457046508,
"num_tokens": 21193712.0,
"step": 10210
},
{
"entropy": 5.8853858470916744,
"epoch": 0.9266146589259797,
"grad_norm": 2.03125,
"learning_rate": 0.0004921454432928908,
"loss": 5.718,
"mean_token_accuracy": 0.14995245784521102,
"num_tokens": 21205104.0,
"step": 10215
},
{
"entropy": 5.846768569946289,
"epoch": 0.9270682148040639,
"grad_norm": 2.578125,
"learning_rate": 0.000492136966405631,
"loss": 5.6111,
"mean_token_accuracy": 0.15548471063375474,
"num_tokens": 21215495.0,
"step": 10220
},
{
"entropy": 5.789591693878174,
"epoch": 0.927521770682148,
"grad_norm": 2.515625,
"learning_rate": 0.0004921284850279233,
"loss": 5.6316,
"mean_token_accuracy": 0.15263270437717438,
"num_tokens": 21226047.0,
"step": 10225
},
{
"entropy": 5.819576454162598,
"epoch": 0.9279753265602322,
"grad_norm": 3.125,
"learning_rate": 0.0004921199991599433,
"loss": 5.6606,
"mean_token_accuracy": 0.1504075489938259,
"num_tokens": 21235784.0,
"step": 10230
},
{
"entropy": 5.873410987854004,
"epoch": 0.9284288824383164,
"grad_norm": 2.203125,
"learning_rate": 0.0004921115088018665,
"loss": 5.6496,
"mean_token_accuracy": 0.15487141907215118,
"num_tokens": 21245493.0,
"step": 10235
},
{
"entropy": 5.891049909591675,
"epoch": 0.9288824383164006,
"grad_norm": 1.8359375,
"learning_rate": 0.000492103013953868,
"loss": 5.7419,
"mean_token_accuracy": 0.14885285943746568,
"num_tokens": 21255541.0,
"step": 10240
},
{
"entropy": 5.756620264053344,
"epoch": 0.9293359941944848,
"grad_norm": 2.328125,
"learning_rate": 0.0004920945146161242,
"loss": 5.6112,
"mean_token_accuracy": 0.1582723706960678,
"num_tokens": 21265427.0,
"step": 10245
},
{
"entropy": 5.780202388763428,
"epoch": 0.9297895500725689,
"grad_norm": 1.96875,
"learning_rate": 0.0004920860107888105,
"loss": 5.605,
"mean_token_accuracy": 0.1546992763876915,
"num_tokens": 21276597.0,
"step": 10250
},
{
"entropy": 5.783038330078125,
"epoch": 0.9302431059506531,
"grad_norm": 1.9765625,
"learning_rate": 0.0004920775024721026,
"loss": 5.604,
"mean_token_accuracy": 0.1555513247847557,
"num_tokens": 21286416.0,
"step": 10255
},
{
"entropy": 5.770952939987183,
"epoch": 0.9306966618287373,
"grad_norm": 2.25,
"learning_rate": 0.0004920689896661767,
"loss": 5.6224,
"mean_token_accuracy": 0.16034480333328247,
"num_tokens": 21296064.0,
"step": 10260
},
{
"entropy": 5.796384191513061,
"epoch": 0.9311502177068215,
"grad_norm": 3.03125,
"learning_rate": 0.0004920604723712088,
"loss": 5.6472,
"mean_token_accuracy": 0.1589336574077606,
"num_tokens": 21306836.0,
"step": 10265
},
{
"entropy": 5.765583276748657,
"epoch": 0.9316037735849056,
"grad_norm": 2.140625,
"learning_rate": 0.0004920519505873751,
"loss": 5.5942,
"mean_token_accuracy": 0.1574314147233963,
"num_tokens": 21316936.0,
"step": 10270
},
{
"entropy": 5.7488853454589846,
"epoch": 0.9320573294629898,
"grad_norm": 3.703125,
"learning_rate": 0.0004920434243148515,
"loss": 5.5772,
"mean_token_accuracy": 0.16025102734565735,
"num_tokens": 21326892.0,
"step": 10275
},
{
"entropy": 5.79693112373352,
"epoch": 0.9325108853410741,
"grad_norm": 2.421875,
"learning_rate": 0.0004920348935538147,
"loss": 5.5968,
"mean_token_accuracy": 0.15238452702760696,
"num_tokens": 21337270.0,
"step": 10280
},
{
"entropy": 5.816615438461303,
"epoch": 0.9329644412191582,
"grad_norm": 2.21875,
"learning_rate": 0.000492026358304441,
"loss": 5.664,
"mean_token_accuracy": 0.15377061069011688,
"num_tokens": 21347968.0,
"step": 10285
},
{
"entropy": 5.80585355758667,
"epoch": 0.9334179970972424,
"grad_norm": 2.578125,
"learning_rate": 0.000492017818566907,
"loss": 5.6084,
"mean_token_accuracy": 0.1599868893623352,
"num_tokens": 21358312.0,
"step": 10290
},
{
"entropy": 5.83113660812378,
"epoch": 0.9338715529753265,
"grad_norm": 2.359375,
"learning_rate": 0.0004920092743413892,
"loss": 5.5822,
"mean_token_accuracy": 0.15949864387512208,
"num_tokens": 21368794.0,
"step": 10295
},
{
"entropy": 5.85403733253479,
"epoch": 0.9343251088534107,
"grad_norm": 1.6875,
"learning_rate": 0.0004920007256280642,
"loss": 5.6545,
"mean_token_accuracy": 0.1595981612801552,
"num_tokens": 21380080.0,
"step": 10300
},
{
"entropy": 5.783421802520752,
"epoch": 0.934778664731495,
"grad_norm": 2.296875,
"learning_rate": 0.000491992172427109,
"loss": 5.6109,
"mean_token_accuracy": 0.15633728057146073,
"num_tokens": 21389961.0,
"step": 10305
},
{
"entropy": 5.724167251586914,
"epoch": 0.9352322206095791,
"grad_norm": 2.640625,
"learning_rate": 0.0004919836147387002,
"loss": 5.6492,
"mean_token_accuracy": 0.15565934926271438,
"num_tokens": 21399869.0,
"step": 10310
},
{
"entropy": 5.795440196990967,
"epoch": 0.9356857764876633,
"grad_norm": 2.71875,
"learning_rate": 0.0004919750525630151,
"loss": 5.6292,
"mean_token_accuracy": 0.15735144168138504,
"num_tokens": 21410343.0,
"step": 10315
},
{
"entropy": 5.877757787704468,
"epoch": 0.9361393323657474,
"grad_norm": 2.75,
"learning_rate": 0.0004919664859002306,
"loss": 5.6479,
"mean_token_accuracy": 0.1598845586180687,
"num_tokens": 21420915.0,
"step": 10320
},
{
"entropy": 5.834699201583862,
"epoch": 0.9365928882438317,
"grad_norm": 2.890625,
"learning_rate": 0.0004919579147505239,
"loss": 5.6066,
"mean_token_accuracy": 0.1559985712170601,
"num_tokens": 21430879.0,
"step": 10325
},
{
"entropy": 5.776730012893677,
"epoch": 0.9370464441219158,
"grad_norm": 2.765625,
"learning_rate": 0.0004919493391140721,
"loss": 5.6739,
"mean_token_accuracy": 0.15353767275810243,
"num_tokens": 21440843.0,
"step": 10330
},
{
"entropy": 5.837963724136353,
"epoch": 0.9375,
"grad_norm": 3.578125,
"learning_rate": 0.0004919407589910529,
"loss": 5.8305,
"mean_token_accuracy": 0.14749686270952225,
"num_tokens": 21452132.0,
"step": 10335
},
{
"entropy": 5.822698879241943,
"epoch": 0.9379535558780842,
"grad_norm": 5.9375,
"learning_rate": 0.0004919321743816434,
"loss": 5.6264,
"mean_token_accuracy": 0.15394114553928376,
"num_tokens": 21461861.0,
"step": 10340
},
{
"entropy": 5.831479740142822,
"epoch": 0.9384071117561683,
"grad_norm": 2.03125,
"learning_rate": 0.0004919235852860213,
"loss": 5.626,
"mean_token_accuracy": 0.15050584822893143,
"num_tokens": 21472244.0,
"step": 10345
},
{
"entropy": 5.858907270431518,
"epoch": 0.9388606676342526,
"grad_norm": 2.203125,
"learning_rate": 0.0004919149917043641,
"loss": 5.6769,
"mean_token_accuracy": 0.15351335778832437,
"num_tokens": 21482394.0,
"step": 10350
},
{
"entropy": 5.798162841796875,
"epoch": 0.9393142235123367,
"grad_norm": 3.5,
"learning_rate": 0.0004919063936368496,
"loss": 5.724,
"mean_token_accuracy": 0.1505119249224663,
"num_tokens": 21493334.0,
"step": 10355
},
{
"entropy": 5.903670835494995,
"epoch": 0.9397677793904209,
"grad_norm": 2.671875,
"learning_rate": 0.0004918977910836557,
"loss": 5.6716,
"mean_token_accuracy": 0.15008047968149185,
"num_tokens": 21504443.0,
"step": 10360
},
{
"entropy": 5.919463634490967,
"epoch": 0.940221335268505,
"grad_norm": 3.265625,
"learning_rate": 0.0004918891840449601,
"loss": 5.7001,
"mean_token_accuracy": 0.1535200819373131,
"num_tokens": 21513908.0,
"step": 10365
},
{
"entropy": 5.819669914245606,
"epoch": 0.9406748911465893,
"grad_norm": 2.515625,
"learning_rate": 0.000491880572520941,
"loss": 5.7129,
"mean_token_accuracy": 0.15204098969697952,
"num_tokens": 21524349.0,
"step": 10370
},
{
"entropy": 5.970065927505493,
"epoch": 0.9411284470246735,
"grad_norm": 1.75,
"learning_rate": 0.0004918719565117763,
"loss": 5.7562,
"mean_token_accuracy": 0.1551732286810875,
"num_tokens": 21535003.0,
"step": 10375
},
{
"entropy": 5.940258646011353,
"epoch": 0.9415820029027576,
"grad_norm": 2.96875,
"learning_rate": 0.0004918633360176444,
"loss": 5.8002,
"mean_token_accuracy": 0.1469562128186226,
"num_tokens": 21545045.0,
"step": 10380
},
{
"entropy": 5.784483289718628,
"epoch": 0.9420355587808418,
"grad_norm": 2.59375,
"learning_rate": 0.0004918547110387234,
"loss": 5.6039,
"mean_token_accuracy": 0.15595303624868392,
"num_tokens": 21554830.0,
"step": 10385
},
{
"entropy": 5.826089000701904,
"epoch": 0.9424891146589259,
"grad_norm": 3.625,
"learning_rate": 0.0004918460815751916,
"loss": 5.5774,
"mean_token_accuracy": 0.15867999494075774,
"num_tokens": 21564693.0,
"step": 10390
},
{
"entropy": 5.778178453445435,
"epoch": 0.9429426705370102,
"grad_norm": 3.859375,
"learning_rate": 0.0004918374476272277,
"loss": 5.5948,
"mean_token_accuracy": 0.15591406226158142,
"num_tokens": 21575475.0,
"step": 10395
},
{
"entropy": 5.782553958892822,
"epoch": 0.9433962264150944,
"grad_norm": 3.21875,
"learning_rate": 0.0004918288091950101,
"loss": 5.6214,
"mean_token_accuracy": 0.15539320111274718,
"num_tokens": 21586944.0,
"step": 10400
},
{
"entropy": 5.872455310821533,
"epoch": 0.9438497822931785,
"grad_norm": 2.546875,
"learning_rate": 0.0004918201662787174,
"loss": 5.6537,
"mean_token_accuracy": 0.1531512513756752,
"num_tokens": 21598170.0,
"step": 10405
},
{
"entropy": 5.838468408584594,
"epoch": 0.9443033381712627,
"grad_norm": 2.59375,
"learning_rate": 0.0004918115188785285,
"loss": 5.6587,
"mean_token_accuracy": 0.15409923940896988,
"num_tokens": 21609170.0,
"step": 10410
},
{
"entropy": 5.841028451919556,
"epoch": 0.9447568940493469,
"grad_norm": 2.15625,
"learning_rate": 0.0004918028669946221,
"loss": 5.7217,
"mean_token_accuracy": 0.15151064842939377,
"num_tokens": 21620171.0,
"step": 10415
},
{
"entropy": 5.8394111633300785,
"epoch": 0.9452104499274311,
"grad_norm": 6.3125,
"learning_rate": 0.0004917942106271771,
"loss": 5.6671,
"mean_token_accuracy": 0.16137124970555305,
"num_tokens": 21631072.0,
"step": 10420
},
{
"entropy": 5.75335693359375,
"epoch": 0.9456640058055152,
"grad_norm": 2.46875,
"learning_rate": 0.0004917855497763727,
"loss": 5.5325,
"mean_token_accuracy": 0.16007351726293564,
"num_tokens": 21641543.0,
"step": 10425
},
{
"entropy": 5.783782625198365,
"epoch": 0.9461175616835994,
"grad_norm": 2.59375,
"learning_rate": 0.0004917768844423878,
"loss": 5.6278,
"mean_token_accuracy": 0.15471049398183823,
"num_tokens": 21651342.0,
"step": 10430
},
{
"entropy": 5.888284301757812,
"epoch": 0.9465711175616836,
"grad_norm": 2.0625,
"learning_rate": 0.0004917682146254017,
"loss": 5.6243,
"mean_token_accuracy": 0.16156017035245895,
