3035 lines
82 KiB
JSON
3035 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.01611984567934403,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 5338482001920000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|