fpadovani/dan-latn-10mb-100mb_seed10
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
bad567f1e2a0eefc995daffb3b5cf1afdb67cd5a
dan-latn-10mb-100mb_seed10/checkpoint-6000/trainer_state.json
ModelHub XC bad567f1e2 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/dan-latn-10mb-100mb_seed10
Source: Original Platform
2026-07-18 01:45:13 +08:00

12057 lines
329 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5011693952555963,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742122554779053,
"epoch": 0.000417641162712997,
"grad_norm": 5.59375,
"learning_rate": 2e-06,
"loss": 10.7426,
"mean_token_accuracy": 0.0,
"num_tokens": 8702.0,
"step": 5
},
{
"entropy": 10.742121601104737,
"epoch": 0.000835282325425994,
"grad_norm": 4.9375,
"learning_rate": 4.5e-06,
"loss": 10.7359,
"mean_token_accuracy": 0.0,
"num_tokens": 17355.0,
"step": 10
},
{
"entropy": 10.742114067077637,
"epoch": 0.001252923488138991,
"grad_norm": 5.1875,
"learning_rate": 7e-06,
"loss": 10.7158,
"mean_token_accuracy": 0.0,
"num_tokens": 25975.0,
"step": 15
},
{
"entropy": 10.742129802703857,
"epoch": 0.001670564650851988,
"grad_norm": 5.125,
"learning_rate": 9.5e-06,
"loss": 10.6794,
"mean_token_accuracy": 0.00029278567526489494,
"num_tokens": 35099.0,
"step": 20
},
{
"entropy": 10.742053031921387,
"epoch": 0.002088205813564985,
"grad_norm": 4.875,
"learning_rate": 1.2e-05,
"loss": 10.5829,
"mean_token_accuracy": 0.002680328907445073,
"num_tokens": 44526.0,
"step": 25
},
{
"entropy": 10.741730117797852,
"epoch": 0.002505846976277982,
"grad_norm": 4.5,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.4495,
"mean_token_accuracy": 0.025214007031172513,
"num_tokens": 53856.0,
"step": 30
},
{
"entropy": 10.740700435638427,
"epoch": 0.002923488138990979,
"grad_norm": 3.3125,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.3326,
"mean_token_accuracy": 0.04184841923415661,
"num_tokens": 63583.0,
"step": 35
},
{
"entropy": 10.737869358062744,
"epoch": 0.003341129301703976,
"grad_norm": 2.796875,
"learning_rate": 1.95e-05,
"loss": 10.1371,
"mean_token_accuracy": 0.04405288808047771,
"num_tokens": 71894.0,
"step": 40
},
{
"entropy": 10.73253116607666,
"epoch": 0.003758770464416973,
"grad_norm": 2.1875,
"learning_rate": 2.2e-05,
"loss": 10.0595,
"mean_token_accuracy": 0.04489131569862366,
"num_tokens": 81234.0,
"step": 45
},
{
"entropy": 10.726473808288574,
"epoch": 0.00417641162712997,
"grad_norm": 2.125,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.9743,
"mean_token_accuracy": 0.04135479014366865,
"num_tokens": 90759.0,
"step": 50
},
{
"entropy": 10.72225399017334,
"epoch": 0.004594052789842967,
"grad_norm": 2.1875,
"learning_rate": 2.7e-05,
"loss": 9.9276,
"mean_token_accuracy": 0.042906027659773825,
"num_tokens": 100765.0,
"step": 55
},
{
"entropy": 10.71838436126709,
"epoch": 0.005011693952555964,
"grad_norm": 1.9609375,
"learning_rate": 2.95e-05,
"loss": 9.8575,
"mean_token_accuracy": 0.04668322093784809,
"num_tokens": 110317.0,
"step": 60
},
{
"entropy": 10.714269351959228,
"epoch": 0.005429335115268961,
"grad_norm": 1.9453125,
"learning_rate": 3.2e-05,
"loss": 9.7809,
"mean_token_accuracy": 0.04387491997331381,
"num_tokens": 119665.0,
"step": 65
},
{
"entropy": 10.711958026885986,
"epoch": 0.005846976277981958,
"grad_norm": 2.015625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.7164,
"mean_token_accuracy": 0.04178145974874496,
"num_tokens": 128083.0,
"step": 70
},
{
"entropy": 10.710527801513672,
"epoch": 0.006264617440694955,
"grad_norm": 1.96875,
"learning_rate": 3.7e-05,
"loss": 9.6655,
"mean_token_accuracy": 0.04346162974834442,
"num_tokens": 137168.0,
"step": 75
},
{
"entropy": 10.705409908294678,
"epoch": 0.006682258603407952,
"grad_norm": 2.03125,
"learning_rate": 3.95e-05,
"loss": 9.5644,
"mean_token_accuracy": 0.04724513851106167,
"num_tokens": 146023.0,
"step": 80
},
{
"entropy": 10.697060108184814,
"epoch": 0.007099899766120949,
"grad_norm": 1.8828125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.539,
"mean_token_accuracy": 0.043902423605322836,
"num_tokens": 154392.0,
"step": 85
},
{
"entropy": 10.686701488494872,
"epoch": 0.007517540928833946,
"grad_norm": 1.8828125,
"learning_rate": 4.45e-05,
"loss": 9.4266,
"mean_token_accuracy": 0.04443326331675053,
"num_tokens": 163673.0,
"step": 90
},
{
"entropy": 10.679195785522461,
"epoch": 0.007935182091546943,
"grad_norm": 1.96875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.3917,
"mean_token_accuracy": 0.04754145070910454,
"num_tokens": 172879.0,
"step": 95
},
{
"entropy": 10.668279361724853,
"epoch": 0.00835282325425994,
"grad_norm": 1.875,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.311,
"mean_token_accuracy": 0.05805562548339367,
"num_tokens": 183378.0,
"step": 100
},
{
"entropy": 10.64778175354004,
"epoch": 0.008770464416972937,
"grad_norm": 1.8984375,
"learning_rate": 5.2e-05,
"loss": 9.1882,
"mean_token_accuracy": 0.058023329451680185,
"num_tokens": 192255.0,
"step": 105
},
{
"entropy": 10.609379386901855,
"epoch": 0.009188105579685933,
"grad_norm": 1.8125,
"learning_rate": 5.45e-05,
"loss": 9.0856,
"mean_token_accuracy": 0.05645543336868286,
"num_tokens": 202118.0,
"step": 110
},
{
"entropy": 10.552488803863525,
"epoch": 0.00960574674239893,
"grad_norm": 1.78125,
"learning_rate": 5.7e-05,
"loss": 8.9239,
"mean_token_accuracy": 0.05940695628523827,
"num_tokens": 210536.0,
"step": 115
},
{
"entropy": 10.494095134735108,
"epoch": 0.010023387905111928,
"grad_norm": 1.78125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.8123,
"mean_token_accuracy": 0.05913706123828888,
"num_tokens": 219577.0,
"step": 120
},
{
"entropy": 10.45734453201294,
"epoch": 0.010441029067824925,
"grad_norm": 1.8984375,
"learning_rate": 6.2e-05,
"loss": 8.7051,
"mean_token_accuracy": 0.057156120985746385,
"num_tokens": 227676.0,
"step": 125
},
{
"entropy": 10.395902347564697,
"epoch": 0.010858670230537921,
"grad_norm": 1.7890625,
"learning_rate": 6.450000000000001e-05,
"loss": 8.5495,
"mean_token_accuracy": 0.06603703089058399,
"num_tokens": 236692.0,
"step": 130
},
{
"entropy": 10.323854541778564,
"epoch": 0.011276311393250919,
"grad_norm": 1.484375,
"learning_rate": 6.7e-05,
"loss": 8.5097,
"mean_token_accuracy": 0.05694400519132614,
"num_tokens": 246143.0,
"step": 135
},
{
"entropy": 10.26358995437622,
"epoch": 0.011693952555963916,
"grad_norm": 1.578125,
"learning_rate": 6.950000000000001e-05,
"loss": 8.3347,
"mean_token_accuracy": 0.06364477723836899,
"num_tokens": 255263.0,
"step": 140
},
{
"entropy": 10.139849853515624,
"epoch": 0.012111593718676913,
"grad_norm": 1.3984375,
"learning_rate": 7.2e-05,
"loss": 8.2469,
"mean_token_accuracy": 0.06181924939155579,
"num_tokens": 264219.0,
"step": 145
},
{
"entropy": 10.00226879119873,
"epoch": 0.01252923488138991,
"grad_norm": 1.34375,
"learning_rate": 7.45e-05,
"loss": 8.1372,
"mean_token_accuracy": 0.06256255879998207,
"num_tokens": 274819.0,
"step": 150
},
{
"entropy": 9.904099082946777,
"epoch": 0.012946876044102907,
"grad_norm": 1.3515625,
"learning_rate": 7.7e-05,
"loss": 8.0018,
"mean_token_accuracy": 0.06529068276286125,
"num_tokens": 284382.0,
"step": 155
},
{
"entropy": 9.72164478302002,
"epoch": 0.013364517206815904,
"grad_norm": 1.28125,
"learning_rate": 7.950000000000001e-05,
"loss": 7.9077,
"mean_token_accuracy": 0.06376104950904846,
"num_tokens": 294457.0,
"step": 160
},
{
"entropy": 9.5483229637146,
"epoch": 0.013782158369528901,
"grad_norm": 1.171875,
"learning_rate": 8.2e-05,
"loss": 7.7723,
"mean_token_accuracy": 0.0630855705589056,
"num_tokens": 303788.0,
"step": 165
},
{
"entropy": 9.329432392120362,
"epoch": 0.014199799532241897,
"grad_norm": 1.0703125,
"learning_rate": 8.450000000000001e-05,
"loss": 7.7114,
"mean_token_accuracy": 0.0626518715173006,
"num_tokens": 312815.0,
"step": 170
},
{
"entropy": 9.140421104431152,
"epoch": 0.014617440694954895,
"grad_norm": 1.015625,
"learning_rate": 8.7e-05,
"loss": 7.6501,
"mean_token_accuracy": 0.06696275398135185,
"num_tokens": 323599.0,
"step": 175
},
{
"entropy": 8.923584079742431,
"epoch": 0.015035081857667892,
"grad_norm": 1.015625,
"learning_rate": 8.95e-05,
"loss": 7.5714,
"mean_token_accuracy": 0.06868262700736523,
"num_tokens": 332788.0,
"step": 180
},
{
"entropy": 8.773529720306396,
"epoch": 0.01545272302038089,
"grad_norm": 1.0234375,
"learning_rate": 9.2e-05,
"loss": 7.557,
"mean_token_accuracy": 0.06944454908370971,
"num_tokens": 342528.0,
"step": 185
},
{
"entropy": 8.585641288757325,
"epoch": 0.015870364183093885,
"grad_norm": 1.0078125,
"learning_rate": 9.45e-05,
"loss": 7.4513,
"mean_token_accuracy": 0.07589564621448516,
"num_tokens": 350404.0,
"step": 190
},
{
"entropy": 8.316099929809571,
"epoch": 0.016288005345806884,
"grad_norm": 0.94921875,
"learning_rate": 9.7e-05,
"loss": 7.3716,
"mean_token_accuracy": 0.0725555706769228,
"num_tokens": 360806.0,
"step": 195
},
{
"entropy": 8.30558967590332,
"epoch": 0.01670564650851988,
"grad_norm": 1.0078125,
"learning_rate": 9.95e-05,
"loss": 7.3737,
"mean_token_accuracy": 0.07712265998125076,
"num_tokens": 370236.0,
"step": 200
},
{
"entropy": 8.26392936706543,
"epoch": 0.017123287671232876,
"grad_norm": 0.828125,
"learning_rate": 0.000102,
"loss": 7.3407,
"mean_token_accuracy": 0.07612420171499253,
"num_tokens": 380409.0,
"step": 205
},
{
"entropy": 8.188503170013428,
"epoch": 0.017540928833945875,
"grad_norm": 1.40625,
"learning_rate": 0.00010449999999999999,
"loss": 7.2554,
"mean_token_accuracy": 0.0819562017917633,
"num_tokens": 389797.0,
"step": 210
},
{
"entropy": 8.090349197387695,
"epoch": 0.01795856999665887,
"grad_norm": 1.734375,
"learning_rate": 0.000107,
"loss": 7.1638,
"mean_token_accuracy": 0.08251158446073532,
"num_tokens": 398867.0,
"step": 215
},
{
"entropy": 7.969277334213257,
"epoch": 0.018376211159371866,
"grad_norm": 0.91796875,
"learning_rate": 0.0001095,
"loss": 7.3557,
"mean_token_accuracy": 0.07762790322303773,
"num_tokens": 409921.0,
"step": 220
},
{
"entropy": 8.04366888999939,
"epoch": 0.018793852322084866,
"grad_norm": 1.4140625,
"learning_rate": 0.000112,
"loss": 7.16,
"mean_token_accuracy": 0.08810438439249993,
"num_tokens": 418962.0,
"step": 225
},
{
"entropy": 7.852949237823486,
"epoch": 0.01921149348479786,
"grad_norm": 1.078125,
"learning_rate": 0.0001145,
"loss": 7.1777,
"mean_token_accuracy": 0.08824707269668579,
"num_tokens": 428111.0,
"step": 230
},
{
"entropy": 7.989979887008667,
"epoch": 0.01962913464751086,
"grad_norm": 0.93359375,
"learning_rate": 0.00011700000000000001,
"loss": 7.2081,
"mean_token_accuracy": 0.08595793023705482,
"num_tokens": 437370.0,
"step": 235
},
{
"entropy": 7.886132049560547,
"epoch": 0.020046775810223856,
"grad_norm": 1.40625,
"learning_rate": 0.00011949999999999999,
"loss": 7.0847,
"mean_token_accuracy": 0.08573980554938317,
"num_tokens": 446687.0,
"step": 240
},
{
"entropy": 7.9483672142028805,
"epoch": 0.020464416972936852,
"grad_norm": 1.03125,
"learning_rate": 0.000122,
"loss": 7.2439,
"mean_token_accuracy": 0.07872043699026107,
"num_tokens": 455306.0,
"step": 245
},
{
"entropy": 7.801475906372071,
"epoch": 0.02088205813564985,
"grad_norm": 0.953125,
"learning_rate": 0.0001245,
"loss": 7.1467,
"mean_token_accuracy": 0.08012107759714127,
"num_tokens": 465491.0,
"step": 250
},
{
"entropy": 7.833319616317749,
"epoch": 0.021299699298362847,
"grad_norm": 0.9453125,
"learning_rate": 0.000127,
"loss": 7.1415,
"mean_token_accuracy": 0.08490735366940498,
"num_tokens": 475782.0,
"step": 255
},
{
"entropy": 7.83586573600769,
"epoch": 0.021717340461075842,
"grad_norm": 1.0625,
"learning_rate": 0.0001295,
"loss": 7.1472,
"mean_token_accuracy": 0.08664263188838958,
"num_tokens": 484901.0,
"step": 260
},
{
"entropy": 7.776985597610474,
"epoch": 0.02213498162378884,
"grad_norm": 0.9375,
"learning_rate": 0.000132,
"loss": 7.072,
"mean_token_accuracy": 0.08647527918219566,
"num_tokens": 493742.0,
"step": 265
},
{
"entropy": 7.73929386138916,
"epoch": 0.022552622786501837,
"grad_norm": 1.046875,
"learning_rate": 0.00013450000000000002,
"loss": 7.0923,
"mean_token_accuracy": 0.08229703679680825,
"num_tokens": 502959.0,
"step": 270
},
{
"entropy": 7.730380439758301,
"epoch": 0.022970263949214833,
"grad_norm": 0.98828125,
"learning_rate": 0.00013700000000000002,
"loss": 7.0222,
"mean_token_accuracy": 0.08907360509037972,
"num_tokens": 512003.0,
"step": 275
},
{
"entropy": 7.7105358123779295,
"epoch": 0.023387905111927832,
"grad_norm": 0.984375,
"learning_rate": 0.0001395,
"loss": 7.1147,
"mean_token_accuracy": 0.0869998462498188,
"num_tokens": 520680.0,
"step": 280
},
{
"entropy": 7.667366695404053,
"epoch": 0.023805546274640828,
"grad_norm": 1.0625,
"learning_rate": 0.00014199999999999998,
"loss": 7.0402,
"mean_token_accuracy": 0.08833886310458183,
"num_tokens": 529001.0,
"step": 285
},
{
"entropy": 7.6311241626739506,
"epoch": 0.024223187437353827,
"grad_norm": 1.0546875,
"learning_rate": 0.0001445,
"loss": 7.0129,
"mean_token_accuracy": 0.0985827162861824,
"num_tokens": 537309.0,
"step": 290
},
{
"entropy": 7.624583101272583,
"epoch": 0.024640828600066823,
"grad_norm": 0.97265625,
"learning_rate": 0.000147,
"loss": 6.9437,
"mean_token_accuracy": 0.09322682172060012,
"num_tokens": 546436.0,
"step": 295
},
{
"entropy": 7.56837830543518,
"epoch": 0.02505846976277982,
"grad_norm": 1.0234375,
"learning_rate": 0.0001495,
"loss": 7.0198,
"mean_token_accuracy": 0.09756411463022233,
"num_tokens": 555634.0,
"step": 300
},
{
"entropy": 7.483178758621216,
"epoch": 0.025476110925492818,
"grad_norm": 1.15625,
"learning_rate": 0.000152,
"loss": 6.8692,
"mean_token_accuracy": 0.09527311623096466,
"num_tokens": 564781.0,
"step": 305
},
{
"entropy": 7.586818838119507,
"epoch": 0.025893752088205813,
"grad_norm": 1.0234375,
"learning_rate": 0.00015450000000000001,
"loss": 7.0743,
"mean_token_accuracy": 0.0929844930768013,
"num_tokens": 575130.0,
"step": 310
},
{
"entropy": 7.626620864868164,
"epoch": 0.02631139325091881,
"grad_norm": 0.9921875,
"learning_rate": 0.000157,
"loss": 7.0136,
"mean_token_accuracy": 0.08612743467092514,
"num_tokens": 584030.0,
"step": 315
},
{
"entropy": 7.515700101852417,
"epoch": 0.026729034413631808,
"grad_norm": 0.96875,
"learning_rate": 0.0001595,
"loss": 6.9211,
"mean_token_accuracy": 0.09688968062400818,
"num_tokens": 593079.0,
"step": 320
},
{
"entropy": 7.4722246646881105,
"epoch": 0.027146675576344804,
"grad_norm": 1.015625,
"learning_rate": 0.000162,
"loss": 6.8694,
"mean_token_accuracy": 0.10736346021294593,
"num_tokens": 601830.0,
"step": 325
},
{
"entropy": 7.49754638671875,
"epoch": 0.027564316739057803,
"grad_norm": 1.2421875,
"learning_rate": 0.00016450000000000001,
"loss": 6.8304,
"mean_token_accuracy": 0.09651741534471511,
"num_tokens": 610296.0,
"step": 330
},
{
"entropy": 7.4197306632995605,
"epoch": 0.0279819579017708,
"grad_norm": 1.078125,
"learning_rate": 0.00016700000000000002,
"loss": 6.8731,
"mean_token_accuracy": 0.09373932555317879,
"num_tokens": 619762.0,
"step": 335
},
{
"entropy": 7.420308685302734,
"epoch": 0.028399599064483794,
"grad_norm": 1.015625,
"learning_rate": 0.00016950000000000003,
"loss": 6.8909,
"mean_token_accuracy": 0.09487934783101082,
"num_tokens": 629271.0,
"step": 340
},
{
"entropy": 7.52911901473999,
"epoch": 0.028817240227196794,
"grad_norm": 1.0078125,
"learning_rate": 0.00017199999999999998,
"loss": 6.9938,
"mean_token_accuracy": 0.08758045360445976,
"num_tokens": 638425.0,
"step": 345
},
{
"entropy": 7.482238721847534,
"epoch": 0.02923488138990979,
"grad_norm": 1.0390625,
"learning_rate": 0.00017449999999999999,
"loss": 6.9674,
"mean_token_accuracy": 0.09015953838825226,
"num_tokens": 647322.0,
"step": 350
},
{
"entropy": 7.503779792785645,
"epoch": 0.029652522552622785,
"grad_norm": 1.265625,
"learning_rate": 0.000177,
"loss": 6.9697,
"mean_token_accuracy": 0.08973397910594941,
"num_tokens": 656511.0,
"step": 355
},
{
"entropy": 7.481411170959473,
"epoch": 0.030070163715335784,
"grad_norm": 1.03125,
"learning_rate": 0.0001795,
"loss": 6.8918,
"mean_token_accuracy": 0.0935517430305481,
"num_tokens": 667039.0,
"step": 360
},
{
"entropy": 7.378106498718262,
"epoch": 0.03048780487804878,
"grad_norm": 1.125,
"learning_rate": 0.000182,
"loss": 6.766,
"mean_token_accuracy": 0.0998542919754982,
"num_tokens": 676065.0,
"step": 365
},
{
"entropy": 7.464444732666015,
"epoch": 0.03090544604076178,
"grad_norm": 1.1328125,
"learning_rate": 0.0001845,
"loss": 6.9181,
"mean_token_accuracy": 0.09695194140076638,
"num_tokens": 684344.0,
"step": 370
},
{
"entropy": 7.335521030426025,
"epoch": 0.03132308720347477,
"grad_norm": 1.1171875,
"learning_rate": 0.000187,
"loss": 6.873,
"mean_token_accuracy": 0.09614600911736489,
"num_tokens": 692839.0,
"step": 375
},
{
"entropy": 7.4365702152252195,
"epoch": 0.03174072836618777,
"grad_norm": 1.1484375,
"learning_rate": 0.0001895,
"loss": 6.845,
"mean_token_accuracy": 0.09973672106862068,
"num_tokens": 702689.0,
"step": 380
},
{
"entropy": 7.325565671920776,
"epoch": 0.03215836952890077,
"grad_norm": 1.1328125,
"learning_rate": 0.000192,
"loss": 6.8415,
"mean_token_accuracy": 0.0916935682296753,
"num_tokens": 711579.0,
"step": 385
},
{
"entropy": 7.428183269500733,
"epoch": 0.03257601069161377,
"grad_norm": 1.140625,
"learning_rate": 0.0001945,
"loss": 6.7867,
"mean_token_accuracy": 0.09845689535140992,
"num_tokens": 720469.0,
"step": 390
},
{
"entropy": 7.209290266036987,
"epoch": 0.03299365185432676,
"grad_norm": 0.9609375,
"learning_rate": 0.00019700000000000002,
"loss": 6.7604,
"mean_token_accuracy": 0.09956807419657707,
"num_tokens": 730326.0,
"step": 395
},
{
"entropy": 7.355399799346924,
"epoch": 0.03341129301703976,
"grad_norm": 1.0390625,
"learning_rate": 0.00019950000000000002,
"loss": 6.7572,
"mean_token_accuracy": 0.10625478997826576,
"num_tokens": 738992.0,
"step": 400
},
{
"entropy": 7.229115772247314,
"epoch": 0.03382893417975276,
"grad_norm": 1.046875,
"learning_rate": 0.000202,
"loss": 6.8151,
"mean_token_accuracy": 0.09731377884745598,
"num_tokens": 748387.0,
"step": 405
},
{
"entropy": 7.369128704071045,
"epoch": 0.03424657534246575,
"grad_norm": 1.1015625,
"learning_rate": 0.00020449999999999998,
"loss": 6.7356,
"mean_token_accuracy": 0.10202734097838402,
"num_tokens": 757712.0,
"step": 410
},
{
"entropy": 7.201074743270874,
"epoch": 0.03466421650517875,
"grad_norm": 1.1875,
"learning_rate": 0.000207,
"loss": 6.7106,
"mean_token_accuracy": 0.09947160556912422,
"num_tokens": 766626.0,
"step": 415
},
{
"entropy": 7.218473052978515,
"epoch": 0.03508185766789175,
"grad_norm": 1.0078125,
"learning_rate": 0.0002095,
"loss": 6.6622,
"mean_token_accuracy": 0.10467798709869384,
"num_tokens": 775942.0,
"step": 420
},
{
"entropy": 7.319445276260376,
"epoch": 0.03549949883060474,
"grad_norm": 1.21875,
"learning_rate": 0.000212,
"loss": 6.743,
"mean_token_accuracy": 0.10339497774839401,
"num_tokens": 784680.0,
"step": 425
},
{
"entropy": 7.324266147613526,
"epoch": 0.03591713999331774,
"grad_norm": 1.2421875,
"learning_rate": 0.0002145,
"loss": 6.8274,
"mean_token_accuracy": 0.09892603307962418,
"num_tokens": 794107.0,
"step": 430
},
{
"entropy": 7.254756164550781,
"epoch": 0.03633478115603074,
"grad_norm": 1.21875,
"learning_rate": 0.00021700000000000002,
"loss": 6.796,
"mean_token_accuracy": 0.10009015426039695,
"num_tokens": 803584.0,
"step": 435
},
{
"entropy": 7.310313606262207,
"epoch": 0.03675242231874373,
"grad_norm": 1.1015625,
"learning_rate": 0.0002195,
"loss": 6.7773,
"mean_token_accuracy": 0.10064135193824768,
"num_tokens": 812460.0,
"step": 440
},
{
"entropy": 7.196012496948242,
"epoch": 0.03717006348145673,
"grad_norm": 1.234375,
"learning_rate": 0.000222,
"loss": 6.7503,
"mean_token_accuracy": 0.10141536369919776,
"num_tokens": 821946.0,
"step": 445
},
{
"entropy": 7.204004621505737,
"epoch": 0.03758770464416973,
"grad_norm": 1.015625,
"learning_rate": 0.0002245,
"loss": 6.6911,
"mean_token_accuracy": 0.09515870362520218,
"num_tokens": 831781.0,
"step": 450
},
{
"entropy": 7.2210071086883545,
"epoch": 0.03800534580688272,
"grad_norm": 1.09375,
"learning_rate": 0.00022700000000000002,
"loss": 6.6405,
"mean_token_accuracy": 0.10319173634052277,
"num_tokens": 840903.0,
"step": 455
},
{
"entropy": 7.1188592433929445,
"epoch": 0.03842298696959572,
"grad_norm": 0.97265625,
"learning_rate": 0.00022950000000000002,
"loss": 6.7495,
"mean_token_accuracy": 0.1028764121234417,
"num_tokens": 850545.0,
"step": 460
},
{
"entropy": 7.177313804626465,
"epoch": 0.03884062813230872,
"grad_norm": 1.2109375,
"learning_rate": 0.00023200000000000003,
"loss": 6.726,
"mean_token_accuracy": 0.10074673295021057,
"num_tokens": 860606.0,
"step": 465
},
{
"entropy": 7.329133081436157,
"epoch": 0.03925826929502172,
"grad_norm": 1.15625,
"learning_rate": 0.00023449999999999998,
"loss": 6.7694,
"mean_token_accuracy": 0.10264237001538276,
"num_tokens": 868767.0,
"step": 470
},
{
"entropy": 7.103873348236084,
"epoch": 0.03967591045773471,
"grad_norm": 1.2109375,
"learning_rate": 0.000237,
"loss": 6.6409,
"mean_token_accuracy": 0.1013137087225914,
"num_tokens": 877820.0,
"step": 475
},
{
"entropy": 7.212535333633423,
"epoch": 0.04009355162044771,
"grad_norm": 1.125,
"learning_rate": 0.0002395,
"loss": 6.7729,
"mean_token_accuracy": 0.09580215588212013,
"num_tokens": 887035.0,
"step": 480
},
{
"entropy": 7.261131954193115,
"epoch": 0.04051119278316071,
"grad_norm": 1.1640625,
"learning_rate": 0.000242,
"loss": 6.7436,
"mean_token_accuracy": 0.10203550904989242,
"num_tokens": 896249.0,
"step": 485
},
{
"entropy": 7.094175338745117,
"epoch": 0.040928833945873704,
"grad_norm": 1.2421875,
"learning_rate": 0.0002445,
"loss": 6.6564,
"mean_token_accuracy": 0.09608230963349343,
"num_tokens": 905307.0,
"step": 490
},
{
"entropy": 7.0745786190032955,
"epoch": 0.0413464751085867,
"grad_norm": 0.984375,
"learning_rate": 0.000247,
"loss": 6.6378,
"mean_token_accuracy": 0.10336728692054749,
"num_tokens": 914587.0,
"step": 495
},
{
"entropy": 7.122777318954467,
"epoch": 0.0417641162712997,
"grad_norm": 0.9453125,
"learning_rate": 0.0002495,
"loss": 6.5584,
"mean_token_accuracy": 0.11078567281365395,
"num_tokens": 923361.0,
"step": 500
},
{
"entropy": 7.154393339157105,
"epoch": 0.042181757434012694,
"grad_norm": 1.1796875,
"learning_rate": 0.000252,
"loss": 6.6615,
"mean_token_accuracy": 0.10329530760645866,
"num_tokens": 931698.0,
"step": 505
},
{
"entropy": 7.1185431480407715,
"epoch": 0.04259939859672569,
"grad_norm": 1.2578125,
"learning_rate": 0.0002545,
"loss": 6.6228,
"mean_token_accuracy": 0.1053171068429947,
"num_tokens": 940141.0,
"step": 510
},
{
"entropy": 6.98853235244751,
"epoch": 0.04301703975943869,
"grad_norm": 1.109375,
"learning_rate": 0.000257,
"loss": 6.5873,
"mean_token_accuracy": 0.10950357466936111,
"num_tokens": 948926.0,
"step": 515
},
{
"entropy": 7.208238315582276,
"epoch": 0.043434680922151685,
"grad_norm": 1.109375,
"learning_rate": 0.0002595,
"loss": 6.6677,
"mean_token_accuracy": 0.10224388837814331,
"num_tokens": 957995.0,
"step": 520
},
{
"entropy": 7.085497188568115,
"epoch": 0.043852322084864684,
"grad_norm": 1.03125,
"learning_rate": 0.000262,
"loss": 6.6741,
"mean_token_accuracy": 0.10617537721991539,
"num_tokens": 968004.0,
"step": 525
},
{
"entropy": 7.1213349342346195,
"epoch": 0.04426996324757768,
"grad_norm": 1.140625,
"learning_rate": 0.00026450000000000003,
"loss": 6.6951,
"mean_token_accuracy": 0.10752663984894753,
"num_tokens": 977931.0,
"step": 530
},
{
"entropy": 6.977365303039551,
"epoch": 0.044687604410290675,
"grad_norm": 1.234375,
"learning_rate": 0.00026700000000000004,
"loss": 6.6561,
"mean_token_accuracy": 0.10034975931048393,
"num_tokens": 986904.0,
"step": 535
},
{
"entropy": 7.063871145248413,
"epoch": 0.045105245573003674,
"grad_norm": 1.0859375,
"learning_rate": 0.00026950000000000005,
"loss": 6.6027,
"mean_token_accuracy": 0.10650393143296241,
"num_tokens": 997482.0,
"step": 540
},
{
"entropy": 7.0778279304504395,
"epoch": 0.045522886735716674,
"grad_norm": 1.015625,
"learning_rate": 0.00027200000000000005,
"loss": 6.6128,
"mean_token_accuracy": 0.10725649818778038,
"num_tokens": 1007196.0,
"step": 545
},
{
"entropy": 7.058020162582397,
"epoch": 0.045940527898429666,
"grad_norm": 1.0703125,
"learning_rate": 0.0002745,
"loss": 6.7021,
"mean_token_accuracy": 0.10655904859304428,
"num_tokens": 1016435.0,
"step": 550
},
{
"entropy": 7.029399967193603,
"epoch": 0.046358169061142665,
"grad_norm": 1.109375,
"learning_rate": 0.000277,
"loss": 6.5814,
"mean_token_accuracy": 0.10815284475684166,
"num_tokens": 1026299.0,
"step": 555
},
{
"entropy": 6.969214200973511,
"epoch": 0.046775810223855664,
"grad_norm": 1.046875,
"learning_rate": 0.0002795,
"loss": 6.5411,
"mean_token_accuracy": 0.11086997389793396,
"num_tokens": 1034672.0,
"step": 560
},
{
"entropy": 6.929189205169678,
"epoch": 0.04719345138656866,
"grad_norm": 1.140625,
"learning_rate": 0.00028199999999999997,
"loss": 6.5294,
"mean_token_accuracy": 0.11677401661872863,
"num_tokens": 1042786.0,
"step": 565
},
{
"entropy": 7.0000075340271,
"epoch": 0.047611092549281656,
"grad_norm": 1.046875,
"learning_rate": 0.0002845,
"loss": 6.5073,
"mean_token_accuracy": 0.11181584969162942,
"num_tokens": 1052709.0,
"step": 570
},
{
"entropy": 7.024741125106812,
"epoch": 0.048028733711994655,
"grad_norm": 1.15625,
"learning_rate": 0.000287,
"loss": 6.5826,
"mean_token_accuracy": 0.10405817180871964,
"num_tokens": 1061880.0,
"step": 575
},
{
"entropy": 7.037390232086182,
"epoch": 0.048446374874707654,
"grad_norm": 1.1953125,
"learning_rate": 0.0002895,
"loss": 6.5537,
"mean_token_accuracy": 0.10765022113919258,
"num_tokens": 1071234.0,
"step": 580
},
{
"entropy": 7.0248270511627195,
"epoch": 0.048864016037420646,
"grad_norm": 1.265625,
"learning_rate": 0.000292,
"loss": 6.6946,
"mean_token_accuracy": 0.10200266987085342,
"num_tokens": 1081252.0,
"step": 585
},
{
"entropy": 6.906812763214111,
"epoch": 0.049281657200133645,
"grad_norm": 1.0546875,
"learning_rate": 0.0002945,
"loss": 6.5343,
"mean_token_accuracy": 0.10761479586362839,
"num_tokens": 1090642.0,
"step": 590
},
{
"entropy": 7.004644918441772,
"epoch": 0.049699298362846644,
"grad_norm": 1.0859375,
"learning_rate": 0.000297,
"loss": 6.5122,
"mean_token_accuracy": 0.1107115812599659,
"num_tokens": 1099918.0,
"step": 595
},
{
"entropy": 7.037489223480224,
"epoch": 0.05011693952555964,
"grad_norm": 1.2109375,
"learning_rate": 0.0002995,
"loss": 6.5498,
"mean_token_accuracy": 0.10938952788710594,
"num_tokens": 1109521.0,
"step": 600
},
{
"entropy": 6.80020923614502,
"epoch": 0.050534580688272636,
"grad_norm": 1.265625,
"learning_rate": 0.000302,
"loss": 6.4223,
"mean_token_accuracy": 0.12316331788897514,
"num_tokens": 1118140.0,
"step": 605
},
{
"entropy": 6.91481761932373,
"epoch": 0.050952221850985635,
"grad_norm": 0.96484375,
"learning_rate": 0.0003045,
"loss": 6.5787,
"mean_token_accuracy": 0.10839839428663253,
"num_tokens": 1127591.0,
"step": 610
},
{
"entropy": 7.027100372314453,
"epoch": 0.05136986301369863,
"grad_norm": 1.140625,
"learning_rate": 0.000307,
"loss": 6.5591,
"mean_token_accuracy": 0.1044753447175026,
"num_tokens": 1136223.0,
"step": 615
},
{
"entropy": 6.935626745223999,
"epoch": 0.051787504176411626,
"grad_norm": 1.1640625,
"learning_rate": 0.0003095,
"loss": 6.4368,
"mean_token_accuracy": 0.11771413609385491,
"num_tokens": 1146367.0,
"step": 620
},
{
"entropy": 6.7960490703582765,
"epoch": 0.052205145339124626,
"grad_norm": 0.96875,
"learning_rate": 0.000312,
"loss": 6.4507,
"mean_token_accuracy": 0.11227771192789078,
"num_tokens": 1156230.0,
"step": 625
},
{
"entropy": 6.978573226928711,
"epoch": 0.05262278650183762,
"grad_norm": 1.0703125,
"learning_rate": 0.0003145,
"loss": 6.5624,
"mean_token_accuracy": 0.10628528222441673,
"num_tokens": 1166106.0,
"step": 630
},
{
"entropy": 6.914755392074585,
"epoch": 0.05304042766455062,
"grad_norm": 1.09375,
"learning_rate": 0.000317,
"loss": 6.5656,
"mean_token_accuracy": 0.11575563699007034,
"num_tokens": 1174887.0,
"step": 635
},
{
"entropy": 6.908324193954468,
"epoch": 0.053458068827263616,
"grad_norm": 1.234375,
"learning_rate": 0.0003195,
"loss": 6.5233,
"mean_token_accuracy": 0.12051590755581856,
"num_tokens": 1184027.0,
"step": 640
},
{
"entropy": 6.857800912857056,
"epoch": 0.053875709989976615,
"grad_norm": 0.921875,
"learning_rate": 0.000322,
"loss": 6.4713,
"mean_token_accuracy": 0.1128120668232441,
"num_tokens": 1194632.0,
"step": 645
},
{
"entropy": 6.883691072463989,
"epoch": 0.05429335115268961,
"grad_norm": 1.265625,
"learning_rate": 0.00032450000000000003,
"loss": 6.4153,
"mean_token_accuracy": 0.12454415932297706,
"num_tokens": 1204601.0,
"step": 650
},
{
"entropy": 6.806664991378784,
"epoch": 0.05471099231540261,
"grad_norm": 1.1484375,
"learning_rate": 0.00032700000000000003,
"loss": 6.6042,
"mean_token_accuracy": 0.11112436130642891,
"num_tokens": 1213288.0,
"step": 655
},
{
"entropy": 6.8806665420532225,
"epoch": 0.055128633478115606,
"grad_norm": 1.0625,
"learning_rate": 0.00032950000000000004,
"loss": 6.5376,
"mean_token_accuracy": 0.11152591183781624,
"num_tokens": 1222768.0,
"step": 660
},
{
"entropy": 6.910279750823975,
"epoch": 0.0555462746408286,
"grad_norm": 1.109375,
"learning_rate": 0.00033200000000000005,
"loss": 6.5191,
"mean_token_accuracy": 0.1134497992694378,
"num_tokens": 1232421.0,
"step": 665
},
{
"entropy": 6.828466987609863,
"epoch": 0.0559639158035416,
"grad_norm": 1.2109375,
"learning_rate": 0.00033450000000000005,
"loss": 6.3929,
"mean_token_accuracy": 0.12303503230214119,
"num_tokens": 1241230.0,
"step": 670
},
{
"entropy": 6.886627149581909,
"epoch": 0.056381556966254597,
"grad_norm": 1.0234375,
"learning_rate": 0.000337,
"loss": 6.5597,
"mean_token_accuracy": 0.11267556771636009,
"num_tokens": 1250879.0,
"step": 675
},
{
"entropy": 6.886648941040039,
"epoch": 0.05679919812896759,
"grad_norm": 1.2421875,
"learning_rate": 0.0003395,
"loss": 6.4169,
"mean_token_accuracy": 0.12256434261798858,
"num_tokens": 1259073.0,
"step": 680
},
{
"entropy": 6.741843509674072,
"epoch": 0.05721683929168059,
"grad_norm": 1.21875,
"learning_rate": 0.000342,
"loss": 6.4493,
"mean_token_accuracy": 0.11314508318901062,
"num_tokens": 1269174.0,
"step": 685
},
{
"entropy": 6.916187334060669,
"epoch": 0.05763448045439359,
"grad_norm": 1.140625,
"learning_rate": 0.00034449999999999997,
"loss": 6.6023,
"mean_token_accuracy": 0.11100954487919808,
"num_tokens": 1279366.0,
"step": 690
},
{
"entropy": 6.768966245651245,
"epoch": 0.05805212161710658,
"grad_norm": 1.203125,
"learning_rate": 0.000347,
"loss": 6.4105,
"mean_token_accuracy": 0.11207879930734635,
"num_tokens": 1288879.0,
"step": 695
},
{
"entropy": 6.743048238754272,
"epoch": 0.05846976277981958,
"grad_norm": 1.1796875,
"learning_rate": 0.0003495,
"loss": 6.3024,
"mean_token_accuracy": 0.11753646731376648,
"num_tokens": 1298098.0,
"step": 700
},
{
"entropy": 6.770110464096069,
"epoch": 0.05888740394253258,
"grad_norm": 1.125,
"learning_rate": 0.000352,
"loss": 6.4606,
"mean_token_accuracy": 0.11356354281306266,
"num_tokens": 1306960.0,
"step": 705
},
{
"entropy": 6.854166841506958,
"epoch": 0.05930504510524557,
"grad_norm": 1.0390625,
"learning_rate": 0.0003545,
"loss": 6.5218,
"mean_token_accuracy": 0.11107228845357894,
"num_tokens": 1317252.0,
"step": 710
},
{
"entropy": 6.823494577407837,
"epoch": 0.05972268626795857,
"grad_norm": 1.28125,
"learning_rate": 0.000357,
"loss": 6.427,
"mean_token_accuracy": 0.11604731008410454,
"num_tokens": 1326712.0,
"step": 715
},
{
"entropy": 6.788588237762451,
"epoch": 0.06014032743067157,
"grad_norm": 1.203125,
"learning_rate": 0.0003595,
"loss": 6.4856,
"mean_token_accuracy": 0.11697607561945915,
"num_tokens": 1336205.0,
"step": 720
},
{
"entropy": 6.823712778091431,
"epoch": 0.06055796859338457,
"grad_norm": 1.0078125,
"learning_rate": 0.000362,
"loss": 6.4533,
"mean_token_accuracy": 0.1154071107506752,
"num_tokens": 1345347.0,
"step": 725
},
{
"entropy": 6.788979196548462,
"epoch": 0.06097560975609756,
"grad_norm": 1.0234375,
"learning_rate": 0.0003645,
"loss": 6.3888,
"mean_token_accuracy": 0.11326988786458969,
"num_tokens": 1354267.0,
"step": 730
},
{
"entropy": 6.707552146911621,
"epoch": 0.06139325091881056,
"grad_norm": 1.1484375,
"learning_rate": 0.000367,
"loss": 6.4253,
"mean_token_accuracy": 0.10984169691801071,
"num_tokens": 1363949.0,
"step": 735
},
{
"entropy": 6.847339773178101,
"epoch": 0.06181089208152356,
"grad_norm": 1.2109375,
"learning_rate": 0.0003695,
"loss": 6.46,
"mean_token_accuracy": 0.11028305515646934,
"num_tokens": 1374405.0,
"step": 740
},
{
"entropy": 6.574991655349732,
"epoch": 0.06222853324423655,
"grad_norm": 1.0078125,
"learning_rate": 0.000372,
"loss": 6.3733,
"mean_token_accuracy": 0.12012771666049957,
"num_tokens": 1384156.0,
"step": 745
},
{
"entropy": 6.7866004467010494,
"epoch": 0.06264617440694954,
"grad_norm": 1.0390625,
"learning_rate": 0.0003745,
"loss": 6.3843,
"mean_token_accuracy": 0.12277914807200432,
"num_tokens": 1393896.0,
"step": 750
},
{
"entropy": 6.672963714599609,
"epoch": 0.06306381556966255,
"grad_norm": 0.9296875,
"learning_rate": 0.000377,
"loss": 6.2705,
"mean_token_accuracy": 0.1256583385169506,
"num_tokens": 1403842.0,
"step": 755
},
{
"entropy": 6.712489080429077,
"epoch": 0.06348145673237554,
"grad_norm": 1.1796875,
"learning_rate": 0.0003795,
"loss": 6.3878,
"mean_token_accuracy": 0.12302643731236458,
"num_tokens": 1412644.0,
"step": 760
},
{
"entropy": 6.6753857135772705,
"epoch": 0.06389909789508853,
"grad_norm": 1.1015625,
"learning_rate": 0.000382,
"loss": 6.3293,
"mean_token_accuracy": 0.12695508673787118,
"num_tokens": 1421815.0,
"step": 765
},
{
"entropy": 6.762657022476196,
"epoch": 0.06431673905780154,
"grad_norm": 1.109375,
"learning_rate": 0.0003845,
"loss": 6.4556,
"mean_token_accuracy": 0.11635951921343804,
"num_tokens": 1431572.0,
"step": 770
},
{
"entropy": 6.64413104057312,
"epoch": 0.06473438022051453,
"grad_norm": 1.046875,
"learning_rate": 0.00038700000000000003,
"loss": 6.3452,
"mean_token_accuracy": 0.12333026081323624,
"num_tokens": 1440609.0,
"step": 775
},
{
"entropy": 6.771637105941773,
"epoch": 0.06515202138322754,
"grad_norm": 1.1484375,
"learning_rate": 0.00038950000000000003,