"num_tokens": 21661043.0,
"step": 10435
},
{
"entropy": 5.774037456512451,
"epoch": 0.9470246734397678,
"grad_norm": 2.453125,
"learning_rate": 0.0004917595403255938,
"loss": 5.4839,
"mean_token_accuracy": 0.1669143870472908,
"num_tokens": 21671197.0,
"step": 10440
},
{
"entropy": 5.731167411804199,
"epoch": 0.947478229317852,
"grad_norm": 2.390625,
"learning_rate": 0.0004917508615431431,
"loss": 5.5763,
"mean_token_accuracy": 0.15985903590917588,
"num_tokens": 21682066.0,
"step": 10445
},
{
"entropy": 5.862526941299438,
"epoch": 0.9479317851959361,
"grad_norm": 2.25,
"learning_rate": 0.0004917421782782296,
"loss": 5.6135,
"mean_token_accuracy": 0.15685460567474366,
"num_tokens": 21692117.0,
"step": 10450
},
{
"entropy": 5.820763969421387,
"epoch": 0.9483853410740203,
"grad_norm": 2.9375,
"learning_rate": 0.0004917334905310325,
"loss": 5.7187,
"mean_token_accuracy": 0.1483694888651371,
"num_tokens": 21703544.0,
"step": 10455
},
{
"entropy": 5.784500217437744,
"epoch": 0.9488388969521045,
"grad_norm": 2.609375,
"learning_rate": 0.0004917247983017317,
"loss": 5.5451,
"mean_token_accuracy": 0.1575774610042572,
"num_tokens": 21712850.0,
"step": 10460
},
{
"entropy": 5.857076168060303,
"epoch": 0.9492924528301887,
"grad_norm": 4.46875,
"learning_rate": 0.0004917161015905067,
"loss": 5.7064,
"mean_token_accuracy": 0.1546853482723236,
"num_tokens": 21723015.0,
"step": 10465
},
{
"entropy": 5.780153369903564,
"epoch": 0.9497460087082729,
"grad_norm": 1.875,
"learning_rate": 0.0004917074003975376,
"loss": 5.6598,
"mean_token_accuracy": 0.15189022123813628,
"num_tokens": 21732453.0,
"step": 10470
},
{
"entropy": 5.802216529846191,
"epoch": 0.950199564586357,
"grad_norm": 2.59375,
"learning_rate": 0.0004916986947230042,
"loss": 5.7187,
"mean_token_accuracy": 0.15118529945611953,
"num_tokens": 21743251.0,
"step": 10475
},
{
"entropy": 5.887509822845459,
"epoch": 0.9506531204644412,
"grad_norm": 1.8125,
"learning_rate": 0.0004916899845670866,
"loss": 5.6239,
"mean_token_accuracy": 0.15586574971675873,
"num_tokens": 21752891.0,
"step": 10480
},
{
"entropy": 5.838762617111206,
"epoch": 0.9511066763425254,
"grad_norm": 1.6953125,
"learning_rate": 0.0004916812699299649,
"loss": 5.6682,
"mean_token_accuracy": 0.15595989674329758,
"num_tokens": 21763230.0,
"step": 10485
},
{
"entropy": 5.8394464492797855,
"epoch": 0.9515602322206096,
"grad_norm": 2.34375,
"learning_rate": 0.0004916725508118193,
"loss": 5.5997,
"mean_token_accuracy": 0.15375525653362274,
"num_tokens": 21773663.0,
"step": 10490
},
{
"entropy": 5.81770887374878,
"epoch": 0.9520137880986937,
"grad_norm": 2.09375,
"learning_rate": 0.0004916638272128301,
"loss": 5.6637,
"mean_token_accuracy": 0.15654143393039704,
"num_tokens": 21784265.0,
"step": 10495
},
{
"entropy": 5.745655345916748,
"epoch": 0.9524673439767779,
"grad_norm": 2.234375,
"learning_rate": 0.0004916550991331777,
"loss": 5.6368,
"mean_token_accuracy": 0.15793596804142,
"num_tokens": 21795056.0,
"step": 10500
},
{
"entropy": 5.8132641315460205,
"epoch": 0.9529208998548622,
"grad_norm": 2.4375,
"learning_rate": 0.0004916463665730427,
"loss": 5.6342,
"mean_token_accuracy": 0.15889952480793,
"num_tokens": 21806204.0,
"step": 10505
},
{
"entropy": 5.8509033679962155,
"epoch": 0.9533744557329463,
"grad_norm": 2.203125,
"learning_rate": 0.0004916376295326057,
"loss": 5.6527,
"mean_token_accuracy": 0.15502676367759705,
"num_tokens": 21816993.0,
"step": 10510
},
{
"entropy": 5.824394702911377,
"epoch": 0.9538280116110305,
"grad_norm": 2.75,
"learning_rate": 0.0004916288880120472,
"loss": 5.5126,
"mean_token_accuracy": 0.16500263810157775,
"num_tokens": 21827457.0,
"step": 10515
},
{
"entropy": 5.768416929244995,
"epoch": 0.9542815674891146,
"grad_norm": 1.9140625,
"learning_rate": 0.0004916201420115482,
"loss": 5.608,
"mean_token_accuracy": 0.15698783695697785,
"num_tokens": 21837555.0,
"step": 10520
},
{
"entropy": 5.842128801345825,
"epoch": 0.9547351233671988,
"grad_norm": 2.234375,
"learning_rate": 0.0004916113915312895,
"loss": 5.6454,
"mean_token_accuracy": 0.16147412061691285,
"num_tokens": 21847074.0,
"step": 10525
},
{
"entropy": 5.8470460891723635,
"epoch": 0.9551886792452831,
"grad_norm": 2.359375,
"learning_rate": 0.0004916026365714519,
"loss": 5.642,
"mean_token_accuracy": 0.1498659148812294,
"num_tokens": 21858333.0,
"step": 10530
},
{
"entropy": 5.770013284683228,
"epoch": 0.9556422351233672,
"grad_norm": 2.03125,
"learning_rate": 0.0004915938771322165,
"loss": 5.653,
"mean_token_accuracy": 0.15716380327939988,
"num_tokens": 21869132.0,
"step": 10535
},
{
"entropy": 5.758356142044067,
"epoch": 0.9560957910014514,
"grad_norm": 2.71875,
"learning_rate": 0.0004915851132137646,
"loss": 5.6858,
"mean_token_accuracy": 0.15203768908977508,
"num_tokens": 21879153.0,
"step": 10540
},
{
"entropy": 5.845894527435303,
"epoch": 0.9565493468795355,
"grad_norm": 1.9765625,
"learning_rate": 0.0004915763448162774,
"loss": 5.5592,
"mean_token_accuracy": 0.15751828402280807,
"num_tokens": 21889063.0,
"step": 10545
},
{
"entropy": 5.8422064781188965,
"epoch": 0.9570029027576198,
"grad_norm": 2.96875,
"learning_rate": 0.0004915675719399362,
"loss": 5.6833,
"mean_token_accuracy": 0.15525020360946656,
"num_tokens": 21899005.0,
"step": 10550
},
{
"entropy": 5.801091575622559,
"epoch": 0.9574564586357039,
"grad_norm": 3.0,
"learning_rate": 0.0004915587945849225,
"loss": 5.5429,
"mean_token_accuracy": 0.15986981689929963,
"num_tokens": 21908391.0,
"step": 10555
},
{
"entropy": 5.783149099349975,
"epoch": 0.9579100145137881,
"grad_norm": 3.890625,
"learning_rate": 0.0004915500127514176,
"loss": 5.626,
"mean_token_accuracy": 0.1556545153260231,
"num_tokens": 21917655.0,
"step": 10560
},
{
"entropy": 5.848690319061279,
"epoch": 0.9583635703918723,
"grad_norm": 2.0625,
"learning_rate": 0.0004915412264396034,
"loss": 5.6832,
"mean_token_accuracy": 0.1582992359995842,
"num_tokens": 21928038.0,
"step": 10565
},
{
"entropy": 5.786609077453614,
"epoch": 0.9588171262699564,
"grad_norm": 2.484375,
"learning_rate": 0.0004915324356496613,
"loss": 5.5746,
"mean_token_accuracy": 0.16353203654289244,
"num_tokens": 21938106.0,
"step": 10570
},
{
"entropy": 5.773234176635742,
"epoch": 0.9592706821480407,
"grad_norm": 2.578125,
"learning_rate": 0.0004915236403817733,
"loss": 5.5702,
"mean_token_accuracy": 0.16138877272605895,
"num_tokens": 21947484.0,
"step": 10575
},
{
"entropy": 5.814287567138672,
"epoch": 0.9597242380261248,
"grad_norm": 2.609375,
"learning_rate": 0.0004915148406361213,
"loss": 5.6499,
"mean_token_accuracy": 0.15784090012311935,
"num_tokens": 21957139.0,
"step": 10580
},
{
"entropy": 5.769003677368164,
"epoch": 0.960177793904209,
"grad_norm": 2.21875,
"learning_rate": 0.0004915060364128872,
"loss": 5.6391,
"mean_token_accuracy": 0.14913761019706726,
"num_tokens": 21967030.0,
"step": 10585
},
{
"entropy": 5.797343826293945,
"epoch": 0.9606313497822931,
"grad_norm": 1.8046875,
"learning_rate": 0.0004914972277122532,
"loss": 5.614,
"mean_token_accuracy": 0.15193152502179147,
"num_tokens": 21976418.0,
"step": 10590
},
{
"entropy": 5.77819242477417,
"epoch": 0.9610849056603774,
"grad_norm": 2.859375,
"learning_rate": 0.0004914884145344012,
"loss": 5.6184,
"mean_token_accuracy": 0.16035711616277695,
"num_tokens": 21987284.0,
"step": 10595
},
{
"entropy": 5.716995763778686,
"epoch": 0.9615384615384616,
"grad_norm": 2.25,
"learning_rate": 0.0004914795968795137,
"loss": 5.5251,
"mean_token_accuracy": 0.16215120255947113,
"num_tokens": 21997106.0,
"step": 10600
},
{
"entropy": 5.741790771484375,
"epoch": 0.9619920174165457,
"grad_norm": 2.625,
"learning_rate": 0.0004914707747477728,
"loss": 5.539,
"mean_token_accuracy": 0.1570558652281761,
"num_tokens": 22007589.0,
"step": 10605
},
{
"entropy": 5.734328365325927,
"epoch": 0.9624455732946299,
"grad_norm": 2.40625,
"learning_rate": 0.0004914619481393613,
"loss": 5.6412,
"mean_token_accuracy": 0.15544965565204621,
"num_tokens": 22017669.0,
"step": 10610
},
{
"entropy": 5.949207544326782,
"epoch": 0.962899129172714,
"grad_norm": 2.515625,
"learning_rate": 0.0004914531170544615,
"loss": 5.8122,
"mean_token_accuracy": 0.14744702577590943,
"num_tokens": 22029330.0,
"step": 10615
},
{
"entropy": 5.87768325805664,
"epoch": 0.9633526850507983,
"grad_norm": 2.484375,
"learning_rate": 0.0004914442814932561,
"loss": 5.5873,
"mean_token_accuracy": 0.15414217859506607,
"num_tokens": 22039877.0,
"step": 10620
},
{
"entropy": 5.795701313018799,
"epoch": 0.9638062409288825,
"grad_norm": 2.890625,
"learning_rate": 0.0004914354414559278,
"loss": 5.6064,
"mean_token_accuracy": 0.15927164107561112,
"num_tokens": 22049579.0,
"step": 10625
},
{
"entropy": 5.655880784988403,
"epoch": 0.9642597968069666,
"grad_norm": 2.125,
"learning_rate": 0.0004914265969426594,
"loss": 5.5232,
"mean_token_accuracy": 0.1695343554019928,
"num_tokens": 22059608.0,
"step": 10630
},
{
"entropy": 5.74952974319458,
"epoch": 0.9647133526850508,
"grad_norm": 2.84375,
"learning_rate": 0.0004914177479536338,
"loss": 5.6462,
"mean_token_accuracy": 0.15356021150946617,
"num_tokens": 22070036.0,
"step": 10635
},
{
"entropy": 5.7686145305633545,
"epoch": 0.965166908563135,
"grad_norm": 2.25,
"learning_rate": 0.000491408894489034,
"loss": 5.641,
"mean_token_accuracy": 0.15436567217111588,
"num_tokens": 22080679.0,
"step": 10640
},
{
"entropy": 5.805910730361939,
"epoch": 0.9656204644412192,
"grad_norm": 3.28125,
"learning_rate": 0.0004914000365490432,
"loss": 5.4927,
"mean_token_accuracy": 0.15782108306884765,
"num_tokens": 22092192.0,
"step": 10645
},
{
"entropy": 5.823235750198364,
"epoch": 0.9660740203193033,
"grad_norm": 2.96875,
"learning_rate": 0.0004913911741338445,
"loss": 5.6783,
"mean_token_accuracy": 0.15351950079202653,
"num_tokens": 22102722.0,
"step": 10650
},
{
"entropy": 5.76548810005188,
"epoch": 0.9665275761973875,
"grad_norm": 2.703125,
"learning_rate": 0.0004913823072436211,
"loss": 5.5783,
"mean_token_accuracy": 0.16051917225122453,
"num_tokens": 22113521.0,
"step": 10655
},
{
"entropy": 5.850063276290894,
"epoch": 0.9669811320754716,
"grad_norm": 2.1875,
"learning_rate": 0.0004913734358785565,
"loss": 5.6703,
"mean_token_accuracy": 0.15622948259115219,
"num_tokens": 22124891.0,
"step": 10660
},
{
"entropy": 5.794833660125732,
"epoch": 0.9674346879535559,
"grad_norm": 1.9765625,
"learning_rate": 0.0004913645600388341,
"loss": 5.6532,
"mean_token_accuracy": 0.15173319876194,
"num_tokens": 22135972.0,
"step": 10665
},
{
"entropy": 5.839829730987549,
"epoch": 0.9678882438316401,
"grad_norm": 2.71875,
"learning_rate": 0.0004913556797246374,
"loss": 5.7049,
"mean_token_accuracy": 0.1470730111002922,
"num_tokens": 22146623.0,
"step": 10670
},
{
"entropy": 5.847836494445801,
"epoch": 0.9683417997097242,
"grad_norm": 3.34375,
"learning_rate": 0.0004913467949361503,
"loss": 5.6013,
"mean_token_accuracy": 0.15360285937786103,
"num_tokens": 22157619.0,
"step": 10675
},
{
"entropy": 5.866746807098389,
"epoch": 0.9687953555878084,
"grad_norm": 3.140625,
"learning_rate": 0.0004913379056735561,
"loss": 5.6142,
"mean_token_accuracy": 0.1510605663061142,
"num_tokens": 22168655.0,
"step": 10680
},
{
"entropy": 5.754996824264526,
"epoch": 0.9692489114658926,
"grad_norm": 2.515625,
"learning_rate": 0.000491329011937039,
"loss": 5.651,
"mean_token_accuracy": 0.1540650822222233,
"num_tokens": 22178500.0,
"step": 10685
},
{
"entropy": 5.83361701965332,
"epoch": 0.9697024673439768,
"grad_norm": 1.9453125,
"learning_rate": 0.0004913201137267828,
"loss": 5.6513,
"mean_token_accuracy": 0.1479468896985054,
"num_tokens": 22188873.0,
"step": 10690
},
{
"entropy": 5.740324068069458,
"epoch": 0.970156023222061,
"grad_norm": 5.125,
"learning_rate": 0.0004913112110429715,
"loss": 5.5849,
"mean_token_accuracy": 0.15538954585790635,
"num_tokens": 22199743.0,
"step": 10695
},
{
"entropy": 5.790153455734253,
"epoch": 0.9706095791001451,
"grad_norm": 3.09375,
"learning_rate": 0.0004913023038857892,
"loss": 5.5389,
"mean_token_accuracy": 0.16194612979888917,
"num_tokens": 22210301.0,
"step": 10700
},
{
"entropy": 5.811965370178223,
"epoch": 0.9710631349782293,
"grad_norm": 2.828125,
"learning_rate": 0.0004912933922554203,
"loss": 5.547,
"mean_token_accuracy": 0.16348409205675124,
"num_tokens": 22220771.0,
"step": 10705
},
{
"entropy": 5.758093738555909,
"epoch": 0.9715166908563135,
"grad_norm": 2.78125,
"learning_rate": 0.0004912844761520487,
"loss": 5.6359,
"mean_token_accuracy": 0.15542892888188362,
"num_tokens": 22231100.0,
"step": 10710
},
{
"entropy": 5.736060333251953,
"epoch": 0.9719702467343977,
"grad_norm": 5.21875,
"learning_rate": 0.0004912755555758592,
"loss": 5.6462,
"mean_token_accuracy": 0.15574013888835908,
"num_tokens": 22240997.0,
"step": 10715
},
{
"entropy": 5.755614709854126,
"epoch": 0.9724238026124818,
"grad_norm": 2.6875,
"learning_rate": 0.000491266630527036,
"loss": 5.6414,
"mean_token_accuracy": 0.15689482688903808,
"num_tokens": 22251795.0,
"step": 10720
},
{
"entropy": 5.897975873947144,
"epoch": 0.972877358490566,
"grad_norm": 5.15625,
"learning_rate": 0.0004912577010057638,
"loss": 5.6928,
"mean_token_accuracy": 0.15143510773777963,
"num_tokens": 22262465.0,
"step": 10725
},
{
"entropy": 5.956141090393066,
"epoch": 0.9733309143686503,
"grad_norm": 3.328125,
"learning_rate": 0.0004912487670122273,
"loss": 5.7718,
"mean_token_accuracy": 0.15309564918279647,
"num_tokens": 22272399.0,
"step": 10730
},
{
"entropy": 5.857673025131225,
"epoch": 0.9737844702467344,
"grad_norm": 2.28125,
"learning_rate": 0.0004912398285466111,
"loss": 5.6959,
"mean_token_accuracy": 0.15608857721090316,
"num_tokens": 22282545.0,
"step": 10735
},
{
"entropy": 5.825756359100342,
"epoch": 0.9742380261248186,
"grad_norm": 2.328125,
"learning_rate": 0.0004912308856091002,
"loss": 5.6295,
"mean_token_accuracy": 0.16046428829431533,
"num_tokens": 22293154.0,
"step": 10740
},
{
"entropy": 5.809566068649292,
"epoch": 0.9746915820029027,
"grad_norm": 1.6484375,
"learning_rate": 0.0004912219381998795,
"loss": 5.5947,
"mean_token_accuracy": 0.15636386275291442,
"num_tokens": 22302309.0,
"step": 10745
},
{
"entropy": 5.797397756576538,
"epoch": 0.9751451378809869,
"grad_norm": 3.109375,
"learning_rate": 0.0004912129863191339,
"loss": 5.6919,
"mean_token_accuracy": 0.14868070632219316,
"num_tokens": 22312316.0,
"step": 10750
},
{
"entropy": 5.913840103149414,
"epoch": 0.9755986937590712,
"grad_norm": 2.171875,
"learning_rate": 0.0004912040299670487,
"loss": 5.67,
"mean_token_accuracy": 0.15575771033763885,
"num_tokens": 22322387.0,
"step": 10755
},
{
"entropy": 5.8599964618682865,
"epoch": 0.9760522496371553,
"grad_norm": 2.265625,
"learning_rate": 0.0004911950691438091,
"loss": 5.6383,
"mean_token_accuracy": 0.16185203492641448,
"num_tokens": 22332759.0,
"step": 10760
},
{
"entropy": 5.787653017044067,
"epoch": 0.9765058055152395,
"grad_norm": 2.140625,
"learning_rate": 0.0004911861038496004,
"loss": 5.6792,
"mean_token_accuracy": 0.15071214362978935,
"num_tokens": 22344056.0,
"step": 10765
},
{
"entropy": 5.744269943237304,
"epoch": 0.9769593613933236,
"grad_norm": 2.828125,
"learning_rate": 0.0004911771340846079,
"loss": 5.6302,
"mean_token_accuracy": 0.15291369706392288,
"num_tokens": 22353877.0,
"step": 10770
},
{
"entropy": 5.848117399215698,
"epoch": 0.9774129172714079,
"grad_norm": 1.7734375,
"learning_rate": 0.0004911681598490172,
"loss": 5.6493,
"mean_token_accuracy": 0.1582115352153778,
"num_tokens": 22363549.0,
"step": 10775
},
{
"entropy": 5.918741703033447,
"epoch": 0.977866473149492,
"grad_norm": 2.015625,
"learning_rate": 0.0004911591811430139,
"loss": 5.7284,
"mean_token_accuracy": 0.15085147470235824,
"num_tokens": 22374040.0,
"step": 10780
},
{
"entropy": 5.852246713638306,
"epoch": 0.9783200290275762,
"grad_norm": 2.140625,
"learning_rate": 0.0004911501979667836,
"loss": 5.7608,
"mean_token_accuracy": 0.14955372214317322,
"num_tokens": 22384819.0,
"step": 10785
},
{
"entropy": 5.825027418136597,
"epoch": 0.9787735849056604,
"grad_norm": 1.7734375,
"learning_rate": 0.0004911412103205122,
"loss": 5.6052,
"mean_token_accuracy": 0.15450286120176315,
"num_tokens": 22395130.0,
"step": 10790
},
{
"entropy": 5.7608222484588625,
"epoch": 0.9792271407837445,
"grad_norm": 3.015625,
"learning_rate": 0.0004911322182043855,
"loss": 5.6357,
"mean_token_accuracy": 0.15577088296413422,
"num_tokens": 22405093.0,
"step": 10795
},
{
"entropy": 5.802627611160278,
"epoch": 0.9796806966618288,
"grad_norm": 2.5,
"learning_rate": 0.0004911232216185893,
"loss": 5.6479,
"mean_token_accuracy": 0.15451014339923858,
"num_tokens": 22415008.0,
"step": 10800
},
{
"entropy": 5.888903570175171,
"epoch": 0.9801342525399129,
"grad_norm": 2.984375,
"learning_rate": 0.0004911142205633099,
"loss": 5.7106,
"mean_token_accuracy": 0.1523520439863205,
"num_tokens": 22426486.0,
"step": 10805
},
{
"entropy": 5.8365579605102536,
"epoch": 0.9805878084179971,
"grad_norm": 2.078125,
"learning_rate": 0.0004911052150387334,
"loss": 5.6146,
"mean_token_accuracy": 0.1567578583955765,
"num_tokens": 22436844.0,
"step": 10810
},
{
"entropy": 5.847742700576783,
"epoch": 0.9810413642960812,
"grad_norm": 2.265625,
"learning_rate": 0.0004910962050450458,
"loss": 5.7577,
"mean_token_accuracy": 0.1515177756547928,
"num_tokens": 22446630.0,
"step": 10815
},
{
"entropy": 5.832299661636353,
"epoch": 0.9814949201741655,
"grad_norm": 2.375,
"learning_rate": 0.0004910871905824338,
"loss": 5.5581,
"mean_token_accuracy": 0.15815195590257644,
"num_tokens": 22456406.0,
"step": 10820
},
{
"entropy": 5.80425329208374,
"epoch": 0.9819484760522497,
"grad_norm": 4.4375,
"learning_rate": 0.0004910781716510837,
"loss": 5.5762,
"mean_token_accuracy": 0.16052157580852508,
"num_tokens": 22466210.0,
"step": 10825
},
{
"entropy": 5.756770372390747,
"epoch": 0.9824020319303338,
"grad_norm": 2.25,
"learning_rate": 0.0004910691482511817,
"loss": 5.6691,
"mean_token_accuracy": 0.1583290547132492,
"num_tokens": 22476978.0,
"step": 10830
},
{
"entropy": 5.80738205909729,
"epoch": 0.982855587808418,
"grad_norm": 2.390625,
"learning_rate": 0.0004910601203829148,
"loss": 5.6339,
"mean_token_accuracy": 0.15294980108737946,
"num_tokens": 22488613.0,
"step": 10835
},
{
"entropy": 5.817235898971558,
"epoch": 0.9833091436865021,
"grad_norm": 3.21875,
"learning_rate": 0.0004910510880464696,
"loss": 5.5322,
"mean_token_accuracy": 0.1689889743924141,
"num_tokens": 22498312.0,
"step": 10840
},
{
"entropy": 5.792183303833008,
"epoch": 0.9837626995645864,
"grad_norm": 3.53125,
"learning_rate": 0.0004910420512420327,
"loss": 5.6531,
"mean_token_accuracy": 0.15933890044689178,
"num_tokens": 22508131.0,
"step": 10845
},
{
"entropy": 5.788613033294678,
"epoch": 0.9842162554426706,
"grad_norm": 2.828125,
"learning_rate": 0.0004910330099697913,
"loss": 5.6496,
"mean_token_accuracy": 0.15769676193594934,
"num_tokens": 22517842.0,
"step": 10850
},
{
"entropy": 5.830097818374634,
"epoch": 0.9846698113207547,
"grad_norm": 2.578125,
"learning_rate": 0.0004910239642299321,
"loss": 5.62,
"mean_token_accuracy": 0.15675847083330155,
"num_tokens": 22528239.0,
"step": 10855
},
{
"entropy": 5.806100749969483,
"epoch": 0.9851233671988389,
"grad_norm": 2.765625,
"learning_rate": 0.0004910149140226422,
"loss": 5.5954,
"mean_token_accuracy": 0.15444757640361786,
"num_tokens": 22539734.0,
"step": 10860
},
{
"entropy": 5.815021419525147,
"epoch": 0.9855769230769231,
"grad_norm": 2.640625,
"learning_rate": 0.0004910058593481089,
"loss": 5.6113,
"mean_token_accuracy": 0.16095956712961196,
"num_tokens": 22549796.0,
"step": 10865
},
{
"entropy": 5.739499568939209,
"epoch": 0.9860304789550073,
"grad_norm": 3.8125,
"learning_rate": 0.0004909968002065195,
"loss": 5.6115,
"mean_token_accuracy": 0.15544536262750625,
"num_tokens": 22559934.0,
"step": 10870
},
{
"entropy": 5.8420631885528564,
"epoch": 0.9864840348330914,
"grad_norm": 1.9921875,
"learning_rate": 0.000490987736598061,
"loss": 5.6889,
"mean_token_accuracy": 0.14803468361496924,
"num_tokens": 22569864.0,
"step": 10875
},
{
"entropy": 5.79959135055542,
"epoch": 0.9869375907111756,
"grad_norm": 4.15625,
"learning_rate": 0.0004909786685229211,
"loss": 5.633,
"mean_token_accuracy": 0.15979243367910384,
"num_tokens": 22580687.0,
"step": 10880
},
{
"entropy": 5.820346117019653,
"epoch": 0.9873911465892597,
"grad_norm": 2.6875,
"learning_rate": 0.0004909695959812875,
"loss": 5.6595,
"mean_token_accuracy": 0.15569808185100556,
"num_tokens": 22590507.0,
"step": 10885
},
{
"entropy": 5.835328531265259,
"epoch": 0.987844702467344,
"grad_norm": 2.859375,
"learning_rate": 0.0004909605189733474,
"loss": 5.709,
"mean_token_accuracy": 0.15876459032297136,
"num_tokens": 22601901.0,
"step": 10890
},
{
"entropy": 5.769439935684204,
"epoch": 0.9882982583454282,
"grad_norm": 2.453125,
"learning_rate": 0.0004909514374992889,
"loss": 5.6411,
"mean_token_accuracy": 0.153889998793602,
"num_tokens": 22613003.0,
"step": 10895
},
{
"entropy": 5.782855749130249,
"epoch": 0.9887518142235123,
"grad_norm": 2.09375,
"learning_rate": 0.0004909423515592996,
"loss": 5.4984,
"mean_token_accuracy": 0.16059116572141646,
"num_tokens": 22623401.0,
"step": 10900
},
{
"entropy": 5.829384994506836,
"epoch": 0.9892053701015965,
"grad_norm": 3.3125,
"learning_rate": 0.0004909332611535675,
"loss": 5.6239,
"mean_token_accuracy": 0.1545998379588127,
"num_tokens": 22633147.0,
"step": 10905
},
{
"entropy": 5.774903059005737,
"epoch": 0.9896589259796807,
"grad_norm": 2.875,
"learning_rate": 0.0004909241662822804,
"loss": 5.492,
"mean_token_accuracy": 0.1706838309764862,
"num_tokens": 22642990.0,
"step": 10910
},
{
"entropy": 5.78953857421875,
"epoch": 0.9901124818577649,
"grad_norm": 3.109375,
"learning_rate": 0.0004909150669456265,
"loss": 5.6892,
"mean_token_accuracy": 0.14994772523641586,
"num_tokens": 22652734.0,
"step": 10915
},
{
"entropy": 5.848347282409668,
"epoch": 0.9905660377358491,
"grad_norm": 3.296875,
"learning_rate": 0.000490905963143794,
"loss": 5.7174,
"mean_token_accuracy": 0.14648480713367462,
"num_tokens": 22662968.0,
"step": 10920
},
{
"entropy": 5.813489961624145,
"epoch": 0.9910195936139332,
"grad_norm": 3.171875,
"learning_rate": 0.0004908968548769713,
"loss": 5.5456,
"mean_token_accuracy": 0.16475504338741304,
"num_tokens": 22673157.0,
"step": 10925
},
{
"entropy": 5.778944778442383,
"epoch": 0.9914731494920174,
"grad_norm": 2.390625,
"learning_rate": 0.0004908877421453466,
"loss": 5.6798,
"mean_token_accuracy": 0.15566922426223756,
"num_tokens": 22683105.0,
"step": 10930
},
{
"entropy": 5.80912823677063,
"epoch": 0.9919267053701016,
"grad_norm": 3.53125,
"learning_rate": 0.0004908786249491082,
"loss": 5.6489,
"mean_token_accuracy": 0.15920983254909515,
"num_tokens": 22693793.0,
"step": 10935
},
{
"entropy": 5.945695972442627,
"epoch": 0.9923802612481858,
"grad_norm": 2.21875,
"learning_rate": 0.000490869503288445,
"loss": 5.6744,
"mean_token_accuracy": 0.15801735818386078,
"num_tokens": 22703884.0,
"step": 10940
},
{
"entropy": 5.860371923446655,
"epoch": 0.99283381712627,
"grad_norm": 1.625,
"learning_rate": 0.0004908603771635454,
"loss": 5.6948,
"mean_token_accuracy": 0.15628961175680162,
"num_tokens": 22713699.0,
"step": 10945
},
{
"entropy": 5.845581674575806,
"epoch": 0.9932873730043541,
"grad_norm": 2.0,
"learning_rate": 0.0004908512465745982,
"loss": 5.5425,
"mean_token_accuracy": 0.16390856802463533,
"num_tokens": 22722871.0,
"step": 10950
},
{
"entropy": 5.7688634395599365,
"epoch": 0.9937409288824384,
"grad_norm": 2.140625,
"learning_rate": 0.0004908421115217922,
"loss": 5.5698,
"mean_token_accuracy": 0.15266856402158738,
"num_tokens": 22733314.0,
"step": 10955
},
{
"entropy": 5.730173349380493,
"epoch": 0.9941944847605225,
"grad_norm": 2.703125,
"learning_rate": 0.0004908329720053164,
"loss": 5.5687,
"mean_token_accuracy": 0.1642436295747757,
"num_tokens": 22743255.0,
"step": 10960
},
{
"entropy": 5.741264533996582,
"epoch": 0.9946480406386067,
"grad_norm": 3.265625,
"learning_rate": 0.0004908238280253597,
"loss": 5.561,
"mean_token_accuracy": 0.16317612677812576,
"num_tokens": 22753857.0,
"step": 10965
},
{
"entropy": 5.8027266502380375,
"epoch": 0.9951015965166908,
"grad_norm": 2.8125,
"learning_rate": 0.0004908146795821112,
"loss": 5.5438,
"mean_token_accuracy": 0.15576230436563493,
"num_tokens": 22764326.0,
"step": 10970
},
{
"entropy": 5.792645835876465,
"epoch": 0.995555152394775,
"grad_norm": 2.359375,
"learning_rate": 0.0004908055266757602,
"loss": 5.6373,
"mean_token_accuracy": 0.1557354211807251,
"num_tokens": 22774958.0,
"step": 10975
},
{
"entropy": 5.760008525848389,
"epoch": 0.9960087082728593,
"grad_norm": 2.65625,
"learning_rate": 0.0004907963693064958,
"loss": 5.5343,
"mean_token_accuracy": 0.15991889238357543,
"num_tokens": 22785616.0,
"step": 10980
},
{
"entropy": 5.808274126052856,
"epoch": 0.9964622641509434,
"grad_norm": 2.703125,
"learning_rate": 0.0004907872074745075,
"loss": 5.7053,
"mean_token_accuracy": 0.14998844712972642,
"num_tokens": 22795967.0,
"step": 10985
},
{
"entropy": 5.737427473068237,
"epoch": 0.9969158200290276,
"grad_norm": 3.640625,
"learning_rate": 0.0004907780411799849,
"loss": 5.7098,
"mean_token_accuracy": 0.1508146420121193,
"num_tokens": 22806335.0,
"step": 10990
},
{
"entropy": 5.877075958251953,
"epoch": 0.9973693759071117,
"grad_norm": 3.390625,
"learning_rate": 0.0004907688704231173,
"loss": 5.6294,
"mean_token_accuracy": 0.1598728522658348,
"num_tokens": 22816390.0,
"step": 10995
},
{
"entropy": 5.855796384811401,
"epoch": 0.997822931785196,
"grad_norm": 1.8984375,
"learning_rate": 0.0004907596952040945,
"loss": 5.7025,
"mean_token_accuracy": 0.15267188623547553,
"num_tokens": 22826395.0,
"step": 11000
},
{
"entropy": 5.836028337478638,
"epoch": 0.9982764876632801,
"grad_norm": 2.515625,
"learning_rate": 0.0004907505155231063,
"loss": 5.685,
"mean_token_accuracy": 0.1540266364812851,
"num_tokens": 22836790.0,
"step": 11005
},
{
"entropy": 5.788539075851441,
"epoch": 0.9987300435413643,
"grad_norm": 2.859375,
"learning_rate": 0.0004907413313803424,
"loss": 5.5793,
"mean_token_accuracy": 0.1512584000825882,
"num_tokens": 22846687.0,
"step": 11010
},
{
"entropy": 5.874964427947998,
"epoch": 0.9991835994194485,
"grad_norm": 2.765625,
"learning_rate": 0.0004907321427759928,
"loss": 5.5969,
"mean_token_accuracy": 0.16132215410470963,
"num_tokens": 22857208.0,
"step": 11015
},
{
"entropy": 5.848848915100097,
"epoch": 0.9996371552975326,
"grad_norm": 2.78125,
"learning_rate": 0.0004907229497102475,
"loss": 5.725,
"mean_token_accuracy": 0.1487257033586502,
"num_tokens": 22868556.0,
"step": 11020
},
{
"entropy": 5.865156602859497,
"epoch": 1.0000907111756168,
"grad_norm": 2.796875,
"learning_rate": 0.0004907137521832966,
"loss": 5.6894,
"mean_token_accuracy": 0.15483292788267136,
"num_tokens": 22880081.0,
"step": 11025
},
{
"entropy": 5.830396747589111,
"epoch": 1.000544267053701,
"grad_norm": 2.453125,
"learning_rate": 0.0004907045501953304,
"loss": 5.5247,
"mean_token_accuracy": 0.16311182975769042,
"num_tokens": 22892141.0,
"step": 11030
},
{
"entropy": 5.803026151657105,
"epoch": 1.0009978229317853,
"grad_norm": 5.34375,
"learning_rate": 0.000490695343746539,
"loss": 5.5888,
"mean_token_accuracy": 0.16324917674064637,
"num_tokens": 22902688.0,
"step": 11035
},
{
"entropy": 5.696486568450927,
"epoch": 1.0014513788098693,
"grad_norm": 2.078125,
"learning_rate": 0.0004906861328371131,
"loss": 5.4996,
"mean_token_accuracy": 0.158633553981781,
"num_tokens": 22912453.0,
"step": 11040
},
{
"entropy": 5.8159349918365475,
"epoch": 1.0019049346879536,
"grad_norm": 2.453125,
"learning_rate": 0.0004906769174672429,
"loss": 5.6483,
"mean_token_accuracy": 0.15360057651996611,
"num_tokens": 22923167.0,
"step": 11045
},
{
"entropy": 5.924277257919312,
"epoch": 1.0023584905660377,
"grad_norm": 4.53125,
"learning_rate": 0.000490667697637119,
"loss": 5.6311,
"mean_token_accuracy": 0.15385877937078477,
"num_tokens": 22934753.0,
"step": 11050
},
{
"entropy": 5.7873869895935055,
"epoch": 1.002812046444122,
"grad_norm": 2.40625,
"learning_rate": 0.0004906584733469324,
"loss": 5.5552,
"mean_token_accuracy": 0.16017444133758546,
"num_tokens": 22945277.0,
"step": 11055
},
{
"entropy": 5.794849157333374,
"epoch": 1.0032656023222062,
"grad_norm": 2.53125,
"learning_rate": 0.0004906492445968735,
"loss": 5.6275,
"mean_token_accuracy": 0.14777041599154472,
"num_tokens": 22955797.0,
"step": 11060
},
{
"entropy": 5.851789331436157,
"epoch": 1.0037191582002902,
"grad_norm": 3.484375,
"learning_rate": 0.0004906400113871332,
"loss": 5.5971,
"mean_token_accuracy": 0.15662909150123597,
"num_tokens": 22965969.0,
"step": 11065
},
{
"entropy": 5.816144752502441,
"epoch": 1.0041727140783745,
"grad_norm": 2.734375,
"learning_rate": 0.0004906307737179026,
"loss": 5.5523,
"mean_token_accuracy": 0.1689614936709404,
"num_tokens": 22975323.0,
"step": 11070
},
{
"entropy": 5.771370553970337,
"epoch": 1.0046262699564585,
"grad_norm": 2.609375,
"learning_rate": 0.0004906215315893726,
"loss": 5.5812,
"mean_token_accuracy": 0.1592397302389145,
"num_tokens": 22985063.0,
"step": 11075
},
{
"entropy": 5.851737546920776,
"epoch": 1.0050798258345428,
"grad_norm": 4.375,
"learning_rate": 0.0004906122850017344,
"loss": 5.6981,
"mean_token_accuracy": 0.15012871325016022,
"num_tokens": 22996001.0,
"step": 11080
},
{
"entropy": 5.851078081130981,
"epoch": 1.005533381712627,
"grad_norm": 2.4375,
"learning_rate": 0.0004906030339551792,
"loss": 5.6094,
"mean_token_accuracy": 0.15733970254659652,
"num_tokens": 23006896.0,
"step": 11085
},
{
"entropy": 5.830320119857788,
"epoch": 1.0059869375907111,
"grad_norm": 2.703125,
"learning_rate": 0.0004905937784498983,
"loss": 5.5859,
"mean_token_accuracy": 0.15726361870765687,
"num_tokens": 23017080.0,
"step": 11090
},
{
"entropy": 5.812807703018189,
"epoch": 1.0064404934687954,
"grad_norm": 3.5625,
"learning_rate": 0.0004905845184860831,
"loss": 5.5377,
"mean_token_accuracy": 0.16178667098283767,
"num_tokens": 23027168.0,
"step": 11095
},
{
"entropy": 5.810239839553833,
"epoch": 1.0068940493468794,
"grad_norm": 2.46875,
"learning_rate": 0.0004905752540639252,
"loss": 5.5612,
"mean_token_accuracy": 0.1557539463043213,
"num_tokens": 23037364.0,
"step": 11100
},
{
"entropy": 5.766785526275635,
"epoch": 1.0073476052249637,
"grad_norm": 2.65625,
"learning_rate": 0.0004905659851836161,
"loss": 5.487,
"mean_token_accuracy": 0.16209156960248947,
"num_tokens": 23047609.0,
"step": 11105
},
{
"entropy": 5.8113995552062985,
"epoch": 1.007801161103048,
"grad_norm": 1.9296875,
"learning_rate": 0.0004905567118453475,
"loss": 5.6088,
"mean_token_accuracy": 0.15477470606565474,
"num_tokens": 23057942.0,
"step": 11110
},
{
"entropy": 5.782913875579834,
"epoch": 1.008254716981132,
"grad_norm": 2.28125,
"learning_rate": 0.0004905474340493111,
"loss": 5.5295,
"mean_token_accuracy": 0.15781660974025727,
"num_tokens": 23069119.0,
"step": 11115
},
{
"entropy": 5.712530899047851,
"epoch": 1.0087082728592163,
"grad_norm": 2.515625,
"learning_rate": 0.0004905381517956988,
"loss": 5.517,
"mean_token_accuracy": 0.16303566098213196,
"num_tokens": 23080060.0,
"step": 11120
},
{
"entropy": 5.689813899993896,
"epoch": 1.0091618287373005,
"grad_norm": 2.640625,
"learning_rate": 0.0004905288650847027,
"loss": 5.5316,
"mean_token_accuracy": 0.15620430260896684,
"num_tokens": 23090645.0,
"step": 11125
},
{
"entropy": 5.779735660552978,
"epoch": 1.0096153846153846,
"grad_norm": 2.203125,
"learning_rate": 0.0004905195739165146,
"loss": 5.5688,
"mean_token_accuracy": 0.16028851717710496,
"num_tokens": 23100449.0,
"step": 11130
},
{
"entropy": 5.780767917633057,
"epoch": 1.0100689404934688,
"grad_norm": 3.015625,
"learning_rate": 0.000490510278291327,
"loss": 5.5475,
"mean_token_accuracy": 0.15569900572299958,
"num_tokens": 23110415.0,
"step": 11135
},
{
"entropy": 5.787471866607666,
"epoch": 1.010522496371553,
"grad_norm": 2.46875,
"learning_rate": 0.0004905009782093318,
"loss": 5.5464,
"mean_token_accuracy": 0.1608660787343979,
"num_tokens": 23120336.0,
"step": 11140
},
{
"entropy": 5.753488683700562,
"epoch": 1.0109760522496372,
"grad_norm": 2.609375,
"learning_rate": 0.0004904916736707214,
"loss": 5.6233,
"mean_token_accuracy": 0.15254369676113128,
"num_tokens": 23130568.0,
"step": 11145
},
{
"entropy": 5.794592237472534,
"epoch": 1.0114296081277214,
"grad_norm": 2.671875,
"learning_rate": 0.0004904823646756885,
"loss": 5.558,
"mean_token_accuracy": 0.1587290346622467,
"num_tokens": 23140080.0,
"step": 11150
},
{
"entropy": 5.794294214248657,
"epoch": 1.0118831640058055,
"grad_norm": 2.96875,
"learning_rate": 0.0004904730512244252,
"loss": 5.5304,
"mean_token_accuracy": 0.1569518581032753,
"num_tokens": 23150360.0,
"step": 11155
},
{
"entropy": 5.770283842086792,
"epoch": 1.0123367198838897,
"grad_norm": 2.140625,
"learning_rate": 0.0004904637333171243,
"loss": 5.5292,
"mean_token_accuracy": 0.1553722381591797,
"num_tokens": 23160620.0,
"step": 11160
},
{
"entropy": 5.8569262504577635,
"epoch": 1.0127902757619738,
"grad_norm": 2.53125,
"learning_rate": 0.0004904544109539785,
"loss": 5.6203,
"mean_token_accuracy": 0.15243971943855286,
"num_tokens": 23171920.0,
"step": 11165
},
{
"entropy": 5.776814937591553,
"epoch": 1.013243831640058,
"grad_norm": 2.640625,
"learning_rate": 0.0004904450841351805,
"loss": 5.5694,
"mean_token_accuracy": 0.1612408071756363,
"num_tokens": 23182929.0,
"step": 11170
},
{
"entropy": 5.814702081680298,
"epoch": 1.0136973875181423,
"grad_norm": 2.234375,
"learning_rate": 0.0004904357528609236,
"loss": 5.5286,
"mean_token_accuracy": 0.15563077628612518,
"num_tokens": 23193020.0,
"step": 11175
},
{
"entropy": 5.7581392288208,
"epoch": 1.0141509433962264,
"grad_norm": 2.515625,
"learning_rate": 0.0004904264171314,
"loss": 5.5425,
"mean_token_accuracy": 0.16192737072706223,
"num_tokens": 23204073.0,
"step": 11180
},
{
"entropy": 5.703367280960083,
"epoch": 1.0146044992743106,
"grad_norm": 3.046875,
"learning_rate": 0.0004904170769468035,
"loss": 5.455,
"mean_token_accuracy": 0.15856247693300246,
"num_tokens": 23214376.0,
"step": 11185
},
{
"entropy": 5.829725122451782,
"epoch": 1.0150580551523947,
"grad_norm": 2.78125,
"learning_rate": 0.0004904077323073268,
"loss": 5.6171,
"mean_token_accuracy": 0.15695907771587372,
"num_tokens": 23224304.0,
"step": 11190
},
{
"entropy": 5.8234271049499515,
"epoch": 1.015511611030479,
"grad_norm": 2.75,
"learning_rate": 0.0004903983832131633,
"loss": 5.6157,
"mean_token_accuracy": 0.1551354482769966,
"num_tokens": 23235324.0,
"step": 11195
},
{
"entropy": 5.830600929260254,
"epoch": 1.0159651669085632,
"grad_norm": 2.890625,
"learning_rate": 0.0004903890296645063,
"loss": 5.6004,
"mean_token_accuracy": 0.15357705280184747,
"num_tokens": 23245774.0,
"step": 11200
},
{
"entropy": 5.828565788269043,
"epoch": 1.0164187227866472,
"grad_norm": 4.8125,
"learning_rate": 0.0004903796716615495,
"loss": 5.544,
"mean_token_accuracy": 0.15702730417251587,
"num_tokens": 23256274.0,
"step": 11205
},
{
"entropy": 5.766544246673584,
"epoch": 1.0168722786647315,
"grad_norm": 2.71875,
"learning_rate": 0.0004903703092044861,
"loss": 5.5615,
"mean_token_accuracy": 0.15786381363868712,
"num_tokens": 23266304.0,
"step": 11210
},
{
"entropy": 5.700562572479248,
"epoch": 1.0173258345428158,
"grad_norm": 3.140625,
"learning_rate": 0.0004903609422935097,
"loss": 5.4626,
"mean_token_accuracy": 0.16196913570165633,
"num_tokens": 23276552.0,
"step": 11215
},
{
"entropy": 5.816535758972168,
"epoch": 1.0177793904208998,
"grad_norm": 2.328125,
"learning_rate": 0.0004903515709288143,
"loss": 5.5879,
"mean_token_accuracy": 0.1515471652150154,
"num_tokens": 23287799.0,
"step": 11220
},
{
"entropy": 5.743448162078858,
"epoch": 1.018232946298984,
"grad_norm": 2.15625,
"learning_rate": 0.0004903421951105933,
"loss": 5.4336,
"mean_token_accuracy": 0.16320459693670272,
"num_tokens": 23297958.0,
"step": 11225
},
{
"entropy": 5.75546817779541,
"epoch": 1.0186865021770681,
"grad_norm": 2.546875,
"learning_rate": 0.0004903328148390411,
"loss": 5.6199,
"mean_token_accuracy": 0.15796598941087722,
"num_tokens": 23307523.0,
"step": 11230
},
{
"entropy": 5.872856283187867,
"epoch": 1.0191400580551524,
"grad_norm": 2.40625,
"learning_rate": 0.0004903234301143513,
"loss": 5.6053,
"mean_token_accuracy": 0.15784354209899903,
"num_tokens": 23317970.0,
"step": 11235
},
{
"entropy": 5.7929973125457765,
"epoch": 1.0195936139332367,
"grad_norm": 1.9296875,
"learning_rate": 0.0004903140409367181,
"loss": 5.5726,
"mean_token_accuracy": 0.16449209153652192,
"num_tokens": 23328612.0,
"step": 11240
},
{
"entropy": 5.77749195098877,
"epoch": 1.0200471698113207,
"grad_norm": 1.7890625,
"learning_rate": 0.0004903046473063357,
"loss": 5.5229,
"mean_token_accuracy": 0.16105701476335527,
"num_tokens": 23338104.0,
"step": 11245
},
{
"entropy": 5.756003093719483,
"epoch": 1.020500725689405,
"grad_norm": 2.109375,
"learning_rate": 0.0004902952492233983,
"loss": 5.481,
"mean_token_accuracy": 0.16933009028434753,
"num_tokens": 23348105.0,
"step": 11250
},
{
"entropy": 5.69498348236084,
"epoch": 1.020954281567489,
"grad_norm": 1.734375,
"learning_rate": 0.0004902858466881002,
"loss": 5.4468,
"mean_token_accuracy": 0.16551368683576584,
"num_tokens": 23357150.0,
"step": 11255
},
{
"entropy": 5.802523422241211,
"epoch": 1.0214078374455733,
"grad_norm": 3.296875,
"learning_rate": 0.000490276439700636,
"loss": 5.6272,
"mean_token_accuracy": 0.15664812475442885,
"num_tokens": 23367223.0,
"step": 11260
},
{
"entropy": 5.800776100158691,
"epoch": 1.0218613933236576,
"grad_norm": 2.875,
"learning_rate": 0.0004902670282612002,
"loss": 5.5596,
"mean_token_accuracy": 0.1589964099228382,
"num_tokens": 23377664.0,
"step": 11265
},
{
"entropy": 5.713521337509155,
"epoch": 1.0223149492017416,
"grad_norm": 2.203125,
"learning_rate": 0.0004902576123699873,
"loss": 5.4541,
"mean_token_accuracy": 0.16042310297489165,
"num_tokens": 23388057.0,
"step": 11270
},
{
"entropy": 5.670312547683716,
"epoch": 1.0227685050798259,
"grad_norm": 2.359375,
"learning_rate": 0.0004902481920271922,
"loss": 5.4181,
"mean_token_accuracy": 0.16238413229584694,
"num_tokens": 23397808.0,
"step": 11275
},
{
"entropy": 5.703090810775757,
"epoch": 1.02322206095791,
"grad_norm": 2.78125,
"learning_rate": 0.0004902387672330097,
"loss": 5.4935,
"mean_token_accuracy": 0.1601045623421669,
"num_tokens": 23406979.0,
"step": 11280
},
{
"entropy": 5.734795808792114,
"epoch": 1.0236756168359942,
"grad_norm": 2.328125,
"learning_rate": 0.0004902293379876345,
"loss": 5.519,
"mean_token_accuracy": 0.16237116158008574,
"num_tokens": 23416545.0,
"step": 11285
},
{
"entropy": 5.746735763549805,
"epoch": 1.0241291727140784,
"grad_norm": 2.21875,
"learning_rate": 0.0004902199042912618,
"loss": 5.5738,
"mean_token_accuracy": 0.15182949900627135,
"num_tokens": 23427464.0,
"step": 11290
},
{
"entropy": 5.789399719238281,
"epoch": 1.0245827285921625,
"grad_norm": 2.421875,
"learning_rate": 0.0004902104661440868,
"loss": 5.5789,
"mean_token_accuracy": 0.1616011753678322,
"num_tokens": 23436690.0,
"step": 11295
},
{
"entropy": 5.785635852813721,
"epoch": 1.0250362844702468,
"grad_norm": 5.75,
"learning_rate": 0.0004902010235463042,
"loss": 5.5021,
"mean_token_accuracy": 0.16081411242485047,
"num_tokens": 23446701.0,
"step": 11300
},
{
"entropy": 5.862888479232788,
"epoch": 1.025489840348331,
"grad_norm": 2.890625,
"learning_rate": 0.0004901915764981098,
"loss": 5.5849,
"mean_token_accuracy": 0.15415136218070985,
"num_tokens": 23456966.0,
"step": 11305
},
{
"entropy": 5.783653259277344,
"epoch": 1.025943396226415,
"grad_norm": 2.109375,
"learning_rate": 0.0004901821249996988,
"loss": 5.5786,
"mean_token_accuracy": 0.15850554257631302,
"num_tokens": 23467300.0,
"step": 11310
},
{
"entropy": 5.791384220123291,
"epoch": 1.0263969521044993,
"grad_norm": 2.515625,
"learning_rate": 0.0004901726690512665,
"loss": 5.5952,
"mean_token_accuracy": 0.15290943831205367,
"num_tokens": 23478457.0,
"step": 11315
},
{
"entropy": 5.7736084938049315,
"epoch": 1.0268505079825834,
"grad_norm": 4.53125,
"learning_rate": 0.0004901632086530087,
"loss": 5.5072,
"mean_token_accuracy": 0.16404951214790345,
"num_tokens": 23488066.0,
"step": 11320
},
{
"entropy": 5.807363986968994,
"epoch": 1.0273040638606676,
"grad_norm": 2.28125,
"learning_rate": 0.0004901537438051207,
"loss": 5.5865,
"mean_token_accuracy": 0.15398266762495041,
"num_tokens": 23498856.0,
"step": 11325
},
{
"entropy": 5.756262969970703,
"epoch": 1.027757619738752,
"grad_norm": 2.8125,
"learning_rate": 0.0004901442745077986,
"loss": 5.5447,
"mean_token_accuracy": 0.15879352241754532,
"num_tokens": 23509845.0,
"step": 11330
},
{
"entropy": 5.823694276809692,
"epoch": 1.028211175616836,
"grad_norm": 2.921875,
"learning_rate": 0.0004901348007612381,
"loss": 5.6804,
"mean_token_accuracy": 0.1574851855635643,
"num_tokens": 23520322.0,
"step": 11335
},
{
"entropy": 5.853392314910889,
"epoch": 1.0286647314949202,
"grad_norm": 4.28125,
"learning_rate": 0.0004901253225656351,
"loss": 5.5748,
"mean_token_accuracy": 0.16039203256368637,
"num_tokens": 23530030.0,
"step": 11340
},
{
"entropy": 5.866197681427002,
"epoch": 1.0291182873730043,
"grad_norm": 2.6875,
"learning_rate": 0.0004901158399211855,
"loss": 5.5978,
"mean_token_accuracy": 0.15863490998744964,
"num_tokens": 23541063.0,
"step": 11345
},
{
"entropy": 5.719363069534301,
"epoch": 1.0295718432510885,
"grad_norm": 2.125,
"learning_rate": 0.0004901063528280856,
"loss": 5.5453,
"mean_token_accuracy": 0.1542106255888939,
"num_tokens": 23551693.0,
"step": 11350
},
{
"entropy": 5.803479194641113,
"epoch": 1.0300253991291728,
"grad_norm": 4.78125,
"learning_rate": 0.0004900968612865314,
"loss": 5.576,
"mean_token_accuracy": 0.1636955976486206,
"num_tokens": 23561758.0,
"step": 11355
},
{
"entropy": 5.760063648223877,
"epoch": 1.0304789550072568,
"grad_norm": 4.90625,
"learning_rate": 0.0004900873652967195,
"loss": 5.5147,
"mean_token_accuracy": 0.1550989605486393,
"num_tokens": 23571912.0,
"step": 11360
},
{
"entropy": 5.739124202728272,
"epoch": 1.030932510885341,
"grad_norm": 2.75,
"learning_rate": 0.0004900778648588459,
"loss": 5.5181,
"mean_token_accuracy": 0.15988562256097794,
"num_tokens": 23582600.0,
"step": 11365
},
{
"entropy": 5.735342407226563,
"epoch": 1.0313860667634251,
"grad_norm": 2.5625,
"learning_rate": 0.0004900683599731073,
"loss": 5.4133,
"mean_token_accuracy": 0.1681668356060982,
"num_tokens": 23592761.0,
"step": 11370
},
{
"entropy": 5.811415195465088,
"epoch": 1.0318396226415094,
"grad_norm": 2.203125,
"learning_rate": 0.0004900588506397002,
"loss": 5.6334,
"mean_token_accuracy": 0.16160874366760253,
"num_tokens": 23603119.0,
"step": 11375
},
{
"entropy": 5.729310464859009,
"epoch": 1.0322931785195937,
"grad_norm": 2.890625,
"learning_rate": 0.0004900493368588213,
"loss": 5.507,
"mean_token_accuracy": 0.16110349595546722,
"num_tokens": 23613916.0,
"step": 11380
},
{
"entropy": 5.793110656738281,
"epoch": 1.0327467343976777,
"grad_norm": 3.96875,
"learning_rate": 0.0004900398186306672,
"loss": 5.6471,
"mean_token_accuracy": 0.15355589240789413,
"num_tokens": 23623956.0,
"step": 11385
},
{
"entropy": 5.777653598785401,
"epoch": 1.033200290275762,
"grad_norm": 2.1875,
"learning_rate": 0.000490030295955435,
"loss": 5.5648,
"mean_token_accuracy": 0.1563473090529442,
"num_tokens": 23634666.0,
"step": 11390
},
{
"entropy": 5.7300341606140135,
"epoch": 1.0336538461538463,
"grad_norm": 2.53125,
"learning_rate": 0.0004900207688333215,
"loss": 5.5063,
"mean_token_accuracy": 0.15261198133230208,
"num_tokens": 23645601.0,
"step": 11395
},
{
"entropy": 5.807222938537597,
"epoch": 1.0341074020319303,
"grad_norm": 2.609375,
"learning_rate": 0.0004900112372645236,
"loss": 5.5892,
"mean_token_accuracy": 0.15825920850038527,
"num_tokens": 23656339.0,
"step": 11400
},
{
"entropy": 5.798036432266235,
"epoch": 1.0345609579100146,
"grad_norm": 2.953125,
"learning_rate": 0.0004900017012492386,
"loss": 5.5612,
"mean_token_accuracy": 0.15829772353172303,
"num_tokens": 23667433.0,
"step": 11405
},
{
"entropy": 5.821787405014038,
"epoch": 1.0350145137880986,
"grad_norm": 4.875,
"learning_rate": 0.0004899921607876634,
"loss": 5.6094,
"mean_token_accuracy": 0.15375540256500245,
"num_tokens": 23678186.0,
"step": 11410
},
{
"entropy": 5.793618154525757,
"epoch": 1.0354680696661829,
"grad_norm": 2.515625,
"learning_rate": 0.0004899826158799958,
"loss": 5.5524,
"mean_token_accuracy": 0.1572679951786995,
"num_tokens": 23688317.0,
"step": 11415
},
{
"entropy": 5.7853141784667965,
"epoch": 1.0359216255442671,
"grad_norm": 2.734375,
"learning_rate": 0.0004899730665264328,
"loss": 5.4839,
"mean_token_accuracy": 0.1556887872517109,
"num_tokens": 23697752.0,
"step": 11420
},
{
"entropy": 5.676204586029053,
"epoch": 1.0363751814223512,
"grad_norm": 2.171875,
"learning_rate": 0.0004899635127271721,
"loss": 5.4778,
"mean_token_accuracy": 0.16474685668945313,
"num_tokens": 23707803.0,
"step": 11425
},
{
"entropy": 5.721381902694702,
"epoch": 1.0368287373004355,
"grad_norm": 2.015625,
"learning_rate": 0.0004899539544824111,
"loss": 5.4961,
"mean_token_accuracy": 0.16447029262781143,
"num_tokens": 23717630.0,
"step": 11430
},
{
"entropy": 5.858443784713745,
"epoch": 1.0372822931785195,
"grad_norm": 2.765625,
"learning_rate": 0.0004899443917923476,
"loss": 5.5957,
"mean_token_accuracy": 0.15795777142047882,
"num_tokens": 23728836.0,
"step": 11435
},
{
"entropy": 5.770029544830322,
"epoch": 1.0377358490566038,
"grad_norm": 3.03125,
"learning_rate": 0.0004899348246571792,
"loss": 5.5849,
"mean_token_accuracy": 0.15624269396066665,
"num_tokens": 23740145.0,
"step": 11440
},
{
"entropy": 5.7938860893249515,
"epoch": 1.038189404934688,
"grad_norm": 2.375,
"learning_rate": 0.000489925253077104,
"loss": 5.527,
"mean_token_accuracy": 0.16054621934890748,
"num_tokens": 23751162.0,
"step": 11445
},
{
"entropy": 5.730824613571167,
"epoch": 1.038642960812772,
"grad_norm": 2.59375,
"learning_rate": 0.0004899156770523196,
"loss": 5.5637,
"mean_token_accuracy": 0.15623183995485307,
"num_tokens": 23761634.0,
"step": 11450
},
{
"entropy": 5.724364185333252,
"epoch": 1.0390965166908563,
"grad_norm": 2.625,
"learning_rate": 0.0004899060965830244,
"loss": 5.5489,
"mean_token_accuracy": 0.15919629484415054,
"num_tokens": 23772483.0,
"step": 11455
},
{
"entropy": 5.785492277145385,
"epoch": 1.0395500725689404,
"grad_norm": 3.234375,
"learning_rate": 0.0004898965116694162,
"loss": 5.4983,
"mean_token_accuracy": 0.1587543696165085,
"num_tokens": 23782298.0,
"step": 11460
},
{
"entropy": 5.805042123794555,
"epoch": 1.0400036284470247,
"grad_norm": 3.4375,
"learning_rate": 0.0004898869223116935,
"loss": 5.5577,
"mean_token_accuracy": 0.15344442874193193,
"num_tokens": 23792492.0,
"step": 11465
},
{
"entropy": 5.877156162261963,
"epoch": 1.040457184325109,
"grad_norm": 2.046875,
"learning_rate": 0.0004898773285100546,
"loss": 5.7103,
"mean_token_accuracy": 0.1547830045223236,
"num_tokens": 23802999.0,
"step": 11470
},
{
"entropy": 5.80629072189331,
"epoch": 1.040910740203193,
"grad_norm": 8.5625,
"learning_rate": 0.0004898677302646975,
"loss": 5.5281,
"mean_token_accuracy": 0.16056863516569136,
"num_tokens": 23812541.0,
"step": 11475
},
{
"entropy": 5.741022062301636,
"epoch": 1.0413642960812772,
"grad_norm": 2.3125,
"learning_rate": 0.0004898581275758211,
"loss": 5.6307,
"mean_token_accuracy": 0.15013584792613982,
"num_tokens": 23823244.0,
"step": 11480
},
{
"entropy": 5.7505106925964355,
"epoch": 1.0418178519593615,
"grad_norm": 2.03125,
"learning_rate": 0.0004898485204436239,
"loss": 5.4879,
"mean_token_accuracy": 0.16015641987323762,
"num_tokens": 23832346.0,
"step": 11485
},
{
"entropy": 5.8216358661651615,
"epoch": 1.0422714078374455,
"grad_norm": 3.109375,
"learning_rate": 0.0004898389088683047,
"loss": 5.6227,
"mean_token_accuracy": 0.16134147346019745,
"num_tokens": 23842554.0,
"step": 11490
},
{
"entropy": 5.718843460083008,
"epoch": 1.0427249637155298,
"grad_norm": 2.984375,
"learning_rate": 0.0004898292928500619,
"loss": 5.4529,
"mean_token_accuracy": 0.16722290962934494,
"num_tokens": 23852938.0,
"step": 11495
},
{
"entropy": 5.760653495788574,
"epoch": 1.0431785195936139,
"grad_norm": 2.859375,
"learning_rate": 0.0004898196723890947,
"loss": 5.4795,
"mean_token_accuracy": 0.1648701697587967,
"num_tokens": 23862367.0,
"step": 11500
},
{
"entropy": 5.74053988456726,
"epoch": 1.0436320754716981,
"grad_norm": 2.765625,
"learning_rate": 0.0004898100474856019,
"loss": 5.5826,
"mean_token_accuracy": 0.15636928826570512,
"num_tokens": 23873249.0,
"step": 11505
},
{
"entropy": 5.753488063812256,
"epoch": 1.0440856313497824,
"grad_norm": 3.0625,
"learning_rate": 0.0004898004181397824,
"loss": 5.5644,
"mean_token_accuracy": 0.15214299708604812,
"num_tokens": 23883830.0,
"step": 11510
},
{
"entropy": 5.803683567047119,
"epoch": 1.0445391872278664,
"grad_norm": 1.96875,
"learning_rate": 0.0004897907843518358,
"loss": 5.5182,
"mean_token_accuracy": 0.15898152813315392,
"num_tokens": 23894359.0,
"step": 11515
},
{
"entropy": 5.731159114837647,
"epoch": 1.0449927431059507,
"grad_norm": 2.40625,
"learning_rate": 0.0004897811461219609,
"loss": 5.5049,
"mean_token_accuracy": 0.15634157806634902,
"num_tokens": 23903779.0,
"step": 11520
},
{
"entropy": 5.750355339050293,
"epoch": 1.0454462989840347,
"grad_norm": 13.5625,
"learning_rate": 0.0004897715034503571,
"loss": 5.5669,
"mean_token_accuracy": 0.15915703177452087,
"num_tokens": 23914523.0,
"step": 11525
},
{
"entropy": 5.707683134078979,
"epoch": 1.045899854862119,
"grad_norm": 2.375,
"learning_rate": 0.000489761856337224,
"loss": 5.5596,
"mean_token_accuracy": 0.16285839378833772,
"num_tokens": 23924996.0,
"step": 11530
},
{
"entropy": 5.734094095230103,
"epoch": 1.0463534107402033,
"grad_norm": 2.34375,
"learning_rate": 0.0004897522047827609,
"loss": 5.5205,
"mean_token_accuracy": 0.16182830184698105,
"num_tokens": 23935843.0,
"step": 11535
},
{
"entropy": 5.776166391372681,
"epoch": 1.0468069666182873,
"grad_norm": 3.234375,
"learning_rate": 0.0004897425487871676,
"loss": 5.4958,
"mean_token_accuracy": 0.1574185460805893,
"num_tokens": 23945044.0,
"step": 11540
},
{
"entropy": 5.748601865768433,
"epoch": 1.0472605224963716,
"grad_norm": 2.65625,
"learning_rate": 0.0004897328883506436,
"loss": 5.5306,
"mean_token_accuracy": 0.15569346994161606,
"num_tokens": 23955933.0,
"step": 11545
},
{
"entropy": 5.775574541091919,
"epoch": 1.0477140783744556,
"grad_norm": 2.875,
"learning_rate": 0.0004897232234733887,
"loss": 5.5918,
"mean_token_accuracy": 0.15481627136468887,
"num_tokens": 23966320.0,
"step": 11550
},
{
"entropy": 5.746476888656616,
"epoch": 1.04816763425254,
"grad_norm": 3.296875,
"learning_rate": 0.0004897135541556028,
"loss": 5.5005,
"mean_token_accuracy": 0.16378279030323029,
"num_tokens": 23976233.0,
"step": 11555
},
{
"entropy": 5.771604919433594,
"epoch": 1.0486211901306242,
"grad_norm": 2.625,
"learning_rate": 0.000489703880397486,
"loss": 5.5711,
"mean_token_accuracy": 0.15494766235351562,
"num_tokens": 23987854.0,
"step": 11560
},
{
"entropy": 5.718685150146484,
"epoch": 1.0490747460087082,
"grad_norm": 2.40625,
"learning_rate": 0.0004896942021992382,
"loss": 5.562,
"mean_token_accuracy": 0.15465296059846878,
"num_tokens": 23997973.0,
"step": 11565
},
{
"entropy": 5.722238397598266,
"epoch": 1.0495283018867925,
"grad_norm": 2.96875,
"learning_rate": 0.0004896845195610595,
"loss": 5.5481,
"mean_token_accuracy": 0.15867221653461455,
"num_tokens": 24007313.0,
"step": 11570
},
{
"entropy": 5.768670654296875,
"epoch": 1.0499818577648767,
"grad_norm": 2.765625,
"learning_rate": 0.0004896748324831504,
"loss": 5.589,
"mean_token_accuracy": 0.15097645074129104,
"num_tokens": 24018254.0,
"step": 11575
},
{
"entropy": 5.832350778579712,
"epoch": 1.0504354136429608,
"grad_norm": 2.40625,
"learning_rate": 0.0004896651409657109,
"loss": 5.6043,
"mean_token_accuracy": 0.15535546690225602,
"num_tokens": 24029799.0,
"step": 11580
},
{
"entropy": 5.746990394592285,
"epoch": 1.050888969521045,
"grad_norm": 2.515625,
"learning_rate": 0.0004896554450089417,
"loss": 5.4668,
"mean_token_accuracy": 0.1698855474591255,
"num_tokens": 24040972.0,
"step": 11585
},
{
"entropy": 5.676478624343872,
"epoch": 1.051342525399129,
"grad_norm": 3.34375,
"learning_rate": 0.0004896457446130431,
"loss": 5.4066,
"mean_token_accuracy": 0.16618487387895584,
"num_tokens": 24050781.0,
"step": 11590
},
{
"entropy": 5.697405385971069,
"epoch": 1.0517960812772134,
"grad_norm": 2.5,
"learning_rate": 0.0004896360397782159,
"loss": 5.5164,
"mean_token_accuracy": 0.15654089152812958,
"num_tokens": 24060184.0,
"step": 11595
},
{
"entropy": 5.852255058288574,
"epoch": 1.0522496371552976,
"grad_norm": 2.546875,
"learning_rate": 0.0004896263305046607,
"loss": 5.5832,
"mean_token_accuracy": 0.1613552138209343,
"num_tokens": 24070777.0,
"step": 11600
},
{
"entropy": 5.763235378265381,
"epoch": 1.0527031930333817,
"grad_norm": 3.734375,
"learning_rate": 0.0004896166167925782,
"loss": 5.5994,
"mean_token_accuracy": 0.15702067017555238,
"num_tokens": 24080878.0,
"step": 11605
},
{
"entropy": 5.739090347290039,
"epoch": 1.053156748911466,
"grad_norm": 2.40625,
"learning_rate": 0.0004896068986421693,
"loss": 5.5704,
"mean_token_accuracy": 0.16213036477565765,
"num_tokens": 24090038.0,
"step": 11610
},
{
"entropy": 5.837324953079223,
"epoch": 1.05361030478955,
"grad_norm": 5.03125,
"learning_rate": 0.0004895971760536351,
"loss": 5.603,
"mean_token_accuracy": 0.16087906658649445,
"num_tokens": 24100425.0,
"step": 11615
},
{
"entropy": 5.869068241119384,
"epoch": 1.0540638606676342,
"grad_norm": 3.59375,
"learning_rate": 0.0004895874490271768,
"loss": 5.6849,
"mean_token_accuracy": 0.15415374487638472,
"num_tokens": 24112182.0,
"step": 11620
},
{
"entropy": 5.788816738128662,
"epoch": 1.0545174165457185,
"grad_norm": 2.453125,
"learning_rate": 0.0004895777175629952,
"loss": 5.5304,
"mean_token_accuracy": 0.1549362249672413,
"num_tokens": 24122182.0,
"step": 11625
},
{
"entropy": 5.797952890396118,
"epoch": 1.0549709724238026,
"grad_norm": 3.625,
"learning_rate": 0.0004895679816612917,
"loss": 5.5376,
"mean_token_accuracy": 0.1583702802658081,
"num_tokens": 24131930.0,
"step": 11630
},
{
"entropy": 5.839129638671875,
"epoch": 1.0554245283018868,
"grad_norm": 2.46875,
"learning_rate": 0.0004895582413222678,
"loss": 5.526,
"mean_token_accuracy": 0.16291525661945344,
"num_tokens": 24141575.0,
"step": 11635
},
{
"entropy": 5.779266023635865,
"epoch": 1.0558780841799709,
"grad_norm": 3.65625,
"learning_rate": 0.0004895484965461247,
"loss": 5.6657,
"mean_token_accuracy": 0.155142679810524,
"num_tokens": 24151720.0,
"step": 11640
},
{
"entropy": 5.754306221008301,
"epoch": 1.0563316400580551,
"grad_norm": 2.5625,
"learning_rate": 0.000489538747333064,
"loss": 5.5678,
"mean_token_accuracy": 0.15946533977985383,
"num_tokens": 24161881.0,
"step": 11645
},
{
"entropy": 5.7893461227417,
"epoch": 1.0567851959361394,
"grad_norm": 3.078125,
"learning_rate": 0.0004895289936832874,
"loss": 5.5077,
"mean_token_accuracy": 0.1613079935312271,
"num_tokens": 24171474.0,
"step": 11650
},
{
"entropy": 5.812449169158936,
"epoch": 1.0572387518142234,
"grad_norm": 3.328125,
"learning_rate": 0.0004895192355969966,
"loss": 5.6141,
"mean_token_accuracy": 0.15436174273490905,
"num_tokens": 24182186.0,
"step": 11655
},
{
"entropy": 5.797398424148559,
"epoch": 1.0576923076923077,
"grad_norm": 3.109375,
"learning_rate": 0.0004895094730743932,
"loss": 5.5539,
"mean_token_accuracy": 0.16231205612421035,
"num_tokens": 24192318.0,
"step": 11660
},
{
"entropy": 5.853780698776245,
"epoch": 1.058145863570392,
"grad_norm": 2.71875,
"learning_rate": 0.0004894997061156793,
"loss": 5.6112,
"mean_token_accuracy": 0.1608656793832779,
"num_tokens": 24202918.0,
"step": 11665
},
{
"entropy": 5.771274995803833,
"epoch": 1.058599419448476,
"grad_norm": 3.453125,
"learning_rate": 0.0004894899347210567,
"loss": 5.5157,
"mean_token_accuracy": 0.16073851138353348,
"num_tokens": 24213218.0,
"step": 11670
},
{
"entropy": 5.7356054306030275,
"epoch": 1.0590529753265603,
"grad_norm": 2.5,
"learning_rate": 0.0004894801588907278,
"loss": 5.5161,
"mean_token_accuracy": 0.16054870188236237,
"num_tokens": 24223229.0,
"step": 11675
},
{
"entropy": 5.812273836135864,
"epoch": 1.0595065312046443,
"grad_norm": 2.640625,
"learning_rate": 0.0004894703786248944,
"loss": 5.581,
"mean_token_accuracy": 0.15022995471954345,
"num_tokens": 24233637.0,
"step": 11680
},
{
"entropy": 5.803047227859497,
"epoch": 1.0599600870827286,
"grad_norm": 2.46875,
"learning_rate": 0.000489460593923759,
"loss": 5.6094,
"mean_token_accuracy": 0.16174269765615462,
"num_tokens": 24244251.0,
"step": 11685
},
{
"entropy": 5.7527000427246096,
"epoch": 1.0604136429608129,
"grad_norm": 3.34375,
"learning_rate": 0.0004894508047875237,
"loss": 5.5103,
"mean_token_accuracy": 0.1596752181649208,
"num_tokens": 24254380.0,
"step": 11690
},
{
"entropy": 5.662437391281128,
"epoch": 1.060867198838897,
"grad_norm": 2.8125,
"learning_rate": 0.0004894410112163913,
"loss": 5.4316,
"mean_token_accuracy": 0.1607103630900383,
"num_tokens": 24264841.0,
"step": 11695
},
{
"entropy": 5.675736808776856,
"epoch": 1.0613207547169812,
"grad_norm": 3.46875,
"learning_rate": 0.0004894312132105641,
"loss": 5.4988,
"mean_token_accuracy": 0.16021037101745605,
"num_tokens": 24274383.0,
"step": 11700
},
{
"entropy": 5.808772087097168,
"epoch": 1.0617743105950652,
"grad_norm": 2.234375,
"learning_rate": 0.0004894214107702447,
"loss": 5.5599,
"mean_token_accuracy": 0.15565950721502303,
"num_tokens": 24284140.0,
"step": 11705
},
{
"entropy": 5.893798446655273,
"epoch": 1.0622278664731495,
"grad_norm": 2.125,
"learning_rate": 0.000489411603895636,
"loss": 5.6526,
"mean_token_accuracy": 0.1567644879221916,
"num_tokens": 24295044.0,
"step": 11710
},
{
"entropy": 5.787691354751587,
"epoch": 1.0626814223512338,
"grad_norm": 2.171875,
"learning_rate": 0.0004894017925869406,
"loss": 5.5477,
"mean_token_accuracy": 0.1604714199900627,
"num_tokens": 24305080.0,
"step": 11715
},
{
"entropy": 5.7589418411254885,
"epoch": 1.0631349782293178,
"grad_norm": 2.1875,
"learning_rate": 0.0004893919768443615,
"loss": 5.5929,
"mean_token_accuracy": 0.15704786777496338,
"num_tokens": 24316306.0,
"step": 11720
},
{
"entropy": 5.6547332286834715,
"epoch": 1.063588534107402,
"grad_norm": 1.96875,
"learning_rate": 0.0004893821566681017,
"loss": 5.5487,
"mean_token_accuracy": 0.16503504812717437,
"num_tokens": 24326740.0,
"step": 11725
},
{
"entropy": 5.787192344665527,
"epoch": 1.0640420899854863,
"grad_norm": 2.203125,
"learning_rate": 0.0004893723320583644,
"loss": 5.5698,
"mean_token_accuracy": 0.15400078892707825,
"num_tokens": 24337621.0,
"step": 11730
},
{
"entropy": 5.7745507717132565,
"epoch": 1.0644956458635704,
"grad_norm": 2.96875,
"learning_rate": 0.0004893625030153525,
"loss": 5.4569,
"mean_token_accuracy": 0.16433921158313752,
"num_tokens": 24347732.0,
"step": 11735
},
{
"entropy": 5.791579294204712,
"epoch": 1.0649492017416546,
"grad_norm": 2.140625,
"learning_rate": 0.0004893526695392694,
"loss": 5.4969,
"mean_token_accuracy": 0.16680119484663009,
"num_tokens": 24359173.0,
"step": 11740
},
{
"entropy": 5.733738374710083,
"epoch": 1.0654027576197387,
"grad_norm": 2.8125,
"learning_rate": 0.0004893428316303185,
"loss": 5.4729,
"mean_token_accuracy": 0.1564803808927536,
"num_tokens": 24369465.0,
"step": 11745
},
{
"entropy": 5.687932920455933,
"epoch": 1.065856313497823,
"grad_norm": 2.359375,
"learning_rate": 0.0004893329892887031,
"loss": 5.5126,
"mean_token_accuracy": 0.1594853550195694,
"num_tokens": 24378899.0,
"step": 11750
},
{
"entropy": 5.714597225189209,
"epoch": 1.0663098693759072,
"grad_norm": 2.515625,
"learning_rate": 0.0004893231425146271,
"loss": 5.6376,
"mean_token_accuracy": 0.1584230750799179,
"num_tokens": 24388919.0,
"step": 11755
},
{
"entropy": 5.748736715316772,
"epoch": 1.0667634252539913,
"grad_norm": 2.46875,
"learning_rate": 0.0004893132913082937,
"loss": 5.4999,
"mean_token_accuracy": 0.16259035617113113,
"num_tokens": 24399632.0,
"step": 11760
},
{
"entropy": 5.816553211212158,
"epoch": 1.0672169811320755,
"grad_norm": 2.5,
"learning_rate": 0.0004893034356699069,
"loss": 5.5458,
"mean_token_accuracy": 0.1614643394947052,
"num_tokens": 24410817.0,
"step": 11765
},
{
"entropy": 5.779141235351562,
"epoch": 1.0676705370101596,
"grad_norm": 1.8984375,
"learning_rate": 0.0004892935755996705,
"loss": 5.5481,
"mean_token_accuracy": 0.15944551825523376,
"num_tokens": 24421058.0,
"step": 11770
},
{
"entropy": 5.704815626144409,
"epoch": 1.0681240928882438,
"grad_norm": 2.796875,
"learning_rate": 0.0004892837110977883,
"loss": 5.4538,
"mean_token_accuracy": 0.15836425572633744,
"num_tokens": 24431522.0,
"step": 11775
},
{
"entropy": 5.716076183319092,
"epoch": 1.068577648766328,
"grad_norm": 2.265625,
"learning_rate": 0.0004892738421644644,
"loss": 5.5085,
"mean_token_accuracy": 0.1596313402056694,
"num_tokens": 24441746.0,
"step": 11780
},
{
"entropy": 5.7126411437988285,
"epoch": 1.0690312046444121,
"grad_norm": 2.609375,
"learning_rate": 0.0004892639687999027,
"loss": 5.414,
"mean_token_accuracy": 0.1638681098818779,
"num_tokens": 24451916.0,
"step": 11785
},
{
"entropy": 5.729473066329956,
"epoch": 1.0694847605224964,
"grad_norm": 1.8671875,
"learning_rate": 0.0004892540910043076,
"loss": 5.5454,
"mean_token_accuracy": 0.1636336326599121,
"num_tokens": 24461935.0,
"step": 11790
},
{
"entropy": 5.703885555267334,
"epoch": 1.0699383164005805,
"grad_norm": 2.8125,
"learning_rate": 0.0004892442087778833,
"loss": 5.4762,
"mean_token_accuracy": 0.15906477123498916,
"num_tokens": 24472437.0,
"step": 11795
},
{
"entropy": 5.71383752822876,
"epoch": 1.0703918722786647,
"grad_norm": 3.359375,
"learning_rate": 0.0004892343221208342,
"loss": 5.5325,
"mean_token_accuracy": 0.16436335295438767,
"num_tokens": 24481726.0,
"step": 11800
},
{
"entropy": 5.8328804016113285,
"epoch": 1.070845428156749,
"grad_norm": 2.640625,
"learning_rate": 0.0004892244310333646,
"loss": 5.6385,
"mean_token_accuracy": 0.14966130405664443,
"num_tokens": 24492340.0,
"step": 11805
},
{
"entropy": 5.780572319030762,
"epoch": 1.071298984034833,
"grad_norm": 1.9765625,
"learning_rate": 0.0004892145355156793,
"loss": 5.518,
"mean_token_accuracy": 0.1623941883444786,
"num_tokens": 24502138.0,
"step": 11810
},
{
"entropy": 5.740274572372437,
"epoch": 1.0717525399129173,
"grad_norm": 2.3125,
"learning_rate": 0.0004892046355679829,
"loss": 5.5688,
"mean_token_accuracy": 0.15533077716827393,
"num_tokens": 24512723.0,
"step": 11815
},
{
"entropy": 5.811892747879028,
"epoch": 1.0722060957910013,
"grad_norm": 3.28125,
"learning_rate": 0.0004891947311904799,
"loss": 5.6049,
"mean_token_accuracy": 0.1485086649656296,
"num_tokens": 24523766.0,
"step": 11820
},
{
"entropy": 5.812064456939697,
"epoch": 1.0726596516690856,
"grad_norm": 5.84375,
"learning_rate": 0.0004891848223833753,
"loss": 5.5622,
"mean_token_accuracy": 0.15951724946498871,
"num_tokens": 24534571.0,
"step": 11825
},
{
"entropy": 5.790804481506347,
"epoch": 1.0731132075471699,
"grad_norm": 4.1875,
"learning_rate": 0.000489174909146874,
"loss": 5.5618,
"mean_token_accuracy": 0.16514019370079042,
"num_tokens": 24545265.0,
"step": 11830
},
{
"entropy": 5.808223009109497,
"epoch": 1.073566763425254,
"grad_norm": 2.40625,
"learning_rate": 0.0004891649914811811,
"loss": 5.6276,
"mean_token_accuracy": 0.1587499588727951,
"num_tokens": 24555524.0,
"step": 11835
},
{
"entropy": 5.773516654968262,
"epoch": 1.0740203193033382,
"grad_norm": 7.75,
"learning_rate": 0.0004891550693865014,
"loss": 5.5005,
"mean_token_accuracy": 0.1684954434633255,
"num_tokens": 24565420.0,
"step": 11840
},
{
"entropy": 5.808403062820434,
"epoch": 1.0744738751814222,
"grad_norm": 3.875,
"learning_rate": 0.0004891451428630405,
"loss": 5.5525,
"mean_token_accuracy": 0.15781337320804595,
"num_tokens": 24576794.0,
"step": 11845
},
{
"entropy": 5.762145185470581,
"epoch": 1.0749274310595065,
"grad_norm": 2.453125,
"learning_rate": 0.0004891352119110033,
"loss": 5.5556,
"mean_token_accuracy": 0.16086098402738572,
"num_tokens": 24586320.0,
"step": 11850
},
{
"entropy": 5.7493363380432125,
"epoch": 1.0753809869375908,
"grad_norm": 2.484375,
"learning_rate": 0.0004891252765305954,
"loss": 5.5305,
"mean_token_accuracy": 0.16228426843881608,
"num_tokens": 24596524.0,
"step": 11855
},
{
"entropy": 5.799742603302002,
"epoch": 1.0758345428156748,
"grad_norm": 2.09375,
"learning_rate": 0.0004891153367220222,
"loss": 5.4769,
"mean_token_accuracy": 0.16120700389146805,
"num_tokens": 24605877.0,
"step": 11860
},
{
"entropy": 5.871392726898193,
"epoch": 1.076288098693759,
"grad_norm": 2.625,
"learning_rate": 0.0004891053924854894,
"loss": 5.6675,
"mean_token_accuracy": 0.15001487284898757,
"num_tokens": 24616719.0,
"step": 11865
},
{
"entropy": 5.732544708251953,
"epoch": 1.0767416545718433,
"grad_norm": 2.15625,
"learning_rate": 0.0004890954438212024,
"loss": 5.4695,
"mean_token_accuracy": 0.16442711055278778,
"num_tokens": 24626882.0,
"step": 11870
},
{
"entropy": 5.762554931640625,
"epoch": 1.0771952104499274,
"grad_norm": 2.640625,
"learning_rate": 0.0004890854907293671,
"loss": 5.5507,
"mean_token_accuracy": 0.15483054518699646,
"num_tokens": 24637884.0,
"step": 11875
},
{
"entropy": 5.720759391784668,
"epoch": 1.0776487663280117,
"grad_norm": 2.5625,
"learning_rate": 0.0004890755332101893,
"loss": 5.5107,
"mean_token_accuracy": 0.15827159583568573,
"num_tokens": 24647389.0,
"step": 11880
},
{
"entropy": 5.756432104110718,
"epoch": 1.0781023222060957,
"grad_norm": 3.203125,
"learning_rate": 0.0004890655712638749,
"loss": 5.5659,
"mean_token_accuracy": 0.15237744450569152,
"num_tokens": 24657595.0,
"step": 11885
},
{
"entropy": 5.75420880317688,
"epoch": 1.07855587808418,
"grad_norm": 2.765625,
"learning_rate": 0.00048905560489063,
"loss": 5.5033,
"mean_token_accuracy": 0.15938466638326645,
"num_tokens": 24667683.0,
"step": 11890
},
{
"entropy": 5.859339427947998,
"epoch": 1.0790094339622642,
"grad_norm": 4.03125,
"learning_rate": 0.0004890456340906605,
"loss": 5.6284,
"mean_token_accuracy": 0.1514022320508957,
"num_tokens": 24678041.0,
"step": 11895
},
{
"entropy": 5.78812313079834,
"epoch": 1.0794629898403483,
"grad_norm": 3.734375,
"learning_rate": 0.0004890356588641729,
"loss": 5.6002,
"mean_token_accuracy": 0.16286241114139557,
"num_tokens": 24688569.0,
"step": 11900
},
{
"entropy": 5.7869555950164795,
"epoch": 1.0799165457184325,
"grad_norm": 2.421875,
"learning_rate": 0.0004890256792113733,
"loss": 5.5975,
"mean_token_accuracy": 0.15714042633771896,
"num_tokens": 24699913.0,
"step": 11905
},
{
"entropy": 5.867937850952148,
"epoch": 1.0803701015965168,
"grad_norm": 4.90625,
"learning_rate": 0.0004890156951324681,
"loss": 5.588,
"mean_token_accuracy": 0.15711037516593934,
"num_tokens": 24710682.0,
"step": 11910
},
{
"entropy": 5.890251636505127,
"epoch": 1.0808236574746009,
"grad_norm": 2.546875,
"learning_rate": 0.0004890057066276636,
"loss": 5.678,
"mean_token_accuracy": 0.15511685758829116,
"num_tokens": 24721087.0,
"step": 11915
},
{
"entropy": 5.749015665054321,
"epoch": 1.0812772133526851,
"grad_norm": 3.09375,
"learning_rate": 0.0004889957136971668,
"loss": 5.5046,
"mean_token_accuracy": 0.16428616046905517,
"num_tokens": 24731323.0,
"step": 11920
},
{
"entropy": 5.734285354614258,
"epoch": 1.0817307692307692,
"grad_norm": 2.390625,
"learning_rate": 0.000488985716341184,
"loss": 5.5549,
"mean_token_accuracy": 0.15688968300819398,
"num_tokens": 24742559.0,
"step": 11925
},
{
"entropy": 5.8322375297546385,
"epoch": 1.0821843251088534,
"grad_norm": 3.421875,
"learning_rate": 0.0004889757145599222,
"loss": 5.5385,
"mean_token_accuracy": 0.158398574590683,
"num_tokens": 24752307.0,
"step": 11930
},
{
"entropy": 5.715606069564819,
"epoch": 1.0826378809869377,
"grad_norm": 3.40625,
"learning_rate": 0.000488965708353588,
"loss": 5.4942,
"mean_token_accuracy": 0.15804069489240646,
"num_tokens": 24762792.0,
"step": 11935
},
{
"entropy": 5.799453639984131,
"epoch": 1.0830914368650217,
"grad_norm": 3.75,
"learning_rate": 0.0004889556977223883,
"loss": 5.6277,
"mean_token_accuracy": 0.1521786704659462,
"num_tokens": 24774377.0,
"step": 11940
},
{
"entropy": 5.814619016647339,
"epoch": 1.083544992743106,
"grad_norm": 3.46875,
"learning_rate": 0.0004889456826665304,
"loss": 5.5527,
"mean_token_accuracy": 0.15799938887357712,
"num_tokens": 24785919.0,
"step": 11945
},
{
"entropy": 5.841064405441284,
"epoch": 1.08399854862119,
"grad_norm": 2.515625,
"learning_rate": 0.0004889356631862211,
"loss": 5.5847,
"mean_token_accuracy": 0.15743543207645416,
"num_tokens": 24796594.0,
"step": 11950
},
{
"entropy": 5.763912773132324,
"epoch": 1.0844521044992743,
"grad_norm": 2.75,
"learning_rate": 0.000488925639281668,
"loss": 5.565,
"mean_token_accuracy": 0.15561186671257018,
"num_tokens": 24807322.0,
"step": 11955
},
{
"entropy": 5.741538190841675,
"epoch": 1.0849056603773586,
"grad_norm": 3.734375,
"learning_rate": 0.0004889156109530781,
"loss": 5.4712,
"mean_token_accuracy": 0.16155950129032134,
"num_tokens": 24817593.0,
"step": 11960
},
{
"entropy": 5.769239807128907,
"epoch": 1.0853592162554426,
"grad_norm": 3.5,
"learning_rate": 0.0004889055782006588,
"loss": 5.4859,
"mean_token_accuracy": 0.15776087194681168,
"num_tokens": 24827577.0,
"step": 11965
},
{
"entropy": 5.742091464996338,
"epoch": 1.085812772133527,
"grad_norm": 3.78125,
"learning_rate": 0.0004888955410246176,
"loss": 5.5279,
"mean_token_accuracy": 0.1618715599179268,
"num_tokens": 24838148.0,
"step": 11970
},
{
"entropy": 5.732159185409546,
"epoch": 1.086266328011611,
"grad_norm": 3.734375,
"learning_rate": 0.0004888854994251621,
"loss": 5.4838,
"mean_token_accuracy": 0.16252688318490982,
"num_tokens": 24848451.0,
"step": 11975
},
{
"entropy": 5.761207103729248,
"epoch": 1.0867198838896952,
"grad_norm": 3.0625,
"learning_rate": 0.0004888754534025,
"loss": 5.5747,
"mean_token_accuracy": 0.15174014270305633,
"num_tokens": 24859619.0,
"step": 11980
},
{
"entropy": 5.761291885375977,
"epoch": 1.0871734397677795,
"grad_norm": 2.296875,
"learning_rate": 0.0004888654029568391,
"loss": 5.6151,
"mean_token_accuracy": 0.14883196651935576,
"num_tokens": 24869863.0,
"step": 11985
},
{
"entropy": 5.742021799087524,
"epoch": 1.0876269956458635,
"grad_norm": 2.484375,
"learning_rate": 0.0004888553480883871,
"loss": 5.5408,
"mean_token_accuracy": 0.15269162356853486,
"num_tokens": 24880524.0,
"step": 11990
},
{
"entropy": 5.733129167556763,
"epoch": 1.0880805515239478,
"grad_norm": 1.9921875,
"learning_rate": 0.000488845288797352,
"loss": 5.5186,
"mean_token_accuracy": 0.16384661942720413,
"num_tokens": 24891514.0,
"step": 11995
},
{
"entropy": 5.739286851882935,
"epoch": 1.0885341074020318,
"grad_norm": 6.0625,
"learning_rate": 0.0004888352250839418,
"loss": 5.5016,
"mean_token_accuracy": 0.16507037729024887,
"num_tokens": 24900842.0,
"step": 12000
},
{
"epoch": 1.0885341074020318,
"eval_entropy": 5.580312384091891,
"eval_loss": 5.6294732093811035,
"eval_mean_token_accuracy": 0.16381040554982504,
"eval_num_tokens": 24900842.0,
"eval_runtime": 26.6163,
"eval_samples_per_second": 1328.511,
"eval_steps_per_second": 166.064,
"step": 12000
}
],
"logging_steps": 5,
"max_steps": 110230,
"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": 3.6456130803456e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink