8035 lines
220 KiB
JSON
8035 lines
220 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.04298625514491741,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692043209075928,
|
|
"epoch": 5.3732818931146765e-05,
|
|
"grad_norm": 12.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8501,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12012.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69202766418457,
|
|
"epoch": 0.00010746563786229353,
|
|
"grad_norm": 14.3125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7855,
|
|
"mean_token_accuracy": 0.00020087850280106067,
|
|
"num_tokens": 23548.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.69138708114624,
|
|
"epoch": 0.0001611984567934403,
|
|
"grad_norm": 9.8125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5161,
|
|
"mean_token_accuracy": 0.016332162456819788,
|
|
"num_tokens": 35237.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.685348415374756,
|
|
"epoch": 0.00021493127572458706,
|
|
"grad_norm": 6.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1997,
|
|
"mean_token_accuracy": 0.035921670868992804,
|
|
"num_tokens": 47773.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.647090339660645,
|
|
"epoch": 0.00026866409465573383,
|
|
"grad_norm": 4.59375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8165,
|
|
"mean_token_accuracy": 0.04100495874881745,
|
|
"num_tokens": 59398.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.562076568603516,
|
|
"epoch": 0.0003223969135868806,
|
|
"grad_norm": 3.328125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6944,
|
|
"mean_token_accuracy": 0.03505560103803873,
|
|
"num_tokens": 70474.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.566006851196288,
|
|
"epoch": 0.00037612973251802736,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6543,
|
|
"mean_token_accuracy": 0.03815589081496,
|
|
"num_tokens": 81441.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.595672798156738,
|
|
"epoch": 0.0004298625514491741,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6209,
|
|
"mean_token_accuracy": 0.0424302164465189,
|
|
"num_tokens": 92819.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.57416124343872,
|
|
"epoch": 0.0004835953703803209,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.477,
|
|
"mean_token_accuracy": 0.04498981647193432,
|
|
"num_tokens": 105565.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.500903701782226,
|
|
"epoch": 0.0005373281893114677,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4751,
|
|
"mean_token_accuracy": 0.03739957753568888,
|
|
"num_tokens": 116424.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.49787940979004,
|
|
"epoch": 0.0005910610082426144,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4268,
|
|
"mean_token_accuracy": 0.0429843544960022,
|
|
"num_tokens": 127807.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.53837537765503,
|
|
"epoch": 0.0006447938271737612,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.2854,
|
|
"mean_token_accuracy": 0.05508757047355175,
|
|
"num_tokens": 139177.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.40587682723999,
|
|
"epoch": 0.000698526646104908,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2319,
|
|
"mean_token_accuracy": 0.05228099822998047,
|
|
"num_tokens": 151296.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.370543670654296,
|
|
"epoch": 0.0007522594650360547,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.0806,
|
|
"mean_token_accuracy": 0.07307359352707862,
|
|
"num_tokens": 160488.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.382234954833985,
|
|
"epoch": 0.0008059922839672015,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.0574,
|
|
"mean_token_accuracy": 0.06776703521609306,
|
|
"num_tokens": 171627.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.367051887512208,
|
|
"epoch": 0.0008597251028983482,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.8925,
|
|
"mean_token_accuracy": 0.07181778848171234,
|
|
"num_tokens": 182978.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.235933113098145,
|
|
"epoch": 0.000913457921829495,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.7913,
|
|
"mean_token_accuracy": 0.07224572598934173,
|
|
"num_tokens": 193362.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.180474948883056,
|
|
"epoch": 0.0009671907407606418,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7247,
|
|
"mean_token_accuracy": 0.07832827009260654,
|
|
"num_tokens": 204896.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.17861442565918,
|
|
"epoch": 0.0010209235596917885,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6001,
|
|
"mean_token_accuracy": 0.07487386986613273,
|
|
"num_tokens": 215828.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.129089546203613,
|
|
"epoch": 0.0010746563786229353,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.454,
|
|
"mean_token_accuracy": 0.0813417136669159,
|
|
"num_tokens": 226311.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.016890907287598,
|
|
"epoch": 0.001128389197554082,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.3892,
|
|
"mean_token_accuracy": 0.07950438261032104,
|
|
"num_tokens": 237817.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.948394012451171,
|
|
"epoch": 0.0011821220164852288,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.2654,
|
|
"mean_token_accuracy": 0.080025664716959,
|
|
"num_tokens": 248778.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.865277194976807,
|
|
"epoch": 0.0012358548354163756,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2215,
|
|
"mean_token_accuracy": 0.07951191291213036,
|
|
"num_tokens": 260638.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.790133190155029,
|
|
"epoch": 0.0012895876543475224,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.0775,
|
|
"mean_token_accuracy": 0.0801257323473692,
|
|
"num_tokens": 272223.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.634031391143798,
|
|
"epoch": 0.0013433204732786691,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.8657,
|
|
"mean_token_accuracy": 0.09108843877911568,
|
|
"num_tokens": 282883.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.479462718963623,
|
|
"epoch": 0.001397053292209816,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8083,
|
|
"mean_token_accuracy": 0.08633529022336006,
|
|
"num_tokens": 293318.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.311678791046143,
|
|
"epoch": 0.0014507861111409627,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.6502,
|
|
"mean_token_accuracy": 0.08342773504555226,
|
|
"num_tokens": 303563.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.093393898010254,
|
|
"epoch": 0.0015045189300721094,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.5358,
|
|
"mean_token_accuracy": 0.09152235686779023,
|
|
"num_tokens": 313968.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.970313549041748,
|
|
"epoch": 0.0015582517490032562,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5692,
|
|
"mean_token_accuracy": 0.08849354460835457,
|
|
"num_tokens": 325261.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.724966049194336,
|
|
"epoch": 0.001611984567934403,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.3732,
|
|
"mean_token_accuracy": 0.09316953867673874,
|
|
"num_tokens": 335803.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.553716468811036,
|
|
"epoch": 0.0016657173868655497,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.2783,
|
|
"mean_token_accuracy": 0.08820799067616462,
|
|
"num_tokens": 346582.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.43663501739502,
|
|
"epoch": 0.0017194502057966965,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3562,
|
|
"mean_token_accuracy": 0.09117907658219337,
|
|
"num_tokens": 357779.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.219385528564453,
|
|
"epoch": 0.0017731830247278433,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2185,
|
|
"mean_token_accuracy": 0.09431424736976624,
|
|
"num_tokens": 369191.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.094470024108887,
|
|
"epoch": 0.00182691584365899,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2579,
|
|
"mean_token_accuracy": 0.09085006713867187,
|
|
"num_tokens": 381374.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.00999412536621,
|
|
"epoch": 0.0018806486625901368,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.2359,
|
|
"mean_token_accuracy": 0.09357822686433792,
|
|
"num_tokens": 392152.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.9514116764068605,
|
|
"epoch": 0.0019343814815212836,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2057,
|
|
"mean_token_accuracy": 0.09208550006151199,
|
|
"num_tokens": 404121.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.96669111251831,
|
|
"epoch": 0.0019881143004524303,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.2056,
|
|
"mean_token_accuracy": 0.0923257291316986,
|
|
"num_tokens": 416144.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.824949645996094,
|
|
"epoch": 0.002041847119383577,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.0888,
|
|
"mean_token_accuracy": 0.09763017818331718,
|
|
"num_tokens": 427141.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.715742874145508,
|
|
"epoch": 0.002095579938314724,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.0035,
|
|
"mean_token_accuracy": 0.0935081422328949,
|
|
"num_tokens": 439113.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.7036233901977536,
|
|
"epoch": 0.0021493127572458706,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0734,
|
|
"mean_token_accuracy": 0.09741963669657708,
|
|
"num_tokens": 450595.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.739162731170654,
|
|
"epoch": 0.0022030455761770174,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1579,
|
|
"mean_token_accuracy": 0.09232115969061852,
|
|
"num_tokens": 462811.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.705861139297485,
|
|
"epoch": 0.002256778395108164,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1276,
|
|
"mean_token_accuracy": 0.09830137938261033,
|
|
"num_tokens": 473629.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.643820524215698,
|
|
"epoch": 0.002310511214039311,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1024,
|
|
"mean_token_accuracy": 0.0988999992609024,
|
|
"num_tokens": 484731.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.66978874206543,
|
|
"epoch": 0.0023642440329704577,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0548,
|
|
"mean_token_accuracy": 0.09935099259018898,
|
|
"num_tokens": 496309.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.658147382736206,
|
|
"epoch": 0.0024179768519016044,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.0923,
|
|
"mean_token_accuracy": 0.1015954852104187,
|
|
"num_tokens": 508715.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.41760745048523,
|
|
"epoch": 0.002471709670832751,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.9405,
|
|
"mean_token_accuracy": 0.1079390250146389,
|
|
"num_tokens": 519791.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.54394679069519,
|
|
"epoch": 0.002525442489763898,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.083,
|
|
"mean_token_accuracy": 0.09995704516768456,
|
|
"num_tokens": 531897.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.555436658859253,
|
|
"epoch": 0.0025791753086950447,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0798,
|
|
"mean_token_accuracy": 0.09950428679585457,
|
|
"num_tokens": 543243.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.442711400985718,
|
|
"epoch": 0.0026329081276261915,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.9189,
|
|
"mean_token_accuracy": 0.10362667217850685,
|
|
"num_tokens": 554139.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.480173254013062,
|
|
"epoch": 0.0026866409465573383,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9273,
|
|
"mean_token_accuracy": 0.10886659622192382,
|
|
"num_tokens": 565378.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.426122951507568,
|
|
"epoch": 0.002740373765488485,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.881,
|
|
"mean_token_accuracy": 0.10440984815359115,
|
|
"num_tokens": 576628.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.444299507141113,
|
|
"epoch": 0.002794106584419632,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0019,
|
|
"mean_token_accuracy": 0.09789407551288605,
|
|
"num_tokens": 588760.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.4239896774292,
|
|
"epoch": 0.0028478394033507786,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9158,
|
|
"mean_token_accuracy": 0.10981944426894188,
|
|
"num_tokens": 600024.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.270383596420288,
|
|
"epoch": 0.0029015722222819253,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.8775,
|
|
"mean_token_accuracy": 0.10582949295639991,
|
|
"num_tokens": 613150.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.346567535400391,
|
|
"epoch": 0.002955305041213072,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.10970969945192337,
|
|
"num_tokens": 624762.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.417334127426147,
|
|
"epoch": 0.003009037860144219,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.892,
|
|
"mean_token_accuracy": 0.10366739556193352,
|
|
"num_tokens": 635761.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.223995733261108,
|
|
"epoch": 0.0030627706790753656,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8555,
|
|
"mean_token_accuracy": 0.10360193029046058,
|
|
"num_tokens": 647882.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.312860345840454,
|
|
"epoch": 0.0031165034980065124,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8848,
|
|
"mean_token_accuracy": 0.1077139988541603,
|
|
"num_tokens": 659749.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.270877933502197,
|
|
"epoch": 0.003170236316937659,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8362,
|
|
"mean_token_accuracy": 0.10846684277057647,
|
|
"num_tokens": 672002.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.204770469665528,
|
|
"epoch": 0.003223969135868806,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.11234059333801269,
|
|
"num_tokens": 682312.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.183935689926147,
|
|
"epoch": 0.0032777019547999527,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8246,
|
|
"mean_token_accuracy": 0.10333632677793503,
|
|
"num_tokens": 692188.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.269667243957519,
|
|
"epoch": 0.0033314347737310995,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.7992,
|
|
"mean_token_accuracy": 0.10754085555672646,
|
|
"num_tokens": 703789.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.259969711303711,
|
|
"epoch": 0.0033851675926622462,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8807,
|
|
"mean_token_accuracy": 0.11048192977905273,
|
|
"num_tokens": 714752.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.137263011932373,
|
|
"epoch": 0.003438900411593393,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.8206,
|
|
"mean_token_accuracy": 0.11233628690242767,
|
|
"num_tokens": 726716.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.213597440719605,
|
|
"epoch": 0.0034926332305245398,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8401,
|
|
"mean_token_accuracy": 0.11278397589921951,
|
|
"num_tokens": 737267.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.222741270065308,
|
|
"epoch": 0.0035463660494556865,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8783,
|
|
"mean_token_accuracy": 0.10917454287409782,
|
|
"num_tokens": 750180.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.1152294158935545,
|
|
"epoch": 0.0036000988683868333,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.6925,
|
|
"mean_token_accuracy": 0.11422673463821412,
|
|
"num_tokens": 761337.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.07347264289856,
|
|
"epoch": 0.00365383168731798,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.6614,
|
|
"mean_token_accuracy": 0.1183428131043911,
|
|
"num_tokens": 771379.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.18135552406311,
|
|
"epoch": 0.003707564506249127,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7586,
|
|
"mean_token_accuracy": 0.11464423760771751,
|
|
"num_tokens": 782508.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.083883142471313,
|
|
"epoch": 0.0037612973251802736,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.766,
|
|
"mean_token_accuracy": 0.1096316896378994,
|
|
"num_tokens": 795541.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.03632378578186,
|
|
"epoch": 0.0038150301441114204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7017,
|
|
"mean_token_accuracy": 0.11446580439805984,
|
|
"num_tokens": 805665.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.182166957855225,
|
|
"epoch": 0.003868762963042567,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8425,
|
|
"mean_token_accuracy": 0.11153884828090668,
|
|
"num_tokens": 817393.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.163691473007202,
|
|
"epoch": 0.003922495781973714,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7986,
|
|
"mean_token_accuracy": 0.10896589532494545,
|
|
"num_tokens": 828436.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.033854150772095,
|
|
"epoch": 0.003976228600904861,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.653,
|
|
"mean_token_accuracy": 0.11405793353915214,
|
|
"num_tokens": 839223.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.135569000244141,
|
|
"epoch": 0.004029961419836007,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6199,
|
|
"mean_token_accuracy": 0.12831516563892365,
|
|
"num_tokens": 848953.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.988350915908813,
|
|
"epoch": 0.004083694238767154,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7725,
|
|
"mean_token_accuracy": 0.10901687815785407,
|
|
"num_tokens": 860349.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.042834091186523,
|
|
"epoch": 0.004137427057698301,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6875,
|
|
"mean_token_accuracy": 0.11479247137904167,
|
|
"num_tokens": 870993.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.062792253494263,
|
|
"epoch": 0.004191159876629448,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7484,
|
|
"mean_token_accuracy": 0.11023281887173653,
|
|
"num_tokens": 880950.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.0711596488952635,
|
|
"epoch": 0.0042448926955605945,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.11763814464211464,
|
|
"num_tokens": 892583.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.909264755249024,
|
|
"epoch": 0.004298625514491741,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5686,
|
|
"mean_token_accuracy": 0.12108325287699699,
|
|
"num_tokens": 902191.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.913547658920288,
|
|
"epoch": 0.004352358333422888,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6222,
|
|
"mean_token_accuracy": 0.11458350345492363,
|
|
"num_tokens": 914274.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.911238813400269,
|
|
"epoch": 0.004406091152354035,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5556,
|
|
"mean_token_accuracy": 0.11596038416028023,
|
|
"num_tokens": 925253.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.00830454826355,
|
|
"epoch": 0.0044598239712851815,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7087,
|
|
"mean_token_accuracy": 0.11384080052375793,
|
|
"num_tokens": 935520.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.927305507659912,
|
|
"epoch": 0.004513556790216328,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6906,
|
|
"mean_token_accuracy": 0.11692763790488243,
|
|
"num_tokens": 947817.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.889912366867065,
|
|
"epoch": 0.004567289609147475,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6066,
|
|
"mean_token_accuracy": 0.11778534352779388,
|
|
"num_tokens": 959474.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.939700126647949,
|
|
"epoch": 0.004621022428078622,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5916,
|
|
"mean_token_accuracy": 0.11906479597091675,
|
|
"num_tokens": 971271.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.882960844039917,
|
|
"epoch": 0.004674755247009769,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.612,
|
|
"mean_token_accuracy": 0.12181256487965583,
|
|
"num_tokens": 981995.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.9154595851898195,
|
|
"epoch": 0.004728488065940915,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6914,
|
|
"mean_token_accuracy": 0.11828593611717224,
|
|
"num_tokens": 994346.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.966398048400879,
|
|
"epoch": 0.004782220884872062,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6802,
|
|
"mean_token_accuracy": 0.1137112945318222,
|
|
"num_tokens": 1005552.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.843932342529297,
|
|
"epoch": 0.004835953703803209,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6247,
|
|
"mean_token_accuracy": 0.11204231306910514,
|
|
"num_tokens": 1017202.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.807126045227051,
|
|
"epoch": 0.004889686522734356,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5839,
|
|
"mean_token_accuracy": 0.12199563458561898,
|
|
"num_tokens": 1027450.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.89690580368042,
|
|
"epoch": 0.004943419341665502,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5637,
|
|
"mean_token_accuracy": 0.11613315865397453,
|
|
"num_tokens": 1038990.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.806377410888672,
|
|
"epoch": 0.004997152160596649,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5786,
|
|
"mean_token_accuracy": 0.11820394694805145,
|
|
"num_tokens": 1050643.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.754554319381714,
|
|
"epoch": 0.005050884979527796,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5004,
|
|
"mean_token_accuracy": 0.12497379332780838,
|
|
"num_tokens": 1061182.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.9261133670806885,
|
|
"epoch": 0.005104617798458943,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6711,
|
|
"mean_token_accuracy": 0.11877285093069076,
|
|
"num_tokens": 1073074.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.839569091796875,
|
|
"epoch": 0.0051583506173900895,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6933,
|
|
"mean_token_accuracy": 0.1182055912911892,
|
|
"num_tokens": 1085061.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.822368907928467,
|
|
"epoch": 0.005212083436321236,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5495,
|
|
"mean_token_accuracy": 0.11974000334739685,
|
|
"num_tokens": 1096148.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.738174152374268,
|
|
"epoch": 0.005265816255252383,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4834,
|
|
"mean_token_accuracy": 0.12004421576857567,
|
|
"num_tokens": 1106159.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.8385029315948485,
|
|
"epoch": 0.00531954907418353,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6323,
|
|
"mean_token_accuracy": 0.12007313966751099,
|
|
"num_tokens": 1117159.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.834402704238892,
|
|
"epoch": 0.0053732818931146765,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5295,
|
|
"mean_token_accuracy": 0.12340744212269783,
|
|
"num_tokens": 1129633.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.888376379013062,
|
|
"epoch": 0.005427014712045823,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.12026305049657822,
|
|
"num_tokens": 1140814.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.8587806701660154,
|
|
"epoch": 0.00548074753097697,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6052,
|
|
"mean_token_accuracy": 0.12095662355422973,
|
|
"num_tokens": 1152963.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.571197032928467,
|
|
"epoch": 0.005534480349908117,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4367,
|
|
"mean_token_accuracy": 0.1233276754617691,
|
|
"num_tokens": 1165092.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.769173955917358,
|
|
"epoch": 0.005588213168839264,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5331,
|
|
"mean_token_accuracy": 0.1194583684206009,
|
|
"num_tokens": 1177403.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.709192943572998,
|
|
"epoch": 0.00564194598777041,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5345,
|
|
"mean_token_accuracy": 0.11615289524197578,
|
|
"num_tokens": 1187436.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.793780326843262,
|
|
"epoch": 0.005695678806701557,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6191,
|
|
"mean_token_accuracy": 0.11863412484526634,
|
|
"num_tokens": 1199906.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.816549777984619,
|
|
"epoch": 0.005749411625632704,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5816,
|
|
"mean_token_accuracy": 0.1212506726384163,
|
|
"num_tokens": 1211411.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.711446523666382,
|
|
"epoch": 0.005803144444563851,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5452,
|
|
"mean_token_accuracy": 0.12302517369389535,
|
|
"num_tokens": 1223669.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.627863168716431,
|
|
"epoch": 0.0058568772634949974,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.12178474590182305,
|
|
"num_tokens": 1234931.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.712806463241577,
|
|
"epoch": 0.005910610082426144,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4942,
|
|
"mean_token_accuracy": 0.12170209959149361,
|
|
"num_tokens": 1245731.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.725653505325317,
|
|
"epoch": 0.005964342901357291,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4796,
|
|
"mean_token_accuracy": 0.12272765785455704,
|
|
"num_tokens": 1257441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.729346084594726,
|
|
"epoch": 0.006018075720288438,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5136,
|
|
"mean_token_accuracy": 0.12274165153503418,
|
|
"num_tokens": 1269169.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.632570362091064,
|
|
"epoch": 0.0060718085392195845,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4972,
|
|
"mean_token_accuracy": 0.12501644492149352,
|
|
"num_tokens": 1280636.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.588498926162719,
|
|
"epoch": 0.006125541358150731,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4233,
|
|
"mean_token_accuracy": 0.1268314391374588,
|
|
"num_tokens": 1292266.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.6426673412323,
|
|
"epoch": 0.006179274177081878,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.12650827020406724,
|
|
"num_tokens": 1304090.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.662691974639893,
|
|
"epoch": 0.006233006996013025,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5546,
|
|
"mean_token_accuracy": 0.12577759027481078,
|
|
"num_tokens": 1314758.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.787952995300293,
|
|
"epoch": 0.0062867398149441716,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.528,
|
|
"mean_token_accuracy": 0.1216070830821991,
|
|
"num_tokens": 1326085.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.702496099472046,
|
|
"epoch": 0.006340472633875318,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4632,
|
|
"mean_token_accuracy": 0.12628382220864295,
|
|
"num_tokens": 1336766.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.507978296279907,
|
|
"epoch": 0.006394205452806465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.384,
|
|
"mean_token_accuracy": 0.12896264716982841,
|
|
"num_tokens": 1346864.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.745914697647095,
|
|
"epoch": 0.006447938271737612,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4981,
|
|
"mean_token_accuracy": 0.1272559054195881,
|
|
"num_tokens": 1357429.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.573749780654907,
|
|
"epoch": 0.006501671090668759,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3802,
|
|
"mean_token_accuracy": 0.12390858009457588,
|
|
"num_tokens": 1367080.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.646589183807373,
|
|
"epoch": 0.006555403909599905,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5166,
|
|
"mean_token_accuracy": 0.1198254756629467,
|
|
"num_tokens": 1379104.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.6281653881073,
|
|
"epoch": 0.006609136728531052,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3401,
|
|
"mean_token_accuracy": 0.1338522858917713,
|
|
"num_tokens": 1390346.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.481263113021851,
|
|
"epoch": 0.006662869547462199,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4714,
|
|
"mean_token_accuracy": 0.12188125848770141,
|
|
"num_tokens": 1401512.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.5761816024780275,
|
|
"epoch": 0.006716602366393346,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4625,
|
|
"mean_token_accuracy": 0.1254078231751919,
|
|
"num_tokens": 1412410.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.563577651977539,
|
|
"epoch": 0.0067703351853244925,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3567,
|
|
"mean_token_accuracy": 0.12992686927318572,
|
|
"num_tokens": 1424028.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.653519344329834,
|
|
"epoch": 0.006824068004255639,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.1232505589723587,
|
|
"num_tokens": 1435275.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.635418176651001,
|
|
"epoch": 0.006877800823186786,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4836,
|
|
"mean_token_accuracy": 0.1284290887415409,
|
|
"num_tokens": 1447336.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.5360087871551515,
|
|
"epoch": 0.006931533642117933,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4281,
|
|
"mean_token_accuracy": 0.12723628357052802,
|
|
"num_tokens": 1459224.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.741295862197876,
|
|
"epoch": 0.0069852664610490795,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5461,
|
|
"mean_token_accuracy": 0.12568391785025596,
|
|
"num_tokens": 1472077.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.6534796237945555,
|
|
"epoch": 0.007038999279980226,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.12574814409017562,
|
|
"num_tokens": 1484296.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.606902265548706,
|
|
"epoch": 0.007092732098911373,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4744,
|
|
"mean_token_accuracy": 0.12749515175819398,
|
|
"num_tokens": 1495723.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.517670679092407,
|
|
"epoch": 0.00714646491784252,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3645,
|
|
"mean_token_accuracy": 0.13283996507525445,
|
|
"num_tokens": 1506610.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.611458110809326,
|
|
"epoch": 0.007200197736773667,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4832,
|
|
"mean_token_accuracy": 0.1267407178878784,
|
|
"num_tokens": 1519216.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.491586589813233,
|
|
"epoch": 0.007253930555704813,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.446,
|
|
"mean_token_accuracy": 0.12697500586509705,
|
|
"num_tokens": 1529925.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.59675145149231,
|
|
"epoch": 0.00730766337463596,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4008,
|
|
"mean_token_accuracy": 0.1306189052760601,
|
|
"num_tokens": 1541329.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.4607322216033936,
|
|
"epoch": 0.007361396193567107,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4265,
|
|
"mean_token_accuracy": 0.12382375597953796,
|
|
"num_tokens": 1552716.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.526829433441162,
|
|
"epoch": 0.007415129012498254,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3576,
|
|
"mean_token_accuracy": 0.12638328224420547,
|
|
"num_tokens": 1563170.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.584331369400024,
|
|
"epoch": 0.0074688618314294,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5071,
|
|
"mean_token_accuracy": 0.13071208000183104,
|
|
"num_tokens": 1574431.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.633341598510742,
|
|
"epoch": 0.007522594650360547,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.12467131316661835,
|
|
"num_tokens": 1586173.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.478550148010254,
|
|
"epoch": 0.007576327469291694,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4009,
|
|
"mean_token_accuracy": 0.12992049753665924,
|
|
"num_tokens": 1596809.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.555624914169312,
|
|
"epoch": 0.007630060288222841,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3558,
|
|
"mean_token_accuracy": 0.12971629872918128,
|
|
"num_tokens": 1608033.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.477349472045899,
|
|
"epoch": 0.0076837931071539875,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4094,
|
|
"mean_token_accuracy": 0.1303440511226654,
|
|
"num_tokens": 1619135.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.490011358261109,
|
|
"epoch": 0.007737525926085134,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4472,
|
|
"mean_token_accuracy": 0.1352962836623192,
|
|
"num_tokens": 1630162.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.639921855926514,
|
|
"epoch": 0.007791258745016281,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4996,
|
|
"mean_token_accuracy": 0.12582269459962844,
|
|
"num_tokens": 1641439.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.511112117767334,
|
|
"epoch": 0.007844991563947428,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4003,
|
|
"mean_token_accuracy": 0.12877567633986473,
|
|
"num_tokens": 1653615.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.463456630706787,
|
|
"epoch": 0.007898724382878575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2878,
|
|
"mean_token_accuracy": 0.1366309866309166,
|
|
"num_tokens": 1663262.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.641633558273315,
|
|
"epoch": 0.007952457201809721,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5463,
|
|
"mean_token_accuracy": 0.11863012239336967,
|
|
"num_tokens": 1676012.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.5580424785614015,
|
|
"epoch": 0.008006190020740869,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4331,
|
|
"mean_token_accuracy": 0.1215569481253624,
|
|
"num_tokens": 1688145.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.420039892196655,
|
|
"epoch": 0.008059922839672015,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.13154705688357354,
|
|
"num_tokens": 1698710.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.487041378021241,
|
|
"epoch": 0.008113655658603162,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3722,
|
|
"mean_token_accuracy": 0.13104054182767869,
|
|
"num_tokens": 1710956.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.517988061904907,
|
|
"epoch": 0.008167388477534308,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4403,
|
|
"mean_token_accuracy": 0.1294276535511017,
|
|
"num_tokens": 1722915.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.449849367141724,
|
|
"epoch": 0.008221121296465456,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3422,
|
|
"mean_token_accuracy": 0.1337424710392952,
|
|
"num_tokens": 1734474.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.4679615020751955,
|
|
"epoch": 0.008274854115396602,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4661,
|
|
"mean_token_accuracy": 0.1281263567507267,
|
|
"num_tokens": 1747871.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.52447919845581,
|
|
"epoch": 0.00832858693432775,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3677,
|
|
"mean_token_accuracy": 0.13014960363507272,
|
|
"num_tokens": 1759368.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.4473429203033445,
|
|
"epoch": 0.008382319753258895,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3219,
|
|
"mean_token_accuracy": 0.12805930003523827,
|
|
"num_tokens": 1770077.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.341373634338379,
|
|
"epoch": 0.008436052572190043,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3127,
|
|
"mean_token_accuracy": 0.12711255848407746,
|
|
"num_tokens": 1781866.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.420151329040527,
|
|
"epoch": 0.008489785391121189,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3181,
|
|
"mean_token_accuracy": 0.12717027962207794,
|
|
"num_tokens": 1792495.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.4637500762939455,
|
|
"epoch": 0.008543518210052337,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3178,
|
|
"mean_token_accuracy": 0.134869834035635,
|
|
"num_tokens": 1803615.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.361494874954223,
|
|
"epoch": 0.008597251028983482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3092,
|
|
"mean_token_accuracy": 0.13040214627981186,
|
|
"num_tokens": 1814302.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.459384536743164,
|
|
"epoch": 0.00865098384791463,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3336,
|
|
"mean_token_accuracy": 0.13680917099118234,
|
|
"num_tokens": 1825883.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.427951526641846,
|
|
"epoch": 0.008704716666845776,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4295,
|
|
"mean_token_accuracy": 0.12659786641597748,
|
|
"num_tokens": 1837863.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.3974611282348635,
|
|
"epoch": 0.008758449485776924,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3164,
|
|
"mean_token_accuracy": 0.12896919026970863,
|
|
"num_tokens": 1848566.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.373238897323608,
|
|
"epoch": 0.00881218230470807,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4152,
|
|
"mean_token_accuracy": 0.13358617648482324,
|
|
"num_tokens": 1862068.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.488036108016968,
|
|
"epoch": 0.008865915123639217,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.13050863072276114,
|
|
"num_tokens": 1873985.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.3769128799438475,
|
|
"epoch": 0.008919647942570363,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3659,
|
|
"mean_token_accuracy": 0.1300532825291157,
|
|
"num_tokens": 1885805.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.389228773117066,
|
|
"epoch": 0.00897338076150151,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3363,
|
|
"mean_token_accuracy": 0.12887792587280272,
|
|
"num_tokens": 1897520.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.381129407882691,
|
|
"epoch": 0.009027113580432657,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.1281817726790905,
|
|
"num_tokens": 1908645.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.377236175537109,
|
|
"epoch": 0.009080846399363804,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3134,
|
|
"mean_token_accuracy": 0.13246509581804275,
|
|
"num_tokens": 1919933.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.407544898986816,
|
|
"epoch": 0.00913457921829495,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3749,
|
|
"mean_token_accuracy": 0.1343151532113552,
|
|
"num_tokens": 1932150.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.402065134048462,
|
|
"epoch": 0.009188312037226098,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3266,
|
|
"mean_token_accuracy": 0.13442369997501374,
|
|
"num_tokens": 1944379.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.432179164886475,
|
|
"epoch": 0.009242044856157244,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.1368368983268738,
|
|
"num_tokens": 1956156.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.499852514266967,
|
|
"epoch": 0.009295777675088391,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4659,
|
|
"mean_token_accuracy": 0.1252061903476715,
|
|
"num_tokens": 1968585.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.327846670150757,
|
|
"epoch": 0.009349510494019537,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3164,
|
|
"mean_token_accuracy": 0.1293796181678772,
|
|
"num_tokens": 1979627.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.394920063018799,
|
|
"epoch": 0.009403243312950685,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3903,
|
|
"mean_token_accuracy": 0.1291674293577671,
|
|
"num_tokens": 1993368.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.483371686935425,
|
|
"epoch": 0.00945697613188183,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.439,
|
|
"mean_token_accuracy": 0.12856033444404602,
|
|
"num_tokens": 2004334.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.250707912445068,
|
|
"epoch": 0.009510708950812978,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2372,
|
|
"mean_token_accuracy": 0.1328311175107956,
|
|
"num_tokens": 2015328.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.345365381240844,
|
|
"epoch": 0.009564441769744124,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3678,
|
|
"mean_token_accuracy": 0.12939499244093894,
|
|
"num_tokens": 2027341.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.272373819351197,
|
|
"epoch": 0.009618174588675272,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2033,
|
|
"mean_token_accuracy": 0.1377152383327484,
|
|
"num_tokens": 2036774.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.438100528717041,
|
|
"epoch": 0.009671907407606418,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2639,
|
|
"mean_token_accuracy": 0.1349859781563282,
|
|
"num_tokens": 2047320.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.199841403961182,
|
|
"epoch": 0.009725640226537565,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2053,
|
|
"mean_token_accuracy": 0.1357021301984787,
|
|
"num_tokens": 2058336.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.342473936080933,
|
|
"epoch": 0.009779373045468711,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.13590859845280648,
|
|
"num_tokens": 2069767.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.282735824584961,
|
|
"epoch": 0.009833105864399859,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.247,
|
|
"mean_token_accuracy": 0.1364621177315712,
|
|
"num_tokens": 2080018.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.3263512134552,
|
|
"epoch": 0.009886838683331005,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3181,
|
|
"mean_token_accuracy": 0.1294290691614151,
|
|
"num_tokens": 2091274.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.404293727874756,
|
|
"epoch": 0.009940571502262152,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4552,
|
|
"mean_token_accuracy": 0.12284724041819572,
|
|
"num_tokens": 2104780.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.404548597335816,
|
|
"epoch": 0.009994304321193298,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.12493002042174339,
|
|
"num_tokens": 2118181.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.420569181442261,
|
|
"epoch": 0.010048037140124446,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3692,
|
|
"mean_token_accuracy": 0.1358141668140888,
|
|
"num_tokens": 2128371.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.301304912567138,
|
|
"epoch": 0.010101769959055592,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1927,
|
|
"mean_token_accuracy": 0.13996189907193185,
|
|
"num_tokens": 2140385.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.357480192184449,
|
|
"epoch": 0.01015550277798674,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4256,
|
|
"mean_token_accuracy": 0.12998471185564994,
|
|
"num_tokens": 2151030.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.157267808914185,
|
|
"epoch": 0.010209235596917885,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1555,
|
|
"mean_token_accuracy": 0.13839885592460632,
|
|
"num_tokens": 2162977.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.45593671798706,
|
|
"epoch": 0.010262968415849033,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4188,
|
|
"mean_token_accuracy": 0.12474074959754944,
|
|
"num_tokens": 2175320.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.31116042137146,
|
|
"epoch": 0.010316701234780179,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2703,
|
|
"mean_token_accuracy": 0.13561318814754486,
|
|
"num_tokens": 2187401.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.256970357894898,
|
|
"epoch": 0.010370434053711327,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.273,
|
|
"mean_token_accuracy": 0.13364465087652205,
|
|
"num_tokens": 2198475.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.364040231704712,
|
|
"epoch": 0.010424166872642473,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2953,
|
|
"mean_token_accuracy": 0.1304472081363201,
|
|
"num_tokens": 2209401.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.179701614379883,
|
|
"epoch": 0.01047789969157362,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2209,
|
|
"mean_token_accuracy": 0.135325139015913,
|
|
"num_tokens": 2220687.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.4695775508880615,
|
|
"epoch": 0.010531632510504766,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3947,
|
|
"mean_token_accuracy": 0.13351753056049348,
|
|
"num_tokens": 2232876.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.207294893264771,
|
|
"epoch": 0.010585365329435914,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.13681197613477708,
|
|
"num_tokens": 2243683.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.2024573802948,
|
|
"epoch": 0.01063909814836706,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2193,
|
|
"mean_token_accuracy": 0.13724667727947235,
|
|
"num_tokens": 2254919.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.28679141998291,
|
|
"epoch": 0.010692830967298207,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2558,
|
|
"mean_token_accuracy": 0.13715462759137154,
|
|
"num_tokens": 2267583.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.299418592453003,
|
|
"epoch": 0.010746563786229353,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3302,
|
|
"mean_token_accuracy": 0.13524263501167297,
|
|
"num_tokens": 2279228.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.1662780284881595,
|
|
"epoch": 0.0108002966051605,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.1837,
|
|
"mean_token_accuracy": 0.142218117415905,
|
|
"num_tokens": 2291213.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.279097414016723,
|
|
"epoch": 0.010854029424091647,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.13057726323604585,
|
|
"num_tokens": 2304080.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.222244453430176,
|
|
"epoch": 0.010907762243022794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.1906,
|
|
"mean_token_accuracy": 0.13947457447648048,
|
|
"num_tokens": 2315266.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.173005819320679,
|
|
"epoch": 0.01096149506195394,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.2185,
|
|
"mean_token_accuracy": 0.13768133595585824,
|
|
"num_tokens": 2327333.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.293088293075561,
|
|
"epoch": 0.011015227880885088,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.2438,
|
|
"mean_token_accuracy": 0.1401859149336815,
|
|
"num_tokens": 2338356.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.11752758026123,
|
|
"epoch": 0.011068960699816234,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.1355143591761589,
|
|
"num_tokens": 2348935.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.3127048969268795,
|
|
"epoch": 0.011122693518747381,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.1934,
|
|
"mean_token_accuracy": 0.13663028106093406,
|
|
"num_tokens": 2359588.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.140753889083863,
|
|
"epoch": 0.011176426337678527,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.1701,
|
|
"mean_token_accuracy": 0.1450001023709774,
|
|
"num_tokens": 2369463.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.160957622528076,
|
|
"epoch": 0.011230159156609675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.1656,
|
|
"mean_token_accuracy": 0.1414645403623581,
|
|
"num_tokens": 2381900.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.255343818664551,
|
|
"epoch": 0.01128389197554082,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.1495,
|
|
"mean_token_accuracy": 0.14251192957162856,
|
|
"num_tokens": 2393709.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.2668678760528564,
|
|
"epoch": 0.011337624794471968,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.3363,
|
|
"mean_token_accuracy": 0.13923534750938416,
|
|
"num_tokens": 2405084.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.220151090621949,
|
|
"epoch": 0.011391357613403114,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.2392,
|
|
"mean_token_accuracy": 0.14038070291280746,
|
|
"num_tokens": 2416174.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.183935499191284,
|
|
"epoch": 0.011445090432334262,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.252,
|
|
"mean_token_accuracy": 0.13655493929982185,
|
|
"num_tokens": 2427222.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.224246263504028,
|
|
"epoch": 0.011498823251265408,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.1835,
|
|
"mean_token_accuracy": 0.14006806984543801,
|
|
"num_tokens": 2437796.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.186404895782471,
|
|
"epoch": 0.011552556070196555,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.2269,
|
|
"mean_token_accuracy": 0.13990388736128806,
|
|
"num_tokens": 2449065.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.23367486000061,
|
|
"epoch": 0.011606288889127701,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.2503,
|
|
"mean_token_accuracy": 0.14235268533229828,
|
|
"num_tokens": 2460652.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.209626150131226,
|
|
"epoch": 0.011660021708058849,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1733,
|
|
"mean_token_accuracy": 0.14472328796982764,
|
|
"num_tokens": 2472364.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.277301216125489,
|
|
"epoch": 0.011713754526989995,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.3776,
|
|
"mean_token_accuracy": 0.1335189960896969,
|
|
"num_tokens": 2485395.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.2589233875274655,
|
|
"epoch": 0.011767487345921143,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.2798,
|
|
"mean_token_accuracy": 0.13312781751155853,
|
|
"num_tokens": 2498081.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.0594724178314205,
|
|
"epoch": 0.011821220164852288,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.1041,
|
|
"mean_token_accuracy": 0.13819077759981155,
|
|
"num_tokens": 2508453.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.371691083908081,
|
|
"epoch": 0.011874952983783436,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.13026287406682968,
|
|
"num_tokens": 2520474.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.189488077163697,
|
|
"epoch": 0.011928685802714582,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.1501,
|
|
"mean_token_accuracy": 0.1381780758500099,
|
|
"num_tokens": 2531809.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.207001543045044,
|
|
"epoch": 0.01198241862164573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.2354,
|
|
"mean_token_accuracy": 0.1409098967909813,
|
|
"num_tokens": 2542561.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.184149646759034,
|
|
"epoch": 0.012036151440576875,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.1751,
|
|
"mean_token_accuracy": 0.13992175906896592,
|
|
"num_tokens": 2554386.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.211317443847657,
|
|
"epoch": 0.012089884259508023,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.268,
|
|
"mean_token_accuracy": 0.13457310497760772,
|
|
"num_tokens": 2565765.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.274694538116455,
|
|
"epoch": 0.012143617078439169,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.337,
|
|
"mean_token_accuracy": 0.12759412080049515,
|
|
"num_tokens": 2577496.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.245618629455566,
|
|
"epoch": 0.012197349897370317,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.1421503022313118,
|
|
"num_tokens": 2588301.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.185357093811035,
|
|
"epoch": 0.012251082716301463,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.13766290843486786,
|
|
"num_tokens": 2599201.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.220892143249512,
|
|
"epoch": 0.01230481553523261,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.1683,
|
|
"mean_token_accuracy": 0.13754888027906417,
|
|
"num_tokens": 2611402.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.261027193069458,
|
|
"epoch": 0.012358548354163756,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.2672,
|
|
"mean_token_accuracy": 0.1381250239908695,
|
|
"num_tokens": 2623976.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.152841472625733,
|
|
"epoch": 0.012412281173094904,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.13515327349305153,
|
|
"num_tokens": 2635093.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.196397590637207,
|
|
"epoch": 0.01246601399202605,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.1445,
|
|
"mean_token_accuracy": 0.15016828179359437,
|
|
"num_tokens": 2646262.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.208798456192016,
|
|
"epoch": 0.012519746810957197,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.2398,
|
|
"mean_token_accuracy": 0.14184571355581282,
|
|
"num_tokens": 2657325.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.073056030273437,
|
|
"epoch": 0.012573479629888343,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.1166,
|
|
"mean_token_accuracy": 0.140309177339077,
|
|
"num_tokens": 2669218.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.194371080398559,
|
|
"epoch": 0.01262721244881949,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.1338,
|
|
"mean_token_accuracy": 0.14102206751704216,
|
|
"num_tokens": 2681034.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.2010109424591064,
|
|
"epoch": 0.012680945267750637,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.2307,
|
|
"mean_token_accuracy": 0.13652921691536904,
|
|
"num_tokens": 2692853.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.0191319465637205,
|
|
"epoch": 0.012734678086681784,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9808,
|
|
"mean_token_accuracy": 0.15345293283462524,
|
|
"num_tokens": 2703470.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.1350805282592775,
|
|
"epoch": 0.01278841090561293,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.1146,
|
|
"mean_token_accuracy": 0.14526959359645844,
|
|
"num_tokens": 2713740.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.040780162811279,
|
|
"epoch": 0.012842143724544078,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.052,
|
|
"mean_token_accuracy": 0.15104619711637496,
|
|
"num_tokens": 2724951.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.057363033294678,
|
|
"epoch": 0.012895876543475224,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.1079,
|
|
"mean_token_accuracy": 0.15360509753227233,
|
|
"num_tokens": 2735545.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.233933353424073,
|
|
"epoch": 0.012949609362406371,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.1821,
|
|
"mean_token_accuracy": 0.14036801978945732,
|
|
"num_tokens": 2747176.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.063135814666748,
|
|
"epoch": 0.013003342181337517,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.0523,
|
|
"mean_token_accuracy": 0.14713248685002328,
|
|
"num_tokens": 2757439.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.069062185287476,
|
|
"epoch": 0.013057075000268665,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.0739,
|
|
"mean_token_accuracy": 0.14628921821713448,
|
|
"num_tokens": 2768832.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.217538452148437,
|
|
"epoch": 0.01311080781919981,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.2231,
|
|
"mean_token_accuracy": 0.14129021465778352,
|
|
"num_tokens": 2780557.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.103370571136475,
|
|
"epoch": 0.013164540638130958,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.1506,
|
|
"mean_token_accuracy": 0.14464109539985656,
|
|
"num_tokens": 2791630.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.203542852401734,
|
|
"epoch": 0.013218273457062104,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.1708,
|
|
"mean_token_accuracy": 0.14704614207148553,
|
|
"num_tokens": 2802920.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.125507640838623,
|
|
"epoch": 0.013272006275993252,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2278,
|
|
"mean_token_accuracy": 0.13306833431124687,
|
|
"num_tokens": 2815119.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.089851427078247,
|
|
"epoch": 0.013325739094924398,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.0673,
|
|
"mean_token_accuracy": 0.15316638201475144,
|
|
"num_tokens": 2825475.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.11448221206665,
|
|
"epoch": 0.013379471913855545,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.0862,
|
|
"mean_token_accuracy": 0.1438266344368458,
|
|
"num_tokens": 2836751.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.065900611877441,
|
|
"epoch": 0.013433204732786691,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.14495786055922508,
|
|
"num_tokens": 2848396.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.175085210800171,
|
|
"epoch": 0.013486937551717839,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.1109,
|
|
"mean_token_accuracy": 0.14748385846614837,
|
|
"num_tokens": 2859635.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.036644649505615,
|
|
"epoch": 0.013540670370648985,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.14415860995650293,
|
|
"num_tokens": 2871294.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.150858688354492,
|
|
"epoch": 0.013594403189580133,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.1425,
|
|
"mean_token_accuracy": 0.14190022721886636,
|
|
"num_tokens": 2881877.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.08399338722229,
|
|
"epoch": 0.013648136008511278,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.0469,
|
|
"mean_token_accuracy": 0.14779358804225923,
|
|
"num_tokens": 2893084.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.018289709091187,
|
|
"epoch": 0.013701868827442426,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.15105676501989365,
|
|
"num_tokens": 2903884.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.206864881515503,
|
|
"epoch": 0.013755601646373572,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.14684298038482665,
|
|
"num_tokens": 2916302.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.958971071243286,
|
|
"epoch": 0.01380933446530472,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.1494239941239357,
|
|
"num_tokens": 2926945.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.012639570236206,
|
|
"epoch": 0.013863067284235865,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.0225,
|
|
"mean_token_accuracy": 0.16030077040195465,
|
|
"num_tokens": 2938305.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.1305427074432375,
|
|
"epoch": 0.013916800103167013,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.1889,
|
|
"mean_token_accuracy": 0.14667668342590331,
|
|
"num_tokens": 2949652.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.179107427597046,
|
|
"epoch": 0.013970532922098159,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.097,
|
|
"mean_token_accuracy": 0.1481576070189476,
|
|
"num_tokens": 2960237.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.984218406677246,
|
|
"epoch": 0.014024265741029307,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.0083,
|
|
"mean_token_accuracy": 0.14653314501047135,
|
|
"num_tokens": 2971360.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.132961654663086,
|
|
"epoch": 0.014077998559960453,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.1534,
|
|
"mean_token_accuracy": 0.1430380217730999,
|
|
"num_tokens": 2981751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.08388123512268,
|
|
"epoch": 0.0141317313788916,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.1837,
|
|
"mean_token_accuracy": 0.13980235084891318,
|
|
"num_tokens": 2993284.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.234730958938599,
|
|
"epoch": 0.014185464197822746,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.1498,
|
|
"mean_token_accuracy": 0.14336772859096528,
|
|
"num_tokens": 3004853.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.123141479492188,
|
|
"epoch": 0.014239197016753894,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.137,
|
|
"mean_token_accuracy": 0.14076138213276862,
|
|
"num_tokens": 3015875.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.958143711090088,
|
|
"epoch": 0.01429292983568504,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.0321,
|
|
"mean_token_accuracy": 0.14536121785640715,
|
|
"num_tokens": 3026180.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.994622421264649,
|
|
"epoch": 0.014346662654616187,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.146102274954319,
|
|
"num_tokens": 3038647.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.036538553237915,
|
|
"epoch": 0.014400395473547333,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.0033,
|
|
"mean_token_accuracy": 0.14610179960727693,
|
|
"num_tokens": 3050087.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.114994859695434,
|
|
"epoch": 0.01445412829247848,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.191,
|
|
"mean_token_accuracy": 0.14669101536273957,
|
|
"num_tokens": 3061962.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.160302257537841,
|
|
"epoch": 0.014507861111409627,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.1876,
|
|
"mean_token_accuracy": 0.1374008759856224,
|
|
"num_tokens": 3074375.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.054080057144165,
|
|
"epoch": 0.014561593930340774,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 5.9516,
|
|
"mean_token_accuracy": 0.14929106086492538,
|
|
"num_tokens": 3085760.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.053500366210938,
|
|
"epoch": 0.01461532674927192,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.1490132227540016,
|
|
"num_tokens": 3096894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.08893837928772,
|
|
"epoch": 0.014669059568203068,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.1199,
|
|
"mean_token_accuracy": 0.14497611969709395,
|
|
"num_tokens": 3108485.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.077220821380616,
|
|
"epoch": 0.014722792387134214,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.14355420991778373,
|
|
"num_tokens": 3119499.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.023743915557861,
|
|
"epoch": 0.014776525206065361,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.986,
|
|
"mean_token_accuracy": 0.14942106306552888,
|
|
"num_tokens": 3130389.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.09236855506897,
|
|
"epoch": 0.014830258024996507,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.1202,
|
|
"mean_token_accuracy": 0.14113230854272843,
|
|
"num_tokens": 3141034.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.072746324539184,
|
|
"epoch": 0.014883990843927655,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.0854,
|
|
"mean_token_accuracy": 0.14430767744779588,
|
|
"num_tokens": 3152438.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.022948694229126,
|
|
"epoch": 0.0149377236628588,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.0776,
|
|
"mean_token_accuracy": 0.1505906477570534,
|
|
"num_tokens": 3163994.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.0805083274841305,
|
|
"epoch": 0.014991456481789948,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.0461,
|
|
"mean_token_accuracy": 0.1505217045545578,
|
|
"num_tokens": 3175330.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.97646312713623,
|
|
"epoch": 0.015045189300721094,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0017,
|
|
"mean_token_accuracy": 0.14700844660401344,
|
|
"num_tokens": 3185981.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.015813255310059,
|
|
"epoch": 0.015098922119652242,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.1522893041372299,
|
|
"num_tokens": 3198267.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.091410112380982,
|
|
"epoch": 0.015152654938583388,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.0322,
|
|
"mean_token_accuracy": 0.14734495803713799,
|
|
"num_tokens": 3210405.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.986245775222779,
|
|
"epoch": 0.015206387757514536,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.0579,
|
|
"mean_token_accuracy": 0.1433064103126526,
|
|
"num_tokens": 3220528.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.014886426925659,
|
|
"epoch": 0.015260120576445681,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0343,
|
|
"mean_token_accuracy": 0.14501605182886124,
|
|
"num_tokens": 3231623.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.983619356155396,
|
|
"epoch": 0.015313853395376829,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.0083,
|
|
"mean_token_accuracy": 0.14880426228046417,
|
|
"num_tokens": 3242067.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.971679782867431,
|
|
"epoch": 0.015367586214307975,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.15313563197851182,
|
|
"num_tokens": 3252898.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.075572872161866,
|
|
"epoch": 0.015421319033239123,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.1571,
|
|
"mean_token_accuracy": 0.14407433122396468,
|
|
"num_tokens": 3265128.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.120305967330933,
|
|
"epoch": 0.015475051852170268,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.0641,
|
|
"mean_token_accuracy": 0.14864805042743684,
|
|
"num_tokens": 3275808.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.003734397888183,
|
|
"epoch": 0.015528784671101416,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.0399,
|
|
"mean_token_accuracy": 0.15222140848636628,
|
|
"num_tokens": 3286682.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.070817852020264,
|
|
"epoch": 0.015582517490032562,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.0783,
|
|
"mean_token_accuracy": 0.153215591609478,
|
|
"num_tokens": 3299402.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.1345799446105955,
|
|
"epoch": 0.01563625030896371,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.1546,
|
|
"mean_token_accuracy": 0.14280287623405458,
|
|
"num_tokens": 3311185.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.167125463485718,
|
|
"epoch": 0.015689983127894856,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.0728,
|
|
"mean_token_accuracy": 0.14637029469013213,
|
|
"num_tokens": 3322489.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.956501626968384,
|
|
"epoch": 0.015743715946826,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.0951,
|
|
"mean_token_accuracy": 0.1488440901041031,
|
|
"num_tokens": 3334731.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.063800287246704,
|
|
"epoch": 0.01579744876575715,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0151,
|
|
"mean_token_accuracy": 0.15109897255897523,
|
|
"num_tokens": 3347107.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.066719198226929,
|
|
"epoch": 0.015851181584688297,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.0086,
|
|
"mean_token_accuracy": 0.1461263604462147,
|
|
"num_tokens": 3358640.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.058962821960449,
|
|
"epoch": 0.015904914403619443,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.1511,
|
|
"mean_token_accuracy": 0.1421753115952015,
|
|
"num_tokens": 3369743.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.13518762588501,
|
|
"epoch": 0.01595864722255059,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.1039,
|
|
"mean_token_accuracy": 0.14061811119318007,
|
|
"num_tokens": 3381413.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.0545531749725345,
|
|
"epoch": 0.016012380041481738,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.9629,
|
|
"mean_token_accuracy": 0.15147602558135986,
|
|
"num_tokens": 3391347.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.090392827987671,
|
|
"epoch": 0.016066112860412884,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.1731,
|
|
"mean_token_accuracy": 0.14226020574569703,
|
|
"num_tokens": 3403348.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.013070344924927,
|
|
"epoch": 0.01611984567934403,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.9502,
|
|
"mean_token_accuracy": 0.15287814140319825,
|
|
"num_tokens": 3414845.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.949804735183716,
|
|
"epoch": 0.016173578498275176,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.15804148465394974,
|
|
"num_tokens": 3426105.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.934489631652832,
|
|
"epoch": 0.016227311317206325,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 5.9319,
|
|
"mean_token_accuracy": 0.14723418205976485,
|
|
"num_tokens": 3438001.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.975189590454102,
|
|
"epoch": 0.01628104413613747,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 5.9712,
|
|
"mean_token_accuracy": 0.15585855692625045,
|
|
"num_tokens": 3448888.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.960818529129028,
|
|
"epoch": 0.016334776955068617,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.14731887802481652,
|
|
"num_tokens": 3460325.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.036577892303467,
|
|
"epoch": 0.016388509773999763,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.14315218701958657,
|
|
"num_tokens": 3471599.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.994147682189942,
|
|
"epoch": 0.016442242592930912,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.0781,
|
|
"mean_token_accuracy": 0.14317068159580232,
|
|
"num_tokens": 3482661.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.0836540222167965,
|
|
"epoch": 0.016495975411862058,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.14395593404769896,
|
|
"num_tokens": 3493111.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.156943607330322,
|
|
"epoch": 0.016549708230793204,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.1462,
|
|
"mean_token_accuracy": 0.14156040474772452,
|
|
"num_tokens": 3505575.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.01388897895813,
|
|
"epoch": 0.01660344104972435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.0175,
|
|
"mean_token_accuracy": 0.14803557544946672,
|
|
"num_tokens": 3515498.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.069819784164428,
|
|
"epoch": 0.0166571738686555,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 5.9977,
|
|
"mean_token_accuracy": 0.15337349474430084,
|
|
"num_tokens": 3527280.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.991217470169067,
|
|
"epoch": 0.016710906687586645,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.0206,
|
|
"mean_token_accuracy": 0.14981672763824463,
|
|
"num_tokens": 3538979.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.008299779891968,
|
|
"epoch": 0.01676463950651779,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.0896,
|
|
"mean_token_accuracy": 0.14561396539211274,
|
|
"num_tokens": 3551193.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.0555739402771,
|
|
"epoch": 0.016818372325448937,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.9618,
|
|
"mean_token_accuracy": 0.15235140174627304,
|
|
"num_tokens": 3562052.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.980268669128418,
|
|
"epoch": 0.016872105144380086,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.0227,
|
|
"mean_token_accuracy": 0.1535588562488556,
|
|
"num_tokens": 3573378.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.064887428283692,
|
|
"epoch": 0.016925837963311232,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.15122793316841127,
|
|
"num_tokens": 3584533.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.035519599914551,
|
|
"epoch": 0.016979570782242378,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.1535245344042778,
|
|
"num_tokens": 3595934.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.846812677383423,
|
|
"epoch": 0.017033303601173524,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 5.9077,
|
|
"mean_token_accuracy": 0.1490027278661728,
|
|
"num_tokens": 3607704.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.140884160995483,
|
|
"epoch": 0.017087036420104673,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.0267,
|
|
"mean_token_accuracy": 0.14754170551896095,
|
|
"num_tokens": 3619135.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.892256498336792,
|
|
"epoch": 0.01714076923903582,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 5.9312,
|
|
"mean_token_accuracy": 0.15658772438764573,
|
|
"num_tokens": 3630128.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.898729991912842,
|
|
"epoch": 0.017194502057966965,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.8594,
|
|
"mean_token_accuracy": 0.15367861837148666,
|
|
"num_tokens": 3642045.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.016676568984986,
|
|
"epoch": 0.01724823487689811,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.9743,
|
|
"mean_token_accuracy": 0.15024706870317459,
|
|
"num_tokens": 3652631.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.975802707672119,
|
|
"epoch": 0.01730196769582926,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.15665287524461746,
|
|
"num_tokens": 3663669.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.988212919235229,
|
|
"epoch": 0.017355700514760406,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 5.964,
|
|
"mean_token_accuracy": 0.15508741587400438,
|
|
"num_tokens": 3675425.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.8765256881713865,
|
|
"epoch": 0.017409433333691552,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 5.9864,
|
|
"mean_token_accuracy": 0.1466473385691643,
|
|
"num_tokens": 3687030.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.067311143875122,
|
|
"epoch": 0.017463166152622698,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.9918,
|
|
"mean_token_accuracy": 0.14988280236721038,
|
|
"num_tokens": 3697597.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.052053213119507,
|
|
"epoch": 0.017516898971553847,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.041,
|
|
"mean_token_accuracy": 0.15012367069721222,
|
|
"num_tokens": 3710967.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.998264408111572,
|
|
"epoch": 0.017570631790484993,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 5.9856,
|
|
"mean_token_accuracy": 0.15023086369037628,
|
|
"num_tokens": 3722452.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.971319723129272,
|
|
"epoch": 0.01762436460941614,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.15278342962265015,
|
|
"num_tokens": 3733256.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.9798520565032955,
|
|
"epoch": 0.017678097428347285,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.15721064656972886,
|
|
"num_tokens": 3744458.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.901952314376831,
|
|
"epoch": 0.017731830247278434,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.15427957475185394,
|
|
"num_tokens": 3756099.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.088839864730835,
|
|
"epoch": 0.01778556306620958,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.0048,
|
|
"mean_token_accuracy": 0.1539461597800255,
|
|
"num_tokens": 3768667.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.9746912002563475,
|
|
"epoch": 0.017839295885140726,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 5.925,
|
|
"mean_token_accuracy": 0.15976884216070175,
|
|
"num_tokens": 3779303.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.987901973724365,
|
|
"epoch": 0.017893028704071872,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.14600329846143723,
|
|
"num_tokens": 3792271.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.981582593917847,
|
|
"epoch": 0.01794676152300302,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.0726,
|
|
"mean_token_accuracy": 0.1465673990547657,
|
|
"num_tokens": 3805387.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.0047464847564695,
|
|
"epoch": 0.018000494341934167,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.14909646064043044,
|
|
"num_tokens": 3817184.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.832823324203491,
|
|
"epoch": 0.018054227160865313,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.16253771334886552,
|
|
"num_tokens": 3827399.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.989438438415528,
|
|
"epoch": 0.01810795997979646,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.8701,
|
|
"mean_token_accuracy": 0.1581664875149727,
|
|
"num_tokens": 3838320.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.9107225894927975,
|
|
"epoch": 0.01816169279872761,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 5.9919,
|
|
"mean_token_accuracy": 0.15036764591932297,
|
|
"num_tokens": 3849431.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.968296241760254,
|
|
"epoch": 0.018215425617658754,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 5.9573,
|
|
"mean_token_accuracy": 0.1487127035856247,
|
|
"num_tokens": 3860265.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.979854774475098,
|
|
"epoch": 0.0182691584365899,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.1496817596256733,
|
|
"num_tokens": 3872906.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.9607490539550785,
|
|
"epoch": 0.018322891255521046,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 5.8952,
|
|
"mean_token_accuracy": 0.16413907706737518,
|
|
"num_tokens": 3883592.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.914381790161133,
|
|
"epoch": 0.018376624074452196,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.150118912756443,
|
|
"num_tokens": 3895827.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.05848798751831,
|
|
"epoch": 0.01843035689338334,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 5.9976,
|
|
"mean_token_accuracy": 0.15300071835517884,
|
|
"num_tokens": 3906414.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.967066144943237,
|
|
"epoch": 0.018484089712314487,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 5.9985,
|
|
"mean_token_accuracy": 0.15606716126203538,
|
|
"num_tokens": 3917546.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.0117936611175535,
|
|
"epoch": 0.018537822531245633,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.15176693648099898,
|
|
"num_tokens": 3930251.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.942166519165039,
|
|
"epoch": 0.018591555350176783,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 5.8854,
|
|
"mean_token_accuracy": 0.15790802389383315,
|
|
"num_tokens": 3940912.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.968822240829468,
|
|
"epoch": 0.01864528816910793,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 5.9738,
|
|
"mean_token_accuracy": 0.15192489475011825,
|
|
"num_tokens": 3952163.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.0719554901123045,
|
|
"epoch": 0.018699020988039074,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.0409,
|
|
"mean_token_accuracy": 0.1489500418305397,
|
|
"num_tokens": 3964747.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.024024391174317,
|
|
"epoch": 0.01875275380697022,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.9648,
|
|
"mean_token_accuracy": 0.15161607265472413,
|
|
"num_tokens": 3975789.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.890661907196045,
|
|
"epoch": 0.01880648662590137,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 5.8881,
|
|
"mean_token_accuracy": 0.15519577413797378,
|
|
"num_tokens": 3987816.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.976252126693725,
|
|
"epoch": 0.018860219444832516,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 5.9422,
|
|
"mean_token_accuracy": 0.16160201877355576,
|
|
"num_tokens": 3999882.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.038746356964111,
|
|
"epoch": 0.01891395226376366,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.0676,
|
|
"mean_token_accuracy": 0.1523244395852089,
|
|
"num_tokens": 4012018.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.886965990066528,
|
|
"epoch": 0.018967685082694807,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.15264154970645905,
|
|
"num_tokens": 4023617.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.93788857460022,
|
|
"epoch": 0.019021417901625957,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.1563568189740181,
|
|
"num_tokens": 4035514.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.018500137329101,
|
|
"epoch": 0.019075150720557103,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.9049,
|
|
"mean_token_accuracy": 0.15637483298778534,
|
|
"num_tokens": 4046353.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.920236587524414,
|
|
"epoch": 0.01912888353948825,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.9225,
|
|
"mean_token_accuracy": 0.15857286751270294,
|
|
"num_tokens": 4056839.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.953616905212402,
|
|
"epoch": 0.019182616358419394,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.9055,
|
|
"mean_token_accuracy": 0.15334571674466133,
|
|
"num_tokens": 4068501.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.896912384033203,
|
|
"epoch": 0.019236349177350544,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.1632518321275711,
|
|
"num_tokens": 4080234.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.949176025390625,
|
|
"epoch": 0.01929008199628169,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 5.9342,
|
|
"mean_token_accuracy": 0.15496103018522261,
|
|
"num_tokens": 4091278.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.872956085205078,
|
|
"epoch": 0.019343814815212836,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 5.8986,
|
|
"mean_token_accuracy": 0.159717920422554,
|
|
"num_tokens": 4104017.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.045824766159058,
|
|
"epoch": 0.01939754763414398,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.15220074355602264,
|
|
"num_tokens": 4115840.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.957116413116455,
|
|
"epoch": 0.01945128045307513,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.9011,
|
|
"mean_token_accuracy": 0.15585386604070664,
|
|
"num_tokens": 4126156.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.921035194396973,
|
|
"epoch": 0.019505013272006277,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 5.9728,
|
|
"mean_token_accuracy": 0.1535612739622593,
|
|
"num_tokens": 4137430.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.0158624172210695,
|
|
"epoch": 0.019558746090937423,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.8915,
|
|
"mean_token_accuracy": 0.1606772392988205,
|
|
"num_tokens": 4148180.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.926541423797607,
|
|
"epoch": 0.01961247890986857,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 5.9404,
|
|
"mean_token_accuracy": 0.15600281357765197,
|
|
"num_tokens": 4159321.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.989021730422974,
|
|
"epoch": 0.019666211728799718,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 5.9789,
|
|
"mean_token_accuracy": 0.150770403444767,
|
|
"num_tokens": 4171701.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.013437128067016,
|
|
"epoch": 0.019719944547730864,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.1528964176774025,
|
|
"num_tokens": 4182471.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.862358665466308,
|
|
"epoch": 0.01977367736666201,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.15291235893964766,
|
|
"num_tokens": 4194524.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.07710747718811,
|
|
"epoch": 0.019827410185593156,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 5.9101,
|
|
"mean_token_accuracy": 0.1592467501759529,
|
|
"num_tokens": 4205400.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.889270830154419,
|
|
"epoch": 0.019881143004524305,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.848,
|
|
"mean_token_accuracy": 0.15857322067022322,
|
|
"num_tokens": 4216128.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.9004319190979,
|
|
"epoch": 0.01993487582345545,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.8819,
|
|
"mean_token_accuracy": 0.16485319882631302,
|
|
"num_tokens": 4227338.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.839734077453613,
|
|
"epoch": 0.019988608642386597,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.15411941409111024,
|
|
"num_tokens": 4238576.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.953181362152099,
|
|
"epoch": 0.020042341461317743,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.8317,
|
|
"mean_token_accuracy": 0.16432886570692062,
|
|
"num_tokens": 4251308.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.920472192764282,
|
|
"epoch": 0.020096074280248892,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.16128864288330078,
|
|
"num_tokens": 4262646.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.963524627685547,
|
|
"epoch": 0.020149807099180038,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.9365,
|
|
"mean_token_accuracy": 0.15799531936645508,
|
|
"num_tokens": 4272665.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.946022081375122,
|
|
"epoch": 0.020203539918111184,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 5.8932,
|
|
"mean_token_accuracy": 0.1565667286515236,
|
|
"num_tokens": 4283474.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.8529167652130125,
|
|
"epoch": 0.02025727273704233,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.15843120515346526,
|
|
"num_tokens": 4294294.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.9457238674163815,
|
|
"epoch": 0.02031100555597348,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.9155,
|
|
"mean_token_accuracy": 0.15780601650476456,
|
|
"num_tokens": 4304331.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.859744358062744,
|
|
"epoch": 0.020364738374904625,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 5.8213,
|
|
"mean_token_accuracy": 0.16632000505924224,
|
|
"num_tokens": 4315921.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.870091199874878,
|
|
"epoch": 0.02041847119383577,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.7803,
|
|
"mean_token_accuracy": 0.16494602113962173,
|
|
"num_tokens": 4326994.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.991072273254394,
|
|
"epoch": 0.020472204012766917,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.9941,
|
|
"mean_token_accuracy": 0.1513021007180214,
|
|
"num_tokens": 4339200.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.863003730773926,
|
|
"epoch": 0.020525936831698066,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.1549378976225853,
|
|
"num_tokens": 4350836.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.78472638130188,
|
|
"epoch": 0.020579669650629212,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.8385,
|
|
"mean_token_accuracy": 0.15767095237970352,
|
|
"num_tokens": 4362270.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.997738265991211,
|
|
"epoch": 0.020633402469560358,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 5.9712,
|
|
"mean_token_accuracy": 0.15071888118982316,
|
|
"num_tokens": 4373342.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.948875188827515,
|
|
"epoch": 0.020687135288491504,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.942,
|
|
"mean_token_accuracy": 0.15640480518341066,
|
|
"num_tokens": 4383773.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.982992839813233,
|
|
"epoch": 0.020740868107422653,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.8646,
|
|
"mean_token_accuracy": 0.1617866724729538,
|
|
"num_tokens": 4395995.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.968017673492431,
|
|
"epoch": 0.0207946009263538,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.0062,
|
|
"mean_token_accuracy": 0.1527876764535904,
|
|
"num_tokens": 4407413.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.014474534988404,
|
|
"epoch": 0.020848333745284945,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.15978933423757552,
|
|
"num_tokens": 4418250.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.874887323379516,
|
|
"epoch": 0.02090206656421609,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.8418,
|
|
"mean_token_accuracy": 0.15898097455501556,
|
|
"num_tokens": 4430719.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.90364408493042,
|
|
"epoch": 0.02095579938314724,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.8066,
|
|
"mean_token_accuracy": 0.16269971579313278,
|
|
"num_tokens": 4441783.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.900510597229004,
|
|
"epoch": 0.021009532202078386,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.8651,
|
|
"mean_token_accuracy": 0.16481055617332457,
|
|
"num_tokens": 4451695.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.877746438980102,
|
|
"epoch": 0.021063265021009532,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 5.908,
|
|
"mean_token_accuracy": 0.15196669101715088,
|
|
"num_tokens": 4463437.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.90544638633728,
|
|
"epoch": 0.021116997839940678,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.1617526188492775,
|
|
"num_tokens": 4474805.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.769601202011108,
|
|
"epoch": 0.021170730658871827,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.7712,
|
|
"mean_token_accuracy": 0.16277728527784346,
|
|
"num_tokens": 4485322.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.03261170387268,
|
|
"epoch": 0.021224463477802973,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.14840409606695176,
|
|
"num_tokens": 4498593.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.92830901145935,
|
|
"epoch": 0.02127819629673412,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.15990394055843354,
|
|
"num_tokens": 4509791.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.804613447189331,
|
|
"epoch": 0.021331929115665265,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.8352,
|
|
"mean_token_accuracy": 0.15364052653312682,
|
|
"num_tokens": 4522500.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.943467807769776,
|
|
"epoch": 0.021385661934596414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.16179392337799073,
|
|
"num_tokens": 4533261.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.8952281951904295,
|
|
"epoch": 0.02143939475352756,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.15579070299863815,
|
|
"num_tokens": 4545115.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.916673755645752,
|
|
"epoch": 0.021493127572458706,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.15585279166698457,
|
|
"num_tokens": 4555652.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.934366464614868,
|
|
"epoch": 0.021546860391389852,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 5.8749,
|
|
"mean_token_accuracy": 0.15709411799907685,
|
|
"num_tokens": 4567002.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.845104026794433,
|
|
"epoch": 0.021600593210321,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 5.8313,
|
|
"mean_token_accuracy": 0.15837447047233583,
|
|
"num_tokens": 4578865.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.793741273880005,
|
|
"epoch": 0.021654326029252147,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 5.7499,
|
|
"mean_token_accuracy": 0.16316841691732406,
|
|
"num_tokens": 4589888.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.894748830795288,
|
|
"epoch": 0.021708058848183293,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 5.9061,
|
|
"mean_token_accuracy": 0.16138454228639604,
|
|
"num_tokens": 4602259.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.909497356414795,
|
|
"epoch": 0.02176179166711444,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 5.8477,
|
|
"mean_token_accuracy": 0.16068296432495116,
|
|
"num_tokens": 4613685.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.823246479034424,
|
|
"epoch": 0.02181552448604559,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 5.7935,
|
|
"mean_token_accuracy": 0.16047248989343643,
|
|
"num_tokens": 4624429.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.907180452346802,
|
|
"epoch": 0.021869257304976734,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.15788697600364685,
|
|
"num_tokens": 4635539.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.977007293701172,
|
|
"epoch": 0.02192299012390788,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 5.8179,
|
|
"mean_token_accuracy": 0.1568484991788864,
|
|
"num_tokens": 4646462.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.903705787658692,
|
|
"epoch": 0.021976722942839026,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.93,
|
|
"mean_token_accuracy": 0.15742650926113128,
|
|
"num_tokens": 4657070.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.974299526214599,
|
|
"epoch": 0.022030455761770176,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 5.9091,
|
|
"mean_token_accuracy": 0.1498878687620163,
|
|
"num_tokens": 4669516.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.746604919433594,
|
|
"epoch": 0.02208418858070132,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.16665473133325576,
|
|
"num_tokens": 4680438.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.9115900039672855,
|
|
"epoch": 0.022137921399632467,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 5.7809,
|
|
"mean_token_accuracy": 0.1592145726084709,
|
|
"num_tokens": 4691406.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.917256927490234,
|
|
"epoch": 0.022191654218563613,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.16325609534978866,
|
|
"num_tokens": 4703001.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.771203660964966,
|
|
"epoch": 0.022245387037494763,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.16295778900384902,
|
|
"num_tokens": 4714565.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.940184879302978,
|
|
"epoch": 0.02229911985642591,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.166159588098526,
|
|
"num_tokens": 4724953.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.889848279953003,
|
|
"epoch": 0.022352852675357054,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 5.831,
|
|
"mean_token_accuracy": 0.16450853198766707,
|
|
"num_tokens": 4736494.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.844532537460327,
|
|
"epoch": 0.0224065854942882,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.1601066008210182,
|
|
"num_tokens": 4748091.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.902660083770752,
|
|
"epoch": 0.02246031831321935,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.16419359147548676,
|
|
"num_tokens": 4760750.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.970547866821289,
|
|
"epoch": 0.022514051132150496,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 5.8665,
|
|
"mean_token_accuracy": 0.15801959484815598,
|
|
"num_tokens": 4771554.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.881400918960571,
|
|
"epoch": 0.02256778395108164,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 5.7908,
|
|
"mean_token_accuracy": 0.1635134130716324,
|
|
"num_tokens": 4782435.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.842329263687134,
|
|
"epoch": 0.022621516770012787,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.15357455015182495,
|
|
"num_tokens": 4794056.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.955543804168701,
|
|
"epoch": 0.022675249588943937,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 5.7678,
|
|
"mean_token_accuracy": 0.1559322863817215,
|
|
"num_tokens": 4804833.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.8961999893188475,
|
|
"epoch": 0.022728982407875083,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.0007,
|
|
"mean_token_accuracy": 0.15544072836637496,
|
|
"num_tokens": 4815302.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.015468263626099,
|
|
"epoch": 0.02278271522680623,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 5.86,
|
|
"mean_token_accuracy": 0.15648248046636581,
|
|
"num_tokens": 4825495.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.820241069793701,
|
|
"epoch": 0.022836448045737374,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 5.8121,
|
|
"mean_token_accuracy": 0.16377327591180801,
|
|
"num_tokens": 4836944.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.793530559539795,
|
|
"epoch": 0.022890180864668524,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 5.7814,
|
|
"mean_token_accuracy": 0.15986590832471848,
|
|
"num_tokens": 4848531.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.774704027175903,
|
|
"epoch": 0.02294391368359967,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 5.8061,
|
|
"mean_token_accuracy": 0.15781699866056442,
|
|
"num_tokens": 4859684.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.844147491455078,
|
|
"epoch": 0.022997646502530816,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 5.6786,
|
|
"mean_token_accuracy": 0.16881223618984223,
|
|
"num_tokens": 4869940.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.888142824172974,
|
|
"epoch": 0.02305137932146196,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.15286178514361382,
|
|
"num_tokens": 4882705.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.019544410705566,
|
|
"epoch": 0.02310511214039311,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.1605281725525856,
|
|
"num_tokens": 4894669.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.882890415191651,
|
|
"epoch": 0.023158844959324257,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 5.8603,
|
|
"mean_token_accuracy": 0.1549423336982727,
|
|
"num_tokens": 4905452.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.814490127563476,
|
|
"epoch": 0.023212577778255403,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 5.7832,
|
|
"mean_token_accuracy": 0.1656615898013115,
|
|
"num_tokens": 4917147.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.900537300109863,
|
|
"epoch": 0.02326631059718655,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.897,
|
|
"mean_token_accuracy": 0.158531653881073,
|
|
"num_tokens": 4927852.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.015939664840698,
|
|
"epoch": 0.023320043416117698,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.15496922582387923,
|
|
"num_tokens": 4940823.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.937225103378296,
|
|
"epoch": 0.023373776235048844,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 5.8259,
|
|
"mean_token_accuracy": 0.16374034732580184,
|
|
"num_tokens": 4953059.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.827975940704346,
|
|
"epoch": 0.02342750905397999,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 5.7998,
|
|
"mean_token_accuracy": 0.1567051202058792,
|
|
"num_tokens": 4963083.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.804965400695801,
|
|
"epoch": 0.023481241872911136,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.6965,
|
|
"mean_token_accuracy": 0.17487651109695435,
|
|
"num_tokens": 4974613.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.873419713973999,
|
|
"epoch": 0.023534974691842285,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 5.7715,
|
|
"mean_token_accuracy": 0.165004463493824,
|
|
"num_tokens": 4985664.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.9600988864898685,
|
|
"epoch": 0.02358870751077343,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.162746462225914,
|
|
"num_tokens": 4996548.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.895655965805053,
|
|
"epoch": 0.023642440329704577,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 5.8219,
|
|
"mean_token_accuracy": 0.15765224620699883,
|
|
"num_tokens": 5008049.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.881889200210571,
|
|
"epoch": 0.023696173148635723,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.16293119490146638,
|
|
"num_tokens": 5018966.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.973626756668091,
|
|
"epoch": 0.023749905967566872,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 5.9199,
|
|
"mean_token_accuracy": 0.15537893921136856,
|
|
"num_tokens": 5031236.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.906442880630493,
|
|
"epoch": 0.023803638786498018,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.17129947394132614,
|
|
"num_tokens": 5043225.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.834283447265625,
|
|
"epoch": 0.023857371605429164,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.1622248560190201,
|
|
"num_tokens": 5054622.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.77225694656372,
|
|
"epoch": 0.02391110442436031,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 5.697,
|
|
"mean_token_accuracy": 0.17152493000030516,
|
|
"num_tokens": 5065733.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.786276865005493,
|
|
"epoch": 0.02396483724329146,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.1677599549293518,
|
|
"num_tokens": 5076916.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.94759202003479,
|
|
"epoch": 0.024018570062222605,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.1546024948358536,
|
|
"num_tokens": 5089219.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.934771537780762,
|
|
"epoch": 0.02407230288115375,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 5.8263,
|
|
"mean_token_accuracy": 0.16091297715902328,
|
|
"num_tokens": 5100713.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.777915096282959,
|
|
"epoch": 0.024126035700084897,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.7414,
|
|
"mean_token_accuracy": 0.16442784667015076,
|
|
"num_tokens": 5112791.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.869255828857422,
|
|
"epoch": 0.024179768519016046,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 5.7968,
|
|
"mean_token_accuracy": 0.16915352642536163,
|
|
"num_tokens": 5124174.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.743664979934692,
|
|
"epoch": 0.024233501337947192,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.1726829767227173,
|
|
"num_tokens": 5135119.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.82419900894165,
|
|
"epoch": 0.024287234156878338,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.7785,
|
|
"mean_token_accuracy": 0.15548328906297684,
|
|
"num_tokens": 5146341.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.856739282608032,
|
|
"epoch": 0.024340966975809484,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.16073938310146332,
|
|
"num_tokens": 5156827.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.877044010162353,
|
|
"epoch": 0.024394699794740633,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.1646682694554329,
|
|
"num_tokens": 5167871.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.8398068904876705,
|
|
"epoch": 0.02444843261367178,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 5.6926,
|
|
"mean_token_accuracy": 0.1736021801829338,
|
|
"num_tokens": 5178912.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.877172994613647,
|
|
"epoch": 0.024502165432602925,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 5.8173,
|
|
"mean_token_accuracy": 0.16081616282463074,
|
|
"num_tokens": 5189665.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.876492500305176,
|
|
"epoch": 0.02455589825153407,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 5.7755,
|
|
"mean_token_accuracy": 0.16354334205389023,
|
|
"num_tokens": 5201928.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.8545441150665285,
|
|
"epoch": 0.02460963107046522,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 5.9475,
|
|
"mean_token_accuracy": 0.16005272269248963,
|
|
"num_tokens": 5214288.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.880595254898071,
|
|
"epoch": 0.024663363889396366,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 5.7083,
|
|
"mean_token_accuracy": 0.16264453828334807,
|
|
"num_tokens": 5224508.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.9032481670379635,
|
|
"epoch": 0.024717096708327512,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 5.7615,
|
|
"mean_token_accuracy": 0.16583350747823716,
|
|
"num_tokens": 5234883.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.757997226715088,
|
|
"epoch": 0.024770829527258658,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.1645049884915352,
|
|
"num_tokens": 5245373.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.9000328540802,
|
|
"epoch": 0.024824562346189807,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.16993815153837205,
|
|
"num_tokens": 5256806.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.838250923156738,
|
|
"epoch": 0.024878295165120953,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 5.6702,
|
|
"mean_token_accuracy": 0.16396461874246598,
|
|
"num_tokens": 5268088.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.776035308837891,
|
|
"epoch": 0.0249320279840521,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.7306,
|
|
"mean_token_accuracy": 0.16490367650985718,
|
|
"num_tokens": 5279339.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.823333358764648,
|
|
"epoch": 0.024985760802983245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 5.6935,
|
|
"mean_token_accuracy": 0.1696757972240448,
|
|
"num_tokens": 5289658.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.773639774322509,
|
|
"epoch": 0.025039493621914394,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.7281,
|
|
"mean_token_accuracy": 0.16653199195861818,
|
|
"num_tokens": 5300230.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.722403240203858,
|
|
"epoch": 0.02509322644084554,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 5.6704,
|
|
"mean_token_accuracy": 0.1663460001349449,
|
|
"num_tokens": 5311308.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.759013605117798,
|
|
"epoch": 0.025146959259776686,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.6438,
|
|
"mean_token_accuracy": 0.17089420258998872,
|
|
"num_tokens": 5321469.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.922854566574097,
|
|
"epoch": 0.025200692078707832,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.15669528990983964,
|
|
"num_tokens": 5334853.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.947241163253784,
|
|
"epoch": 0.02525442489763898,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 5.8178,
|
|
"mean_token_accuracy": 0.16408377140760422,
|
|
"num_tokens": 5345770.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.845251941680909,
|
|
"epoch": 0.025308157716570127,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 5.7921,
|
|
"mean_token_accuracy": 0.1645584896206856,
|
|
"num_tokens": 5357405.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.837396574020386,
|
|
"epoch": 0.025361890535501273,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.7584,
|
|
"mean_token_accuracy": 0.16327665448188783,
|
|
"num_tokens": 5369067.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.879635524749756,
|
|
"epoch": 0.02541562335443242,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 5.8431,
|
|
"mean_token_accuracy": 0.16264996975660323,
|
|
"num_tokens": 5380117.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.81128339767456,
|
|
"epoch": 0.02546935617336357,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 5.7605,
|
|
"mean_token_accuracy": 0.17122591882944108,
|
|
"num_tokens": 5393468.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.7529360294342045,
|
|
"epoch": 0.025523088992294714,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 5.7003,
|
|
"mean_token_accuracy": 0.16343179047107698,
|
|
"num_tokens": 5403712.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.805402183532715,
|
|
"epoch": 0.02557682181122586,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 5.697,
|
|
"mean_token_accuracy": 0.17064484506845473,
|
|
"num_tokens": 5414696.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.847384977340698,
|
|
"epoch": 0.025630554630157006,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 5.7572,
|
|
"mean_token_accuracy": 0.16016057580709459,
|
|
"num_tokens": 5425391.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.905871629714966,
|
|
"epoch": 0.025684287449088156,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.8393,
|
|
"mean_token_accuracy": 0.15981703624129295,
|
|
"num_tokens": 5437680.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.862912988662719,
|
|
"epoch": 0.0257380202680193,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.1575775146484375,
|
|
"num_tokens": 5448346.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.889981698989868,
|
|
"epoch": 0.025791753086950447,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.7787,
|
|
"mean_token_accuracy": 0.16920848041772843,
|
|
"num_tokens": 5461150.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.834473419189453,
|
|
"epoch": 0.025845485905881593,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.8376,
|
|
"mean_token_accuracy": 0.16043857261538505,
|
|
"num_tokens": 5473414.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.889663791656494,
|
|
"epoch": 0.025899218724812743,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 5.7165,
|
|
"mean_token_accuracy": 0.16560643464326857,
|
|
"num_tokens": 5484132.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.89264235496521,
|
|
"epoch": 0.02595295154374389,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.1657539889216423,
|
|
"num_tokens": 5495385.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.760320472717285,
|
|
"epoch": 0.026006684362675034,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 5.6835,
|
|
"mean_token_accuracy": 0.16800438314676286,
|
|
"num_tokens": 5506740.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.894057369232177,
|
|
"epoch": 0.02606041718160618,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.6918,
|
|
"mean_token_accuracy": 0.16996567100286483,
|
|
"num_tokens": 5518626.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.769460916519165,
|
|
"epoch": 0.02611415000053733,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 5.7068,
|
|
"mean_token_accuracy": 0.16771385669708253,
|
|
"num_tokens": 5529564.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.824889659881592,
|
|
"epoch": 0.026167882819468476,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 5.7681,
|
|
"mean_token_accuracy": 0.16524617820978166,
|
|
"num_tokens": 5541288.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.830038118362427,
|
|
"epoch": 0.02622161563839962,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.7364,
|
|
"mean_token_accuracy": 0.17004459947347642,
|
|
"num_tokens": 5551910.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.87531304359436,
|
|
"epoch": 0.026275348457330767,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 5.7712,
|
|
"mean_token_accuracy": 0.1673659548163414,
|
|
"num_tokens": 5562639.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.865144729614258,
|
|
"epoch": 0.026329081276261917,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.7446,
|
|
"mean_token_accuracy": 0.16738333106040953,
|
|
"num_tokens": 5573668.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.841930532455445,
|
|
"epoch": 0.026382814095193063,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.7847,
|
|
"mean_token_accuracy": 0.16369713842868805,
|
|
"num_tokens": 5586278.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.742566108703613,
|
|
"epoch": 0.02643654691412421,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.6992,
|
|
"mean_token_accuracy": 0.16847637295722961,
|
|
"num_tokens": 5596428.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.964372205734253,
|
|
"epoch": 0.026490279733055355,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 5.8889,
|
|
"mean_token_accuracy": 0.1610577628016472,
|
|
"num_tokens": 5609465.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.779795455932617,
|
|
"epoch": 0.026544012551986504,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 5.6764,
|
|
"mean_token_accuracy": 0.17105079293251038,
|
|
"num_tokens": 5621160.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.813624668121338,
|
|
"epoch": 0.02659774537091765,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.7573,
|
|
"mean_token_accuracy": 0.1618871957063675,
|
|
"num_tokens": 5633508.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.97997145652771,
|
|
"epoch": 0.026651478189848796,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.893,
|
|
"mean_token_accuracy": 0.1559185042977333,
|
|
"num_tokens": 5645762.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.839811944961548,
|
|
"epoch": 0.02670521100877994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 5.8108,
|
|
"mean_token_accuracy": 0.15802412182092668,
|
|
"num_tokens": 5656743.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.877661418914795,
|
|
"epoch": 0.02675894382771109,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.7867,
|
|
"mean_token_accuracy": 0.16386277824640275,
|
|
"num_tokens": 5668236.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.845249271392822,
|
|
"epoch": 0.026812676646642237,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 5.7663,
|
|
"mean_token_accuracy": 0.16282822340726852,
|
|
"num_tokens": 5680122.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.811351537704468,
|
|
"epoch": 0.026866409465573383,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.6952,
|
|
"mean_token_accuracy": 0.17274193465709686,
|
|
"num_tokens": 5691953.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.830418825149536,
|
|
"epoch": 0.02692014228450453,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 5.7449,
|
|
"mean_token_accuracy": 0.16308906078338622,
|
|
"num_tokens": 5703784.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.907339477539063,
|
|
"epoch": 0.026973875103435678,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 5.7023,
|
|
"mean_token_accuracy": 0.1719420537352562,
|
|
"num_tokens": 5714562.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.729317712783813,
|
|
"epoch": 0.027027607922366824,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 5.7346,
|
|
"mean_token_accuracy": 0.15874211341142655,
|
|
"num_tokens": 5726710.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.7773449420928955,
|
|
"epoch": 0.02708134074129797,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 5.722,
|
|
"mean_token_accuracy": 0.17372216433286666,
|
|
"num_tokens": 5737576.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.816823673248291,
|
|
"epoch": 0.027135073560229116,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 5.618,
|
|
"mean_token_accuracy": 0.1709082156419754,
|
|
"num_tokens": 5747961.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.802463388442993,
|
|
"epoch": 0.027188806379160265,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 5.6864,
|
|
"mean_token_accuracy": 0.17425455302000045,
|
|
"num_tokens": 5760361.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.82440767288208,
|
|
"epoch": 0.02724253919809141,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.16925353258848191,
|
|
"num_tokens": 5770506.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.823035717010498,
|
|
"epoch": 0.027296272017022557,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.16615234464406967,
|
|
"num_tokens": 5781568.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.964355993270874,
|
|
"epoch": 0.027350004835953703,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.15283150970935822,
|
|
"num_tokens": 5794829.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.861364936828613,
|
|
"epoch": 0.027403737654884852,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 5.6422,
|
|
"mean_token_accuracy": 0.17635300680994986,
|
|
"num_tokens": 5806847.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.787832498550415,
|
|
"epoch": 0.027457470473815998,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 5.6971,
|
|
"mean_token_accuracy": 0.1635927215218544,
|
|
"num_tokens": 5819214.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.815683078765869,
|
|
"epoch": 0.027511203292747144,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.163364876806736,
|
|
"num_tokens": 5830549.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.781494951248169,
|
|
"epoch": 0.02756493611167829,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 5.6615,
|
|
"mean_token_accuracy": 0.17006005346775055,
|
|
"num_tokens": 5842040.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.790893030166626,
|
|
"epoch": 0.02761866893060944,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 5.7145,
|
|
"mean_token_accuracy": 0.16810346096754075,
|
|
"num_tokens": 5854203.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.790723896026611,
|
|
"epoch": 0.027672401749540585,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 5.7404,
|
|
"mean_token_accuracy": 0.17112241089344024,
|
|
"num_tokens": 5866174.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.751659297943116,
|
|
"epoch": 0.02772613456847173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.7007,
|
|
"mean_token_accuracy": 0.17084870487451553,
|
|
"num_tokens": 5877427.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.867400598526001,
|
|
"epoch": 0.027779867387402877,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 5.7611,
|
|
"mean_token_accuracy": 0.16156804710626602,
|
|
"num_tokens": 5889089.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.807675743103028,
|
|
"epoch": 0.027833600206334026,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.7237,
|
|
"mean_token_accuracy": 0.16828633844852448,
|
|
"num_tokens": 5899638.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.89572172164917,
|
|
"epoch": 0.027887333025265172,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 5.8002,
|
|
"mean_token_accuracy": 0.16591026559472083,
|
|
"num_tokens": 5910953.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.86466498374939,
|
|
"epoch": 0.027941065844196318,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 5.7385,
|
|
"mean_token_accuracy": 0.1696275532245636,
|
|
"num_tokens": 5920767.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.974060535430908,
|
|
"epoch": 0.027994798663127464,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.9842,
|
|
"mean_token_accuracy": 0.155734646320343,
|
|
"num_tokens": 5933551.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.864262056350708,
|
|
"epoch": 0.028048531482058613,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 5.6967,
|
|
"mean_token_accuracy": 0.16535083651542665,
|
|
"num_tokens": 5944613.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.714416646957398,
|
|
"epoch": 0.02810226430098976,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.17641158998012543,
|
|
"num_tokens": 5956008.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.843830633163452,
|
|
"epoch": 0.028155997119920905,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 5.7116,
|
|
"mean_token_accuracy": 0.16503191590309144,
|
|
"num_tokens": 5967955.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.763233184814453,
|
|
"epoch": 0.02820972993885205,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 5.7026,
|
|
"mean_token_accuracy": 0.17177481353282928,
|
|
"num_tokens": 5979026.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.865948104858399,
|
|
"epoch": 0.0282634627577832,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.8077,
|
|
"mean_token_accuracy": 0.1681867554783821,
|
|
"num_tokens": 5990728.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.889879369735718,
|
|
"epoch": 0.028317195576714346,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.6878,
|
|
"mean_token_accuracy": 0.17442910820245744,
|
|
"num_tokens": 6003071.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.793350028991699,
|
|
"epoch": 0.028370928395645492,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.16766123175621034,
|
|
"num_tokens": 6014032.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.777959251403809,
|
|
"epoch": 0.028424661214576638,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 5.6644,
|
|
"mean_token_accuracy": 0.17170659005641936,
|
|
"num_tokens": 6024674.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.843295955657959,
|
|
"epoch": 0.028478394033507787,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 5.7395,
|
|
"mean_token_accuracy": 0.16094683408737182,
|
|
"num_tokens": 6037743.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.810171365737915,
|
|
"epoch": 0.028532126852438933,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.6552,
|
|
"mean_token_accuracy": 0.17111349254846572,
|
|
"num_tokens": 6049547.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.835242128372192,
|
|
"epoch": 0.02858585967137008,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.8632,
|
|
"mean_token_accuracy": 0.1614356316626072,
|
|
"num_tokens": 6062431.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.9649711608886715,
|
|
"epoch": 0.028639592490301225,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.8275,
|
|
"mean_token_accuracy": 0.16581905484199524,
|
|
"num_tokens": 6073728.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.8523218631744385,
|
|
"epoch": 0.028693325309232375,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.7007,
|
|
"mean_token_accuracy": 0.1676059991121292,
|
|
"num_tokens": 6085348.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.729227018356323,
|
|
"epoch": 0.02874705812816352,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 5.6908,
|
|
"mean_token_accuracy": 0.16612711995840074,
|
|
"num_tokens": 6095482.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.90185809135437,
|
|
"epoch": 0.028800790947094666,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.7635,
|
|
"mean_token_accuracy": 0.16939902752637864,
|
|
"num_tokens": 6107811.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.895412349700928,
|
|
"epoch": 0.028854523766025812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 5.704,
|
|
"mean_token_accuracy": 0.16830794066190718,
|
|
"num_tokens": 6118781.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.727690076828003,
|
|
"epoch": 0.02890825658495696,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.6942,
|
|
"mean_token_accuracy": 0.1738647624850273,
|
|
"num_tokens": 6129939.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.778635168075562,
|
|
"epoch": 0.028961989403888107,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.6864,
|
|
"mean_token_accuracy": 0.17006654441356658,
|
|
"num_tokens": 6140840.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.899269533157349,
|
|
"epoch": 0.029015722222819253,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 5.8274,
|
|
"mean_token_accuracy": 0.16288660019636153,
|
|
"num_tokens": 6152787.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.841928863525391,
|
|
"epoch": 0.0290694550417504,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.6858,
|
|
"mean_token_accuracy": 0.16882029473781585,
|
|
"num_tokens": 6164422.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.769184017181397,
|
|
"epoch": 0.02912318786068155,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.6329,
|
|
"mean_token_accuracy": 0.17495559453964232,
|
|
"num_tokens": 6175690.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.8205819606781,
|
|
"epoch": 0.029176920679612695,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 5.7193,
|
|
"mean_token_accuracy": 0.1662059471011162,
|
|
"num_tokens": 6187525.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.793738746643067,
|
|
"epoch": 0.02923065349854384,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.16315963119268417,
|
|
"num_tokens": 6199020.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.831929397583008,
|
|
"epoch": 0.029284386317474986,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.17417470663785933,
|
|
"num_tokens": 6210390.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.845892810821534,
|
|
"epoch": 0.029338119136406136,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 5.7524,
|
|
"mean_token_accuracy": 0.1651771053671837,
|
|
"num_tokens": 6221571.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.839717197418213,
|
|
"epoch": 0.02939185195533728,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.7071,
|
|
"mean_token_accuracy": 0.1717137351632118,
|
|
"num_tokens": 6232173.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.714461135864258,
|
|
"epoch": 0.029445584774268427,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 5.5408,
|
|
"mean_token_accuracy": 0.17940844148397445,
|
|
"num_tokens": 6242940.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.790823936462402,
|
|
"epoch": 0.029499317593199573,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.15964574813842775,
|
|
"num_tokens": 6254214.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.896879386901856,
|
|
"epoch": 0.029553050412130723,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.16712495535612107,
|
|
"num_tokens": 6266312.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.720922756195068,
|
|
"epoch": 0.02960678323106187,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 5.5601,
|
|
"mean_token_accuracy": 0.17704130113124847,
|
|
"num_tokens": 6276952.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.681634092330933,
|
|
"epoch": 0.029660516049993015,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.5425,
|
|
"mean_token_accuracy": 0.18228785395622255,
|
|
"num_tokens": 6288451.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.692604637145996,
|
|
"epoch": 0.02971424886892416,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.17026182562112807,
|
|
"num_tokens": 6300428.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.8079735279083256,
|
|
"epoch": 0.02976798168785531,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.17361425310373307,
|
|
"num_tokens": 6311076.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.854108428955078,
|
|
"epoch": 0.029821714506786456,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 5.7559,
|
|
"mean_token_accuracy": 0.1648738592863083,
|
|
"num_tokens": 6321488.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.868263912200928,
|
|
"epoch": 0.0298754473257176,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.7258,
|
|
"mean_token_accuracy": 0.16696933656930923,
|
|
"num_tokens": 6333391.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.776430988311768,
|
|
"epoch": 0.029929180144648748,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.6171,
|
|
"mean_token_accuracy": 0.17402495592832565,
|
|
"num_tokens": 6343366.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.770546245574951,
|
|
"epoch": 0.029982912963579897,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.7035,
|
|
"mean_token_accuracy": 0.16388753950595855,
|
|
"num_tokens": 6355118.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.840235853195191,
|
|
"epoch": 0.030036645782511043,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.16506970226764678,
|
|
"num_tokens": 6366592.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.801573181152344,
|
|
"epoch": 0.03009037860144219,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 5.6489,
|
|
"mean_token_accuracy": 0.16679887473583221,
|
|
"num_tokens": 6377356.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.935713386535644,
|
|
"epoch": 0.030144111420373335,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.7311,
|
|
"mean_token_accuracy": 0.16755720525979995,
|
|
"num_tokens": 6389853.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.819516944885254,
|
|
"epoch": 0.030197844239304484,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.6573,
|
|
"mean_token_accuracy": 0.17161076068878173,
|
|
"num_tokens": 6401971.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.782439947128296,
|
|
"epoch": 0.03025157705823563,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.6414,
|
|
"mean_token_accuracy": 0.17304342985153198,
|
|
"num_tokens": 6414522.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.771671438217163,
|
|
"epoch": 0.030305309877166776,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.6,
|
|
"mean_token_accuracy": 0.17084032744169236,
|
|
"num_tokens": 6425689.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.762628126144409,
|
|
"epoch": 0.03035904269609792,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.6508,
|
|
"mean_token_accuracy": 0.17409223765134813,
|
|
"num_tokens": 6435956.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.820963716506958,
|
|
"epoch": 0.03041277551502907,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.7355,
|
|
"mean_token_accuracy": 0.16871976107358932,
|
|
"num_tokens": 6448008.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.782035779953003,
|
|
"epoch": 0.030466508333960217,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.5977,
|
|
"mean_token_accuracy": 0.17293880730867386,
|
|
"num_tokens": 6459158.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.873963832855225,
|
|
"epoch": 0.030520241152891363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.7457,
|
|
"mean_token_accuracy": 0.16532379537820815,
|
|
"num_tokens": 6470021.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.882900381088257,
|
|
"epoch": 0.03057397397182251,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.6629,
|
|
"mean_token_accuracy": 0.17378174662590026,
|
|
"num_tokens": 6481073.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.76137409210205,
|
|
"epoch": 0.030627706790753658,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.6988,
|
|
"mean_token_accuracy": 0.16973501890897752,
|
|
"num_tokens": 6491206.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.790130376815796,
|
|
"epoch": 0.030681439609684804,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.689,
|
|
"mean_token_accuracy": 0.1693686291575432,
|
|
"num_tokens": 6503866.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.951168394088745,
|
|
"epoch": 0.03073517242861595,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 5.7766,
|
|
"mean_token_accuracy": 0.16254181861877443,
|
|
"num_tokens": 6516675.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.872813940048218,
|
|
"epoch": 0.030788905247547096,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.6967,
|
|
"mean_token_accuracy": 0.1711062654852867,
|
|
"num_tokens": 6527745.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.8216368675231935,
|
|
"epoch": 0.030842638066478245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.7151,
|
|
"mean_token_accuracy": 0.17183676958084107,
|
|
"num_tokens": 6540345.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.787891435623169,
|
|
"epoch": 0.03089637088540939,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.6882,
|
|
"mean_token_accuracy": 0.17021293342113494,
|
|
"num_tokens": 6551318.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.7962048053741455,
|
|
"epoch": 0.030950103704340537,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.6039,
|
|
"mean_token_accuracy": 0.17215406000614167,
|
|
"num_tokens": 6562644.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.814161491394043,
|
|
"epoch": 0.031003836523271683,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.6876,
|
|
"mean_token_accuracy": 0.17021351903676987,
|
|
"num_tokens": 6574482.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.787248516082764,
|
|
"epoch": 0.031057569342202832,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.5807,
|
|
"mean_token_accuracy": 0.17575011104345323,
|
|
"num_tokens": 6583981.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.745701742172241,
|
|
"epoch": 0.031111302161133978,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.5538,
|
|
"mean_token_accuracy": 0.17817822694778443,
|
|
"num_tokens": 6595822.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.687450790405274,
|
|
"epoch": 0.031165034980065124,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.5734,
|
|
"mean_token_accuracy": 0.1703382194042206,
|
|
"num_tokens": 6606845.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.75296540260315,
|
|
"epoch": 0.03121876779899627,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.17386566251516342,
|
|
"num_tokens": 6618205.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.84825644493103,
|
|
"epoch": 0.03127250061792742,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.7194,
|
|
"mean_token_accuracy": 0.17248747050762175,
|
|
"num_tokens": 6629674.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.825883388519287,
|
|
"epoch": 0.031326233436858565,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.6873,
|
|
"mean_token_accuracy": 0.17095083147287368,
|
|
"num_tokens": 6641322.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.823066997528076,
|
|
"epoch": 0.03137996625578971,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.7467,
|
|
"mean_token_accuracy": 0.1705140456557274,
|
|
"num_tokens": 6653757.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.814864587783814,
|
|
"epoch": 0.03143369907472086,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.6443,
|
|
"mean_token_accuracy": 0.17340247631072997,
|
|
"num_tokens": 6664822.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.811478853225708,
|
|
"epoch": 0.031487431893652,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.6877,
|
|
"mean_token_accuracy": 0.17079921811819077,
|
|
"num_tokens": 6676020.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.771947622299194,
|
|
"epoch": 0.03154116471258315,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.5625,
|
|
"mean_token_accuracy": 0.17955648005008698,
|
|
"num_tokens": 6687920.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.814941024780273,
|
|
"epoch": 0.0315948975315143,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.7455,
|
|
"mean_token_accuracy": 0.1686069130897522,
|
|
"num_tokens": 6699615.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.77859902381897,
|
|
"epoch": 0.03164863035044545,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.16882998496294022,
|
|
"num_tokens": 6710618.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.739752912521363,
|
|
"epoch": 0.03170236316937659,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.5908,
|
|
"mean_token_accuracy": 0.17599669694900513,
|
|
"num_tokens": 6722140.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.827819919586181,
|
|
"epoch": 0.03175609598830774,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.676,
|
|
"mean_token_accuracy": 0.1703364685177803,
|
|
"num_tokens": 6733732.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.767013025283814,
|
|
"epoch": 0.031809828807238885,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.6357,
|
|
"mean_token_accuracy": 0.1753098413348198,
|
|
"num_tokens": 6745488.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.748724889755249,
|
|
"epoch": 0.03186356162617003,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.17133375257253647,
|
|
"num_tokens": 6756972.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.8181640148162845,
|
|
"epoch": 0.03191729444510118,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 5.6871,
|
|
"mean_token_accuracy": 0.17716166377067566,
|
|
"num_tokens": 6768223.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.847131967544556,
|
|
"epoch": 0.03197102726403232,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.7234,
|
|
"mean_token_accuracy": 0.17008933573961257,
|
|
"num_tokens": 6778840.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.786890029907227,
|
|
"epoch": 0.032024760082963476,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.6755,
|
|
"mean_token_accuracy": 0.1672493740916252,
|
|
"num_tokens": 6790408.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.754341411590576,
|
|
"epoch": 0.03207849290189462,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.5667,
|
|
"mean_token_accuracy": 0.17534744888544082,
|
|
"num_tokens": 6802523.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.7777200698852536,
|
|
"epoch": 0.03213222572082577,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.6491,
|
|
"mean_token_accuracy": 0.17468351870775223,
|
|
"num_tokens": 6812665.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.749584913253784,
|
|
"epoch": 0.03218595853975691,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.5756,
|
|
"mean_token_accuracy": 0.17782480120658875,
|
|
"num_tokens": 6822691.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.814772415161133,
|
|
"epoch": 0.03223969135868806,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.7451,
|
|
"mean_token_accuracy": 0.16809259802103044,
|
|
"num_tokens": 6834832.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.879122114181518,
|
|
"epoch": 0.032293424177619205,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 5.7432,
|
|
"mean_token_accuracy": 0.1686186045408249,
|
|
"num_tokens": 6847099.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.809540796279907,
|
|
"epoch": 0.03234715699655035,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001606685517817114,
|
|
"loss": 5.6662,
|
|
"mean_token_accuracy": 0.172978737950325,
|
|
"num_tokens": 6859383.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.75958514213562,
|
|
"epoch": 0.0324008898154815,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00015965545838587592,
|
|
"loss": 5.5918,
|
|
"mean_token_accuracy": 0.17844594419002532,
|
|
"num_tokens": 6872107.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.701963949203491,
|
|
"epoch": 0.03245462263441265,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00015864552721891467,
|
|
"loss": 5.5756,
|
|
"mean_token_accuracy": 0.17398642003536224,
|
|
"num_tokens": 6882737.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.754208183288574,
|
|
"epoch": 0.032508355453343796,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00015763878596860076,
|
|
"loss": 5.6366,
|
|
"mean_token_accuracy": 0.1801183134317398,
|
|
"num_tokens": 6893955.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.7247887134552,
|
|
"epoch": 0.03256208827227494,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00015663526223525412,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.1824837163090706,
|
|
"num_tokens": 6904446.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.722404098510742,
|
|
"epoch": 0.03261582109120609,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001556349835309848,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.17439327239990235,
|
|
"num_tokens": 6915755.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.8725292682647705,
|
|
"epoch": 0.03266955391013723,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001546379772789389,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.16626499742269515,
|
|
"num_tokens": 6928801.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.904501581192017,
|
|
"epoch": 0.03272328672906838,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00015364427081254622,
|
|
"loss": 5.7857,
|
|
"mean_token_accuracy": 0.16541238874197006,
|
|
"num_tokens": 6941294.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.788457822799683,
|
|
"epoch": 0.032777019547999525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00015265389137477165,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.17478950172662736,
|
|
"num_tokens": 6954432.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.76705846786499,
|
|
"epoch": 0.03283075236693067,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00015166686611736786,
|
|
"loss": 5.6491,
|
|
"mean_token_accuracy": 0.1729201301932335,
|
|
"num_tokens": 6965335.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.823248195648193,
|
|
"epoch": 0.032884485185861824,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00015068322210013064,
|
|
"loss": 5.7631,
|
|
"mean_token_accuracy": 0.16607172042131424,
|
|
"num_tokens": 6977440.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.735596942901611,
|
|
"epoch": 0.03293821800479297,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 5.6161,
|
|
"mean_token_accuracy": 0.17930434346199037,
|
|
"num_tokens": 6989210.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.866488742828369,
|
|
"epoch": 0.032991950823724116,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00014872618556110905,
|
|
"loss": 5.6381,
|
|
"mean_token_accuracy": 0.17192607820034028,
|
|
"num_tokens": 6999916.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.782448720932007,
|
|
"epoch": 0.03304568364265526,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00014775284669246992,
|
|
"loss": 5.5544,
|
|
"mean_token_accuracy": 0.17496816217899322,
|
|
"num_tokens": 7011533.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.813140058517456,
|
|
"epoch": 0.03309941646158641,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00014678299636881716,
|
|
"loss": 5.7025,
|
|
"mean_token_accuracy": 0.17206587344408036,
|
|
"num_tokens": 7023407.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.8577446937561035,
|
|
"epoch": 0.03315314928051755,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001458166611790873,
|
|
"loss": 5.6627,
|
|
"mean_token_accuracy": 0.16903851330280303,
|
|
"num_tokens": 7034233.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.7842785835266115,
|
|
"epoch": 0.0332068820994487,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00014485386761584773,
|
|
"loss": 5.6231,
|
|
"mean_token_accuracy": 0.17471387684345246,
|
|
"num_tokens": 7046941.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.754002046585083,
|
|
"epoch": 0.033260614918379845,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00014389464207457042,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.17619651556015015,
|
|
"num_tokens": 7057828.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.936915636062622,
|
|
"epoch": 0.033314347737311,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00014293901085290795,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.16599795669317247,
|
|
"num_tokens": 7070417.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.807844352722168,
|
|
"epoch": 0.033368080556242144,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00014198700014997307,
|
|
"loss": 5.651,
|
|
"mean_token_accuracy": 0.1716989740729332,
|
|
"num_tokens": 7082284.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.752397727966309,
|
|
"epoch": 0.03342181337517329,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00014103863606562016,
|
|
"loss": 5.5705,
|
|
"mean_token_accuracy": 0.17725420147180557,
|
|
"num_tokens": 7093586.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.797366619110107,
|
|
"epoch": 0.033475546194104436,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00014009394459972964,
|
|
"loss": 5.6056,
|
|
"mean_token_accuracy": 0.18163676410913468,
|
|
"num_tokens": 7104547.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.806517124176025,
|
|
"epoch": 0.03352927901303558,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00013915295165149513,
|
|
"loss": 5.6489,
|
|
"mean_token_accuracy": 0.17738565504550935,
|
|
"num_tokens": 7115412.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.755319452285766,
|
|
"epoch": 0.03358301183196673,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.17332856655120848,
|
|
"num_tokens": 7126116.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.779922246932983,
|
|
"epoch": 0.03363674465089787,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00013728216439707862,
|
|
"loss": 5.6088,
|
|
"mean_token_accuracy": 0.1755072921514511,
|
|
"num_tokens": 7137085.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.773222780227661,
|
|
"epoch": 0.03369047746982902,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00013635242137947419,
|
|
"loss": 5.671,
|
|
"mean_token_accuracy": 0.17120273262262345,
|
|
"num_tokens": 7149166.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.8588690757751465,
|
|
"epoch": 0.03374421028876017,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00013542647945527498,
|
|
"loss": 5.741,
|
|
"mean_token_accuracy": 0.16475639045238494,
|
|
"num_tokens": 7161753.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.765697622299195,
|
|
"epoch": 0.03379794310769132,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001345043640096465,
|
|
"loss": 5.5746,
|
|
"mean_token_accuracy": 0.18044933080673217,
|
|
"num_tokens": 7172846.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.804821014404297,
|
|
"epoch": 0.033851675926622464,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00013358610032284957,
|
|
"loss": 5.6505,
|
|
"mean_token_accuracy": 0.16912970095872878,
|
|
"num_tokens": 7185362.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.7809232711792,
|
|
"epoch": 0.03390540874555361,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000132671713569547,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.17532532811164855,
|
|
"num_tokens": 7195499.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.746611261367798,
|
|
"epoch": 0.033959141564484756,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001317612288181136,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.17372410297393798,
|
|
"num_tokens": 7205933.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.761715030670166,
|
|
"epoch": 0.0340128743834159,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00013085467102994864,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.16941637992858888,
|
|
"num_tokens": 7217291.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.829409122467041,
|
|
"epoch": 0.03406660720234705,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00012995206505879198,
|
|
"loss": 5.6814,
|
|
"mean_token_accuracy": 0.17496579736471177,
|
|
"num_tokens": 7228683.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.700780582427979,
|
|
"epoch": 0.034120340021278193,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0001290534356500421,
|
|
"loss": 5.5274,
|
|
"mean_token_accuracy": 0.17498208582401276,
|
|
"num_tokens": 7240417.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.7706043243408205,
|
|
"epoch": 0.034174072840209346,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00012815880744007827,
|
|
"loss": 5.5975,
|
|
"mean_token_accuracy": 0.17494024485349655,
|
|
"num_tokens": 7251295.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.818415975570678,
|
|
"epoch": 0.03422780565914049,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 5.6336,
|
|
"mean_token_accuracy": 0.17472454756498337,
|
|
"num_tokens": 7261338.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.771653413772583,
|
|
"epoch": 0.03428153847807164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00012638165261287868,
|
|
"loss": 5.5941,
|
|
"mean_token_accuracy": 0.17674591839313508,
|
|
"num_tokens": 7273010.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.782175970077515,
|
|
"epoch": 0.034335271297002784,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001254991747172402,
|
|
"loss": 5.6524,
|
|
"mean_token_accuracy": 0.1713513270020485,
|
|
"num_tokens": 7284995.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.79613471031189,
|
|
"epoch": 0.03438900411593393,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00012462079546224662,
|
|
"loss": 5.6585,
|
|
"mean_token_accuracy": 0.17008266299962999,
|
|
"num_tokens": 7296155.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.7490638256072994,
|
|
"epoch": 0.034442736934865076,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00012374653892910896,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.17523309141397475,
|
|
"num_tokens": 7307035.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.831938219070435,
|
|
"epoch": 0.03449646975379622,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00012287642908601166,
|
|
"loss": 5.6464,
|
|
"mean_token_accuracy": 0.1761499047279358,
|
|
"num_tokens": 7318849.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.825429964065552,
|
|
"epoch": 0.03455020257272737,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00012201048978745569,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.17673419117927552,
|
|
"num_tokens": 7329458.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.799201917648316,
|
|
"epoch": 0.03460393539165852,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00012114874477360427,
|
|
"loss": 5.7209,
|
|
"mean_token_accuracy": 0.16919921189546586,
|
|
"num_tokens": 7340588.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.804809856414795,
|
|
"epoch": 0.034657668210589666,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00012029121766963236,
|
|
"loss": 5.6913,
|
|
"mean_token_accuracy": 0.1678548648953438,
|
|
"num_tokens": 7351211.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.84837646484375,
|
|
"epoch": 0.03471140102952081,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011943793198507858,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.17412487864494325,
|
|
"num_tokens": 7363254.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.826323699951172,
|
|
"epoch": 0.03476513384845196,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00011858891111320104,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.17825008630752565,
|
|
"num_tokens": 7375139.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.876450872421264,
|
|
"epoch": 0.034818866667383104,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001177441783303359,
|
|
"loss": 5.6919,
|
|
"mean_token_accuracy": 0.16714323312044144,
|
|
"num_tokens": 7386475.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.7469034671783445,
|
|
"epoch": 0.03487259948631425,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.17687939554452897,
|
|
"num_tokens": 7397423.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.82618727684021,
|
|
"epoch": 0.034926332305245396,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00011606766954855124,
|
|
"loss": 5.7281,
|
|
"mean_token_accuracy": 0.1714285656809807,
|
|
"num_tokens": 7410062.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.887579727172851,
|
|
"epoch": 0.03498006512417654,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011523593951196702,
|
|
"loss": 5.6291,
|
|
"mean_token_accuracy": 0.17846414148807527,
|
|
"num_tokens": 7421353.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.822369956970215,
|
|
"epoch": 0.035033797943107695,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011440858948780523,
|
|
"loss": 5.6408,
|
|
"mean_token_accuracy": 0.1749933511018753,
|
|
"num_tokens": 7432413.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.753646183013916,
|
|
"epoch": 0.03508753076203884,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00011358564215828484,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.17465341687202454,
|
|
"num_tokens": 7443804.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.782634449005127,
|
|
"epoch": 0.035141263580969986,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00011276712008492254,
|
|
"loss": 5.6561,
|
|
"mean_token_accuracy": 0.16764227598905562,
|
|
"num_tokens": 7455668.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.745923662185669,
|
|
"epoch": 0.03519499639990113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011195304570791451,
|
|
"loss": 5.5339,
|
|
"mean_token_accuracy": 0.179062819480896,
|
|
"num_tokens": 7467557.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.802323675155639,
|
|
"epoch": 0.03524872921883228,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00011114344134552094,
|
|
"loss": 5.5955,
|
|
"mean_token_accuracy": 0.17347132861614228,
|
|
"num_tokens": 7478099.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.8616753101348875,
|
|
"epoch": 0.035302462037763424,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001103383291934545,
|
|
"loss": 5.6677,
|
|
"mean_token_accuracy": 0.16931651681661605,
|
|
"num_tokens": 7490065.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.802161121368409,
|
|
"epoch": 0.03535619485669457,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00010953773132427141,
|
|
"loss": 5.6849,
|
|
"mean_token_accuracy": 0.17054295390844346,
|
|
"num_tokens": 7502727.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.754947185516357,
|
|
"epoch": 0.035409927675625716,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00010874166968676677,
|
|
"loss": 5.644,
|
|
"mean_token_accuracy": 0.17221480011940002,
|
|
"num_tokens": 7514295.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.854808378219604,
|
|
"epoch": 0.03546366049455687,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00010795016610537251,
|
|
"loss": 5.5843,
|
|
"mean_token_accuracy": 0.17890419960021972,
|
|
"num_tokens": 7524939.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.727210426330567,
|
|
"epoch": 0.035517393313488015,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.17440043687820433,
|
|
"num_tokens": 7536201.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.775128269195557,
|
|
"epoch": 0.03557112613241916,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001063809197832406,
|
|
"loss": 5.5964,
|
|
"mean_token_accuracy": 0.1797577530145645,
|
|
"num_tokens": 7547457.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.629309749603271,
|
|
"epoch": 0.035624858951350306,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00010560322006418368,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.17570652812719345,
|
|
"num_tokens": 7558359.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.793405199050904,
|
|
"epoch": 0.03567859177028145,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00010483016444341887,
|
|
"loss": 5.707,
|
|
"mean_token_accuracy": 0.16954949051141738,
|
|
"num_tokens": 7571021.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.7984151363372805,
|
|
"epoch": 0.0357323245892126,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00010406177411465654,
|
|
"loss": 5.6361,
|
|
"mean_token_accuracy": 0.1689628168940544,
|
|
"num_tokens": 7582400.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.83315076828003,
|
|
"epoch": 0.035786057408143744,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00010329807014370562,
|
|
"loss": 5.6391,
|
|
"mean_token_accuracy": 0.17905323654413224,
|
|
"num_tokens": 7593579.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.8320074558258055,
|
|
"epoch": 0.03583979022707489,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00010253907346789632,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.17714599519968033,
|
|
"num_tokens": 7605034.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.885298061370849,
|
|
"epoch": 0.03589352304600604,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010178480489550596,
|
|
"loss": 5.6957,
|
|
"mean_token_accuracy": 0.17411292642354964,
|
|
"num_tokens": 7616349.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.794470024108887,
|
|
"epoch": 0.03594725586493719,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00010103528510518836,
|
|
"loss": 5.6984,
|
|
"mean_token_accuracy": 0.17186133563518524,
|
|
"num_tokens": 7627065.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.879992866516114,
|
|
"epoch": 0.036000988683868335,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001002905346454073,
|
|
"loss": 5.6445,
|
|
"mean_token_accuracy": 0.17213908731937408,
|
|
"num_tokens": 7638461.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.7958722591400145,
|
|
"epoch": 0.03605472150279948,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 9.955057393387285e-05,
|
|
"loss": 5.6895,
|
|
"mean_token_accuracy": 0.16624573320150376,
|
|
"num_tokens": 7651159.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.725649976730347,
|
|
"epoch": 0.036108454321730626,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.88154232569816e-05,
|
|
"loss": 5.5473,
|
|
"mean_token_accuracy": 0.17729577124118806,
|
|
"num_tokens": 7663391.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.780076742172241,
|
|
"epoch": 0.03616218714066177,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 5.6865,
|
|
"mean_token_accuracy": 0.17146698832511903,
|
|
"num_tokens": 7676476.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.877428340911865,
|
|
"epoch": 0.03621591995959292,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 9.735963249281549e-05,
|
|
"loss": 5.6852,
|
|
"mean_token_accuracy": 0.1688380479812622,
|
|
"num_tokens": 7688398.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.889744424819947,
|
|
"epoch": 0.036269652778524064,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 9.663903231677974e-05,
|
|
"loss": 5.6981,
|
|
"mean_token_accuracy": 0.1683603510260582,
|
|
"num_tokens": 7700375.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.774162483215332,
|
|
"epoch": 0.03632338559745522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.592332199677145e-05,
|
|
"loss": 5.5592,
|
|
"mean_token_accuracy": 0.17962543070316314,
|
|
"num_tokens": 7711353.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.767346572875977,
|
|
"epoch": 0.03637711841638636,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 9.521252115435061e-05,
|
|
"loss": 5.6003,
|
|
"mean_token_accuracy": 0.17228911817073822,
|
|
"num_tokens": 7722943.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.7835533618927,
|
|
"epoch": 0.03643085123531751,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 9.450664927648126e-05,
|
|
"loss": 5.6445,
|
|
"mean_token_accuracy": 0.16676168739795685,
|
|
"num_tokens": 7735861.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.767013740539551,
|
|
"epoch": 0.036484584054248655,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 9.380572571499758e-05,
|
|
"loss": 5.5983,
|
|
"mean_token_accuracy": 0.17627354115247726,
|
|
"num_tokens": 7747657.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.811153173446655,
|
|
"epoch": 0.0365383168731798,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.310976968607307e-05,
|
|
"loss": 5.6362,
|
|
"mean_token_accuracy": 0.17387233227491378,
|
|
"num_tokens": 7758550.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.8585106372833256,
|
|
"epoch": 0.036592049692110946,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.241880026969381e-05,
|
|
"loss": 5.6643,
|
|
"mean_token_accuracy": 0.16933151781558992,
|
|
"num_tokens": 7769960.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.807639217376709,
|
|
"epoch": 0.03664578251104209,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.173283640913537e-05,
|
|
"loss": 5.5443,
|
|
"mean_token_accuracy": 0.17946419566869737,
|
|
"num_tokens": 7780944.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.841978120803833,
|
|
"epoch": 0.03669951532997324,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.10518969104436e-05,
|
|
"loss": 5.7299,
|
|
"mean_token_accuracy": 0.1680121436715126,
|
|
"num_tokens": 7793036.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.768467807769776,
|
|
"epoch": 0.03675324814890439,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.037600044191868e-05,
|
|
"loss": 5.5485,
|
|
"mean_token_accuracy": 0.18244272321462632,
|
|
"num_tokens": 7804073.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.842762088775634,
|
|
"epoch": 0.03680698096783554,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.17053049504756929,
|
|
"num_tokens": 7816288.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.767947149276734,
|
|
"epoch": 0.03686071378676668,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 8.903941057677692e-05,
|
|
"loss": 5.6282,
|
|
"mean_token_accuracy": 0.17155197709798814,
|
|
"num_tokens": 7827158.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.770988702774048,
|
|
"epoch": 0.03691444660569783,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 8.837875382344635e-05,
|
|
"loss": 5.5902,
|
|
"mean_token_accuracy": 0.17665452212095262,
|
|
"num_tokens": 7838696.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.80814151763916,
|
|
"epoch": 0.036968179424628975,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.772321338585076e-05,
|
|
"loss": 5.6133,
|
|
"mean_token_accuracy": 0.17915503233671187,
|
|
"num_tokens": 7851079.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.864059543609619,
|
|
"epoch": 0.03702191224356012,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.707280723596242e-05,
|
|
"loss": 5.6215,
|
|
"mean_token_accuracy": 0.17463277131319047,
|
|
"num_tokens": 7861910.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.80171217918396,
|
|
"epoch": 0.037075645062491266,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 8.64275532049944e-05,
|
|
"loss": 5.6432,
|
|
"mean_token_accuracy": 0.17207258343696594,
|
|
"num_tokens": 7874510.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.817556524276734,
|
|
"epoch": 0.03712937788142241,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 8.578746898291198e-05,
|
|
"loss": 5.6627,
|
|
"mean_token_accuracy": 0.17624083608388902,
|
|
"num_tokens": 7885824.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.83292145729065,
|
|
"epoch": 0.037183110700353565,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 8.515257211794742e-05,
|
|
"loss": 5.6517,
|
|
"mean_token_accuracy": 0.18418766111135482,
|
|
"num_tokens": 7897499.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.743706321716308,
|
|
"epoch": 0.03723684351928471,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.452288001611896e-05,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.17062459141016006,
|
|
"num_tokens": 7908275.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.854253244400025,
|
|
"epoch": 0.03729057633821586,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.389840994075379e-05,
|
|
"loss": 5.6386,
|
|
"mean_token_accuracy": 0.17409906834363936,
|
|
"num_tokens": 7920491.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.893707275390625,
|
|
"epoch": 0.037344309157147,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 8.327917901201435e-05,
|
|
"loss": 5.632,
|
|
"mean_token_accuracy": 0.17331487983465194,
|
|
"num_tokens": 7931983.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.759643268585205,
|
|
"epoch": 0.03739804197607815,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 8.266520420642931e-05,
|
|
"loss": 5.6381,
|
|
"mean_token_accuracy": 0.17134664058685303,
|
|
"num_tokens": 7945087.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.744393682479858,
|
|
"epoch": 0.037451774795009295,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 5.5063,
|
|
"mean_token_accuracy": 0.18321227431297302,
|
|
"num_tokens": 7956549.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.889737319946289,
|
|
"epoch": 0.03750550761394044,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.145309014987978e-05,
|
|
"loss": 5.6793,
|
|
"mean_token_accuracy": 0.17564765959978104,
|
|
"num_tokens": 7968002.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.7995827198028564,
|
|
"epoch": 0.037559240432871586,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 8.085498412963437e-05,
|
|
"loss": 5.6604,
|
|
"mean_token_accuracy": 0.17089386582374572,
|
|
"num_tokens": 7980666.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.72282452583313,
|
|
"epoch": 0.03761297325180274,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.026220069307078e-05,
|
|
"loss": 5.4693,
|
|
"mean_token_accuracy": 0.18240586519241334,
|
|
"num_tokens": 7990619.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.824265718460083,
|
|
"epoch": 0.037666706070733885,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.967475609164621e-05,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.17272981256246567,
|
|
"num_tokens": 8003184.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.746862268447876,
|
|
"epoch": 0.03772043888966503,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7.909266643045124e-05,
|
|
"loss": 5.5488,
|
|
"mean_token_accuracy": 0.18024033904075623,
|
|
"num_tokens": 8014329.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.746979808807373,
|
|
"epoch": 0.03777417170859618,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 7.851594766776802e-05,
|
|
"loss": 5.6289,
|
|
"mean_token_accuracy": 0.1784014567732811,
|
|
"num_tokens": 8026362.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.810842418670655,
|
|
"epoch": 0.03782790452752732,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 7.794461561463265e-05,
|
|
"loss": 5.6743,
|
|
"mean_token_accuracy": 0.17631909102201462,
|
|
"num_tokens": 8037362.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.8603888034820555,
|
|
"epoch": 0.03788163734645847,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 7.7378685934402e-05,
|
|
"loss": 5.7505,
|
|
"mean_token_accuracy": 0.1612520270049572,
|
|
"num_tokens": 8049482.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.7806295394897464,
|
|
"epoch": 0.037935370165389615,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.68181741423242e-05,
|
|
"loss": 5.5252,
|
|
"mean_token_accuracy": 0.18271390795707704,
|
|
"num_tokens": 8060364.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.740333223342896,
|
|
"epoch": 0.03798910298432076,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.626309560511313e-05,
|
|
"loss": 5.5889,
|
|
"mean_token_accuracy": 0.18059088289737701,
|
|
"num_tokens": 8071092.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.851498174667358,
|
|
"epoch": 0.03804283580325191,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 7.571346554052724e-05,
|
|
"loss": 5.6867,
|
|
"mean_token_accuracy": 0.17384813129901885,
|
|
"num_tokens": 8081522.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.758526754379273,
|
|
"epoch": 0.03809656862218306,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 7.516929901695249e-05,
|
|
"loss": 5.6254,
|
|
"mean_token_accuracy": 0.17677592635154724,
|
|
"num_tokens": 8092860.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.8255006790161135,
|
|
"epoch": 0.038150301441114205,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 7.463061095298893e-05,
|
|
"loss": 5.6699,
|
|
"mean_token_accuracy": 0.1656552657485008,
|
|
"num_tokens": 8105472.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.807037448883056,
|
|
"epoch": 0.03820403426004535,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 7.409741611704198e-05,
|
|
"loss": 5.6033,
|
|
"mean_token_accuracy": 0.17878384441137313,
|
|
"num_tokens": 8116843.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.7983581066131595,
|
|
"epoch": 0.0382577670789765,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 7.35697291269174e-05,
|
|
"loss": 5.5139,
|
|
"mean_token_accuracy": 0.1741447627544403,
|
|
"num_tokens": 8128369.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.785706186294556,
|
|
"epoch": 0.03831149989790764,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 7.304756444942056e-05,
|
|
"loss": 5.6086,
|
|
"mean_token_accuracy": 0.17580868154764176,
|
|
"num_tokens": 8140700.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.794817876815796,
|
|
"epoch": 0.03836523271683879,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7.253093639995994e-05,
|
|
"loss": 5.5874,
|
|
"mean_token_accuracy": 0.17722944021224976,
|
|
"num_tokens": 8150561.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.788495349884033,
|
|
"epoch": 0.038418965535769935,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 7.20198591421544e-05,
|
|
"loss": 5.6176,
|
|
"mean_token_accuracy": 0.17027283161878587,
|
|
"num_tokens": 8163534.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.8203753471374515,
|
|
"epoch": 0.03847269835470109,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.151434668744517e-05,
|
|
"loss": 5.6604,
|
|
"mean_token_accuracy": 0.17337529361248016,
|
|
"num_tokens": 8175304.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.8935041427612305,
|
|
"epoch": 0.03852643117363223,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.101441289471153e-05,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.16810978204011917,
|
|
"num_tokens": 8186796.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.888511705398559,
|
|
"epoch": 0.03858016399256338,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 7.052007146989098e-05,
|
|
"loss": 5.5974,
|
|
"mean_token_accuracy": 0.17807104885578157,
|
|
"num_tokens": 8197660.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.87249174118042,
|
|
"epoch": 0.038633896811494525,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.003133596560341e-05,
|
|
"loss": 5.6622,
|
|
"mean_token_accuracy": 0.17983966916799546,
|
|
"num_tokens": 8208340.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.813354730606079,
|
|
"epoch": 0.03868762963042567,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.954821978077952e-05,
|
|
"loss": 5.621,
|
|
"mean_token_accuracy": 0.18003249615430833,
|
|
"num_tokens": 8219263.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.821247959136963,
|
|
"epoch": 0.03874136244935682,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 6.907073616029356e-05,
|
|
"loss": 5.66,
|
|
"mean_token_accuracy": 0.1719284772872925,
|
|
"num_tokens": 8230883.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.808666372299195,
|
|
"epoch": 0.03879509526828796,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 6.85988981946002e-05,
|
|
"loss": 5.6595,
|
|
"mean_token_accuracy": 0.1744724154472351,
|
|
"num_tokens": 8242260.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.7251993179321286,
|
|
"epoch": 0.03884882808721911,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.813271881937564e-05,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.17448703944683075,
|
|
"num_tokens": 8253895.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.801276636123657,
|
|
"epoch": 0.03890256090615026,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.767221081516286e-05,
|
|
"loss": 5.6788,
|
|
"mean_token_accuracy": 0.1719422921538353,
|
|
"num_tokens": 8265458.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.821825838088989,
|
|
"epoch": 0.03895629372508141,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 6.72173868070215e-05,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.17571719735860825,
|
|
"num_tokens": 8277685.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.768735551834107,
|
|
"epoch": 0.039010026544012553,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.676825926418149e-05,
|
|
"loss": 5.482,
|
|
"mean_token_accuracy": 0.17471762746572495,
|
|
"num_tokens": 8289218.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.7693956851959225,
|
|
"epoch": 0.0390637593629437,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 6.632484049970122e-05,
|
|
"loss": 5.5836,
|
|
"mean_token_accuracy": 0.1779899999499321,
|
|
"num_tokens": 8299683.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.740936708450318,
|
|
"epoch": 0.039117492181874845,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 6.588714267013019e-05,
|
|
"loss": 5.5226,
|
|
"mean_token_accuracy": 0.18451472520828247,
|
|
"num_tokens": 8311582.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.751329851150513,
|
|
"epoch": 0.03917122500080599,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 6.545517777517544e-05,
|
|
"loss": 5.636,
|
|
"mean_token_accuracy": 0.1704813465476036,
|
|
"num_tokens": 8323835.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.800146436691284,
|
|
"epoch": 0.03922495781973714,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 6.502895765737281e-05,
|
|
"loss": 5.6464,
|
|
"mean_token_accuracy": 0.17180621176958083,
|
|
"num_tokens": 8335599.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.689224481582642,
|
|
"epoch": 0.03927869063866828,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.460849400176212e-05,
|
|
"loss": 5.4794,
|
|
"mean_token_accuracy": 0.18157135546207429,
|
|
"num_tokens": 8346651.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.801444673538208,
|
|
"epoch": 0.039332423457599436,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 6.419379833556694e-05,
|
|
"loss": 5.6728,
|
|
"mean_token_accuracy": 0.17194918543100357,
|
|
"num_tokens": 8358013.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.819459056854248,
|
|
"epoch": 0.03938615627653058,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.378488202787835e-05,
|
|
"loss": 5.6362,
|
|
"mean_token_accuracy": 0.1684793710708618,
|
|
"num_tokens": 8369611.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.761497831344604,
|
|
"epoch": 0.03943988909546173,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 6.33817562893435e-05,
|
|
"loss": 5.5918,
|
|
"mean_token_accuracy": 0.18165694177150726,
|
|
"num_tokens": 8380514.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.774696111679077,
|
|
"epoch": 0.039493621914392874,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.29844321718582e-05,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.1809420645236969,
|
|
"num_tokens": 8391291.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.896197128295898,
|
|
"epoch": 0.03954735473332402,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 6.259292056826383e-05,
|
|
"loss": 5.6944,
|
|
"mean_token_accuracy": 0.17697725743055343,
|
|
"num_tokens": 8402521.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.816521263122558,
|
|
"epoch": 0.039601087552255165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 6.220723221204873e-05,
|
|
"loss": 5.6053,
|
|
"mean_token_accuracy": 0.17517307102680207,
|
|
"num_tokens": 8414217.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.849808406829834,
|
|
"epoch": 0.03965482037118631,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.182737767705406e-05,
|
|
"loss": 5.6606,
|
|
"mean_token_accuracy": 0.17844959944486619,
|
|
"num_tokens": 8425812.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.725746393203735,
|
|
"epoch": 0.03970855319011746,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 6.145336737718375e-05,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.18177959322929382,
|
|
"num_tokens": 8436865.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.781955242156982,
|
|
"epoch": 0.03976228600904861,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 6.10852115661191e-05,
|
|
"loss": 5.547,
|
|
"mean_token_accuracy": 0.18054633289575578,
|
|
"num_tokens": 8448754.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.793501853942871,
|
|
"epoch": 0.039816018827979756,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 6.072292033703766e-05,
|
|
"loss": 5.5597,
|
|
"mean_token_accuracy": 0.1750421479344368,
|
|
"num_tokens": 8460251.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.81627140045166,
|
|
"epoch": 0.0398697516469109,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 6.036650362233648e-05,
|
|
"loss": 5.584,
|
|
"mean_token_accuracy": 0.17421112060546876,
|
|
"num_tokens": 8470655.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.8224467754364015,
|
|
"epoch": 0.03992348446584205,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 6.0015971193359824e-05,
|
|
"loss": 5.7104,
|
|
"mean_token_accuracy": 0.17625889480113982,
|
|
"num_tokens": 8481292.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.729261875152588,
|
|
"epoch": 0.039977217284773194,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 5.9671332660131306e-05,
|
|
"loss": 5.5002,
|
|
"mean_token_accuracy": 0.17888855189085007,
|
|
"num_tokens": 8491986.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.874344396591186,
|
|
"epoch": 0.04003095010370434,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 5.933259747109042e-05,
|
|
"loss": 5.7567,
|
|
"mean_token_accuracy": 0.17169555574655532,
|
|
"num_tokens": 8504312.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.823251438140869,
|
|
"epoch": 0.040084682922635485,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 5.899977491283351e-05,
|
|
"loss": 5.6447,
|
|
"mean_token_accuracy": 0.17388678789138795,
|
|
"num_tokens": 8516022.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.774986839294433,
|
|
"epoch": 0.04013841574156663,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 5.867287410985908e-05,
|
|
"loss": 5.6324,
|
|
"mean_token_accuracy": 0.17721940726041793,
|
|
"num_tokens": 8527016.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.787354850769043,
|
|
"epoch": 0.040192148560497784,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 5.835190402431779e-05,
|
|
"loss": 5.529,
|
|
"mean_token_accuracy": 0.1780991807579994,
|
|
"num_tokens": 8537817.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.822180891036988,
|
|
"epoch": 0.04024588137942893,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.803687345576673e-05,
|
|
"loss": 5.6106,
|
|
"mean_token_accuracy": 0.1760574534535408,
|
|
"num_tokens": 8548726.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.742981004714966,
|
|
"epoch": 0.040299614198360076,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 5.7727791040927977e-05,
|
|
"loss": 5.4964,
|
|
"mean_token_accuracy": 0.18106072694063186,
|
|
"num_tokens": 8559788.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.759645891189575,
|
|
"epoch": 0.04035334701729122,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 5.742466525345213e-05,
|
|
"loss": 5.5723,
|
|
"mean_token_accuracy": 0.17885070890188218,
|
|
"num_tokens": 8571649.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.7875265121459964,
|
|
"epoch": 0.04040707983622237,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 5.7127504403685775e-05,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.17578454911708832,
|
|
"num_tokens": 8583287.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.772876501083374,
|
|
"epoch": 0.040460812655153514,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.6836316638443664e-05,
|
|
"loss": 5.5651,
|
|
"mean_token_accuracy": 0.17683877944946289,
|
|
"num_tokens": 8594049.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.812844944000244,
|
|
"epoch": 0.04051454547408466,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.655110994078553e-05,
|
|
"loss": 5.6017,
|
|
"mean_token_accuracy": 0.176737941801548,
|
|
"num_tokens": 8605329.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.794150114059448,
|
|
"epoch": 0.040568278293015805,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 5.6271892129797056e-05,
|
|
"loss": 5.6048,
|
|
"mean_token_accuracy": 0.1782495990395546,
|
|
"num_tokens": 8616846.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.733209323883057,
|
|
"epoch": 0.04062201111194696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.599867086037556e-05,
|
|
"loss": 5.4389,
|
|
"mean_token_accuracy": 0.184093177318573,
|
|
"num_tokens": 8626559.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.758758211135865,
|
|
"epoch": 0.040675743930878104,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 5.573145362302012e-05,
|
|
"loss": 5.6268,
|
|
"mean_token_accuracy": 0.17258528470993043,
|
|
"num_tokens": 8639420.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.789173936843872,
|
|
"epoch": 0.04072947674980925,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.5470247743626404e-05,
|
|
"loss": 5.5539,
|
|
"mean_token_accuracy": 0.17462630420923234,
|
|
"num_tokens": 8650545.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.688164186477661,
|
|
"epoch": 0.040783209568740396,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 5.5215060383285414e-05,
|
|
"loss": 5.423,
|
|
"mean_token_accuracy": 0.1844472587108612,
|
|
"num_tokens": 8661437.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.900033855438233,
|
|
"epoch": 0.04083694238767154,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 5.496589853808759e-05,
|
|
"loss": 5.6283,
|
|
"mean_token_accuracy": 0.1767179325222969,
|
|
"num_tokens": 8672738.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.8547381401062015,
|
|
"epoch": 0.04089067520660269,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 5.47227690389308e-05,
|
|
"loss": 5.6686,
|
|
"mean_token_accuracy": 0.16542310863733292,
|
|
"num_tokens": 8685001.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.790684938430786,
|
|
"epoch": 0.040944408025533834,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 5.448567855133306e-05,
|
|
"loss": 5.5835,
|
|
"mean_token_accuracy": 0.17481715381145477,
|
|
"num_tokens": 8697195.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.816874742507935,
|
|
"epoch": 0.04099814084446498,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.425463357524986e-05,
|
|
"loss": 5.6337,
|
|
"mean_token_accuracy": 0.1752805545926094,
|
|
"num_tokens": 8709979.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.831140470504761,
|
|
"epoch": 0.04105187366339613,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 5.402964044489591e-05,
|
|
"loss": 5.7038,
|
|
"mean_token_accuracy": 0.1666072502732277,
|
|
"num_tokens": 8723019.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.792020082473755,
|
|
"epoch": 0.04110560648232728,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.381070532857153e-05,
|
|
"loss": 5.5883,
|
|
"mean_token_accuracy": 0.17005713880062104,
|
|
"num_tokens": 8734465.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.733161783218383,
|
|
"epoch": 0.041159339301258424,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 5.359783422849357e-05,
|
|
"loss": 5.5507,
|
|
"mean_token_accuracy": 0.1810339704155922,
|
|
"num_tokens": 8745655.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.746679401397705,
|
|
"epoch": 0.04121307212018957,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 5.3391032980630736e-05,
|
|
"loss": 5.5689,
|
|
"mean_token_accuracy": 0.1752975642681122,
|
|
"num_tokens": 8758026.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.854162979125976,
|
|
"epoch": 0.041266804939120716,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 5.31903072545437e-05,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.17121294736862183,
|
|
"num_tokens": 8768810.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.800893878936767,
|
|
"epoch": 0.04132053775805186,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 5.29956625532297e-05,
|
|
"loss": 5.6329,
|
|
"mean_token_accuracy": 0.1720083862543106,
|
|
"num_tokens": 8780230.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.733534240722657,
|
|
"epoch": 0.04137427057698301,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 5.280710421297146e-05,
|
|
"loss": 5.4748,
|
|
"mean_token_accuracy": 0.1846863657236099,
|
|
"num_tokens": 8791037.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.821543121337891,
|
|
"epoch": 0.041428003395914154,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.2624637403191165e-05,
|
|
"loss": 5.6164,
|
|
"mean_token_accuracy": 0.17735685259103776,
|
|
"num_tokens": 8801780.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.784174203872681,
|
|
"epoch": 0.041481736214845306,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 5.2448267126308605e-05,
|
|
"loss": 5.6157,
|
|
"mean_token_accuracy": 0.16746658831834793,
|
|
"num_tokens": 8813953.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.862273216247559,
|
|
"epoch": 0.04153546903377645,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.2277998217603954e-05,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.17257596999406816,
|
|
"num_tokens": 8824657.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.8266082286834715,
|
|
"epoch": 0.0415892018527076,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 5.211383534508541e-05,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.1792323261499405,
|
|
"num_tokens": 8835987.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.838964033126831,
|
|
"epoch": 0.041642934671638744,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 5.1955783009361044e-05,
|
|
"loss": 5.596,
|
|
"mean_token_accuracy": 0.18123795986175537,
|
|
"num_tokens": 8847002.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.761981344223022,
|
|
"epoch": 0.04169666749056989,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 5.180384554351543e-05,
|
|
"loss": 5.5594,
|
|
"mean_token_accuracy": 0.17555866092443467,
|
|
"num_tokens": 8858375.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.7682453155517575,
|
|
"epoch": 0.041750400309501036,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.1658027112990976e-05,
|
|
"loss": 5.5825,
|
|
"mean_token_accuracy": 0.1820688247680664,
|
|
"num_tokens": 8869562.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.750371503829956,
|
|
"epoch": 0.04180413312843218,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 5.151833171547365e-05,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.17593167573213578,
|
|
"num_tokens": 8882402.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.732252979278565,
|
|
"epoch": 0.04185786594736333,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 5.1384763180783274e-05,
|
|
"loss": 5.5449,
|
|
"mean_token_accuracy": 0.17476944327354432,
|
|
"num_tokens": 8894349.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.762109994888306,
|
|
"epoch": 0.04191159876629448,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 5.125732517076876e-05,
|
|
"loss": 5.6305,
|
|
"mean_token_accuracy": 0.17056868225336075,
|
|
"num_tokens": 8907481.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.807485580444336,
|
|
"epoch": 0.041965331585225626,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 5.113602117920747e-05,
|
|
"loss": 5.6359,
|
|
"mean_token_accuracy": 0.17412620335817336,
|
|
"num_tokens": 8920165.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.813747406005859,
|
|
"epoch": 0.04201906440415677,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 5.102085453170966e-05,
|
|
"loss": 5.6432,
|
|
"mean_token_accuracy": 0.17479592561721802,
|
|
"num_tokens": 8932662.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.832588815689087,
|
|
"epoch": 0.04207279722308792,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 5.091182838562709e-05,
|
|
"loss": 5.662,
|
|
"mean_token_accuracy": 0.16547612696886063,
|
|
"num_tokens": 8944556.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.727107572555542,
|
|
"epoch": 0.042126530042019064,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.0808945729966716e-05,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.18496476113796234,
|
|
"num_tokens": 8955099.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.739648675918579,
|
|
"epoch": 0.04218026286095021,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 5.071220938530844e-05,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.17657755315303802,
|
|
"num_tokens": 8966327.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.8771507263183596,
|
|
"epoch": 0.042233995679881356,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 5.0621622003728026e-05,
|
|
"loss": 5.7252,
|
|
"mean_token_accuracy": 0.17451791167259217,
|
|
"num_tokens": 8977926.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.762888622283936,
|
|
"epoch": 0.0422877284988125,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 5.053718606872433e-05,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.17950422465801238,
|
|
"num_tokens": 8989504.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.802980279922485,
|
|
"epoch": 0.042341461317743655,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 5.0458903895151134e-05,
|
|
"loss": 5.6631,
|
|
"mean_token_accuracy": 0.17326463162899017,
|
|
"num_tokens": 9000855.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.8287492275238035,
|
|
"epoch": 0.0423951941366748,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 5.038677762915381e-05,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.17053717523813247,
|
|
"num_tokens": 9012858.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.799716854095459,
|
|
"epoch": 0.042448926955605946,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 5.032080924811033e-05,
|
|
"loss": 5.6264,
|
|
"mean_token_accuracy": 0.17711709886789323,
|
|
"num_tokens": 9024809.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.783935594558716,
|
|
"epoch": 0.04250265977453709,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.026100056057718e-05,
|
|
"loss": 5.6482,
|
|
"mean_token_accuracy": 0.17101871520280837,
|
|
"num_tokens": 9035698.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.706310701370239,
|
|
"epoch": 0.04255639259346824,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.0207353206239764e-05,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.18749566078186036,
|
|
"num_tokens": 9046418.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.771052885055542,
|
|
"epoch": 0.042610125412399384,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 5.0159868655867436e-05,
|
|
"loss": 5.5726,
|
|
"mean_token_accuracy": 0.18165841996669768,
|
|
"num_tokens": 9057648.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.896209955215454,
|
|
"epoch": 0.04266385823133053,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 5.011854821127305e-05,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.16683439835906028,
|
|
"num_tokens": 9069663.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.714368534088135,
|
|
"epoch": 0.042717591050261676,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.008339300527755e-05,
|
|
"loss": 5.4837,
|
|
"mean_token_accuracy": 0.18830896466970443,
|
|
"num_tokens": 9080036.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.765442943572998,
|
|
"epoch": 0.04277132386919283,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.005440400167859e-05,
|
|
"loss": 5.5973,
|
|
"mean_token_accuracy": 0.1691672086715698,
|
|
"num_tokens": 9090623.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.809631299972534,
|
|
"epoch": 0.042825056688123975,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.003158199522442e-05,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.17250185757875441,
|
|
"num_tokens": 9102142.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.894583082199096,
|
|
"epoch": 0.04287878950705512,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 5.0014927611591806e-05,
|
|
"loss": 5.7427,
|
|
"mean_token_accuracy": 0.17117976993322373,
|
|
"num_tokens": 9113937.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.746591329574585,
|
|
"epoch": 0.042932522325986266,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.000444130736916e-05,
|
|
"loss": 5.546,
|
|
"mean_token_accuracy": 0.17259600162506103,
|
|
"num_tokens": 9125735.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.712372446060181,
|
|
"epoch": 0.04298625514491741,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 5.0000123370043736e-05,
|
|
"loss": 5.4786,
|
|
"mean_token_accuracy": 0.18034083098173143,
|
|
"num_tokens": 9137307.0,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.4204912689152e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|