"loss": 6.4553,
"mean_token_accuracy": 0.12362872958183288,
"num_tokens": 1451193.0,
"step": 780
},
{
"entropy": 6.616492509841919,
"epoch": 0.06556966254594053,
"grad_norm": 1.1484375,
"learning_rate": 0.00039200000000000004,
"loss": 6.263,
"mean_token_accuracy": 0.12763497680425645,
"num_tokens": 1459772.0,
"step": 785
},
{
"entropy": 6.700885200500489,
"epoch": 0.06598730370865352,
"grad_norm": 1.015625,
"learning_rate": 0.00039450000000000005,
"loss": 6.3668,
"mean_token_accuracy": 0.1201256386935711,
"num_tokens": 1469251.0,
"step": 790
},
{
"entropy": 6.637189674377441,
"epoch": 0.06640494487136653,
"grad_norm": 0.98046875,
"learning_rate": 0.00039700000000000005,
"loss": 6.486,
"mean_token_accuracy": 0.11636328250169754,
"num_tokens": 1478090.0,
"step": 795
},
{
"entropy": 6.766718864440918,
"epoch": 0.06682258603407952,
"grad_norm": 1.1328125,
"learning_rate": 0.0003995,
"loss": 6.3289,
"mean_token_accuracy": 0.11908229663968087,
"num_tokens": 1488231.0,
"step": 800
},
{
"entropy": 6.748629426956176,
"epoch": 0.06724022719679251,
"grad_norm": 1.1875,
"learning_rate": 0.000402,
"loss": 6.4014,
"mean_token_accuracy": 0.12175856158137321,
"num_tokens": 1496904.0,
"step": 805
},
{
"entropy": 6.523228120803833,
"epoch": 0.06765786835950552,
"grad_norm": 1.09375,
"learning_rate": 0.0004045,
"loss": 6.334,
"mean_token_accuracy": 0.12687446549534798,
"num_tokens": 1506006.0,
"step": 810
},
{
"entropy": 6.733132934570312,
"epoch": 0.06807550952221851,
"grad_norm": 0.91796875,
"learning_rate": 0.00040699999999999997,
"loss": 6.2813,
"mean_token_accuracy": 0.12203963398933411,
"num_tokens": 1515826.0,
"step": 815
},
{
"entropy": 6.6297523975372314,
"epoch": 0.0684931506849315,
"grad_norm": 1.265625,
"learning_rate": 0.0004095,
"loss": 6.3552,
"mean_token_accuracy": 0.11122297570109367,
"num_tokens": 1525001.0,
"step": 820
},
{
"entropy": 6.583107852935791,
"epoch": 0.06891079184764451,
"grad_norm": 1.09375,
"learning_rate": 0.000412,
"loss": 6.3229,
"mean_token_accuracy": 0.11932475045323372,
"num_tokens": 1533647.0,
"step": 825
},
{
"entropy": 6.6441779136657715,
"epoch": 0.0693284330103575,
"grad_norm": 1.2265625,
"learning_rate": 0.0004145,
"loss": 6.3445,
"mean_token_accuracy": 0.11783274039626121,
"num_tokens": 1542461.0,
"step": 830
},
{
"entropy": 6.660567283630371,
"epoch": 0.0697460741730705,
"grad_norm": 1.1328125,
"learning_rate": 0.000417,
"loss": 6.2262,
"mean_token_accuracy": 0.1264207936823368,
"num_tokens": 1552249.0,
"step": 835
},
{
"entropy": 6.576170206069946,
"epoch": 0.0701637153357835,
"grad_norm": 0.99609375,
"learning_rate": 0.0004195,
"loss": 6.3921,
"mean_token_accuracy": 0.12855046465992928,
"num_tokens": 1561570.0,
"step": 840
},
{
"entropy": 6.734837198257447,
"epoch": 0.07058135649849649,
"grad_norm": 0.9296875,
"learning_rate": 0.000422,
"loss": 6.4302,
"mean_token_accuracy": 0.11865505427122117,
"num_tokens": 1572263.0,
"step": 845
},
{
"entropy": 6.5386943340301515,
"epoch": 0.07099899766120948,
"grad_norm": 1.0703125,
"learning_rate": 0.0004245,
"loss": 6.2651,
"mean_token_accuracy": 0.12368358746170997,
"num_tokens": 1580828.0,
"step": 850
},
{
"entropy": 6.6218860149383545,
"epoch": 0.07141663882392249,
"grad_norm": 1.0,
"learning_rate": 0.000427,
"loss": 6.2414,
"mean_token_accuracy": 0.11890774369239807,
"num_tokens": 1589858.0,
"step": 855
},
{
"entropy": 6.687988328933716,
"epoch": 0.07183427998663548,
"grad_norm": 1.046875,
"learning_rate": 0.0004295,
"loss": 6.3765,
"mean_token_accuracy": 0.12400484904646873,
"num_tokens": 1598822.0,
"step": 860
},
{
"entropy": 6.580365896224976,
"epoch": 0.07225192114934847,
"grad_norm": 1.1171875,
"learning_rate": 0.000432,
"loss": 6.2517,
"mean_token_accuracy": 0.1287302441895008,
"num_tokens": 1607508.0,
"step": 865
},
{
"entropy": 6.569807338714599,
"epoch": 0.07266956231206148,
"grad_norm": 1.0,
"learning_rate": 0.0004345,
"loss": 6.2682,
"mean_token_accuracy": 0.1310380905866623,
"num_tokens": 1616986.0,
"step": 870
},
{
"entropy": 6.714363670349121,
"epoch": 0.07308720347477447,
"grad_norm": 1.1171875,
"learning_rate": 0.000437,
"loss": 6.3465,
"mean_token_accuracy": 0.12610832080245019,
"num_tokens": 1626940.0,
"step": 875
},
{
"entropy": 6.540983772277832,
"epoch": 0.07350484463748747,
"grad_norm": 1.0625,
"learning_rate": 0.0004395,
"loss": 6.3218,
"mean_token_accuracy": 0.12229821383953095,
"num_tokens": 1636249.0,
"step": 880
},
{
"entropy": 6.62174711227417,
"epoch": 0.07392248580020047,
"grad_norm": 0.984375,
"learning_rate": 0.000442,
"loss": 6.3601,
"mean_token_accuracy": 0.12111141234636306,
"num_tokens": 1645943.0,
"step": 885
},
{
"entropy": 6.664224576950073,
"epoch": 0.07434012696291346,
"grad_norm": 1.1328125,
"learning_rate": 0.0004445,
"loss": 6.402,
"mean_token_accuracy": 0.12018121778964996,
"num_tokens": 1655119.0,
"step": 890
},
{
"entropy": 6.53070068359375,
"epoch": 0.07475776812562646,
"grad_norm": 1.1953125,
"learning_rate": 0.000447,
"loss": 6.3106,
"mean_token_accuracy": 0.124847911298275,
"num_tokens": 1663635.0,
"step": 895
},
{
"entropy": 6.53735842704773,
"epoch": 0.07517540928833946,
"grad_norm": 1.390625,
"learning_rate": 0.00044950000000000003,
"loss": 6.1585,
"mean_token_accuracy": 0.13482338264584542,
"num_tokens": 1672360.0,
"step": 900
},
{
"entropy": 6.471534395217896,
"epoch": 0.07559305045105245,
"grad_norm": 1.0546875,
"learning_rate": 0.00045200000000000004,
"loss": 6.2349,
"mean_token_accuracy": 0.1260572887957096,
"num_tokens": 1681750.0,
"step": 905
},
{
"entropy": 6.578565979003907,
"epoch": 0.07601069161376545,
"grad_norm": 1.0859375,
"learning_rate": 0.00045450000000000004,
"loss": 6.2329,
"mean_token_accuracy": 0.13175752609968186,
"num_tokens": 1690434.0,
"step": 910
},
{
"entropy": 6.559924745559693,
"epoch": 0.07642833277647845,
"grad_norm": 1.15625,
"learning_rate": 0.00045700000000000005,
"loss": 6.2595,
"mean_token_accuracy": 0.12138056978583336,
"num_tokens": 1699982.0,
"step": 915
},
{
"entropy": 6.432953596115112,
"epoch": 0.07684597393919144,
"grad_norm": 1.203125,
"learning_rate": 0.00045950000000000006,
"loss": 6.187,
"mean_token_accuracy": 0.12513865381479264,
"num_tokens": 1709251.0,
"step": 920
},
{
"entropy": 6.492142963409424,
"epoch": 0.07726361510190444,
"grad_norm": 1.1796875,
"learning_rate": 0.000462,
"loss": 6.2698,
"mean_token_accuracy": 0.1200482964515686,
"num_tokens": 1718556.0,
"step": 925
},
{
"entropy": 6.459371662139892,
"epoch": 0.07768125626461744,
"grad_norm": 1.0703125,
"learning_rate": 0.0004645,
"loss": 6.236,
"mean_token_accuracy": 0.1269206404685974,
"num_tokens": 1728265.0,
"step": 930
},
{
"entropy": 6.594856071472168,
"epoch": 0.07809889742733044,
"grad_norm": 1.1171875,
"learning_rate": 0.000467,
"loss": 6.2492,
"mean_token_accuracy": 0.12795129865407945,
"num_tokens": 1737537.0,
"step": 935
},
{
"entropy": 6.351608943939209,
"epoch": 0.07851653859004344,
"grad_norm": 1.0390625,
"learning_rate": 0.0004695,
"loss": 6.1095,
"mean_token_accuracy": 0.12889909520745277,
"num_tokens": 1747357.0,
"step": 940
},
{
"entropy": 6.573435544967651,
"epoch": 0.07893417975275643,
"grad_norm": 1.0859375,
"learning_rate": 0.000472,
"loss": 6.3365,
"mean_token_accuracy": 0.11498595103621483,
"num_tokens": 1757462.0,
"step": 945
},
{
"entropy": 6.536618280410766,
"epoch": 0.07935182091546943,
"grad_norm": 1.2109375,
"learning_rate": 0.0004745,
"loss": 6.175,
"mean_token_accuracy": 0.13465738147497178,
"num_tokens": 1766127.0,
"step": 950
},
{
"entropy": 6.49476056098938,
"epoch": 0.07976946207818243,
"grad_norm": 1.09375,
"learning_rate": 0.000477,
"loss": 6.2778,
"mean_token_accuracy": 0.12881955578923227,
"num_tokens": 1775106.0,
"step": 955
},
{
"entropy": 6.541385173797607,
"epoch": 0.08018710324089542,
"grad_norm": 0.96875,
"learning_rate": 0.0004795,
"loss": 6.2576,
"mean_token_accuracy": 0.13110605105757714,
"num_tokens": 1785626.0,
"step": 960
},
{
"entropy": 6.4958192825317385,
"epoch": 0.08060474440360842,
"grad_norm": 1.078125,
"learning_rate": 0.000482,
"loss": 6.2885,
"mean_token_accuracy": 0.12439391762018204,
"num_tokens": 1795064.0,
"step": 965
},
{
"entropy": 6.43884506225586,
"epoch": 0.08102238556632142,
"grad_norm": 1.1640625,
"learning_rate": 0.0004845,
"loss": 6.2553,
"mean_token_accuracy": 0.12753869816660882,
"num_tokens": 1804429.0,
"step": 970
},
{
"entropy": 6.597222757339478,
"epoch": 0.08144002672903441,
"grad_norm": 1.1015625,
"learning_rate": 0.000487,
"loss": 6.2732,
"mean_token_accuracy": 0.12425616905093192,
"num_tokens": 1813542.0,
"step": 975
},
{
"entropy": 6.474799776077271,
"epoch": 0.08185766789174741,
"grad_norm": 0.91796875,
"learning_rate": 0.0004895,
"loss": 6.1995,
"mean_token_accuracy": 0.12659965306520463,
"num_tokens": 1824172.0,
"step": 980
},
{
"entropy": 6.40195951461792,
"epoch": 0.08227530905446041,
"grad_norm": 1.09375,
"learning_rate": 0.000492,
"loss": 6.2122,
"mean_token_accuracy": 0.12306850850582123,
"num_tokens": 1832989.0,
"step": 985
},
{
"entropy": 6.536772537231445,
"epoch": 0.0826929502171734,
"grad_norm": 1.0546875,
"learning_rate": 0.0004945,
"loss": 6.1649,
"mean_token_accuracy": 0.1325046107172966,
"num_tokens": 1842345.0,
"step": 990
},
{
"entropy": 6.416742944717408,
"epoch": 0.0831105913798864,
"grad_norm": 1.0390625,
"learning_rate": 0.000497,
"loss": 6.2391,
"mean_token_accuracy": 0.12276924774050713,
"num_tokens": 1851952.0,
"step": 995
},
{
"entropy": 6.37835545539856,
"epoch": 0.0835282325425994,
"grad_norm": 1.140625,
"learning_rate": 0.0004995,
"loss": 6.2235,
"mean_token_accuracy": 0.12877754718065262,
"num_tokens": 1861029.0,
"step": 1000
},
{
"entropy": 6.685376405715942,
"epoch": 0.0839458737053124,
"grad_norm": 1.078125,
"learning_rate": 0.0004999999987395534,
"loss": 6.296,
"mean_token_accuracy": 0.13085904866456985,
"num_tokens": 1869650.0,
"step": 1005
},
{
"entropy": 6.526375818252563,
"epoch": 0.08436351486802539,
"grad_norm": 1.03125,
"learning_rate": 0.0004999999936189888,
"loss": 6.35,
"mean_token_accuracy": 0.13077539652585984,
"num_tokens": 1879078.0,
"step": 1010
},
{
"entropy": 6.410852146148682,
"epoch": 0.0847811560307384,
"grad_norm": 1.109375,
"learning_rate": 0.0004999999845595285,
"loss": 6.1781,
"mean_token_accuracy": 0.12249526754021645,
"num_tokens": 1888596.0,
"step": 1015
},
{
"entropy": 6.411129570007324,
"epoch": 0.08519879719345139,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999999715611727,
"loss": 6.1918,
"mean_token_accuracy": 0.13483786955475807,
"num_tokens": 1897342.0,
"step": 1020
},
{
"entropy": 6.401776599884033,
"epoch": 0.08561643835616438,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999999546239215,
"loss": 6.1331,
"mean_token_accuracy": 0.13173556253314017,
"num_tokens": 1905685.0,
"step": 1025
},
{
"entropy": 6.575545740127564,
"epoch": 0.08603407951887738,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999999337477753,
"loss": 6.3381,
"mean_token_accuracy": 0.11911348029971122,
"num_tokens": 1915195.0,
"step": 1030
},
{
"entropy": 6.471239280700684,
"epoch": 0.08645172068159038,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999999089327344,
"loss": 6.3375,
"mean_token_accuracy": 0.11769160553812981,
"num_tokens": 1925156.0,
"step": 1035
},
{
"entropy": 6.4332513332366945,
"epoch": 0.08686936184430337,
"grad_norm": 1.1875,
"learning_rate": 0.0004999998801787993,
"loss": 6.2073,
"mean_token_accuracy": 0.1246685691177845,
"num_tokens": 1934521.0,
"step": 1040
},
{
"entropy": 6.542106246948242,
"epoch": 0.08728700300701638,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999998474859704,
"loss": 6.1181,
"mean_token_accuracy": 0.12874981313943862,
"num_tokens": 1944160.0,
"step": 1045
},
{
"entropy": 6.339631366729736,
"epoch": 0.08770464416972937,
"grad_norm": 1.03125,
"learning_rate": 0.0004999998108542483,
"loss": 6.1041,
"mean_token_accuracy": 0.13465228974819182,
"num_tokens": 1953022.0,
"step": 1050
},
{
"entropy": 6.239454984664917,
"epoch": 0.08812228533244236,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999997702836339,
"loss": 6.0884,
"mean_token_accuracy": 0.12914091050624849,
"num_tokens": 1961544.0,
"step": 1055
},
{
"entropy": 6.530628108978272,
"epoch": 0.08853992649515537,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999997257741274,
"loss": 6.199,
"mean_token_accuracy": 0.1303819626569748,
"num_tokens": 1970319.0,
"step": 1060
},
{
"entropy": 6.2712071418762205,
"epoch": 0.08895756765786836,
"grad_norm": 1.0625,
"learning_rate": 0.00049999967732573,
"loss": 6.0488,
"mean_token_accuracy": 0.13570455610752105,
"num_tokens": 1979582.0,
"step": 1065
},
{
"entropy": 6.4226930141448975,
"epoch": 0.08937520882058135,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999996249384425,
"loss": 6.2109,
"mean_token_accuracy": 0.12984167486429216,
"num_tokens": 1988422.0,
"step": 1070
},
{
"entropy": 6.30324649810791,
"epoch": 0.08979284998329436,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999995686122657,
"loss": 6.1043,
"mean_token_accuracy": 0.13544947057962417,
"num_tokens": 1998125.0,
"step": 1075
},
{
"entropy": 6.3886956691741945,
"epoch": 0.09021049114600735,
"grad_norm": 1.171875,
"learning_rate": 0.0004999995083472006,
"loss": 6.0843,
"mean_token_accuracy": 0.13154266104102136,
"num_tokens": 2006993.0,
"step": 1080
},
{
"entropy": 6.239620876312256,
"epoch": 0.09062813230872034,
"grad_norm": 1.015625,
"learning_rate": 0.0004999994441432481,
"loss": 6.1494,
"mean_token_accuracy": 0.1386062189936638,
"num_tokens": 2016901.0,
"step": 1085
},
{
"entropy": 6.429760503768921,
"epoch": 0.09104577347143335,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999993760004097,
"loss": 6.1803,
"mean_token_accuracy": 0.1347517654299736,
"num_tokens": 2026397.0,
"step": 1090
},
{
"entropy": 6.428375577926635,
"epoch": 0.09146341463414634,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999993039186864,
"loss": 6.143,
"mean_token_accuracy": 0.12609727829694747,
"num_tokens": 2035241.0,
"step": 1095
},
{
"entropy": 6.35336651802063,
"epoch": 0.09188105579685933,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999992278980794,
"loss": 6.1031,
"mean_token_accuracy": 0.13159844800829887,
"num_tokens": 2044559.0,
"step": 1100
},
{
"entropy": 6.362043714523315,
"epoch": 0.09229869695957234,
"grad_norm": 1.0,
"learning_rate": 0.0004999991479385901,
"loss": 6.1496,
"mean_token_accuracy": 0.1316295139491558,
"num_tokens": 2055368.0,
"step": 1105
},
{
"entropy": 6.2911652565002445,
"epoch": 0.09271633812228533,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999990640402198,
"loss": 6.0811,
"mean_token_accuracy": 0.13263901248574256,
"num_tokens": 2064405.0,
"step": 1110
},
{
"entropy": 6.407223510742187,
"epoch": 0.09313397928499834,
"grad_norm": 1.25,
"learning_rate": 0.0004999989762029701,
"loss": 6.1391,
"mean_token_accuracy": 0.13383666202425956,
"num_tokens": 2073341.0,
"step": 1115
},
{
"entropy": 6.365416622161865,
"epoch": 0.09355162044771133,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999988844268425,
"loss": 6.2715,
"mean_token_accuracy": 0.1288045458495617,
"num_tokens": 2081701.0,
"step": 1120
},
{
"entropy": 6.401083040237427,
"epoch": 0.09396926161042432,
"grad_norm": 1.09375,
"learning_rate": 0.0004999987887118387,
"loss": 6.0827,
"mean_token_accuracy": 0.13317954689264297,
"num_tokens": 2090662.0,
"step": 1125
},
{
"entropy": 6.307847595214843,
"epoch": 0.09438690277313733,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999986890579602,
"loss": 6.1208,
"mean_token_accuracy": 0.12922341972589493,
"num_tokens": 2100351.0,
"step": 1130
},
{
"entropy": 6.389047288894654,
"epoch": 0.09480454393585032,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999985854652087,
"loss": 6.1506,
"mean_token_accuracy": 0.13551958799362182,
"num_tokens": 2108776.0,
"step": 1135
},
{
"entropy": 6.26150426864624,
"epoch": 0.09522218509856331,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999984779335862,
"loss": 6.1954,
"mean_token_accuracy": 0.14278710186481475,
"num_tokens": 2117235.0,
"step": 1140
},
{
"entropy": 6.395204067230225,
"epoch": 0.09563982626127632,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999983664630945,
"loss": 6.0463,
"mean_token_accuracy": 0.13623337671160698,
"num_tokens": 2126482.0,
"step": 1145
},
{
"entropy": 6.3989664077758786,
"epoch": 0.09605746742398931,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999982510537357,
"loss": 6.1735,
"mean_token_accuracy": 0.13455943241715432,
"num_tokens": 2135949.0,
"step": 1150
},
{
"entropy": 6.277780437469483,
"epoch": 0.0964751085867023,
"grad_norm": 1.03125,
"learning_rate": 0.0004999981317055116,
"loss": 6.1089,
"mean_token_accuracy": 0.13122595921158792,
"num_tokens": 2144887.0,
"step": 1155
},
{
"entropy": 6.3384411334991455,
"epoch": 0.09689274974941531,
"grad_norm": 1.125,
"learning_rate": 0.0004999980084184244,
"loss": 6.0851,
"mean_token_accuracy": 0.1300405539572239,
"num_tokens": 2153159.0,
"step": 1160
},
{
"entropy": 6.365826511383057,
"epoch": 0.0973103909121283,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999978811924762,
"loss": 6.1517,
"mean_token_accuracy": 0.1322521060705185,
"num_tokens": 2162408.0,
"step": 1165
},
{
"entropy": 6.431914615631103,
"epoch": 0.09772803207484129,
"grad_norm": 1.0625,
"learning_rate": 0.0004999977500276693,
"loss": 6.0947,
"mean_token_accuracy": 0.13279680758714676,
"num_tokens": 2171407.0,
"step": 1170
},
{
"entropy": 6.156321907043457,
"epoch": 0.0981456732375543,
"grad_norm": 1.25,
"learning_rate": 0.0004999976149240058,
"loss": 5.9727,
"mean_token_accuracy": 0.1389105349779129,
"num_tokens": 2180281.0,
"step": 1175
},
{
"entropy": 6.37581090927124,
"epoch": 0.09856331440026729,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999974758814884,
"loss": 6.1001,
"mean_token_accuracy": 0.13524875566363334,
"num_tokens": 2188901.0,
"step": 1180
},
{
"entropy": 6.332090520858765,
"epoch": 0.09898095556298028,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999973329001193,
"loss": 6.2314,
"mean_token_accuracy": 0.12800901457667352,
"num_tokens": 2199284.0,
"step": 1185
},
{
"entropy": 6.2976422786712645,
"epoch": 0.09939859672569329,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999971859799009,
"loss": 6.0547,
"mean_token_accuracy": 0.13661887347698212,
"num_tokens": 2208013.0,
"step": 1190
},
{
"entropy": 6.394473075866699,
"epoch": 0.09981623788840628,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999970351208361,
"loss": 6.0996,
"mean_token_accuracy": 0.13562223240733146,
"num_tokens": 2217444.0,
"step": 1195
},
{
"entropy": 6.087979173660278,
"epoch": 0.10023387905111927,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999968803229275,
"loss": 6.0029,
"mean_token_accuracy": 0.1400437444448471,
"num_tokens": 2226596.0,
"step": 1200
},
{
"entropy": 6.317812299728393,
"epoch": 0.10065152021383228,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999967215861775,
"loss": 6.0634,
"mean_token_accuracy": 0.12716382518410682,
"num_tokens": 2235441.0,
"step": 1205
},
{
"entropy": 6.3419664859771725,
"epoch": 0.10106916137654527,
"grad_norm": 1.046875,
"learning_rate": 0.000499996558910589,
"loss": 6.1267,
"mean_token_accuracy": 0.12739901766180992,
"num_tokens": 2244783.0,
"step": 1210
},
{
"entropy": 6.319285345077515,
"epoch": 0.10148680253925826,
"grad_norm": 0.99609375,
"learning_rate": 0.000499996392296165,
"loss": 6.0018,
"mean_token_accuracy": 0.13457274362444877,
"num_tokens": 2254484.0,
"step": 1215
},
{
"entropy": 6.156066513061523,
"epoch": 0.10190444370197127,
"grad_norm": 1.234375,
"learning_rate": 0.0004999962217429083,
"loss": 6.0126,
"mean_token_accuracy": 0.14332070723176002,
"num_tokens": 2263394.0,
"step": 1220
},
{
"entropy": 6.1780284404754635,
"epoch": 0.10232208486468426,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999960472508219,
"loss": 6.001,
"mean_token_accuracy": 0.13802936673164368,
"num_tokens": 2272826.0,
"step": 1225
},
{
"entropy": 6.242110347747802,
"epoch": 0.10273972602739725,
"grad_norm": 1.0390625,
"learning_rate": 0.000499995868819909,
"loss": 6.0882,
"mean_token_accuracy": 0.14013530537486077,
"num_tokens": 2282169.0,
"step": 1230
},
{
"entropy": 6.300645256042481,
"epoch": 0.10315736719011026,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999956864501725,
"loss": 6.063,
"mean_token_accuracy": 0.13503486216068267,
"num_tokens": 2291432.0,
"step": 1235
},
{
"entropy": 6.316603946685791,
"epoch": 0.10357500835282325,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999955001416156,
"loss": 6.0485,
"mean_token_accuracy": 0.1410436287522316,
"num_tokens": 2300576.0,
"step": 1240
},
{
"entropy": 6.175787353515625,
"epoch": 0.10399264951553625,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999953098942417,
"loss": 6.1101,
"mean_token_accuracy": 0.13297224342823027,
"num_tokens": 2310790.0,
"step": 1245
},
{
"entropy": 6.3335785388946535,
"epoch": 0.10441029067824925,
"grad_norm": 1.046875,
"learning_rate": 0.0004999951157080541,
"loss": 6.0415,
"mean_token_accuracy": 0.1348338879644871,
"num_tokens": 2319831.0,
"step": 1250
},
{
"entropy": 6.1294896602630615,
"epoch": 0.10482793184096224,
"grad_norm": 1.09375,
"learning_rate": 0.0004999949175830561,
"loss": 5.8914,
"mean_token_accuracy": 0.14019689410924913,
"num_tokens": 2329474.0,
"step": 1255
},
{
"entropy": 6.3479657649993895,
"epoch": 0.10524557300367524,
"grad_norm": 1.09375,
"learning_rate": 0.0004999947155192514,
"loss": 6.1823,
"mean_token_accuracy": 0.1338234931230545,
"num_tokens": 2338179.0,
"step": 1260
},
{
"entropy": 6.144778203964234,
"epoch": 0.10566321416638824,
"grad_norm": 1.109375,
"learning_rate": 0.0004999945095166433,
"loss": 5.9713,
"mean_token_accuracy": 0.13504180908203126,
"num_tokens": 2346899.0,
"step": 1265
},
{
"entropy": 6.26592984199524,
"epoch": 0.10608085532910123,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999942995752354,
"loss": 6.118,
"mean_token_accuracy": 0.12911219522356987,
"num_tokens": 2356077.0,
"step": 1270
},
{
"entropy": 6.347704839706421,
"epoch": 0.10649849649181423,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999940856950315,
"loss": 6.0588,
"mean_token_accuracy": 0.1411923922598362,
"num_tokens": 2365224.0,
"step": 1275
},
{
"entropy": 6.158240222930909,
"epoch": 0.10691613765452723,
"grad_norm": 1.03125,
"learning_rate": 0.0004999938678760354,
"loss": 5.9556,
"mean_token_accuracy": 0.1429154209792614,
"num_tokens": 2375187.0,
"step": 1280
},
{
"entropy": 6.348350143432617,
"epoch": 0.10733377881724022,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999936461182507,
"loss": 6.1621,
"mean_token_accuracy": 0.13689570799469947,
"num_tokens": 2383460.0,
"step": 1285
},
{
"entropy": 6.20832200050354,
"epoch": 0.10775141997995323,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999934204216814,
"loss": 6.0613,
"mean_token_accuracy": 0.13686455488204957,
"num_tokens": 2392701.0,
"step": 1290
},
{
"entropy": 6.254377174377441,
"epoch": 0.10816906114266622,
"grad_norm": 1.03125,
"learning_rate": 0.0004999931907863315,
"loss": 5.9995,
"mean_token_accuracy": 0.13979897350072862,
"num_tokens": 2401516.0,
"step": 1295
},
{
"entropy": 6.236419677734375,
"epoch": 0.10858670230537922,
"grad_norm": 1.1328125,
"learning_rate": 0.000499992957212205,
"loss": 5.98,
"mean_token_accuracy": 0.14022000283002853,
"num_tokens": 2410799.0,
"step": 1300
},
{
"entropy": 6.272280979156494,
"epoch": 0.10900434346809222,
"grad_norm": 1.140625,
"learning_rate": 0.0004999927196993059,
"loss": 5.947,
"mean_token_accuracy": 0.14832693412899972,
"num_tokens": 2419431.0,
"step": 1305
},
{
"entropy": 6.114010667800903,
"epoch": 0.10942198463080521,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999924782476384,
"loss": 5.9768,
"mean_token_accuracy": 0.14567074179649353,
"num_tokens": 2427848.0,
"step": 1310
},
{
"entropy": 6.1843818664550785,
"epoch": 0.1098396257935182,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999922328572068,
"loss": 5.9391,
"mean_token_accuracy": 0.14341116473078727,
"num_tokens": 2437960.0,
"step": 1315
},
{
"entropy": 6.290063714981079,
"epoch": 0.11025726695623121,
"grad_norm": 1.15625,
"learning_rate": 0.0004999919835280154,
"loss": 6.063,
"mean_token_accuracy": 0.14577179551124572,
"num_tokens": 2445835.0,
"step": 1320
},
{
"entropy": 6.189918756484985,
"epoch": 0.1106749081189442,
"grad_norm": 1.09375,
"learning_rate": 0.0004999917302600684,
"loss": 5.956,
"mean_token_accuracy": 0.13776196390390397,
"num_tokens": 2454556.0,
"step": 1325
},
{
"entropy": 6.167909431457519,
"epoch": 0.1110925492816572,
"grad_norm": 0.96875,
"learning_rate": 0.0004999914730533704,
"loss": 6.0471,
"mean_token_accuracy": 0.1329791449010372,
"num_tokens": 2465301.0,
"step": 1330
},
{
"entropy": 6.278769207000733,
"epoch": 0.1115101904443702,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999912119079258,
"loss": 6.0267,
"mean_token_accuracy": 0.13931921273469924,
"num_tokens": 2474085.0,
"step": 1335
},
{
"entropy": 6.241355133056641,
"epoch": 0.1119278316070832,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999909468237393,
"loss": 5.9682,
"mean_token_accuracy": 0.14178281426429748,
"num_tokens": 2482439.0,
"step": 1340
},
{
"entropy": 6.11001501083374,
"epoch": 0.11234547276979619,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999906778008153,
"loss": 6.0129,
"mean_token_accuracy": 0.14232075661420823,
"num_tokens": 2491877.0,
"step": 1345
},
{
"entropy": 6.363374376296997,
"epoch": 0.11276311393250919,
"grad_norm": 0.890625,
"learning_rate": 0.0004999904048391588,
"loss": 6.0974,
"mean_token_accuracy": 0.14235304296016693,
"num_tokens": 2501669.0,
"step": 1350
},
{
"entropy": 6.059836149215698,
"epoch": 0.11318075509522219,
"grad_norm": 0.96875,
"learning_rate": 0.0004999901279387743,
"loss": 5.9852,
"mean_token_accuracy": 0.13875525668263436,
"num_tokens": 2511612.0,
"step": 1355
},
{
"entropy": 6.278777408599853,
"epoch": 0.11359839625793518,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999898470996669,
"loss": 5.9083,
"mean_token_accuracy": 0.14018528833985328,
"num_tokens": 2520791.0,
"step": 1360
},
{
"entropy": 6.085299110412597,
"epoch": 0.11401603742064818,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999895623218415,
"loss": 5.9904,
"mean_token_accuracy": 0.1424546256661415,
"num_tokens": 2529834.0,
"step": 1365
},
{
"entropy": 6.2312211990356445,
"epoch": 0.11443367858336118,
"grad_norm": 1.0625,
"learning_rate": 0.0004999892736053027,
"loss": 5.9862,
"mean_token_accuracy": 0.14232583120465278,
"num_tokens": 2539053.0,
"step": 1370
},
{
"entropy": 6.0757904052734375,
"epoch": 0.11485131974607417,
"grad_norm": 1.046875,
"learning_rate": 0.0004999889809500561,
"loss": 6.012,
"mean_token_accuracy": 0.14125642478466033,
"num_tokens": 2549475.0,
"step": 1375
},
{
"entropy": 6.2677689552307125,
"epoch": 0.11526896090878717,
"grad_norm": 1.046875,
"learning_rate": 0.0004999886843561065,
"loss": 6.0769,
"mean_token_accuracy": 0.14093926027417183,
"num_tokens": 2558718.0,
"step": 1380
},
{
"entropy": 6.167285919189453,
"epoch": 0.11568660207150017,
"grad_norm": 1.15625,
"learning_rate": 0.0004999883838234592,
"loss": 5.9037,
"mean_token_accuracy": 0.14511679708957673,
"num_tokens": 2567454.0,
"step": 1385
},
{
"entropy": 6.093648147583008,
"epoch": 0.11610424323421316,
"grad_norm": 1.15625,
"learning_rate": 0.0004999880793521194,
"loss": 5.82,
"mean_token_accuracy": 0.14054204970598222,
"num_tokens": 2575989.0,
"step": 1390
},
{
"entropy": 6.19893708229065,
"epoch": 0.11652188439692616,
"grad_norm": 1.140625,
"learning_rate": 0.0004999877709420924,
"loss": 6.0154,
"mean_token_accuracy": 0.1437713012099266,
"num_tokens": 2585240.0,
"step": 1395
},
{
"entropy": 6.164418172836304,
"epoch": 0.11693952555963916,
"grad_norm": 1.15625,
"learning_rate": 0.0004999874585933837,
"loss": 6.0533,
"mean_token_accuracy": 0.13939456194639205,
"num_tokens": 2593286.0,
"step": 1400
},
{
"entropy": 6.257520866394043,
"epoch": 0.11735716672235215,
"grad_norm": 0.984375,
"learning_rate": 0.0004999871423059987,
"loss": 6.0063,
"mean_token_accuracy": 0.14244570657610894,
"num_tokens": 2602992.0,
"step": 1405
},
{
"entropy": 6.109090948104859,
"epoch": 0.11777480788506516,
"grad_norm": 1.1015625,
"learning_rate": 0.000499986822079943,
"loss": 5.9937,
"mean_token_accuracy": 0.1422979399561882,
"num_tokens": 2611469.0,
"step": 1410
},
{
"entropy": 6.140593338012695,
"epoch": 0.11819244904777815,
"grad_norm": 1.109375,
"learning_rate": 0.0004999864979152221,
"loss": 5.8956,
"mean_token_accuracy": 0.1449924021959305,
"num_tokens": 2621456.0,
"step": 1415
},
{
"entropy": 6.058706188201905,
"epoch": 0.11861009021049114,
"grad_norm": 1.03125,
"learning_rate": 0.0004999861698118417,
"loss": 5.893,
"mean_token_accuracy": 0.13982913196086882,
"num_tokens": 2630355.0,
"step": 1420
},
{
"entropy": 6.172034072875976,
"epoch": 0.11902773137320415,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999858377698076,
"loss": 5.9123,
"mean_token_accuracy": 0.14697495326399804,
"num_tokens": 2639508.0,
"step": 1425
},
{
"entropy": 6.0646039009094235,
"epoch": 0.11944537253591714,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999855017891256,
"loss": 5.8397,
"mean_token_accuracy": 0.15053340196609497,
"num_tokens": 2648056.0,
"step": 1430
},
{
"entropy": 6.182350111007691,
"epoch": 0.11986301369863013,
"grad_norm": 1.015625,
"learning_rate": 0.0004999851618698017,
"loss": 6.0378,
"mean_token_accuracy": 0.13799886107444764,
"num_tokens": 2657731.0,
"step": 1435
},
{
"entropy": 6.104360628128052,
"epoch": 0.12028065486134314,
"grad_norm": 1.1875,
"learning_rate": 0.0004999848180118416,
"loss": 5.878,
"mean_token_accuracy": 0.14412474036216735,
"num_tokens": 2666424.0,
"step": 1440
},
{
"entropy": 5.992441177368164,
"epoch": 0.12069829602405613,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999844702152516,
"loss": 5.793,
"mean_token_accuracy": 0.1518073007464409,
"num_tokens": 2674646.0,
"step": 1445
},
{
"entropy": 6.237054443359375,
"epoch": 0.12111593718676913,
"grad_norm": 1.078125,
"learning_rate": 0.0004999841184800374,
"loss": 5.9863,
"mean_token_accuracy": 0.14161835610866547,
"num_tokens": 2683295.0,
"step": 1450
},
{
"entropy": 6.116791200637818,
"epoch": 0.12153357834948213,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999837628062055,
"loss": 5.8928,
"mean_token_accuracy": 0.14603603705763818,
"num_tokens": 2692220.0,
"step": 1455
},
{
"entropy": 6.111183929443359,
"epoch": 0.12195121951219512,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999834031937621,
"loss": 6.0896,
"mean_token_accuracy": 0.13455943539738655,
"num_tokens": 2702095.0,
"step": 1460
},
{
"entropy": 6.3252228736877445,
"epoch": 0.12236886067490813,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999830396427133,
"loss": 6.0157,
"mean_token_accuracy": 0.1376993291079998,
"num_tokens": 2711440.0,
"step": 1465
},
{
"entropy": 6.169790554046631,
"epoch": 0.12278650183762112,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999826721530657,
"loss": 6.001,
"mean_token_accuracy": 0.14647723585367203,
"num_tokens": 2720564.0,
"step": 1470
},
{
"entropy": 6.193677091598511,
"epoch": 0.12320414300033411,
"grad_norm": 1.09375,
"learning_rate": 0.0004999823007248255,
"loss": 5.8517,
"mean_token_accuracy": 0.13836075663566588,
"num_tokens": 2729442.0,
"step": 1475
},
{
"entropy": 6.122766494750977,
"epoch": 0.12362178416304712,
"grad_norm": 1.109375,
"learning_rate": 0.0004999819253579992,
"loss": 5.9647,
"mean_token_accuracy": 0.1364038832485676,
"num_tokens": 2738683.0,
"step": 1480
},
{
"entropy": 6.1442817687988285,
"epoch": 0.12403942532576011,
"grad_norm": 1.09375,
"learning_rate": 0.0004999815460525938,
"loss": 5.9333,
"mean_token_accuracy": 0.14273701012134551,
"num_tokens": 2747888.0,
"step": 1485
},
{
"entropy": 6.067212295532227,
"epoch": 0.1244570664884731,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999811628086155,
"loss": 5.7995,
"mean_token_accuracy": 0.14842113852500916,
"num_tokens": 2756667.0,
"step": 1490
},
{
"entropy": 6.137423133850097,
"epoch": 0.1248747076511861,
"grad_norm": 1.1484375,
"learning_rate": 0.000499980775626071,
"loss": 5.978,
"mean_token_accuracy": 0.14387312531471252,
"num_tokens": 2765116.0,
"step": 1495
},
{
"entropy": 5.987923860549927,
"epoch": 0.12529234881389908,
"grad_norm": 1.0,
"learning_rate": 0.0004999803845049674,
"loss": 5.8348,
"mean_token_accuracy": 0.148624786734581,
"num_tokens": 2774259.0,
"step": 1500
},
{
"entropy": 6.0781763076782225,
"epoch": 0.1257099899766121,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999799894453112,
"loss": 5.869,
"mean_token_accuracy": 0.14260915145277978,
"num_tokens": 2782988.0,
"step": 1505
},
{
"entropy": 6.074441432952881,
"epoch": 0.1261276311393251,
"grad_norm": 1.09375,
"learning_rate": 0.0004999795904471095,
"loss": 5.8764,
"mean_token_accuracy": 0.1461164064705372,
"num_tokens": 2791409.0,
"step": 1510
},
{
"entropy": 6.199710178375244,
"epoch": 0.12654527230203808,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999791875103694,
"loss": 5.9489,
"mean_token_accuracy": 0.143809075653553,
"num_tokens": 2800644.0,
"step": 1515
},
{
"entropy": 6.061881637573242,
"epoch": 0.12696291346475108,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999787806350977,
"loss": 5.8425,
"mean_token_accuracy": 0.14784278273582457,
"num_tokens": 2809838.0,
"step": 1520
},
{
"entropy": 6.122160530090332,
"epoch": 0.1273805546274641,
"grad_norm": 1.015625,
"learning_rate": 0.0004999783698213016,
"loss": 5.9717,
"mean_token_accuracy": 0.1433075912296772,
"num_tokens": 2818972.0,
"step": 1525
},
{
"entropy": 6.13506383895874,
"epoch": 0.12779819579017707,
"grad_norm": 1.109375,
"learning_rate": 0.0004999779550689885,
"loss": 5.8962,
"mean_token_accuracy": 0.1441321402788162,
"num_tokens": 2827733.0,
"step": 1530
},
{
"entropy": 6.1400655746459964,
"epoch": 0.12821583695289007,
"grad_norm": 1.046875,
"learning_rate": 0.0004999775363781654,
"loss": 5.9921,
"mean_token_accuracy": 0.14137880951166154,
"num_tokens": 2837300.0,
"step": 1535
},
{
"entropy": 6.183238124847412,
"epoch": 0.12863347811560308,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999771137488395,
"loss": 5.8972,
"mean_token_accuracy": 0.14587430506944657,
"num_tokens": 2846076.0,
"step": 1540
},
{
"entropy": 5.954459524154663,
"epoch": 0.12905111927831608,
"grad_norm": 1.0625,
"learning_rate": 0.0004999766871810188,
"loss": 5.8199,
"mean_token_accuracy": 0.14452776312828064,
"num_tokens": 2855523.0,
"step": 1545
},
{
"entropy": 6.148645496368408,
"epoch": 0.12946876044102906,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999762566747101,
"loss": 5.9546,
"mean_token_accuracy": 0.14385624825954438,
"num_tokens": 2864277.0,
"step": 1550
},
{
"entropy": 6.116501188278198,
"epoch": 0.12988640160374207,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999758222299214,
"loss": 5.9374,
"mean_token_accuracy": 0.14204028993844986,
"num_tokens": 2873561.0,
"step": 1555
},
{
"entropy": 6.12092695236206,
"epoch": 0.13030404276645507,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999753838466601,
"loss": 5.8148,
"mean_token_accuracy": 0.1522620737552643,
"num_tokens": 2882449.0,
"step": 1560
},
{
"entropy": 6.0730757236480715,
"epoch": 0.13072168392916805,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999749415249338,
"loss": 5.7752,
"mean_token_accuracy": 0.15969083160161973,
"num_tokens": 2890862.0,
"step": 1565
},
{
"entropy": 5.995173978805542,
"epoch": 0.13113932509188106,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999744952647505,
"loss": 5.8679,
"mean_token_accuracy": 0.15128796994686128,
"num_tokens": 2899611.0,
"step": 1570
},
{
"entropy": 6.126881837844849,
"epoch": 0.13155696625459407,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999740450661176,
"loss": 5.917,
"mean_token_accuracy": 0.14792831093072892,
"num_tokens": 2908111.0,
"step": 1575
},
{
"entropy": 6.092712545394898,
"epoch": 0.13197460741730704,
"grad_norm": 1.015625,
"learning_rate": 0.0004999735909290436,
"loss": 5.8916,
"mean_token_accuracy": 0.1501033529639244,
"num_tokens": 2917194.0,
"step": 1580
},
{
"entropy": 6.04348578453064,
"epoch": 0.13239224858002005,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999731328535358,
"loss": 5.8566,
"mean_token_accuracy": 0.14753102511167526,
"num_tokens": 2926684.0,
"step": 1585
},
{
"entropy": 5.9717113971710205,
"epoch": 0.13280988974273306,
"grad_norm": 1.03125,
"learning_rate": 0.0004999726708396026,
"loss": 5.8318,
"mean_token_accuracy": 0.14780840948224067,
"num_tokens": 2936094.0,
"step": 1590
},
{
"entropy": 6.083980226516724,
"epoch": 0.13322753090544603,
"grad_norm": 1.078125,
"learning_rate": 0.000499972204887252,
"loss": 5.8373,
"mean_token_accuracy": 0.14772946015000343,
"num_tokens": 2944389.0,
"step": 1595
},
{
"entropy": 6.005502843856812,
"epoch": 0.13364517206815904,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999717349964922,
"loss": 5.8579,
"mean_token_accuracy": 0.14418584555387498,
"num_tokens": 2954466.0,
"step": 1600
},
{
"entropy": 6.135638904571533,
"epoch": 0.13406281323087205,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999712611673313,
"loss": 6.0455,
"mean_token_accuracy": 0.13582754135131836,
"num_tokens": 2965403.0,
"step": 1605
},
{
"entropy": 6.272896146774292,
"epoch": 0.13448045439358502,
"grad_norm": 1.0,
"learning_rate": 0.0004999707833997777,
"loss": 6.044,
"mean_token_accuracy": 0.14214518070220947,
"num_tokens": 2975896.0,
"step": 1610
},
{
"entropy": 5.984560871124268,
"epoch": 0.13489809555629803,
"grad_norm": 1.109375,
"learning_rate": 0.0004999703016938398,
"loss": 5.8884,
"mean_token_accuracy": 0.15345285460352898,
"num_tokens": 2984937.0,
"step": 1615
},
{
"entropy": 5.952976226806641,
"epoch": 0.13531573671901104,
"grad_norm": 1.03125,
"learning_rate": 0.000499969816049526,
"loss": 5.8512,
"mean_token_accuracy": 0.14834006428718566,
"num_tokens": 2994052.0,
"step": 1620
},
{
"entropy": 6.1258875846862795,
"epoch": 0.13573337788172402,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999693264668446,
"loss": 5.7352,
"mean_token_accuracy": 0.1505567364394665,
"num_tokens": 3003274.0,
"step": 1625
},
{
"entropy": 6.0712028503417965,
"epoch": 0.13615101904443702,
"grad_norm": 1.0625,
"learning_rate": 0.0004999688329458045,
"loss": 5.84,
"mean_token_accuracy": 0.15118275731801986,
"num_tokens": 3011440.0,
"step": 1630
},
{
"entropy": 6.008664512634278,
"epoch": 0.13656866020715003,
"grad_norm": 1.015625,
"learning_rate": 0.0004999683354864141,
"loss": 5.9329,
"mean_token_accuracy": 0.14488905742764474,
"num_tokens": 3021032.0,
"step": 1635
},
{
"entropy": 6.072646713256836,
"epoch": 0.136986301369863,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999678340886823,
"loss": 5.9002,
"mean_token_accuracy": 0.1485132485628128,
"num_tokens": 3030160.0,
"step": 1640
},
{
"entropy": 6.0518828392028805,
"epoch": 0.137403942532576,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999673287526176,
"loss": 5.9321,
"mean_token_accuracy": 0.1511251837015152,
"num_tokens": 3039098.0,
"step": 1645
},
{
"entropy": 6.145869588851928,
"epoch": 0.13782158369528902,
"grad_norm": 1.015625,
"learning_rate": 0.0004999668194782291,
"loss": 5.8652,
"mean_token_accuracy": 0.14529599547386168,
"num_tokens": 3048679.0,
"step": 1650
},
{
"entropy": 6.002886247634888,
"epoch": 0.138239224858002,
"grad_norm": 1.171875,
"learning_rate": 0.0004999663062655257,
"loss": 5.7778,
"mean_token_accuracy": 0.15449400395154952,
"num_tokens": 3057613.0,
"step": 1655
},
{
"entropy": 6.0733765125274655,
"epoch": 0.138656866020715,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999657891145162,
"loss": 5.9725,
"mean_token_accuracy": 0.14817949309945105,
"num_tokens": 3066291.0,
"step": 1660
},
{
"entropy": 6.028089237213135,
"epoch": 0.139074507183428,
"grad_norm": 1.09375,
"learning_rate": 0.0004999652680252099,
"loss": 5.902,
"mean_token_accuracy": 0.14239186346530913,
"num_tokens": 3075897.0,
"step": 1665
},
{
"entropy": 6.171479940414429,
"epoch": 0.139492148346141,
"grad_norm": 1.0625,
"learning_rate": 0.0004999647429976157,
"loss": 5.8825,
"mean_token_accuracy": 0.14334002286195754,
"num_tokens": 3085193.0,
"step": 1670
},
{
"entropy": 6.060310125350952,
"epoch": 0.139909789508854,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999642140317429,
"loss": 5.8657,
"mean_token_accuracy": 0.14457208216190337,
"num_tokens": 3094335.0,
"step": 1675
},
{
"entropy": 6.0583384990692135,
"epoch": 0.140327430671567,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999636811276007,
"loss": 5.9671,
"mean_token_accuracy": 0.14252406731247902,
"num_tokens": 3103964.0,
"step": 1680
},
{
"entropy": 6.042476081848145,
"epoch": 0.14074507183427998,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999631442851986,
"loss": 5.8487,
"mean_token_accuracy": 0.14683347940444946,
"num_tokens": 3113321.0,
"step": 1685
},
{
"entropy": 6.0762725353240965,
"epoch": 0.14116271299699298,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999626035045458,
"loss": 5.8548,
"mean_token_accuracy": 0.14716964587569237,
"num_tokens": 3122467.0,
"step": 1690
},
{
"entropy": 6.074911594390869,
"epoch": 0.141580354159706,
"grad_norm": 1.015625,
"learning_rate": 0.0004999620587856519,
"loss": 5.8106,
"mean_token_accuracy": 0.1543986439704895,
"num_tokens": 3131239.0,
"step": 1695
},
{
"entropy": 5.974051237106323,
"epoch": 0.14199799532241897,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999615101285262,
"loss": 5.9463,
"mean_token_accuracy": 0.14293202459812165,
"num_tokens": 3140568.0,
"step": 1700
},
{
"entropy": 6.088217878341675,
"epoch": 0.14241563648513197,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999609575331786,
"loss": 5.8839,
"mean_token_accuracy": 0.14690518230199814,
"num_tokens": 3151227.0,
"step": 1705
},
{
"entropy": 6.073013973236084,
"epoch": 0.14283327764784498,
"grad_norm": 1.0625,
"learning_rate": 0.0004999604009996186,
"loss": 5.7842,
"mean_token_accuracy": 0.1575349062681198,
"num_tokens": 3159865.0,
"step": 1710
},
{
"entropy": 5.97011547088623,
"epoch": 0.14325091881055796,
"grad_norm": 0.96875,
"learning_rate": 0.000499959840527856,
"loss": 5.7554,
"mean_token_accuracy": 0.1552349552512169,
"num_tokens": 3169832.0,
"step": 1715
},
{
"entropy": 5.988092088699341,
"epoch": 0.14366855997327097,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999592761179007,
"loss": 5.9937,
"mean_token_accuracy": 0.14358080923557281,
"num_tokens": 3180357.0,
"step": 1720
},
{
"entropy": 6.048899412155151,
"epoch": 0.14408620113598397,
"grad_norm": 1.03125,
"learning_rate": 0.0004999587077697624,
"loss": 5.8213,
"mean_token_accuracy": 0.15338835269212722,
"num_tokens": 3189663.0,
"step": 1725
},
{
"entropy": 6.034728240966797,
"epoch": 0.14450384229869695,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999581354834512,
"loss": 5.9016,
"mean_token_accuracy": 0.1513874851167202,
"num_tokens": 3199578.0,
"step": 1730
},
{
"entropy": 6.077489566802979,
"epoch": 0.14492148346140996,
"grad_norm": 1.09375,
"learning_rate": 0.000499957559258977,
"loss": 5.8504,
"mean_token_accuracy": 0.15805275589227677,
"num_tokens": 3208839.0,
"step": 1735
},
{
"entropy": 6.081794166564942,
"epoch": 0.14533912462412296,
"grad_norm": 1.125,
"learning_rate": 0.00049995697909635,
"loss": 5.8594,
"mean_token_accuracy": 0.15180393382906915,
"num_tokens": 3217449.0,
"step": 1740
},
{
"entropy": 5.9614564895629885,
"epoch": 0.14575676578683594,
"grad_norm": 1.125,
"learning_rate": 0.0004999563949955803,
"loss": 5.7557,
"mean_token_accuracy": 0.1466016449034214,
"num_tokens": 3226561.0,
"step": 1745
},
{
"entropy": 6.04267988204956,
"epoch": 0.14617440694954895,
"grad_norm": 1.0,
"learning_rate": 0.000499955806956678,
"loss": 5.8522,
"mean_token_accuracy": 0.15752413123846054,
"num_tokens": 3235400.0,
"step": 1750
},
{
"entropy": 6.0401026725769045,
"epoch": 0.14659204811226195,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999552149796536,
"loss": 5.9144,
"mean_token_accuracy": 0.14410170838236808,
"num_tokens": 3244862.0,
"step": 1755
},
{
"entropy": 5.979519271850586,
"epoch": 0.14700968927497493,
"grad_norm": 0.984375,
"learning_rate": 0.0004999546190645175,
"loss": 5.8129,
"mean_token_accuracy": 0.1537217117846012,
"num_tokens": 3254370.0,
"step": 1760
},
{
"entropy": 6.0241940975189205,
"epoch": 0.14742733043768794,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999540192112799,
"loss": 5.8457,
"mean_token_accuracy": 0.15348097831010818,
"num_tokens": 3264079.0,
"step": 1765
},
{
"entropy": 5.9918759822845455,
"epoch": 0.14784497160040094,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999534154199514,
"loss": 5.8969,
"mean_token_accuracy": 0.15254206955432892,
"num_tokens": 3273315.0,
"step": 1770
},
{
"entropy": 5.959523487091064,
"epoch": 0.14826261276311392,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999528076905426,
"loss": 5.7799,
"mean_token_accuracy": 0.1521797373890877,
"num_tokens": 3282045.0,
"step": 1775
},
{
"entropy": 6.010910081863403,
"epoch": 0.14868025392582693,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999521960230641,
"loss": 5.8839,
"mean_token_accuracy": 0.1504328355193138,
"num_tokens": 3291596.0,
"step": 1780
},
{
"entropy": 5.929302835464478,
"epoch": 0.14909789508853993,
"grad_norm": 1.0625,
"learning_rate": 0.0004999515804175268,
"loss": 5.7192,
"mean_token_accuracy": 0.15593315809965133,
"num_tokens": 3300407.0,
"step": 1785
},
{
"entropy": 6.025093269348145,
"epoch": 0.1495155362512529,
"grad_norm": 1.0390625,
"learning_rate": 0.000499950960873941,
"loss": 5.7526,
"mean_token_accuracy": 0.15117358267307282,
"num_tokens": 3309848.0,
"step": 1790
},
{
"entropy": 5.93266453742981,
"epoch": 0.14993317741396592,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999503373923182,
"loss": 5.8559,
"mean_token_accuracy": 0.1462218925356865,
"num_tokens": 3319031.0,
"step": 1795
},
{
"entropy": 5.971647119522094,
"epoch": 0.15035081857667892,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999497099726687,
"loss": 5.7953,
"mean_token_accuracy": 0.1453866645693779,
"num_tokens": 3327918.0,
"step": 1800
},
{
"entropy": 5.944482469558716,
"epoch": 0.1507684597393919,
"grad_norm": 1.046875,
"learning_rate": 0.0004999490786150039,
"loss": 5.8234,
"mean_token_accuracy": 0.14495486468076707,
"num_tokens": 3337071.0,
"step": 1805
},
{
"entropy": 6.112966394424438,
"epoch": 0.1511861009021049,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999484433193348,
"loss": 5.8881,
"mean_token_accuracy": 0.1472978837788105,
"num_tokens": 3346334.0,
"step": 1810
},
{
"entropy": 6.038641357421875,
"epoch": 0.15160374206481791,
"grad_norm": 1.0625,
"learning_rate": 0.0004999478040856722,
"loss": 5.8042,
"mean_token_accuracy": 0.14679064601659775,
"num_tokens": 3354932.0,
"step": 1815
},
{
"entropy": 5.838101720809936,
"epoch": 0.1520213832275309,
"grad_norm": 1.015625,
"learning_rate": 0.0004999471609140276,
"loss": 5.6583,
"mean_token_accuracy": 0.151202093064785,
"num_tokens": 3364345.0,
"step": 1820
},
{
"entropy": 6.046137475967408,
"epoch": 0.1524390243902439,
"grad_norm": 1.125,
"learning_rate": 0.0004999465138044122,
"loss": 5.837,
"mean_token_accuracy": 0.14820541590452194,
"num_tokens": 3372918.0,
"step": 1825
},
{
"entropy": 5.906940031051636,
"epoch": 0.1528566655529569,
"grad_norm": 1.140625,
"learning_rate": 0.0004999458627568373,
"loss": 5.71,
"mean_token_accuracy": 0.15286548137664796,
"num_tokens": 3381815.0,
"step": 1830
},
{
"entropy": 5.9043457984924315,
"epoch": 0.15327430671566988,
"grad_norm": 1.03125,
"learning_rate": 0.0004999452077713144,
"loss": 5.7599,
"mean_token_accuracy": 0.1523844078183174,
"num_tokens": 3391057.0,
"step": 1835
},
{
"entropy": 6.079875421524048,
"epoch": 0.1536919478783829,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999445488478547,
"loss": 5.9567,
"mean_token_accuracy": 0.14450442418456078,
"num_tokens": 3401254.0,
"step": 1840
},
{
"entropy": 5.981129789352417,
"epoch": 0.1541095890410959,
"grad_norm": 0.9921875,
"learning_rate": 0.00049994388598647,
"loss": 5.8614,
"mean_token_accuracy": 0.14819915741682052,
"num_tokens": 3411031.0,
"step": 1845
},
{
"entropy": 5.982936429977417,
"epoch": 0.15452723020380887,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999432191871719,
"loss": 5.7968,
"mean_token_accuracy": 0.15687229484319687,
"num_tokens": 3420154.0,
"step": 1850
},
{
"entropy": 6.114473152160644,
"epoch": 0.15494487136652188,
"grad_norm": 1.203125,
"learning_rate": 0.0004999425484499718,
"loss": 5.8862,
"mean_token_accuracy": 0.15098146200180054,
"num_tokens": 3429285.0,
"step": 1855
},
{
"entropy": 5.987294626235962,
"epoch": 0.1553625125292349,
"grad_norm": 1.03125,
"learning_rate": 0.0004999418737748817,
"loss": 5.8558,
"mean_token_accuracy": 0.14990614205598832,
"num_tokens": 3438065.0,
"step": 1860
},
{
"entropy": 6.067724609375,
"epoch": 0.15578015369194786,
"grad_norm": 1.046875,
"learning_rate": 0.0004999411951619134,
"loss": 5.8615,
"mean_token_accuracy": 0.14882780537009238,
"num_tokens": 3447534.0,
"step": 1865
},
{
"entropy": 6.009346771240234,
"epoch": 0.15619779485466087,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999405126110786,
"loss": 5.8194,
"mean_token_accuracy": 0.15343883335590364,
"num_tokens": 3456435.0,
"step": 1870
},
{
"entropy": 5.931367492675781,
"epoch": 0.15661543601737388,
"grad_norm": 1.0,
"learning_rate": 0.0004999398261223895,
"loss": 5.7965,
"mean_token_accuracy": 0.154514442384243,
"num_tokens": 3464982.0,
"step": 1875
},
{
"entropy": 6.091391706466675,
"epoch": 0.15703307718008688,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999391356958579,
"loss": 5.9261,
"mean_token_accuracy": 0.14851265177130699,
"num_tokens": 3474585.0,
"step": 1880
},
{
"entropy": 5.937274599075318,
"epoch": 0.15745071834279986,
"grad_norm": 1.078125,
"learning_rate": 0.000499938441331496,
"loss": 5.7652,
"mean_token_accuracy": 0.15298966467380523,
"num_tokens": 3483164.0,
"step": 1885
},
{
"entropy": 5.942174434661865,
"epoch": 0.15786835950551287,
"grad_norm": 1.046875,
"learning_rate": 0.0004999377430293159,
"loss": 5.7712,
"mean_token_accuracy": 0.14635431468486787,
"num_tokens": 3492984.0,
"step": 1890
},
{
"entropy": 5.909766054153442,
"epoch": 0.15828600066822587,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999370407893299,
"loss": 5.6107,
"mean_token_accuracy": 0.16215786188840867,
"num_tokens": 3500356.0,
"step": 1895
},
{
"entropy": 6.018685245513916,
"epoch": 0.15870364183093885,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999363346115501,
"loss": 5.8125,
"mean_token_accuracy": 0.15466036051511764,
"num_tokens": 3509792.0,
"step": 1900
},
{
"entropy": 5.9935344696044925,
"epoch": 0.15912128299365186,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999356244959893,
"loss": 5.7378,
"mean_token_accuracy": 0.149660687148571,
"num_tokens": 3519340.0,
"step": 1905
},
{
"entropy": 5.8559294700622555,
"epoch": 0.15953892415636486,
"grad_norm": 1.078125,
"learning_rate": 0.0004999349104426594,
"loss": 5.7346,
"mean_token_accuracy": 0.15643487572669984,
"num_tokens": 3528846.0,
"step": 1910
},
{
"entropy": 5.969501399993897,
"epoch": 0.15995656531907784,
"grad_norm": 1.09375,
"learning_rate": 0.0004999341924515732,
"loss": 5.723,
"mean_token_accuracy": 0.15309956967830657,
"num_tokens": 3537766.0,
"step": 1915
},
{
"entropy": 5.955421113967896,
"epoch": 0.16037420648179085,
"grad_norm": 1.0625,
"learning_rate": 0.0004999334705227431,
"loss": 5.796,
"mean_token_accuracy": 0.1499815657734871,
"num_tokens": 3546724.0,
"step": 1920
},
{
"entropy": 5.901809024810791,
"epoch": 0.16079184764450385,
"grad_norm": 1.0546875,
"learning_rate": 0.000499932744656182,
"loss": 5.6755,
"mean_token_accuracy": 0.16025683730840684,
"num_tokens": 3555599.0,
"step": 1925
},
{
"entropy": 5.8585106372833256,
"epoch": 0.16120948880721683,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999320148519024,
"loss": 5.7395,
"mean_token_accuracy": 0.15616288557648658,
"num_tokens": 3564754.0,
"step": 1930
},
{
"entropy": 5.821812677383423,
"epoch": 0.16162712996992984,
"grad_norm": 1.015625,
"learning_rate": 0.0004999312811099171,
"loss": 5.597,
"mean_token_accuracy": 0.16239829510450363,
"num_tokens": 3573166.0,
"step": 1935
},
{
"entropy": 5.954546213150024,
"epoch": 0.16204477113264285,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999305434302391,
"loss": 5.8643,
"mean_token_accuracy": 0.1504362255334854,
"num_tokens": 3581602.0,
"step": 1940
},
{
"entropy": 5.977283334732055,
"epoch": 0.16246241229535582,
"grad_norm": 0.9921875,
"learning_rate": 0.000499929801812881,
"loss": 5.6727,
"mean_token_accuracy": 0.16076785027980806,
"num_tokens": 3590946.0,
"step": 1945
},
{
"entropy": 5.840662717819214,
"epoch": 0.16288005345806883,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999290562578561,
"loss": 5.7435,
"mean_token_accuracy": 0.15856219083070755,
"num_tokens": 3599888.0,
"step": 1950
},
{
"entropy": 5.941868877410888,
"epoch": 0.16329769462078184,
"grad_norm": 1.015625,
"learning_rate": 0.0004999283067651773,
"loss": 5.7591,
"mean_token_accuracy": 0.16239314377307892,
"num_tokens": 3609773.0,
"step": 1955
},
{
"entropy": 5.869628143310547,
"epoch": 0.16371533578349481,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999275533348577,
"loss": 5.7349,
"mean_token_accuracy": 0.1552025258541107,
"num_tokens": 3618820.0,
"step": 1960
},
{
"entropy": 5.990446376800537,
"epoch": 0.16413297694620782,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999267959669106,
"loss": 5.8281,
"mean_token_accuracy": 0.1533038690686226,
"num_tokens": 3629010.0,
"step": 1965
},
{
"entropy": 5.895765590667724,
"epoch": 0.16455061810892083,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999260346613491,
"loss": 5.747,
"mean_token_accuracy": 0.15518127232789994,
"num_tokens": 3636915.0,
"step": 1970
},
{
"entropy": 5.916001415252685,
"epoch": 0.1649682592716338,
"grad_norm": 1.0625,
"learning_rate": 0.0004999252694181867,
"loss": 5.7295,
"mean_token_accuracy": 0.1585697278380394,
"num_tokens": 3646011.0,
"step": 1975
},
{
"entropy": 5.943479585647583,
"epoch": 0.1653859004343468,
"grad_norm": 1.0,
"learning_rate": 0.0004999245002374367,
"loss": 5.7033,
"mean_token_accuracy": 0.16055430471897125,
"num_tokens": 3655059.0,
"step": 1980
},
{
"entropy": 5.9830348014831545,
"epoch": 0.16580354159705982,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999237271191125,
"loss": 5.7979,
"mean_token_accuracy": 0.15375116169452668,
"num_tokens": 3664553.0,
"step": 1985
},
{
"entropy": 5.9315876960754395,
"epoch": 0.1662211827597728,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999229500632279,
"loss": 5.738,
"mean_token_accuracy": 0.1533743694424629,
"num_tokens": 3673604.0,
"step": 1990
},
{
"entropy": 5.930443811416626,
"epoch": 0.1666388239224858,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999221690697961,
"loss": 5.7126,
"mean_token_accuracy": 0.1625876694917679,
"num_tokens": 3682634.0,
"step": 1995
},
{
"entropy": 5.919408082962036,
"epoch": 0.1670564650851988,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999213841388312,
"loss": 5.7982,
"mean_token_accuracy": 0.15079522579908372,
"num_tokens": 3691961.0,
"step": 2000
},
{
"entropy": 5.88884482383728,
"epoch": 0.16747410624791179,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999205952703466,
"loss": 5.7503,
"mean_token_accuracy": 0.1490720458328724,
"num_tokens": 3701412.0,
"step": 2005
},
{
"entropy": 5.956575727462768,
"epoch": 0.1678917474106248,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999198024643563,
"loss": 5.7988,
"mean_token_accuracy": 0.15698140263557434,
"num_tokens": 3710449.0,
"step": 2010
},
{
"entropy": 5.961668586730957,
"epoch": 0.1683093885733378,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999190057208741,
"loss": 5.7591,
"mean_token_accuracy": 0.14847119450569152,
"num_tokens": 3718654.0,
"step": 2015
},
{
"entropy": 5.898903131484985,
"epoch": 0.16872702973605078,
"grad_norm": 0.99609375,
"learning_rate": 0.000499918205039914,
"loss": 5.7487,
"mean_token_accuracy": 0.15195512697100638,
"num_tokens": 3728660.0,
"step": 2020
},
{
"entropy": 5.887713956832886,
"epoch": 0.16914467089876378,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999174004214899,
"loss": 5.7612,
"mean_token_accuracy": 0.1582832634449005,
"num_tokens": 3738085.0,
"step": 2025
},
{
"entropy": 5.943373489379883,
"epoch": 0.1695623120614768,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999165918656161,
"loss": 5.698,
"mean_token_accuracy": 0.15785950943827629,
"num_tokens": 3747727.0,
"step": 2030
},
{
"entropy": 5.963870048522949,
"epoch": 0.16997995322418977,
"grad_norm": 1.046875,
"learning_rate": 0.0004999157793723065,
"loss": 5.8168,
"mean_token_accuracy": 0.15320221334695816,
"num_tokens": 3757864.0,
"step": 2035
},
{
"entropy": 5.827042865753174,
"epoch": 0.17039759438690277,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999149629415755,
"loss": 5.6983,
"mean_token_accuracy": 0.16073038354516028,
"num_tokens": 3766843.0,
"step": 2040
},
{
"entropy": 5.94908447265625,
"epoch": 0.17081523554961578,
"grad_norm": 1.03125,
"learning_rate": 0.0004999141425734374,
"loss": 5.7476,
"mean_token_accuracy": 0.1535307839512825,
"num_tokens": 3776171.0,
"step": 2045
},
{
"entropy": 5.944053077697754,
"epoch": 0.17123287671232876,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999133182679064,
"loss": 5.7253,
"mean_token_accuracy": 0.16170434057712554,
"num_tokens": 3785545.0,
"step": 2050
},
{
"entropy": 5.784623289108277,
"epoch": 0.17165051787504176,
"grad_norm": 1.046875,
"learning_rate": 0.0004999124900249971,
"loss": 5.6729,
"mean_token_accuracy": 0.15819989293813705,
"num_tokens": 3794110.0,
"step": 2055
},
{
"entropy": 5.945992040634155,
"epoch": 0.17206815903775477,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999116578447239,
"loss": 5.8108,
"mean_token_accuracy": 0.15511984676122664,
"num_tokens": 3803727.0,
"step": 2060
},
{
"entropy": 5.9064085483551025,
"epoch": 0.17248580020046775,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999108217271015,
"loss": 5.7648,
"mean_token_accuracy": 0.1535521849989891,
"num_tokens": 3812904.0,
"step": 2065
},
{
"entropy": 6.095454597473145,
"epoch": 0.17290344136318075,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999099816721444,
"loss": 5.851,
"mean_token_accuracy": 0.15227640569210052,
"num_tokens": 3822009.0,
"step": 2070
},
{
"entropy": 5.841128206253051,
"epoch": 0.17332108252589376,
"grad_norm": 1.046875,
"learning_rate": 0.0004999091376798673,
"loss": 5.7599,
"mean_token_accuracy": 0.15635585486888887,
"num_tokens": 3830877.0,
"step": 2075
},
{
"entropy": 6.060049676895142,
"epoch": 0.17373872368860674,
"grad_norm": 1.0625,
"learning_rate": 0.0004999082897502849,
"loss": 5.8234,
"mean_token_accuracy": 0.1512978546321392,
"num_tokens": 3839876.0,
"step": 2080
},
{
"entropy": 6.048352336883545,
"epoch": 0.17415636485131974,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999074378834123,
"loss": 5.8174,
"mean_token_accuracy": 0.16038154661655427,
"num_tokens": 3848769.0,
"step": 2085
},
{
"entropy": 5.854233169555664,
"epoch": 0.17457400601403275,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999065820792644,
"loss": 5.631,
"mean_token_accuracy": 0.15877358615398407,
"num_tokens": 3857676.0,
"step": 2090
},
{
"entropy": 5.918405389785766,
"epoch": 0.17499164717674573,
"grad_norm": 1.109375,
"learning_rate": 0.0004999057223378559,
"loss": 5.7287,
"mean_token_accuracy": 0.15598385334014891,
"num_tokens": 3867098.0,
"step": 2095
},
{
"entropy": 5.797727441787719,
"epoch": 0.17540928833945874,
"grad_norm": 1.0078125,
"learning_rate": 0.000499904858659202,
"loss": 5.6714,
"mean_token_accuracy": 0.1513415813446045,
"num_tokens": 3876247.0,
"step": 2100
},
{
"entropy": 5.859252500534057,
"epoch": 0.17582692950217174,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999039910433179,
"loss": 5.7278,
"mean_token_accuracy": 0.16455016881227494,
"num_tokens": 3885747.0,
"step": 2105
},
{
"entropy": 6.052870464324951,
"epoch": 0.17624457066488472,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999031194902187,
"loss": 5.8473,
"mean_token_accuracy": 0.15237005054950714,
"num_tokens": 3896077.0,
"step": 2110
},
{
"entropy": 5.972765064239502,
"epoch": 0.17666221182759773,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999022439999198,
"loss": 5.8471,
"mean_token_accuracy": 0.14567215442657472,
"num_tokens": 3906323.0,
"step": 2115
},
{
"entropy": 5.924546098709106,
"epoch": 0.17707985299031073,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999013645724362,
"loss": 5.744,
"mean_token_accuracy": 0.15510283634066582,
"num_tokens": 3915510.0,
"step": 2120
},
{
"entropy": 5.93877215385437,
"epoch": 0.1774974941530237,
"grad_norm": 1.03125,
"learning_rate": 0.0004999004812077836,
"loss": 5.8143,
"mean_token_accuracy": 0.1571462519466877,
"num_tokens": 3924799.0,
"step": 2125
},
{
"entropy": 5.834187316894531,
"epoch": 0.17791513531573672,
"grad_norm": 1.1875,
"learning_rate": 0.0004998995939059773,
"loss": 5.7565,
"mean_token_accuracy": 0.15311417132616043,
"num_tokens": 3933675.0,
"step": 2130
},
{
"entropy": 5.885807180404663,
"epoch": 0.17833277647844972,
"grad_norm": 1.0234375,
"learning_rate": 0.000499898702667033,
"loss": 5.6939,
"mean_token_accuracy": 0.15966519266366958,
"num_tokens": 3942086.0,
"step": 2135
},
{
"entropy": 5.9234106063842775,
"epoch": 0.1787504176411627,
"grad_norm": 1.171875,
"learning_rate": 0.0004998978074909661,
"loss": 5.7162,
"mean_token_accuracy": 0.15536481142044067,
"num_tokens": 3950626.0,
"step": 2140
},
{
"entropy": 5.733146715164184,
"epoch": 0.1791680588038757,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998969083777925,
"loss": 5.6178,
"mean_token_accuracy": 0.17237166315317154,
"num_tokens": 3959006.0,
"step": 2145
},
{
"entropy": 5.853768444061279,
"epoch": 0.1795856999665887,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998960053275277,
"loss": 5.7,
"mean_token_accuracy": 0.16328830420970916,
"num_tokens": 3968177.0,
"step": 2150
},
{
"entropy": 5.902877569198608,
"epoch": 0.1800033411293017,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998950983401875,
"loss": 5.6651,
"mean_token_accuracy": 0.15961136817932128,
"num_tokens": 3977052.0,
"step": 2155
},
{
"entropy": 5.8690708637237545,
"epoch": 0.1804209822920147,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998941874157882,
"loss": 5.6858,
"mean_token_accuracy": 0.15438530892133712,
"num_tokens": 3985101.0,
"step": 2160
},
{
"entropy": 5.850992012023926,
"epoch": 0.1808386234547277,
"grad_norm": 1.078125,
"learning_rate": 0.0004998932725543453,
"loss": 5.6098,
"mean_token_accuracy": 0.15906111001968384,
"num_tokens": 3992936.0,
"step": 2165
},
{
"entropy": 5.916439151763916,
"epoch": 0.18125626461744068,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998923537558749,
"loss": 5.7801,
"mean_token_accuracy": 0.1568411409854889,
"num_tokens": 4001702.0,
"step": 2170
},
{
"entropy": 5.8718656539917,
"epoch": 0.1816739057801537,
"grad_norm": 1.078125,
"learning_rate": 0.0004998914310203932,
"loss": 5.7721,
"mean_token_accuracy": 0.157361763715744,
"num_tokens": 4011639.0,
"step": 2175
},
{
"entropy": 5.805581378936767,
"epoch": 0.1820915469428667,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998905043479163,
"loss": 5.5848,
"mean_token_accuracy": 0.16170865297317505,
"num_tokens": 4021420.0,
"step": 2180
},
{
"entropy": 5.7755042552948,
"epoch": 0.18250918810557967,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998895737384605,
"loss": 5.6565,
"mean_token_accuracy": 0.15489371567964555,
"num_tokens": 4030869.0,
"step": 2185
},
{
"entropy": 5.772803878784179,
"epoch": 0.18292682926829268,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998886391920418,
"loss": 5.5327,
"mean_token_accuracy": 0.1654495880007744,
"num_tokens": 4039573.0,
"step": 2190
},
{
"entropy": 5.7726646900177006,
"epoch": 0.18334447043100568,
"grad_norm": 1.0,
"learning_rate": 0.0004998877007086769,
"loss": 5.6365,
"mean_token_accuracy": 0.16419821083545685,
"num_tokens": 4048684.0,
"step": 2195
},
{
"entropy": 5.77266206741333,
"epoch": 0.18376211159371866,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998867582883821,
"loss": 5.7919,
"mean_token_accuracy": 0.152945114672184,
"num_tokens": 4057968.0,
"step": 2200
},
{
"entropy": 5.963563871383667,
"epoch": 0.18417975275643167,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998858119311739,
"loss": 5.6945,
"mean_token_accuracy": 0.15852373093366623,
"num_tokens": 4068057.0,
"step": 2205
},
{
"entropy": 5.830620002746582,
"epoch": 0.18459739391914468,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998848616370689,
"loss": 5.6105,
"mean_token_accuracy": 0.16643038839101792,
"num_tokens": 4077141.0,
"step": 2210
},
{
"entropy": 5.784256935119629,
"epoch": 0.18501503508185768,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998839074060837,
"loss": 5.6668,
"mean_token_accuracy": 0.15795958489179612,
"num_tokens": 4085227.0,
"step": 2215
},
{
"entropy": 5.864757537841797,
"epoch": 0.18543267624457066,
"grad_norm": 1.09375,
"learning_rate": 0.000499882949238235,
"loss": 5.6295,
"mean_token_accuracy": 0.1627066969871521,
"num_tokens": 4093674.0,
"step": 2220
},
{
"entropy": 5.904892301559448,
"epoch": 0.18585031740728367,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998819871335396,
"loss": 5.7977,
"mean_token_accuracy": 0.15320228040218353,
"num_tokens": 4103466.0,
"step": 2225
},
{
"entropy": 5.9529468536376955,
"epoch": 0.18626795856999667,
"grad_norm": 1.09375,
"learning_rate": 0.0004998810210920142,
"loss": 5.6308,
"mean_token_accuracy": 0.15609087347984313,
"num_tokens": 4111833.0,
"step": 2230
},
{
"entropy": 5.831648969650269,
"epoch": 0.18668559973270965,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998800511136759,
"loss": 5.696,
"mean_token_accuracy": 0.15844447761774064,
"num_tokens": 4121546.0,
"step": 2235
},
{
"entropy": 5.88375301361084,
"epoch": 0.18710324089542266,
"grad_norm": 1.015625,
"learning_rate": 0.0004998790771985417,
"loss": 5.6594,
"mean_token_accuracy": 0.16582350432872772,
"num_tokens": 4130408.0,
"step": 2240
},
{
"entropy": 5.788675308227539,
"epoch": 0.18752088205813566,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998780993466284,
"loss": 5.7884,
"mean_token_accuracy": 0.15529951304197312,
"num_tokens": 4139793.0,
"step": 2245
},
{
"entropy": 5.9381426811218265,
"epoch": 0.18793852322084864,
"grad_norm": 1.0625,
"learning_rate": 0.0004998771175579534,
"loss": 5.6301,
"mean_token_accuracy": 0.1638265699148178,
"num_tokens": 4148192.0,
"step": 2250
},
{
"entropy": 5.749383401870728,
"epoch": 0.18835616438356165,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998761318325337,
"loss": 5.583,
"mean_token_accuracy": 0.1587271884083748,
"num_tokens": 4157960.0,
"step": 2255
},
{
"entropy": 5.599839687347412,
"epoch": 0.18877380554627465,
"grad_norm": 1.125,
"learning_rate": 0.0004998751421703868,
"loss": 5.4448,
"mean_token_accuracy": 0.17087434977293015,
"num_tokens": 4166356.0,
"step": 2260
},
{
"entropy": 5.921038627624512,
"epoch": 0.18919144670898763,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998741485715297,
"loss": 5.7834,
"mean_token_accuracy": 0.1493260845541954,
"num_tokens": 4176139.0,
"step": 2265
},
{
"entropy": 5.862062454223633,
"epoch": 0.18960908787170064,
"grad_norm": 1.0859375,
"learning_rate": 0.00049987315103598,
"loss": 5.6537,
"mean_token_accuracy": 0.1571880027651787,
"num_tokens": 4185583.0,
"step": 2270
},
{
"entropy": 5.7853264808654785,
"epoch": 0.19002672903441364,
"grad_norm": 1.109375,
"learning_rate": 0.0004998721495637551,
"loss": 5.7499,
"mean_token_accuracy": 0.15288184583187103,
"num_tokens": 4194275.0,
"step": 2275
},
{
"entropy": 5.86413426399231,
"epoch": 0.19044437019712662,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998711441548726,
"loss": 5.6825,
"mean_token_accuracy": 0.16003127843141557,
"num_tokens": 4203424.0,
"step": 2280
},
{
"entropy": 5.735157155990601,
"epoch": 0.19086201135983963,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998701348093499,
"loss": 5.6521,
"mean_token_accuracy": 0.16366421282291413,
"num_tokens": 4212452.0,
"step": 2285
},
{
"entropy": 5.897738647460938,
"epoch": 0.19127965252255263,
"grad_norm": 0.96875,
"learning_rate": 0.000499869121527205,
"loss": 5.7611,
"mean_token_accuracy": 0.15444075167179108,
"num_tokens": 4221639.0,
"step": 2290
},
{
"entropy": 5.865130138397217,
"epoch": 0.1916972936852656,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998681043084553,
"loss": 5.6302,
"mean_token_accuracy": 0.1672831192612648,
"num_tokens": 4231362.0,
"step": 2295
},
{
"entropy": 5.815743017196655,
"epoch": 0.19211493484797862,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998670831531188,
"loss": 5.7266,
"mean_token_accuracy": 0.15458433628082274,
"num_tokens": 4241035.0,
"step": 2300
},
{
"entropy": 5.94687819480896,
"epoch": 0.19253257601069163,
"grad_norm": 1.109375,
"learning_rate": 0.0004998660580612135,
"loss": 5.6458,
"mean_token_accuracy": 0.1604814574122429,
"num_tokens": 4250275.0,
"step": 2305
},
{
"entropy": 5.761688947677612,
"epoch": 0.1929502171734046,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998650290327569,
"loss": 5.7459,
"mean_token_accuracy": 0.15605543106794356,
"num_tokens": 4259952.0,
"step": 2310
},
{
"entropy": 5.860372161865234,
"epoch": 0.1933678583361176,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998639960677675,
"loss": 5.58,
"mean_token_accuracy": 0.16590958833694458,
"num_tokens": 4269985.0,
"step": 2315
},
{
"entropy": 5.8047374248504635,
"epoch": 0.19378549949883062,
"grad_norm": 1.046875,
"learning_rate": 0.0004998629591662631,
"loss": 5.5889,
"mean_token_accuracy": 0.16162675321102143,
"num_tokens": 4278317.0,
"step": 2320
},
{
"entropy": 5.785900402069092,
"epoch": 0.1942031406615436,
"grad_norm": 1.046875,
"learning_rate": 0.000499861918328262,
"loss": 5.6248,
"mean_token_accuracy": 0.16352908909320832,
"num_tokens": 4286633.0,
"step": 2325
},
{
"entropy": 5.809439373016358,
"epoch": 0.1946207818242566,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998608735537822,
"loss": 5.6043,
"mean_token_accuracy": 0.16824746429920195,
"num_tokens": 4295696.0,
"step": 2330
},
{
"entropy": 5.761776733398437,
"epoch": 0.1950384229869696,
"grad_norm": 0.984375,
"learning_rate": 0.0004998598248428423,
"loss": 5.6942,
"mean_token_accuracy": 0.16295794695615767,
"num_tokens": 4305139.0,
"step": 2335
},
{
"entropy": 5.801323413848877,
"epoch": 0.19545606414968258,
"grad_norm": 1.0,
"learning_rate": 0.0004998587721954604,
"loss": 5.5569,
"mean_token_accuracy": 0.16570183932781218,
"num_tokens": 4314029.0,
"step": 2340
},
{
"entropy": 5.803794860839844,
"epoch": 0.1958737053123956,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998577156116551,
"loss": 5.7027,
"mean_token_accuracy": 0.15905870646238326,
"num_tokens": 4323667.0,
"step": 2345
},
{
"entropy": 5.898095417022705,
"epoch": 0.1962913464751086,
"grad_norm": 1.046875,
"learning_rate": 0.0004998566550914448,
"loss": 5.7514,
"mean_token_accuracy": 0.1614445775747299,
"num_tokens": 4333456.0,
"step": 2350
},
{
"entropy": 5.882632923126221,
"epoch": 0.19670898763782158,
"grad_norm": 1.09375,
"learning_rate": 0.0004998555906348481,
"loss": 5.7133,
"mean_token_accuracy": 0.15647562146186828,
"num_tokens": 4342527.0,
"step": 2355
},
{
"entropy": 5.8123023986816404,
"epoch": 0.19712662880053458,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998545222418836,
"loss": 5.6222,
"mean_token_accuracy": 0.15761913359165192,
"num_tokens": 4351294.0,
"step": 2360
},
{
"entropy": 5.791871070861816,
"epoch": 0.1975442699632476,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998534499125701,
"loss": 5.6326,
"mean_token_accuracy": 0.1602482318878174,
"num_tokens": 4361596.0,
"step": 2365
},
{
"entropy": 5.866342163085937,
"epoch": 0.19796191112596057,
"grad_norm": 1.0,
"learning_rate": 0.0004998523736469263,
"loss": 5.6464,
"mean_token_accuracy": 0.16265444308519364,
"num_tokens": 4370917.0,
"step": 2370
},
{
"entropy": 5.719123411178589,
"epoch": 0.19837955228867357,
"grad_norm": 1.0,
"learning_rate": 0.000499851293444971,
"loss": 5.6783,
"mean_token_accuracy": 0.1592626839876175,
"num_tokens": 4380244.0,
"step": 2375
},
{
"entropy": 5.787483501434326,
"epoch": 0.19879719345138658,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998502093067231,
"loss": 5.6553,
"mean_token_accuracy": 0.15903472751379014,
"num_tokens": 4389852.0,
"step": 2380
},
{
"entropy": 5.896140050888062,
"epoch": 0.19921483461409956,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998491212322017,
"loss": 5.7254,
"mean_token_accuracy": 0.1548086531460285,
"num_tokens": 4399743.0,
"step": 2385
},
{
"entropy": 5.879378080368042,
"epoch": 0.19963247577681256,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998480292214258,
"loss": 5.6432,
"mean_token_accuracy": 0.15781110972166063,
"num_tokens": 4408661.0,
"step": 2390
},
{
"entropy": 5.790725517272949,
"epoch": 0.20005011693952557,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998469332744144,
"loss": 5.6009,
"mean_token_accuracy": 0.16551696062088012,
"num_tokens": 4417671.0,
"step": 2395
},
{
"entropy": 5.77877402305603,
"epoch": 0.20046775810223855,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998458333911869,
"loss": 5.6896,
"mean_token_accuracy": 0.1598905384540558,
"num_tokens": 4426850.0,
"step": 2400
},
{
"entropy": 5.955265283584595,
"epoch": 0.20088539926495155,
"grad_norm": 1.125,
"learning_rate": 0.0004998447295717624,
"loss": 5.7931,
"mean_token_accuracy": 0.15605985820293428,
"num_tokens": 4435769.0,
"step": 2405
},
{
"entropy": 5.819901657104492,
"epoch": 0.20130304042766456,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998436218161602,
"loss": 5.7201,
"mean_token_accuracy": 0.16049996465444566,
"num_tokens": 4445521.0,
"step": 2410
},
{
"entropy": 5.86675992012024,
"epoch": 0.20172068159037754,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998425101243999,
"loss": 5.7061,
"mean_token_accuracy": 0.15795376002788544,
"num_tokens": 4454674.0,
"step": 2415
},
{
"entropy": 5.996614456176758,
"epoch": 0.20213832275309054,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998413944965007,
"loss": 5.7236,
"mean_token_accuracy": 0.1618449531495571,
"num_tokens": 4464035.0,
"step": 2420
},
{
"entropy": 5.7577736377716064,
"epoch": 0.20255596391580355,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998402749324823,
"loss": 5.5729,
"mean_token_accuracy": 0.163533054292202,
"num_tokens": 4472055.0,
"step": 2425
},
{
"entropy": 5.785602712631226,
"epoch": 0.20297360507851653,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998391514323642,
"loss": 5.6749,
"mean_token_accuracy": 0.1660549208521843,
"num_tokens": 4480669.0,
"step": 2430
},
{
"entropy": 5.895810174942016,
"epoch": 0.20339124624122953,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998380239961661,
"loss": 5.6533,
"mean_token_accuracy": 0.1660724997520447,
"num_tokens": 4489481.0,
"step": 2435
},
{
"entropy": 5.792815732955932,
"epoch": 0.20380888740394254,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998368926239078,
"loss": 5.555,
"mean_token_accuracy": 0.1635495975613594,
"num_tokens": 4497508.0,
"step": 2440
},
{
"entropy": 5.710590028762818,
"epoch": 0.20422652856665552,
"grad_norm": 0.92578125,
"learning_rate": 0.000499835757315609,
"loss": 5.6292,
"mean_token_accuracy": 0.16281557828187943,
"num_tokens": 4506373.0,
"step": 2445
},
{
"entropy": 5.8337420463562015,
"epoch": 0.20464416972936852,
"grad_norm": 1.03125,
"learning_rate": 0.0004998346180712898,
"loss": 5.6954,
"mean_token_accuracy": 0.1564403623342514,
"num_tokens": 4516283.0,
"step": 2450
},
{
"entropy": 5.770689964294434,
"epoch": 0.20506181089208153,
"grad_norm": 1.125,
"learning_rate": 0.0004998334748909698,
"loss": 5.5398,
"mean_token_accuracy": 0.1744811698794365,
"num_tokens": 4524276.0,
"step": 2455
},
{
"entropy": 5.7850175380706785,
"epoch": 0.2054794520547945,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998323277746694,
"loss": 5.6454,
"mean_token_accuracy": 0.15986775159835814,
"num_tokens": 4532951.0,
"step": 2460
},
{
"entropy": 5.787815380096435,
"epoch": 0.20589709321750752,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998311767224081,
"loss": 5.624,
"mean_token_accuracy": 0.16870736479759216,
"num_tokens": 4542229.0,
"step": 2465
},
{
"entropy": 5.749476480484009,
"epoch": 0.20631473438022052,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998300217342067,
"loss": 5.5336,
"mean_token_accuracy": 0.16806766539812087,
"num_tokens": 4551126.0,
"step": 2470
},
{
"entropy": 5.787161350250244,
"epoch": 0.2067323755429335,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998288628100851,
"loss": 5.6474,
"mean_token_accuracy": 0.1591607302427292,
"num_tokens": 4560628.0,
"step": 2475
},
{
"entropy": 5.809577608108521,
"epoch": 0.2071500167056465,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998276999500636,
"loss": 5.6891,
"mean_token_accuracy": 0.16426692605018617,
"num_tokens": 4570184.0,
"step": 2480
},
{
"entropy": 5.848023366928101,
"epoch": 0.2075676578683595,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998265331541627,
"loss": 5.5526,
"mean_token_accuracy": 0.16781152188777923,
"num_tokens": 4580357.0,
"step": 2485
},
{
"entropy": 5.827300310134888,
"epoch": 0.2079852990310725,
"grad_norm": 1.0,
"learning_rate": 0.0004998253624224026,
"loss": 5.7448,
"mean_token_accuracy": 0.1515749141573906,
"num_tokens": 4589965.0,
"step": 2490
},
{
"entropy": 5.801667642593384,
"epoch": 0.2084029401937855,
"grad_norm": 1.0,
"learning_rate": 0.000499824187754804,
"loss": 5.5621,
"mean_token_accuracy": 0.1672935664653778,
"num_tokens": 4598753.0,
"step": 2495
},
{
"entropy": 5.7748055934906,
"epoch": 0.2088205813564985,
"grad_norm": 1.0625,
"learning_rate": 0.0004998230091513872,
"loss": 5.6224,
"mean_token_accuracy": 0.16568318605422974,
"num_tokens": 4608698.0,
"step": 2500
},
{
"entropy": 5.810699701309204,
"epoch": 0.20923822251921148,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998218266121731,
"loss": 5.7281,
"mean_token_accuracy": 0.16196652129292488,
"num_tokens": 4618137.0,
"step": 2505
},
{
"entropy": 5.762399625778198,
"epoch": 0.2096558636819245,
"grad_norm": 1.046875,
"learning_rate": 0.0004998206401371823,
"loss": 5.5543,
"mean_token_accuracy": 0.16459099799394608,
"num_tokens": 4627468.0,
"step": 2510
},
{
"entropy": 5.700483942031861,
"epoch": 0.2100735048446375,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998194497264357,
"loss": 5.465,
"mean_token_accuracy": 0.17145939618349076,
"num_tokens": 4636659.0,
"step": 2515
},
{
"entropy": 5.712591743469238,
"epoch": 0.21049114600735047,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998182553799538,
"loss": 5.5657,
"mean_token_accuracy": 0.1616992771625519,
"num_tokens": 4644969.0,
"step": 2520
},
{
"entropy": 5.8478742122650145,
"epoch": 0.21090878717006348,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998170570977579,
"loss": 5.6626,
"mean_token_accuracy": 0.15977111607789993,
"num_tokens": 4653930.0,
"step": 2525
},
{
"entropy": 5.778926277160645,
"epoch": 0.21132642833277648,
"grad_norm": 1.0625,
"learning_rate": 0.0004998158548798687,
"loss": 5.5438,
"mean_token_accuracy": 0.16683215498924256,
"num_tokens": 4663146.0,
"step": 2530
},
{
"entropy": 5.731309604644776,
"epoch": 0.21174406949548946,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998146487263074,
"loss": 5.6999,
"mean_token_accuracy": 0.15984612554311753,
"num_tokens": 4671874.0,
"step": 2535
},
{
"entropy": 5.6937836647033695,
"epoch": 0.21216171065820247,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998134386370951,
"loss": 5.5375,
"mean_token_accuracy": 0.1655649721622467,
"num_tokens": 4680811.0,
"step": 2540
},
{
"entropy": 5.822449207305908,
"epoch": 0.21257935182091547,
"grad_norm": 1.03125,
"learning_rate": 0.0004998122246122529,
"loss": 5.6168,
"mean_token_accuracy": 0.1640414848923683,
"num_tokens": 4689977.0,
"step": 2545
},
{
"entropy": 5.72489562034607,
"epoch": 0.21299699298362845,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998110066518021,
"loss": 5.4963,
"mean_token_accuracy": 0.1763449266552925,
"num_tokens": 4698479.0,
"step": 2550
},
{
"entropy": 5.640039730072021,
"epoch": 0.21341463414634146,
"grad_norm": 0.92578125,
"learning_rate": 0.000499809784755764,
"loss": 5.5826,
"mean_token_accuracy": 0.161933995783329,
"num_tokens": 4707575.0,
"step": 2555
},
{
"entropy": 5.783615350723267,
"epoch": 0.21383227530905446,
"grad_norm": 1.046875,
"learning_rate": 0.00049980855892416,
"loss": 5.6148,
"mean_token_accuracy": 0.1656864881515503,
"num_tokens": 4717015.0,
"step": 2560
},
{
"entropy": 5.783946561813354,
"epoch": 0.21424991647176747,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998073291570116,
"loss": 5.6292,
"mean_token_accuracy": 0.15882773846387863,
"num_tokens": 4726466.0,
"step": 2565
},
{
"entropy": 5.801740312576294,
"epoch": 0.21466755763448045,
"grad_norm": 1.125,
"learning_rate": 0.0004998060954543404,
"loss": 5.6929,
"mean_token_accuracy": 0.16479280441999436,
"num_tokens": 4736497.0,
"step": 2570
},
{
"entropy": 5.713510656356812,
"epoch": 0.21508519879719346,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998048578161678,
"loss": 5.6212,
"mean_token_accuracy": 0.1654522955417633,
"num_tokens": 4745693.0,
"step": 2575
},
{
"entropy": 5.744898414611816,
"epoch": 0.21550283995990646,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998036162425155,
"loss": 5.5405,
"mean_token_accuracy": 0.1734490066766739,
"num_tokens": 4754274.0,
"step": 2580
},
{
"entropy": 5.825760412216186,
"epoch": 0.21592048112261944,
"grad_norm": 0.921875,
"learning_rate": 0.0004998023707334055,
"loss": 5.76,
"mean_token_accuracy": 0.15564341247081756,
"num_tokens": 4765607.0,
"step": 2585
},
{
"entropy": 5.813382291793824,
"epoch": 0.21633812228533245,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998011212888592,
"loss": 5.5185,
"mean_token_accuracy": 0.16956669241189956,
"num_tokens": 4775291.0,
"step": 2590
},
{
"entropy": 5.656440782546997,
"epoch": 0.21675576344804545,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997998679088988,
"loss": 5.4658,
"mean_token_accuracy": 0.1743773937225342,
"num_tokens": 4783859.0,
"step": 2595
},
{
"entropy": 5.670311164855957,
"epoch": 0.21717340461075843,
"grad_norm": 1.03125,
"learning_rate": 0.0004997986105935461,
"loss": 5.5265,
"mean_token_accuracy": 0.16651082336902617,
"num_tokens": 4792891.0,
"step": 2600
},
{
"entropy": 5.757329988479614,
"epoch": 0.21759104577347144,
"grad_norm": 1.125,
"learning_rate": 0.0004997973493428231,
"loss": 5.606,
"mean_token_accuracy": 0.16225490868091583,
"num_tokens": 4801551.0,
"step": 2605
},
{
"entropy": 5.799970865249634,
"epoch": 0.21800868693618444,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997960841567519,
"loss": 5.5832,
"mean_token_accuracy": 0.16481925845146178,
"num_tokens": 4810430.0,
"step": 2610
},
{
"entropy": 5.712379360198975,
"epoch": 0.21842632809889742,
"grad_norm": 1.0,
"learning_rate": 0.0004997948150353548,
"loss": 5.568,
"mean_token_accuracy": 0.1724429875612259,
"num_tokens": 4819156.0,
"step": 2615
},
{
"entropy": 5.778680372238159,
"epoch": 0.21884396926161043,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997935419786537,
"loss": 5.6167,
"mean_token_accuracy": 0.16232399195432662,
"num_tokens": 4827724.0,
"step": 2620
},
{
"entropy": 5.7654613018035885,
"epoch": 0.21926161042432343,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997922649866712,
"loss": 5.5331,
"mean_token_accuracy": 0.17342215925455093,
"num_tokens": 4836954.0,
"step": 2625
},
{
"entropy": 5.770890951156616,
"epoch": 0.2196792515870364,
"grad_norm": 0.953125,
"learning_rate": 0.0004997909840594294,
"loss": 5.6689,
"mean_token_accuracy": 0.16099561005830765,
"num_tokens": 4847504.0,
"step": 2630
},
{
"entropy": 5.784782838821411,
"epoch": 0.22009689274974942,
"grad_norm": 1.046875,
"learning_rate": 0.0004997896991969509,
"loss": 5.6018,
"mean_token_accuracy": 0.17029968053102493,
"num_tokens": 4856105.0,
"step": 2635
},
{
"entropy": 5.7090918064117435,
"epoch": 0.22051453391246242,
"grad_norm": 1.0,
"learning_rate": 0.0004997884103992581,
"loss": 5.5947,
"mean_token_accuracy": 0.16394108533859253,
"num_tokens": 4866118.0,
"step": 2640
},
{
"entropy": 5.75405969619751,
"epoch": 0.2209321750751754,
"grad_norm": 1.0625,
"learning_rate": 0.0004997871176663736,
"loss": 5.6109,
"mean_token_accuracy": 0.16301430761814117,
"num_tokens": 4875242.0,
"step": 2645
},
{
"entropy": 5.890598392486572,
"epoch": 0.2213498162378884,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997858209983201,
"loss": 5.6817,
"mean_token_accuracy": 0.16772009283304215,
"num_tokens": 4884726.0,
"step": 2650
},
{
"entropy": 5.734208250045777,
"epoch": 0.22176745740060141,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997845203951202,
"loss": 5.6117,
"mean_token_accuracy": 0.16239058673381807,
"num_tokens": 4893178.0,
"step": 2655
},
{
"entropy": 5.790639019012451,
"epoch": 0.2221850985633144,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997832158567967,
"loss": 5.7144,
"mean_token_accuracy": 0.16555942147970198,
"num_tokens": 4903126.0,
"step": 2660
},
{
"entropy": 5.7078839302062985,
"epoch": 0.2226027397260274,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997819073833724,
"loss": 5.4287,
"mean_token_accuracy": 0.17669259011745453,
"num_tokens": 4912069.0,
"step": 2665
},
{
"entropy": 5.6767395496368405,
"epoch": 0.2230203808887404,
"grad_norm": 1.078125,
"learning_rate": 0.0004997805949748702,
"loss": 5.5,
"mean_token_accuracy": 0.16776227653026582,
"num_tokens": 4921014.0,
"step": 2670
},
{
"entropy": 5.594740676879883,
"epoch": 0.22343802205145338,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997792786313132,
"loss": 5.5565,
"mean_token_accuracy": 0.16906140074133874,
"num_tokens": 4930382.0,
"step": 2675
},
{
"entropy": 5.7800782203674315,
"epoch": 0.2238556632141664,
"grad_norm": 1.03125,
"learning_rate": 0.0004997779583527243,
"loss": 5.6108,
"mean_token_accuracy": 0.15652296096086502,
"num_tokens": 4939672.0,
"step": 2680
},
{
"entropy": 5.809805393218994,
"epoch": 0.2242733043768794,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997766341391268,
"loss": 5.6483,
"mean_token_accuracy": 0.15791001468896865,
"num_tokens": 4949186.0,
"step": 2685
},
{
"entropy": 5.787170314788819,
"epoch": 0.22469094553959237,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997753059905437,
"loss": 5.6182,
"mean_token_accuracy": 0.1618640184402466,
"num_tokens": 4958155.0,
"step": 2690
},
{
"entropy": 5.640371799468994,
"epoch": 0.22510858670230538,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997739739069981,
"loss": 5.4923,
"mean_token_accuracy": 0.17097348123788833,
"num_tokens": 4966524.0,
"step": 2695
},
{
"entropy": 5.768008184432984,
"epoch": 0.22552622786501839,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997726378885138,
"loss": 5.5607,
"mean_token_accuracy": 0.16756032705307006,
"num_tokens": 4976066.0,
"step": 2700
},
{
"entropy": 5.716037225723267,
"epoch": 0.22594386902773136,
"grad_norm": 1.09375,
"learning_rate": 0.0004997712979351137,
"loss": 5.4664,
"mean_token_accuracy": 0.17585382759571075,
"num_tokens": 4984957.0,
"step": 2705
},
{
"entropy": 5.6190448760986325,
"epoch": 0.22636151019044437,
"grad_norm": 1.015625,
"learning_rate": 0.0004997699540468216,
"loss": 5.4433,
"mean_token_accuracy": 0.17340092957019806,
"num_tokens": 4994430.0,
"step": 2710
},
{
"entropy": 5.646725225448608,
"epoch": 0.22677915135315738,
"grad_norm": 1.046875,
"learning_rate": 0.0004997686062236609,
"loss": 5.5724,
"mean_token_accuracy": 0.1680037721991539,
"num_tokens": 5003893.0,
"step": 2715
},
{
"entropy": 5.751096105575561,
"epoch": 0.22719679251587035,
"grad_norm": 1.09375,
"learning_rate": 0.0004997672544656553,
"loss": 5.5873,
"mean_token_accuracy": 0.16746485978364944,
"num_tokens": 5014083.0,
"step": 2720
},
{
"entropy": 5.6994177341461185,
"epoch": 0.22761443367858336,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997658987728283,
"loss": 5.4538,
"mean_token_accuracy": 0.1704857677221298,
"num_tokens": 5023025.0,
"step": 2725
},
{
"entropy": 5.653778743743897,
"epoch": 0.22803207484129637,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997645391452035,
"loss": 5.5722,
"mean_token_accuracy": 0.1677939623594284,
"num_tokens": 5032620.0,
"step": 2730
},
{
"entropy": 5.802613639831543,
"epoch": 0.22844971600400935,
"grad_norm": 1.015625,
"learning_rate": 0.000499763175582805,
"loss": 5.5867,
"mean_token_accuracy": 0.16344683468341828,
"num_tokens": 5041773.0,
"step": 2735
},
{
"entropy": 5.8078845024108885,
"epoch": 0.22886735716672235,
"grad_norm": 1.078125,
"learning_rate": 0.0004997618080856566,
"loss": 5.6512,
"mean_token_accuracy": 0.15850055813789368,
"num_tokens": 5051450.0,
"step": 2740
},
{
"entropy": 5.704273319244384,
"epoch": 0.22928499832943536,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997604366537823,
"loss": 5.6019,
"mean_token_accuracy": 0.16153014898300172,
"num_tokens": 5061172.0,
"step": 2745
},
{
"entropy": 5.68063178062439,
"epoch": 0.22970263949214834,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997590612872058,
"loss": 5.5001,
"mean_token_accuracy": 0.17272624373435974,
"num_tokens": 5071161.0,
"step": 2750
},
{
"entropy": 5.735128879547119,
"epoch": 0.23012028065486134,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997576819859515,
"loss": 5.5911,
"mean_token_accuracy": 0.1705945312976837,
"num_tokens": 5080884.0,
"step": 2755
},
{
"entropy": 5.698446464538574,
"epoch": 0.23053792181757435,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997562987500434,
"loss": 5.5478,
"mean_token_accuracy": 0.16551766842603682,
"num_tokens": 5090164.0,
"step": 2760
},
{
"entropy": 5.622687101364136,
"epoch": 0.23095556298028733,
"grad_norm": 1.015625,
"learning_rate": 0.0004997549115795058,
"loss": 5.492,
"mean_token_accuracy": 0.1630670577287674,
"num_tokens": 5098700.0,
"step": 2765
},
{
"entropy": 5.801355028152466,
"epoch": 0.23137320414300033,
"grad_norm": 1.0390625,
"learning_rate": 0.000499753520474363,
"loss": 5.7264,
"mean_token_accuracy": 0.15362178534269333,
"num_tokens": 5107713.0,
"step": 2770
},
{
"entropy": 5.88943395614624,
"epoch": 0.23179084530571334,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997521254346391,
"loss": 5.7254,
"mean_token_accuracy": 0.15206045657396317,
"num_tokens": 5117532.0,
"step": 2775
},
{
"entropy": 5.686231327056885,
"epoch": 0.23220848646842632,
"grad_norm": 0.953125,
"learning_rate": 0.0004997507264603589,
"loss": 5.4773,
"mean_token_accuracy": 0.16964103877544404,
"num_tokens": 5126891.0,
"step": 2780
},
{
"entropy": 5.638346815109253,
"epoch": 0.23262612763113932,
"grad_norm": 0.984375,
"learning_rate": 0.0004997493235515466,
"loss": 5.4362,
"mean_token_accuracy": 0.1693371891975403,
"num_tokens": 5135608.0,
"step": 2785
},
{
"entropy": 5.718230295181274,
"epoch": 0.23304376879385233,
"grad_norm": 0.9375,
"learning_rate": 0.0004997479167082269,
"loss": 5.5464,
"mean_token_accuracy": 0.16581456810235978,
"num_tokens": 5145055.0,
"step": 2790
},
{
"entropy": 5.780206775665283,
"epoch": 0.2334614099565653,
"grad_norm": 1.0625,
"learning_rate": 0.0004997465059304243,
"loss": 5.5928,
"mean_token_accuracy": 0.16677365899086,
"num_tokens": 5154742.0,
"step": 2795
},
{
"entropy": 5.755462265014648,
"epoch": 0.23387905111927831,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997450912181637,
"loss": 5.5588,
"mean_token_accuracy": 0.1658760577440262,
"num_tokens": 5163876.0,
"step": 2800
},
{
"entropy": 5.690133333206177,
"epoch": 0.23429669228199132,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997436725714696,
"loss": 5.5556,
"mean_token_accuracy": 0.16277918666601182,
"num_tokens": 5173614.0,
"step": 2805
},
{
"entropy": 5.723607540130615,
"epoch": 0.2347143334447043,
"grad_norm": 1.046875,
"learning_rate": 0.0004997422499903671,
"loss": 5.5742,
"mean_token_accuracy": 0.17018153071403502,
"num_tokens": 5182061.0,
"step": 2810
},
{
"entropy": 5.859146022796631,
"epoch": 0.2351319746074173,
"grad_norm": 1.03125,
"learning_rate": 0.000499740823474881,
"loss": 5.7519,
"mean_token_accuracy": 0.15502672716975213,
"num_tokens": 5191661.0,
"step": 2815
},
{
"entropy": 5.838399457931518,
"epoch": 0.2355496157701303,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997393930250363,
"loss": 5.668,
"mean_token_accuracy": 0.15951003283262252,
"num_tokens": 5200624.0,
"step": 2820
},
{
"entropy": 5.732665061950684,
"epoch": 0.2359672569328433,
"grad_norm": 1.046875,
"learning_rate": 0.000499737958640858,
"loss": 5.5837,
"mean_token_accuracy": 0.16468401551246642,
"num_tokens": 5209741.0,
"step": 2825
},
{
"entropy": 5.711571788787841,
"epoch": 0.2363848980955563,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997365203223711,
"loss": 5.5124,
"mean_token_accuracy": 0.17061730325222016,
"num_tokens": 5218877.0,
"step": 2830
},
{
"entropy": 5.721763801574707,
"epoch": 0.2368025392582693,
"grad_norm": 0.890625,
"learning_rate": 0.000499735078069601,
"loss": 5.6584,
"mean_token_accuracy": 0.15691242963075638,
"num_tokens": 5229153.0,
"step": 2835
},
{
"entropy": 5.712396860122681,
"epoch": 0.23722018042098228,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997336318825728,
"loss": 5.5358,
"mean_token_accuracy": 0.16605440974235536,
"num_tokens": 5238561.0,
"step": 2840
},
{
"entropy": 5.745545673370361,
"epoch": 0.23763782158369529,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997321817613119,
"loss": 5.6068,
"mean_token_accuracy": 0.16150942146778108,
"num_tokens": 5248135.0,
"step": 2845
},
{
"entropy": 5.727228450775146,
"epoch": 0.2380554627464083,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997307277058436,
"loss": 5.5193,
"mean_token_accuracy": 0.1711078479886055,
"num_tokens": 5257286.0,
"step": 2850
},
{
"entropy": 5.73480453491211,
"epoch": 0.23847310390912127,
"grad_norm": 1.0,
"learning_rate": 0.0004997292697161935,
"loss": 5.7028,
"mean_token_accuracy": 0.1597718745470047,
"num_tokens": 5266595.0,
"step": 2855
},
{
"entropy": 5.738870573043823,
"epoch": 0.23889074507183428,
"grad_norm": 0.95703125,
"learning_rate": 0.000499727807792387,
"loss": 5.4908,
"mean_token_accuracy": 0.16441358774900436,
"num_tokens": 5275651.0,
"step": 2860
},
{
"entropy": 5.6004722118377686,
"epoch": 0.23930838623454728,
"grad_norm": 0.984375,
"learning_rate": 0.0004997263419344495,
"loss": 5.4236,
"mean_token_accuracy": 0.17445990592241287,
"num_tokens": 5284596.0,
"step": 2865
},
{
"entropy": 5.684853649139404,
"epoch": 0.23972602739726026,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997248721424071,
"loss": 5.5716,
"mean_token_accuracy": 0.16735316216945648,
"num_tokens": 5292410.0,
"step": 2870
},
{
"entropy": 5.689931201934814,
"epoch": 0.24014366855997327,
"grad_norm": 1.03125,
"learning_rate": 0.0004997233984162853,
"loss": 5.4656,
"mean_token_accuracy": 0.17355552911758423,
"num_tokens": 5301124.0,
"step": 2875
},
{
"entropy": 5.752901887893676,
"epoch": 0.24056130972268627,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997219207561099,
"loss": 5.5634,
"mean_token_accuracy": 0.16944514065980912,
"num_tokens": 5310571.0,
"step": 2880
},
{
"entropy": 5.711241912841797,
"epoch": 0.24097895088539925,
"grad_norm": 1.015625,
"learning_rate": 0.0004997204391619068,
"loss": 5.5289,
"mean_token_accuracy": 0.16994390040636062,
"num_tokens": 5319746.0,
"step": 2885
},
{
"entropy": 5.681296682357788,
"epoch": 0.24139659204811226,
"grad_norm": 1.046875,
"learning_rate": 0.0004997189536337018,
"loss": 5.6115,
"mean_token_accuracy": 0.16615889891982077,
"num_tokens": 5329734.0,
"step": 2890
},
{
"entropy": 5.638192129135132,
"epoch": 0.24181423321082526,
"grad_norm": 1.015625,
"learning_rate": 0.0004997174641715211,
"loss": 5.4221,
"mean_token_accuracy": 0.174007411301136,
"num_tokens": 5338405.0,
"step": 2895
},
{
"entropy": 5.7107789516448975,
"epoch": 0.24223187437353827,
"grad_norm": 1.015625,
"learning_rate": 0.0004997159707753906,
"loss": 5.6188,
"mean_token_accuracy": 0.16282730400562287,
"num_tokens": 5348179.0,
"step": 2900
},
{
"entropy": 5.830678081512451,
"epoch": 0.24264951553625125,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997144734453367,
"loss": 5.6737,
"mean_token_accuracy": 0.15729831159114838,
"num_tokens": 5358035.0,
"step": 2905
},
{
"entropy": 5.681098127365113,
"epoch": 0.24306715669896425,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997129721813853,
"loss": 5.55,
"mean_token_accuracy": 0.17312878519296646,
"num_tokens": 5366713.0,
"step": 2910
},
{
"entropy": 5.59452486038208,
"epoch": 0.24348479786167726,
"grad_norm": 1.078125,
"learning_rate": 0.0004997114669835629,
"loss": 5.4361,
"mean_token_accuracy": 0.17589147090911866,
"num_tokens": 5375245.0,
"step": 2915
},
{
"entropy": 5.598801755905152,
"epoch": 0.24390243902439024,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997099578518959,
"loss": 5.516,
"mean_token_accuracy": 0.17066216617822647,
"num_tokens": 5384097.0,
"step": 2920
},
{
"entropy": 5.769566297531128,
"epoch": 0.24432008018710324,
"grad_norm": 0.921875,
"learning_rate": 0.0004997084447864105,
"loss": 5.6232,
"mean_token_accuracy": 0.15925524681806563,
"num_tokens": 5393611.0,
"step": 2925
},
{
"entropy": 5.702077054977417,
"epoch": 0.24473772134981625,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997069277871333,
"loss": 5.5554,
"mean_token_accuracy": 0.16924689710140228,
"num_tokens": 5402332.0,
"step": 2930
},
{
"entropy": 5.572508668899536,
"epoch": 0.24515536251252923,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997054068540908,
"loss": 5.5032,
"mean_token_accuracy": 0.1812253177165985,
"num_tokens": 5411415.0,
"step": 2935
},
{
"entropy": 5.762411594390869,
"epoch": 0.24557300367524224,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997038819873098,
"loss": 5.406,
"mean_token_accuracy": 0.17928455621004105,
"num_tokens": 5419742.0,
"step": 2940
},
{
"entropy": 5.63980073928833,
"epoch": 0.24599064483795524,
"grad_norm": 1.03125,
"learning_rate": 0.0004997023531868167,
"loss": 5.4431,
"mean_token_accuracy": 0.17695468217134475,
"num_tokens": 5428689.0,
"step": 2945
},
{
"entropy": 5.7062506675720215,
"epoch": 0.24640828600066822,
"grad_norm": 0.953125,
"learning_rate": 0.0004997008204526386,
"loss": 5.636,
"mean_token_accuracy": 0.159965880215168,
"num_tokens": 5438096.0,
"step": 2950
},
{
"entropy": 5.713601589202881,
"epoch": 0.24682592716338123,
"grad_norm": 1.03125,
"learning_rate": 0.0004996992837848021,
"loss": 5.5199,
"mean_token_accuracy": 0.16694851964712143,
"num_tokens": 5446726.0,
"step": 2955
},
{
"entropy": 5.689603424072265,
"epoch": 0.24724356832609423,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996977431833342,
"loss": 5.6214,
"mean_token_accuracy": 0.164705727994442,
"num_tokens": 5456275.0,
"step": 2960
},
{
"entropy": 5.7204803943634035,
"epoch": 0.2476612094888072,
"grad_norm": 1.109375,
"learning_rate": 0.0004996961986482618,
"loss": 5.5755,
"mean_token_accuracy": 0.17396539449691772,
"num_tokens": 5466135.0,
"step": 2965
},
{
"entropy": 5.666129779815674,
"epoch": 0.24807885065152022,
"grad_norm": 1.03125,
"learning_rate": 0.0004996946501796119,
"loss": 5.445,
"mean_token_accuracy": 0.1775720924139023,
"num_tokens": 5474882.0,
"step": 2970
},
{
"entropy": 5.536153888702392,
"epoch": 0.24849649181423322,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996930977774118,
"loss": 5.3817,
"mean_token_accuracy": 0.17584052830934524,
"num_tokens": 5484856.0,
"step": 2975
},
{
"entropy": 5.7396951675415036,
"epoch": 0.2489141329769462,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996915414416885,
"loss": 5.5855,
"mean_token_accuracy": 0.16176697015762329,
"num_tokens": 5493840.0,
"step": 2980
},
{
"entropy": 5.690059757232666,
"epoch": 0.2493317741396592,
"grad_norm": 1.0,
"learning_rate": 0.0004996899811724693,
"loss": 5.5683,
"mean_token_accuracy": 0.1691231533885002,
"num_tokens": 5502655.0,
"step": 2985
},
{
"entropy": 5.6382838726043705,
"epoch": 0.2497494153023722,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996884169697816,
"loss": 5.4792,
"mean_token_accuracy": 0.17606000751256942,
"num_tokens": 5511640.0,
"step": 2990
},
{
"entropy": 5.601326131820679,
"epoch": 0.2501670564650852,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996868488336526,
"loss": 5.4391,
"mean_token_accuracy": 0.1702730119228363,
"num_tokens": 5520704.0,
"step": 2995
},
{
"entropy": 5.68818998336792,
"epoch": 0.25058469762779817,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996852767641099,
"loss": 5.5449,
"mean_token_accuracy": 0.16632438600063323,
"num_tokens": 5531090.0,
"step": 3000
},
{
"epoch": 0.25058469762779817,
"eval_entropy": 5.435018019059077,
"eval_loss": 5.5041375160217285,
"eval_mean_token_accuracy": 0.17884081382335204,
"eval_num_tokens": 5531090.0,
"eval_runtime": 45.7906,
"eval_samples_per_second": 824.689,
"eval_steps_per_second": 103.1,
"step": 3000
},
{
"entropy": 5.701263189315796,
"epoch": 0.2510023387905112,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996837007611811,
"loss": 5.5868,
"mean_token_accuracy": 0.1650678113102913,
"num_tokens": 5539905.0,
"step": 3005
},
{
"entropy": 5.780577039718628,
"epoch": 0.2514199799532242,
"grad_norm": 1.171875,
"learning_rate": 0.0004996821208248935,
"loss": 5.6192,
"mean_token_accuracy": 0.1613453820347786,
"num_tokens": 5549052.0,
"step": 3010
},
{
"entropy": 5.770159578323364,
"epoch": 0.25183762111593716,
"grad_norm": 1.109375,
"learning_rate": 0.0004996805369552749,
"loss": 5.6166,
"mean_token_accuracy": 0.16313306540250777,
"num_tokens": 5558156.0,
"step": 3015
},
{
"entropy": 5.691643190383911,
"epoch": 0.2522552622786502,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996789491523532,
"loss": 5.5177,
"mean_token_accuracy": 0.1727276027202606,
"num_tokens": 5567367.0,
"step": 3020
},
{
"entropy": 5.689290523529053,
"epoch": 0.2526729034413632,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996773574161559,
"loss": 5.613,
"mean_token_accuracy": 0.1634739249944687,
"num_tokens": 5576883.0,
"step": 3025
},
{
"entropy": 5.735215950012207,
"epoch": 0.25309054460407615,
"grad_norm": 1.09375,
"learning_rate": 0.0004996757617467112,
"loss": 5.5376,
"mean_token_accuracy": 0.16875770837068557,
"num_tokens": 5586050.0,
"step": 3030
},
{
"entropy": 5.781308889389038,
"epoch": 0.2535081857667892,
"grad_norm": 1.0,
"learning_rate": 0.0004996741621440466,
"loss": 5.6752,
"mean_token_accuracy": 0.16329652369022368,
"num_tokens": 5595933.0,
"step": 3035
},
{
"entropy": 5.716435384750366,
"epoch": 0.25392582692950216,
"grad_norm": 0.953125,
"learning_rate": 0.0004996725586081906,
"loss": 5.5353,
"mean_token_accuracy": 0.16573665738105775,
"num_tokens": 5604832.0,
"step": 3040
},
{
"entropy": 5.676588869094848,
"epoch": 0.25434346809221514,
"grad_norm": 1.03125,
"learning_rate": 0.0004996709511391709,
"loss": 5.5299,
"mean_token_accuracy": 0.16873539686203004,
"num_tokens": 5614077.0,
"step": 3045
},
{
"entropy": 5.658464765548706,
"epoch": 0.2547611092549282,
"grad_norm": 1.0,
"learning_rate": 0.0004996693397370158,
"loss": 5.4967,
"mean_token_accuracy": 0.17220488786697388,
"num_tokens": 5622971.0,
"step": 3050
},
{
"entropy": 5.5848020076751705,
"epoch": 0.25517875041764115,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996677244017534,
"loss": 5.4154,
"mean_token_accuracy": 0.17932918518781663,
"num_tokens": 5631356.0,
"step": 3055
},
{
"entropy": 5.64474573135376,
"epoch": 0.25559639158035413,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996661051334121,
"loss": 5.4696,
"mean_token_accuracy": 0.1699503093957901,
"num_tokens": 5641095.0,
"step": 3060
},
{
"entropy": 5.645469522476196,
"epoch": 0.25601403274306717,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996644819320201,
"loss": 5.4605,
"mean_token_accuracy": 0.1748073071241379,
"num_tokens": 5651311.0,
"step": 3065
},
{
"entropy": 5.704125118255615,
"epoch": 0.25643167390578014,
"grad_norm": 1.03125,
"learning_rate": 0.0004996628547976061,
"loss": 5.4973,
"mean_token_accuracy": 0.173713319003582,
"num_tokens": 5659957.0,
"step": 3070
},
{
"entropy": 5.599669027328491,
"epoch": 0.2568493150684932,
"grad_norm": 1.015625,
"learning_rate": 0.0004996612237301983,
"loss": 5.4842,
"mean_token_accuracy": 0.16966650635004044,
"num_tokens": 5668876.0,
"step": 3075
},
{
"entropy": 5.591058588027954,
"epoch": 0.25726695623120616,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996595887298254,
"loss": 5.446,
"mean_token_accuracy": 0.1702518045902252,
"num_tokens": 5677809.0,
"step": 3080
},
{
"entropy": 5.600033569335937,
"epoch": 0.25768459739391913,
"grad_norm": 1.0078125,
"learning_rate": 0.000499657949796516,
"loss": 5.4119,
"mean_token_accuracy": 0.1776470348238945,
"num_tokens": 5687055.0,
"step": 3085
},
{
"entropy": 5.57922329902649,
"epoch": 0.25810223855663217,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996563069302986,
"loss": 5.4714,
"mean_token_accuracy": 0.17120731472969056,
"num_tokens": 5696394.0,
"step": 3090
},
{
"entropy": 5.701501893997192,
"epoch": 0.25851987971934515,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996546601312023,
"loss": 5.5305,
"mean_token_accuracy": 0.17498694807291032,
"num_tokens": 5704372.0,
"step": 3095
},
{
"entropy": 5.607770776748657,
"epoch": 0.2589375208820581,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996530093992558,
"loss": 5.3904,
"mean_token_accuracy": 0.17966010719537734,
"num_tokens": 5713535.0,
"step": 3100
},
{
"entropy": 5.569789981842041,
"epoch": 0.25935516204477116,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996513547344877,
"loss": 5.4381,
"mean_token_accuracy": 0.17471472620964051,
"num_tokens": 5722559.0,
"step": 3105
},
{
"entropy": 5.640872287750244,
"epoch": 0.25977280320748414,
"grad_norm": 1.015625,
"learning_rate": 0.0004996496961369274,
"loss": 5.5028,
"mean_token_accuracy": 0.1675395280122757,
"num_tokens": 5731765.0,
"step": 3110
},
{
"entropy": 5.562532901763916,
"epoch": 0.2601904443701971,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996480336066037,
"loss": 5.3818,
"mean_token_accuracy": 0.18473481386899948,
"num_tokens": 5739989.0,
"step": 3115
},
{
"entropy": 5.575654935836792,
"epoch": 0.26060808553291015,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996463671435458,
"loss": 5.4262,
"mean_token_accuracy": 0.17603574693202972,
"num_tokens": 5748461.0,
"step": 3120
},
{
"entropy": 5.696073913574219,
"epoch": 0.26102572669562313,
"grad_norm": 1.0,
"learning_rate": 0.0004996446967477829,
"loss": 5.5447,
"mean_token_accuracy": 0.17175565510988236,
"num_tokens": 5757495.0,
"step": 3125
},
{
"entropy": 5.760733699798584,
"epoch": 0.2614433678583361,
"grad_norm": 1.0,
"learning_rate": 0.0004996430224193441,
"loss": 5.6074,
"mean_token_accuracy": 0.16298272460699081,
"num_tokens": 5767951.0,
"step": 3130
},
{
"entropy": 5.7048746109008786,
"epoch": 0.26186100902104914,
"grad_norm": 1.015625,
"learning_rate": 0.0004996413441582589,
"loss": 5.5531,
"mean_token_accuracy": 0.17121414095163345,
"num_tokens": 5777021.0,
"step": 3135
},
{
"entropy": 5.621953725814819,
"epoch": 0.2622786501837621,
"grad_norm": 1.078125,
"learning_rate": 0.0004996396619645564,
"loss": 5.5088,
"mean_token_accuracy": 0.17355370968580247,
"num_tokens": 5785794.0,
"step": 3140
},
{
"entropy": 5.62881326675415,
"epoch": 0.2626962913464751,
"grad_norm": 1.078125,
"learning_rate": 0.0004996379758382662,
"loss": 5.3888,
"mean_token_accuracy": 0.17763975858688355,
"num_tokens": 5794480.0,
"step": 3145
},
{
"entropy": 5.504114103317261,
"epoch": 0.26311393250918813,
"grad_norm": 1.09375,
"learning_rate": 0.000499636285779418,
"loss": 5.3339,
"mean_token_accuracy": 0.18272120505571365,
"num_tokens": 5802341.0,
"step": 3150
},
{
"entropy": 5.616283178329468,
"epoch": 0.2635315736719011,
"grad_norm": 1.015625,
"learning_rate": 0.0004996345917880412,
"loss": 5.4833,
"mean_token_accuracy": 0.1707792580127716,
"num_tokens": 5811365.0,
"step": 3155
},
{
"entropy": 5.590710544586182,
"epoch": 0.2639492148346141,
"grad_norm": 1.03125,
"learning_rate": 0.0004996328938641654,
"loss": 5.4102,
"mean_token_accuracy": 0.18022541850805282,
"num_tokens": 5819957.0,
"step": 3160
},
{
"entropy": 5.6348718166351315,
"epoch": 0.2643668559973271,
"grad_norm": 1.125,
"learning_rate": 0.0004996311920078204,
"loss": 5.4697,
"mean_token_accuracy": 0.17249127775430678,
"num_tokens": 5828378.0,
"step": 3165
},
{
"entropy": 5.701619005203247,
"epoch": 0.2647844971600401,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996294862190361,
"loss": 5.536,
"mean_token_accuracy": 0.1653009533882141,
"num_tokens": 5837310.0,
"step": 3170
},
{
"entropy": 5.640773677825928,
"epoch": 0.2652021383227531,
"grad_norm": 0.953125,
"learning_rate": 0.000499627776497842,
"loss": 5.4713,
"mean_token_accuracy": 0.17431499511003495,
"num_tokens": 5845945.0,
"step": 3175
},
{
"entropy": 5.565975904464722,
"epoch": 0.2656197794854661,
"grad_norm": 1.03125,
"learning_rate": 0.0004996260628442686,
"loss": 5.476,
"mean_token_accuracy": 0.16953630596399308,
"num_tokens": 5855954.0,
"step": 3180
},
{
"entropy": 5.73816499710083,
"epoch": 0.2660374206481791,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996243452583454,
"loss": 5.5884,
"mean_token_accuracy": 0.15719056725502015,
"num_tokens": 5865327.0,
"step": 3185
},
{
"entropy": 5.765272760391236,
"epoch": 0.26645506181089207,
"grad_norm": 1.015625,
"learning_rate": 0.0004996226237401028,
"loss": 5.5667,
"mean_token_accuracy": 0.17207725644111632,
"num_tokens": 5873959.0,
"step": 3190
},
{
"entropy": 5.617573165893555,
"epoch": 0.2668727029736051,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996208982895706,
"loss": 5.4019,
"mean_token_accuracy": 0.17430177181959153,
"num_tokens": 5882566.0,
"step": 3195
},
{
"entropy": 5.6079028129577635,
"epoch": 0.2672903441363181,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996191689067794,
"loss": 5.4558,
"mean_token_accuracy": 0.171030256152153,
"num_tokens": 5891407.0,
"step": 3200
},
{
"entropy": 5.680180215835572,
"epoch": 0.26770798529903106,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996174355917591,
"loss": 5.4195,
"mean_token_accuracy": 0.1772758662700653,
"num_tokens": 5900123.0,
"step": 3205
},
{
"entropy": 5.59296555519104,
"epoch": 0.2681256264617441,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996156983445404,
"loss": 5.5092,
"mean_token_accuracy": 0.17191927433013915,
"num_tokens": 5909043.0,
"step": 3210
},
{
"entropy": 5.644886541366577,
"epoch": 0.26854326762445707,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996139571651534,
"loss": 5.5085,
"mean_token_accuracy": 0.17081481367349624,
"num_tokens": 5918728.0,
"step": 3215
},
{
"entropy": 5.700129461288452,
"epoch": 0.26896090878717005,
"grad_norm": 1.046875,
"learning_rate": 0.0004996122120536288,
"loss": 5.5211,
"mean_token_accuracy": 0.1704002872109413,
"num_tokens": 5927281.0,
"step": 3220
},
{
"entropy": 5.654440546035767,
"epoch": 0.2693785499498831,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996104630099969,
"loss": 5.4243,
"mean_token_accuracy": 0.18245238363742827,
"num_tokens": 5935696.0,
"step": 3225
},
{
"entropy": 5.487210369110107,
"epoch": 0.26979619111259606,
"grad_norm": 1.0,
"learning_rate": 0.0004996087100342885,
"loss": 5.475,
"mean_token_accuracy": 0.17293272614479066,
"num_tokens": 5944858.0,
"step": 3230
},
{
"entropy": 5.786357736587524,
"epoch": 0.27021383227530904,
"grad_norm": 1.0,
"learning_rate": 0.0004996069531265345,
"loss": 5.5756,
"mean_token_accuracy": 0.1659095987677574,
"num_tokens": 5954243.0,
"step": 3235
},
{
"entropy": 5.619866609573364,
"epoch": 0.2706314734380221,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996051922867653,
"loss": 5.4607,
"mean_token_accuracy": 0.1738550290465355,
"num_tokens": 5963318.0,
"step": 3240
},
{
"entropy": 5.562462186813354,
"epoch": 0.27104911460073505,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996034275150118,
"loss": 5.3642,
"mean_token_accuracy": 0.18603187501430513,
"num_tokens": 5971487.0,
"step": 3245
},
{
"entropy": 5.563692235946656,
"epoch": 0.27146675576344803,
"grad_norm": 0.99609375,
"learning_rate": 0.000499601658811305,
"loss": 5.4388,
"mean_token_accuracy": 0.18158252984285356,
"num_tokens": 5979981.0,
"step": 3250
},
{
"entropy": 5.720085859298706,
"epoch": 0.27188439692616106,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995998861756758,
"loss": 5.4731,
"mean_token_accuracy": 0.1740057200193405,
"num_tokens": 5989427.0,
"step": 3255
},
{
"entropy": 5.600606250762939,
"epoch": 0.27230203808887404,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995981096081552,
"loss": 5.4389,
"mean_token_accuracy": 0.17140828669071198,
"num_tokens": 5999434.0,
"step": 3260
},
{
"entropy": 5.588597822189331,
"epoch": 0.272719679251587,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995963291087744,
"loss": 5.4958,
"mean_token_accuracy": 0.1686915397644043,
"num_tokens": 6008963.0,
"step": 3265
},
{
"entropy": 5.639681959152222,
"epoch": 0.27313732041430006,
"grad_norm": 0.97265625,
"learning_rate": 0.0004995945446775645,
"loss": 5.5255,
"mean_token_accuracy": 0.16593142300844194,
"num_tokens": 6017693.0,
"step": 3270
},
{
"entropy": 5.689112806320191,
"epoch": 0.27355496157701303,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995927563145566,
"loss": 5.5641,
"mean_token_accuracy": 0.1642194002866745,
"num_tokens": 6027943.0,
"step": 3275
},
{
"entropy": 5.699401807785034,
"epoch": 0.273972602739726,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995909640197825,
"loss": 5.4968,
"mean_token_accuracy": 0.17256641536951065,
"num_tokens": 6036750.0,
"step": 3280
},
{
"entropy": 5.606420516967773,
"epoch": 0.27439024390243905,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995891677932729,
"loss": 5.3693,
"mean_token_accuracy": 0.1757910281419754,
"num_tokens": 6045722.0,
"step": 3285
},
{
"entropy": 5.575706672668457,
"epoch": 0.274807885065152,
"grad_norm": 1.046875,
"learning_rate": 0.0004995873676350597,
"loss": 5.5376,
"mean_token_accuracy": 0.1688882127404213,
"num_tokens": 6055001.0,
"step": 3290
},
{
"entropy": 5.714236783981323,
"epoch": 0.275225526227865,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995855635451743,
"loss": 5.4956,
"mean_token_accuracy": 0.16770867705345155,
"num_tokens": 6063420.0,
"step": 3295
},
{
"entropy": 5.56790246963501,
"epoch": 0.27564316739057804,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995837555236484,
"loss": 5.4715,
"mean_token_accuracy": 0.1717872366309166,
"num_tokens": 6073075.0,
"step": 3300
},
{
"entropy": 5.686523342132569,
"epoch": 0.276060808553291,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995819435705133,
"loss": 5.5323,
"mean_token_accuracy": 0.16897699981927872,
"num_tokens": 6083035.0,
"step": 3305
},
{
"entropy": 5.562081241607666,
"epoch": 0.276478449716004,
"grad_norm": 1.0625,
"learning_rate": 0.0004995801276858011,
"loss": 5.4245,
"mean_token_accuracy": 0.17127932012081146,
"num_tokens": 6090980.0,
"step": 3310
},
{
"entropy": 5.583674812316895,
"epoch": 0.276896090878717,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995783078695434,
"loss": 5.4192,
"mean_token_accuracy": 0.1810470089316368,
"num_tokens": 6099402.0,
"step": 3315
},
{
"entropy": 5.670659017562866,
"epoch": 0.27731373204143,
"grad_norm": 0.88671875,
"learning_rate": 0.000499576484121772,
"loss": 5.4785,
"mean_token_accuracy": 0.17860162109136582,
"num_tokens": 6109251.0,
"step": 3320
},
{
"entropy": 5.685530805587769,
"epoch": 0.277731373204143,
"grad_norm": 1.0,
"learning_rate": 0.0004995746564425191,
"loss": 5.6013,
"mean_token_accuracy": 0.16376227289438247,
"num_tokens": 6119005.0,
"step": 3325
},
{
"entropy": 5.680942440032959,
"epoch": 0.278149014366856,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995728248318163,
"loss": 5.4578,
"mean_token_accuracy": 0.17641908675432205,
"num_tokens": 6127566.0,
"step": 3330
},
{
"entropy": 5.582499313354492,
"epoch": 0.278566655529569,
"grad_norm": 1.0,
"learning_rate": 0.0004995709892896961,
"loss": 5.4739,
"mean_token_accuracy": 0.1744324892759323,
"num_tokens": 6137189.0,
"step": 3335
},
{
"entropy": 5.691714143753051,
"epoch": 0.278984296692282,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995691498161903,
"loss": 5.5094,
"mean_token_accuracy": 0.1700959786772728,
"num_tokens": 6147038.0,
"step": 3340
},
{
"entropy": 5.683166360855102,
"epoch": 0.279401937854995,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995673064113313,
"loss": 5.5137,
"mean_token_accuracy": 0.17305740118026733,
"num_tokens": 6155770.0,
"step": 3345
},
{
"entropy": 5.587224531173706,
"epoch": 0.279819579017708,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995654590751513,
"loss": 5.4766,
"mean_token_accuracy": 0.17449014633893967,
"num_tokens": 6164030.0,
"step": 3350
},
{
"entropy": 5.572806549072266,
"epoch": 0.28023722018042097,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995636078076825,
"loss": 5.4072,
"mean_token_accuracy": 0.17433962225914001,
"num_tokens": 6173046.0,
"step": 3355
},
{
"entropy": 5.666071510314941,
"epoch": 0.280654861343134,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995617526089575,
"loss": 5.4332,
"mean_token_accuracy": 0.179418683052063,
"num_tokens": 6182193.0,
"step": 3360
},
{
"entropy": 5.60524845123291,
"epoch": 0.281072502505847,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995598934790087,
"loss": 5.4363,
"mean_token_accuracy": 0.17807191610336304,
"num_tokens": 6190791.0,
"step": 3365
},
{
"entropy": 5.6347860336303714,
"epoch": 0.28149014366855996,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995580304178687,
"loss": 5.4834,
"mean_token_accuracy": 0.16796986013650894,
"num_tokens": 6199245.0,
"step": 3370
},
{
"entropy": 5.605140256881714,
"epoch": 0.281907784831273,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995561634255701,
"loss": 5.4762,
"mean_token_accuracy": 0.17431096434593202,
"num_tokens": 6208420.0,
"step": 3375
},
{
"entropy": 5.610596609115601,
"epoch": 0.28232542599398597,
"grad_norm": 1.03125,
"learning_rate": 0.0004995542925021455,
"loss": 5.5092,
"mean_token_accuracy": 0.1715408131480217,
"num_tokens": 6217059.0,
"step": 3380
},
{
"entropy": 5.581323337554932,
"epoch": 0.28274306715669895,
"grad_norm": 1.046875,
"learning_rate": 0.0004995524176476277,
"loss": 5.3765,
"mean_token_accuracy": 0.17685751914978026,
"num_tokens": 6225955.0,
"step": 3385
},
{
"entropy": 5.584299468994141,
"epoch": 0.283160708319412,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995505388620497,
"loss": 5.4535,
"mean_token_accuracy": 0.17205138504505157,
"num_tokens": 6235989.0,
"step": 3390
},
{
"entropy": 5.671328830718994,
"epoch": 0.28357834948212496,
"grad_norm": 1.0703125,
"learning_rate": 0.000499548656145444,
"loss": 5.482,
"mean_token_accuracy": 0.16483559012413024,
"num_tokens": 6244821.0,
"step": 3395
},
{
"entropy": 5.658933019638061,
"epoch": 0.28399599064483794,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995467694978439,
"loss": 5.6208,
"mean_token_accuracy": 0.16611048877239226,
"num_tokens": 6254157.0,
"step": 3400
},
{
"entropy": 5.750556373596192,
"epoch": 0.28441363180755097,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995448789192823,
"loss": 5.5468,
"mean_token_accuracy": 0.17158015072345734,
"num_tokens": 6263807.0,
"step": 3405
},
{
"entropy": 5.580238723754883,
"epoch": 0.28483127297026395,
"grad_norm": 1.03125,
"learning_rate": 0.0004995429844097922,
"loss": 5.4675,
"mean_token_accuracy": 0.17550105154514312,
"num_tokens": 6273075.0,
"step": 3410
},
{
"entropy": 5.660174798965454,
"epoch": 0.2852489141329769,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995410859694071,
"loss": 5.4353,
"mean_token_accuracy": 0.17570087164640427,
"num_tokens": 6282079.0,
"step": 3415
},
{
"entropy": 5.564072990417481,
"epoch": 0.28566655529568996,
"grad_norm": 0.96875,
"learning_rate": 0.0004995391835981599,
"loss": 5.4799,
"mean_token_accuracy": 0.1690463125705719,
"num_tokens": 6291957.0,
"step": 3420
},
{
"entropy": 5.637568235397339,
"epoch": 0.28608419645840294,
"grad_norm": 1.078125,
"learning_rate": 0.000499537277296084,
"loss": 5.4848,
"mean_token_accuracy": 0.16970505118370055,
"num_tokens": 6301144.0,
"step": 3425
},
{
"entropy": 5.621378087997437,
"epoch": 0.2865018376211159,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995353670632127,
"loss": 5.4694,
"mean_token_accuracy": 0.17622729688882827,
"num_tokens": 6310654.0,
"step": 3430
},
{
"entropy": 5.651818752288818,
"epoch": 0.28691947878382895,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995334528995797,
"loss": 5.4097,
"mean_token_accuracy": 0.1763767495751381,
"num_tokens": 6319562.0,
"step": 3435
},
{
"entropy": 5.61013503074646,
"epoch": 0.28733711994654193,
"grad_norm": 0.9375,
"learning_rate": 0.0004995315348052182,
"loss": 5.4635,
"mean_token_accuracy": 0.17209085673093796,
"num_tokens": 6329056.0,
"step": 3440
},
{
"entropy": 5.562315130233765,
"epoch": 0.2877547611092549,
"grad_norm": 1.03125,
"learning_rate": 0.000499529612780162,
"loss": 5.4353,
"mean_token_accuracy": 0.1756098523736,
"num_tokens": 6337857.0,
"step": 3445
},
{
"entropy": 5.648341941833496,
"epoch": 0.28817240227196794,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995276868244446,
"loss": 5.3831,
"mean_token_accuracy": 0.1731391057372093,
"num_tokens": 6346876.0,
"step": 3450
},
{
"entropy": 5.526121950149536,
"epoch": 0.2885900434346809,
"grad_norm": 1.015625,
"learning_rate": 0.0004995257569380998,
"loss": 5.4226,
"mean_token_accuracy": 0.1779223069548607,
"num_tokens": 6356200.0,
"step": 3455
},
{
"entropy": 5.66925311088562,
"epoch": 0.2890076845973939,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995238231211614,
"loss": 5.4707,
"mean_token_accuracy": 0.17246579229831696,
"num_tokens": 6365761.0,
"step": 3460
},
{
"entropy": 5.6093944072723385,
"epoch": 0.28942532576010693,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995218853736631,
"loss": 5.5215,
"mean_token_accuracy": 0.1674227923154831,
"num_tokens": 6375326.0,
"step": 3465
},
{
"entropy": 5.691514873504639,
"epoch": 0.2898429669228199,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995199436956391,
"loss": 5.5508,
"mean_token_accuracy": 0.17709466367959975,
"num_tokens": 6384662.0,
"step": 3470
},
{
"entropy": 5.5763490200042725,
"epoch": 0.2902606080855329,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995179980871231,
"loss": 5.4526,
"mean_token_accuracy": 0.17378984689712523,
"num_tokens": 6393576.0,
"step": 3475
},
{
"entropy": 5.612979507446289,
"epoch": 0.2906782492482459,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995160485481493,
"loss": 5.3858,
"mean_token_accuracy": 0.1743532732129097,
"num_tokens": 6402436.0,
"step": 3480
},
{
"entropy": 5.647745513916016,
"epoch": 0.2910958904109589,
"grad_norm": 1.0390625,
"learning_rate": 0.000499514095078752,
"loss": 5.449,
"mean_token_accuracy": 0.1738475278019905,
"num_tokens": 6411471.0,
"step": 3485
},
{
"entropy": 5.667367744445801,
"epoch": 0.2915135315736719,
"grad_norm": 0.9765625,
"learning_rate": 0.000499512137678965,
"loss": 5.5783,
"mean_token_accuracy": 0.16805141121149064,
"num_tokens": 6421355.0,
"step": 3490
},
{
"entropy": 5.580398082733154,
"epoch": 0.2919311727363849,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995101763488229,
"loss": 5.5103,
"mean_token_accuracy": 0.1668199583888054,
"num_tokens": 6430579.0,
"step": 3495
},
{
"entropy": 5.577182912826538,
"epoch": 0.2923488138990979,
"grad_norm": 0.9453125,
"learning_rate": 0.0004995082110883599,
"loss": 5.4551,
"mean_token_accuracy": 0.1728660747408867,
"num_tokens": 6439939.0,
"step": 3500
},
{
"entropy": 5.515458440780639,
"epoch": 0.29276645506181087,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995062418976104,
"loss": 5.2876,
"mean_token_accuracy": 0.17504774332046508,
"num_tokens": 6448598.0,
"step": 3505
},
{
"entropy": 5.620071649551392,
"epoch": 0.2931840962245239,
"grad_norm": 0.921875,
"learning_rate": 0.0004995042687766089,
"loss": 5.5019,
"mean_token_accuracy": 0.1640665978193283,
"num_tokens": 6457647.0,
"step": 3510
},
{
"entropy": 5.542760372161865,
"epoch": 0.2936017373872369,
"grad_norm": 1.015625,
"learning_rate": 0.0004995022917253899,
"loss": 5.318,
"mean_token_accuracy": 0.1857729509472847,
"num_tokens": 6466391.0,
"step": 3515
},
{
"entropy": 5.564976930618286,
"epoch": 0.29401937854994986,
"grad_norm": 1.015625,
"learning_rate": 0.0004995003107439882,
"loss": 5.4214,
"mean_token_accuracy": 0.1815667122602463,
"num_tokens": 6475252.0,
"step": 3520
},
{
"entropy": 5.613153648376465,
"epoch": 0.2944370197126629,
"grad_norm": 1.046875,
"learning_rate": 0.0004994983258324383,
"loss": 5.4737,
"mean_token_accuracy": 0.17397726625204085,
"num_tokens": 6484959.0,
"step": 3525
},
{
"entropy": 5.652011346817017,
"epoch": 0.2948546608753759,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994963369907748,
"loss": 5.5589,
"mean_token_accuracy": 0.17536231279373168,
"num_tokens": 6495221.0,
"step": 3530
},
{
"entropy": 5.596998786926269,
"epoch": 0.29527230203808885,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994943442190328,
"loss": 5.367,
"mean_token_accuracy": 0.1758079096674919,
"num_tokens": 6504390.0,
"step": 3535
},
{
"entropy": 5.49628758430481,
"epoch": 0.2956899432008019,
"grad_norm": 0.890625,
"learning_rate": 0.0004994923475172471,
"loss": 5.3862,
"mean_token_accuracy": 0.17681221067905425,
"num_tokens": 6514105.0,
"step": 3540
},
{
"entropy": 5.497510385513306,
"epoch": 0.29610758436351486,
"grad_norm": 1.0,
"learning_rate": 0.0004994903468854527,
"loss": 5.4067,
"mean_token_accuracy": 0.1738963931798935,
"num_tokens": 6523673.0,
"step": 3545
},
{
"entropy": 5.567109155654907,
"epoch": 0.29652522552622784,
"grad_norm": 0.953125,
"learning_rate": 0.0004994883423236843,
"loss": 5.3705,
"mean_token_accuracy": 0.18165491968393327,
"num_tokens": 6532158.0,
"step": 3550
},
{
"entropy": 5.618200397491455,
"epoch": 0.2969428666889409,
"grad_norm": 0.95703125,
"learning_rate": 0.0004994863338319775,
"loss": 5.4994,
"mean_token_accuracy": 0.16905129104852676,
"num_tokens": 6541580.0,
"step": 3555
},
{
"entropy": 5.614306402206421,
"epoch": 0.29736050785165385,
"grad_norm": 0.92578125,
"learning_rate": 0.0004994843214103672,
"loss": 5.4248,
"mean_token_accuracy": 0.17545585483312606,
"num_tokens": 6550298.0,
"step": 3560
},
{
"entropy": 5.739443969726563,
"epoch": 0.29777814901436683,
"grad_norm": 0.921875,
"learning_rate": 0.0004994823050588886,
"loss": 5.598,
"mean_token_accuracy": 0.17042788714170456,
"num_tokens": 6560774.0,
"step": 3565
},
{
"entropy": 5.604328870773315,
"epoch": 0.29819579017707987,
"grad_norm": 1.0625,
"learning_rate": 0.000499480284777577,
"loss": 5.4048,
"mean_token_accuracy": 0.17681997120380402,
"num_tokens": 6568946.0,
"step": 3570
},
{
"entropy": 5.623058176040649,
"epoch": 0.29861343133979285,
"grad_norm": 1.015625,
"learning_rate": 0.0004994782605664677,
"loss": 5.5405,
"mean_token_accuracy": 0.1736734241247177,
"num_tokens": 6579798.0,
"step": 3575
},
{
"entropy": 5.587291431427002,
"epoch": 0.2990310725025058,
"grad_norm": 1.015625,
"learning_rate": 0.0004994762324255965,
"loss": 5.3898,
"mean_token_accuracy": 0.17470444291830062,
"num_tokens": 6589309.0,
"step": 3580
},
{
"entropy": 5.675253868103027,
"epoch": 0.29944871366521886,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994742003549986,
"loss": 5.5437,
"mean_token_accuracy": 0.16898644268512725,
"num_tokens": 6598835.0,
"step": 3585
},
{
"entropy": 5.6550679206848145,
"epoch": 0.29986635482793184,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994721643547095,
"loss": 5.4385,
"mean_token_accuracy": 0.1754730075597763,
"num_tokens": 6608167.0,
"step": 3590
},
{
"entropy": 5.511614847183227,
"epoch": 0.3002839959906448,
"grad_norm": 1.03125,
"learning_rate": 0.000499470124424765,
"loss": 5.3531,
"mean_token_accuracy": 0.18357801884412767,
"num_tokens": 6616904.0,
"step": 3595
},
{
"entropy": 5.507685136795044,
"epoch": 0.30070163715335785,
"grad_norm": 1.1640625,
"learning_rate": 0.0004994680805652008,
"loss": 5.4475,
"mean_token_accuracy": 0.17928119003772736,
"num_tokens": 6625841.0,
"step": 3600
},
{
"entropy": 5.5269406795501705,
"epoch": 0.3011192783160708,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994660327760527,
"loss": 5.2841,
"mean_token_accuracy": 0.18719015121459961,
"num_tokens": 6633755.0,
"step": 3605
},
{
"entropy": 5.544970369338989,
"epoch": 0.3015369194787838,
"grad_norm": 1.0625,
"learning_rate": 0.0004994639810573565,
"loss": 5.4278,
"mean_token_accuracy": 0.1831843376159668,
"num_tokens": 6642372.0,
"step": 3610
},
{
"entropy": 5.626782369613648,
"epoch": 0.30195456064149684,
"grad_norm": 0.93359375,
"learning_rate": 0.000499461925409148,
"loss": 5.4002,
"mean_token_accuracy": 0.18004399687051773,
"num_tokens": 6652271.0,
"step": 3615
},
{
"entropy": 5.643217420578003,
"epoch": 0.3023722018042098,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994598658314634,
"loss": 5.5107,
"mean_token_accuracy": 0.16798514425754546,
"num_tokens": 6661463.0,
"step": 3620
},
{
"entropy": 5.589829206466675,
"epoch": 0.3027898429669228,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994578023243386,
"loss": 5.4099,
"mean_token_accuracy": 0.17615225911140442,
"num_tokens": 6671423.0,
"step": 3625
},
{
"entropy": 5.581682109832764,
"epoch": 0.30320748412963583,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994557348878098,
"loss": 5.4016,
"mean_token_accuracy": 0.1665767788887024,
"num_tokens": 6680218.0,
"step": 3630
},
{
"entropy": 5.519573068618774,
"epoch": 0.3036251252923488,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994536635219133,
"loss": 5.4024,
"mean_token_accuracy": 0.17173747718334198,
"num_tokens": 6689117.0,
"step": 3635
},
{
"entropy": 5.585209655761719,
"epoch": 0.3040427664550618,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994515882266852,
"loss": 5.4655,
"mean_token_accuracy": 0.17823539227247237,
"num_tokens": 6698024.0,
"step": 3640
},
{
"entropy": 5.65181794166565,
"epoch": 0.3044604076177748,
"grad_norm": 1.03125,
"learning_rate": 0.0004994495090021618,
"loss": 5.5942,
"mean_token_accuracy": 0.1686693847179413,
"num_tokens": 6706253.0,
"step": 3645
},
{
"entropy": 5.648300361633301,
"epoch": 0.3048780487804878,
"grad_norm": 1.0,
"learning_rate": 0.0004994474258483796,
"loss": 5.538,
"mean_token_accuracy": 0.17750954627990723,
"num_tokens": 6716440.0,
"step": 3650
},
{
"entropy": 5.619204807281494,
"epoch": 0.3052956899432008,
"grad_norm": 1.03125,
"learning_rate": 0.0004994453387653751,
"loss": 5.3662,
"mean_token_accuracy": 0.17830655425786973,
"num_tokens": 6725263.0,
"step": 3655
},
{
"entropy": 5.542618083953857,
"epoch": 0.3057133311059138,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994432477531847,
"loss": 5.5249,
"mean_token_accuracy": 0.17351234704256058,
"num_tokens": 6735023.0,
"step": 3660
},
{
"entropy": 5.61643500328064,
"epoch": 0.3061309722686268,
"grad_norm": 1.0,
"learning_rate": 0.0004994411528118452,
"loss": 5.4538,
"mean_token_accuracy": 0.1697386994957924,
"num_tokens": 6744659.0,
"step": 3665
},
{
"entropy": 5.551114273071289,
"epoch": 0.30654861343133977,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994390539413931,
"loss": 5.4562,
"mean_token_accuracy": 0.1664675086736679,
"num_tokens": 6753844.0,
"step": 3670
},
{
"entropy": 5.598834800720215,
"epoch": 0.3069662545940528,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994369511418652,
"loss": 5.4429,
"mean_token_accuracy": 0.17445214986801147,
"num_tokens": 6762574.0,
"step": 3675
},
{
"entropy": 5.6650525569915775,
"epoch": 0.3073838957567658,
"grad_norm": 1.015625,
"learning_rate": 0.0004994348444132983,
"loss": 5.5809,
"mean_token_accuracy": 0.17015109211206436,
"num_tokens": 6772991.0,
"step": 3680
},
{
"entropy": 5.548781204223633,
"epoch": 0.30780153691947876,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994327337557293,
"loss": 5.3976,
"mean_token_accuracy": 0.17374803274869918,
"num_tokens": 6781380.0,
"step": 3685
},
{
"entropy": 5.564214038848877,
"epoch": 0.3082191780821918,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994306191691952,
"loss": 5.4297,
"mean_token_accuracy": 0.17173753976821898,
"num_tokens": 6791125.0,
"step": 3690
},
{
"entropy": 5.718353986740112,
"epoch": 0.30863681924490477,
"grad_norm": 0.8671875,
"learning_rate": 0.0004994285006537329,
"loss": 5.4633,
"mean_token_accuracy": 0.17240104228258132,
"num_tokens": 6801038.0,
"step": 3695
},
{
"entropy": 5.48997654914856,
"epoch": 0.30905446040761775,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994263782093796,
"loss": 5.3667,
"mean_token_accuracy": 0.17592186182737352,
"num_tokens": 6810049.0,
"step": 3700
},
{
"entropy": 5.526648473739624,
"epoch": 0.3094721015703308,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994242518361723,
"loss": 5.3635,
"mean_token_accuracy": 0.17937501668930053,
"num_tokens": 6819375.0,
"step": 3705
},
{
"entropy": 5.6430739402771,
"epoch": 0.30988974273304376,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994221215341484,
"loss": 5.458,
"mean_token_accuracy": 0.17287019789218902,
"num_tokens": 6828872.0,
"step": 3710
},
{
"entropy": 5.544311571121216,
"epoch": 0.31030738389575674,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994199873033452,
"loss": 5.3312,
"mean_token_accuracy": 0.1801329791545868,
"num_tokens": 6838698.0,
"step": 3715
},
{
"entropy": 5.492436122894287,
"epoch": 0.3107250250584698,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994178491437999,
"loss": 5.4537,
"mean_token_accuracy": 0.18416499048471452,
"num_tokens": 6848147.0,
"step": 3720
},
{
"entropy": 5.596598291397095,
"epoch": 0.31114266622118275,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994157070555501,
"loss": 5.4418,
"mean_token_accuracy": 0.17777412086725236,
"num_tokens": 6857777.0,
"step": 3725
},
{
"entropy": 5.618748950958252,
"epoch": 0.31156030738389573,
"grad_norm": 1.046875,
"learning_rate": 0.0004994135610386331,
"loss": 5.3921,
"mean_token_accuracy": 0.17851509898900986,
"num_tokens": 6866340.0,
"step": 3730
},
{
"entropy": 5.623656368255615,
"epoch": 0.31197794854660876,
"grad_norm": 1.046875,
"learning_rate": 0.0004994114110930867,
"loss": 5.4169,
"mean_token_accuracy": 0.18082705289125442,
"num_tokens": 6875670.0,
"step": 3735
},
{
"entropy": 5.6321512222290036,
"epoch": 0.31239558970932174,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994092572189483,
"loss": 5.4428,
"mean_token_accuracy": 0.17251480966806412,
"num_tokens": 6885722.0,
"step": 3740
},
{
"entropy": 5.589204740524292,
"epoch": 0.3128132308720348,
"grad_norm": 0.953125,
"learning_rate": 0.0004994070994162558,
"loss": 5.4977,
"mean_token_accuracy": 0.17485132813453674,
"num_tokens": 6894267.0,
"step": 3745
},
{
"entropy": 5.648172521591187,
"epoch": 0.31323087203474775,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994049376850471,
"loss": 5.4553,
"mean_token_accuracy": 0.16601836010813714,
"num_tokens": 6904488.0,
"step": 3750
},
{
"entropy": 5.583740615844727,
"epoch": 0.31364851319746073,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994027720253596,
"loss": 5.4254,
"mean_token_accuracy": 0.1724591687321663,
"num_tokens": 6913329.0,
"step": 3755
},
{
"entropy": 5.544614505767822,
"epoch": 0.31406615436017377,
"grad_norm": 1.078125,
"learning_rate": 0.0004994006024372315,
"loss": 5.3702,
"mean_token_accuracy": 0.1843334659934044,
"num_tokens": 6921722.0,
"step": 3760
},
{
"entropy": 5.602997827529907,
"epoch": 0.31448379552288674,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993984289207007,
"loss": 5.4567,
"mean_token_accuracy": 0.17751626223325728,
"num_tokens": 6931183.0,
"step": 3765
},
{
"entropy": 5.635328531265259,
"epoch": 0.3149014366855997,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993962514758054,
"loss": 5.5229,
"mean_token_accuracy": 0.17839431315660476,
"num_tokens": 6939813.0,
"step": 3770
},
{
"entropy": 5.642183876037597,
"epoch": 0.31531907784831276,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993940701025836,
"loss": 5.5787,
"mean_token_accuracy": 0.1723739504814148,
"num_tokens": 6950227.0,
"step": 3775
},
{
"entropy": 5.620023775100708,
"epoch": 0.31573671901102573,
"grad_norm": 1.015625,
"learning_rate": 0.0004993918848010734,
"loss": 5.4013,
"mean_token_accuracy": 0.17621598690748214,
"num_tokens": 6959971.0,
"step": 3780
},
{
"entropy": 5.537133884429932,
"epoch": 0.3161543601737387,
"grad_norm": 0.96875,
"learning_rate": 0.0004993896955713133,
"loss": 5.3342,
"mean_token_accuracy": 0.1891302391886711,
"num_tokens": 6968557.0,
"step": 3785
},
{
"entropy": 5.552752256393433,
"epoch": 0.31657200133645175,
"grad_norm": 1.046875,
"learning_rate": 0.0004993875024133414,
"loss": 5.4015,
"mean_token_accuracy": 0.17565442025661468,
"num_tokens": 6978058.0,
"step": 3790
},
{
"entropy": 5.566929817199707,
"epoch": 0.3169896424991647,
"grad_norm": 1.078125,
"learning_rate": 0.0004993853053271962,
"loss": 5.451,
"mean_token_accuracy": 0.17600191235542298,
"num_tokens": 6986549.0,
"step": 3795
},
{
"entropy": 5.570564937591553,
"epoch": 0.3174072836618777,
"grad_norm": 0.96875,
"learning_rate": 0.0004993831043129161,
"loss": 5.4316,
"mean_token_accuracy": 0.1723635047674179,
"num_tokens": 6995907.0,
"step": 3800
},
{
"entropy": 5.569065189361572,
"epoch": 0.31782492482459074,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993808993705397,
"loss": 5.3498,
"mean_token_accuracy": 0.1742982655763626,
"num_tokens": 7006195.0,
"step": 3805
},
{
"entropy": 5.447622632980346,
"epoch": 0.3182425659873037,
"grad_norm": 1.015625,
"learning_rate": 0.0004993786905001055,
"loss": 5.3432,
"mean_token_accuracy": 0.17850562781095505,
"num_tokens": 7015223.0,
"step": 3810
},
{
"entropy": 5.57822813987732,
"epoch": 0.3186602071500167,
"grad_norm": 0.9609375,
"learning_rate": 0.0004993764777016524,
"loss": 5.3828,
"mean_token_accuracy": 0.17754828035831452,
"num_tokens": 7024191.0,
"step": 3815
},
{
"entropy": 5.438897752761841,
"epoch": 0.31907784831272973,
"grad_norm": 1.0,
"learning_rate": 0.0004993742609752188,
"loss": 5.3995,
"mean_token_accuracy": 0.17499469965696335,
"num_tokens": 7032947.0,
"step": 3820
},
{
"entropy": 5.548297119140625,
"epoch": 0.3194954894754427,
"grad_norm": 1.046875,
"learning_rate": 0.0004993720403208438,
"loss": 5.2831,
"mean_token_accuracy": 0.18875021934509278,
"num_tokens": 7041422.0,
"step": 3825
},
{
"entropy": 5.603101539611816,
"epoch": 0.3199131306381557,
"grad_norm": 0.9609375,
"learning_rate": 0.000499369815738566,
"loss": 5.5387,
"mean_token_accuracy": 0.16446553170681,
"num_tokens": 7051051.0,
"step": 3830
},
{
"entropy": 5.542109632492066,
"epoch": 0.3203307718008687,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993675872284248,
"loss": 5.4217,
"mean_token_accuracy": 0.17837198227643966,
"num_tokens": 7060116.0,
"step": 3835
},
{
"entropy": 5.634469318389892,
"epoch": 0.3207484129635817,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993653547904586,
"loss": 5.4017,
"mean_token_accuracy": 0.17458738535642623,
"num_tokens": 7068759.0,
"step": 3840
},
{
"entropy": 5.415265274047852,
"epoch": 0.3211660541262947,
"grad_norm": 1.03125,
"learning_rate": 0.000499363118424707,
"loss": 5.3133,
"mean_token_accuracy": 0.1825884237885475,
"num_tokens": 7078321.0,
"step": 3845
},
{
"entropy": 5.624850416183472,
"epoch": 0.3215836952890077,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993608781312088,
"loss": 5.3847,
"mean_token_accuracy": 0.178279247879982,
"num_tokens": 7088126.0,
"step": 3850
},
{
"entropy": 5.579383516311646,
"epoch": 0.3220013364517207,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993586339100035,
"loss": 5.3573,
"mean_token_accuracy": 0.17236804068088532,
"num_tokens": 7096948.0,
"step": 3855
},
{
"entropy": 5.429286098480224,
"epoch": 0.32241897761443367,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993563857611301,
"loss": 5.3408,
"mean_token_accuracy": 0.18273018449544906,
"num_tokens": 7106867.0,
"step": 3860
},
{
"entropy": 5.467058324813843,
"epoch": 0.3228366187771467,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993541336846283,
"loss": 5.4563,
"mean_token_accuracy": 0.17537071853876113,
"num_tokens": 7115971.0,
"step": 3865
},
{
"entropy": 5.578650236129761,
"epoch": 0.3232542599398597,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993518776805372,
"loss": 5.303,
"mean_token_accuracy": 0.18472694903612136,
"num_tokens": 7124659.0,
"step": 3870
},
{
"entropy": 5.605439043045044,
"epoch": 0.32367190110257266,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993496177488965,
"loss": 5.6036,
"mean_token_accuracy": 0.16980562806129457,
"num_tokens": 7133456.0,
"step": 3875
},
{
"entropy": 5.511414003372193,
"epoch": 0.3240895422652857,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993473538897457,
"loss": 5.3837,
"mean_token_accuracy": 0.17757457196712495,
"num_tokens": 7142649.0,
"step": 3880
},
{
"entropy": 5.648666000366211,
"epoch": 0.32450718342799867,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993450861031243,
"loss": 5.5084,
"mean_token_accuracy": 0.17098803967237472,
"num_tokens": 7153151.0,
"step": 3885
},
{
"entropy": 5.503868484497071,
"epoch": 0.32492482459071165,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993428143890722,
"loss": 5.3833,
"mean_token_accuracy": 0.1792224019765854,
"num_tokens": 7161638.0,
"step": 3890
},
{
"entropy": 5.519910860061645,
"epoch": 0.3253424657534247,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993405387476291,
"loss": 5.3676,
"mean_token_accuracy": 0.180894111096859,
"num_tokens": 7170832.0,
"step": 3895
},
{
"entropy": 5.567269706726075,
"epoch": 0.32576010691613766,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993382591788349,
"loss": 5.3633,
"mean_token_accuracy": 0.18134671151638032,
"num_tokens": 7180025.0,
"step": 3900
},
{
"entropy": 5.553269243240356,
"epoch": 0.32617774807885064,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993359756827293,
"loss": 5.4006,
"mean_token_accuracy": 0.17944396585226058,
"num_tokens": 7189287.0,
"step": 3905
},
{
"entropy": 5.601120281219482,
"epoch": 0.32659538924156367,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993336882593524,
"loss": 5.4187,
"mean_token_accuracy": 0.1689503476023674,
"num_tokens": 7198438.0,
"step": 3910
},
{
"entropy": 5.528104448318482,
"epoch": 0.32701303040427665,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993313969087444,
"loss": 5.4449,
"mean_token_accuracy": 0.1743347465991974,
"num_tokens": 7208006.0,
"step": 3915
},
{
"entropy": 5.602669095993042,
"epoch": 0.32743067156698963,
"grad_norm": 1.046875,
"learning_rate": 0.0004993291016309452,
"loss": 5.4289,
"mean_token_accuracy": 0.17532382905483246,
"num_tokens": 7216975.0,
"step": 3920
},
{
"entropy": 5.513580417633056,
"epoch": 0.32784831272970266,
"grad_norm": 1.0234375,
"learning_rate": 0.000499326802425995,
"loss": 5.4445,
"mean_token_accuracy": 0.17284868955612182,
"num_tokens": 7225664.0,
"step": 3925
},
{
"entropy": 5.578778839111328,
"epoch": 0.32826595389241564,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993244992939342,
"loss": 5.4803,
"mean_token_accuracy": 0.17680099755525588,
"num_tokens": 7234780.0,
"step": 3930
},
{
"entropy": 5.590681791305542,
"epoch": 0.3286835950551286,
"grad_norm": 1.015625,
"learning_rate": 0.000499322192234803,
"loss": 5.2954,
"mean_token_accuracy": 0.18491931557655333,
"num_tokens": 7242648.0,
"step": 3935
},
{
"entropy": 5.542868185043335,
"epoch": 0.32910123621784165,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993198812486418,
"loss": 5.4138,
"mean_token_accuracy": 0.18120619356632234,
"num_tokens": 7251856.0,
"step": 3940
},
{
"entropy": 5.669962549209595,
"epoch": 0.32951887738055463,
"grad_norm": 1.0390625,
"learning_rate": 0.000499317566335491,
"loss": 5.4361,
"mean_token_accuracy": 0.17906445413827896,
"num_tokens": 7260552.0,
"step": 3945
},
{
"entropy": 5.527570629119873,
"epoch": 0.3299365185432676,
"grad_norm": 0.921875,
"learning_rate": 0.0004993152474953913,
"loss": 5.4093,
"mean_token_accuracy": 0.17863740175962448,
"num_tokens": 7270328.0,
"step": 3950
},
{
"entropy": 5.501397800445557,
"epoch": 0.33035415970598064,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993129247283831,
"loss": 5.3743,
"mean_token_accuracy": 0.18057112395763397,
"num_tokens": 7279797.0,
"step": 3955
},
{
"entropy": 5.573873615264892,
"epoch": 0.3307718008686936,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993105980345071,
"loss": 5.4583,
"mean_token_accuracy": 0.17486949414014816,
"num_tokens": 7290205.0,
"step": 3960
},
{
"entropy": 5.548024415969849,
"epoch": 0.3311894420314066,
"grad_norm": 1.078125,
"learning_rate": 0.0004993082674138042,
"loss": 5.4016,
"mean_token_accuracy": 0.18294740617275237,
"num_tokens": 7298544.0,
"step": 3965
},
{
"entropy": 5.442582082748413,
"epoch": 0.33160708319411963,
"grad_norm": 1.09375,
"learning_rate": 0.000499305932866315,
"loss": 5.3753,
"mean_token_accuracy": 0.18276675790548325,
"num_tokens": 7307662.0,
"step": 3970
},
{
"entropy": 5.542536497116089,
"epoch": 0.3320247243568326,
"grad_norm": 0.9609375,
"learning_rate": 0.0004993035943920805,
"loss": 5.4341,
"mean_token_accuracy": 0.17841993570327758,
"num_tokens": 7317085.0,
"step": 3975
},
{
"entropy": 5.643019580841065,
"epoch": 0.3324423655195456,
"grad_norm": 1.0,
"learning_rate": 0.0004993012519911416,
"loss": 5.4074,
"mean_token_accuracy": 0.1812269002199173,
"num_tokens": 7326190.0,
"step": 3980
},
{
"entropy": 5.5581700801849365,
"epoch": 0.3328600066822586,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992989056635392,
"loss": 5.4291,
"mean_token_accuracy": 0.18142253309488296,
"num_tokens": 7336143.0,
"step": 3985
},
{
"entropy": 5.557201528549195,
"epoch": 0.3332776478449716,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992965554093146,
"loss": 5.4159,
"mean_token_accuracy": 0.1773979902267456,
"num_tokens": 7346024.0,
"step": 3990
},
{
"entropy": 5.588890838623047,
"epoch": 0.3336952890076846,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992942012285086,
"loss": 5.4644,
"mean_token_accuracy": 0.1687189981341362,
"num_tokens": 7355477.0,
"step": 3995
},
{
"entropy": 5.5849528312683105,
"epoch": 0.3341129301703976,
"grad_norm": 1.078125,
"learning_rate": 0.0004992918431211629,
"loss": 5.3011,
"mean_token_accuracy": 0.1911688417196274,
"num_tokens": 7364720.0,
"step": 4000
},
{
"entropy": 5.504607343673706,
"epoch": 0.3345305713331106,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992894810873183,
"loss": 5.3315,
"mean_token_accuracy": 0.18405580669641494,
"num_tokens": 7373793.0,
"step": 4005
},
{
"entropy": 5.550771713256836,
"epoch": 0.33494821249582357,
"grad_norm": 0.87109375,
"learning_rate": 0.0004992871151270165,
"loss": 5.48,
"mean_token_accuracy": 0.1778610572218895,
"num_tokens": 7383886.0,
"step": 4010
},
{
"entropy": 5.655559968948364,
"epoch": 0.3353658536585366,
"grad_norm": 0.859375,
"learning_rate": 0.0004992847452402986,
"loss": 5.5231,
"mean_token_accuracy": 0.1737416908144951,
"num_tokens": 7393617.0,
"step": 4015
},
{
"entropy": 5.571028566360473,
"epoch": 0.3357834948212496,
"grad_norm": 1.1171875,
"learning_rate": 0.0004992823714272064,
"loss": 5.2894,
"mean_token_accuracy": 0.1906258374452591,
"num_tokens": 7401949.0,
"step": 4020
},
{
"entropy": 5.399551343917847,
"epoch": 0.33620113598396256,
"grad_norm": 1.0,
"learning_rate": 0.0004992799936877813,
"loss": 5.3085,
"mean_token_accuracy": 0.18554448932409287,
"num_tokens": 7410180.0,
"step": 4025
},
{
"entropy": 5.669768762588501,
"epoch": 0.3366187771466756,
"grad_norm": 0.921875,
"learning_rate": 0.0004992776120220649,
"loss": 5.5245,
"mean_token_accuracy": 0.16831523180007935,
"num_tokens": 7420506.0,
"step": 4030
},
{
"entropy": 5.654279279708862,
"epoch": 0.3370364183093886,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992752264300989,
"loss": 5.428,
"mean_token_accuracy": 0.1782596752047539,
"num_tokens": 7429644.0,
"step": 4035
},
{
"entropy": 5.593759536743164,
"epoch": 0.33745405947210155,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992728369119252,
"loss": 5.473,
"mean_token_accuracy": 0.17445121109485626,
"num_tokens": 7439129.0,
"step": 4040
},
{
"entropy": 5.529770135879517,
"epoch": 0.3378717006348146,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992704434675855,
"loss": 5.4463,
"mean_token_accuracy": 0.16825538277626037,
"num_tokens": 7449355.0,
"step": 4045
},
{
"entropy": 5.601033926010132,
"epoch": 0.33828934179752757,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992680460971218,
"loss": 5.4313,
"mean_token_accuracy": 0.1750248908996582,
"num_tokens": 7458763.0,
"step": 4050
},
{
"entropy": 5.5343467712402346,
"epoch": 0.33870698296024054,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992656448005758,
"loss": 5.3246,
"mean_token_accuracy": 0.18176914155483245,
"num_tokens": 7467697.0,
"step": 4055
},
{
"entropy": 5.548629426956177,
"epoch": 0.3391246241229536,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992632395779899,
"loss": 5.474,
"mean_token_accuracy": 0.1769605129957199,
"num_tokens": 7476728.0,
"step": 4060
},
{
"entropy": 5.527476406097412,
"epoch": 0.33954226528566656,
"grad_norm": 0.9375,
"learning_rate": 0.0004992608304294061,
"loss": 5.4385,
"mean_token_accuracy": 0.1807983100414276,
"num_tokens": 7486794.0,
"step": 4065
},
{
"entropy": 5.648890638351441,
"epoch": 0.33995990644837953,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992584173548664,
"loss": 5.3964,
"mean_token_accuracy": 0.17890453785657884,
"num_tokens": 7496286.0,
"step": 4070
},
{
"entropy": 5.540819454193115,
"epoch": 0.34037754761109257,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992560003544131,
"loss": 5.3963,
"mean_token_accuracy": 0.1823621667921543,
"num_tokens": 7505670.0,
"step": 4075
},
{
"entropy": 5.524999904632568,
"epoch": 0.34079518877380555,
"grad_norm": 0.97265625,
"learning_rate": 0.0004992535794280887,
"loss": 5.4657,
"mean_token_accuracy": 0.17998970299959183,
"num_tokens": 7514702.0,
"step": 4080
},
{
"entropy": 5.573918151855469,
"epoch": 0.3412128299365185,
"grad_norm": 1.015625,
"learning_rate": 0.0004992511545759354,
"loss": 5.4059,
"mean_token_accuracy": 0.1779135674238205,
"num_tokens": 7523705.0,
"step": 4085
},
{
"entropy": 5.477626943588257,
"epoch": 0.34163047109923156,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992487257979958,
"loss": 5.2937,
"mean_token_accuracy": 0.18465856313705445,
"num_tokens": 7532491.0,
"step": 4090
},
{
"entropy": 5.512922763824463,
"epoch": 0.34204811226194454,
"grad_norm": 0.9765625,
"learning_rate": 0.0004992462930943123,
"loss": 5.3333,
"mean_token_accuracy": 0.17840403020381929,
"num_tokens": 7540761.0,
"step": 4095
},
{
"entropy": 5.6294543743133545,
"epoch": 0.3424657534246575,
"grad_norm": 1.015625,
"learning_rate": 0.0004992438564649275,
"loss": 5.414,
"mean_token_accuracy": 0.17790439277887343,
"num_tokens": 7550897.0,
"step": 4100
},
{
"entropy": 5.4592883586883545,
"epoch": 0.34288339458737055,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992414159098841,
"loss": 5.306,
"mean_token_accuracy": 0.17544154673814774,
"num_tokens": 7559982.0,
"step": 4105
},
{
"entropy": 5.524592304229737,
"epoch": 0.3433010357500835,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992389714292247,
"loss": 5.3798,
"mean_token_accuracy": 0.17102374136447906,
"num_tokens": 7569358.0,
"step": 4110
},
{
"entropy": 5.565312910079956,
"epoch": 0.3437186769127965,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992365230229922,
"loss": 5.3843,
"mean_token_accuracy": 0.173369462788105,
"num_tokens": 7578437.0,
"step": 4115
},
{
"entropy": 5.4965414047241214,
"epoch": 0.34413631807550954,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992340706912295,
"loss": 5.3223,
"mean_token_accuracy": 0.18511273562908173,
"num_tokens": 7587528.0,
"step": 4120
},
{
"entropy": 5.589574480056763,
"epoch": 0.3445539592382225,
"grad_norm": 0.984375,
"learning_rate": 0.0004992316144339794,
"loss": 5.3593,
"mean_token_accuracy": 0.1848417639732361,
"num_tokens": 7595991.0,
"step": 4125
},
{
"entropy": 5.51519889831543,
"epoch": 0.3449716004009355,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992291542512852,
"loss": 5.3253,
"mean_token_accuracy": 0.18367814123630524,
"num_tokens": 7604538.0,
"step": 4130
},
{
"entropy": 5.524887561798096,
"epoch": 0.34538924156364853,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992266901431895,
"loss": 5.3594,
"mean_token_accuracy": 0.1797987386584282,
"num_tokens": 7613853.0,
"step": 4135
},
{
"entropy": 5.478319311141968,
"epoch": 0.3458068827263615,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992242221097358,
"loss": 5.2927,
"mean_token_accuracy": 0.1872032940387726,
"num_tokens": 7623538.0,
"step": 4140
},
{
"entropy": 5.511484241485595,
"epoch": 0.3462245238890745,
"grad_norm": 1.015625,
"learning_rate": 0.0004992217501509672,
"loss": 5.3232,
"mean_token_accuracy": 0.1825394406914711,
"num_tokens": 7632121.0,
"step": 4145
},
{
"entropy": 5.602395820617676,
"epoch": 0.3466421650517875,
"grad_norm": 1.0390625,
"learning_rate": 0.000499219274266927,
"loss": 5.4051,
"mean_token_accuracy": 0.1751413092017174,
"num_tokens": 7640779.0,
"step": 4150
},
{
"entropy": 5.487886953353882,
"epoch": 0.3470598062145005,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992167944576584,
"loss": 5.3108,
"mean_token_accuracy": 0.17689754217863082,
"num_tokens": 7649138.0,
"step": 4155
},
{
"entropy": 5.451353406906128,
"epoch": 0.3474774473772135,
"grad_norm": 0.9296875,
"learning_rate": 0.000499214310723205,
"loss": 5.3063,
"mean_token_accuracy": 0.1837406873703003,
"num_tokens": 7658510.0,
"step": 4160
},
{
"entropy": 5.49662823677063,
"epoch": 0.3478950885399265,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992118230636102,
"loss": 5.2671,
"mean_token_accuracy": 0.18834259063005448,
"num_tokens": 7666946.0,
"step": 4165
},
{
"entropy": 5.512709474563598,
"epoch": 0.3483127297026395,
"grad_norm": 1.0,
"learning_rate": 0.0004992093314789176,
"loss": 5.2836,
"mean_token_accuracy": 0.1844685673713684,
"num_tokens": 7675572.0,
"step": 4170
},
{
"entropy": 5.409683656692505,
"epoch": 0.34873037086535247,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992068359691707,
"loss": 5.3178,
"mean_token_accuracy": 0.1820487394928932,
"num_tokens": 7684719.0,
"step": 4175
},
{
"entropy": 5.430762386322021,
"epoch": 0.3491480120280655,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992043365344132,
"loss": 5.3443,
"mean_token_accuracy": 0.17550996094942092,
"num_tokens": 7694370.0,
"step": 4180
},
{
"entropy": 5.593293333053589,
"epoch": 0.3495656531907785,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992018331746891,
"loss": 5.4183,
"mean_token_accuracy": 0.1738266259431839,
"num_tokens": 7703572.0,
"step": 4185
},
{
"entropy": 5.579589700698852,
"epoch": 0.34998329435349146,
"grad_norm": 1.015625,
"learning_rate": 0.0004991993258900418,
"loss": 5.4031,
"mean_token_accuracy": 0.18230030536651612,
"num_tokens": 7712732.0,
"step": 4190
},
{
"entropy": 5.567245578765869,
"epoch": 0.3504009355162045,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991968146805156,
"loss": 5.469,
"mean_token_accuracy": 0.17016618698835373,
"num_tokens": 7722213.0,
"step": 4195
},
{
"entropy": 5.586042642593384,
"epoch": 0.35081857667891747,
"grad_norm": 0.9375,
"learning_rate": 0.0004991942995461542,
"loss": 5.3873,
"mean_token_accuracy": 0.17417784184217452,
"num_tokens": 7732228.0,
"step": 4200
},
{
"entropy": 5.545481777191162,
"epoch": 0.35123621784163045,
"grad_norm": 0.97265625,
"learning_rate": 0.0004991917804870018,
"loss": 5.4492,
"mean_token_accuracy": 0.17282902151346208,
"num_tokens": 7741662.0,
"step": 4205
},
{
"entropy": 5.571476173400879,
"epoch": 0.3516538590043435,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991892575031024,
"loss": 5.3611,
"mean_token_accuracy": 0.18452260494232178,
"num_tokens": 7749608.0,
"step": 4210
},
{
"entropy": 5.485206890106201,
"epoch": 0.35207150016705646,
"grad_norm": 1.0625,
"learning_rate": 0.0004991867305945001,
"loss": 5.3603,
"mean_token_accuracy": 0.17790669053792954,
"num_tokens": 7758387.0,
"step": 4215
},
{
"entropy": 5.433868741989135,
"epoch": 0.35248914132976944,
"grad_norm": 1.03125,
"learning_rate": 0.0004991841997612394,
"loss": 5.3858,
"mean_token_accuracy": 0.17614639103412627,
"num_tokens": 7768103.0,
"step": 4220
},
{
"entropy": 5.48126449584961,
"epoch": 0.3529067824924825,
"grad_norm": 1.046875,
"learning_rate": 0.0004991816650033643,
"loss": 5.233,
"mean_token_accuracy": 0.18586599677801133,
"num_tokens": 7776826.0,
"step": 4225
},
{
"entropy": 5.478780460357666,
"epoch": 0.35332442365519545,
"grad_norm": 1.09375,
"learning_rate": 0.0004991791263209194,
"loss": 5.3538,
"mean_token_accuracy": 0.17375344932079315,
"num_tokens": 7785866.0,
"step": 4230
},
{
"entropy": 5.607062244415284,
"epoch": 0.35374206481790843,
"grad_norm": 0.921875,
"learning_rate": 0.0004991765837139491,
"loss": 5.3845,
"mean_token_accuracy": 0.1705208495259285,
"num_tokens": 7795140.0,
"step": 4235
},
{
"entropy": 5.4969398975372314,
"epoch": 0.35415970598062146,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991740371824977,
"loss": 5.3626,
"mean_token_accuracy": 0.18425862342119217,
"num_tokens": 7804256.0,
"step": 4240
},
{
"entropy": 5.484107542037964,
"epoch": 0.35457734714333444,
"grad_norm": 1.0078125,
"learning_rate": 0.00049917148672661,
"loss": 5.3062,
"mean_token_accuracy": 0.18591732233762742,
"num_tokens": 7812295.0,
"step": 4245
},
{
"entropy": 5.597751331329346,
"epoch": 0.3549949883060474,
"grad_norm": 1.03125,
"learning_rate": 0.0004991689323463306,
"loss": 5.3691,
"mean_token_accuracy": 0.1824835479259491,
"num_tokens": 7821208.0,
"step": 4250
},
{
"entropy": 5.499298477172852,
"epoch": 0.35541262946876045,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991663740417042,
"loss": 5.4201,
"mean_token_accuracy": 0.1770219698548317,
"num_tokens": 7830542.0,
"step": 4255
},
{
"entropy": 5.423828172683716,
"epoch": 0.35583027063147343,
"grad_norm": 0.953125,
"learning_rate": 0.0004991638118127757,
"loss": 5.2156,
"mean_token_accuracy": 0.18910087496042252,
"num_tokens": 7839469.0,
"step": 4260
},
{
"entropy": 5.571734666824341,
"epoch": 0.3562479117941864,
"grad_norm": 0.8828125,
"learning_rate": 0.0004991612456595896,
"loss": 5.399,
"mean_token_accuracy": 0.1752399519085884,
"num_tokens": 7849350.0,
"step": 4265
},
{
"entropy": 5.493672275543213,
"epoch": 0.35666555295689945,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991586755821913,
"loss": 5.3277,
"mean_token_accuracy": 0.18433552831411362,
"num_tokens": 7858882.0,
"step": 4270
},
{
"entropy": 5.481722164154053,
"epoch": 0.3570831941196124,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991561015806254,
"loss": 5.3898,
"mean_token_accuracy": 0.1779269754886627,
"num_tokens": 7868704.0,
"step": 4275
},
{
"entropy": 5.612700986862182,
"epoch": 0.3575008352823254,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991535236549372,
"loss": 5.4024,
"mean_token_accuracy": 0.1754518121480942,
"num_tokens": 7878103.0,
"step": 4280
},
{
"entropy": 5.37280764579773,
"epoch": 0.35791847644503844,
"grad_norm": 0.90625,
"learning_rate": 0.0004991509418051718,
"loss": 5.1519,
"mean_token_accuracy": 0.19291636049747468,
"num_tokens": 7887228.0,
"step": 4285
},
{
"entropy": 5.498071479797363,
"epoch": 0.3583361176077514,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991483560313743,
"loss": 5.3952,
"mean_token_accuracy": 0.17669123709201812,
"num_tokens": 7897205.0,
"step": 4290
},
{
"entropy": 5.4981996536254885,
"epoch": 0.3587537587704644,
"grad_norm": 0.99609375,
"learning_rate": 0.00049914576633359,
"loss": 5.2836,
"mean_token_accuracy": 0.18551304191350937,
"num_tokens": 7906336.0,
"step": 4295
},
{
"entropy": 5.440997743606568,
"epoch": 0.3591713999331774,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991431727118643,
"loss": 5.329,
"mean_token_accuracy": 0.17580654919147493,
"num_tokens": 7914773.0,
"step": 4300
},
{
"entropy": 5.467397689819336,
"epoch": 0.3595890410958904,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991405751662426,
"loss": 5.2677,
"mean_token_accuracy": 0.18639442771673204,
"num_tokens": 7922759.0,
"step": 4305
},
{
"entropy": 5.485037755966187,
"epoch": 0.3600066822586034,
"grad_norm": 0.86328125,
"learning_rate": 0.0004991379736967703,
"loss": 5.2533,
"mean_token_accuracy": 0.18302679061889648,
"num_tokens": 7932199.0,
"step": 4310
},
{
"entropy": 5.428512048721314,
"epoch": 0.3604243234213164,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991353683034929,
"loss": 5.2824,
"mean_token_accuracy": 0.18623681366443634,
"num_tokens": 7941045.0,
"step": 4315
},
{
"entropy": 5.493168306350708,
"epoch": 0.3608419645840294,
"grad_norm": 1.109375,
"learning_rate": 0.0004991327589864563,
"loss": 5.2616,
"mean_token_accuracy": 0.19039287716150283,
"num_tokens": 7949339.0,
"step": 4320
},
{
"entropy": 5.488691711425782,
"epoch": 0.3612596057467424,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991301457457057,
"loss": 5.3045,
"mean_token_accuracy": 0.18072604089975358,
"num_tokens": 7958182.0,
"step": 4325
},
{
"entropy": 5.510261535644531,
"epoch": 0.3616772469094554,
"grad_norm": 1.1328125,
"learning_rate": 0.0004991275285812873,
"loss": 5.3744,
"mean_token_accuracy": 0.18515653908252716,
"num_tokens": 7966711.0,
"step": 4330
},
{
"entropy": 5.407047462463379,
"epoch": 0.3620948880721684,
"grad_norm": 1.0625,
"learning_rate": 0.0004991249074932466,
"loss": 5.226,
"mean_token_accuracy": 0.18636159896850585,
"num_tokens": 7974864.0,
"step": 4335
},
{
"entropy": 5.485093259811402,
"epoch": 0.36251252923488136,
"grad_norm": 0.953125,
"learning_rate": 0.0004991222824816296,
"loss": 5.4021,
"mean_token_accuracy": 0.17070340663194655,
"num_tokens": 7984812.0,
"step": 4340
},
{
"entropy": 5.510895109176635,
"epoch": 0.3629301703975944,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991196535464824,
"loss": 5.3762,
"mean_token_accuracy": 0.17756010890007018,
"num_tokens": 7994862.0,
"step": 4345
},
{
"entropy": 5.519192886352539,
"epoch": 0.3633478115603074,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991170206878508,
"loss": 5.2958,
"mean_token_accuracy": 0.17955076843500137,
"num_tokens": 8003943.0,
"step": 4350
},
{
"entropy": 5.489389324188233,
"epoch": 0.36376545272302035,
"grad_norm": 0.9609375,
"learning_rate": 0.000499114383905781,
"loss": 5.3249,
"mean_token_accuracy": 0.18052455335855483,
"num_tokens": 8012841.0,
"step": 4355
},
{
"entropy": 5.4510892868042,
"epoch": 0.3641830938857334,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991117432003192,
"loss": 5.2826,
"mean_token_accuracy": 0.18153723031282426,
"num_tokens": 8021766.0,
"step": 4360
},
{
"entropy": 5.5102602481842045,
"epoch": 0.36460073504844637,
"grad_norm": 1.015625,
"learning_rate": 0.0004991090985715116,
"loss": 5.2785,
"mean_token_accuracy": 0.1822043165564537,
"num_tokens": 8030491.0,
"step": 4365
},
{
"entropy": 5.442203187942505,
"epoch": 0.36501837621115935,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991064500194043,
"loss": 5.2758,
"mean_token_accuracy": 0.18632659167051316,
"num_tokens": 8039325.0,
"step": 4370
},
{
"entropy": 5.487106895446777,
"epoch": 0.3654360173738724,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991037975440439,
"loss": 5.3648,
"mean_token_accuracy": 0.18059566766023635,
"num_tokens": 8048673.0,
"step": 4375
},
{
"entropy": 5.651902198791504,
"epoch": 0.36585365853658536,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991011411454769,
"loss": 5.5032,
"mean_token_accuracy": 0.17421940341591835,
"num_tokens": 8057934.0,
"step": 4380
},
{
"entropy": 5.448347330093384,
"epoch": 0.36627129969929834,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990984808237495,
"loss": 5.3021,
"mean_token_accuracy": 0.18487884253263473,
"num_tokens": 8066923.0,
"step": 4385
},
{
"entropy": 5.462883567810058,
"epoch": 0.36668894086201137,
"grad_norm": 0.984375,
"learning_rate": 0.0004990958165789086,
"loss": 5.3978,
"mean_token_accuracy": 0.1760290041565895,
"num_tokens": 8076263.0,
"step": 4390
},
{
"entropy": 5.514085721969605,
"epoch": 0.36710658202472435,
"grad_norm": 0.90234375,
"learning_rate": 0.0004990931484110005,
"loss": 5.4095,
"mean_token_accuracy": 0.1810617759823799,
"num_tokens": 8085741.0,
"step": 4395
},
{
"entropy": 5.486966371536255,
"epoch": 0.3675242231874373,
"grad_norm": 0.984375,
"learning_rate": 0.0004990904763200722,
"loss": 5.3771,
"mean_token_accuracy": 0.178292016685009,
"num_tokens": 8094646.0,
"step": 4400
},
{
"entropy": 5.475282049179077,
"epoch": 0.36794186435015036,
"grad_norm": 0.94140625,
"learning_rate": 0.0004990878003061705,
"loss": 5.3496,
"mean_token_accuracy": 0.177623650431633,
"num_tokens": 8103215.0,
"step": 4405
},
{
"entropy": 5.467202425003052,
"epoch": 0.36835950551286334,
"grad_norm": 0.9765625,
"learning_rate": 0.0004990851203693419,
"loss": 5.1819,
"mean_token_accuracy": 0.18833307027816773,
"num_tokens": 8112381.0,
"step": 4410
},
{
"entropy": 5.486483478546143,
"epoch": 0.3687771466755763,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990824365096336,
"loss": 5.2997,
"mean_token_accuracy": 0.1806489571928978,
"num_tokens": 8121139.0,
"step": 4415
},
{
"entropy": 5.561320209503174,
"epoch": 0.36919478783828935,
"grad_norm": 1.03125,
"learning_rate": 0.0004990797487270926,
"loss": 5.4055,
"mean_token_accuracy": 0.18097275197505952,
"num_tokens": 8130011.0,
"step": 4420
},
{
"entropy": 5.4849504947662355,
"epoch": 0.36961242900100233,
"grad_norm": 0.953125,
"learning_rate": 0.0004990770570217659,
"loss": 5.3462,
"mean_token_accuracy": 0.18215416222810746,
"num_tokens": 8138757.0,
"step": 4425
},
{
"entropy": 5.463592290878296,
"epoch": 0.37003007016371536,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990743613937005,
"loss": 5.3316,
"mean_token_accuracy": 0.18118430972099303,
"num_tokens": 8147862.0,
"step": 4430
},
{
"entropy": 5.491280555725098,
"epoch": 0.37044771132642834,
"grad_norm": 0.9765625,
"learning_rate": 0.0004990716618429436,
"loss": 5.2773,
"mean_token_accuracy": 0.1887040838599205,
"num_tokens": 8156754.0,
"step": 4435
},
{
"entropy": 5.466277074813843,
"epoch": 0.3708653524891413,
"grad_norm": 0.984375,
"learning_rate": 0.0004990689583695428,
"loss": 5.3265,
"mean_token_accuracy": 0.18729740530252456,
"num_tokens": 8165647.0,
"step": 4440
},
{
"entropy": 5.428556489944458,
"epoch": 0.37128299365185435,
"grad_norm": 1.0,
"learning_rate": 0.000499066250973545,
"loss": 5.2823,
"mean_token_accuracy": 0.18351357579231262,
"num_tokens": 8174811.0,
"step": 4445
},
{
"entropy": 5.435256910324097,
"epoch": 0.37170063481456733,
"grad_norm": 0.98828125,
"learning_rate": 0.0004990635396549978,
"loss": 5.2514,
"mean_token_accuracy": 0.19050270766019822,
"num_tokens": 8184270.0,
"step": 4450
},
{
"entropy": 5.578146076202392,
"epoch": 0.3721182759772803,
"grad_norm": 1.015625,
"learning_rate": 0.0004990608244139486,
"loss": 5.3904,
"mean_token_accuracy": 0.17651860117912294,
"num_tokens": 8193209.0,
"step": 4455
},
{
"entropy": 5.605398416519165,
"epoch": 0.37253591713999334,
"grad_norm": 1.046875,
"learning_rate": 0.000499058105250445,
"loss": 5.4663,
"mean_token_accuracy": 0.17623425722122193,
"num_tokens": 8202502.0,
"step": 4460
},
{
"entropy": 5.56442608833313,
"epoch": 0.3729535583027063,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990553821645346,
"loss": 5.427,
"mean_token_accuracy": 0.17833691239356994,
"num_tokens": 8212226.0,
"step": 4465
},
{
"entropy": 5.532606315612793,
"epoch": 0.3733711994654193,
"grad_norm": 0.984375,
"learning_rate": 0.000499052655156265,
"loss": 5.3418,
"mean_token_accuracy": 0.17988564670085908,
"num_tokens": 8220729.0,
"step": 4470
},
{
"entropy": 5.487145471572876,
"epoch": 0.37378884062813234,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990499242256839,
"loss": 5.3848,
"mean_token_accuracy": 0.18228236883878707,
"num_tokens": 8230385.0,
"step": 4475
},
{
"entropy": 5.559069967269897,
"epoch": 0.3742064817908453,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990471893728394,
"loss": 5.3262,
"mean_token_accuracy": 0.18018670082092286,
"num_tokens": 8240057.0,
"step": 4480
},
{
"entropy": 5.324749755859375,
"epoch": 0.3746241229535583,
"grad_norm": 0.95703125,
"learning_rate": 0.000499044450597779,
"loss": 5.1832,
"mean_token_accuracy": 0.1948035776615143,
"num_tokens": 8249903.0,
"step": 4485
},
{
"entropy": 5.474420166015625,
"epoch": 0.3750417641162713,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990417079005508,
"loss": 5.3195,
"mean_token_accuracy": 0.18153468817472457,
"num_tokens": 8259334.0,
"step": 4490
},
{
"entropy": 5.384976768493653,
"epoch": 0.3754594052789843,
"grad_norm": 0.97265625,
"learning_rate": 0.000499038961281203,
"loss": 5.2248,
"mean_token_accuracy": 0.19038648456335067,
"num_tokens": 8268543.0,
"step": 4495
},
{
"entropy": 5.525278043746948,
"epoch": 0.3758770464416973,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990362107397834,
"loss": 5.3814,
"mean_token_accuracy": 0.18458831012248994,
"num_tokens": 8278257.0,
"step": 4500
},
{
"entropy": 5.512914323806763,
"epoch": 0.3762946876044103,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990334562763401,
"loss": 5.2567,
"mean_token_accuracy": 0.19528784602880478,
"num_tokens": 8287506.0,
"step": 4505
},
{
"entropy": 5.434405660629272,
"epoch": 0.3767123287671233,
"grad_norm": 1.03125,
"learning_rate": 0.0004990306978909216,
"loss": 5.2564,
"mean_token_accuracy": 0.19175124168395996,
"num_tokens": 8295803.0,
"step": 4510
},
{
"entropy": 5.4258969783782955,
"epoch": 0.3771299699298363,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990279355835761,
"loss": 5.3151,
"mean_token_accuracy": 0.1786240041255951,
"num_tokens": 8305344.0,
"step": 4515
},
{
"entropy": 5.55896520614624,
"epoch": 0.3775476110925493,
"grad_norm": 0.90625,
"learning_rate": 0.0004990251693543519,
"loss": 5.4226,
"mean_token_accuracy": 0.17490887641906738,
"num_tokens": 8315226.0,
"step": 4520
},
{
"entropy": 5.575095748901367,
"epoch": 0.3779652522552623,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990223992032975,
"loss": 5.3775,
"mean_token_accuracy": 0.17822618186473846,
"num_tokens": 8324614.0,
"step": 4525
},
{
"entropy": 5.371170139312744,
"epoch": 0.37838289341797526,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990196251304612,
"loss": 5.1793,
"mean_token_accuracy": 0.18501548767089843,
"num_tokens": 8334014.0,
"step": 4530
},
{
"entropy": 5.452558135986328,
"epoch": 0.3788005345806883,
"grad_norm": 0.98046875,
"learning_rate": 0.000499016847135892,
"loss": 5.2613,
"mean_token_accuracy": 0.1895100936293602,
"num_tokens": 8343038.0,
"step": 4535
},
{
"entropy": 5.53317608833313,
"epoch": 0.3792181757434013,
"grad_norm": 0.94140625,
"learning_rate": 0.000499014065219638,
"loss": 5.4311,
"mean_token_accuracy": 0.17636892199516296,
"num_tokens": 8353170.0,
"step": 4540
},
{
"entropy": 5.456049823760987,
"epoch": 0.37963581690611425,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990112793817481,
"loss": 5.2466,
"mean_token_accuracy": 0.19051716327667237,
"num_tokens": 8362240.0,
"step": 4545
},
{
"entropy": 5.4160411834716795,
"epoch": 0.3800534580688273,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990084896222713,
"loss": 5.2783,
"mean_token_accuracy": 0.18670295178890228,
"num_tokens": 8370723.0,
"step": 4550
},
{
"entropy": 5.404092025756836,
"epoch": 0.38047109923154027,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990056959412562,
"loss": 5.1605,
"mean_token_accuracy": 0.1899563893675804,
"num_tokens": 8379282.0,
"step": 4555
},
{
"entropy": 5.432742595672607,
"epoch": 0.38088874039425324,
"grad_norm": 0.921875,
"learning_rate": 0.0004990028983387517,
"loss": 5.2442,
"mean_token_accuracy": 0.19815326035022734,
"num_tokens": 8387918.0,
"step": 4560
},
{
"entropy": 5.567809247970581,
"epoch": 0.3813063815569663,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990000968148068,
"loss": 5.3888,
"mean_token_accuracy": 0.17480838000774385,
"num_tokens": 8397755.0,
"step": 4565
},
{
"entropy": 5.5175190448760985,
"epoch": 0.38172402271967926,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989972913694707,
"loss": 5.3956,
"mean_token_accuracy": 0.17644692361354827,
"num_tokens": 8408229.0,
"step": 4570
},
{
"entropy": 5.441759347915649,
"epoch": 0.38214166388239224,
"grad_norm": 1.015625,
"learning_rate": 0.0004989944820027923,
"loss": 5.324,
"mean_token_accuracy": 0.17874940633773803,
"num_tokens": 8417039.0,
"step": 4575
},
{
"entropy": 5.417681932449341,
"epoch": 0.38255930504510527,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989916687148209,
"loss": 5.2101,
"mean_token_accuracy": 0.18702251017093657,
"num_tokens": 8426656.0,
"step": 4580
},
{
"entropy": 5.533075571060181,
"epoch": 0.38297694620781825,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989888515056057,
"loss": 5.267,
"mean_token_accuracy": 0.18273862898349763,
"num_tokens": 8435254.0,
"step": 4585
},
{
"entropy": 5.374875497817993,
"epoch": 0.3833945873705312,
"grad_norm": 1.0390625,
"learning_rate": 0.000498986030375196,
"loss": 5.1947,
"mean_token_accuracy": 0.18701607435941697,
"num_tokens": 8444594.0,
"step": 4590
},
{
"entropy": 5.345734453201294,
"epoch": 0.38381222853324426,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989832053236413,
"loss": 5.198,
"mean_token_accuracy": 0.1889701634645462,
"num_tokens": 8453737.0,
"step": 4595
},
{
"entropy": 5.496078252792358,
"epoch": 0.38422986969595724,
"grad_norm": 1.0390625,
"learning_rate": 0.000498980376350991,
"loss": 5.2865,
"mean_token_accuracy": 0.18156538754701615,
"num_tokens": 8463033.0,
"step": 4600
},
{
"entropy": 5.508194017410278,
"epoch": 0.3846475108586702,
"grad_norm": 0.984375,
"learning_rate": 0.0004989775434572945,
"loss": 5.3063,
"mean_token_accuracy": 0.17999700754880904,
"num_tokens": 8472875.0,
"step": 4605
},
{
"entropy": 5.5385175228118895,
"epoch": 0.38506515202138325,
"grad_norm": 0.94140625,
"learning_rate": 0.0004989747066426015,
"loss": 5.4145,
"mean_token_accuracy": 0.17085468024015427,
"num_tokens": 8482281.0,
"step": 4610
},
{
"entropy": 5.502962446212768,
"epoch": 0.38548279318409623,
"grad_norm": 0.98046875,
"learning_rate": 0.0004989718659069618,
"loss": 5.3521,
"mean_token_accuracy": 0.18520839661359786,
"num_tokens": 8490871.0,
"step": 4615
},
{
"entropy": 5.437960529327393,
"epoch": 0.3859004343468092,
"grad_norm": 0.95703125,
"learning_rate": 0.000498969021250425,
"loss": 5.2898,
"mean_token_accuracy": 0.1898990362882614,
"num_tokens": 8499765.0,
"step": 4620
},
{
"entropy": 5.38298659324646,
"epoch": 0.38631807550952224,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989661726730407,
"loss": 5.2825,
"mean_token_accuracy": 0.18729780614376068,
"num_tokens": 8508163.0,
"step": 4625
},
{
"entropy": 5.498678398132324,
"epoch": 0.3867357166722352,
"grad_norm": 1.0,
"learning_rate": 0.0004989633201748591,
"loss": 5.3476,
"mean_token_accuracy": 0.1796739086508751,
"num_tokens": 8517791.0,
"step": 4630
},
{
"entropy": 5.476361989974976,
"epoch": 0.3871533578349482,
"grad_norm": 0.96875,
"learning_rate": 0.0004989604637559299,
"loss": 5.2706,
"mean_token_accuracy": 0.1868073597550392,
"num_tokens": 8526051.0,
"step": 4635
},
{
"entropy": 5.493523216247558,
"epoch": 0.38757099899766123,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989576034163033,
"loss": 5.3115,
"mean_token_accuracy": 0.17812628000974656,
"num_tokens": 8536102.0,
"step": 4640
},
{
"entropy": 5.487749004364014,
"epoch": 0.3879886401603742,
"grad_norm": 0.99609375,
"learning_rate": 0.0004989547391560293,
"loss": 5.3189,
"mean_token_accuracy": 0.17128999531269073,
"num_tokens": 8544566.0,
"step": 4645
},
{
"entropy": 5.517955350875854,
"epoch": 0.3884062813230872,
"grad_norm": 0.984375,
"learning_rate": 0.000498951870975158,
"loss": 5.4864,
"mean_token_accuracy": 0.17243936210870742,
"num_tokens": 8555397.0,
"step": 4650
},
{
"entropy": 5.521270132064819,
"epoch": 0.3888239224858002,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989489988737395,
"loss": 5.3229,
"mean_token_accuracy": 0.186031574010849,
"num_tokens": 8564184.0,
"step": 4655
},
{
"entropy": 5.612167549133301,
"epoch": 0.3892415636485132,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989461228518243,
"loss": 5.3263,
"mean_token_accuracy": 0.18221308887004853,
"num_tokens": 8572967.0,
"step": 4660
},
{
"entropy": 5.474199867248535,
"epoch": 0.3896592048112262,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989432429094627,
"loss": 5.3105,
"mean_token_accuracy": 0.18588981926441192,
"num_tokens": 8581946.0,
"step": 4665
},
{
"entropy": 5.408122158050537,
"epoch": 0.3900768459739392,
"grad_norm": 0.97265625,
"learning_rate": 0.0004989403590467049,
"loss": 5.2165,
"mean_token_accuracy": 0.19287759214639663,
"num_tokens": 8590528.0,
"step": 4670
},
{
"entropy": 5.42282018661499,
"epoch": 0.3904944871366522,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989374712636017,
"loss": 5.2163,
"mean_token_accuracy": 0.188861246407032,
"num_tokens": 8600399.0,
"step": 4675
},
{
"entropy": 5.484813785552978,
"epoch": 0.39091212829936517,
"grad_norm": 1.1484375,
"learning_rate": 0.0004989345795602036,
"loss": 5.2769,
"mean_token_accuracy": 0.18671162724494933,
"num_tokens": 8609620.0,
"step": 4680
},
{
"entropy": 5.41406741142273,
"epoch": 0.3913297694620782,
"grad_norm": 1.046875,
"learning_rate": 0.0004989316839365609,
"loss": 5.3121,
"mean_token_accuracy": 0.18761529624462128,
"num_tokens": 8618077.0,
"step": 4685
},
{
"entropy": 5.373091411590576,
"epoch": 0.3917474106247912,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989287843927246,
"loss": 5.2369,
"mean_token_accuracy": 0.18883857280015945,
"num_tokens": 8626672.0,
"step": 4690
},
{
"entropy": 5.42592487335205,
"epoch": 0.39216505178750416,
"grad_norm": 0.8984375,
"learning_rate": 0.0004989258809287455,
"loss": 5.2432,
"mean_token_accuracy": 0.1877009853720665,
"num_tokens": 8635997.0,
"step": 4695
},
{
"entropy": 5.49935212135315,
"epoch": 0.3925826929502172,
"grad_norm": 1.015625,
"learning_rate": 0.0004989229735446743,
"loss": 5.2962,
"mean_token_accuracy": 0.17100426703691482,
"num_tokens": 8645296.0,
"step": 4700
},
{
"entropy": 5.515685272216797,
"epoch": 0.39300033411293017,
"grad_norm": 0.9140625,
"learning_rate": 0.0004989200622405619,
"loss": 5.4064,
"mean_token_accuracy": 0.17950820028781891,
"num_tokens": 8654772.0,
"step": 4705
},
{
"entropy": 5.4163219928741455,
"epoch": 0.39341797527564315,
"grad_norm": 0.984375,
"learning_rate": 0.0004989171470164592,
"loss": 5.2638,
"mean_token_accuracy": 0.1816800579428673,
"num_tokens": 8664279.0,
"step": 4710
},
{
"entropy": 5.476864528656006,
"epoch": 0.3938356164383562,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989142278724173,
"loss": 5.3503,
"mean_token_accuracy": 0.17603901475667955,
"num_tokens": 8673405.0,
"step": 4715
},
{
"entropy": 5.490294027328491,
"epoch": 0.39425325760106916,
"grad_norm": 1.0234375,
"learning_rate": 0.0004989113048084874,
"loss": 5.3073,
"mean_token_accuracy": 0.1824507713317871,
"num_tokens": 8682254.0,
"step": 4720
},
{
"entropy": 5.45321569442749,
"epoch": 0.39467089876378214,
"grad_norm": 0.9765625,
"learning_rate": 0.0004989083778247206,
"loss": 5.2041,
"mean_token_accuracy": 0.1958479955792427,
"num_tokens": 8690751.0,
"step": 4725
},
{
"entropy": 5.516774129867554,
"epoch": 0.3950885399264952,
"grad_norm": 0.99609375,
"learning_rate": 0.000498905446921168,
"loss": 5.4066,
"mean_token_accuracy": 0.1741472974419594,
"num_tokens": 8700485.0,
"step": 4730
},
{
"entropy": 5.4783166408538815,
"epoch": 0.39550618108920815,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989025120978812,
"loss": 5.3031,
"mean_token_accuracy": 0.17581946849823,
"num_tokens": 8709650.0,
"step": 4735
},
{
"entropy": 5.425529527664184,
"epoch": 0.39592382225192113,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988995733549113,
"loss": 5.2515,
"mean_token_accuracy": 0.18366322964429854,
"num_tokens": 8719090.0,
"step": 4740
},
{
"entropy": 5.426125860214233,
"epoch": 0.39634146341463417,
"grad_norm": 0.94921875,
"learning_rate": 0.0004988966306923099,
"loss": 5.4147,
"mean_token_accuracy": 0.18398146331310272,
"num_tokens": 8728563.0,
"step": 4745
},
{
"entropy": 5.576057195663452,
"epoch": 0.39675910457734714,
"grad_norm": 1.09375,
"learning_rate": 0.0004988936841101284,
"loss": 5.3112,
"mean_token_accuracy": 0.1861005201935768,
"num_tokens": 8737587.0,
"step": 4750
},
{
"entropy": 5.541537809371948,
"epoch": 0.3971767457400601,
"grad_norm": 1.03125,
"learning_rate": 0.0004988907336084186,
"loss": 5.3726,
"mean_token_accuracy": 0.18668136298656463,
"num_tokens": 8747377.0,
"step": 4755
},
{
"entropy": 5.566713142395019,
"epoch": 0.39759438690277316,
"grad_norm": 0.953125,
"learning_rate": 0.0004988877791872319,
"loss": 5.3374,
"mean_token_accuracy": 0.18346888571977615,
"num_tokens": 8756471.0,
"step": 4760
},
{
"entropy": 5.406263446807861,
"epoch": 0.39801202806548613,
"grad_norm": 1.0,
"learning_rate": 0.0004988848208466202,
"loss": 5.3573,
"mean_token_accuracy": 0.18126556873321534,
"num_tokens": 8765445.0,
"step": 4765
},
{
"entropy": 5.578355693817139,
"epoch": 0.3984296692281991,
"grad_norm": 0.921875,
"learning_rate": 0.0004988818585866351,
"loss": 5.3813,
"mean_token_accuracy": 0.18272889852523805,
"num_tokens": 8774946.0,
"step": 4770
},
{
"entropy": 5.536053323745728,
"epoch": 0.39884731039091215,
"grad_norm": 0.984375,
"learning_rate": 0.0004988788924073288,
"loss": 5.3036,
"mean_token_accuracy": 0.17975222170352936,
"num_tokens": 8783652.0,
"step": 4775
},
{
"entropy": 5.3968977451324465,
"epoch": 0.3992649515536251,
"grad_norm": 1.0,
"learning_rate": 0.0004988759223087528,
"loss": 5.2166,
"mean_token_accuracy": 0.19014027565717698,
"num_tokens": 8792633.0,
"step": 4780
},
{
"entropy": 5.366938066482544,
"epoch": 0.3996825927163381,
"grad_norm": 1.0859375,
"learning_rate": 0.0004988729482909595,
"loss": 5.2527,
"mean_token_accuracy": 0.18786804527044296,
"num_tokens": 8802376.0,
"step": 4785
},
{
"entropy": 5.423903322219848,
"epoch": 0.40010023387905114,
"grad_norm": 0.98828125,
"learning_rate": 0.0004988699703540006,
"loss": 5.2833,
"mean_token_accuracy": 0.1882595971226692,
"num_tokens": 8812122.0,
"step": 4790
},
{
"entropy": 5.41010046005249,
"epoch": 0.4005178750417641,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988669884979284,
"loss": 5.2057,
"mean_token_accuracy": 0.1879386231303215,
"num_tokens": 8821111.0,
"step": 4795
},
{
"entropy": 5.447687149047852,
"epoch": 0.4009355162044771,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988640027227951,
"loss": 5.356,
"mean_token_accuracy": 0.18124178498983384,
"num_tokens": 8830354.0,
"step": 4800
},
{
"entropy": 5.403280639648438,
"epoch": 0.4013531573671901,
"grad_norm": 0.9140625,
"learning_rate": 0.000498861013028653,
"loss": 5.2161,
"mean_token_accuracy": 0.1910982608795166,
"num_tokens": 8839500.0,
"step": 4805
},
{
"entropy": 5.484344387054444,
"epoch": 0.4017707985299031,
"grad_norm": 1.0390625,
"learning_rate": 0.0004988580194155544,
"loss": 5.2894,
"mean_token_accuracy": 0.1856370285153389,
"num_tokens": 8849632.0,
"step": 4810
},
{
"entropy": 5.5780744552612305,
"epoch": 0.4021884396926161,
"grad_norm": 1.03125,
"learning_rate": 0.0004988550218835517,
"loss": 5.4133,
"mean_token_accuracy": 0.17849408388137816,
"num_tokens": 8858383.0,
"step": 4815
},
{
"entropy": 5.385653781890869,
"epoch": 0.4026060808553291,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988520204326975,
"loss": 5.1983,
"mean_token_accuracy": 0.19090892970561982,
"num_tokens": 8866755.0,
"step": 4820
},
{
"entropy": 5.454571104049682,
"epoch": 0.4030237220180421,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988490150630441,
"loss": 5.3995,
"mean_token_accuracy": 0.17596109658479692,
"num_tokens": 8876896.0,
"step": 4825
},
{
"entropy": 5.548031997680664,
"epoch": 0.4034413631807551,
"grad_norm": 0.98046875,
"learning_rate": 0.0004988460057746443,
"loss": 5.334,
"mean_token_accuracy": 0.18507929295301437,
"num_tokens": 8886029.0,
"step": 4830
},
{
"entropy": 5.5046196460723875,
"epoch": 0.4038590043434681,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988429925675507,
"loss": 5.224,
"mean_token_accuracy": 0.18310321122407913,
"num_tokens": 8895396.0,
"step": 4835
},
{
"entropy": 5.496002531051635,
"epoch": 0.4042766455061811,
"grad_norm": 0.96875,
"learning_rate": 0.0004988399754418161,
"loss": 5.2947,
"mean_token_accuracy": 0.19374214261770248,
"num_tokens": 8904125.0,
"step": 4840
},
{
"entropy": 5.4126872539520265,
"epoch": 0.40469428666889407,
"grad_norm": 1.0625,
"learning_rate": 0.0004988369543974932,
"loss": 5.3743,
"mean_token_accuracy": 0.1803189903497696,
"num_tokens": 8913011.0,
"step": 4845
},
{
"entropy": 5.402322721481323,
"epoch": 0.4051119278316071,
"grad_norm": 0.97265625,
"learning_rate": 0.0004988339294346351,
"loss": 5.2873,
"mean_token_accuracy": 0.18455570340156555,
"num_tokens": 8921596.0,
"step": 4850
},
{
"entropy": 5.496550750732422,
"epoch": 0.4055295689943201,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988309005532946,
"loss": 5.2637,
"mean_token_accuracy": 0.18543683290481566,
"num_tokens": 8931267.0,
"step": 4855
},
{
"entropy": 5.358983516693115,
"epoch": 0.40594721015703306,
"grad_norm": 0.9375,
"learning_rate": 0.0004988278677535247,
"loss": 5.1291,
"mean_token_accuracy": 0.19053051471710206,
"num_tokens": 8940082.0,
"step": 4860
},
{
"entropy": 5.505894899368286,
"epoch": 0.4063648513197461,
"grad_norm": 0.953125,
"learning_rate": 0.0004988248310353788,
"loss": 5.3916,
"mean_token_accuracy": 0.1844634771347046,
"num_tokens": 8949838.0,
"step": 4865
},
{
"entropy": 5.494176912307739,
"epoch": 0.40678249248245907,
"grad_norm": 1.03125,
"learning_rate": 0.0004988217903989095,
"loss": 5.3029,
"mean_token_accuracy": 0.1894363820552826,
"num_tokens": 8958510.0,
"step": 4870
},
{
"entropy": 5.51385908126831,
"epoch": 0.40720013364517205,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988187458441704,
"loss": 5.3238,
"mean_token_accuracy": 0.18134814202785493,
"num_tokens": 8968372.0,
"step": 4875
},
{
"entropy": 5.3543768405914305,
"epoch": 0.4076177748078851,
"grad_norm": 1.0,
"learning_rate": 0.0004988156973712149,
"loss": 5.1652,
"mean_token_accuracy": 0.19353510141372682,
"num_tokens": 8977121.0,
"step": 4880
},
{
"entropy": 5.433396863937378,
"epoch": 0.40803541597059806,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988126449800961,
"loss": 5.3042,
"mean_token_accuracy": 0.18651665449142457,
"num_tokens": 8986643.0,
"step": 4885
},
{
"entropy": 5.452758121490478,
"epoch": 0.40845305713331104,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988095886708676,
"loss": 5.1491,
"mean_token_accuracy": 0.19153372049331666,
"num_tokens": 8994925.0,
"step": 4890
},
{
"entropy": 5.3938392162322994,
"epoch": 0.40887069829602407,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988065284435828,
"loss": 5.2535,
"mean_token_accuracy": 0.18742894232273102,
"num_tokens": 9003844.0,
"step": 4895
},
{
"entropy": 5.459982872009277,
"epoch": 0.40928833945873705,
"grad_norm": 0.984375,
"learning_rate": 0.0004988034642982954,
"loss": 5.2691,
"mean_token_accuracy": 0.1847367376089096,
"num_tokens": 9013351.0,
"step": 4900
},
{
"entropy": 5.510900974273682,
"epoch": 0.40970598062145,
"grad_norm": 0.8828125,
"learning_rate": 0.0004988003962350589,
"loss": 5.3575,
"mean_token_accuracy": 0.1836380034685135,
"num_tokens": 9023488.0,
"step": 4905
},
{
"entropy": 5.4314319610595705,
"epoch": 0.41012362178416306,
"grad_norm": 0.96875,
"learning_rate": 0.000498797324253927,
"loss": 5.2696,
"mean_token_accuracy": 0.1866202875971794,
"num_tokens": 9033018.0,
"step": 4910
},
{
"entropy": 5.427098226547241,
"epoch": 0.41054126294687604,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987942483549537,
"loss": 5.2602,
"mean_token_accuracy": 0.18624649941921234,
"num_tokens": 9041572.0,
"step": 4915
},
{
"entropy": 5.573683738708496,
"epoch": 0.410958904109589,
"grad_norm": 0.953125,
"learning_rate": 0.0004987911685381926,
"loss": 5.4781,
"mean_token_accuracy": 0.17627662271261216,
"num_tokens": 9051505.0,
"step": 4920
},
{
"entropy": 5.479363441467285,
"epoch": 0.41137654527230205,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987880848036977,
"loss": 5.2731,
"mean_token_accuracy": 0.18796220421791077,
"num_tokens": 9061168.0,
"step": 4925
},
{
"entropy": 5.3330975532531735,
"epoch": 0.41179418643501503,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987849971515232,
"loss": 5.1533,
"mean_token_accuracy": 0.18812801390886308,
"num_tokens": 9069776.0,
"step": 4930
},
{
"entropy": 5.484102582931518,
"epoch": 0.412211827597728,
"grad_norm": 0.96875,
"learning_rate": 0.0004987819055817227,
"loss": 5.2336,
"mean_token_accuracy": 0.18646983206272125,
"num_tokens": 9078995.0,
"step": 4935
},
{
"entropy": 5.394393873214722,
"epoch": 0.41262946876044104,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987788100943507,
"loss": 5.2675,
"mean_token_accuracy": 0.19072956889867781,
"num_tokens": 9088291.0,
"step": 4940
},
{
"entropy": 5.4095031261444095,
"epoch": 0.413047109923154,
"grad_norm": 1.03125,
"learning_rate": 0.0004987757106894613,
"loss": 5.1844,
"mean_token_accuracy": 0.18989295959472657,
"num_tokens": 9097013.0,
"step": 4945
},
{
"entropy": 5.435576009750366,
"epoch": 0.413464751085867,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987726073671086,
"loss": 5.323,
"mean_token_accuracy": 0.17917950749397277,
"num_tokens": 9106909.0,
"step": 4950
},
{
"entropy": 5.417089748382568,
"epoch": 0.41388239224858003,
"grad_norm": 0.93359375,
"learning_rate": 0.0004987695001273471,
"loss": 5.3114,
"mean_token_accuracy": 0.1776151493191719,
"num_tokens": 9116056.0,
"step": 4955
},
{
"entropy": 5.45604944229126,
"epoch": 0.414300033411293,
"grad_norm": 0.953125,
"learning_rate": 0.0004987663889702313,
"loss": 5.271,
"mean_token_accuracy": 0.18915201723575592,
"num_tokens": 9124968.0,
"step": 4960
},
{
"entropy": 5.457709074020386,
"epoch": 0.414717674574006,
"grad_norm": 0.95703125,
"learning_rate": 0.0004987632738958155,
"loss": 5.231,
"mean_token_accuracy": 0.19009630084037782,
"num_tokens": 9134423.0,
"step": 4965
},
{
"entropy": 5.541339588165283,
"epoch": 0.415135315736719,
"grad_norm": 0.89453125,
"learning_rate": 0.0004987601549041541,
"loss": 5.3458,
"mean_token_accuracy": 0.1850879430770874,
"num_tokens": 9144179.0,
"step": 4970
},
{
"entropy": 5.420493459701538,
"epoch": 0.415552956899432,
"grad_norm": 0.953125,
"learning_rate": 0.0004987570319953018,
"loss": 5.3669,
"mean_token_accuracy": 0.18252476155757905,
"num_tokens": 9154045.0,
"step": 4975
},
{
"entropy": 5.308528327941895,
"epoch": 0.415970598062145,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987539051693134,
"loss": 5.165,
"mean_token_accuracy": 0.19155644923448562,
"num_tokens": 9162922.0,
"step": 4980
},
{
"entropy": 5.401633548736572,
"epoch": 0.416388239224858,
"grad_norm": 1.046875,
"learning_rate": 0.0004987507744262436,
"loss": 5.1935,
"mean_token_accuracy": 0.18865952342748643,
"num_tokens": 9172327.0,
"step": 4985
},
{
"entropy": 5.382492542266846,
"epoch": 0.416805880387571,
"grad_norm": 0.96484375,
"learning_rate": 0.0004987476397661472,
"loss": 5.1728,
"mean_token_accuracy": 0.19583573937416077,
"num_tokens": 9180830.0,
"step": 4990
},
{
"entropy": 5.490254497528076,
"epoch": 0.41722352155028397,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987445011890789,
"loss": 5.4038,
"mean_token_accuracy": 0.18453311920166016,
"num_tokens": 9190787.0,
"step": 4995
},
{
"entropy": 5.508502006530762,
"epoch": 0.417641162712997,
"grad_norm": 0.91796875,
"learning_rate": 0.000498741358695094,
"loss": 5.3215,
"mean_token_accuracy": 0.18619910329580308,
"num_tokens": 9199554.0,
"step": 5000
},
{
"entropy": 5.374854564666748,
"epoch": 0.41805880387571,
"grad_norm": 0.97265625,
"learning_rate": 0.000498738212284247,
"loss": 5.2194,
"mean_token_accuracy": 0.18740950971841813,
"num_tokens": 9209309.0,
"step": 5005
},
{
"entropy": 5.488752555847168,
"epoch": 0.41847644503842296,
"grad_norm": 0.921875,
"learning_rate": 0.0004987350619565934,
"loss": 5.2969,
"mean_token_accuracy": 0.1809329703450203,
"num_tokens": 9218529.0,
"step": 5010
},
{
"entropy": 5.462771558761597,
"epoch": 0.418894086201136,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987319077121884,
"loss": 5.2268,
"mean_token_accuracy": 0.19021305441856384,
"num_tokens": 9226924.0,
"step": 5015
},
{
"entropy": 5.348441743850708,
"epoch": 0.419311727363849,
"grad_norm": 0.890625,
"learning_rate": 0.0004987287495510869,
"loss": 5.2795,
"mean_token_accuracy": 0.17676490694284439,
"num_tokens": 9237602.0,
"step": 5020
},
{
"entropy": 5.500708532333374,
"epoch": 0.41972936852656195,
"grad_norm": 0.984375,
"learning_rate": 0.0004987255874733444,
"loss": 5.2708,
"mean_token_accuracy": 0.18865429162979125,
"num_tokens": 9246209.0,
"step": 5025
},
{
"entropy": 5.421678066253662,
"epoch": 0.420147009689275,
"grad_norm": 0.96484375,
"learning_rate": 0.0004987224214790161,
"loss": 5.2258,
"mean_token_accuracy": 0.18645316511392593,
"num_tokens": 9255464.0,
"step": 5030
},
{
"entropy": 5.35802845954895,
"epoch": 0.42056465085198796,
"grad_norm": 1.0234375,
"learning_rate": 0.0004987192515681575,
"loss": 5.2716,
"mean_token_accuracy": 0.1853409081697464,
"num_tokens": 9265520.0,
"step": 5035
},
{
"entropy": 5.481031227111816,
"epoch": 0.42098229201470094,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987160777408243,
"loss": 5.2367,
"mean_token_accuracy": 0.1859673097729683,
"num_tokens": 9274916.0,
"step": 5040
},
{
"entropy": 5.484265422821045,
"epoch": 0.421399933177414,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987128999970717,
"loss": 5.3157,
"mean_token_accuracy": 0.17960992455482483,
"num_tokens": 9284684.0,
"step": 5045
},
{
"entropy": 5.4706159114837645,
"epoch": 0.42181757434012696,
"grad_norm": 0.9375,
"learning_rate": 0.0004987097183369555,
"loss": 5.3478,
"mean_token_accuracy": 0.1874099686741829,
"num_tokens": 9294020.0,
"step": 5050
},
{
"entropy": 5.45339674949646,
"epoch": 0.42223521550283993,
"grad_norm": 0.96875,
"learning_rate": 0.0004987065327605315,
"loss": 5.3207,
"mean_token_accuracy": 0.18615740090608596,
"num_tokens": 9303730.0,
"step": 5055
},
{
"entropy": 5.468004512786865,
"epoch": 0.42265285666555297,
"grad_norm": 0.9609375,
"learning_rate": 0.0004987033432678553,
"loss": 5.2454,
"mean_token_accuracy": 0.18701883554458618,
"num_tokens": 9312851.0,
"step": 5060
},
{
"entropy": 5.352798128128052,
"epoch": 0.42307049782826595,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987001498589828,
"loss": 5.1149,
"mean_token_accuracy": 0.1964243084192276,
"num_tokens": 9321143.0,
"step": 5065
},
{
"entropy": 5.454397535324096,
"epoch": 0.4234881389909789,
"grad_norm": 1.0234375,
"learning_rate": 0.00049869695253397,
"loss": 5.2397,
"mean_token_accuracy": 0.1905997335910797,
"num_tokens": 9329843.0,
"step": 5070
},
{
"entropy": 5.540484428405762,
"epoch": 0.42390578015369196,
"grad_norm": 0.9453125,
"learning_rate": 0.0004986937512928727,
"loss": 5.3858,
"mean_token_accuracy": 0.17478944063186647,
"num_tokens": 9339549.0,
"step": 5075
},
{
"entropy": 5.48171215057373,
"epoch": 0.42432342131640494,
"grad_norm": 0.9453125,
"learning_rate": 0.000498690546135747,
"loss": 5.2764,
"mean_token_accuracy": 0.18295445591211318,
"num_tokens": 9348220.0,
"step": 5080
},
{
"entropy": 5.44566707611084,
"epoch": 0.4247410624791179,
"grad_norm": 1.0234375,
"learning_rate": 0.0004986873370626491,
"loss": 5.2448,
"mean_token_accuracy": 0.1865879163146019,
"num_tokens": 9357440.0,
"step": 5085
},
{
"entropy": 5.469370603561401,
"epoch": 0.42515870364183095,
"grad_norm": 1.0234375,
"learning_rate": 0.000498684124073635,
"loss": 5.2668,
"mean_token_accuracy": 0.18817803263664246,
"num_tokens": 9366457.0,
"step": 5090
},
{
"entropy": 5.437319898605347,
"epoch": 0.4255763448045439,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986809071687612,
"loss": 5.3515,
"mean_token_accuracy": 0.18234614580869674,
"num_tokens": 9375432.0,
"step": 5095
},
{
"entropy": 5.428205251693726,
"epoch": 0.4259939859672569,
"grad_norm": 1.09375,
"learning_rate": 0.0004986776863480838,
"loss": 5.2131,
"mean_token_accuracy": 0.18737971186637878,
"num_tokens": 9383353.0,
"step": 5100
},
{
"entropy": 5.512642192840576,
"epoch": 0.42641162712996994,
"grad_norm": 0.9609375,
"learning_rate": 0.0004986744616116593,
"loss": 5.2686,
"mean_token_accuracy": 0.1873912274837494,
"num_tokens": 9393012.0,
"step": 5105
},
{
"entropy": 5.437117862701416,
"epoch": 0.4268292682926829,
"grad_norm": 1.0390625,
"learning_rate": 0.000498671232959544,
"loss": 5.2801,
"mean_token_accuracy": 0.191709765791893,
"num_tokens": 9401394.0,
"step": 5110
},
{
"entropy": 5.370872163772583,
"epoch": 0.42724690945539595,
"grad_norm": 0.875,
"learning_rate": 0.0004986680003917947,
"loss": 5.2166,
"mean_token_accuracy": 0.18689843267202377,
"num_tokens": 9410310.0,
"step": 5115
},
{
"entropy": 5.393358850479126,
"epoch": 0.42766455061810893,
"grad_norm": 0.953125,
"learning_rate": 0.0004986647639084677,
"loss": 5.1861,
"mean_token_accuracy": 0.19294835776090621,
"num_tokens": 9418625.0,
"step": 5120
},
{
"entropy": 5.455751991271972,
"epoch": 0.4280821917808219,
"grad_norm": 1.0,
"learning_rate": 0.00049866152350962,
"loss": 5.3259,
"mean_token_accuracy": 0.18329329937696456,
"num_tokens": 9428722.0,
"step": 5125
},
{
"entropy": 5.482298088073731,
"epoch": 0.42849983294353494,
"grad_norm": 0.92578125,
"learning_rate": 0.0004986582791953078,
"loss": 5.2909,
"mean_token_accuracy": 0.18959519267082214,
"num_tokens": 9438447.0,
"step": 5130
},
{
"entropy": 5.418071269989014,
"epoch": 0.4289174741062479,
"grad_norm": 1.03125,
"learning_rate": 0.0004986550309655884,
"loss": 5.2217,
"mean_token_accuracy": 0.18700263798236846,
"num_tokens": 9447457.0,
"step": 5135
},
{
"entropy": 5.475923204421997,
"epoch": 0.4293351152689609,
"grad_norm": 1.015625,
"learning_rate": 0.0004986517788205184,
"loss": 5.3571,
"mean_token_accuracy": 0.1782189592719078,
"num_tokens": 9456837.0,
"step": 5140
},
{
"entropy": 5.547563076019287,
"epoch": 0.42975275643167393,
"grad_norm": 1.015625,
"learning_rate": 0.0004986485227601548,
"loss": 5.3784,
"mean_token_accuracy": 0.17853771895170212,
"num_tokens": 9465733.0,
"step": 5145
},
{
"entropy": 5.43075156211853,
"epoch": 0.4301703975943869,
"grad_norm": 1.03125,
"learning_rate": 0.0004986452627845547,
"loss": 5.232,
"mean_token_accuracy": 0.1913004770874977,
"num_tokens": 9473978.0,
"step": 5150
},
{
"entropy": 5.379370880126953,
"epoch": 0.4305880387570999,
"grad_norm": 0.96875,
"learning_rate": 0.000498641998893775,
"loss": 5.2881,
"mean_token_accuracy": 0.18594589680433274,
"num_tokens": 9482914.0,
"step": 5155
},
{
"entropy": 5.438448047637939,
"epoch": 0.4310056799198129,
"grad_norm": 0.94921875,
"learning_rate": 0.0004986387310878728,
"loss": 5.2442,
"mean_token_accuracy": 0.19495287388563157,
"num_tokens": 9492060.0,
"step": 5160
},
{
"entropy": 5.466447591781616,
"epoch": 0.4314233210825259,
"grad_norm": 0.88671875,
"learning_rate": 0.0004986354593669056,
"loss": 5.3145,
"mean_token_accuracy": 0.1829022854566574,
"num_tokens": 9501186.0,
"step": 5165
},
{
"entropy": 5.484063053131104,
"epoch": 0.4318409622452389,
"grad_norm": 0.92578125,
"learning_rate": 0.0004986321837309304,
"loss": 5.2933,
"mean_token_accuracy": 0.18105715066194533,
"num_tokens": 9509727.0,
"step": 5170
},
{
"entropy": 5.478100299835205,
"epoch": 0.4322586034079519,
"grad_norm": 0.96484375,
"learning_rate": 0.0004986289041800046,
"loss": 5.237,
"mean_token_accuracy": 0.18081933557987212,
"num_tokens": 9519203.0,
"step": 5175
},
{
"entropy": 5.443422412872314,
"epoch": 0.4326762445706649,
"grad_norm": 1.015625,
"learning_rate": 0.0004986256207141857,
"loss": 5.2751,
"mean_token_accuracy": 0.18555376678705215,
"num_tokens": 9529277.0,
"step": 5180
},
{
"entropy": 5.4498817920684814,
"epoch": 0.43309388573337787,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986223333335311,
"loss": 5.2232,
"mean_token_accuracy": 0.1830058738589287,
"num_tokens": 9538215.0,
"step": 5185
},
{
"entropy": 5.301967334747315,
"epoch": 0.4335115268960909,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986190420380985,
"loss": 5.187,
"mean_token_accuracy": 0.1891431137919426,
"num_tokens": 9547570.0,
"step": 5190
},
{
"entropy": 5.2756694793701175,
"epoch": 0.4339291680588039,
"grad_norm": 0.83203125,
"learning_rate": 0.0004986157468279453,
"loss": 5.1053,
"mean_token_accuracy": 0.19198305606842042,
"num_tokens": 9556383.0,
"step": 5195
},
{
"entropy": 5.383520793914795,
"epoch": 0.43434680922151686,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986124477031293,
"loss": 5.2461,
"mean_token_accuracy": 0.18692761212587355,
"num_tokens": 9566793.0,
"step": 5200
},
{
"entropy": 5.489547538757324,
"epoch": 0.4347644503842299,
"grad_norm": 0.921875,
"learning_rate": 0.0004986091446637083,
"loss": 5.261,
"mean_token_accuracy": 0.18457262217998505,
"num_tokens": 9576079.0,
"step": 5205
},
{
"entropy": 5.438872385025024,
"epoch": 0.4351820915469429,
"grad_norm": 0.87890625,
"learning_rate": 0.00049860583770974,
"loss": 5.2583,
"mean_token_accuracy": 0.18729605972766877,
"num_tokens": 9586318.0,
"step": 5210
},
{
"entropy": 5.373680973052979,
"epoch": 0.43559973270965585,
"grad_norm": 0.9765625,
"learning_rate": 0.0004986025268412823,
"loss": 5.1746,
"mean_token_accuracy": 0.1897333726286888,
"num_tokens": 9594453.0,
"step": 5215
},
{
"entropy": 5.303825521469117,
"epoch": 0.4360173738723689,
"grad_norm": 0.9609375,
"learning_rate": 0.0004985992120583933,
"loss": 5.1023,
"mean_token_accuracy": 0.19857619404792787,
"num_tokens": 9603677.0,
"step": 5220
},
{
"entropy": 5.422148132324219,
"epoch": 0.43643501503508186,
"grad_norm": 1.0390625,
"learning_rate": 0.000498595893361131,
"loss": 5.2916,
"mean_token_accuracy": 0.18710941970348358,
"num_tokens": 9612353.0,
"step": 5225
},
{
"entropy": 5.506691551208496,
"epoch": 0.43685265619779484,
"grad_norm": 0.92578125,
"learning_rate": 0.0004985925707495533,
"loss": 5.2885,
"mean_token_accuracy": 0.18843815475702286,
"num_tokens": 9621607.0,
"step": 5230
},
{
"entropy": 5.44722580909729,
"epoch": 0.4372702973605079,
"grad_norm": 1.046875,
"learning_rate": 0.0004985892442237186,
"loss": 5.2681,
"mean_token_accuracy": 0.1900441288948059,
"num_tokens": 9630155.0,
"step": 5235
},
{
"entropy": 5.399506902694702,
"epoch": 0.43768793852322085,
"grad_norm": 1.015625,
"learning_rate": 0.0004985859137836851,
"loss": 5.218,
"mean_token_accuracy": 0.18862127661705017,
"num_tokens": 9639418.0,
"step": 5240
},
{
"entropy": 5.438596439361572,
"epoch": 0.43810557968593383,
"grad_norm": 1.0703125,
"learning_rate": 0.0004985825794295109,
"loss": 5.2455,
"mean_token_accuracy": 0.18817966133356095,
"num_tokens": 9648030.0,
"step": 5245
},
{
"entropy": 5.414564228057861,
"epoch": 0.43852322084864687,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985792411612546,
"loss": 5.221,
"mean_token_accuracy": 0.18779697865247727,
"num_tokens": 9657746.0,
"step": 5250
},
{
"entropy": 5.388850307464599,
"epoch": 0.43894086201135984,
"grad_norm": 0.9609375,
"learning_rate": 0.0004985758989789746,
"loss": 5.2583,
"mean_token_accuracy": 0.19324512630701066,
"num_tokens": 9667840.0,
"step": 5255
},
{
"entropy": 5.486927175521851,
"epoch": 0.4393585031740728,
"grad_norm": 0.953125,
"learning_rate": 0.0004985725528827293,
"loss": 5.3114,
"mean_token_accuracy": 0.18287472426891327,
"num_tokens": 9677595.0,
"step": 5260
},
{
"entropy": 5.330494594573975,
"epoch": 0.43977614433678586,
"grad_norm": 0.98046875,
"learning_rate": 0.0004985692028725774,
"loss": 5.1812,
"mean_token_accuracy": 0.19438521265983583,
"num_tokens": 9685515.0,
"step": 5265
},
{
"entropy": 5.343545579910279,
"epoch": 0.44019378549949884,
"grad_norm": 1.0859375,
"learning_rate": 0.0004985658489485774,
"loss": 5.1982,
"mean_token_accuracy": 0.1855165183544159,
"num_tokens": 9694238.0,
"step": 5270
},
{
"entropy": 5.490376043319702,
"epoch": 0.4406114266622118,
"grad_norm": 0.98046875,
"learning_rate": 0.0004985624911107882,
"loss": 5.3041,
"mean_token_accuracy": 0.18229396641254425,
"num_tokens": 9703388.0,
"step": 5275
},
{
"entropy": 5.453785371780396,
"epoch": 0.44102906782492485,
"grad_norm": 1.0234375,
"learning_rate": 0.0004985591293592685,
"loss": 5.2273,
"mean_token_accuracy": 0.18630390614271164,
"num_tokens": 9713085.0,
"step": 5280
},
{
"entropy": 5.477219820022583,
"epoch": 0.4414467089876378,
"grad_norm": 0.98046875,
"learning_rate": 0.0004985557636940771,
"loss": 5.2545,
"mean_token_accuracy": 0.18936343640089034,
"num_tokens": 9721733.0,
"step": 5285
},
{
"entropy": 5.489253044128418,
"epoch": 0.4418643501503508,
"grad_norm": 1.0078125,
"learning_rate": 0.000498552394115273,
"loss": 5.345,
"mean_token_accuracy": 0.18463747203350067,
"num_tokens": 9731478.0,
"step": 5290
},
{
"entropy": 5.4287919998168945,
"epoch": 0.44228199131306384,
"grad_norm": 1.0546875,
"learning_rate": 0.0004985490206229151,
"loss": 5.268,
"mean_token_accuracy": 0.18442467004060745,
"num_tokens": 9742058.0,
"step": 5295
},
{
"entropy": 5.340026950836181,
"epoch": 0.4426996324757768,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985456432170626,
"loss": 5.2287,
"mean_token_accuracy": 0.18886782228946686,
"num_tokens": 9751034.0,
"step": 5300
},
{
"entropy": 5.395762872695923,
"epoch": 0.4431172736384898,
"grad_norm": 0.87890625,
"learning_rate": 0.0004985422618977744,
"loss": 5.2981,
"mean_token_accuracy": 0.18440112471580505,
"num_tokens": 9760692.0,
"step": 5305
},
{
"entropy": 5.432795906066895,
"epoch": 0.44353491480120283,
"grad_norm": 1.0078125,
"learning_rate": 0.00049853887666511,
"loss": 5.1569,
"mean_token_accuracy": 0.18824788480997084,
"num_tokens": 9769759.0,
"step": 5310
},
{
"entropy": 5.376852655410767,
"epoch": 0.4439525559639158,
"grad_norm": 1.1328125,
"learning_rate": 0.0004985354875191283,
"loss": 5.2153,
"mean_token_accuracy": 0.19364855140447618,
"num_tokens": 9778597.0,
"step": 5315
},
{
"entropy": 5.2799968242645265,
"epoch": 0.4443701971266288,
"grad_norm": 1.0,
"learning_rate": 0.0004985320944598889,
"loss": 5.1343,
"mean_token_accuracy": 0.2005080282688141,
"num_tokens": 9787259.0,
"step": 5320
},
{
"entropy": 5.418052053451538,
"epoch": 0.4447878382893418,
"grad_norm": 1.1171875,
"learning_rate": 0.0004985286974874512,
"loss": 5.3293,
"mean_token_accuracy": 0.18076300323009492,
"num_tokens": 9796297.0,
"step": 5325
},
{
"entropy": 5.460309982299805,
"epoch": 0.4452054794520548,
"grad_norm": 1.0703125,
"learning_rate": 0.0004985252966018745,
"loss": 5.1262,
"mean_token_accuracy": 0.20076993554830552,
"num_tokens": 9805217.0,
"step": 5330
},
{
"entropy": 5.400050354003906,
"epoch": 0.4456231206147678,
"grad_norm": 0.9375,
"learning_rate": 0.0004985218918032184,
"loss": 5.2831,
"mean_token_accuracy": 0.18437316864728928,
"num_tokens": 9814788.0,
"step": 5335
},
{
"entropy": 5.306721639633179,
"epoch": 0.4460407617774808,
"grad_norm": 0.9765625,
"learning_rate": 0.0004985184830915426,
"loss": 5.1314,
"mean_token_accuracy": 0.18621596246957778,
"num_tokens": 9823238.0,
"step": 5340
},
{
"entropy": 5.397016334533691,
"epoch": 0.4464584029401938,
"grad_norm": 0.8984375,
"learning_rate": 0.0004985150704669067,
"loss": 5.283,
"mean_token_accuracy": 0.1887279123067856,
"num_tokens": 9832619.0,
"step": 5345
},
{
"entropy": 5.4020778179168705,
"epoch": 0.44687604410290677,
"grad_norm": 0.96484375,
"learning_rate": 0.0004985116539293704,
"loss": 5.1927,
"mean_token_accuracy": 0.19023136645555497,
"num_tokens": 9841505.0,
"step": 5350
},
{
"entropy": 5.439421319961548,
"epoch": 0.4472936852656198,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985082334789936,
"loss": 5.2556,
"mean_token_accuracy": 0.18449438065290452,
"num_tokens": 9851062.0,
"step": 5355
},
{
"entropy": 5.3722827434539795,
"epoch": 0.4477113264283328,
"grad_norm": 0.9609375,
"learning_rate": 0.0004985048091158361,
"loss": 5.1143,
"mean_token_accuracy": 0.18671787679195403,
"num_tokens": 9860080.0,
"step": 5360
},
{
"entropy": 5.377421188354492,
"epoch": 0.44812896759104576,
"grad_norm": 0.90625,
"learning_rate": 0.0004985013808399578,
"loss": 5.308,
"mean_token_accuracy": 0.1794202744960785,
"num_tokens": 9870968.0,
"step": 5365
},
{
"entropy": 5.3914794921875,
"epoch": 0.4485466087537588,
"grad_norm": 0.953125,
"learning_rate": 0.0004984979486514189,
"loss": 5.1939,
"mean_token_accuracy": 0.20115971118211745,
"num_tokens": 9879243.0,
"step": 5370
},
{
"entropy": 5.391350984573364,
"epoch": 0.44896424991647177,
"grad_norm": 0.8984375,
"learning_rate": 0.0004984945125502795,
"loss": 5.198,
"mean_token_accuracy": 0.19181129783391954,
"num_tokens": 9889203.0,
"step": 5375
},
{
"entropy": 5.410650396347046,
"epoch": 0.44938189107918475,
"grad_norm": 0.91796875,
"learning_rate": 0.0004984910725365994,
"loss": 5.2604,
"mean_token_accuracy": 0.19484663009643555,
"num_tokens": 9898989.0,
"step": 5380
},
{
"entropy": 5.385343742370606,
"epoch": 0.4497995322418978,
"grad_norm": 0.8671875,
"learning_rate": 0.0004984876286104392,
"loss": 5.1857,
"mean_token_accuracy": 0.19111303240060806,
"num_tokens": 9908030.0,
"step": 5385
},
{
"entropy": 5.515639925003052,
"epoch": 0.45021717340461076,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984841807718591,
"loss": 5.4202,
"mean_token_accuracy": 0.18028317391872406,
"num_tokens": 9917331.0,
"step": 5390
},
{
"entropy": 5.442778778076172,
"epoch": 0.45063481456732374,
"grad_norm": 1.0,
"learning_rate": 0.0004984807290209193,
"loss": 5.1812,
"mean_token_accuracy": 0.19573354125022888,
"num_tokens": 9925418.0,
"step": 5395
},
{
"entropy": 5.4272825717926025,
"epoch": 0.45105245573003677,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984772733576803,
"loss": 5.2878,
"mean_token_accuracy": 0.17937994450330735,
"num_tokens": 9934525.0,
"step": 5400
},
{
"entropy": 5.421660232543945,
"epoch": 0.45147009689274975,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984738137822028,
"loss": 5.3251,
"mean_token_accuracy": 0.18079611659049988,
"num_tokens": 9944087.0,
"step": 5405
},
{
"entropy": 5.493865871429444,
"epoch": 0.45188773805546273,
"grad_norm": 0.93359375,
"learning_rate": 0.000498470350294547,
"loss": 5.2337,
"mean_token_accuracy": 0.19009149372577666,
"num_tokens": 9953005.0,
"step": 5410
},
{
"entropy": 5.445831775665283,
"epoch": 0.45230537921817576,
"grad_norm": 0.98046875,
"learning_rate": 0.0004984668828947739,
"loss": 5.1957,
"mean_token_accuracy": 0.18674670457839965,
"num_tokens": 9961963.0,
"step": 5415
},
{
"entropy": 5.354967498779297,
"epoch": 0.45272302038088874,
"grad_norm": 0.98046875,
"learning_rate": 0.0004984634115829439,
"loss": 5.2189,
"mean_token_accuracy": 0.1852552056312561,
"num_tokens": 9970654.0,
"step": 5420
},
{
"entropy": 5.432764148712158,
"epoch": 0.4531406615436017,
"grad_norm": 0.9921875,
"learning_rate": 0.000498459936359118,
"loss": 5.2775,
"mean_token_accuracy": 0.18723644316196442,
"num_tokens": 9979625.0,
"step": 5425
},
{
"entropy": 5.433414268493652,
"epoch": 0.45355830270631475,
"grad_norm": 0.9140625,
"learning_rate": 0.0004984564572233568,
"loss": 5.201,
"mean_token_accuracy": 0.19804619997739792,
"num_tokens": 9988584.0,
"step": 5430
},
{
"entropy": 5.314892339706421,
"epoch": 0.45397594386902773,
"grad_norm": 0.98046875,
"learning_rate": 0.0004984529741757216,
"loss": 5.1175,
"mean_token_accuracy": 0.19850447922945022,
"num_tokens": 9997315.0,
"step": 5435
},
{
"entropy": 5.330080842971801,
"epoch": 0.4543935850317407,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984494872162728,
"loss": 5.2569,
"mean_token_accuracy": 0.18722862750291824,
"num_tokens": 10006431.0,
"step": 5440
},
{
"entropy": 5.435089254379273,
"epoch": 0.45481122619445374,
"grad_norm": 1.0390625,
"learning_rate": 0.000498445996345072,
"loss": 5.2344,
"mean_token_accuracy": 0.19351187199354172,
"num_tokens": 10015960.0,
"step": 5445
},
{
"entropy": 5.407324361801147,
"epoch": 0.4552288673571667,
"grad_norm": 0.859375,
"learning_rate": 0.0004984425015621799,
"loss": 5.1885,
"mean_token_accuracy": 0.19092332422733307,
"num_tokens": 10025948.0,
"step": 5450
},
{
"entropy": 5.432583522796631,
"epoch": 0.4556465085198797,
"grad_norm": 0.98828125,
"learning_rate": 0.000498439002867658,
"loss": 5.2369,
"mean_token_accuracy": 0.1858166828751564,
"num_tokens": 10034358.0,
"step": 5455
},
{
"entropy": 5.421167230606079,
"epoch": 0.45606414968259273,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984355002615673,
"loss": 5.2892,
"mean_token_accuracy": 0.18035876601934434,
"num_tokens": 10044887.0,
"step": 5460
},
{
"entropy": 5.349170207977295,
"epoch": 0.4564817908453057,
"grad_norm": 0.91796875,
"learning_rate": 0.0004984319937439693,
"loss": 5.1712,
"mean_token_accuracy": 0.1863957315683365,
"num_tokens": 10054026.0,
"step": 5465
},
{
"entropy": 5.409258699417114,
"epoch": 0.4568994320080187,
"grad_norm": 0.96875,
"learning_rate": 0.0004984284833149252,
"loss": 5.2769,
"mean_token_accuracy": 0.1844397231936455,
"num_tokens": 10063089.0,
"step": 5470
},
{
"entropy": 5.588518762588501,
"epoch": 0.4573170731707317,
"grad_norm": 1.0,
"learning_rate": 0.0004984249689744966,
"loss": 5.3104,
"mean_token_accuracy": 0.1768183022737503,
"num_tokens": 10073816.0,
"step": 5475
},
{
"entropy": 5.40412712097168,
"epoch": 0.4577347143334447,
"grad_norm": 0.9921875,
"learning_rate": 0.000498421450722745,
"loss": 5.2455,
"mean_token_accuracy": 0.189549458026886,
"num_tokens": 10082407.0,
"step": 5480
},
{
"entropy": 5.391772985458374,
"epoch": 0.4581523554961577,
"grad_norm": 1.0390625,
"learning_rate": 0.000498417928559732,
"loss": 5.2202,
"mean_token_accuracy": 0.18563029915094376,
"num_tokens": 10091423.0,
"step": 5485
},
{
"entropy": 5.433389282226562,
"epoch": 0.4585699966588707,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984144024855191,
"loss": 5.2309,
"mean_token_accuracy": 0.18567567467689514,
"num_tokens": 10101381.0,
"step": 5490
},
{
"entropy": 5.424247074127197,
"epoch": 0.4589876378215837,
"grad_norm": 0.8515625,
"learning_rate": 0.0004984108725001683,
"loss": 5.2853,
"mean_token_accuracy": 0.18847638964653016,
"num_tokens": 10111000.0,
"step": 5495
},
{
"entropy": 5.465272045135498,
"epoch": 0.45940527898429667,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984073386037413,
"loss": 5.2686,
"mean_token_accuracy": 0.18563397377729415,
"num_tokens": 10120380.0,
"step": 5500
},
{
"entropy": 5.40173716545105,
"epoch": 0.4598229201470097,
"grad_norm": 0.90234375,
"learning_rate": 0.0004984038007962997,
"loss": 5.228,
"mean_token_accuracy": 0.18884543776512147,
"num_tokens": 10129236.0,
"step": 5505
},
{
"entropy": 5.479322004318237,
"epoch": 0.4602405613097227,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984002590779059,
"loss": 5.3713,
"mean_token_accuracy": 0.18686946481466293,
"num_tokens": 10138851.0,
"step": 5510
},
{
"entropy": 5.427707624435425,
"epoch": 0.46065820247243566,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983967134486215,
"loss": 5.239,
"mean_token_accuracy": 0.1890336573123932,
"num_tokens": 10148030.0,
"step": 5515
},
{
"entropy": 5.394572162628174,
"epoch": 0.4610758436351487,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983931639085088,
"loss": 5.1859,
"mean_token_accuracy": 0.19900597482919694,
"num_tokens": 10156362.0,
"step": 5520
},
{
"entropy": 5.4444725036621096,
"epoch": 0.4614934847978617,
"grad_norm": 0.91015625,
"learning_rate": 0.0004983896104576298,
"loss": 5.255,
"mean_token_accuracy": 0.1866983264684677,
"num_tokens": 10165603.0,
"step": 5525
},
{
"entropy": 5.396441125869751,
"epoch": 0.46191112596057465,
"grad_norm": 0.890625,
"learning_rate": 0.000498386053096047,
"loss": 5.2353,
"mean_token_accuracy": 0.18723201602697373,
"num_tokens": 10175577.0,
"step": 5530
},
{
"entropy": 5.396557235717774,
"epoch": 0.4623287671232877,
"grad_norm": 1.0,
"learning_rate": 0.0004983824918238221,
"loss": 5.1474,
"mean_token_accuracy": 0.19485459923744203,
"num_tokens": 10184736.0,
"step": 5535
},
{
"entropy": 5.470610809326172,
"epoch": 0.46274640828600067,
"grad_norm": 1.0078125,
"learning_rate": 0.000498378926641018,
"loss": 5.2521,
"mean_token_accuracy": 0.19185058474540712,
"num_tokens": 10194154.0,
"step": 5540
},
{
"entropy": 5.423891258239746,
"epoch": 0.46316404944871364,
"grad_norm": 1.046875,
"learning_rate": 0.0004983753575476969,
"loss": 5.2859,
"mean_token_accuracy": 0.18525027632713317,
"num_tokens": 10203773.0,
"step": 5545
},
{
"entropy": 5.381209850311279,
"epoch": 0.4635816906114267,
"grad_norm": 1.0625,
"learning_rate": 0.0004983717845439212,
"loss": 5.2654,
"mean_token_accuracy": 0.18513274788856507,
"num_tokens": 10212260.0,
"step": 5550
},
{
"entropy": 5.422849082946778,
"epoch": 0.46399933177413966,
"grad_norm": 0.859375,
"learning_rate": 0.0004983682076297536,
"loss": 5.3042,
"mean_token_accuracy": 0.18700378984212876,
"num_tokens": 10221042.0,
"step": 5555
},
{
"entropy": 5.414604806900025,
"epoch": 0.46441697293685263,
"grad_norm": 1.015625,
"learning_rate": 0.0004983646268052567,
"loss": 5.1815,
"mean_token_accuracy": 0.18983226120471955,
"num_tokens": 10230336.0,
"step": 5560
},
{
"entropy": 5.450517988204956,
"epoch": 0.46483461409956567,
"grad_norm": 0.83203125,
"learning_rate": 0.0004983610420704931,
"loss": 5.3707,
"mean_token_accuracy": 0.18141137361526488,
"num_tokens": 10241655.0,
"step": 5565
},
{
"entropy": 5.348510074615478,
"epoch": 0.46525225526227865,
"grad_norm": 0.9609375,
"learning_rate": 0.0004983574534255256,
"loss": 5.2713,
"mean_token_accuracy": 0.18565696328878403,
"num_tokens": 10250926.0,
"step": 5570
},
{
"entropy": 5.3349449634552,
"epoch": 0.4656698964249916,
"grad_norm": 0.97265625,
"learning_rate": 0.000498353860870417,
"loss": 5.2191,
"mean_token_accuracy": 0.18590866774320602,
"num_tokens": 10260034.0,
"step": 5575
},
{
"entropy": 5.482433462142945,
"epoch": 0.46608753758770466,
"grad_norm": 1.078125,
"learning_rate": 0.0004983502644052303,
"loss": 5.3123,
"mean_token_accuracy": 0.1779246985912323,
"num_tokens": 10269170.0,
"step": 5580
},
{
"entropy": 5.5078449726104735,
"epoch": 0.46650517875041764,
"grad_norm": 1.03125,
"learning_rate": 0.0004983466640300282,
"loss": 5.2016,
"mean_token_accuracy": 0.18893269151449205,
"num_tokens": 10278385.0,
"step": 5585
},
{
"entropy": 5.4656421661376955,
"epoch": 0.4669228199131306,
"grad_norm": 0.9609375,
"learning_rate": 0.0004983430597448741,
"loss": 5.241,
"mean_token_accuracy": 0.19075634181499482,
"num_tokens": 10287678.0,
"step": 5590
},
{
"entropy": 5.403865623474121,
"epoch": 0.46734046107584365,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983394515498308,
"loss": 5.2327,
"mean_token_accuracy": 0.18925031572580336,
"num_tokens": 10296579.0,
"step": 5595
},
{
"entropy": 5.437282562255859,
"epoch": 0.46775810223855663,
"grad_norm": 0.95703125,
"learning_rate": 0.0004983358394449616,
"loss": 5.2735,
"mean_token_accuracy": 0.1898487016558647,
"num_tokens": 10305812.0,
"step": 5600
},
{
"entropy": 5.448018407821655,
"epoch": 0.4681757434012696,
"grad_norm": 0.8984375,
"learning_rate": 0.0004983322234303297,
"loss": 5.2729,
"mean_token_accuracy": 0.18164938688278198,
"num_tokens": 10314735.0,
"step": 5605
},
{
"entropy": 5.4351646423339846,
"epoch": 0.46859338456398264,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983286035059984,
"loss": 5.3636,
"mean_token_accuracy": 0.18386204689741134,
"num_tokens": 10323415.0,
"step": 5610
},
{
"entropy": 5.418152618408203,
"epoch": 0.4690110257266956,
"grad_norm": 1.078125,
"learning_rate": 0.0004983249796720311,
"loss": 5.2145,
"mean_token_accuracy": 0.1970824718475342,
"num_tokens": 10331670.0,
"step": 5615
},
{
"entropy": 5.456290531158447,
"epoch": 0.4694286668894086,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983213519284912,
"loss": 5.3228,
"mean_token_accuracy": 0.1823984920978546,
"num_tokens": 10340609.0,
"step": 5620
},
{
"entropy": 5.348497343063355,
"epoch": 0.46984630805212163,
"grad_norm": 1.015625,
"learning_rate": 0.0004983177202754421,
"loss": 5.0314,
"mean_token_accuracy": 0.2066277638077736,
"num_tokens": 10349035.0,
"step": 5625
},
{
"entropy": 5.330918788909912,
"epoch": 0.4702639492148346,
"grad_norm": 0.984375,
"learning_rate": 0.0004983140847129476,
"loss": 5.1219,
"mean_token_accuracy": 0.19656471014022828,
"num_tokens": 10358005.0,
"step": 5630
},
{
"entropy": 5.360467433929443,
"epoch": 0.4706815903775476,
"grad_norm": 1.0703125,
"learning_rate": 0.0004983104452410713,
"loss": 5.2602,
"mean_token_accuracy": 0.18540138155221939,
"num_tokens": 10366791.0,
"step": 5635
},
{
"entropy": 5.40042724609375,
"epoch": 0.4710992315402606,
"grad_norm": 0.98828125,
"learning_rate": 0.0004983068018598768,
"loss": 5.356,
"mean_token_accuracy": 0.1813494935631752,
"num_tokens": 10376485.0,
"step": 5640
},
{
"entropy": 5.396445274353027,
"epoch": 0.4715168727029736,
"grad_norm": 0.94921875,
"learning_rate": 0.000498303154569428,
"loss": 5.2309,
"mean_token_accuracy": 0.19248452484607698,
"num_tokens": 10385962.0,
"step": 5645
},
{
"entropy": 5.391742277145386,
"epoch": 0.4719345138656866,
"grad_norm": 0.96875,
"learning_rate": 0.0004982995033697887,
"loss": 5.2076,
"mean_token_accuracy": 0.19472643584012986,
"num_tokens": 10394314.0,
"step": 5650
},
{
"entropy": 5.394843530654907,
"epoch": 0.4723521550283996,
"grad_norm": 1.0546875,
"learning_rate": 0.0004982958482610226,
"loss": 5.194,
"mean_token_accuracy": 0.1960946276783943,
"num_tokens": 10403733.0,
"step": 5655
},
{
"entropy": 5.353224611282348,
"epoch": 0.4727697961911126,
"grad_norm": 0.953125,
"learning_rate": 0.000498292189243194,
"loss": 5.2987,
"mean_token_accuracy": 0.17926349639892578,
"num_tokens": 10413979.0,
"step": 5660
},
{
"entropy": 5.4293018817901615,
"epoch": 0.47318743735382557,
"grad_norm": 1.03125,
"learning_rate": 0.000498288526316367,
"loss": 5.1776,
"mean_token_accuracy": 0.19092903882265091,
"num_tokens": 10422630.0,
"step": 5665
},
{
"entropy": 5.391359853744507,
"epoch": 0.4736050785165386,
"grad_norm": 1.078125,
"learning_rate": 0.0004982848594806054,
"loss": 5.1785,
"mean_token_accuracy": 0.19484773129224778,
"num_tokens": 10431422.0,
"step": 5670
},
{
"entropy": 5.439520025253296,
"epoch": 0.4740227196792516,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982811887359736,
"loss": 5.2962,
"mean_token_accuracy": 0.1824900820851326,
"num_tokens": 10440953.0,
"step": 5675
},
{
"entropy": 5.3710533618927006,
"epoch": 0.47444036084196456,
"grad_norm": 1.015625,
"learning_rate": 0.0004982775140825359,
"loss": 5.2116,
"mean_token_accuracy": 0.1896195650100708,
"num_tokens": 10449331.0,
"step": 5680
},
{
"entropy": 5.394535493850708,
"epoch": 0.4748580020046776,
"grad_norm": 0.953125,
"learning_rate": 0.0004982738355203565,
"loss": 5.2097,
"mean_token_accuracy": 0.19098381251096724,
"num_tokens": 10458021.0,
"step": 5685
},
{
"entropy": 5.43800573348999,
"epoch": 0.47527564316739057,
"grad_norm": 0.95703125,
"learning_rate": 0.0004982701530494998,
"loss": 5.2131,
"mean_token_accuracy": 0.19253524988889695,
"num_tokens": 10466743.0,
"step": 5690
},
{
"entropy": 5.374477672576904,
"epoch": 0.47569328433010355,
"grad_norm": 1.0546875,
"learning_rate": 0.0004982664666700304,
"loss": 5.2137,
"mean_token_accuracy": 0.18603936731815338,
"num_tokens": 10475537.0,
"step": 5695
},
{
"entropy": 5.513244009017944,
"epoch": 0.4761109254928166,
"grad_norm": 1.046875,
"learning_rate": 0.0004982627763820128,
"loss": 5.3724,
"mean_token_accuracy": 0.1832273319363594,
"num_tokens": 10485257.0,
"step": 5700
},
{
"entropy": 5.525562906265259,
"epoch": 0.47652856665552956,
"grad_norm": 0.984375,
"learning_rate": 0.0004982590821855115,
"loss": 5.2968,
"mean_token_accuracy": 0.1894189313054085,
"num_tokens": 10494449.0,
"step": 5705
},
{
"entropy": 5.402962779998779,
"epoch": 0.47694620781824254,
"grad_norm": 0.8984375,
"learning_rate": 0.0004982553840805912,
"loss": 5.2778,
"mean_token_accuracy": 0.1898958593606949,
"num_tokens": 10504216.0,
"step": 5710
},
{
"entropy": 5.4967145919799805,
"epoch": 0.4773638489809556,
"grad_norm": 1.0625,
"learning_rate": 0.0004982516820673167,
"loss": 5.2703,
"mean_token_accuracy": 0.1894870176911354,
"num_tokens": 10512676.0,
"step": 5715
},
{
"entropy": 5.364314126968384,
"epoch": 0.47778149014366855,
"grad_norm": 1.0390625,
"learning_rate": 0.0004982479761457528,
"loss": 5.1251,
"mean_token_accuracy": 0.1935805857181549,
"num_tokens": 10521587.0,
"step": 5720
},
{
"entropy": 5.404315710067749,
"epoch": 0.47819913130638153,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982442663159643,
"loss": 5.2866,
"mean_token_accuracy": 0.18175797611474992,
"num_tokens": 10530043.0,
"step": 5725
},
{
"entropy": 5.368263149261475,
"epoch": 0.47861677246909456,
"grad_norm": 0.9765625,
"learning_rate": 0.0004982405525780162,
"loss": 5.1154,
"mean_token_accuracy": 0.19503370821475982,
"num_tokens": 10539491.0,
"step": 5730
},
{
"entropy": 5.354542064666748,
"epoch": 0.47903441363180754,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982368349319736,
"loss": 5.1194,
"mean_token_accuracy": 0.18833290040493011,
"num_tokens": 10547893.0,
"step": 5735
},
{
"entropy": 5.469902372360229,
"epoch": 0.4794520547945205,
"grad_norm": 0.9921875,
"learning_rate": 0.0004982331133779014,
"loss": 5.3036,
"mean_token_accuracy": 0.17868206202983855,
"num_tokens": 10557597.0,
"step": 5740
},
{
"entropy": 5.465248727798462,
"epoch": 0.47986969595723356,
"grad_norm": 1.0625,
"learning_rate": 0.0004982293879158649,
"loss": 5.2167,
"mean_token_accuracy": 0.1897121325135231,
"num_tokens": 10566349.0,
"step": 5745
},
{
"entropy": 5.403199195861816,
"epoch": 0.48028733711994653,
"grad_norm": 1.015625,
"learning_rate": 0.0004982256585459293,
"loss": 5.2872,
"mean_token_accuracy": 0.18665742874145508,
"num_tokens": 10574897.0,
"step": 5750
},
{
"entropy": 5.447471380233765,
"epoch": 0.4807049782826595,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982219252681597,
"loss": 5.2176,
"mean_token_accuracy": 0.18940539360046388,
"num_tokens": 10583233.0,
"step": 5755
},
{
"entropy": 5.332688522338867,
"epoch": 0.48112261944537255,
"grad_norm": 1.0,
"learning_rate": 0.0004982181880826218,
"loss": 5.1276,
"mean_token_accuracy": 0.1920537531375885,
"num_tokens": 10592202.0,
"step": 5760
},
{
"entropy": 5.337571334838867,
"epoch": 0.4815402606080855,
"grad_norm": 0.89453125,
"learning_rate": 0.0004982144469893807,
"loss": 5.231,
"mean_token_accuracy": 0.19041890501976014,
"num_tokens": 10601221.0,
"step": 5765
},
{
"entropy": 5.455085945129395,
"epoch": 0.4819579017707985,
"grad_norm": 0.88671875,
"learning_rate": 0.000498210701988502,
"loss": 5.2974,
"mean_token_accuracy": 0.18375370800495147,
"num_tokens": 10610983.0,
"step": 5770
},
{
"entropy": 5.457437992095947,
"epoch": 0.48237554293351154,
"grad_norm": 0.9140625,
"learning_rate": 0.0004982069530800514,
"loss": 5.3049,
"mean_token_accuracy": 0.18623082041740419,
"num_tokens": 10619378.0,
"step": 5775
},
{
"entropy": 5.470373058319092,
"epoch": 0.4827931840962245,
"grad_norm": 0.9375,
"learning_rate": 0.0004982032002640943,
"loss": 5.1988,
"mean_token_accuracy": 0.19191256910562515,
"num_tokens": 10628288.0,
"step": 5780
},
{
"entropy": 5.457940578460693,
"epoch": 0.48321082525893755,
"grad_norm": 0.8984375,
"learning_rate": 0.0004981994435406965,
"loss": 5.3574,
"mean_token_accuracy": 0.17868073433637618,
"num_tokens": 10639433.0,
"step": 5785
},
{
"entropy": 5.347267723083496,
"epoch": 0.4836284664216505,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981956829099239,
"loss": 5.1572,
"mean_token_accuracy": 0.18723085671663284,
"num_tokens": 10649654.0,
"step": 5790
},
{
"entropy": 5.403138923645019,
"epoch": 0.4840461075843635,
"grad_norm": 0.8984375,
"learning_rate": 0.0004981919183718422,
"loss": 5.2951,
"mean_token_accuracy": 0.18621639013290406,
"num_tokens": 10658620.0,
"step": 5795
},
{
"entropy": 5.3692464351654055,
"epoch": 0.48446374874707654,
"grad_norm": 0.9765625,
"learning_rate": 0.0004981881499265173,
"loss": 5.1033,
"mean_token_accuracy": 0.19129590839147567,
"num_tokens": 10667942.0,
"step": 5800
},
{
"entropy": 5.305265188217163,
"epoch": 0.4848813899097895,
"grad_norm": 0.9296875,
"learning_rate": 0.0004981843775740151,
"loss": 5.1846,
"mean_token_accuracy": 0.18814138174057007,
"num_tokens": 10676895.0,
"step": 5805
},
{
"entropy": 5.496080636978149,
"epoch": 0.4852990310725025,
"grad_norm": 0.9921875,
"learning_rate": 0.0004981806013144017,
"loss": 5.3384,
"mean_token_accuracy": 0.18187302947044373,
"num_tokens": 10685986.0,
"step": 5810
},
{
"entropy": 5.4939916133880615,
"epoch": 0.48571667223521553,
"grad_norm": 0.9921875,
"learning_rate": 0.0004981768211477434,
"loss": 5.2342,
"mean_token_accuracy": 0.19719823598861694,
"num_tokens": 10694430.0,
"step": 5815
},
{
"entropy": 5.368886184692383,
"epoch": 0.4861343133979285,
"grad_norm": 0.97265625,
"learning_rate": 0.000498173037074106,
"loss": 5.1212,
"mean_token_accuracy": 0.19116412997245788,
"num_tokens": 10702997.0,
"step": 5820
},
{
"entropy": 5.374971628189087,
"epoch": 0.4865519545606415,
"grad_norm": 0.98046875,
"learning_rate": 0.0004981692490935561,
"loss": 5.3167,
"mean_token_accuracy": 0.18619992285966874,
"num_tokens": 10711652.0,
"step": 5825
},
{
"entropy": 5.4620719909667965,
"epoch": 0.4869695957233545,
"grad_norm": 0.94921875,
"learning_rate": 0.0004981654572061597,
"loss": 5.293,
"mean_token_accuracy": 0.18306376338005065,
"num_tokens": 10721749.0,
"step": 5830
},
{
"entropy": 5.36485071182251,
"epoch": 0.4873872368860675,
"grad_norm": 1.0,
"learning_rate": 0.0004981616614119836,
"loss": 5.229,
"mean_token_accuracy": 0.18836989849805832,
"num_tokens": 10729678.0,
"step": 5835
},
{
"entropy": 5.4476886749267575,
"epoch": 0.4878048780487805,
"grad_norm": 0.97265625,
"learning_rate": 0.0004981578617110938,
"loss": 5.2584,
"mean_token_accuracy": 0.18365153819322586,
"num_tokens": 10740276.0,
"step": 5840
},
{
"entropy": 5.3470542430877686,
"epoch": 0.4882225192114935,
"grad_norm": 0.91015625,
"learning_rate": 0.0004981540581035571,
"loss": 5.2722,
"mean_token_accuracy": 0.18048646599054335,
"num_tokens": 10749249.0,
"step": 5845
},
{
"entropy": 5.426681041717529,
"epoch": 0.4886401603742065,
"grad_norm": 0.890625,
"learning_rate": 0.00049815025058944,
"loss": 5.1276,
"mean_token_accuracy": 0.19819484800100326,
"num_tokens": 10758888.0,
"step": 5850
},
{
"entropy": 5.296361207962036,
"epoch": 0.48905780153691947,
"grad_norm": 1.0546875,
"learning_rate": 0.0004981464391688092,
"loss": 5.2043,
"mean_token_accuracy": 0.19552705734968184,
"num_tokens": 10768156.0,
"step": 5855
},
{
"entropy": 5.347362995147705,
"epoch": 0.4894754426996325,
"grad_norm": 1.0859375,
"learning_rate": 0.0004981426238417312,
"loss": 5.1635,
"mean_token_accuracy": 0.20008436739444732,
"num_tokens": 10776762.0,
"step": 5860
},
{
"entropy": 5.441961288452148,
"epoch": 0.4898930838623455,
"grad_norm": 0.96875,
"learning_rate": 0.0004981388046082731,
"loss": 5.2665,
"mean_token_accuracy": 0.17891191840171813,
"num_tokens": 10785495.0,
"step": 5865
},
{
"entropy": 5.368473482131958,
"epoch": 0.49031072502505846,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981349814685016,
"loss": 5.1282,
"mean_token_accuracy": 0.191314297914505,
"num_tokens": 10794400.0,
"step": 5870
},
{
"entropy": 5.359164381027222,
"epoch": 0.4907283661877715,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981311544224838,
"loss": 5.1942,
"mean_token_accuracy": 0.19437927901744842,
"num_tokens": 10803138.0,
"step": 5875
},
{
"entropy": 5.462613439559936,
"epoch": 0.49114600735048447,
"grad_norm": 0.94140625,
"learning_rate": 0.0004981273234702864,
"loss": 5.2361,
"mean_token_accuracy": 0.18240292370319366,
"num_tokens": 10813583.0,
"step": 5880
},
{
"entropy": 5.348620653152466,
"epoch": 0.49156364851319745,
"grad_norm": 0.93359375,
"learning_rate": 0.0004981234886119767,
"loss": 5.1446,
"mean_token_accuracy": 0.19907235205173493,
"num_tokens": 10822532.0,
"step": 5885
},
{
"entropy": 5.289391088485718,
"epoch": 0.4919812896759105,
"grad_norm": 0.98828125,
"learning_rate": 0.0004981196498476217,
"loss": 5.2331,
"mean_token_accuracy": 0.19357739835977555,
"num_tokens": 10832256.0,
"step": 5890
},
{
"entropy": 5.422555160522461,
"epoch": 0.49239893083862346,
"grad_norm": 0.98828125,
"learning_rate": 0.0004981158071772887,
"loss": 5.1194,
"mean_token_accuracy": 0.1932843953371048,
"num_tokens": 10840909.0,
"step": 5895
},
{
"entropy": 5.400418472290039,
"epoch": 0.49281657200133644,
"grad_norm": 0.9453125,
"learning_rate": 0.0004981119606010449,
"loss": 5.1791,
"mean_token_accuracy": 0.19052398800849915,
"num_tokens": 10849931.0,
"step": 5900
},
{
"entropy": 5.398821496963501,
"epoch": 0.4932342131640495,
"grad_norm": 0.92578125,
"learning_rate": 0.0004981081101189577,
"loss": 5.2242,
"mean_token_accuracy": 0.190839883685112,
"num_tokens": 10859518.0,
"step": 5905
},
{
"entropy": 5.397838973999024,
"epoch": 0.49365185432676245,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981042557310945,
"loss": 5.3032,
"mean_token_accuracy": 0.18702766299247742,
"num_tokens": 10868567.0,
"step": 5910
},
{
"entropy": 5.270372676849365,
"epoch": 0.49406949548947543,
"grad_norm": 0.8984375,
"learning_rate": 0.0004981003974375227,
"loss": 5.0597,
"mean_token_accuracy": 0.19807375967502594,
"num_tokens": 10877463.0,
"step": 5915
},
{
"entropy": 5.339721965789795,
"epoch": 0.49448713665218846,
"grad_norm": 0.94140625,
"learning_rate": 0.0004980965352383099,
"loss": 5.1939,
"mean_token_accuracy": 0.1899080440402031,
"num_tokens": 10886228.0,
"step": 5920
},
{
"entropy": 5.493758630752564,
"epoch": 0.49490477781490144,
"grad_norm": 0.9375,
"learning_rate": 0.0004980926691335238,
"loss": 5.2877,
"mean_token_accuracy": 0.18432832062244414,
"num_tokens": 10895415.0,
"step": 5925
},
{
"entropy": 5.414480495452881,
"epoch": 0.4953224189776144,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980887991232318,
"loss": 5.2015,
"mean_token_accuracy": 0.1809064745903015,
"num_tokens": 10904146.0,
"step": 5930
},
{
"entropy": 5.369438552856446,
"epoch": 0.49574006014032745,
"grad_norm": 0.93359375,
"learning_rate": 0.000498084925207502,
"loss": 5.235,
"mean_token_accuracy": 0.19040572196245192,
"num_tokens": 10914027.0,
"step": 5935
},
{
"entropy": 5.367545557022095,
"epoch": 0.49615770130304043,
"grad_norm": 0.91796875,
"learning_rate": 0.000498081047386402,
"loss": 5.1854,
"mean_token_accuracy": 0.19287265688180924,
"num_tokens": 10923065.0,
"step": 5940
},
{
"entropy": 5.493957376480102,
"epoch": 0.4965753424657534,
"grad_norm": 0.89453125,
"learning_rate": 0.0004980771656599998,
"loss": 5.3274,
"mean_token_accuracy": 0.17854325026273726,
"num_tokens": 10932752.0,
"step": 5945
},
{
"entropy": 5.504323673248291,
"epoch": 0.49699298362846644,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980732800283633,
"loss": 5.2632,
"mean_token_accuracy": 0.1884702205657959,
"num_tokens": 10942536.0,
"step": 5950
},
{
"entropy": 5.376228141784668,
"epoch": 0.4974106247911794,
"grad_norm": 0.9375,
"learning_rate": 0.0004980693904915604,
"loss": 5.1704,
"mean_token_accuracy": 0.18781334161758423,
"num_tokens": 10952258.0,
"step": 5955
},
{
"entropy": 5.376414632797241,
"epoch": 0.4978282659538924,
"grad_norm": 0.95703125,
"learning_rate": 0.0004980654970496595,
"loss": 5.2256,
"mean_token_accuracy": 0.18580455482006072,
"num_tokens": 10961590.0,
"step": 5960
},
{
"entropy": 5.405714321136474,
"epoch": 0.49824590711660544,
"grad_norm": 0.921875,
"learning_rate": 0.0004980615997027284,
"loss": 5.184,
"mean_token_accuracy": 0.19644657373428345,
"num_tokens": 10971380.0,
"step": 5965
},
{
"entropy": 5.374588346481323,
"epoch": 0.4986635482793184,
"grad_norm": 1.0390625,
"learning_rate": 0.0004980576984508356,
"loss": 5.1929,
"mean_token_accuracy": 0.1910658985376358,
"num_tokens": 10979510.0,
"step": 5970
},
{
"entropy": 5.364955043792724,
"epoch": 0.4990811894420314,
"grad_norm": 0.94921875,
"learning_rate": 0.0004980537932940492,
"loss": 5.199,
"mean_token_accuracy": 0.18778235465288162,
"num_tokens": 10988921.0,
"step": 5975
},
{
"entropy": 5.428235054016113,
"epoch": 0.4994988306047444,
"grad_norm": 0.875,
"learning_rate": 0.0004980498842324377,
"loss": 5.1811,
"mean_token_accuracy": 0.19132654815912248,
"num_tokens": 10998644.0,
"step": 5980
},
{
"entropy": 5.309868288040161,
"epoch": 0.4999164717674574,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980459712660694,
"loss": 5.0925,
"mean_token_accuracy": 0.19584756791591645,
"num_tokens": 11006975.0,
"step": 5985
},
{
"entropy": 5.3369608402252195,
"epoch": 0.5003341129301704,
"grad_norm": 0.96484375,
"learning_rate": 0.0004980420543950131,
"loss": 5.1628,
"mean_token_accuracy": 0.1886674866080284,
"num_tokens": 11016011.0,
"step": 5990
},
{
"entropy": 5.51916651725769,
"epoch": 0.5007517540928834,
"grad_norm": 0.9375,
"learning_rate": 0.0004980381336193371,
"loss": 5.363,
"mean_token_accuracy": 0.17856939733028412,
"num_tokens": 11025600.0,
"step": 5995
},
{
"entropy": 5.369285678863525,
"epoch": 0.5011693952555963,
"grad_norm": 0.96875,
"learning_rate": 0.0004980342089391099,
"loss": 5.1041,
"mean_token_accuracy": 0.19820183664560317,
"num_tokens": 11034747.0,
"step": 6000
},
{
"epoch": 0.5011693952555963,
"eval_entropy": 5.074534042486067,
"eval_loss": 5.171754360198975,
"eval_mean_token_accuracy": 0.20148191507373087,
"eval_num_tokens": 11034747.0,
"eval_runtime": 45.7091,
"eval_samples_per_second": 826.159,
"eval_steps_per_second": 103.284,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 119720,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2419717884641280.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink