18233 lines
498 KiB
JSON
18233 lines
498 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 8.174466151749204,
|
|
"eval_steps": 500,
|
|
"global_step": 9000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692006397247315,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 13.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8349,
|
|
"mean_token_accuracy": 9.124087519012392e-05,
|
|
"num_tokens": 10156.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691936588287353,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 13.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.768,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20669.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691073989868164,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 9.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5071,
|
|
"mean_token_accuracy": 0.02231150579173118,
|
|
"num_tokens": 30952.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.680047035217285,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 6.46875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1516,
|
|
"mean_token_accuracy": 0.04495340101420879,
|
|
"num_tokens": 41429.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.620328998565673,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 4.09375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8702,
|
|
"mean_token_accuracy": 0.04405966177582741,
|
|
"num_tokens": 52166.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.600391387939453,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 3.125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7103,
|
|
"mean_token_accuracy": 0.04457313194870949,
|
|
"num_tokens": 61670.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.599573707580566,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6383,
|
|
"mean_token_accuracy": 0.04294479861855507,
|
|
"num_tokens": 71702.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.582702255249023,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5438,
|
|
"mean_token_accuracy": 0.04814234748482704,
|
|
"num_tokens": 82840.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.578528881072998,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5291,
|
|
"mean_token_accuracy": 0.04308051839470863,
|
|
"num_tokens": 92290.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.577267169952393,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5231,
|
|
"mean_token_accuracy": 0.04617781527340412,
|
|
"num_tokens": 101842.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.557446765899659,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5235,
|
|
"mean_token_accuracy": 0.044288964942097664,
|
|
"num_tokens": 114030.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.576010799407959,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3777,
|
|
"mean_token_accuracy": 0.053348609060049054,
|
|
"num_tokens": 124349.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.424556732177734,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.227,
|
|
"mean_token_accuracy": 0.062253265827894214,
|
|
"num_tokens": 133534.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.427367973327637,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.1253,
|
|
"mean_token_accuracy": 0.0669670920819044,
|
|
"num_tokens": 144383.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.475873756408692,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1105,
|
|
"mean_token_accuracy": 0.06663424372673035,
|
|
"num_tokens": 155467.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.44368600845337,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.9769,
|
|
"mean_token_accuracy": 0.06910263895988464,
|
|
"num_tokens": 165903.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.294152164459229,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.8596,
|
|
"mean_token_accuracy": 0.07446017190814018,
|
|
"num_tokens": 175812.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.241645622253419,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7369,
|
|
"mean_token_accuracy": 0.0734786469489336,
|
|
"num_tokens": 186357.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.233689880371093,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6529,
|
|
"mean_token_accuracy": 0.08008331432938576,
|
|
"num_tokens": 196124.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.130806350708008,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.5378,
|
|
"mean_token_accuracy": 0.0810570053756237,
|
|
"num_tokens": 205655.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.153574180603027,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4974,
|
|
"mean_token_accuracy": 0.07750472910702229,
|
|
"num_tokens": 215870.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.066059398651124,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.3739,
|
|
"mean_token_accuracy": 0.07989286929368973,
|
|
"num_tokens": 225027.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.00927906036377,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2865,
|
|
"mean_token_accuracy": 0.07930834442377091,
|
|
"num_tokens": 236352.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.885906505584718,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1373,
|
|
"mean_token_accuracy": 0.08147214129567146,
|
|
"num_tokens": 246713.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.743410396575928,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.9555,
|
|
"mean_token_accuracy": 0.08192854598164559,
|
|
"num_tokens": 256950.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.601629638671875,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9226,
|
|
"mean_token_accuracy": 0.0807999774813652,
|
|
"num_tokens": 267345.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.43809471130371,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7251,
|
|
"mean_token_accuracy": 0.0855719804763794,
|
|
"num_tokens": 276782.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.163784599304199,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.5904,
|
|
"mean_token_accuracy": 0.08386219590902329,
|
|
"num_tokens": 287024.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.970656967163086,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6436,
|
|
"mean_token_accuracy": 0.0813270166516304,
|
|
"num_tokens": 296597.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.746164703369141,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4616,
|
|
"mean_token_accuracy": 0.07949798852205277,
|
|
"num_tokens": 307709.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.559226799011231,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.4771,
|
|
"mean_token_accuracy": 0.08536540046334266,
|
|
"num_tokens": 317976.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.463272094726562,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3407,
|
|
"mean_token_accuracy": 0.08832384943962097,
|
|
"num_tokens": 327319.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.277487754821777,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3296,
|
|
"mean_token_accuracy": 0.09019008278846741,
|
|
"num_tokens": 337337.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.112997436523438,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3544,
|
|
"mean_token_accuracy": 0.08267815336585045,
|
|
"num_tokens": 347779.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.123576450347901,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3699,
|
|
"mean_token_accuracy": 0.08613633513450622,
|
|
"num_tokens": 358088.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.9902650833129885,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2808,
|
|
"mean_token_accuracy": 0.08979431986808777,
|
|
"num_tokens": 368314.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.845839357376098,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.1745,
|
|
"mean_token_accuracy": 0.09477217569947242,
|
|
"num_tokens": 378406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.823360013961792,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1909,
|
|
"mean_token_accuracy": 0.08993544280529023,
|
|
"num_tokens": 388912.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.795131635665894,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1864,
|
|
"mean_token_accuracy": 0.09252966791391373,
|
|
"num_tokens": 399290.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.745543479919434,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.1118,
|
|
"mean_token_accuracy": 0.09492666274309158,
|
|
"num_tokens": 409757.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.725181293487549,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2908,
|
|
"mean_token_accuracy": 0.08956578224897385,
|
|
"num_tokens": 420182.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.697661828994751,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.164,
|
|
"mean_token_accuracy": 0.08962696269154549,
|
|
"num_tokens": 429897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.651192474365234,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.0723,
|
|
"mean_token_accuracy": 0.09363478198647499,
|
|
"num_tokens": 441131.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.565522766113281,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0423,
|
|
"mean_token_accuracy": 0.09191863983869553,
|
|
"num_tokens": 451894.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.576038503646851,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.9726,
|
|
"mean_token_accuracy": 0.09775904789566994,
|
|
"num_tokens": 462613.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.570877313613892,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0725,
|
|
"mean_token_accuracy": 0.09952598586678504,
|
|
"num_tokens": 471606.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.4385573387146,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.9811,
|
|
"mean_token_accuracy": 0.09829605221748353,
|
|
"num_tokens": 481559.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.5122908592224125,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0112,
|
|
"mean_token_accuracy": 0.08943306282162666,
|
|
"num_tokens": 492766.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.515708589553833,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0791,
|
|
"mean_token_accuracy": 0.09232381284236908,
|
|
"num_tokens": 503357.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.483732652664185,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9049,
|
|
"mean_token_accuracy": 0.09773992225527764,
|
|
"num_tokens": 513098.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.434525108337402,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.9856,
|
|
"mean_token_accuracy": 0.09069462195038795,
|
|
"num_tokens": 522807.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.545422697067261,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9915,
|
|
"mean_token_accuracy": 0.0919874556362629,
|
|
"num_tokens": 532670.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.408900785446167,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9358,
|
|
"mean_token_accuracy": 0.09355607330799103,
|
|
"num_tokens": 543647.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.463206338882446,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9815,
|
|
"mean_token_accuracy": 0.10008879154920577,
|
|
"num_tokens": 553047.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.350521087646484,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.04,
|
|
"mean_token_accuracy": 0.09540624544024467,
|
|
"num_tokens": 563413.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.391801500320435,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9979,
|
|
"mean_token_accuracy": 0.09288603365421295,
|
|
"num_tokens": 573790.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.465456533432007,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8925,
|
|
"mean_token_accuracy": 0.09841871336102485,
|
|
"num_tokens": 582958.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.248134994506836,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8605,
|
|
"mean_token_accuracy": 0.09925618246197701,
|
|
"num_tokens": 592886.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.357699966430664,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.09516463130712509,
|
|
"num_tokens": 603692.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.398810338973999,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9375,
|
|
"mean_token_accuracy": 0.09712802171707154,
|
|
"num_tokens": 613534.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.381405019760132,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8936,
|
|
"mean_token_accuracy": 0.09751264378428459,
|
|
"num_tokens": 623862.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.37616286277771,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9753,
|
|
"mean_token_accuracy": 0.09758875817060471,
|
|
"num_tokens": 633759.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.254412126541138,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8595,
|
|
"mean_token_accuracy": 0.09413841366767883,
|
|
"num_tokens": 643553.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.26377305984497,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7828,
|
|
"mean_token_accuracy": 0.10154594853520393,
|
|
"num_tokens": 653714.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.18550124168396,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.7995,
|
|
"mean_token_accuracy": 0.10076900273561477,
|
|
"num_tokens": 662817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.303860998153686,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.84,
|
|
"mean_token_accuracy": 0.0980252668261528,
|
|
"num_tokens": 672381.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.084452152252197,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.704,
|
|
"mean_token_accuracy": 0.1016760528087616,
|
|
"num_tokens": 682525.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.245214605331421,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8214,
|
|
"mean_token_accuracy": 0.09809606671333312,
|
|
"num_tokens": 691826.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.1477008819580075,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7884,
|
|
"mean_token_accuracy": 0.10888883322477341,
|
|
"num_tokens": 701319.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.180845546722412,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8811,
|
|
"mean_token_accuracy": 0.09638292342424393,
|
|
"num_tokens": 710888.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.259013700485229,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8189,
|
|
"mean_token_accuracy": 0.10254016146063805,
|
|
"num_tokens": 722516.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.177930450439453,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7834,
|
|
"mean_token_accuracy": 0.0994310162961483,
|
|
"num_tokens": 733058.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.135624599456787,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8795,
|
|
"mean_token_accuracy": 0.10592572912573814,
|
|
"num_tokens": 744832.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.182344436645508,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7888,
|
|
"mean_token_accuracy": 0.10330469757318497,
|
|
"num_tokens": 755387.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.133783912658691,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7362,
|
|
"mean_token_accuracy": 0.10057889819145202,
|
|
"num_tokens": 764455.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.0702344417572025,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7728,
|
|
"mean_token_accuracy": 0.09861194267868996,
|
|
"num_tokens": 774513.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.239872455596924,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8665,
|
|
"mean_token_accuracy": 0.10181790813803673,
|
|
"num_tokens": 785540.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.1300865650177006,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7207,
|
|
"mean_token_accuracy": 0.09868112578988075,
|
|
"num_tokens": 795233.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.091469669342041,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7836,
|
|
"mean_token_accuracy": 0.09853761792182922,
|
|
"num_tokens": 804855.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.121715068817139,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.10705440938472748,
|
|
"num_tokens": 814818.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.0411852359771725,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7698,
|
|
"mean_token_accuracy": 0.1102868601679802,
|
|
"num_tokens": 824622.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.094270896911621,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6485,
|
|
"mean_token_accuracy": 0.1126321442425251,
|
|
"num_tokens": 834813.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.969181776046753,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7691,
|
|
"mean_token_accuracy": 0.10265972688794137,
|
|
"num_tokens": 844887.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.0739898681640625,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7507,
|
|
"mean_token_accuracy": 0.10284761190414429,
|
|
"num_tokens": 855430.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.955200242996216,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5928,
|
|
"mean_token_accuracy": 0.10530942752957344,
|
|
"num_tokens": 866052.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.931502008438111,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7609,
|
|
"mean_token_accuracy": 0.09737498015165329,
|
|
"num_tokens": 877961.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.038836812973022,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7166,
|
|
"mean_token_accuracy": 0.10936553180217742,
|
|
"num_tokens": 888386.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.057555246353149,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7618,
|
|
"mean_token_accuracy": 0.10554658025503158,
|
|
"num_tokens": 898272.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.007745981216431,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7134,
|
|
"mean_token_accuracy": 0.10867258682847022,
|
|
"num_tokens": 908317.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.9758936882019045,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6975,
|
|
"mean_token_accuracy": 0.10497988387942314,
|
|
"num_tokens": 917970.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.033078050613403,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7406,
|
|
"mean_token_accuracy": 0.10012255981564522,
|
|
"num_tokens": 928132.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.922481346130371,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6654,
|
|
"mean_token_accuracy": 0.10554703250527382,
|
|
"num_tokens": 938092.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.0776440620422365,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7696,
|
|
"mean_token_accuracy": 0.10158471092581749,
|
|
"num_tokens": 948650.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.936448192596435,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6879,
|
|
"mean_token_accuracy": 0.10931911468505859,
|
|
"num_tokens": 960195.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0439310550689695,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.714,
|
|
"mean_token_accuracy": 0.10808822214603424,
|
|
"num_tokens": 970608.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.9404487133026125,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6986,
|
|
"mean_token_accuracy": 0.10617228373885154,
|
|
"num_tokens": 981053.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.817537832260132,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.623,
|
|
"mean_token_accuracy": 0.11259457916021347,
|
|
"num_tokens": 991560.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.957720375061035,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6487,
|
|
"mean_token_accuracy": 0.10093178153038025,
|
|
"num_tokens": 1002008.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.949663352966309,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6577,
|
|
"mean_token_accuracy": 0.11253990605473518,
|
|
"num_tokens": 1012349.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.866103410720825,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6536,
|
|
"mean_token_accuracy": 0.10712857395410538,
|
|
"num_tokens": 1022886.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 7.01205497120702,
|
|
"eval_loss": 6.782735824584961,
|
|
"eval_mean_token_accuracy": 0.10699583826543287,
|
|
"eval_num_tokens": 1022886.0,
|
|
"eval_runtime": 2.6346,
|
|
"eval_samples_per_second": 1303.802,
|
|
"eval_steps_per_second": 163.213,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.038756370544434,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7025,
|
|
"mean_token_accuracy": 0.10537334904074669,
|
|
"num_tokens": 1033129.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.824175500869751,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5941,
|
|
"mean_token_accuracy": 0.11075264886021614,
|
|
"num_tokens": 1043069.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.944571828842163,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6249,
|
|
"mean_token_accuracy": 0.10747886225581169,
|
|
"num_tokens": 1054049.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.737239837646484,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6107,
|
|
"mean_token_accuracy": 0.10490327775478363,
|
|
"num_tokens": 1065155.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8596274852752686,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.602,
|
|
"mean_token_accuracy": 0.1115146167576313,
|
|
"num_tokens": 1075510.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.839829826354981,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.11251503229141235,
|
|
"num_tokens": 1086494.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.885312938690186,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6622,
|
|
"mean_token_accuracy": 0.11076548621058464,
|
|
"num_tokens": 1097311.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.8867042541503904,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6121,
|
|
"mean_token_accuracy": 0.10653524398803711,
|
|
"num_tokens": 1107757.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.7683220386505125,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5174,
|
|
"mean_token_accuracy": 0.1121369518339634,
|
|
"num_tokens": 1118889.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.904896688461304,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7133,
|
|
"mean_token_accuracy": 0.10461575463414192,
|
|
"num_tokens": 1128951.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.827801132202149,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5235,
|
|
"mean_token_accuracy": 0.11478598713874817,
|
|
"num_tokens": 1138583.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.823013877868652,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5837,
|
|
"mean_token_accuracy": 0.11256127953529357,
|
|
"num_tokens": 1149572.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.808546161651611,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5508,
|
|
"mean_token_accuracy": 0.11078241392970085,
|
|
"num_tokens": 1160532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.7923932552337645,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5579,
|
|
"mean_token_accuracy": 0.11136969849467278,
|
|
"num_tokens": 1170915.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.768967342376709,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5276,
|
|
"mean_token_accuracy": 0.11916670873761177,
|
|
"num_tokens": 1180822.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.733667182922363,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.568,
|
|
"mean_token_accuracy": 0.11458625867962838,
|
|
"num_tokens": 1190447.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.752501106262207,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6129,
|
|
"mean_token_accuracy": 0.10823755264282227,
|
|
"num_tokens": 1200951.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.777053880691528,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.11973060294985771,
|
|
"num_tokens": 1211846.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.6049951076507565,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.11239737868309022,
|
|
"num_tokens": 1222074.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.6186809062957765,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.11210898384451866,
|
|
"num_tokens": 1231741.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.734645938873291,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4613,
|
|
"mean_token_accuracy": 0.11268012151122093,
|
|
"num_tokens": 1241078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.815666151046753,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6588,
|
|
"mean_token_accuracy": 0.10641585364937782,
|
|
"num_tokens": 1251685.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.6285426139831545,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5087,
|
|
"mean_token_accuracy": 0.11794958114624024,
|
|
"num_tokens": 1262044.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.700363206863403,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5172,
|
|
"mean_token_accuracy": 0.11475807651877404,
|
|
"num_tokens": 1273647.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.803307104110718,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5712,
|
|
"mean_token_accuracy": 0.11094894334673881,
|
|
"num_tokens": 1284396.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.678705930709839,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5534,
|
|
"mean_token_accuracy": 0.1111487403512001,
|
|
"num_tokens": 1294339.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.718799018859864,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4672,
|
|
"mean_token_accuracy": 0.11378256231546402,
|
|
"num_tokens": 1304572.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.6692344665527346,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5118,
|
|
"mean_token_accuracy": 0.11725670620799064,
|
|
"num_tokens": 1314780.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.713592195510865,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5254,
|
|
"mean_token_accuracy": 0.11340317949652672,
|
|
"num_tokens": 1325215.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.4963737487792965,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4567,
|
|
"mean_token_accuracy": 0.1154212512075901,
|
|
"num_tokens": 1334745.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.773891592025757,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5655,
|
|
"mean_token_accuracy": 0.11454486474394798,
|
|
"num_tokens": 1345150.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.740270376205444,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.11400953680276871,
|
|
"num_tokens": 1355619.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.560915565490722,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4409,
|
|
"mean_token_accuracy": 0.11688014417886734,
|
|
"num_tokens": 1365933.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.635832834243774,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.118435288220644,
|
|
"num_tokens": 1376867.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.725541782379151,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6275,
|
|
"mean_token_accuracy": 0.11254403442144394,
|
|
"num_tokens": 1386739.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.641561841964721,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4586,
|
|
"mean_token_accuracy": 0.1176810882985592,
|
|
"num_tokens": 1397138.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.742542171478272,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5663,
|
|
"mean_token_accuracy": 0.11449027881026268,
|
|
"num_tokens": 1408171.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.55933403968811,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.11086875796318055,
|
|
"num_tokens": 1418259.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.658524894714356,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.502,
|
|
"mean_token_accuracy": 0.11733507215976716,
|
|
"num_tokens": 1428080.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.703834104537964,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5327,
|
|
"mean_token_accuracy": 0.11690907329320907,
|
|
"num_tokens": 1438502.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.5558753490448,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4643,
|
|
"mean_token_accuracy": 0.11556085273623466,
|
|
"num_tokens": 1448396.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.670946264266968,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.1103983074426651,
|
|
"num_tokens": 1459728.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.634099054336548,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4278,
|
|
"mean_token_accuracy": 0.12270175665616989,
|
|
"num_tokens": 1469733.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.553995990753174,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4202,
|
|
"mean_token_accuracy": 0.12040546089410782,
|
|
"num_tokens": 1479004.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.548476982116699,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3915,
|
|
"mean_token_accuracy": 0.11926131621003151,
|
|
"num_tokens": 1489848.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.4792413234710695,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.11599399968981743,
|
|
"num_tokens": 1500877.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.701153993606567,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5473,
|
|
"mean_token_accuracy": 0.11609271615743637,
|
|
"num_tokens": 1511840.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.552907514572143,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4516,
|
|
"mean_token_accuracy": 0.11515182629227638,
|
|
"num_tokens": 1521839.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.5626030445098875,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3459,
|
|
"mean_token_accuracy": 0.11943092346191406,
|
|
"num_tokens": 1531285.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.370995140075683,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3877,
|
|
"mean_token_accuracy": 0.11523654535412789,
|
|
"num_tokens": 1542027.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.586507654190063,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5361,
|
|
"mean_token_accuracy": 0.11796742305159569,
|
|
"num_tokens": 1552527.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.724661540985108,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5582,
|
|
"mean_token_accuracy": 0.11669881269335747,
|
|
"num_tokens": 1562605.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.477885150909424,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4395,
|
|
"mean_token_accuracy": 0.11811442151665688,
|
|
"num_tokens": 1572128.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.537688159942627,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3727,
|
|
"mean_token_accuracy": 0.11771526038646699,
|
|
"num_tokens": 1582352.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.62556095123291,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4766,
|
|
"mean_token_accuracy": 0.11757447347044944,
|
|
"num_tokens": 1592887.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.465275049209595,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4,
|
|
"mean_token_accuracy": 0.11981132328510284,
|
|
"num_tokens": 1603564.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.42244291305542,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2554,
|
|
"mean_token_accuracy": 0.12776268124580384,
|
|
"num_tokens": 1613270.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.381896257400513,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3224,
|
|
"mean_token_accuracy": 0.12746695652604104,
|
|
"num_tokens": 1623009.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.510463762283325,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3784,
|
|
"mean_token_accuracy": 0.11881706863641739,
|
|
"num_tokens": 1633209.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.507561254501343,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3955,
|
|
"mean_token_accuracy": 0.12002938464283944,
|
|
"num_tokens": 1643708.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.467615747451783,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4803,
|
|
"mean_token_accuracy": 0.11377913728356362,
|
|
"num_tokens": 1653973.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.51342453956604,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4546,
|
|
"mean_token_accuracy": 0.12055374309420586,
|
|
"num_tokens": 1664856.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.435442161560059,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3602,
|
|
"mean_token_accuracy": 0.11813876554369926,
|
|
"num_tokens": 1675935.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.53614912033081,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4056,
|
|
"mean_token_accuracy": 0.12431632876396179,
|
|
"num_tokens": 1686597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.432868528366089,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3541,
|
|
"mean_token_accuracy": 0.12400377839803696,
|
|
"num_tokens": 1697385.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.584589052200317,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4077,
|
|
"mean_token_accuracy": 0.11550647392868996,
|
|
"num_tokens": 1706680.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.440202140808106,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.401,
|
|
"mean_token_accuracy": 0.126813330501318,
|
|
"num_tokens": 1716130.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.433648681640625,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.332,
|
|
"mean_token_accuracy": 0.11563118770718575,
|
|
"num_tokens": 1726835.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.501465082168579,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3745,
|
|
"mean_token_accuracy": 0.1284072481095791,
|
|
"num_tokens": 1737708.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.4302887439727785,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3703,
|
|
"mean_token_accuracy": 0.1250189833343029,
|
|
"num_tokens": 1747649.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.467341136932373,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3986,
|
|
"mean_token_accuracy": 0.12090553492307662,
|
|
"num_tokens": 1758327.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.422776794433593,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3411,
|
|
"mean_token_accuracy": 0.12042633518576622,
|
|
"num_tokens": 1769767.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.451702117919922,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.12029828205704689,
|
|
"num_tokens": 1780139.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.539723587036133,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3194,
|
|
"mean_token_accuracy": 0.12847061827778816,
|
|
"num_tokens": 1790783.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.328425884246826,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3449,
|
|
"mean_token_accuracy": 0.11921576634049416,
|
|
"num_tokens": 1801349.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.39916582107544,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3231,
|
|
"mean_token_accuracy": 0.11750098913908005,
|
|
"num_tokens": 1811990.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.380044221878052,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.1286027878522873,
|
|
"num_tokens": 1821725.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.408110952377319,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3034,
|
|
"mean_token_accuracy": 0.11874947100877761,
|
|
"num_tokens": 1831794.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.4103223323822025,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3795,
|
|
"mean_token_accuracy": 0.11963622644543648,
|
|
"num_tokens": 1842002.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.414848804473877,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3629,
|
|
"mean_token_accuracy": 0.127029949426651,
|
|
"num_tokens": 1852282.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.416449689865113,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.11340886801481247,
|
|
"num_tokens": 1862830.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.356419038772583,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3355,
|
|
"mean_token_accuracy": 0.12167628929018974,
|
|
"num_tokens": 1872746.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.417472887039184,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3499,
|
|
"mean_token_accuracy": 0.11554814651608467,
|
|
"num_tokens": 1884353.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.450801038742066,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3923,
|
|
"mean_token_accuracy": 0.11636136919260025,
|
|
"num_tokens": 1895719.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.386584901809693,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.331,
|
|
"mean_token_accuracy": 0.12908836230635642,
|
|
"num_tokens": 1905905.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.452653694152832,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.12015582546591759,
|
|
"num_tokens": 1916010.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.408344888687134,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3198,
|
|
"mean_token_accuracy": 0.1261664181947708,
|
|
"num_tokens": 1926429.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.20701494216919,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2619,
|
|
"mean_token_accuracy": 0.12468230649828911,
|
|
"num_tokens": 1936322.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.447282505035401,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4247,
|
|
"mean_token_accuracy": 0.11766760200262069,
|
|
"num_tokens": 1947885.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.424139404296875,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3787,
|
|
"mean_token_accuracy": 0.11931429132819175,
|
|
"num_tokens": 1957876.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.3488563060760494,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4145,
|
|
"mean_token_accuracy": 0.11656479313969612,
|
|
"num_tokens": 1969531.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.349594783782959,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.12435748800635338,
|
|
"num_tokens": 1980787.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.337444829940796,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3126,
|
|
"mean_token_accuracy": 0.12260773479938507,
|
|
"num_tokens": 1991645.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.310835170745849,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3806,
|
|
"mean_token_accuracy": 0.12270210534334183,
|
|
"num_tokens": 2002544.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.440783357620239,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4326,
|
|
"mean_token_accuracy": 0.11328092440962792,
|
|
"num_tokens": 2012813.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.404192399978638,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2975,
|
|
"mean_token_accuracy": 0.12434757873415947,
|
|
"num_tokens": 2023756.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.372990131378174,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3826,
|
|
"mean_token_accuracy": 0.12353581041097642,
|
|
"num_tokens": 2034245.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.430953598022461,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4207,
|
|
"mean_token_accuracy": 0.11968368738889694,
|
|
"num_tokens": 2045635.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.354709815979004,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3894,
|
|
"mean_token_accuracy": 0.12116859331727028,
|
|
"num_tokens": 2056501.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.398514699935913,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3851,
|
|
"mean_token_accuracy": 0.12161954492330551,
|
|
"num_tokens": 2066540.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.234273269564606,
|
|
"eval_loss": 6.430207252502441,
|
|
"eval_mean_token_accuracy": 0.12242212169911973,
|
|
"eval_num_tokens": 2066540.0,
|
|
"eval_runtime": 2.795,
|
|
"eval_samples_per_second": 1228.966,
|
|
"eval_steps_per_second": 153.844,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.331130456924439,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.2664,
|
|
"mean_token_accuracy": 0.1238655298948288,
|
|
"num_tokens": 2077008.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.280630779266358,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.189,
|
|
"mean_token_accuracy": 0.1289375126361847,
|
|
"num_tokens": 2087314.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.301973485946656,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.3612,
|
|
"mean_token_accuracy": 0.12495536580681801,
|
|
"num_tokens": 2097928.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.324389410018921,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.2924,
|
|
"mean_token_accuracy": 0.12183087170124055,
|
|
"num_tokens": 2108662.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.28196439743042,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.3099,
|
|
"mean_token_accuracy": 0.12506715580821037,
|
|
"num_tokens": 2118546.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.3526708602905275,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.3328,
|
|
"mean_token_accuracy": 0.12695364654064178,
|
|
"num_tokens": 2128012.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.305598783493042,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.1869,
|
|
"mean_token_accuracy": 0.12489267513155937,
|
|
"num_tokens": 2138241.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.392855167388916,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.12088349610567092,
|
|
"num_tokens": 2150780.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.251961183547974,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.12670218348503112,
|
|
"num_tokens": 2161918.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.268582820892334,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.2656,
|
|
"mean_token_accuracy": 0.12303844839334488,
|
|
"num_tokens": 2172742.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.2342870235443115,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.266,
|
|
"mean_token_accuracy": 0.13134688064455985,
|
|
"num_tokens": 2182122.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.323199510574341,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.3078,
|
|
"mean_token_accuracy": 0.12292804196476936,
|
|
"num_tokens": 2193703.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.212486028671265,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.2103,
|
|
"mean_token_accuracy": 0.1271343432366848,
|
|
"num_tokens": 2204041.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.330165958404541,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.261,
|
|
"mean_token_accuracy": 0.12701819017529486,
|
|
"num_tokens": 2214276.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.206712341308593,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.12525501921772958,
|
|
"num_tokens": 2225390.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.3518956184387205,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.12725881561636926,
|
|
"num_tokens": 2235708.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.255891609191894,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.2203,
|
|
"mean_token_accuracy": 0.12782707437872887,
|
|
"num_tokens": 2246284.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.263408994674682,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.2457,
|
|
"mean_token_accuracy": 0.12486742287874222,
|
|
"num_tokens": 2257219.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.181901979446411,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.2442,
|
|
"mean_token_accuracy": 0.12457836270332337,
|
|
"num_tokens": 2268019.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.2217254638671875,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.2471,
|
|
"mean_token_accuracy": 0.12563535794615746,
|
|
"num_tokens": 2278382.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.287105083465576,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.0881,
|
|
"mean_token_accuracy": 0.12760736958848107,
|
|
"num_tokens": 2288515.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.165887403488159,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.1394278571009636,
|
|
"num_tokens": 2298396.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.156573009490967,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.13033899739384652,
|
|
"num_tokens": 2308707.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.189375972747802,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 5.9642,
|
|
"mean_token_accuracy": 0.13605893775820732,
|
|
"num_tokens": 2318981.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.058369588851929,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 6.0609,
|
|
"mean_token_accuracy": 0.1345390684902668,
|
|
"num_tokens": 2329647.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.180541086196899,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 5.9868,
|
|
"mean_token_accuracy": 0.13808555603027345,
|
|
"num_tokens": 2339449.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.1869316577911375,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 6.0855,
|
|
"mean_token_accuracy": 0.13349030911922455,
|
|
"num_tokens": 2349904.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.193985223770142,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 5.9959,
|
|
"mean_token_accuracy": 0.14620960205793382,
|
|
"num_tokens": 2359833.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.21331262588501,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 6.0311,
|
|
"mean_token_accuracy": 0.13320064768195153,
|
|
"num_tokens": 2369857.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.141865062713623,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.14015598893165587,
|
|
"num_tokens": 2380372.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.128895378112793,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 6.0318,
|
|
"mean_token_accuracy": 0.13270985037088395,
|
|
"num_tokens": 2391239.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.043548440933227,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 6.0133,
|
|
"mean_token_accuracy": 0.14425679594278334,
|
|
"num_tokens": 2401178.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.236666345596314,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.1047,
|
|
"mean_token_accuracy": 0.13325065821409227,
|
|
"num_tokens": 2412496.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.181797790527344,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 6.0895,
|
|
"mean_token_accuracy": 0.13301095142960548,
|
|
"num_tokens": 2423281.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.063150691986084,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.0226,
|
|
"mean_token_accuracy": 0.1412420690059662,
|
|
"num_tokens": 2434078.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.052438020706177,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 5.9671,
|
|
"mean_token_accuracy": 0.13784583508968354,
|
|
"num_tokens": 2443808.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.057710981369018,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.144547900557518,
|
|
"num_tokens": 2453397.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.177269363403321,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 6.0958,
|
|
"mean_token_accuracy": 0.13413558155298233,
|
|
"num_tokens": 2464295.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.149464750289917,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 6.1265,
|
|
"mean_token_accuracy": 0.13496065139770508,
|
|
"num_tokens": 2475311.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.225644922256469,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.1369169607758522,
|
|
"num_tokens": 2485968.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.049617290496826,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 6.0647,
|
|
"mean_token_accuracy": 0.13855249360203742,
|
|
"num_tokens": 2496183.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.143985891342163,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.0173,
|
|
"mean_token_accuracy": 0.13463684692978858,
|
|
"num_tokens": 2506242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.115947341918945,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 6.0327,
|
|
"mean_token_accuracy": 0.1393121913075447,
|
|
"num_tokens": 2516443.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.08875412940979,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.14255313724279403,
|
|
"num_tokens": 2525795.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.09457483291626,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.0394,
|
|
"mean_token_accuracy": 0.13587739318609238,
|
|
"num_tokens": 2535046.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.1398111343383786,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.1354116380214691,
|
|
"num_tokens": 2546489.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.889804172515869,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.13921836167573928,
|
|
"num_tokens": 2556335.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.192789316177368,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 5.9891,
|
|
"mean_token_accuracy": 0.13417705819010733,
|
|
"num_tokens": 2567647.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.187864875793457,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 6.1887,
|
|
"mean_token_accuracy": 0.12929189652204515,
|
|
"num_tokens": 2578618.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.992059373855591,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 5.9961,
|
|
"mean_token_accuracy": 0.14164262935519217,
|
|
"num_tokens": 2587797.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.117119789123535,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.0335,
|
|
"mean_token_accuracy": 0.13243043646216393,
|
|
"num_tokens": 2598403.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.133496522903442,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 6.0459,
|
|
"mean_token_accuracy": 0.13184897378087043,
|
|
"num_tokens": 2608296.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.116444063186646,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 6.07,
|
|
"mean_token_accuracy": 0.13852708488702775,
|
|
"num_tokens": 2617906.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.15281834602356,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 6.0024,
|
|
"mean_token_accuracy": 0.13931000232696533,
|
|
"num_tokens": 2627699.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.051550340652466,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 5.9807,
|
|
"mean_token_accuracy": 0.1385290205478668,
|
|
"num_tokens": 2638384.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.953644514083862,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.9161,
|
|
"mean_token_accuracy": 0.14108227714896202,
|
|
"num_tokens": 2648279.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.1206437110900875,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 5.9976,
|
|
"mean_token_accuracy": 0.12919843792915345,
|
|
"num_tokens": 2657000.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.018803453445434,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 6.0253,
|
|
"mean_token_accuracy": 0.13838643431663514,
|
|
"num_tokens": 2667760.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.1206374168396,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 5.9601,
|
|
"mean_token_accuracy": 0.14021410644054413,
|
|
"num_tokens": 2678450.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.103529644012451,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 6.0707,
|
|
"mean_token_accuracy": 0.12993507608771324,
|
|
"num_tokens": 2689695.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.15411376953125,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 6.0735,
|
|
"mean_token_accuracy": 0.14039076045155524,
|
|
"num_tokens": 2700414.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.982687044143677,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 5.909,
|
|
"mean_token_accuracy": 0.15069574266672134,
|
|
"num_tokens": 2709985.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.035314989089966,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 6.052,
|
|
"mean_token_accuracy": 0.13802565932273864,
|
|
"num_tokens": 2720990.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.01940016746521,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.14141381531953812,
|
|
"num_tokens": 2731285.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.055226230621338,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 6.025,
|
|
"mean_token_accuracy": 0.1359126351773739,
|
|
"num_tokens": 2741024.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.041060304641723,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.9621,
|
|
"mean_token_accuracy": 0.14628610089421273,
|
|
"num_tokens": 2751447.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.027036523818969,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 5.9773,
|
|
"mean_token_accuracy": 0.14544942378997802,
|
|
"num_tokens": 2761283.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.946582794189453,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.14272905886173248,
|
|
"num_tokens": 2771354.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.146433973312378,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.0976,
|
|
"mean_token_accuracy": 0.13832786679267883,
|
|
"num_tokens": 2781063.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.09057092666626,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 6.0078,
|
|
"mean_token_accuracy": 0.13548447340726852,
|
|
"num_tokens": 2791103.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.041825675964356,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 6.0989,
|
|
"mean_token_accuracy": 0.13598124980926513,
|
|
"num_tokens": 2802382.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.058695602416992,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.977,
|
|
"mean_token_accuracy": 0.13661930784583093,
|
|
"num_tokens": 2813740.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.0406389236450195,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.9964,
|
|
"mean_token_accuracy": 0.13657273650169371,
|
|
"num_tokens": 2824121.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.056123876571656,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.13741858452558517,
|
|
"num_tokens": 2833356.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.054589462280274,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.9707,
|
|
"mean_token_accuracy": 0.1379225343465805,
|
|
"num_tokens": 2843839.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.064083385467529,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.14091560244560242,
|
|
"num_tokens": 2853523.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.984784507751465,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 5.9936,
|
|
"mean_token_accuracy": 0.13991155698895455,
|
|
"num_tokens": 2863226.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.107791566848755,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 6.049,
|
|
"mean_token_accuracy": 0.14434949383139611,
|
|
"num_tokens": 2873822.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.959303569793701,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.9023,
|
|
"mean_token_accuracy": 0.14439940378069877,
|
|
"num_tokens": 2883645.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.090818786621094,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.13296996280550957,
|
|
"num_tokens": 2894397.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.072311925888061,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.9994,
|
|
"mean_token_accuracy": 0.14064613804221154,
|
|
"num_tokens": 2904947.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.963215160369873,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9573,
|
|
"mean_token_accuracy": 0.146417498588562,
|
|
"num_tokens": 2916183.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.004500913619995,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.8718,
|
|
"mean_token_accuracy": 0.13890110701322556,
|
|
"num_tokens": 2925429.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.009654474258423,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.9446,
|
|
"mean_token_accuracy": 0.14482987374067308,
|
|
"num_tokens": 2935142.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.021552324295044,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 6.0069,
|
|
"mean_token_accuracy": 0.14064541980624198,
|
|
"num_tokens": 2945151.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.149744176864624,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.13899255618453027,
|
|
"num_tokens": 2956015.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.96809663772583,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.9446,
|
|
"mean_token_accuracy": 0.14438059628009797,
|
|
"num_tokens": 2965879.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.064753627777099,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.13743096590042114,
|
|
"num_tokens": 2977021.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.9406397342681885,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.14179753810167312,
|
|
"num_tokens": 2987169.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.093822956085205,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 6.03,
|
|
"mean_token_accuracy": 0.13984452411532403,
|
|
"num_tokens": 2997666.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.0086753368377686,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.8793,
|
|
"mean_token_accuracy": 0.1382022827863693,
|
|
"num_tokens": 3007021.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.953811836242676,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.1476055011153221,
|
|
"num_tokens": 3016557.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.012690639495849,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.14077139645814896,
|
|
"num_tokens": 3026146.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.085404682159424,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 6.0,
|
|
"mean_token_accuracy": 0.13663262203335763,
|
|
"num_tokens": 3037234.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.953345060348511,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.9299,
|
|
"mean_token_accuracy": 0.1439826861023903,
|
|
"num_tokens": 3047282.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.962337970733643,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.9372,
|
|
"mean_token_accuracy": 0.14204188585281372,
|
|
"num_tokens": 3058648.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.117942428588867,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 6.1158,
|
|
"mean_token_accuracy": 0.13780555948615075,
|
|
"num_tokens": 3068996.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.006206035614014,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.9323,
|
|
"mean_token_accuracy": 0.14948996603488923,
|
|
"num_tokens": 3078991.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.948741769790649,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.14248866513371466,
|
|
"num_tokens": 3090200.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.98205418586731,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 5.9613,
|
|
"mean_token_accuracy": 0.13956795409321784,
|
|
"num_tokens": 3101971.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 5.8623645771381465,
|
|
"eval_loss": 6.217444896697998,
|
|
"eval_mean_token_accuracy": 0.13667063372599525,
|
|
"eval_num_tokens": 3101971.0,
|
|
"eval_runtime": 2.5881,
|
|
"eval_samples_per_second": 1327.248,
|
|
"eval_steps_per_second": 166.148,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.013748121261597,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 6.0524,
|
|
"mean_token_accuracy": 0.13690527454018592,
|
|
"num_tokens": 3113381.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.06113076210022,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 6.0186,
|
|
"mean_token_accuracy": 0.14175765812397004,
|
|
"num_tokens": 3125099.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.9092991828918455,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.14090894535183907,
|
|
"num_tokens": 3135007.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.067477083206176,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.14204106777906417,
|
|
"num_tokens": 3145031.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.966224002838135,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 5.9897,
|
|
"mean_token_accuracy": 0.141945631057024,
|
|
"num_tokens": 3156044.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.921667242050171,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.14390116408467293,
|
|
"num_tokens": 3165124.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.919968223571777,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.14834231361746789,
|
|
"num_tokens": 3175246.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.0169219970703125,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.15311520099639891,
|
|
"num_tokens": 3184817.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.924556303024292,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.13362885937094687,
|
|
"num_tokens": 3195142.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.937474060058594,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.1425031989812851,
|
|
"num_tokens": 3205724.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.995885753631592,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 6.0391,
|
|
"mean_token_accuracy": 0.1389890730381012,
|
|
"num_tokens": 3216919.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.09629340171814,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 6.0424,
|
|
"mean_token_accuracy": 0.13747062683105468,
|
|
"num_tokens": 3228929.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.9776817798614506,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 6.006,
|
|
"mean_token_accuracy": 0.1392039619386196,
|
|
"num_tokens": 3239541.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.988026666641235,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.1428993433713913,
|
|
"num_tokens": 3249527.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.922267484664917,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.7935,
|
|
"mean_token_accuracy": 0.1436563551425934,
|
|
"num_tokens": 3260497.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.053193283081055,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 5.9922,
|
|
"mean_token_accuracy": 0.1404685541987419,
|
|
"num_tokens": 3270865.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.845764875411987,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.9144,
|
|
"mean_token_accuracy": 0.14149952158331872,
|
|
"num_tokens": 3281531.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.028441143035889,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.8969,
|
|
"mean_token_accuracy": 0.14223238006234168,
|
|
"num_tokens": 3291970.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.906997346878052,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.14648545160889626,
|
|
"num_tokens": 3301638.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.10058822631836,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.14588833823800088,
|
|
"num_tokens": 3311780.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.916336250305176,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.8669,
|
|
"mean_token_accuracy": 0.139421559125185,
|
|
"num_tokens": 3322096.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.094791984558105,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 6.0503,
|
|
"mean_token_accuracy": 0.13523295298218727,
|
|
"num_tokens": 3332324.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.867384910583496,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.8217,
|
|
"mean_token_accuracy": 0.15053588896989822,
|
|
"num_tokens": 3342037.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.047509002685547,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.1480130821466446,
|
|
"num_tokens": 3353047.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.838294744491577,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.8337,
|
|
"mean_token_accuracy": 0.14561289623379708,
|
|
"num_tokens": 3363841.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.936257076263428,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.9275,
|
|
"mean_token_accuracy": 0.14525468051433563,
|
|
"num_tokens": 3374086.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.940940475463867,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.9071,
|
|
"mean_token_accuracy": 0.14053232818841935,
|
|
"num_tokens": 3384303.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.96299033164978,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.14530465453863145,
|
|
"num_tokens": 3394858.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.863584804534912,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.888,
|
|
"mean_token_accuracy": 0.15014157742261885,
|
|
"num_tokens": 3405252.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.005671787261963,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.966,
|
|
"mean_token_accuracy": 0.14669394940137864,
|
|
"num_tokens": 3416007.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.060011148452759,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.9903,
|
|
"mean_token_accuracy": 0.13971186727285384,
|
|
"num_tokens": 3426932.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.884844446182251,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.8543,
|
|
"mean_token_accuracy": 0.14739785045385362,
|
|
"num_tokens": 3437262.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.934502410888672,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.8547,
|
|
"mean_token_accuracy": 0.1494481533765793,
|
|
"num_tokens": 3447525.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.959396505355835,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.14804420322179795,
|
|
"num_tokens": 3458198.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.922992897033692,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.8252,
|
|
"mean_token_accuracy": 0.1456183172762394,
|
|
"num_tokens": 3468149.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.9007514953613285,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.8103,
|
|
"mean_token_accuracy": 0.1417421281337738,
|
|
"num_tokens": 3478355.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.9374470710754395,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.8568,
|
|
"mean_token_accuracy": 0.14517948627471924,
|
|
"num_tokens": 3488726.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.8628528118133545,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.14774489551782607,
|
|
"num_tokens": 3498346.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.858314132690429,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.14736251682043075,
|
|
"num_tokens": 3508213.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.035370874404907,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 6.0256,
|
|
"mean_token_accuracy": 0.1460537701845169,
|
|
"num_tokens": 3518978.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.963231611251831,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.9474,
|
|
"mean_token_accuracy": 0.14724959209561347,
|
|
"num_tokens": 3529328.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.8825325012207035,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.14066557437181473,
|
|
"num_tokens": 3539347.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.017431735992432,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.9176,
|
|
"mean_token_accuracy": 0.1371845468878746,
|
|
"num_tokens": 3549798.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.934460401535034,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.13514597415924073,
|
|
"num_tokens": 3560363.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.943513965606689,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.9249,
|
|
"mean_token_accuracy": 0.14672969430685043,
|
|
"num_tokens": 3570080.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.8888551712036135,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.9155,
|
|
"mean_token_accuracy": 0.14709983617067338,
|
|
"num_tokens": 3580053.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.00450325012207,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 6.0062,
|
|
"mean_token_accuracy": 0.13961637243628502,
|
|
"num_tokens": 3590119.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.010839223861694,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.1382573813199997,
|
|
"num_tokens": 3601809.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.0190709114074705,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.8557,
|
|
"mean_token_accuracy": 0.14752294570207597,
|
|
"num_tokens": 3611752.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.89147367477417,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.9062,
|
|
"mean_token_accuracy": 0.14509522318840026,
|
|
"num_tokens": 3621137.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.961899852752685,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8875,
|
|
"mean_token_accuracy": 0.14446437284350394,
|
|
"num_tokens": 3631567.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.900608921051026,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.8352,
|
|
"mean_token_accuracy": 0.1415010467171669,
|
|
"num_tokens": 3641145.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.935684490203857,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.8936,
|
|
"mean_token_accuracy": 0.14037201851606368,
|
|
"num_tokens": 3651716.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.9081618309021,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.14923332259058952,
|
|
"num_tokens": 3662577.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.862698030471802,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.822,
|
|
"mean_token_accuracy": 0.15272447615861892,
|
|
"num_tokens": 3672984.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.935724306106567,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.92,
|
|
"mean_token_accuracy": 0.1462317407131195,
|
|
"num_tokens": 3682634.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.881990909576416,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.14524282217025758,
|
|
"num_tokens": 3693022.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.828023529052734,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.7496,
|
|
"mean_token_accuracy": 0.1534799262881279,
|
|
"num_tokens": 3702234.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.916232109069824,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.14201805889606475,
|
|
"num_tokens": 3712235.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.982822275161743,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.15048668384552003,
|
|
"num_tokens": 3722745.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.823061275482178,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.9155,
|
|
"mean_token_accuracy": 0.14683941900730133,
|
|
"num_tokens": 3732942.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.965871524810791,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.9002,
|
|
"mean_token_accuracy": 0.1446148782968521,
|
|
"num_tokens": 3743806.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.8511744976043705,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.9095,
|
|
"mean_token_accuracy": 0.14839880764484406,
|
|
"num_tokens": 3755115.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.951373863220215,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.1507785961031914,
|
|
"num_tokens": 3765867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.834669971466065,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.1484524592757225,
|
|
"num_tokens": 3777071.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.917655420303345,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.8222,
|
|
"mean_token_accuracy": 0.14832532852888108,
|
|
"num_tokens": 3786554.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.903369140625,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.14705911800265312,
|
|
"num_tokens": 3797627.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.870697450637818,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.8612,
|
|
"mean_token_accuracy": 0.14128738567233085,
|
|
"num_tokens": 3808841.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.941121673583984,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.901,
|
|
"mean_token_accuracy": 0.1427225686609745,
|
|
"num_tokens": 3818995.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.906935453414917,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.9651,
|
|
"mean_token_accuracy": 0.1378663718700409,
|
|
"num_tokens": 3828762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.938121604919433,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.14356590881943704,
|
|
"num_tokens": 3838752.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.8538342952728275,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.15533295422792434,
|
|
"num_tokens": 3849664.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.86170129776001,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.15005387961864472,
|
|
"num_tokens": 3859053.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.939327001571655,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8863,
|
|
"mean_token_accuracy": 0.14259825721383096,
|
|
"num_tokens": 3869654.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.915075254440308,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.884,
|
|
"mean_token_accuracy": 0.1454161912202835,
|
|
"num_tokens": 3879678.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.875418329238892,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.7833,
|
|
"mean_token_accuracy": 0.15064050406217575,
|
|
"num_tokens": 3889950.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.8717896938323975,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.14944285452365874,
|
|
"num_tokens": 3900095.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.997308301925659,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.14228387475013732,
|
|
"num_tokens": 3911034.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.86603946685791,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.8936,
|
|
"mean_token_accuracy": 0.14501007050275802,
|
|
"num_tokens": 3922219.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.895516061782837,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.15025294423103333,
|
|
"num_tokens": 3932235.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.874377489089966,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.7768,
|
|
"mean_token_accuracy": 0.15339176058769227,
|
|
"num_tokens": 3941845.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.899298191070557,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.9012,
|
|
"mean_token_accuracy": 0.1398027829825878,
|
|
"num_tokens": 3953392.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.858644628524781,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.7436,
|
|
"mean_token_accuracy": 0.14781643599271774,
|
|
"num_tokens": 3963294.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.785109663009644,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.1467917636036873,
|
|
"num_tokens": 3973660.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.896430444717407,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.8054,
|
|
"mean_token_accuracy": 0.14625525921583177,
|
|
"num_tokens": 3983587.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.910705471038819,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8745,
|
|
"mean_token_accuracy": 0.14467306286096573,
|
|
"num_tokens": 3994346.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.903839254379273,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.14791052490472795,
|
|
"num_tokens": 4005234.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.809622955322266,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.7882,
|
|
"mean_token_accuracy": 0.1488805055618286,
|
|
"num_tokens": 4015542.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.8069682121276855,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.14918827936053275,
|
|
"num_tokens": 4024874.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.929554414749146,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.938,
|
|
"mean_token_accuracy": 0.1463377967476845,
|
|
"num_tokens": 4035880.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.912116003036499,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.8063,
|
|
"mean_token_accuracy": 0.14950322061777116,
|
|
"num_tokens": 4046484.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.939999675750732,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.1482144773006439,
|
|
"num_tokens": 4057037.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.892278623580933,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.9619,
|
|
"mean_token_accuracy": 0.14090994000434875,
|
|
"num_tokens": 4066169.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.8747725009918215,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.14311319589614868,
|
|
"num_tokens": 4076697.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.814601230621338,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.7538,
|
|
"mean_token_accuracy": 0.15333430469036102,
|
|
"num_tokens": 4087372.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.960805749893188,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.1429354578256607,
|
|
"num_tokens": 4097653.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.89245457649231,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.1475401774048805,
|
|
"num_tokens": 4107381.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.825225067138672,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.9037,
|
|
"mean_token_accuracy": 0.1437897801399231,
|
|
"num_tokens": 4118762.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.885867309570313,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7797,
|
|
"mean_token_accuracy": 0.1499905467033386,
|
|
"num_tokens": 4129011.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.9297198295593265,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.14318699836730958,
|
|
"num_tokens": 4138975.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.665801915457083,
|
|
"eval_loss": 6.055540561676025,
|
|
"eval_mean_token_accuracy": 0.14372314550848894,
|
|
"eval_num_tokens": 4138975.0,
|
|
"eval_runtime": 2.5799,
|
|
"eval_samples_per_second": 1331.439,
|
|
"eval_steps_per_second": 166.672,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.850006914138794,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.14215461909770966,
|
|
"num_tokens": 4150164.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.848128604888916,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.7942,
|
|
"mean_token_accuracy": 0.1535679019987583,
|
|
"num_tokens": 4159872.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.87857141494751,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.8633,
|
|
"mean_token_accuracy": 0.14554824233055114,
|
|
"num_tokens": 4170665.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.9005872249603275,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.7454,
|
|
"mean_token_accuracy": 0.15765557885169984,
|
|
"num_tokens": 4180476.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.901723432540893,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8885,
|
|
"mean_token_accuracy": 0.14066040962934495,
|
|
"num_tokens": 4190740.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.853602123260498,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.7754,
|
|
"mean_token_accuracy": 0.14906278252601624,
|
|
"num_tokens": 4201504.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.717321538925171,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.1566212296485901,
|
|
"num_tokens": 4211614.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.875101900100708,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.14518614783883094,
|
|
"num_tokens": 4221141.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.839630126953125,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.8753,
|
|
"mean_token_accuracy": 0.1478387877345085,
|
|
"num_tokens": 4231877.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.781256294250488,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.6927,
|
|
"mean_token_accuracy": 0.1567344546318054,
|
|
"num_tokens": 4242331.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.85308198928833,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.7201,
|
|
"mean_token_accuracy": 0.14589863792061805,
|
|
"num_tokens": 4252406.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.801605272293091,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.7936,
|
|
"mean_token_accuracy": 0.149433696269989,
|
|
"num_tokens": 4263036.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.796260404586792,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.1521616980433464,
|
|
"num_tokens": 4273032.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.857989358901977,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.8452,
|
|
"mean_token_accuracy": 0.15023571699857713,
|
|
"num_tokens": 4283360.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.809034204483032,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.7843,
|
|
"mean_token_accuracy": 0.1533666417002678,
|
|
"num_tokens": 4292623.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.8134842872619625,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.7965,
|
|
"mean_token_accuracy": 0.15429155230522157,
|
|
"num_tokens": 4302817.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.876812982559204,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.8095,
|
|
"mean_token_accuracy": 0.14984023720026016,
|
|
"num_tokens": 4313049.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.797815656661987,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.6837,
|
|
"mean_token_accuracy": 0.1588327094912529,
|
|
"num_tokens": 4322873.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.746478509902954,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.8496,
|
|
"mean_token_accuracy": 0.1438209369778633,
|
|
"num_tokens": 4333339.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.87603907585144,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.8166,
|
|
"mean_token_accuracy": 0.15194771736860274,
|
|
"num_tokens": 4344917.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.70527286529541,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.6632,
|
|
"mean_token_accuracy": 0.1579922318458557,
|
|
"num_tokens": 4354140.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.827756881713867,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.8947,
|
|
"mean_token_accuracy": 0.15089598596096038,
|
|
"num_tokens": 4364976.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.831982851028442,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.6882,
|
|
"mean_token_accuracy": 0.15953390002250672,
|
|
"num_tokens": 4376423.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.826807928085327,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.8343,
|
|
"mean_token_accuracy": 0.15021495670080184,
|
|
"num_tokens": 4387865.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.803925037384033,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.14990636855363845,
|
|
"num_tokens": 4398275.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.842619180679321,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.7401,
|
|
"mean_token_accuracy": 0.15128927528858185,
|
|
"num_tokens": 4407396.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.016067266464233,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 6.0015,
|
|
"mean_token_accuracy": 0.1389892116189003,
|
|
"num_tokens": 4419682.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.844279813766479,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.8686,
|
|
"mean_token_accuracy": 0.14963124543428422,
|
|
"num_tokens": 4430590.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.866333293914795,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7741,
|
|
"mean_token_accuracy": 0.15342569425702096,
|
|
"num_tokens": 4441979.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.817252540588379,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.7839,
|
|
"mean_token_accuracy": 0.15365557819604875,
|
|
"num_tokens": 4451975.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.743074512481689,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.7384,
|
|
"mean_token_accuracy": 0.1501651734113693,
|
|
"num_tokens": 4462778.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.854450845718384,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.7804,
|
|
"mean_token_accuracy": 0.16065036952495576,
|
|
"num_tokens": 4472461.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.789333438873291,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.7424,
|
|
"mean_token_accuracy": 0.1531789280474186,
|
|
"num_tokens": 4482507.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.851453542709351,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.7577,
|
|
"mean_token_accuracy": 0.1508588284254074,
|
|
"num_tokens": 4492601.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.819383144378662,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.14919489920139312,
|
|
"num_tokens": 4503696.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.857596826553345,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.7553,
|
|
"mean_token_accuracy": 0.15510910004377365,
|
|
"num_tokens": 4513649.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.83114333152771,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.7761,
|
|
"mean_token_accuracy": 0.15262338742613793,
|
|
"num_tokens": 4524054.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.82422456741333,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.1471053257584572,
|
|
"num_tokens": 4535549.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.895087671279907,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.803,
|
|
"mean_token_accuracy": 0.1454395830631256,
|
|
"num_tokens": 4545654.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.807495403289795,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.15400820821523667,
|
|
"num_tokens": 4556230.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.776730272505018,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.15552351044283974,
|
|
"num_tokens": 4564486.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.818914890289307,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.15918837636709213,
|
|
"num_tokens": 4575026.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.6877833843231205,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.4734,
|
|
"mean_token_accuracy": 0.15992515832185744,
|
|
"num_tokens": 4585902.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.878614091873169,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.5867,
|
|
"mean_token_accuracy": 0.1587267853319645,
|
|
"num_tokens": 4597609.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.757115221023559,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.4537,
|
|
"mean_token_accuracy": 0.15991983264684678,
|
|
"num_tokens": 4607531.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.649359035491943,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.15773818045854568,
|
|
"num_tokens": 4618826.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.81802773475647,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.16191642731428146,
|
|
"num_tokens": 4629526.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.770311784744263,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.15500493198633195,
|
|
"num_tokens": 4639930.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.765829420089721,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.1719053119421005,
|
|
"num_tokens": 4649661.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.6249712944030765,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.16899963915348054,
|
|
"num_tokens": 4660178.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.808508110046387,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.4678,
|
|
"mean_token_accuracy": 0.16111961156129836,
|
|
"num_tokens": 4670417.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.640920877456665,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.4193,
|
|
"mean_token_accuracy": 0.17042539566755294,
|
|
"num_tokens": 4680221.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.732649278640747,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.15786078870296477,
|
|
"num_tokens": 4691136.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.62737078666687,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.331,
|
|
"mean_token_accuracy": 0.17443092912435532,
|
|
"num_tokens": 4700879.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.730466222763061,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.4986,
|
|
"mean_token_accuracy": 0.15935180932283402,
|
|
"num_tokens": 4711477.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.717178630828857,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.4775,
|
|
"mean_token_accuracy": 0.16150320321321487,
|
|
"num_tokens": 4721973.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.600577306747437,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.4641,
|
|
"mean_token_accuracy": 0.163180872797966,
|
|
"num_tokens": 4732425.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.735846757888794,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.1671282097697258,
|
|
"num_tokens": 4743377.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.7213092803955075,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.5086,
|
|
"mean_token_accuracy": 0.1578649789094925,
|
|
"num_tokens": 4753534.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.66147928237915,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.473,
|
|
"mean_token_accuracy": 0.16131037175655366,
|
|
"num_tokens": 4763159.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.664205121994018,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.16794643849134444,
|
|
"num_tokens": 4773497.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.693774890899658,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.5337,
|
|
"mean_token_accuracy": 0.1682440832257271,
|
|
"num_tokens": 4784291.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.674150276184082,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4914,
|
|
"mean_token_accuracy": 0.16157547384500504,
|
|
"num_tokens": 4794762.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.823915672302246,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.1603628158569336,
|
|
"num_tokens": 4804449.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.680734157562256,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.4526,
|
|
"mean_token_accuracy": 0.1679395079612732,
|
|
"num_tokens": 4814172.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.713951063156128,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.5447,
|
|
"mean_token_accuracy": 0.15825924128293992,
|
|
"num_tokens": 4824909.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.787033605575561,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5809,
|
|
"mean_token_accuracy": 0.16325388252735137,
|
|
"num_tokens": 4834546.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.660865402221679,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.16167585253715516,
|
|
"num_tokens": 4844809.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.695849227905273,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.4994,
|
|
"mean_token_accuracy": 0.16390514522790908,
|
|
"num_tokens": 4855883.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.6910813331604,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.5156,
|
|
"mean_token_accuracy": 0.16123198270797728,
|
|
"num_tokens": 4865936.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.661603689193726,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.4457,
|
|
"mean_token_accuracy": 0.16277610659599304,
|
|
"num_tokens": 4876621.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.550080060958862,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.415,
|
|
"mean_token_accuracy": 0.16018703430891038,
|
|
"num_tokens": 4886315.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.736575651168823,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4893,
|
|
"mean_token_accuracy": 0.15953089892864228,
|
|
"num_tokens": 4897214.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.727215814590454,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4881,
|
|
"mean_token_accuracy": 0.15968473702669145,
|
|
"num_tokens": 4907360.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.634088802337646,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.16416353285312651,
|
|
"num_tokens": 4917591.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.664371538162231,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.15389032512903214,
|
|
"num_tokens": 4928075.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.737252044677734,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.4289,
|
|
"mean_token_accuracy": 0.16773451566696168,
|
|
"num_tokens": 4937898.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.642275476455689,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.15992920845746994,
|
|
"num_tokens": 4948152.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.719000434875488,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.16493412107229233,
|
|
"num_tokens": 4957973.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.5868752002716064,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.4466,
|
|
"mean_token_accuracy": 0.16483312994241714,
|
|
"num_tokens": 4967375.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.713901090621948,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.17020453214645387,
|
|
"num_tokens": 4977286.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.581647205352783,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.534,
|
|
"mean_token_accuracy": 0.15994520634412765,
|
|
"num_tokens": 4987930.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.749277925491333,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.5706,
|
|
"mean_token_accuracy": 0.15547981709241868,
|
|
"num_tokens": 4998707.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.6243288040161135,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.4206,
|
|
"mean_token_accuracy": 0.17375861257314681,
|
|
"num_tokens": 5008540.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.60424427986145,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4885,
|
|
"mean_token_accuracy": 0.16451358944177627,
|
|
"num_tokens": 5018199.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.764675283432007,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.5024,
|
|
"mean_token_accuracy": 0.1648518167436123,
|
|
"num_tokens": 5028251.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.605207777023315,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.15732037276029587,
|
|
"num_tokens": 5038290.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.6687627792358395,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.4552,
|
|
"mean_token_accuracy": 0.16057046353816987,
|
|
"num_tokens": 5048337.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.694064283370972,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.16235274076461792,
|
|
"num_tokens": 5058533.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.7049397945404055,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.5072,
|
|
"mean_token_accuracy": 0.16149232983589173,
|
|
"num_tokens": 5068556.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.601208972930908,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.5945,
|
|
"mean_token_accuracy": 0.15727268159389496,
|
|
"num_tokens": 5079317.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.795121812820435,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.15894168838858605,
|
|
"num_tokens": 5089544.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.669192790985107,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.15734344869852065,
|
|
"num_tokens": 5099491.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.727242374420166,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.1640946477651596,
|
|
"num_tokens": 5109523.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.66324372291565,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.5058,
|
|
"mean_token_accuracy": 0.16412295550107955,
|
|
"num_tokens": 5119940.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.644521474838257,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.4485,
|
|
"mean_token_accuracy": 0.16864456087350846,
|
|
"num_tokens": 5129789.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.670347404479981,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.4734,
|
|
"mean_token_accuracy": 0.16284793466329575,
|
|
"num_tokens": 5139570.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.645919847488403,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.4486,
|
|
"mean_token_accuracy": 0.15870859622955322,
|
|
"num_tokens": 5150526.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.616074180603027,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.4081,
|
|
"mean_token_accuracy": 0.15834681540727616,
|
|
"num_tokens": 5160895.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.644442319869995,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.1621962457895279,
|
|
"num_tokens": 5171054.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.555895900726318,
|
|
"eval_loss": 5.985411643981934,
|
|
"eval_mean_token_accuracy": 0.1481271107529485,
|
|
"eval_num_tokens": 5171054.0,
|
|
"eval_runtime": 2.7841,
|
|
"eval_samples_per_second": 1233.778,
|
|
"eval_steps_per_second": 154.447,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.674577713012695,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.5594,
|
|
"mean_token_accuracy": 0.1488342896103859,
|
|
"num_tokens": 5180870.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.648743724822998,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.1596286565065384,
|
|
"num_tokens": 5190388.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.680267333984375,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4876,
|
|
"mean_token_accuracy": 0.15813228935003282,
|
|
"num_tokens": 5200582.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.6623788356781,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.16308542788028718,
|
|
"num_tokens": 5211079.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.674690771102905,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.15394625142216684,
|
|
"num_tokens": 5221481.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.70333890914917,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.4554,
|
|
"mean_token_accuracy": 0.1672426626086235,
|
|
"num_tokens": 5231387.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.532264280319214,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.4233,
|
|
"mean_token_accuracy": 0.16957077980041504,
|
|
"num_tokens": 5240758.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.610846138000488,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.4277,
|
|
"mean_token_accuracy": 0.1651925578713417,
|
|
"num_tokens": 5250839.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.714931344985962,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.554,
|
|
"mean_token_accuracy": 0.1592802345752716,
|
|
"num_tokens": 5261554.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.582263612747193,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.506,
|
|
"mean_token_accuracy": 0.15975702106952666,
|
|
"num_tokens": 5272422.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.690386199951172,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4529,
|
|
"mean_token_accuracy": 0.16031704768538474,
|
|
"num_tokens": 5282522.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.668948554992676,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.16167995631694793,
|
|
"num_tokens": 5293100.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.651067543029785,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.5332,
|
|
"mean_token_accuracy": 0.15431610196828843,
|
|
"num_tokens": 5303103.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.64556474685669,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.5145,
|
|
"mean_token_accuracy": 0.1608550265431404,
|
|
"num_tokens": 5313765.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.759136438369751,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.5774,
|
|
"mean_token_accuracy": 0.16175244748592377,
|
|
"num_tokens": 5324816.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.651179790496826,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.1617589548230171,
|
|
"num_tokens": 5334882.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.596815586090088,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.4726,
|
|
"mean_token_accuracy": 0.16438071429729462,
|
|
"num_tokens": 5344623.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.64653058052063,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.491,
|
|
"mean_token_accuracy": 0.1641414776444435,
|
|
"num_tokens": 5355786.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.584691762924194,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.4818,
|
|
"mean_token_accuracy": 0.1628663197159767,
|
|
"num_tokens": 5366139.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.799052381515503,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.5787,
|
|
"mean_token_accuracy": 0.15531208068132402,
|
|
"num_tokens": 5376890.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.639358854293823,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.16475296020507812,
|
|
"num_tokens": 5386602.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.651406955718994,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.15713362246751786,
|
|
"num_tokens": 5397749.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.688648891448975,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.5156,
|
|
"mean_token_accuracy": 0.15555270165205,
|
|
"num_tokens": 5407666.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.601361083984375,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.6045,
|
|
"mean_token_accuracy": 0.15360901057720183,
|
|
"num_tokens": 5418668.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.653815793991089,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.4685,
|
|
"mean_token_accuracy": 0.16762498021125793,
|
|
"num_tokens": 5429098.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.654842329025269,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.4137,
|
|
"mean_token_accuracy": 0.17132281064987182,
|
|
"num_tokens": 5438997.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.630882740020752,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.15896065831184386,
|
|
"num_tokens": 5449315.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.582456922531128,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.5051,
|
|
"mean_token_accuracy": 0.1627444103360176,
|
|
"num_tokens": 5460523.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.717349052429199,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.1615318164229393,
|
|
"num_tokens": 5470012.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.672448635101318,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.16089946180582046,
|
|
"num_tokens": 5480364.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.635152530670166,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.6339,
|
|
"mean_token_accuracy": 0.1500776916742325,
|
|
"num_tokens": 5492138.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.684290075302124,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.3933,
|
|
"mean_token_accuracy": 0.17207786440849304,
|
|
"num_tokens": 5501935.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.6565260887146,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.5972,
|
|
"mean_token_accuracy": 0.154171322286129,
|
|
"num_tokens": 5513993.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.686792659759521,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.16697083562612533,
|
|
"num_tokens": 5524822.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.64971694946289,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.1643294706940651,
|
|
"num_tokens": 5535583.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.625869369506836,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.4771,
|
|
"mean_token_accuracy": 0.1663047343492508,
|
|
"num_tokens": 5546199.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.676605463027954,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.4318,
|
|
"mean_token_accuracy": 0.1690910592675209,
|
|
"num_tokens": 5556409.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.598950433731079,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.17030078768730164,
|
|
"num_tokens": 5565854.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.700905990600586,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.1541795924305916,
|
|
"num_tokens": 5576194.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.649804401397705,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.16591859608888626,
|
|
"num_tokens": 5586679.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.603459119796753,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.1541197583079338,
|
|
"num_tokens": 5597268.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.687775325775147,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4882,
|
|
"mean_token_accuracy": 0.16045346409082412,
|
|
"num_tokens": 5608046.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.627590370178223,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.3887,
|
|
"mean_token_accuracy": 0.1690044417977333,
|
|
"num_tokens": 5617541.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.662360763549804,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.4378,
|
|
"mean_token_accuracy": 0.16200482249259948,
|
|
"num_tokens": 5628428.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.656942558288574,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.6359,
|
|
"mean_token_accuracy": 0.14996092095971109,
|
|
"num_tokens": 5639557.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.696690559387207,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.16147277355194092,
|
|
"num_tokens": 5650586.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.693271589279175,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.516,
|
|
"mean_token_accuracy": 0.16273925602436065,
|
|
"num_tokens": 5661525.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.631182050704956,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.16119054555892945,
|
|
"num_tokens": 5671272.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.637885189056396,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.16459427624940873,
|
|
"num_tokens": 5680384.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.655943727493286,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.16809071302413942,
|
|
"num_tokens": 5690983.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.598234844207764,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.501,
|
|
"mean_token_accuracy": 0.16032744497060775,
|
|
"num_tokens": 5702144.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.667960977554321,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.5737,
|
|
"mean_token_accuracy": 0.15668509304523467,
|
|
"num_tokens": 5713283.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.65684142112732,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.15997955054044724,
|
|
"num_tokens": 5724533.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.610548067092895,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.4874,
|
|
"mean_token_accuracy": 0.16011036485433577,
|
|
"num_tokens": 5735186.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.672524642944336,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.5376,
|
|
"mean_token_accuracy": 0.15489088892936706,
|
|
"num_tokens": 5746537.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.68559250831604,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.5192,
|
|
"mean_token_accuracy": 0.162811042368412,
|
|
"num_tokens": 5756666.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.63301682472229,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.5241,
|
|
"mean_token_accuracy": 0.15739674717187882,
|
|
"num_tokens": 5766870.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.640361738204956,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.1637111410498619,
|
|
"num_tokens": 5776775.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.733455991744995,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.15455977171659468,
|
|
"num_tokens": 5786993.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.6326195240020756,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.16543082892894745,
|
|
"num_tokens": 5796230.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.689797687530517,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.1628267914056778,
|
|
"num_tokens": 5807086.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.6216188907623295,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.5469,
|
|
"mean_token_accuracy": 0.15895194411277772,
|
|
"num_tokens": 5817897.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.678765535354614,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.4283,
|
|
"mean_token_accuracy": 0.17189880907535554,
|
|
"num_tokens": 5827794.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.579028034210205,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.1584237337112427,
|
|
"num_tokens": 5837603.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.595484399795533,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.4373,
|
|
"mean_token_accuracy": 0.1674399733543396,
|
|
"num_tokens": 5847013.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.663733386993409,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.1666719898581505,
|
|
"num_tokens": 5856783.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.560027599334717,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.4447,
|
|
"mean_token_accuracy": 0.16549039483070374,
|
|
"num_tokens": 5866241.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.573923921585083,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.1657933548092842,
|
|
"num_tokens": 5876135.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.559563684463501,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.5299,
|
|
"mean_token_accuracy": 0.16448313742876053,
|
|
"num_tokens": 5886841.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.668076229095459,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.16217845380306245,
|
|
"num_tokens": 5896884.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.638378667831421,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.1620452508330345,
|
|
"num_tokens": 5906194.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.620361804962158,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.4327,
|
|
"mean_token_accuracy": 0.1661418780684471,
|
|
"num_tokens": 5917217.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.617027521133423,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.389,
|
|
"mean_token_accuracy": 0.1645030215382576,
|
|
"num_tokens": 5926307.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.57777681350708,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.485,
|
|
"mean_token_accuracy": 0.15858202278614045,
|
|
"num_tokens": 5936718.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.558912324905395,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.4469,
|
|
"mean_token_accuracy": 0.16518836319446564,
|
|
"num_tokens": 5947870.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.647074413299561,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.6199,
|
|
"mean_token_accuracy": 0.15892106890678406,
|
|
"num_tokens": 5959090.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.696274280548096,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.5931,
|
|
"mean_token_accuracy": 0.1565147966146469,
|
|
"num_tokens": 5970435.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.607390022277832,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.4102,
|
|
"mean_token_accuracy": 0.17200414836406708,
|
|
"num_tokens": 5979732.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.582248306274414,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.5628,
|
|
"mean_token_accuracy": 0.16025504767894744,
|
|
"num_tokens": 5989965.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.720489931106568,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.16599828898906707,
|
|
"num_tokens": 6000752.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.633650016784668,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.1575797587633133,
|
|
"num_tokens": 6011318.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.657891941070557,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.16536905020475387,
|
|
"num_tokens": 6022024.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.595902967453003,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.16532984972000123,
|
|
"num_tokens": 6032118.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.676761198043823,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.4882,
|
|
"mean_token_accuracy": 0.1679046556353569,
|
|
"num_tokens": 6042607.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.571296691894531,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.4921,
|
|
"mean_token_accuracy": 0.16592631489038467,
|
|
"num_tokens": 6053625.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.616884326934814,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.4375,
|
|
"mean_token_accuracy": 0.17182671129703522,
|
|
"num_tokens": 6063823.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.633875751495362,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.4494,
|
|
"mean_token_accuracy": 0.16598895341157913,
|
|
"num_tokens": 6073914.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.594155645370483,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.4362,
|
|
"mean_token_accuracy": 0.16640295535326005,
|
|
"num_tokens": 6085121.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.530791234970093,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.3531,
|
|
"mean_token_accuracy": 0.17730343341827393,
|
|
"num_tokens": 6095324.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.575316858291626,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.1687125161290169,
|
|
"num_tokens": 6104704.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.626064872741699,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.15955478847026824,
|
|
"num_tokens": 6116569.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.770713663101196,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.5709,
|
|
"mean_token_accuracy": 0.15991768836975098,
|
|
"num_tokens": 6126166.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.641952037811279,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.1669522613286972,
|
|
"num_tokens": 6136169.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.6181262016296385,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.5145,
|
|
"mean_token_accuracy": 0.16996010690927504,
|
|
"num_tokens": 6146249.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.739789962768555,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.5842,
|
|
"mean_token_accuracy": 0.16424653083086013,
|
|
"num_tokens": 6156892.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.615502405166626,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.465,
|
|
"mean_token_accuracy": 0.17193833142518997,
|
|
"num_tokens": 6166915.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.523143291473389,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.3964,
|
|
"mean_token_accuracy": 0.17168645113706588,
|
|
"num_tokens": 6176837.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.564775657653809,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.16342125385999678,
|
|
"num_tokens": 6188038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.622974061965943,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.16570041030645372,
|
|
"num_tokens": 6199835.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.682276248931885,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.4372,
|
|
"mean_token_accuracy": 0.1666649267077446,
|
|
"num_tokens": 6209122.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.535053278679071,
|
|
"eval_loss": 5.9162821769714355,
|
|
"eval_mean_token_accuracy": 0.15202350129568298,
|
|
"eval_num_tokens": 6209122.0,
|
|
"eval_runtime": 2.5831,
|
|
"eval_samples_per_second": 1329.775,
|
|
"eval_steps_per_second": 166.464,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.57192440032959,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4876,
|
|
"mean_token_accuracy": 0.16483847349882125,
|
|
"num_tokens": 6219317.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.6103438377380375,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.5222,
|
|
"mean_token_accuracy": 0.16368789076805115,
|
|
"num_tokens": 6229561.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.685789155960083,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.514,
|
|
"mean_token_accuracy": 0.16079635322093963,
|
|
"num_tokens": 6240150.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.599572849273682,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.4052,
|
|
"mean_token_accuracy": 0.16487381011247634,
|
|
"num_tokens": 6250641.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.585576963424683,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.4708,
|
|
"mean_token_accuracy": 0.16177722662687302,
|
|
"num_tokens": 6261239.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.574017000198364,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.1596275195479393,
|
|
"num_tokens": 6270919.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.68271541595459,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.15828007757663726,
|
|
"num_tokens": 6283007.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.516647815704346,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.1702626973390579,
|
|
"num_tokens": 6293596.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.570477151870728,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.1650572270154953,
|
|
"num_tokens": 6304568.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.611292982101441,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.4443,
|
|
"mean_token_accuracy": 0.1659901887178421,
|
|
"num_tokens": 6315275.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.621387338638305,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.16356482207775117,
|
|
"num_tokens": 6324986.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.626633882522583,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.1616196423768997,
|
|
"num_tokens": 6336318.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.606946229934692,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.4214,
|
|
"mean_token_accuracy": 0.17164656668901443,
|
|
"num_tokens": 6345986.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.619322729110718,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.5134,
|
|
"mean_token_accuracy": 0.16361067295074463,
|
|
"num_tokens": 6357025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.622792339324951,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.5053,
|
|
"mean_token_accuracy": 0.1639904797077179,
|
|
"num_tokens": 6367885.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.483027648925781,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.4151,
|
|
"mean_token_accuracy": 0.16503288745880126,
|
|
"num_tokens": 6377729.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.534521055221558,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.17016067653894423,
|
|
"num_tokens": 6387368.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.653550815582276,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.5398,
|
|
"mean_token_accuracy": 0.15785185843706132,
|
|
"num_tokens": 6397771.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.654391050338745,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.17247918248176575,
|
|
"num_tokens": 6408238.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.591338062286377,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.5645,
|
|
"mean_token_accuracy": 0.162879978120327,
|
|
"num_tokens": 6419196.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.63900990486145,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.16121173799037933,
|
|
"num_tokens": 6430342.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.599709701538086,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.399,
|
|
"mean_token_accuracy": 0.16971081644296646,
|
|
"num_tokens": 6440218.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.546879053115845,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.1728322461247444,
|
|
"num_tokens": 6450128.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.5785768032073975,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.421,
|
|
"mean_token_accuracy": 0.1764913856983185,
|
|
"num_tokens": 6459916.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.616512584686279,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.4565,
|
|
"mean_token_accuracy": 0.16466539949178696,
|
|
"num_tokens": 6470250.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.620808506011963,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.15890538394451142,
|
|
"num_tokens": 6481623.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.59339017868042,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.4539,
|
|
"mean_token_accuracy": 0.16420140564441682,
|
|
"num_tokens": 6492143.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.686143970489502,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.1655203655362129,
|
|
"num_tokens": 6501643.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.530097484588623,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.396,
|
|
"mean_token_accuracy": 0.17422914057970046,
|
|
"num_tokens": 6511442.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.546724271774292,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.3295,
|
|
"mean_token_accuracy": 0.17118811905384063,
|
|
"num_tokens": 6521303.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.541709566116333,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.475,
|
|
"mean_token_accuracy": 0.16334498822689056,
|
|
"num_tokens": 6531143.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.609386014938354,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.5254,
|
|
"mean_token_accuracy": 0.16524550169706345,
|
|
"num_tokens": 6540899.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.740256404876709,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.5532,
|
|
"mean_token_accuracy": 0.16046922653913498,
|
|
"num_tokens": 6552131.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.570868825912475,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.4969,
|
|
"mean_token_accuracy": 0.16289133280515672,
|
|
"num_tokens": 6563275.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.581732654571534,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.355,
|
|
"mean_token_accuracy": 0.17269828915596008,
|
|
"num_tokens": 6573109.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.5417101860046385,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.471,
|
|
"mean_token_accuracy": 0.16080929785966874,
|
|
"num_tokens": 6584978.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.518564605712891,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.4834,
|
|
"mean_token_accuracy": 0.1639021337032318,
|
|
"num_tokens": 6595608.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.5223010063171385,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.371,
|
|
"mean_token_accuracy": 0.16502223908901215,
|
|
"num_tokens": 6605983.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.614612340927124,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.1647695019841194,
|
|
"num_tokens": 6616682.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.722541761398316,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.16358522474765777,
|
|
"num_tokens": 6628715.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.623310518264771,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.16629290133714675,
|
|
"num_tokens": 6639219.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.511093616485596,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.3862,
|
|
"mean_token_accuracy": 0.1678118109703064,
|
|
"num_tokens": 6649516.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.6276873588562015,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.502,
|
|
"mean_token_accuracy": 0.16501290798187257,
|
|
"num_tokens": 6659470.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.589895868301392,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.4786,
|
|
"mean_token_accuracy": 0.1676536723971367,
|
|
"num_tokens": 6669637.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.533270168304443,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.4241,
|
|
"mean_token_accuracy": 0.16494055092334747,
|
|
"num_tokens": 6680450.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.630631971359253,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.16183138638734818,
|
|
"num_tokens": 6691526.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.629131555557251,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.4568,
|
|
"mean_token_accuracy": 0.1689591586589813,
|
|
"num_tokens": 6702543.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.596776628494263,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.5303,
|
|
"mean_token_accuracy": 0.15820375829935074,
|
|
"num_tokens": 6712374.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.595458602905273,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.16538548320531846,
|
|
"num_tokens": 6722091.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.571884536743164,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.4069,
|
|
"mean_token_accuracy": 0.17532299160957338,
|
|
"num_tokens": 6732024.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.610106992721557,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.5437,
|
|
"mean_token_accuracy": 0.1645393580198288,
|
|
"num_tokens": 6741480.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.612904214859009,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.4091,
|
|
"mean_token_accuracy": 0.16806740015745164,
|
|
"num_tokens": 6752504.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.640478181838989,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.1602922037243843,
|
|
"num_tokens": 6762945.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.551138257980346,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.16794790625572203,
|
|
"num_tokens": 6772862.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.576385736465454,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.4469,
|
|
"mean_token_accuracy": 0.16942329555749894,
|
|
"num_tokens": 6782912.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.643446207046509,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.16744696944952012,
|
|
"num_tokens": 6793249.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.593337774276733,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.1653804063796997,
|
|
"num_tokens": 6803721.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.695056200027466,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.4738,
|
|
"mean_token_accuracy": 0.16928725838661193,
|
|
"num_tokens": 6813637.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.636111545562744,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.16312997192144393,
|
|
"num_tokens": 6824171.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.440194320678711,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.3167,
|
|
"mean_token_accuracy": 0.17693583071231841,
|
|
"num_tokens": 6833668.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.507216612497966,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.2892,
|
|
"mean_token_accuracy": 0.17582226461834377,
|
|
"num_tokens": 6842297.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.667935991287232,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.2536,
|
|
"mean_token_accuracy": 0.166383495926857,
|
|
"num_tokens": 6853124.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.571122550964356,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.1184,
|
|
"mean_token_accuracy": 0.18871196508407592,
|
|
"num_tokens": 6861990.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.530049896240234,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.18157308101654052,
|
|
"num_tokens": 6872024.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.512989568710327,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.18028032034635544,
|
|
"num_tokens": 6881205.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.466656017303467,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.1498,
|
|
"mean_token_accuracy": 0.1791322261095047,
|
|
"num_tokens": 6891741.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.503334999084473,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.25,
|
|
"mean_token_accuracy": 0.1725217118859291,
|
|
"num_tokens": 6902122.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.564512491226196,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.1908,
|
|
"mean_token_accuracy": 0.1777113676071167,
|
|
"num_tokens": 6912129.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.552100563049317,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 5.0951,
|
|
"mean_token_accuracy": 0.18567386865615845,
|
|
"num_tokens": 6921409.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.501531600952148,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.2116,
|
|
"mean_token_accuracy": 0.1765879675745964,
|
|
"num_tokens": 6932346.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.529948282241821,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.1994,
|
|
"mean_token_accuracy": 0.1752462640404701,
|
|
"num_tokens": 6942139.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.562569904327392,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.2323,
|
|
"mean_token_accuracy": 0.17586120814085007,
|
|
"num_tokens": 6952344.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.6356312274932865,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.17907728552818297,
|
|
"num_tokens": 6962430.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.567461109161377,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.26,
|
|
"mean_token_accuracy": 0.16686835438013076,
|
|
"num_tokens": 6973213.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.484992551803589,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.17580418139696122,
|
|
"num_tokens": 6983896.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.519098854064941,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.152,
|
|
"mean_token_accuracy": 0.1856158643960953,
|
|
"num_tokens": 6993549.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.505546045303345,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.2066,
|
|
"mean_token_accuracy": 0.17739834636449814,
|
|
"num_tokens": 7003893.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.6110554218292235,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.3069,
|
|
"mean_token_accuracy": 0.17104706764221192,
|
|
"num_tokens": 7014826.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.606646394729614,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.2595,
|
|
"mean_token_accuracy": 0.17323583513498306,
|
|
"num_tokens": 7024758.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.500548696517944,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.182036592066288,
|
|
"num_tokens": 7034063.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.503037214279175,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.2152,
|
|
"mean_token_accuracy": 0.17064046561717988,
|
|
"num_tokens": 7045137.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.4756406307220455,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.1527,
|
|
"mean_token_accuracy": 0.17823233008384703,
|
|
"num_tokens": 7056381.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.575207281112671,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.287,
|
|
"mean_token_accuracy": 0.16938237994909286,
|
|
"num_tokens": 7066462.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.568540906906128,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.3135,
|
|
"mean_token_accuracy": 0.1692761242389679,
|
|
"num_tokens": 7078438.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.555954837799073,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.17892565131187438,
|
|
"num_tokens": 7088981.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.497908401489258,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.2238,
|
|
"mean_token_accuracy": 0.17320664823055268,
|
|
"num_tokens": 7099407.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.479109477996826,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 5.108,
|
|
"mean_token_accuracy": 0.18698589950799943,
|
|
"num_tokens": 7109454.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.400720262527466,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.2288,
|
|
"mean_token_accuracy": 0.17490898668766022,
|
|
"num_tokens": 7119639.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.527888250350952,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.1821,
|
|
"mean_token_accuracy": 0.1765451103448868,
|
|
"num_tokens": 7129872.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.506334066390991,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 5.119,
|
|
"mean_token_accuracy": 0.18061938136816025,
|
|
"num_tokens": 7140456.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.409275054931641,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.1559,
|
|
"mean_token_accuracy": 0.17900101989507675,
|
|
"num_tokens": 7149787.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.455431985855102,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.1803710490465164,
|
|
"num_tokens": 7160482.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.5051344871521,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.1626,
|
|
"mean_token_accuracy": 0.17794452011585235,
|
|
"num_tokens": 7170818.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.470101547241211,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.203,
|
|
"mean_token_accuracy": 0.1818641021847725,
|
|
"num_tokens": 7180770.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.398298263549805,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 5.1269,
|
|
"mean_token_accuracy": 0.18532654345035554,
|
|
"num_tokens": 7190751.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.476519536972046,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.183659166097641,
|
|
"num_tokens": 7200093.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.593797397613526,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.2342,
|
|
"mean_token_accuracy": 0.17738402932882308,
|
|
"num_tokens": 7210526.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.42172327041626,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 5.1997,
|
|
"mean_token_accuracy": 0.17890812158584596,
|
|
"num_tokens": 7222127.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.393723869323731,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.1622,
|
|
"mean_token_accuracy": 0.18311763405799866,
|
|
"num_tokens": 7233418.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.511578607559204,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.2105,
|
|
"mean_token_accuracy": 0.17556524127721787,
|
|
"num_tokens": 7243743.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.421748930908913,
|
|
"eval_loss": 5.898947715759277,
|
|
"eval_mean_token_accuracy": 0.15511425153114075,
|
|
"eval_num_tokens": 7243743.0,
|
|
"eval_runtime": 2.7746,
|
|
"eval_samples_per_second": 1237.999,
|
|
"eval_steps_per_second": 154.975,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.491481637954712,
|
|
"epoch": 3.1835529304861425,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004338989704678364,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.16904491633176805,
|
|
"num_tokens": 7254906.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.477366542816162,
|
|
"epoch": 3.1880963198546115,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00043364854852027566,
|
|
"loss": 5.198,
|
|
"mean_token_accuracy": 0.17554290741682052,
|
|
"num_tokens": 7264193.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.5823265552520756,
|
|
"epoch": 3.1926397092230805,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004333977351231924,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.17801016867160796,
|
|
"num_tokens": 7273723.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.519900178909301,
|
|
"epoch": 3.1971830985915495,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043314653089544336,
|
|
"loss": 5.2803,
|
|
"mean_token_accuracy": 0.16871743649244308,
|
|
"num_tokens": 7284142.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.49244327545166,
|
|
"epoch": 3.201726487960018,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004328949364568504,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.18007822036743165,
|
|
"num_tokens": 7294228.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.472740888595581,
|
|
"epoch": 3.206269877328487,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004326429524281975,
|
|
"loss": 5.1788,
|
|
"mean_token_accuracy": 0.1800582841038704,
|
|
"num_tokens": 7304555.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.547172164916992,
|
|
"epoch": 3.210813266696956,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004323905794312303,
|
|
"loss": 5.2954,
|
|
"mean_token_accuracy": 0.1646373063325882,
|
|
"num_tokens": 7315955.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.535283470153809,
|
|
"epoch": 3.215356656065425,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00043213781808865414,
|
|
"loss": 5.2393,
|
|
"mean_token_accuracy": 0.17261512726545333,
|
|
"num_tokens": 7326757.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.473321342468262,
|
|
"epoch": 3.2199000454338935,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004318846690241324,
|
|
"loss": 5.2012,
|
|
"mean_token_accuracy": 0.1816797137260437,
|
|
"num_tokens": 7335969.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.397165012359619,
|
|
"epoch": 3.2244434348023625,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00043163113286228533,
|
|
"loss": 5.2153,
|
|
"mean_token_accuracy": 0.17270189821720122,
|
|
"num_tokens": 7345824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.573737287521363,
|
|
"epoch": 3.2289868241708315,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004313772102286882,
|
|
"loss": 5.2789,
|
|
"mean_token_accuracy": 0.16862437278032302,
|
|
"num_tokens": 7356292.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.508597660064697,
|
|
"epoch": 3.2335302135393005,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004311229017498698,
|
|
"loss": 5.2618,
|
|
"mean_token_accuracy": 0.17158734053373337,
|
|
"num_tokens": 7366809.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.5379533767700195,
|
|
"epoch": 3.238073602907769,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000430868208053311,
|
|
"loss": 5.1697,
|
|
"mean_token_accuracy": 0.18673526048660277,
|
|
"num_tokens": 7376076.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.500352191925049,
|
|
"epoch": 3.242616992276238,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00043061312976744326,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.17213442623615266,
|
|
"num_tokens": 7386338.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.469956588745117,
|
|
"epoch": 3.247160381644707,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004303576675216469,
|
|
"loss": 5.1772,
|
|
"mean_token_accuracy": 0.1779381349682808,
|
|
"num_tokens": 7395792.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.526382780075073,
|
|
"epoch": 3.251703771013176,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00043010182194624957,
|
|
"loss": 5.1889,
|
|
"mean_token_accuracy": 0.17896528244018556,
|
|
"num_tokens": 7405765.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.416803455352783,
|
|
"epoch": 3.2562471603816445,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004298455936725248,
|
|
"loss": 5.265,
|
|
"mean_token_accuracy": 0.1754458099603653,
|
|
"num_tokens": 7415213.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.521268463134765,
|
|
"epoch": 3.2607905497501135,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004295889833326903,
|
|
"loss": 5.2362,
|
|
"mean_token_accuracy": 0.1713758572936058,
|
|
"num_tokens": 7425722.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.59187068939209,
|
|
"epoch": 3.2653339391185825,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00042933199155990675,
|
|
"loss": 5.2926,
|
|
"mean_token_accuracy": 0.17655708491802216,
|
|
"num_tokens": 7436240.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.514054679870606,
|
|
"epoch": 3.2698773284870515,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0004290746189882757,
|
|
"loss": 5.2557,
|
|
"mean_token_accuracy": 0.16693031489849092,
|
|
"num_tokens": 7447878.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.561192512512207,
|
|
"epoch": 3.27442071785552,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00042881686625283844,
|
|
"loss": 5.1997,
|
|
"mean_token_accuracy": 0.18229420483112335,
|
|
"num_tokens": 7458635.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.5268950939178465,
|
|
"epoch": 3.278964107223989,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004285587339895742,
|
|
"loss": 5.2451,
|
|
"mean_token_accuracy": 0.17078938633203505,
|
|
"num_tokens": 7468740.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.559709692001343,
|
|
"epoch": 3.283507496592458,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00042830022283539863,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.17038103491067885,
|
|
"num_tokens": 7480977.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.507879018783569,
|
|
"epoch": 3.288050885960927,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004280413334281623,
|
|
"loss": 5.3131,
|
|
"mean_token_accuracy": 0.17209591269493102,
|
|
"num_tokens": 7491758.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.555095243453979,
|
|
"epoch": 3.292594275329396,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00042778206640664915,
|
|
"loss": 5.2325,
|
|
"mean_token_accuracy": 0.17546641379594802,
|
|
"num_tokens": 7500486.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.450375604629516,
|
|
"epoch": 3.2971376646978645,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004275224224105748,
|
|
"loss": 5.2366,
|
|
"mean_token_accuracy": 0.18374965339899063,
|
|
"num_tokens": 7510371.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.502903175354004,
|
|
"epoch": 3.3016810540663335,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00042726240208058484,
|
|
"loss": 5.2506,
|
|
"mean_token_accuracy": 0.181112340092659,
|
|
"num_tokens": 7520697.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.495673036575317,
|
|
"epoch": 3.3062244434348025,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004270020060582538,
|
|
"loss": 5.1874,
|
|
"mean_token_accuracy": 0.1753991052508354,
|
|
"num_tokens": 7530757.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.545063781738281,
|
|
"epoch": 3.310767832803271,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00042674123498608285,
|
|
"loss": 5.2293,
|
|
"mean_token_accuracy": 0.17249100506305695,
|
|
"num_tokens": 7541810.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.600866889953613,
|
|
"epoch": 3.31531122217174,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00042648008950749877,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.16002558320760726,
|
|
"num_tokens": 7552754.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.5452343940734865,
|
|
"epoch": 3.319854611540209,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000426218570266852,
|
|
"loss": 5.1885,
|
|
"mean_token_accuracy": 0.17581411600112914,
|
|
"num_tokens": 7562906.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.412493753433227,
|
|
"epoch": 3.324398000908678,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042595667790941536,
|
|
"loss": 5.207,
|
|
"mean_token_accuracy": 0.17695267945528032,
|
|
"num_tokens": 7574149.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.484325265884399,
|
|
"epoch": 3.328941390277147,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042569441308138204,
|
|
"loss": 5.2028,
|
|
"mean_token_accuracy": 0.18317578434944154,
|
|
"num_tokens": 7584610.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.4527918815612795,
|
|
"epoch": 3.3334847796456155,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004254317764298645,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.17830838561058043,
|
|
"num_tokens": 7594954.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.5265521049499515,
|
|
"epoch": 3.3380281690140845,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004251687686028927,
|
|
"loss": 5.302,
|
|
"mean_token_accuracy": 0.16644491553306578,
|
|
"num_tokens": 7606069.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.457864093780517,
|
|
"epoch": 3.3425715583825535,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00042490539024941216,
|
|
"loss": 5.1405,
|
|
"mean_token_accuracy": 0.18162267357110978,
|
|
"num_tokens": 7615671.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.482856416702271,
|
|
"epoch": 3.347114947751022,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00042464164201928286,
|
|
"loss": 5.3101,
|
|
"mean_token_accuracy": 0.16735140979290009,
|
|
"num_tokens": 7626740.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.47478551864624,
|
|
"epoch": 3.351658337119491,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00042437752456327734,
|
|
"loss": 5.248,
|
|
"mean_token_accuracy": 0.17754929959774018,
|
|
"num_tokens": 7636459.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.595248031616211,
|
|
"epoch": 3.35620172648796,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00042411303853307907,
|
|
"loss": 5.1965,
|
|
"mean_token_accuracy": 0.17650400400161742,
|
|
"num_tokens": 7646974.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.451615333557129,
|
|
"epoch": 3.360745115856429,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004238481845812811,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.17921001315116883,
|
|
"num_tokens": 7657372.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.417365789413452,
|
|
"epoch": 3.365288505224898,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00042358296336138436,
|
|
"loss": 5.2285,
|
|
"mean_token_accuracy": 0.17238256335258484,
|
|
"num_tokens": 7668972.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.541148567199707,
|
|
"epoch": 3.3698318945933665,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0004233173755277956,
|
|
"loss": 5.2213,
|
|
"mean_token_accuracy": 0.19162176847457885,
|
|
"num_tokens": 7679742.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.485103034973145,
|
|
"epoch": 3.3743752839618355,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00042305142173582657,
|
|
"loss": 5.2533,
|
|
"mean_token_accuracy": 0.17657138854265214,
|
|
"num_tokens": 7690759.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.451731777191162,
|
|
"epoch": 3.3789186733303045,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042278510264169165,
|
|
"loss": 5.2203,
|
|
"mean_token_accuracy": 0.17793505787849426,
|
|
"num_tokens": 7700482.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.4929040431976315,
|
|
"epoch": 3.3834620626987735,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00042251841890250696,
|
|
"loss": 5.1898,
|
|
"mean_token_accuracy": 0.1733878269791603,
|
|
"num_tokens": 7710414.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.457860040664673,
|
|
"epoch": 3.388005452067242,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000422251371176288,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.17793077528476714,
|
|
"num_tokens": 7721214.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.431334590911865,
|
|
"epoch": 3.392548841435711,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00042198396012194854,
|
|
"loss": 5.203,
|
|
"mean_token_accuracy": 0.17829247564077377,
|
|
"num_tokens": 7732551.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.578602695465088,
|
|
"epoch": 3.39709223080418,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00042171618639929865,
|
|
"loss": 5.2921,
|
|
"mean_token_accuracy": 0.1765536904335022,
|
|
"num_tokens": 7742266.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.412631940841675,
|
|
"epoch": 3.401635620172649,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00042144805066904345,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.17770766466856003,
|
|
"num_tokens": 7753152.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.4080945491790775,
|
|
"epoch": 3.4061790095411175,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042117955359278125,
|
|
"loss": 5.2409,
|
|
"mean_token_accuracy": 0.17369688600301741,
|
|
"num_tokens": 7764420.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.519856739044189,
|
|
"epoch": 3.4107223989095865,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042091069583300177,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.17681717574596406,
|
|
"num_tokens": 7776044.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.438783264160156,
|
|
"epoch": 3.4152657882780555,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.0004206414780530849,
|
|
"loss": 5.2443,
|
|
"mean_token_accuracy": 0.17244109362363816,
|
|
"num_tokens": 7786737.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.413301563262939,
|
|
"epoch": 3.4198091776465245,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0004203719009172987,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.17546848803758622,
|
|
"num_tokens": 7796974.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.564588212966919,
|
|
"epoch": 3.424352567014993,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000420101965090798,
|
|
"loss": 5.3258,
|
|
"mean_token_accuracy": 0.16958121657371522,
|
|
"num_tokens": 7808221.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.443160724639893,
|
|
"epoch": 3.428895956383462,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004198316712396226,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.1806100457906723,
|
|
"num_tokens": 7818422.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.461479949951172,
|
|
"epoch": 3.433439345751931,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00041956102003069573,
|
|
"loss": 5.1762,
|
|
"mean_token_accuracy": 0.1816395878791809,
|
|
"num_tokens": 7828313.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.465687417984009,
|
|
"epoch": 3.4379827351204,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004192900121318223,
|
|
"loss": 5.214,
|
|
"mean_token_accuracy": 0.18616935908794402,
|
|
"num_tokens": 7839493.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.493979549407959,
|
|
"epoch": 3.4425261244888685,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.0004190186482116874,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.1765694797039032,
|
|
"num_tokens": 7849796.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.477693510055542,
|
|
"epoch": 3.4470695138573375,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004187469289398545,
|
|
"loss": 5.2827,
|
|
"mean_token_accuracy": 0.17442628592252732,
|
|
"num_tokens": 7860865.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.438766431808472,
|
|
"epoch": 3.4516129032258065,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00041847485498676385,
|
|
"loss": 5.2097,
|
|
"mean_token_accuracy": 0.18299593925476074,
|
|
"num_tokens": 7871144.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.5044474601745605,
|
|
"epoch": 3.4561562925942755,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00041820242702373094,
|
|
"loss": 5.2522,
|
|
"mean_token_accuracy": 0.17985574156045914,
|
|
"num_tokens": 7880916.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.427869892120361,
|
|
"epoch": 3.460699681962744,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004179296457229448,
|
|
"loss": 5.1828,
|
|
"mean_token_accuracy": 0.17983298748731613,
|
|
"num_tokens": 7891813.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.468002080917358,
|
|
"epoch": 3.465243071331213,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004176565117574659,
|
|
"loss": 5.2435,
|
|
"mean_token_accuracy": 0.17421369403600692,
|
|
"num_tokens": 7902697.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.515365171432495,
|
|
"epoch": 3.469786460699682,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004173830258012255,
|
|
"loss": 5.2172,
|
|
"mean_token_accuracy": 0.17739980965852736,
|
|
"num_tokens": 7912777.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.456302642822266,
|
|
"epoch": 3.474329850068151,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004171091885290226,
|
|
"loss": 5.2647,
|
|
"mean_token_accuracy": 0.17829198390245438,
|
|
"num_tokens": 7923255.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.48848147392273,
|
|
"epoch": 3.4788732394366195,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004168350006165238,
|
|
"loss": 5.1487,
|
|
"mean_token_accuracy": 0.1818780317902565,
|
|
"num_tokens": 7934268.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.420705509185791,
|
|
"epoch": 3.4834166288050885,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00041656046274026034,
|
|
"loss": 5.1668,
|
|
"mean_token_accuracy": 0.18092650771141053,
|
|
"num_tokens": 7944526.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.420863771438599,
|
|
"epoch": 3.4879600181735575,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004162855755776272,
|
|
"loss": 5.2612,
|
|
"mean_token_accuracy": 0.17779920548200606,
|
|
"num_tokens": 7954512.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.49200177192688,
|
|
"epoch": 3.4925034075420265,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00041601033980688106,
|
|
"loss": 5.2774,
|
|
"mean_token_accuracy": 0.1649803787469864,
|
|
"num_tokens": 7965703.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.5496741771698,
|
|
"epoch": 3.497046796910495,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004157347561071389,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.17156354188919068,
|
|
"num_tokens": 7977822.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.430872964859009,
|
|
"epoch": 3.501590186278964,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00041545882515837604,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.17870686799287797,
|
|
"num_tokens": 7988318.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.441584682464599,
|
|
"epoch": 3.506133575647433,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00041518254764142477,
|
|
"loss": 5.1498,
|
|
"mean_token_accuracy": 0.1806313157081604,
|
|
"num_tokens": 7998076.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.4998860359191895,
|
|
"epoch": 3.510676965015902,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00041490592423797226,
|
|
"loss": 5.2511,
|
|
"mean_token_accuracy": 0.17413922846317292,
|
|
"num_tokens": 8007913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.4388714790344235,
|
|
"epoch": 3.515220354384371,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004146289556305594,
|
|
"loss": 5.2183,
|
|
"mean_token_accuracy": 0.1709863767027855,
|
|
"num_tokens": 8018340.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.543228721618652,
|
|
"epoch": 3.5197637437528395,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004143516425025786,
|
|
"loss": 5.2599,
|
|
"mean_token_accuracy": 0.17260538786649704,
|
|
"num_tokens": 8028312.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.55478343963623,
|
|
"epoch": 3.5243071331213085,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004140739855382724,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.1744476556777954,
|
|
"num_tokens": 8038580.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.47581205368042,
|
|
"epoch": 3.5288505224897775,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00041379598542273197,
|
|
"loss": 5.2076,
|
|
"mean_token_accuracy": 0.17815140783786773,
|
|
"num_tokens": 8048365.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.484137010574341,
|
|
"epoch": 3.533393911858246,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00041351764284189464,
|
|
"loss": 5.2733,
|
|
"mean_token_accuracy": 0.17120714485645294,
|
|
"num_tokens": 8059638.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.51491265296936,
|
|
"epoch": 3.537937301226715,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004132389584825433,
|
|
"loss": 5.3429,
|
|
"mean_token_accuracy": 0.16780453324317932,
|
|
"num_tokens": 8070993.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.429760265350342,
|
|
"epoch": 3.542480690595184,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00041295993303230383,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.1754154935479164,
|
|
"num_tokens": 8081203.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.419435739517212,
|
|
"epoch": 3.547024079963653,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004126805671796438,
|
|
"loss": 5.1974,
|
|
"mean_token_accuracy": 0.17271972000598906,
|
|
"num_tokens": 8091564.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.484901094436646,
|
|
"epoch": 3.551567469332122,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004124008616138707,
|
|
"loss": 5.1668,
|
|
"mean_token_accuracy": 0.17794379591941833,
|
|
"num_tokens": 8100675.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.497135400772095,
|
|
"epoch": 3.5561108587005905,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00041212081702513015,
|
|
"loss": 5.2292,
|
|
"mean_token_accuracy": 0.18242530077695845,
|
|
"num_tokens": 8110605.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.480234670639038,
|
|
"epoch": 3.5606542480690595,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004118404341044044,
|
|
"loss": 5.2689,
|
|
"mean_token_accuracy": 0.17364573180675508,
|
|
"num_tokens": 8120910.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.4553961753845215,
|
|
"epoch": 3.5651976374375285,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00041155971354351034,
|
|
"loss": 5.2264,
|
|
"mean_token_accuracy": 0.17740676999092103,
|
|
"num_tokens": 8131827.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.483454847335816,
|
|
"epoch": 3.569741026805997,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00041127865603509815,
|
|
"loss": 5.1909,
|
|
"mean_token_accuracy": 0.17578980177640915,
|
|
"num_tokens": 8141785.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.489233016967773,
|
|
"epoch": 3.574284416174466,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004109972622726493,
|
|
"loss": 5.2144,
|
|
"mean_token_accuracy": 0.1773121565580368,
|
|
"num_tokens": 8151732.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.51270694732666,
|
|
"epoch": 3.578827805542935,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 0.00041071553295047485,
|
|
"loss": 5.2996,
|
|
"mean_token_accuracy": 0.17181696891784667,
|
|
"num_tokens": 8162334.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.486226224899292,
|
|
"epoch": 3.583371194911404,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004104334687637139,
|
|
"loss": 5.1821,
|
|
"mean_token_accuracy": 0.17809890657663346,
|
|
"num_tokens": 8172810.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.438502073287964,
|
|
"epoch": 3.587914584279873,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00041015107040833187,
|
|
"loss": 5.2417,
|
|
"mean_token_accuracy": 0.17247770428657533,
|
|
"num_tokens": 8183240.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.441746044158935,
|
|
"epoch": 3.5924579736483415,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0004098683385811185,
|
|
"loss": 5.1776,
|
|
"mean_token_accuracy": 0.1753559708595276,
|
|
"num_tokens": 8193824.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.453190898895263,
|
|
"epoch": 3.5970013630168105,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00040958527397968666,
|
|
"loss": 5.1617,
|
|
"mean_token_accuracy": 0.18426349461078645,
|
|
"num_tokens": 8203792.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.521420812606811,
|
|
"epoch": 3.6015447523852795,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 0.0004093018773024699,
|
|
"loss": 5.3447,
|
|
"mean_token_accuracy": 0.1704270750284195,
|
|
"num_tokens": 8214436.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.492675733566284,
|
|
"epoch": 3.606088141753748,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004090181492487215,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.17590948790311814,
|
|
"num_tokens": 8224518.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.508065032958984,
|
|
"epoch": 3.610631531122217,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00040873409051851223,
|
|
"loss": 5.3343,
|
|
"mean_token_accuracy": 0.17109122574329377,
|
|
"num_tokens": 8235234.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.454956531524658,
|
|
"epoch": 3.615174920490686,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004084497018127286,
|
|
"loss": 5.2413,
|
|
"mean_token_accuracy": 0.17276171147823333,
|
|
"num_tokens": 8246612.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.532601261138916,
|
|
"epoch": 3.619718309859155,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004081649838330717,
|
|
"loss": 5.188,
|
|
"mean_token_accuracy": 0.17954647839069365,
|
|
"num_tokens": 8256720.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.435756731033325,
|
|
"epoch": 3.624261699227624,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00040787993728205464,
|
|
"loss": 5.2505,
|
|
"mean_token_accuracy": 0.17694082707166672,
|
|
"num_tokens": 8267666.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.512888956069946,
|
|
"epoch": 3.6288050885960925,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00040759456286300154,
|
|
"loss": 5.223,
|
|
"mean_token_accuracy": 0.17581545263528825,
|
|
"num_tokens": 8277882.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.506636905670166,
|
|
"epoch": 3.6333484779645615,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00040730886128004544,
|
|
"loss": 5.161,
|
|
"mean_token_accuracy": 0.17869097590446473,
|
|
"num_tokens": 8286968.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.6333484779645615,
|
|
"eval_entropy": 5.416537083026975,
|
|
"eval_loss": 5.8548760414123535,
|
|
"eval_mean_token_accuracy": 0.1578681156039238,
|
|
"eval_num_tokens": 8286968.0,
|
|
"eval_runtime": 2.5759,
|
|
"eval_samples_per_second": 1333.537,
|
|
"eval_steps_per_second": 166.935,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.463347959518432,
|
|
"epoch": 3.6378918673330305,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004070228332381266,
|
|
"loss": 5.2245,
|
|
"mean_token_accuracy": 0.17888124883174897,
|
|
"num_tokens": 8297154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.4683431625366214,
|
|
"epoch": 3.6424352567014995,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004067364794429908,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.1753451108932495,
|
|
"num_tokens": 8307954.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.426838445663452,
|
|
"epoch": 3.646978646069968,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004064498006011874,
|
|
"loss": 5.2,
|
|
"mean_token_accuracy": 0.17675285041332245,
|
|
"num_tokens": 8317714.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.370013093948364,
|
|
"epoch": 3.651522035438437,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00040616279742006813,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.18006695806980133,
|
|
"num_tokens": 8327963.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.426497983932495,
|
|
"epoch": 3.656065424806906,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00040587547060778466,
|
|
"loss": 5.1257,
|
|
"mean_token_accuracy": 0.18595771491527557,
|
|
"num_tokens": 8338190.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.469983434677124,
|
|
"epoch": 3.660608814175375,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004055878208732874,
|
|
"loss": 5.2848,
|
|
"mean_token_accuracy": 0.17206220477819442,
|
|
"num_tokens": 8348564.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.470478677749634,
|
|
"epoch": 3.665152203543844,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00040529984892632363,
|
|
"loss": 5.1898,
|
|
"mean_token_accuracy": 0.1818190649151802,
|
|
"num_tokens": 8359463.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.440707111358643,
|
|
"epoch": 3.6696955929123125,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 0.0004050115554774353,
|
|
"loss": 5.2505,
|
|
"mean_token_accuracy": 0.1725202575325966,
|
|
"num_tokens": 8369676.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.449108982086182,
|
|
"epoch": 3.6742389822807815,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0004047229412379578,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.19099992364645005,
|
|
"num_tokens": 8379579.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.506185054779053,
|
|
"epoch": 3.6787823716492505,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0004044340069200183,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.17859074622392654,
|
|
"num_tokens": 8390164.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.3734677791595455,
|
|
"epoch": 3.683325761017719,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00040414475323653326,
|
|
"loss": 5.1253,
|
|
"mean_token_accuracy": 0.18645142316818236,
|
|
"num_tokens": 8399220.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.436134576797485,
|
|
"epoch": 3.687869150386188,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004038551809012076,
|
|
"loss": 5.2089,
|
|
"mean_token_accuracy": 0.17811324447393417,
|
|
"num_tokens": 8409488.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.4844177722930905,
|
|
"epoch": 3.692412539754657,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004035652906285322,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.18525615632534026,
|
|
"num_tokens": 8418817.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.467234992980957,
|
|
"epoch": 3.696955929123126,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00040327508313378244,
|
|
"loss": 5.2587,
|
|
"mean_token_accuracy": 0.1782330244779587,
|
|
"num_tokens": 8428742.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.441283941268921,
|
|
"epoch": 3.701499318491595,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 0.0004029845591330164,
|
|
"loss": 5.143,
|
|
"mean_token_accuracy": 0.18853491693735122,
|
|
"num_tokens": 8438016.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.456124401092529,
|
|
"epoch": 3.7060427078600635,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040269371934307336,
|
|
"loss": 5.2791,
|
|
"mean_token_accuracy": 0.17426258623600005,
|
|
"num_tokens": 8448428.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.438787364959717,
|
|
"epoch": 3.7105860972285325,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00040240256448157146,
|
|
"loss": 5.095,
|
|
"mean_token_accuracy": 0.18299925625324248,
|
|
"num_tokens": 8458444.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.428674221038818,
|
|
"epoch": 3.7151294865970015,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004021110952669065,
|
|
"loss": 5.2376,
|
|
"mean_token_accuracy": 0.1792539656162262,
|
|
"num_tokens": 8467721.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.4132627010345455,
|
|
"epoch": 3.71967287596547,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004018193124182497,
|
|
"loss": 5.1713,
|
|
"mean_token_accuracy": 0.18348006010055543,
|
|
"num_tokens": 8477484.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.473674535751343,
|
|
"epoch": 3.724216265333939,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004015272166555462,
|
|
"loss": 5.177,
|
|
"mean_token_accuracy": 0.1792135238647461,
|
|
"num_tokens": 8486902.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.44284987449646,
|
|
"epoch": 3.728759654702408,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004012348086995132,
|
|
"loss": 5.149,
|
|
"mean_token_accuracy": 0.17300424724817276,
|
|
"num_tokens": 8497626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.4643172264099125,
|
|
"epoch": 3.733303044070877,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0004009420892716385,
|
|
"loss": 5.257,
|
|
"mean_token_accuracy": 0.16694689989089967,
|
|
"num_tokens": 8507518.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.4655187129974365,
|
|
"epoch": 3.737846433439346,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0004006490590941781,
|
|
"loss": 5.2167,
|
|
"mean_token_accuracy": 0.1741604745388031,
|
|
"num_tokens": 8517835.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.498819828033447,
|
|
"epoch": 3.7423898228078145,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0004003557188901547,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.17425093054771423,
|
|
"num_tokens": 8527552.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.488222217559814,
|
|
"epoch": 3.7469332121762835,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004000620693833562,
|
|
"loss": 5.2697,
|
|
"mean_token_accuracy": 0.18052448481321334,
|
|
"num_tokens": 8538082.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.446111440658569,
|
|
"epoch": 3.7514766015447525,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003997681112983335,
|
|
"loss": 5.178,
|
|
"mean_token_accuracy": 0.1773810312151909,
|
|
"num_tokens": 8548558.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.3694253921508786,
|
|
"epoch": 3.756019990913221,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00039947384536039905,
|
|
"loss": 5.1927,
|
|
"mean_token_accuracy": 0.1787725195288658,
|
|
"num_tokens": 8559406.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.511724472045898,
|
|
"epoch": 3.76056338028169,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00039917927229562474,
|
|
"loss": 5.24,
|
|
"mean_token_accuracy": 0.1728445753455162,
|
|
"num_tokens": 8569519.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.5260356903076175,
|
|
"epoch": 3.765106769650159,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00039888439283084035,
|
|
"loss": 5.2525,
|
|
"mean_token_accuracy": 0.17743061780929564,
|
|
"num_tokens": 8579658.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.398521089553833,
|
|
"epoch": 3.769650159018628,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.00039858920769363164,
|
|
"loss": 5.1538,
|
|
"mean_token_accuracy": 0.1841245785355568,
|
|
"num_tokens": 8589681.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.4935249328613285,
|
|
"epoch": 3.774193548387097,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003982937176123386,
|
|
"loss": 5.2365,
|
|
"mean_token_accuracy": 0.17604116648435592,
|
|
"num_tokens": 8600310.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.445260000228882,
|
|
"epoch": 3.7787369377555655,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00039799792331605364,
|
|
"loss": 5.2622,
|
|
"mean_token_accuracy": 0.172903873026371,
|
|
"num_tokens": 8610920.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.436249113082885,
|
|
"epoch": 3.7832803271240345,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00039770182553461965,
|
|
"loss": 5.161,
|
|
"mean_token_accuracy": 0.18358988761901857,
|
|
"num_tokens": 8620536.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.446696519851685,
|
|
"epoch": 3.7878237164925035,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.00039740542499862866,
|
|
"loss": 5.2766,
|
|
"mean_token_accuracy": 0.17727812379598618,
|
|
"num_tokens": 8631647.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.4316620349884035,
|
|
"epoch": 3.792367105860972,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003971087224394194,
|
|
"loss": 5.247,
|
|
"mean_token_accuracy": 0.17368111610412598,
|
|
"num_tokens": 8642659.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.461775779724121,
|
|
"epoch": 3.796910495229441,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0003968117185890761,
|
|
"loss": 5.2201,
|
|
"mean_token_accuracy": 0.1846265584230423,
|
|
"num_tokens": 8651775.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.4933911800384525,
|
|
"epoch": 3.80145388459791,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00039651441418042606,
|
|
"loss": 5.2075,
|
|
"mean_token_accuracy": 0.17223869860172272,
|
|
"num_tokens": 8662520.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.41958441734314,
|
|
"epoch": 3.805997273966379,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003962168099470384,
|
|
"loss": 5.2166,
|
|
"mean_token_accuracy": 0.17913652211427689,
|
|
"num_tokens": 8672630.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.496546125411987,
|
|
"epoch": 3.810540663334848,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000395918906623222,
|
|
"loss": 5.2996,
|
|
"mean_token_accuracy": 0.1642081305384636,
|
|
"num_tokens": 8683348.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.475314283370972,
|
|
"epoch": 3.8150840527033165,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0003956207049440238,
|
|
"loss": 5.1582,
|
|
"mean_token_accuracy": 0.17829646319150924,
|
|
"num_tokens": 8693423.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.479533529281616,
|
|
"epoch": 3.8196274420717855,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.0003953222056452265,
|
|
"loss": 5.2203,
|
|
"mean_token_accuracy": 0.17824296206235885,
|
|
"num_tokens": 8703766.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.521615695953369,
|
|
"epoch": 3.8241708314402545,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003950234094633479,
|
|
"loss": 5.3254,
|
|
"mean_token_accuracy": 0.17608374506235122,
|
|
"num_tokens": 8715330.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.441294240951538,
|
|
"epoch": 3.828714220808723,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00039472431713563746,
|
|
"loss": 5.2167,
|
|
"mean_token_accuracy": 0.18135753273963928,
|
|
"num_tokens": 8725732.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.513021421432495,
|
|
"epoch": 3.833257610177192,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000394424929400076,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.18354632407426835,
|
|
"num_tokens": 8736377.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.549785947799682,
|
|
"epoch": 3.837800999545661,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003941252469953729,
|
|
"loss": 5.2094,
|
|
"mean_token_accuracy": 0.17763325572013855,
|
|
"num_tokens": 8746991.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.427412843704223,
|
|
"epoch": 3.84234438891413,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0003938252706609647,
|
|
"loss": 5.2197,
|
|
"mean_token_accuracy": 0.17716552168130875,
|
|
"num_tokens": 8756920.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.403647565841675,
|
|
"epoch": 3.846887778282599,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003935250011370132,
|
|
"loss": 5.2716,
|
|
"mean_token_accuracy": 0.1683622881770134,
|
|
"num_tokens": 8768617.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.466759777069091,
|
|
"epoch": 3.851431167651068,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003932244391644037,
|
|
"loss": 5.2192,
|
|
"mean_token_accuracy": 0.1872796192765236,
|
|
"num_tokens": 8778427.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.472915887832642,
|
|
"epoch": 3.8559745570195365,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0003929235854847428,
|
|
"loss": 5.2645,
|
|
"mean_token_accuracy": 0.17913682609796525,
|
|
"num_tokens": 8789093.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.441914319992065,
|
|
"epoch": 3.8605179463880055,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00039262244084035724,
|
|
"loss": 5.2063,
|
|
"mean_token_accuracy": 0.17416508942842485,
|
|
"num_tokens": 8799832.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.45185227394104,
|
|
"epoch": 3.8650613357564745,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00039232100597429143,
|
|
"loss": 5.2187,
|
|
"mean_token_accuracy": 0.1795914202928543,
|
|
"num_tokens": 8810149.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.504850816726685,
|
|
"epoch": 3.869604725124943,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00039201928163030584,
|
|
"loss": 5.2441,
|
|
"mean_token_accuracy": 0.18068795502185822,
|
|
"num_tokens": 8820087.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.480374765396118,
|
|
"epoch": 3.874148114493412,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00039171726855287544,
|
|
"loss": 5.2935,
|
|
"mean_token_accuracy": 0.16966182589530945,
|
|
"num_tokens": 8830959.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.503117465972901,
|
|
"epoch": 3.878691503861881,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00039141496748718735,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.1754981115460396,
|
|
"num_tokens": 8842522.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.43432354927063,
|
|
"epoch": 3.88323489323035,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003911123791791395,
|
|
"loss": 5.1287,
|
|
"mean_token_accuracy": 0.18792554289102553,
|
|
"num_tokens": 8852521.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.489029836654663,
|
|
"epoch": 3.887778282598819,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003908095043753384,
|
|
"loss": 5.2575,
|
|
"mean_token_accuracy": 0.17396518737077712,
|
|
"num_tokens": 8862156.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.490391111373901,
|
|
"epoch": 3.8923216719672875,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039050634382309753,
|
|
"loss": 5.1443,
|
|
"mean_token_accuracy": 0.1735390916466713,
|
|
"num_tokens": 8872772.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.503467607498169,
|
|
"epoch": 3.8968650613357565,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003902028982704355,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.18087769597768782,
|
|
"num_tokens": 8883229.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.386239862442016,
|
|
"epoch": 3.9014084507042255,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003898991684660738,
|
|
"loss": 5.2066,
|
|
"mean_token_accuracy": 0.18216447085142135,
|
|
"num_tokens": 8892403.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.390725231170654,
|
|
"epoch": 3.905951840072694,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003895951551594358,
|
|
"loss": 5.211,
|
|
"mean_token_accuracy": 0.17370553761720658,
|
|
"num_tokens": 8902617.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.524963998794556,
|
|
"epoch": 3.910495229441163,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.000389290859100644,
|
|
"loss": 5.2585,
|
|
"mean_token_accuracy": 0.1818625420331955,
|
|
"num_tokens": 8913099.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.424187517166137,
|
|
"epoch": 3.915038618809632,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003889862810405187,
|
|
"loss": 5.169,
|
|
"mean_token_accuracy": 0.1802969232201576,
|
|
"num_tokens": 8922469.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.473340177536011,
|
|
"epoch": 3.919582008178101,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000388681421730576,
|
|
"loss": 5.2547,
|
|
"mean_token_accuracy": 0.1793914958834648,
|
|
"num_tokens": 8933261.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.5170642852783205,
|
|
"epoch": 3.92412539754657,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003883762819230259,
|
|
"loss": 5.2537,
|
|
"mean_token_accuracy": 0.18159746676683425,
|
|
"num_tokens": 8943873.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.423811912536621,
|
|
"epoch": 3.9286687869150385,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00038807086237077057,
|
|
"loss": 5.1365,
|
|
"mean_token_accuracy": 0.1875864088535309,
|
|
"num_tokens": 8953802.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.50324854850769,
|
|
"epoch": 3.9332121762835075,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00038776516382740237,
|
|
"loss": 5.3203,
|
|
"mean_token_accuracy": 0.17752646654844284,
|
|
"num_tokens": 8964309.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.4383080959320065,
|
|
"epoch": 3.9377555656519765,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00038745918704720205,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.18131357580423355,
|
|
"num_tokens": 8973611.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.439136791229248,
|
|
"epoch": 3.942298955020445,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00038715293278513693,
|
|
"loss": 5.2378,
|
|
"mean_token_accuracy": 0.1740028142929077,
|
|
"num_tokens": 8985783.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.516194438934326,
|
|
"epoch": 3.946842344388914,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00038684640179685896,
|
|
"loss": 5.2773,
|
|
"mean_token_accuracy": 0.17066926509141922,
|
|
"num_tokens": 8996736.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.51853756904602,
|
|
"epoch": 3.951385733757383,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003865395948387029,
|
|
"loss": 5.3544,
|
|
"mean_token_accuracy": 0.16442483365535737,
|
|
"num_tokens": 9007635.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.462783432006836,
|
|
"epoch": 3.955929123125852,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00038623251266768437,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.1769654259085655,
|
|
"num_tokens": 9017892.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.425413417816162,
|
|
"epoch": 3.960472512494321,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00038592515604149815,
|
|
"loss": 5.2328,
|
|
"mean_token_accuracy": 0.1802668899297714,
|
|
"num_tokens": 9027594.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.44905571937561,
|
|
"epoch": 3.9650159018627895,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003856175257185162,
|
|
"loss": 5.2476,
|
|
"mean_token_accuracy": 0.17574162483215333,
|
|
"num_tokens": 9037083.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.47435736656189,
|
|
"epoch": 3.9695592912312585,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00038530962245778564,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.18591096848249436,
|
|
"num_tokens": 9047176.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.452231359481812,
|
|
"epoch": 3.9741026805997275,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003850014470190272,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.17814017981290817,
|
|
"num_tokens": 9057810.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.473272466659546,
|
|
"epoch": 3.978646069968196,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003846930001626332,
|
|
"loss": 5.279,
|
|
"mean_token_accuracy": 0.1733193039894104,
|
|
"num_tokens": 9067765.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.539614295959472,
|
|
"epoch": 3.983189459336665,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003843842826496656,
|
|
"loss": 5.2382,
|
|
"mean_token_accuracy": 0.17567098885774612,
|
|
"num_tokens": 9078796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.51374192237854,
|
|
"epoch": 3.987732848705134,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000384075295241854,
|
|
"loss": 5.1652,
|
|
"mean_token_accuracy": 0.1850729137659073,
|
|
"num_tokens": 9088368.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.388001155853272,
|
|
"epoch": 3.992276238073603,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003837660387015944,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.18201180100440978,
|
|
"num_tokens": 9099082.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.439921140670776,
|
|
"epoch": 3.996819627442072,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003834565137919463,
|
|
"loss": 5.2664,
|
|
"mean_token_accuracy": 0.17681502997875215,
|
|
"num_tokens": 9110543.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.4597129291958275,
|
|
"epoch": 4.000908677873694,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003831467212766318,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.17833825780285728,
|
|
"num_tokens": 9119347.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.405972909927368,
|
|
"epoch": 4.005452067242163,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003828366619200331,
|
|
"loss": 4.921,
|
|
"mean_token_accuracy": 0.19136818796396254,
|
|
"num_tokens": 9129154.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.401428365707398,
|
|
"epoch": 4.009995456610632,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003825263364871908,
|
|
"loss": 4.9492,
|
|
"mean_token_accuracy": 0.1984136536717415,
|
|
"num_tokens": 9139877.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.384312629699707,
|
|
"epoch": 4.0145388459791,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003822157457438022,
|
|
"loss": 5.0364,
|
|
"mean_token_accuracy": 0.18836853206157683,
|
|
"num_tokens": 9150508.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.412083768844605,
|
|
"epoch": 4.019082235347569,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00038190489045621893,
|
|
"loss": 4.9612,
|
|
"mean_token_accuracy": 0.192303566634655,
|
|
"num_tokens": 9160403.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.505183458328247,
|
|
"epoch": 4.023625624716038,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0003815937713914455,
|
|
"loss": 5.0197,
|
|
"mean_token_accuracy": 0.18638781160116197,
|
|
"num_tokens": 9171026.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.422731113433838,
|
|
"epoch": 4.028169014084507,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038128238931713743,
|
|
"loss": 4.9975,
|
|
"mean_token_accuracy": 0.1880093812942505,
|
|
"num_tokens": 9181252.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.531588077545166,
|
|
"epoch": 4.032712403452976,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00038097074500159883,
|
|
"loss": 5.0656,
|
|
"mean_token_accuracy": 0.1856105700135231,
|
|
"num_tokens": 9190803.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.469673490524292,
|
|
"epoch": 4.037255792821445,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00038065883921378123,
|
|
"loss": 5.041,
|
|
"mean_token_accuracy": 0.18504955172538756,
|
|
"num_tokens": 9201077.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.377871036529541,
|
|
"epoch": 4.041799182189914,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003803466727232811,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.18186455219984055,
|
|
"num_tokens": 9211239.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.473480987548828,
|
|
"epoch": 4.046342571558383,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003800342463003383,
|
|
"loss": 5.0262,
|
|
"mean_token_accuracy": 0.19146362841129302,
|
|
"num_tokens": 9222683.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.485651683807373,
|
|
"epoch": 4.050885960926851,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00037972156071583386,
|
|
"loss": 5.0009,
|
|
"mean_token_accuracy": 0.18599868863821029,
|
|
"num_tokens": 9232099.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.392268180847168,
|
|
"epoch": 4.05542935029532,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003794086167412884,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.1902068719267845,
|
|
"num_tokens": 9242660.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.380642080307007,
|
|
"epoch": 4.059972739663789,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0003790954151488601,
|
|
"loss": 5.0666,
|
|
"mean_token_accuracy": 0.1762397140264511,
|
|
"num_tokens": 9253981.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.4578444480896,
|
|
"epoch": 4.064516129032258,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0003787819567113427,
|
|
"loss": 5.0103,
|
|
"mean_token_accuracy": 0.18754592090845107,
|
|
"num_tokens": 9264262.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.386116170883179,
|
|
"epoch": 4.069059518400727,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0003784682422021638,
|
|
"loss": 5.0129,
|
|
"mean_token_accuracy": 0.18826246708631517,
|
|
"num_tokens": 9273562.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.410690402984619,
|
|
"epoch": 4.073602907769196,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003781542723953828,
|
|
"loss": 5.0279,
|
|
"mean_token_accuracy": 0.18637023419141768,
|
|
"num_tokens": 9283171.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.423978281021118,
|
|
"epoch": 4.078146297137665,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.00037784004806568885,
|
|
"loss": 5.0083,
|
|
"mean_token_accuracy": 0.19023032039403914,
|
|
"num_tokens": 9292624.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.505765295028686,
|
|
"epoch": 4.082689686506134,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.00037752556998839934,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.18407783806324005,
|
|
"num_tokens": 9303159.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.444028282165528,
|
|
"epoch": 4.087233075874602,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0003772108389394576,
|
|
"loss": 5.0259,
|
|
"mean_token_accuracy": 0.19229615926742555,
|
|
"num_tokens": 9314340.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.087233075874602,
|
|
"eval_entropy": 5.336979123049004,
|
|
"eval_loss": 5.831993579864502,
|
|
"eval_mean_token_accuracy": 0.15985156477883805,
|
|
"eval_num_tokens": 9314340.0,
|
|
"eval_runtime": 2.5806,
|
|
"eval_samples_per_second": 1331.087,
|
|
"eval_steps_per_second": 166.628,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.48622875213623,
|
|
"epoch": 4.091776465243071,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00037689585569543137,
|
|
"loss": 5.0985,
|
|
"mean_token_accuracy": 0.1843963623046875,
|
|
"num_tokens": 9325381.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.413348150253296,
|
|
"epoch": 4.09631985461154,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003765806210335103,
|
|
"loss": 5.0849,
|
|
"mean_token_accuracy": 0.17930045872926711,
|
|
"num_tokens": 9336639.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.41552243232727,
|
|
"epoch": 4.100863243980009,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00037626513573150473,
|
|
"loss": 4.9686,
|
|
"mean_token_accuracy": 0.19257283508777617,
|
|
"num_tokens": 9346794.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.390832948684692,
|
|
"epoch": 4.105406633348478,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00037594940056784315,
|
|
"loss": 5.0452,
|
|
"mean_token_accuracy": 0.19416338950395584,
|
|
"num_tokens": 9356326.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.4480572700500485,
|
|
"epoch": 4.109950022716947,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003756334163215708,
|
|
"loss": 5.0219,
|
|
"mean_token_accuracy": 0.19045811891555786,
|
|
"num_tokens": 9366402.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.430761623382568,
|
|
"epoch": 4.114493412085416,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00037531718377234723,
|
|
"loss": 4.9997,
|
|
"mean_token_accuracy": 0.18785753101110458,
|
|
"num_tokens": 9377067.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.45697660446167,
|
|
"epoch": 4.119036801453885,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 0.0003750007037004451,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.1859073355793953,
|
|
"num_tokens": 9387206.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.396628093719483,
|
|
"epoch": 4.123580190822353,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003746839768867473,
|
|
"loss": 5.0141,
|
|
"mean_token_accuracy": 0.19374891221523285,
|
|
"num_tokens": 9397195.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.355070972442627,
|
|
"epoch": 4.128123580190822,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003743670041127458,
|
|
"loss": 4.9846,
|
|
"mean_token_accuracy": 0.1864893466234207,
|
|
"num_tokens": 9407611.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.46947979927063,
|
|
"epoch": 4.132666969559291,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003740497861605395,
|
|
"loss": 5.0655,
|
|
"mean_token_accuracy": 0.18512188643217087,
|
|
"num_tokens": 9417897.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.454879856109619,
|
|
"epoch": 4.13721035892776,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000373732323812832,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.17811979949474335,
|
|
"num_tokens": 9428896.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.395510625839234,
|
|
"epoch": 4.141753748296229,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00037341461785293025,
|
|
"loss": 4.9629,
|
|
"mean_token_accuracy": 0.19342249035835266,
|
|
"num_tokens": 9438854.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.334596061706543,
|
|
"epoch": 4.146297137664698,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 0.0003730966690647421,
|
|
"loss": 5.021,
|
|
"mean_token_accuracy": 0.18667362928390502,
|
|
"num_tokens": 9447798.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.37627649307251,
|
|
"epoch": 4.150840527033167,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00037277847823277444,
|
|
"loss": 4.9998,
|
|
"mean_token_accuracy": 0.19394131898880004,
|
|
"num_tokens": 9458930.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.3850785255432125,
|
|
"epoch": 4.155383916401636,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00037246004614213164,
|
|
"loss": 4.9588,
|
|
"mean_token_accuracy": 0.19461548775434495,
|
|
"num_tokens": 9469740.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.372608089447022,
|
|
"epoch": 4.159927305770104,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003721413735785132,
|
|
"loss": 4.9855,
|
|
"mean_token_accuracy": 0.19197811037302018,
|
|
"num_tokens": 9479823.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.398150539398193,
|
|
"epoch": 4.164470695138573,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000371822461328212,
|
|
"loss": 4.9567,
|
|
"mean_token_accuracy": 0.19695383608341216,
|
|
"num_tokens": 9489995.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.376239347457886,
|
|
"epoch": 4.169014084507042,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00037150331017811233,
|
|
"loss": 5.0512,
|
|
"mean_token_accuracy": 0.1880466490983963,
|
|
"num_tokens": 9501311.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.407603788375854,
|
|
"epoch": 4.173557473875511,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003711839209156878,
|
|
"loss": 5.0915,
|
|
"mean_token_accuracy": 0.18310328274965287,
|
|
"num_tokens": 9511901.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.463621711730957,
|
|
"epoch": 4.17810086324398,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003708642943289999,
|
|
"loss": 5.0486,
|
|
"mean_token_accuracy": 0.17985265403985978,
|
|
"num_tokens": 9522908.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.4789927959442135,
|
|
"epoch": 4.182644252612449,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003705444312066953,
|
|
"loss": 5.1162,
|
|
"mean_token_accuracy": 0.18482071310281753,
|
|
"num_tokens": 9533363.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.406308650970459,
|
|
"epoch": 4.187187641980918,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003702243323380045,
|
|
"loss": 4.9778,
|
|
"mean_token_accuracy": 0.191676589846611,
|
|
"num_tokens": 9543302.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.363091850280762,
|
|
"epoch": 4.191731031349387,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00036990399851273965,
|
|
"loss": 5.0285,
|
|
"mean_token_accuracy": 0.17811331748962403,
|
|
"num_tokens": 9553143.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.504148960113525,
|
|
"epoch": 4.196274420717856,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003695834305212925,
|
|
"loss": 5.0408,
|
|
"mean_token_accuracy": 0.18852516412734985,
|
|
"num_tokens": 9564236.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.466602325439453,
|
|
"epoch": 4.200817810086324,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 0.00036926262915463294,
|
|
"loss": 4.9886,
|
|
"mean_token_accuracy": 0.19565124660730362,
|
|
"num_tokens": 9574363.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.363614511489868,
|
|
"epoch": 4.205361199454793,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00036894159520430616,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.1840808853507042,
|
|
"num_tokens": 9585631.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.387624359130859,
|
|
"epoch": 4.209904588823262,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00036862032946243173,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.18139443993568422,
|
|
"num_tokens": 9596826.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.3544410228729244,
|
|
"epoch": 4.214447978191731,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036829883272170087,
|
|
"loss": 4.9748,
|
|
"mean_token_accuracy": 0.189482244849205,
|
|
"num_tokens": 9608158.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.425682830810547,
|
|
"epoch": 4.2189913675602,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00036797710577537486,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.1841538205742836,
|
|
"num_tokens": 9619437.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.435922527313233,
|
|
"epoch": 4.223534756928669,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00036765514941728295,
|
|
"loss": 5.0655,
|
|
"mean_token_accuracy": 0.19180352985858917,
|
|
"num_tokens": 9629913.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.374925231933593,
|
|
"epoch": 4.228078146297138,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003673329644418204,
|
|
"loss": 5.0184,
|
|
"mean_token_accuracy": 0.19069213569164276,
|
|
"num_tokens": 9640033.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.41509518623352,
|
|
"epoch": 4.232621535665607,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00036701055164394664,
|
|
"loss": 5.0763,
|
|
"mean_token_accuracy": 0.18921703100204468,
|
|
"num_tokens": 9651010.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.390908432006836,
|
|
"epoch": 4.237164925034075,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00036668791181918313,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.1904071420431137,
|
|
"num_tokens": 9661549.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.394793462753296,
|
|
"epoch": 4.241708314402544,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003663650457636117,
|
|
"loss": 5.0426,
|
|
"mean_token_accuracy": 0.1784576952457428,
|
|
"num_tokens": 9671619.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.4752177715301515,
|
|
"epoch": 4.246251703771013,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003660419542738722,
|
|
"loss": 5.0803,
|
|
"mean_token_accuracy": 0.1901138663291931,
|
|
"num_tokens": 9682582.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.337952327728272,
|
|
"epoch": 4.250795093139482,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003657186381471608,
|
|
"loss": 5.0075,
|
|
"mean_token_accuracy": 0.19203671663999558,
|
|
"num_tokens": 9691997.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.409979009628296,
|
|
"epoch": 4.255338482507951,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00036539509818122765,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.19195705354213716,
|
|
"num_tokens": 9702105.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.433075380325318,
|
|
"epoch": 4.25988187187642,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003650713351743758,
|
|
"loss": 5.0507,
|
|
"mean_token_accuracy": 0.18967791944742202,
|
|
"num_tokens": 9712347.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.401245641708374,
|
|
"epoch": 4.264425261244889,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003647473499254581,
|
|
"loss": 5.0519,
|
|
"mean_token_accuracy": 0.18831434100866318,
|
|
"num_tokens": 9723116.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.362901496887207,
|
|
"epoch": 4.268968650613358,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000364423143233876,
|
|
"loss": 4.9953,
|
|
"mean_token_accuracy": 0.18855696469545363,
|
|
"num_tokens": 9732656.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.4293131828308105,
|
|
"epoch": 4.273512039981826,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00036409871589957723,
|
|
"loss": 5.0967,
|
|
"mean_token_accuracy": 0.1841566503047943,
|
|
"num_tokens": 9743312.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.394158458709716,
|
|
"epoch": 4.278055429350295,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000363774068723054,
|
|
"loss": 5.0143,
|
|
"mean_token_accuracy": 0.1891046717762947,
|
|
"num_tokens": 9753358.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.378567123413086,
|
|
"epoch": 4.282598818718764,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.00036344920250534106,
|
|
"loss": 4.9472,
|
|
"mean_token_accuracy": 0.1971525013446808,
|
|
"num_tokens": 9763472.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.408444118499756,
|
|
"epoch": 4.287142208087233,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003631241180480133,
|
|
"loss": 4.9745,
|
|
"mean_token_accuracy": 0.18902794420719146,
|
|
"num_tokens": 9773700.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.37966685295105,
|
|
"epoch": 4.291685597455702,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003627988161531845,
|
|
"loss": 4.997,
|
|
"mean_token_accuracy": 0.191023550927639,
|
|
"num_tokens": 9784042.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.42055606842041,
|
|
"epoch": 4.296228986824171,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003624732976235046,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.18493931591510773,
|
|
"num_tokens": 9794464.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.4372021675109865,
|
|
"epoch": 4.30077237619264,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003621475632621584,
|
|
"loss": 5.0426,
|
|
"mean_token_accuracy": 0.18617953956127167,
|
|
"num_tokens": 9804459.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.405148220062256,
|
|
"epoch": 4.305315765561109,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003618216138728627,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.1883162662386894,
|
|
"num_tokens": 9814962.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.402033996582031,
|
|
"epoch": 4.309859154929577,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00036149545025986567,
|
|
"loss": 5.058,
|
|
"mean_token_accuracy": 0.18682121038436889,
|
|
"num_tokens": 9825184.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.440119886398316,
|
|
"epoch": 4.314402544298046,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00036116907322794335,
|
|
"loss": 5.1257,
|
|
"mean_token_accuracy": 0.1844857543706894,
|
|
"num_tokens": 9835184.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.410508108139038,
|
|
"epoch": 4.318945933666515,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00036084248358239857,
|
|
"loss": 5.0438,
|
|
"mean_token_accuracy": 0.18420273214578628,
|
|
"num_tokens": 9845358.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.415203952789307,
|
|
"epoch": 4.323489323034984,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00036051568212905897,
|
|
"loss": 4.9704,
|
|
"mean_token_accuracy": 0.19004779160022736,
|
|
"num_tokens": 9856519.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.4190620422363285,
|
|
"epoch": 4.328032712403453,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003601886696742745,
|
|
"loss": 5.0439,
|
|
"mean_token_accuracy": 0.18474730104207993,
|
|
"num_tokens": 9867478.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.363249158859253,
|
|
"epoch": 4.332576101771922,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.000359861447024916,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.18894974291324615,
|
|
"num_tokens": 9877709.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.404053163528443,
|
|
"epoch": 4.337119491140391,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00035953401498837285,
|
|
"loss": 4.9811,
|
|
"mean_token_accuracy": 0.18991907835006713,
|
|
"num_tokens": 9888134.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.3620765209198,
|
|
"epoch": 4.34166288050886,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035920637437255096,
|
|
"loss": 5.023,
|
|
"mean_token_accuracy": 0.19009646624326706,
|
|
"num_tokens": 9899150.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.412943172454834,
|
|
"epoch": 4.346206269877328,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000358878525985871,
|
|
"loss": 5.0518,
|
|
"mean_token_accuracy": 0.18982967734336853,
|
|
"num_tokens": 9909599.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.385846090316773,
|
|
"epoch": 4.350749659245797,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003585504706372663,
|
|
"loss": 4.9874,
|
|
"mean_token_accuracy": 0.1935717687010765,
|
|
"num_tokens": 9919607.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.409936904907227,
|
|
"epoch": 4.355293048614266,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003582222091361808,
|
|
"loss": 5.0548,
|
|
"mean_token_accuracy": 0.18502305895090104,
|
|
"num_tokens": 9930431.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.404749059677124,
|
|
"epoch": 4.359836437982735,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0003578937422925671,
|
|
"loss": 5.0228,
|
|
"mean_token_accuracy": 0.1898396745324135,
|
|
"num_tokens": 9940647.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.346047639846802,
|
|
"epoch": 4.364379827351204,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003575650709168846,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.18138931691646576,
|
|
"num_tokens": 9951110.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.41140308380127,
|
|
"epoch": 4.368923216719673,
|
|
"grad_norm": 2.953125,
|
|
"learning_rate": 0.00035723619582009715,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.18979807198047638,
|
|
"num_tokens": 9961158.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.408469247817993,
|
|
"epoch": 4.373466606088142,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00035690711781367137,
|
|
"loss": 5.0551,
|
|
"mean_token_accuracy": 0.18401383608579636,
|
|
"num_tokens": 9971051.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.328940105438233,
|
|
"epoch": 4.378009995456611,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0003565778377095744,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.1820178285241127,
|
|
"num_tokens": 9981990.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.3872167587280275,
|
|
"epoch": 4.38255338482508,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003562483563202724,
|
|
"loss": 4.975,
|
|
"mean_token_accuracy": 0.19290047585964204,
|
|
"num_tokens": 9991668.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.33818564414978,
|
|
"epoch": 4.387096774193548,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00035591867445872784,
|
|
"loss": 4.9135,
|
|
"mean_token_accuracy": 0.19649587869644164,
|
|
"num_tokens": 10001519.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.335766553878784,
|
|
"epoch": 4.391640163562017,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000355588792938398,
|
|
"loss": 4.9672,
|
|
"mean_token_accuracy": 0.1931387186050415,
|
|
"num_tokens": 10010856.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.329331541061402,
|
|
"epoch": 4.396183552930486,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.0003552587125732326,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.19200100600719452,
|
|
"num_tokens": 10020647.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.322201871871949,
|
|
"epoch": 4.400726942298955,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00035492843417767224,
|
|
"loss": 4.9177,
|
|
"mean_token_accuracy": 0.1978983238339424,
|
|
"num_tokens": 10030114.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.401615333557129,
|
|
"epoch": 4.405270331667424,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.000354597958566646,
|
|
"loss": 4.9923,
|
|
"mean_token_accuracy": 0.19287185668945311,
|
|
"num_tokens": 10039863.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.351267719268799,
|
|
"epoch": 4.409813721035893,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00035426728655556954,
|
|
"loss": 4.9712,
|
|
"mean_token_accuracy": 0.18719333559274673,
|
|
"num_tokens": 10049111.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.386077976226806,
|
|
"epoch": 4.414357110404362,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003539364189603433,
|
|
"loss": 5.0435,
|
|
"mean_token_accuracy": 0.19106572717428208,
|
|
"num_tokens": 10058762.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.431659460067749,
|
|
"epoch": 4.418900499772831,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 0.00035360535659735006,
|
|
"loss": 5.0443,
|
|
"mean_token_accuracy": 0.18679135739803315,
|
|
"num_tokens": 10068587.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.431622886657715,
|
|
"epoch": 4.423443889141299,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 0.0003532741002834534,
|
|
"loss": 5.0205,
|
|
"mean_token_accuracy": 0.18628098964691162,
|
|
"num_tokens": 10079386.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.329840612411499,
|
|
"epoch": 4.427987278509768,
|
|
"grad_norm": 2.78125,
|
|
"learning_rate": 0.00035294265083599526,
|
|
"loss": 4.9652,
|
|
"mean_token_accuracy": 0.19565689712762832,
|
|
"num_tokens": 10089490.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.333083772659302,
|
|
"epoch": 4.432530667878237,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0003526110090727942,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.18788003772497178,
|
|
"num_tokens": 10100465.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.400637531280518,
|
|
"epoch": 4.437074057246706,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.0003522791758121434,
|
|
"loss": 4.9928,
|
|
"mean_token_accuracy": 0.19708962738513947,
|
|
"num_tokens": 10110919.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.384515428543091,
|
|
"epoch": 4.441617446615175,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00035194715187280835,
|
|
"loss": 5.0322,
|
|
"mean_token_accuracy": 0.1874363452196121,
|
|
"num_tokens": 10121116.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.490482950210572,
|
|
"epoch": 4.446160835983644,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003516149380740252,
|
|
"loss": 5.145,
|
|
"mean_token_accuracy": 0.18648931086063386,
|
|
"num_tokens": 10132269.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.402000379562378,
|
|
"epoch": 4.450704225352113,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0003512825352354983,
|
|
"loss": 5.0015,
|
|
"mean_token_accuracy": 0.19632937163114547,
|
|
"num_tokens": 10142856.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.296314239501953,
|
|
"epoch": 4.455247614720582,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00035094994417739884,
|
|
"loss": 4.9945,
|
|
"mean_token_accuracy": 0.18683343678712844,
|
|
"num_tokens": 10152377.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.392163419723511,
|
|
"epoch": 4.45979100408905,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000350617165720362,
|
|
"loss": 5.0596,
|
|
"mean_token_accuracy": 0.1892980754375458,
|
|
"num_tokens": 10161961.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.457519626617431,
|
|
"epoch": 4.464334393457519,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00035028420068548567,
|
|
"loss": 5.0137,
|
|
"mean_token_accuracy": 0.1922242283821106,
|
|
"num_tokens": 10171624.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.505021286010742,
|
|
"epoch": 4.468877782825988,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003499510498943279,
|
|
"loss": 5.1066,
|
|
"mean_token_accuracy": 0.18466755151748657,
|
|
"num_tokens": 10182642.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.380368566513061,
|
|
"epoch": 4.473421172194457,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00034961771416890526,
|
|
"loss": 5.0736,
|
|
"mean_token_accuracy": 0.19002138525247575,
|
|
"num_tokens": 10193307.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.3974409103393555,
|
|
"epoch": 4.477964561562926,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0003492841943316903,
|
|
"loss": 5.0663,
|
|
"mean_token_accuracy": 0.18487306535243989,
|
|
"num_tokens": 10203506.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.372326803207398,
|
|
"epoch": 4.482507950931395,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003489504912056103,
|
|
"loss": 4.9622,
|
|
"mean_token_accuracy": 0.1915032833814621,
|
|
"num_tokens": 10214266.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.314327955245972,
|
|
"epoch": 4.487051340299864,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0003486166056140445,
|
|
"loss": 4.9512,
|
|
"mean_token_accuracy": 0.19726131856441498,
|
|
"num_tokens": 10224953.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.331363582611084,
|
|
"epoch": 4.491594729668333,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00034828253838082223,
|
|
"loss": 5.0089,
|
|
"mean_token_accuracy": 0.18707254379987717,
|
|
"num_tokens": 10235435.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.339622497558594,
|
|
"epoch": 4.496138119036801,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034794829033022134,
|
|
"loss": 5.0039,
|
|
"mean_token_accuracy": 0.19525942653417588,
|
|
"num_tokens": 10245677.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.293358850479126,
|
|
"epoch": 4.50068150840527,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003476138622869657,
|
|
"loss": 4.9793,
|
|
"mean_token_accuracy": 0.1920677050948143,
|
|
"num_tokens": 10256135.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.3947117805480955,
|
|
"epoch": 4.505224897773739,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0003472792550762232,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.18711859434843064,
|
|
"num_tokens": 10266465.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.462158918380737,
|
|
"epoch": 4.509768287142208,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003469444695236038,
|
|
"loss": 5.0731,
|
|
"mean_token_accuracy": 0.1863392025232315,
|
|
"num_tokens": 10278123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.417126226425171,
|
|
"epoch": 4.514311676510677,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00034660950645515765,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.1874504953622818,
|
|
"num_tokens": 10287486.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.422345781326294,
|
|
"epoch": 4.518855065879146,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003462743666973728,
|
|
"loss": 5.0245,
|
|
"mean_token_accuracy": 0.19026759415864944,
|
|
"num_tokens": 10298100.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.407691860198975,
|
|
"epoch": 4.523398455247615,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003459390510771734,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.1926530808210373,
|
|
"num_tokens": 10308549.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.387669372558594,
|
|
"epoch": 4.527941844616084,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003456035604219172,
|
|
"loss": 5.0394,
|
|
"mean_token_accuracy": 0.18225712478160858,
|
|
"num_tokens": 10319030.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.324537992477417,
|
|
"epoch": 4.532485233984552,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0003452678955593942,
|
|
"loss": 5.0835,
|
|
"mean_token_accuracy": 0.18861700892448424,
|
|
"num_tokens": 10329613.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.414364242553711,
|
|
"epoch": 4.537028623353021,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000344932057317824,
|
|
"loss": 5.074,
|
|
"mean_token_accuracy": 0.1876576602458954,
|
|
"num_tokens": 10339844.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.435067510604858,
|
|
"epoch": 4.54157201272149,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003445960465258542,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.1784687116742134,
|
|
"num_tokens": 10350906.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.54157201272149,
|
|
"eval_entropy": 5.31851795329604,
|
|
"eval_loss": 5.808477878570557,
|
|
"eval_mean_token_accuracy": 0.1610874231644841,
|
|
"eval_num_tokens": 10350906.0,
|
|
"eval_runtime": 2.7592,
|
|
"eval_samples_per_second": 1244.94,
|
|
"eval_steps_per_second": 155.844,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.343249273300171,
|
|
"epoch": 4.546115402089959,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003442598640125578,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.18497340083122255,
|
|
"num_tokens": 10361574.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.4460381984710695,
|
|
"epoch": 4.550658791458428,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000343923510607432,
|
|
"loss": 5.0737,
|
|
"mean_token_accuracy": 0.18416446894407273,
|
|
"num_tokens": 10372714.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.334274339675903,
|
|
"epoch": 4.555202180826897,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003435869871403952,
|
|
"loss": 4.9511,
|
|
"mean_token_accuracy": 0.19398237615823746,
|
|
"num_tokens": 10382143.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.385885381698609,
|
|
"epoch": 4.559745570195366,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00034325029444178565,
|
|
"loss": 5.0344,
|
|
"mean_token_accuracy": 0.18356402069330216,
|
|
"num_tokens": 10392040.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.3987328052520756,
|
|
"epoch": 4.564288959563835,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00034291343334235915,
|
|
"loss": 5.0739,
|
|
"mean_token_accuracy": 0.1771576374769211,
|
|
"num_tokens": 10403230.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.288701105117798,
|
|
"epoch": 4.568832348932304,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003425764046732869,
|
|
"loss": 4.9873,
|
|
"mean_token_accuracy": 0.18944236040115356,
|
|
"num_tokens": 10413466.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.411571455001831,
|
|
"epoch": 4.573375738300772,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00034223920926615376,
|
|
"loss": 4.9939,
|
|
"mean_token_accuracy": 0.19113776981830596,
|
|
"num_tokens": 10422559.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.385260009765625,
|
|
"epoch": 4.577919127669241,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00034190184795295577,
|
|
"loss": 4.9713,
|
|
"mean_token_accuracy": 0.19657502621412276,
|
|
"num_tokens": 10432544.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.370634603500366,
|
|
"epoch": 4.58246251703771,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034156432156609844,
|
|
"loss": 5.0537,
|
|
"mean_token_accuracy": 0.18853103816509248,
|
|
"num_tokens": 10442014.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.3384675025939945,
|
|
"epoch": 4.587005906406179,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00034122663093839466,
|
|
"loss": 4.9112,
|
|
"mean_token_accuracy": 0.19477406293153762,
|
|
"num_tokens": 10451093.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.36351261138916,
|
|
"epoch": 4.591549295774648,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003408887769030624,
|
|
"loss": 4.9417,
|
|
"mean_token_accuracy": 0.19278023689985274,
|
|
"num_tokens": 10461678.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.2931910991668705,
|
|
"epoch": 4.596092685143117,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000340550760293723,
|
|
"loss": 5.0079,
|
|
"mean_token_accuracy": 0.1893777996301651,
|
|
"num_tokens": 10471607.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.389155435562134,
|
|
"epoch": 4.600636074511586,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00034021258194439876,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.19169071167707444,
|
|
"num_tokens": 10482453.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.2998288631439205,
|
|
"epoch": 4.605179463880054,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00033987424268951117,
|
|
"loss": 4.9469,
|
|
"mean_token_accuracy": 0.19418544620275496,
|
|
"num_tokens": 10492424.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.34158182144165,
|
|
"epoch": 4.609722853248523,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00033953574336387867,
|
|
"loss": 4.9691,
|
|
"mean_token_accuracy": 0.18988646566867828,
|
|
"num_tokens": 10500958.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.38178448677063,
|
|
"epoch": 4.614266242616992,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003391970848027148,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.19647185653448104,
|
|
"num_tokens": 10510906.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.381689357757568,
|
|
"epoch": 4.618809631985461,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003388582678416257,
|
|
"loss": 4.9722,
|
|
"mean_token_accuracy": 0.18955312073230743,
|
|
"num_tokens": 10521832.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.394029521942139,
|
|
"epoch": 4.62335302135393,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00033851929331660876,
|
|
"loss": 4.9602,
|
|
"mean_token_accuracy": 0.194271220266819,
|
|
"num_tokens": 10532310.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.324763631820678,
|
|
"epoch": 4.627896410722399,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033818016206404983,
|
|
"loss": 5.0087,
|
|
"mean_token_accuracy": 0.1862783282995224,
|
|
"num_tokens": 10543979.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.433263349533081,
|
|
"epoch": 4.632439800090868,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003378408749207215,
|
|
"loss": 5.1029,
|
|
"mean_token_accuracy": 0.1808059260249138,
|
|
"num_tokens": 10553757.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.412091207504273,
|
|
"epoch": 4.636983189459337,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003375014327237812,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.18536376506090163,
|
|
"num_tokens": 10563575.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.432438468933105,
|
|
"epoch": 4.641526578827806,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0003371618363107686,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.18551895916461944,
|
|
"num_tokens": 10572828.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.350871515274048,
|
|
"epoch": 4.646069968196274,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00033682208651960444,
|
|
"loss": 5.0412,
|
|
"mean_token_accuracy": 0.18091315180063247,
|
|
"num_tokens": 10582567.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.346854496002197,
|
|
"epoch": 4.650613357564743,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00033648218418858716,
|
|
"loss": 5.0195,
|
|
"mean_token_accuracy": 0.19157850593328477,
|
|
"num_tokens": 10594046.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.38481068611145,
|
|
"epoch": 4.655156746933212,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0003361421301563922,
|
|
"loss": 4.9895,
|
|
"mean_token_accuracy": 0.18942970931529998,
|
|
"num_tokens": 10604343.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.332149362564087,
|
|
"epoch": 4.659700136301681,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003358019252620691,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.1842874988913536,
|
|
"num_tokens": 10615140.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.389857435226441,
|
|
"epoch": 4.66424352567015,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00033546157034503954,
|
|
"loss": 5.0391,
|
|
"mean_token_accuracy": 0.1931970715522766,
|
|
"num_tokens": 10624768.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.35596022605896,
|
|
"epoch": 4.668786915038619,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003351210662450955,
|
|
"loss": 4.9532,
|
|
"mean_token_accuracy": 0.19302236884832383,
|
|
"num_tokens": 10635246.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.2959836483001705,
|
|
"epoch": 4.673330304407088,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003347804138023968,
|
|
"loss": 4.9496,
|
|
"mean_token_accuracy": 0.19981496632099152,
|
|
"num_tokens": 10645118.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.379540729522705,
|
|
"epoch": 4.677873693775557,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003344396138574698,
|
|
"loss": 4.9714,
|
|
"mean_token_accuracy": 0.1937935993075371,
|
|
"num_tokens": 10655846.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.382776355743408,
|
|
"epoch": 4.682417083144026,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003340986672512042,
|
|
"loss": 5.0669,
|
|
"mean_token_accuracy": 0.1852417215704918,
|
|
"num_tokens": 10665741.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.397711944580078,
|
|
"epoch": 4.686960472512494,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003337575748248521,
|
|
"loss": 5.027,
|
|
"mean_token_accuracy": 0.18821866810321808,
|
|
"num_tokens": 10676120.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.374432325363159,
|
|
"epoch": 4.691503861880963,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00033341633742002503,
|
|
"loss": 4.9963,
|
|
"mean_token_accuracy": 0.1893892005085945,
|
|
"num_tokens": 10686195.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.342378807067871,
|
|
"epoch": 4.696047251249432,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003330749558786924,
|
|
"loss": 5.018,
|
|
"mean_token_accuracy": 0.19185762256383895,
|
|
"num_tokens": 10697278.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.407970571517945,
|
|
"epoch": 4.700590640617901,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0003327334310431791,
|
|
"loss": 5.0373,
|
|
"mean_token_accuracy": 0.19496458023786545,
|
|
"num_tokens": 10707937.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.3728227615356445,
|
|
"epoch": 4.70513402998637,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0003323917637561638,
|
|
"loss": 4.9406,
|
|
"mean_token_accuracy": 0.18923003375530242,
|
|
"num_tokens": 10717963.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.339050531387329,
|
|
"epoch": 4.709677419354839,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.00033204995486067664,
|
|
"loss": 5.0306,
|
|
"mean_token_accuracy": 0.1859367996454239,
|
|
"num_tokens": 10728294.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.427348613739014,
|
|
"epoch": 4.714220808723308,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.00033170800520009703,
|
|
"loss": 5.0765,
|
|
"mean_token_accuracy": 0.1842735379934311,
|
|
"num_tokens": 10739617.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.386172199249268,
|
|
"epoch": 4.718764198091776,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003313659156181518,
|
|
"loss": 5.1031,
|
|
"mean_token_accuracy": 0.1830982968211174,
|
|
"num_tokens": 10750704.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.406493186950684,
|
|
"epoch": 4.723307587460245,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.00033102368695891283,
|
|
"loss": 4.9849,
|
|
"mean_token_accuracy": 0.19748236387968063,
|
|
"num_tokens": 10760106.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.4709350109100345,
|
|
"epoch": 4.727850976828714,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00033068132006679554,
|
|
"loss": 5.1548,
|
|
"mean_token_accuracy": 0.18328567147254943,
|
|
"num_tokens": 10770963.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.3829206943511965,
|
|
"epoch": 4.732394366197183,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.00033033881578655605,
|
|
"loss": 5.0862,
|
|
"mean_token_accuracy": 0.18608589619398117,
|
|
"num_tokens": 10781441.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.429991769790649,
|
|
"epoch": 4.736937755565652,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003299961749632898,
|
|
"loss": 5.0719,
|
|
"mean_token_accuracy": 0.1905667781829834,
|
|
"num_tokens": 10791548.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.414743518829345,
|
|
"epoch": 4.741481144934121,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.00032965339844242864,
|
|
"loss": 5.0242,
|
|
"mean_token_accuracy": 0.18863395601511002,
|
|
"num_tokens": 10801525.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.334980249404907,
|
|
"epoch": 4.74602453430259,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003293104870697399,
|
|
"loss": 4.9996,
|
|
"mean_token_accuracy": 0.1833615481853485,
|
|
"num_tokens": 10811907.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.3371936798095705,
|
|
"epoch": 4.750567923671059,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00032896744169132305,
|
|
"loss": 4.9831,
|
|
"mean_token_accuracy": 0.19588419497013093,
|
|
"num_tokens": 10822996.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.371931076049805,
|
|
"epoch": 4.755111313039528,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003286242631536087,
|
|
"loss": 4.9357,
|
|
"mean_token_accuracy": 0.19018313139677048,
|
|
"num_tokens": 10832561.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.378829765319824,
|
|
"epoch": 4.759654702407996,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.00032828095230335555,
|
|
"loss": 5.0346,
|
|
"mean_token_accuracy": 0.18584325164556503,
|
|
"num_tokens": 10842591.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.326767778396606,
|
|
"epoch": 4.764198091776465,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003279375099876491,
|
|
"loss": 5.0559,
|
|
"mean_token_accuracy": 0.184274423122406,
|
|
"num_tokens": 10852888.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.4165050983428955,
|
|
"epoch": 4.768741481144934,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003275939370538992,
|
|
"loss": 5.0653,
|
|
"mean_token_accuracy": 0.19067742973566054,
|
|
"num_tokens": 10863792.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.360617446899414,
|
|
"epoch": 4.773284870513403,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00032725023434983767,
|
|
"loss": 4.9371,
|
|
"mean_token_accuracy": 0.1989527702331543,
|
|
"num_tokens": 10873206.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.380719900131226,
|
|
"epoch": 4.777828259881872,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00032690640272351715,
|
|
"loss": 5.0882,
|
|
"mean_token_accuracy": 0.18273740112781525,
|
|
"num_tokens": 10883686.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.256427192687989,
|
|
"epoch": 4.782371649250341,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003265624430233076,
|
|
"loss": 4.9958,
|
|
"mean_token_accuracy": 0.19738392978906633,
|
|
"num_tokens": 10894416.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.448296689987183,
|
|
"epoch": 4.78691503861881,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0003262183560978956,
|
|
"loss": 5.1027,
|
|
"mean_token_accuracy": 0.19178464412689208,
|
|
"num_tokens": 10904405.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.494251441955567,
|
|
"epoch": 4.791458427987278,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003258741427962814,
|
|
"loss": 5.0181,
|
|
"mean_token_accuracy": 0.19102512001991273,
|
|
"num_tokens": 10915339.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.3709653377532955,
|
|
"epoch": 4.796001817355747,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00032552980396777714,
|
|
"loss": 5.0179,
|
|
"mean_token_accuracy": 0.1877876877784729,
|
|
"num_tokens": 10926373.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.370393514633179,
|
|
"epoch": 4.800545206724216,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00032518534046200464,
|
|
"loss": 5.0798,
|
|
"mean_token_accuracy": 0.18923916965723037,
|
|
"num_tokens": 10936695.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.471764755249024,
|
|
"epoch": 4.805088596092685,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003248407531288932,
|
|
"loss": 5.1399,
|
|
"mean_token_accuracy": 0.17721383571624755,
|
|
"num_tokens": 10949741.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.420957708358765,
|
|
"epoch": 4.809631985461154,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000324496042818678,
|
|
"loss": 5.071,
|
|
"mean_token_accuracy": 0.19088889360427858,
|
|
"num_tokens": 10960786.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.3931543827056885,
|
|
"epoch": 4.814175374829623,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00032415121038189746,
|
|
"loss": 5.0022,
|
|
"mean_token_accuracy": 0.19492767304182052,
|
|
"num_tokens": 10970423.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.361189699172973,
|
|
"epoch": 4.818718764198092,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00032380625666939125,
|
|
"loss": 5.0035,
|
|
"mean_token_accuracy": 0.19339211285114288,
|
|
"num_tokens": 10980196.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.350902080535889,
|
|
"epoch": 4.823262153566561,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00032346118253229834,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.19087057262659074,
|
|
"num_tokens": 10989857.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.401997900009155,
|
|
"epoch": 4.82780554293503,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00032311598882205496,
|
|
"loss": 5.0285,
|
|
"mean_token_accuracy": 0.1812736228108406,
|
|
"num_tokens": 11000276.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.379430198669434,
|
|
"epoch": 4.832348932303498,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00032277067639039216,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.17555409222841262,
|
|
"num_tokens": 11012502.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.35689344406128,
|
|
"epoch": 4.836892321671967,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032242524608933423,
|
|
"loss": 4.956,
|
|
"mean_token_accuracy": 0.18902430087327957,
|
|
"num_tokens": 11023042.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.388507080078125,
|
|
"epoch": 4.841435711040436,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00032207969877119587,
|
|
"loss": 4.994,
|
|
"mean_token_accuracy": 0.1952795132994652,
|
|
"num_tokens": 11032259.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.452269649505615,
|
|
"epoch": 4.845979100408905,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032173403528858093,
|
|
"loss": 5.128,
|
|
"mean_token_accuracy": 0.1850266233086586,
|
|
"num_tokens": 11042324.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.36343560218811,
|
|
"epoch": 4.850522489777374,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003213882564943796,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.1890823647379875,
|
|
"num_tokens": 11053519.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.392101812362671,
|
|
"epoch": 4.855065879145843,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003210423632417666,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.19582831412553786,
|
|
"num_tokens": 11063413.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.3411932468414305,
|
|
"epoch": 4.859609268514312,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003206963563841993,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.18519364446401596,
|
|
"num_tokens": 11073232.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.316302919387818,
|
|
"epoch": 4.86415265788278,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00032035023677541515,
|
|
"loss": 4.943,
|
|
"mean_token_accuracy": 0.1915128529071808,
|
|
"num_tokens": 11083408.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.340404415130616,
|
|
"epoch": 4.868696047251249,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003200040052694299,
|
|
"loss": 5.0708,
|
|
"mean_token_accuracy": 0.18527871370315552,
|
|
"num_tokens": 11094125.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.390915250778198,
|
|
"epoch": 4.873239436619718,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00031965766272053534,
|
|
"loss": 5.0498,
|
|
"mean_token_accuracy": 0.18285363763570786,
|
|
"num_tokens": 11106293.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.362427139282227,
|
|
"epoch": 4.877782825988187,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00031931120998329737,
|
|
"loss": 4.9463,
|
|
"mean_token_accuracy": 0.2036087691783905,
|
|
"num_tokens": 11116485.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.395703458786011,
|
|
"epoch": 4.882326215356656,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003189646479125536,
|
|
"loss": 5.1053,
|
|
"mean_token_accuracy": 0.18283714205026627,
|
|
"num_tokens": 11127876.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.446658182144165,
|
|
"epoch": 4.886869604725125,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0003186179773634115,
|
|
"loss": 5.1496,
|
|
"mean_token_accuracy": 0.18170513808727265,
|
|
"num_tokens": 11138871.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.409112215042114,
|
|
"epoch": 4.891412994093594,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031827119919124614,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.18656474053859712,
|
|
"num_tokens": 11148872.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.2837296485900875,
|
|
"epoch": 4.895956383462063,
|
|
"grad_norm": 2.984375,
|
|
"learning_rate": 0.00031792431425169826,
|
|
"loss": 4.919,
|
|
"mean_token_accuracy": 0.20070744603872298,
|
|
"num_tokens": 11159716.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.363523149490357,
|
|
"epoch": 4.900499772830532,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00031757732340067195,
|
|
"loss": 5.0043,
|
|
"mean_token_accuracy": 0.19453287571668626,
|
|
"num_tokens": 11169187.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.3966357707977295,
|
|
"epoch": 4.905043162199,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00031723022749433267,
|
|
"loss": 4.9591,
|
|
"mean_token_accuracy": 0.1971052661538124,
|
|
"num_tokens": 11179889.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.441716814041138,
|
|
"epoch": 4.909586551567469,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000316883027389105,
|
|
"loss": 5.1477,
|
|
"mean_token_accuracy": 0.18433551043272017,
|
|
"num_tokens": 11189439.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.424556493759155,
|
|
"epoch": 4.914129940935938,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003165357239416706,
|
|
"loss": 5.0746,
|
|
"mean_token_accuracy": 0.18542537242174148,
|
|
"num_tokens": 11200834.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.376781845092774,
|
|
"epoch": 4.918673330304407,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003161883180089665,
|
|
"loss": 4.9811,
|
|
"mean_token_accuracy": 0.197477388381958,
|
|
"num_tokens": 11210475.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.347079610824585,
|
|
"epoch": 4.923216719672876,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003158408104481821,
|
|
"loss": 5.0572,
|
|
"mean_token_accuracy": 0.18596005141735078,
|
|
"num_tokens": 11220744.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.422362852096557,
|
|
"epoch": 4.927760109041345,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00031549320211675776,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.1806068405508995,
|
|
"num_tokens": 11231108.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.454173564910889,
|
|
"epoch": 4.932303498409814,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00031514549387238246,
|
|
"loss": 5.0693,
|
|
"mean_token_accuracy": 0.18910963088274002,
|
|
"num_tokens": 11241160.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.398426294326782,
|
|
"epoch": 4.936846887778283,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003147976865729918,
|
|
"loss": 5.0941,
|
|
"mean_token_accuracy": 0.18037124574184418,
|
|
"num_tokens": 11251711.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.388494443893433,
|
|
"epoch": 4.941390277146752,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031444978107676567,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.19094152748584747,
|
|
"num_tokens": 11263025.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.374826669692993,
|
|
"epoch": 4.94593366651522,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00031410177824212635,
|
|
"loss": 5.0398,
|
|
"mean_token_accuracy": 0.1876499354839325,
|
|
"num_tokens": 11273927.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.3767012596130375,
|
|
"epoch": 4.950477055883689,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003137536789277362,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.18956393450498582,
|
|
"num_tokens": 11284108.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.403408861160278,
|
|
"epoch": 4.955020445252158,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003134054839924957,
|
|
"loss": 4.9447,
|
|
"mean_token_accuracy": 0.19584997296333312,
|
|
"num_tokens": 11294530.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.302359580993652,
|
|
"epoch": 4.959563834620627,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00031305719429554124,
|
|
"loss": 5.0041,
|
|
"mean_token_accuracy": 0.18805685341358186,
|
|
"num_tokens": 11304803.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.3792112350463865,
|
|
"epoch": 4.964107223989096,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003127088106962429,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.18987852931022645,
|
|
"num_tokens": 11314567.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.343109846115112,
|
|
"epoch": 4.968650613357565,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00031236033405420284,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.1883157804608345,
|
|
"num_tokens": 11324452.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.395601558685303,
|
|
"epoch": 4.973194002726034,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00031201176522925236,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.19120240211486816,
|
|
"num_tokens": 11334930.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.341062450408936,
|
|
"epoch": 4.977737392094502,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0003116631050814503,
|
|
"loss": 4.9541,
|
|
"mean_token_accuracy": 0.19832389056682587,
|
|
"num_tokens": 11345487.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.409185123443604,
|
|
"epoch": 4.982280781462971,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00031131435447108116,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.18907360583543778,
|
|
"num_tokens": 11355150.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.402652597427368,
|
|
"epoch": 4.98682417083144,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003109655142586521,
|
|
"loss": 5.0227,
|
|
"mean_token_accuracy": 0.1914368525147438,
|
|
"num_tokens": 11366642.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.333501958847046,
|
|
"epoch": 4.991367560199909,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00031061658530489195,
|
|
"loss": 5.0291,
|
|
"mean_token_accuracy": 0.18604944944381713,
|
|
"num_tokens": 11377229.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.304904127120972,
|
|
"epoch": 4.995910949568378,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00031026756847074804,
|
|
"loss": 5.0762,
|
|
"mean_token_accuracy": 0.18903617411851883,
|
|
"num_tokens": 11387444.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.995910949568378,
|
|
"eval_entropy": 5.2989203164743826,
|
|
"eval_loss": 5.773827075958252,
|
|
"eval_mean_token_accuracy": 0.16412060635034428,
|
|
"eval_num_tokens": 11387444.0,
|
|
"eval_runtime": 2.5787,
|
|
"eval_samples_per_second": 1332.079,
|
|
"eval_steps_per_second": 166.752,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.411107699076335,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0003099184646173849,
|
|
"loss": 5.0289,
|
|
"mean_token_accuracy": 0.19265816112359366,
|
|
"num_tokens": 11396985.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.387855100631714,
|
|
"epoch": 5.004543389368469,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00030956927460618126,
|
|
"loss": 4.8433,
|
|
"mean_token_accuracy": 0.19894777983427048,
|
|
"num_tokens": 11408747.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.3322971820831295,
|
|
"epoch": 5.009086778736938,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003092199992987291,
|
|
"loss": 4.7724,
|
|
"mean_token_accuracy": 0.20479318350553513,
|
|
"num_tokens": 11418695.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.31050477027893,
|
|
"epoch": 5.013630168105407,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00030887063955683044,
|
|
"loss": 4.851,
|
|
"mean_token_accuracy": 0.1978992700576782,
|
|
"num_tokens": 11428143.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.387004852294922,
|
|
"epoch": 5.018173557473876,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030852119624249553,
|
|
"loss": 4.9095,
|
|
"mean_token_accuracy": 0.19582756608724594,
|
|
"num_tokens": 11439610.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.390898275375366,
|
|
"epoch": 5.022716946842344,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003081716702179412,
|
|
"loss": 4.8606,
|
|
"mean_token_accuracy": 0.20140896290540694,
|
|
"num_tokens": 11450035.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.3226916790008545,
|
|
"epoch": 5.027260336210813,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000307822062345588,
|
|
"loss": 4.8216,
|
|
"mean_token_accuracy": 0.2095102399587631,
|
|
"num_tokens": 11460736.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.316918849945068,
|
|
"epoch": 5.031803725579282,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003074723734880588,
|
|
"loss": 4.8072,
|
|
"mean_token_accuracy": 0.20276584178209306,
|
|
"num_tokens": 11471319.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.291299819946289,
|
|
"epoch": 5.036347114947751,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000307122604508176,
|
|
"loss": 4.8103,
|
|
"mean_token_accuracy": 0.2026289388537407,
|
|
"num_tokens": 11481052.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.323419046401978,
|
|
"epoch": 5.04089050431622,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003067727562689598,
|
|
"loss": 4.8181,
|
|
"mean_token_accuracy": 0.20073846131563186,
|
|
"num_tokens": 11491166.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.354222965240479,
|
|
"epoch": 5.045433893684689,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003064228296336259,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20521838068962098,
|
|
"num_tokens": 11501007.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.349460124969482,
|
|
"epoch": 5.049977283053158,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003060728254655836,
|
|
"loss": 4.8905,
|
|
"mean_token_accuracy": 0.19606309831142427,
|
|
"num_tokens": 11512197.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.351277923583984,
|
|
"epoch": 5.054520672421627,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00030572274462843327,
|
|
"loss": 4.852,
|
|
"mean_token_accuracy": 0.19559213966131211,
|
|
"num_tokens": 11522396.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.354201030731201,
|
|
"epoch": 5.059064061790095,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030537258798596465,
|
|
"loss": 4.8763,
|
|
"mean_token_accuracy": 0.2025006130337715,
|
|
"num_tokens": 11533034.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.434548187255859,
|
|
"epoch": 5.063607451158564,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0003050223564021543,
|
|
"loss": 4.8962,
|
|
"mean_token_accuracy": 0.20179938524961472,
|
|
"num_tokens": 11543281.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.322049713134765,
|
|
"epoch": 5.068150840527033,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003046720507411641,
|
|
"loss": 4.7949,
|
|
"mean_token_accuracy": 0.20763815194368362,
|
|
"num_tokens": 11552905.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.296164464950562,
|
|
"epoch": 5.072694229895502,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003043216718673383,
|
|
"loss": 4.8538,
|
|
"mean_token_accuracy": 0.20509768277406693,
|
|
"num_tokens": 11562397.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.419384002685547,
|
|
"epoch": 5.077237619263971,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030397122064520186,
|
|
"loss": 4.9208,
|
|
"mean_token_accuracy": 0.19034814983606338,
|
|
"num_tokens": 11573375.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.398184108734131,
|
|
"epoch": 5.08178100863244,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00030362069793945856,
|
|
"loss": 4.8405,
|
|
"mean_token_accuracy": 0.19762754589319229,
|
|
"num_tokens": 11584155.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.322801494598389,
|
|
"epoch": 5.086324398000909,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003032701046149881,
|
|
"loss": 4.8284,
|
|
"mean_token_accuracy": 0.20432326793670655,
|
|
"num_tokens": 11593929.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.344111585617066,
|
|
"epoch": 5.090867787369378,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003029194415368448,
|
|
"loss": 4.8568,
|
|
"mean_token_accuracy": 0.20097170919179916,
|
|
"num_tokens": 11604891.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.3820703506469725,
|
|
"epoch": 5.095411176737846,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.0003025687095702548,
|
|
"loss": 4.9318,
|
|
"mean_token_accuracy": 0.18907665610313415,
|
|
"num_tokens": 11616377.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.262263345718384,
|
|
"epoch": 5.099954566106315,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003022179095806146,
|
|
"loss": 4.866,
|
|
"mean_token_accuracy": 0.19518474489450455,
|
|
"num_tokens": 11627079.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.3414954662323,
|
|
"epoch": 5.104497955474784,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.000301867042433488,
|
|
"loss": 4.8443,
|
|
"mean_token_accuracy": 0.20371268689632416,
|
|
"num_tokens": 11638014.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.439047336578369,
|
|
"epoch": 5.109041344843253,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030151610899460505,
|
|
"loss": 4.9133,
|
|
"mean_token_accuracy": 0.2010794013738632,
|
|
"num_tokens": 11648759.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.335438680648804,
|
|
"epoch": 5.113584734211722,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030116511012985916,
|
|
"loss": 4.8615,
|
|
"mean_token_accuracy": 0.20259177386760713,
|
|
"num_tokens": 11659050.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.329847574234009,
|
|
"epoch": 5.118128123580191,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003008140467053048,
|
|
"loss": 4.8993,
|
|
"mean_token_accuracy": 0.19453393220901488,
|
|
"num_tokens": 11670131.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.358381557464599,
|
|
"epoch": 5.12267151294866,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003004629195871565,
|
|
"loss": 4.7948,
|
|
"mean_token_accuracy": 0.21002567261457444,
|
|
"num_tokens": 11680564.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.414137840270996,
|
|
"epoch": 5.127214902317129,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003001117296417852,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.1947482094168663,
|
|
"num_tokens": 11691910.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.281868410110474,
|
|
"epoch": 5.131758291685597,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002997604777357173,
|
|
"loss": 4.8599,
|
|
"mean_token_accuracy": 0.19462326318025588,
|
|
"num_tokens": 11702174.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.402305793762207,
|
|
"epoch": 5.136301681054066,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000299409164735632,
|
|
"loss": 4.9894,
|
|
"mean_token_accuracy": 0.1917511999607086,
|
|
"num_tokens": 11712841.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.422934436798096,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002990577915083591,
|
|
"loss": 4.8852,
|
|
"mean_token_accuracy": 0.196043398976326,
|
|
"num_tokens": 11722781.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.289358806610108,
|
|
"epoch": 5.145388459791004,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00029870635892087716,
|
|
"loss": 4.7607,
|
|
"mean_token_accuracy": 0.2109365165233612,
|
|
"num_tokens": 11732774.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.27865195274353,
|
|
"epoch": 5.149931849159473,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002983548678403112,
|
|
"loss": 4.8597,
|
|
"mean_token_accuracy": 0.19955627024173736,
|
|
"num_tokens": 11743308.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.360081386566162,
|
|
"epoch": 5.154475238527942,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002980033191339305,
|
|
"loss": 4.945,
|
|
"mean_token_accuracy": 0.1885734722018242,
|
|
"num_tokens": 11754572.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.3304070949554445,
|
|
"epoch": 5.159018627896411,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002976517136691465,
|
|
"loss": 4.8581,
|
|
"mean_token_accuracy": 0.2005347266793251,
|
|
"num_tokens": 11764406.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.333689641952515,
|
|
"epoch": 5.16356201726488,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00029730005231351074,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.19650254696607589,
|
|
"num_tokens": 11774121.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.363025283813476,
|
|
"epoch": 5.168105406633348,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002969483359347127,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.1935826361179352,
|
|
"num_tokens": 11784820.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.336814832687378,
|
|
"epoch": 5.172648796001817,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0002965965654005776,
|
|
"loss": 4.8471,
|
|
"mean_token_accuracy": 0.2042616680264473,
|
|
"num_tokens": 11794891.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.34317946434021,
|
|
"epoch": 5.177192185370286,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002962447415790642,
|
|
"loss": 4.8303,
|
|
"mean_token_accuracy": 0.20232591778039932,
|
|
"num_tokens": 11804842.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.3444806098937985,
|
|
"epoch": 5.181735574738755,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00029589286533826276,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.19800789505243302,
|
|
"num_tokens": 11815893.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.280011367797852,
|
|
"epoch": 5.186278964107224,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0002955409375463931,
|
|
"loss": 4.7963,
|
|
"mean_token_accuracy": 0.20849560350179672,
|
|
"num_tokens": 11826332.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.363341426849365,
|
|
"epoch": 5.190822353475693,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002951889590718019,
|
|
"loss": 4.909,
|
|
"mean_token_accuracy": 0.19804689884185792,
|
|
"num_tokens": 11836977.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.36493935585022,
|
|
"epoch": 5.195365742844162,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0002948369307829612,
|
|
"loss": 4.8265,
|
|
"mean_token_accuracy": 0.20093726813793183,
|
|
"num_tokens": 11846483.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.3642572402954105,
|
|
"epoch": 5.199909132212631,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002944848535484656,
|
|
"loss": 4.8809,
|
|
"mean_token_accuracy": 0.20075899958610535,
|
|
"num_tokens": 11857055.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.306966018676758,
|
|
"epoch": 5.2044525215811,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00029413272823703085,
|
|
"loss": 4.7563,
|
|
"mean_token_accuracy": 0.20914650112390518,
|
|
"num_tokens": 11866929.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.326145076751709,
|
|
"epoch": 5.208995910949568,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002937805557174911,
|
|
"loss": 4.8963,
|
|
"mean_token_accuracy": 0.1951029196381569,
|
|
"num_tokens": 11875973.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.342927885055542,
|
|
"epoch": 5.213539300318037,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.00029342833685879713,
|
|
"loss": 4.9027,
|
|
"mean_token_accuracy": 0.20408157259225845,
|
|
"num_tokens": 11886273.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.444299411773682,
|
|
"epoch": 5.218082689686506,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002930760725300137,
|
|
"loss": 4.9152,
|
|
"mean_token_accuracy": 0.1972661316394806,
|
|
"num_tokens": 11897462.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.374633932113648,
|
|
"epoch": 5.222626079054975,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00029272376360031833,
|
|
"loss": 4.8747,
|
|
"mean_token_accuracy": 0.19535194635391234,
|
|
"num_tokens": 11908116.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.353086376190186,
|
|
"epoch": 5.227169468423444,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00029237141093899804,
|
|
"loss": 4.9464,
|
|
"mean_token_accuracy": 0.1922533318400383,
|
|
"num_tokens": 11918990.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.371299314498901,
|
|
"epoch": 5.231712857791913,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00029201901541544824,
|
|
"loss": 4.9382,
|
|
"mean_token_accuracy": 0.19789233207702636,
|
|
"num_tokens": 11929201.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.3460596084594725,
|
|
"epoch": 5.236256247160382,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0002916665778991696,
|
|
"loss": 4.8978,
|
|
"mean_token_accuracy": 0.19113261699676515,
|
|
"num_tokens": 11939002.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.288290309906006,
|
|
"epoch": 5.240799636528851,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002913140992597668,
|
|
"loss": 4.7563,
|
|
"mean_token_accuracy": 0.20476611852645873,
|
|
"num_tokens": 11949255.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.306636810302734,
|
|
"epoch": 5.245343025897319,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0002909615803669458,
|
|
"loss": 4.8271,
|
|
"mean_token_accuracy": 0.2065007731318474,
|
|
"num_tokens": 11959108.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.253688859939575,
|
|
"epoch": 5.249886415265788,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0002906090220905119,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.20501685589551927,
|
|
"num_tokens": 11969196.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.289860391616822,
|
|
"epoch": 5.254429804634257,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002902564253003677,
|
|
"loss": 4.8031,
|
|
"mean_token_accuracy": 0.20607158839702605,
|
|
"num_tokens": 11979101.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.307155752182007,
|
|
"epoch": 5.258973194002726,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002899037908665106,
|
|
"loss": 4.8534,
|
|
"mean_token_accuracy": 0.19790517538785934,
|
|
"num_tokens": 11988396.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.332606315612793,
|
|
"epoch": 5.263516583371195,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002895511196590312,
|
|
"loss": 4.8446,
|
|
"mean_token_accuracy": 0.2031342327594757,
|
|
"num_tokens": 11998355.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.312914848327637,
|
|
"epoch": 5.268059972739664,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002891984125481104,
|
|
"loss": 4.8417,
|
|
"mean_token_accuracy": 0.20249273329973222,
|
|
"num_tokens": 12008938.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.337293100357056,
|
|
"epoch": 5.272603362108133,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028884567040401814,
|
|
"loss": 4.9196,
|
|
"mean_token_accuracy": 0.19253465831279754,
|
|
"num_tokens": 12019875.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.328073120117187,
|
|
"epoch": 5.277146751476602,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002884928940971105,
|
|
"loss": 4.8582,
|
|
"mean_token_accuracy": 0.19893306344747544,
|
|
"num_tokens": 12029368.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.353769159317016,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000288140084497828,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.20066138952970505,
|
|
"num_tokens": 12040191.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.263271856307983,
|
|
"epoch": 5.286233530213539,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002877872424766932,
|
|
"loss": 4.8237,
|
|
"mean_token_accuracy": 0.2112629994750023,
|
|
"num_tokens": 12050087.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.277795076370239,
|
|
"epoch": 5.290776919582008,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002874343689043086,
|
|
"loss": 4.8652,
|
|
"mean_token_accuracy": 0.20840881764888763,
|
|
"num_tokens": 12061014.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.317911958694458,
|
|
"epoch": 5.295320308950477,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000287081464651355,
|
|
"loss": 4.8696,
|
|
"mean_token_accuracy": 0.2076299086213112,
|
|
"num_tokens": 12071055.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.338882160186768,
|
|
"epoch": 5.299863698318946,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002867285305885882,
|
|
"loss": 4.855,
|
|
"mean_token_accuracy": 0.2016809105873108,
|
|
"num_tokens": 12081323.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.2865822315216064,
|
|
"epoch": 5.304407087687415,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00028637556758683816,
|
|
"loss": 4.8187,
|
|
"mean_token_accuracy": 0.20795361548662186,
|
|
"num_tokens": 12091482.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.24774694442749,
|
|
"epoch": 5.308950477055884,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002860225765170058,
|
|
"loss": 4.7594,
|
|
"mean_token_accuracy": 0.20919761955738067,
|
|
"num_tokens": 12101960.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.302752256393433,
|
|
"epoch": 5.313493866424353,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00028566955825006174,
|
|
"loss": 4.851,
|
|
"mean_token_accuracy": 0.19834230244159698,
|
|
"num_tokens": 12111654.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.397878360748291,
|
|
"epoch": 5.318037255792821,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00028531651365704325,
|
|
"loss": 4.9449,
|
|
"mean_token_accuracy": 0.19536163061857223,
|
|
"num_tokens": 12123096.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.318438720703125,
|
|
"epoch": 5.32258064516129,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00028496344360905287,
|
|
"loss": 4.8306,
|
|
"mean_token_accuracy": 0.20456456243991852,
|
|
"num_tokens": 12133064.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.286007642745972,
|
|
"epoch": 5.327124034529759,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028461034897725587,
|
|
"loss": 4.8567,
|
|
"mean_token_accuracy": 0.20500928312540054,
|
|
"num_tokens": 12144386.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.346109294891358,
|
|
"epoch": 5.331667423898228,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028425723063287803,
|
|
"loss": 4.9075,
|
|
"mean_token_accuracy": 0.19663746803998947,
|
|
"num_tokens": 12156111.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.412694978713989,
|
|
"epoch": 5.336210813266697,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028390408944720385,
|
|
"loss": 4.9547,
|
|
"mean_token_accuracy": 0.19140411913394928,
|
|
"num_tokens": 12166462.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.314660596847534,
|
|
"epoch": 5.340754202635166,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002835509262915741,
|
|
"loss": 4.9705,
|
|
"mean_token_accuracy": 0.1867436096072197,
|
|
"num_tokens": 12178635.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.293686056137085,
|
|
"epoch": 5.345297592003635,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00028319774203738375,
|
|
"loss": 4.8283,
|
|
"mean_token_accuracy": 0.19898282438516618,
|
|
"num_tokens": 12188880.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.261138439178467,
|
|
"epoch": 5.349840981372104,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002828445375560798,
|
|
"loss": 4.8175,
|
|
"mean_token_accuracy": 0.1975273683667183,
|
|
"num_tokens": 12199733.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.2514612674713135,
|
|
"epoch": 5.354384370740572,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00028249131371915927,
|
|
"loss": 4.885,
|
|
"mean_token_accuracy": 0.19852383732795714,
|
|
"num_tokens": 12209925.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.396322107315063,
|
|
"epoch": 5.358927760109041,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002821380713981668,
|
|
"loss": 4.9262,
|
|
"mean_token_accuracy": 0.20044863671064378,
|
|
"num_tokens": 12219870.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.311792707443237,
|
|
"epoch": 5.36347114947751,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00028178481146469273,
|
|
"loss": 4.8436,
|
|
"mean_token_accuracy": 0.20829369723796845,
|
|
"num_tokens": 12230295.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.3799491882324215,
|
|
"epoch": 5.368014538845979,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0002814315347903709,
|
|
"loss": 4.8603,
|
|
"mean_token_accuracy": 0.2039797991514206,
|
|
"num_tokens": 12239700.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.324176168441772,
|
|
"epoch": 5.372557928214448,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002810782422468763,
|
|
"loss": 4.936,
|
|
"mean_token_accuracy": 0.19343703240156174,
|
|
"num_tokens": 12249529.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.313447904586792,
|
|
"epoch": 5.377101317582917,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00028072493470592317,
|
|
"loss": 4.8049,
|
|
"mean_token_accuracy": 0.20480545461177826,
|
|
"num_tokens": 12259774.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.3471416473388675,
|
|
"epoch": 5.381644706951386,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00028037161303926265,
|
|
"loss": 4.8567,
|
|
"mean_token_accuracy": 0.19918009042739868,
|
|
"num_tokens": 12270484.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.3432659149169925,
|
|
"epoch": 5.386188096319855,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000280018278118681,
|
|
"loss": 4.8438,
|
|
"mean_token_accuracy": 0.19865224808454512,
|
|
"num_tokens": 12281195.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.306499195098877,
|
|
"epoch": 5.390731485688324,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000279664930815997,
|
|
"loss": 4.8278,
|
|
"mean_token_accuracy": 0.20193144977092742,
|
|
"num_tokens": 12292052.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.287984085083008,
|
|
"epoch": 5.395274875056792,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002793115720030599,
|
|
"loss": 4.9064,
|
|
"mean_token_accuracy": 0.20178543776273727,
|
|
"num_tokens": 12302562.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.288982629776001,
|
|
"epoch": 5.399818264425261,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0002789582025517474,
|
|
"loss": 4.9118,
|
|
"mean_token_accuracy": 0.20169538706541063,
|
|
"num_tokens": 12313875.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.376831436157227,
|
|
"epoch": 5.40436165379373,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0002786048233339637,
|
|
"loss": 4.9069,
|
|
"mean_token_accuracy": 0.19293956458568573,
|
|
"num_tokens": 12324500.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.333746004104614,
|
|
"epoch": 5.408905043162199,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 0.00027825143522163675,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.19926834404468535,
|
|
"num_tokens": 12335474.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.306619358062744,
|
|
"epoch": 5.413448432530668,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002778980390867165,
|
|
"loss": 4.8205,
|
|
"mean_token_accuracy": 0.20654188990592956,
|
|
"num_tokens": 12346174.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.3261433124542235,
|
|
"epoch": 5.417991821899137,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00027754463580117306,
|
|
"loss": 4.8477,
|
|
"mean_token_accuracy": 0.2084801122546196,
|
|
"num_tokens": 12356108.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.322306394577026,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00027719122623699353,
|
|
"loss": 4.9023,
|
|
"mean_token_accuracy": 0.2038418844342232,
|
|
"num_tokens": 12366699.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.415395212173462,
|
|
"epoch": 5.427078600636074,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002768378112661813,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.19175861775875092,
|
|
"num_tokens": 12377225.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.368198537826538,
|
|
"epoch": 5.431621990004543,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002764843917607523,
|
|
"loss": 4.9237,
|
|
"mean_token_accuracy": 0.20467139780521393,
|
|
"num_tokens": 12388936.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.313368082046509,
|
|
"epoch": 5.436165379373012,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002761309685927343,
|
|
"loss": 4.8481,
|
|
"mean_token_accuracy": 0.20026197284460068,
|
|
"num_tokens": 12399130.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.2981579303741455,
|
|
"epoch": 5.440708768741481,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002757775426341636,
|
|
"loss": 4.8436,
|
|
"mean_token_accuracy": 0.20730389952659606,
|
|
"num_tokens": 12410181.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.335815668106079,
|
|
"epoch": 5.44525215810995,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00027542411475708385,
|
|
"loss": 4.9314,
|
|
"mean_token_accuracy": 0.19259196668863296,
|
|
"num_tokens": 12421173.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.401829195022583,
|
|
"epoch": 5.449795547478419,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027507068583354304,
|
|
"loss": 4.9403,
|
|
"mean_token_accuracy": 0.18719758540391923,
|
|
"num_tokens": 12433195.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.449795547478419,
|
|
"eval_entropy": 5.3129775668299475,
|
|
"eval_loss": 5.767161846160889,
|
|
"eval_mean_token_accuracy": 0.1651711739253166,
|
|
"eval_num_tokens": 12433195.0,
|
|
"eval_runtime": 2.5793,
|
|
"eval_samples_per_second": 1331.777,
|
|
"eval_steps_per_second": 166.714,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.353214073181152,
|
|
"epoch": 5.454338936846888,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00027471725673559205,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.19766356199979782,
|
|
"num_tokens": 12442944.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.277512407302856,
|
|
"epoch": 5.458882326215357,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00027436382833528184,
|
|
"loss": 4.8083,
|
|
"mean_token_accuracy": 0.20863053649663926,
|
|
"num_tokens": 12453133.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.282884120941162,
|
|
"epoch": 5.463425715583826,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.0002740104015046621,
|
|
"loss": 4.9337,
|
|
"mean_token_accuracy": 0.1940288156270981,
|
|
"num_tokens": 12464606.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.300756406784058,
|
|
"epoch": 5.467969104952294,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 0.00027365697711577815,
|
|
"loss": 4.8424,
|
|
"mean_token_accuracy": 0.1936388224363327,
|
|
"num_tokens": 12474490.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.301889657974243,
|
|
"epoch": 5.472512494320763,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002733035560406697,
|
|
"loss": 4.849,
|
|
"mean_token_accuracy": 0.19900547713041306,
|
|
"num_tokens": 12485158.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.306543064117432,
|
|
"epoch": 5.477055883689232,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00027295013915136805,
|
|
"loss": 4.8931,
|
|
"mean_token_accuracy": 0.1927631378173828,
|
|
"num_tokens": 12496265.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.295451307296753,
|
|
"epoch": 5.481599273057701,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027259672731989436,
|
|
"loss": 4.7726,
|
|
"mean_token_accuracy": 0.20599951297044755,
|
|
"num_tokens": 12506719.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.29598445892334,
|
|
"epoch": 5.48614266242617,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000272243321418257,
|
|
"loss": 4.865,
|
|
"mean_token_accuracy": 0.2000267833471298,
|
|
"num_tokens": 12515990.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.31119966506958,
|
|
"epoch": 5.490686051794639,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002718899223184501,
|
|
"loss": 4.8817,
|
|
"mean_token_accuracy": 0.19871461540460586,
|
|
"num_tokens": 12526557.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.278953170776367,
|
|
"epoch": 5.495229441163108,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00027153653089245067,
|
|
"loss": 4.8004,
|
|
"mean_token_accuracy": 0.21179110556840897,
|
|
"num_tokens": 12536657.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.302927780151367,
|
|
"epoch": 5.499772830531577,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000271183148012217,
|
|
"loss": 4.8344,
|
|
"mean_token_accuracy": 0.20206410735845565,
|
|
"num_tokens": 12546958.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.3024204730987545,
|
|
"epoch": 5.504316219900045,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002708297745496861,
|
|
"loss": 4.8388,
|
|
"mean_token_accuracy": 0.19798460304737092,
|
|
"num_tokens": 12556978.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.2924378395080565,
|
|
"epoch": 5.508859609268514,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00027047641137677213,
|
|
"loss": 4.8192,
|
|
"mean_token_accuracy": 0.20626424700021745,
|
|
"num_tokens": 12566954.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.316256093978882,
|
|
"epoch": 5.513402998636983,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002701230593653632,
|
|
"loss": 4.8591,
|
|
"mean_token_accuracy": 0.20182436108589172,
|
|
"num_tokens": 12577129.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.3046129703521725,
|
|
"epoch": 5.517946388005452,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0002697697193873207,
|
|
"loss": 4.861,
|
|
"mean_token_accuracy": 0.19280418157577514,
|
|
"num_tokens": 12586802.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.332994508743286,
|
|
"epoch": 5.522489777373921,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026941639231447547,
|
|
"loss": 4.9053,
|
|
"mean_token_accuracy": 0.19478945732116698,
|
|
"num_tokens": 12597732.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.328513479232788,
|
|
"epoch": 5.52703316674239,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002690630790186273,
|
|
"loss": 4.7674,
|
|
"mean_token_accuracy": 0.20590104013681412,
|
|
"num_tokens": 12607185.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.347691631317138,
|
|
"epoch": 5.531576556110859,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026870978037154137,
|
|
"loss": 4.9004,
|
|
"mean_token_accuracy": 0.20062846690416336,
|
|
"num_tokens": 12616821.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.34202823638916,
|
|
"epoch": 5.536119945479328,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002683564972449472,
|
|
"loss": 4.9076,
|
|
"mean_token_accuracy": 0.19329741597175598,
|
|
"num_tokens": 12626698.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.328075313568116,
|
|
"epoch": 5.540663334847796,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002680032305105355,
|
|
"loss": 4.8452,
|
|
"mean_token_accuracy": 0.20078007727861405,
|
|
"num_tokens": 12636588.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.396503543853759,
|
|
"epoch": 5.545206724216265,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0002676499810399572,
|
|
"loss": 4.9024,
|
|
"mean_token_accuracy": 0.19747193157672882,
|
|
"num_tokens": 12647196.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.3326927661895756,
|
|
"epoch": 5.549750113584734,
|
|
"grad_norm": 3.75,
|
|
"learning_rate": 0.00026729674970481985,
|
|
"loss": 4.8872,
|
|
"mean_token_accuracy": 0.19725604057312013,
|
|
"num_tokens": 12658631.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.300666093826294,
|
|
"epoch": 5.554293502953203,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00026694353737668705,
|
|
"loss": 4.9033,
|
|
"mean_token_accuracy": 0.20321848690509797,
|
|
"num_tokens": 12668905.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.261510848999023,
|
|
"epoch": 5.558836892321672,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00026659034492707473,
|
|
"loss": 4.8363,
|
|
"mean_token_accuracy": 0.20147041082382203,
|
|
"num_tokens": 12678270.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.329799842834473,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002662371732274504,
|
|
"loss": 4.8446,
|
|
"mean_token_accuracy": 0.20283385664224624,
|
|
"num_tokens": 12688248.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.360124683380127,
|
|
"epoch": 5.56792367105861,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00026588402314923,
|
|
"loss": 4.9285,
|
|
"mean_token_accuracy": 0.194251412153244,
|
|
"num_tokens": 12698418.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.263089370727539,
|
|
"epoch": 5.572467060427079,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0002655308955637763,
|
|
"loss": 4.778,
|
|
"mean_token_accuracy": 0.21052460372447968,
|
|
"num_tokens": 12708578.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.331230926513672,
|
|
"epoch": 5.577010449795548,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002651777913423965,
|
|
"loss": 4.8876,
|
|
"mean_token_accuracy": 0.19488443583250045,
|
|
"num_tokens": 12718899.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.26721887588501,
|
|
"epoch": 5.581553839164016,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002648247113563402,
|
|
"loss": 4.8231,
|
|
"mean_token_accuracy": 0.19611486196517944,
|
|
"num_tokens": 12728342.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.367319202423095,
|
|
"epoch": 5.586097228532485,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00026447165647679703,
|
|
"loss": 4.8801,
|
|
"mean_token_accuracy": 0.20532096475362777,
|
|
"num_tokens": 12739337.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.302419996261596,
|
|
"epoch": 5.590640617900954,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.000264118627574895,
|
|
"loss": 4.8737,
|
|
"mean_token_accuracy": 0.19580366164445878,
|
|
"num_tokens": 12749650.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.337950325012207,
|
|
"epoch": 5.595184007269423,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00026376562552169776,
|
|
"loss": 4.9208,
|
|
"mean_token_accuracy": 0.19134819656610488,
|
|
"num_tokens": 12760573.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.451846885681152,
|
|
"epoch": 5.599727396637892,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00026341265118820284,
|
|
"loss": 4.9452,
|
|
"mean_token_accuracy": 0.1949541077017784,
|
|
"num_tokens": 12771070.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.400966024398803,
|
|
"epoch": 5.604270786006361,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.00026305970544533903,
|
|
"loss": 4.8593,
|
|
"mean_token_accuracy": 0.20434089750051498,
|
|
"num_tokens": 12781896.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.285603618621826,
|
|
"epoch": 5.60881417537483,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00026270678916396527,
|
|
"loss": 4.9154,
|
|
"mean_token_accuracy": 0.20134656578302385,
|
|
"num_tokens": 12792675.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.311793947219849,
|
|
"epoch": 5.613357564743298,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00026235390321486705,
|
|
"loss": 4.8163,
|
|
"mean_token_accuracy": 0.20536674708127975,
|
|
"num_tokens": 12802371.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.3678020477294925,
|
|
"epoch": 5.617900954111767,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00026200104846875574,
|
|
"loss": 4.8691,
|
|
"mean_token_accuracy": 0.20340487062931062,
|
|
"num_tokens": 12813281.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.29765682220459,
|
|
"epoch": 5.622444343480236,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000261648225796265,
|
|
"loss": 4.906,
|
|
"mean_token_accuracy": 0.19416456520557404,
|
|
"num_tokens": 12823859.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.297624158859253,
|
|
"epoch": 5.626987732848705,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002612954360679499,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.2087327539920807,
|
|
"num_tokens": 12833288.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.303289556503296,
|
|
"epoch": 5.631531122217174,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.000260942680154284,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.2011386051774025,
|
|
"num_tokens": 12843698.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.266569757461548,
|
|
"epoch": 5.636074511585643,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002605899589256574,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.21375961899757384,
|
|
"num_tokens": 12853956.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.357098579406738,
|
|
"epoch": 5.640617900954112,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00026023727325237463,
|
|
"loss": 4.8957,
|
|
"mean_token_accuracy": 0.20028781294822692,
|
|
"num_tokens": 12864680.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.340889406204224,
|
|
"epoch": 5.645161290322581,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002598846240046527,
|
|
"loss": 4.7711,
|
|
"mean_token_accuracy": 0.20858325362205504,
|
|
"num_tokens": 12873790.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.359616136550903,
|
|
"epoch": 5.64970467969105,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002595320120526183,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.20702067762613297,
|
|
"num_tokens": 12884559.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.274465703964234,
|
|
"epoch": 5.654248069059518,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00025917943826630666,
|
|
"loss": 4.8936,
|
|
"mean_token_accuracy": 0.20152451545000077,
|
|
"num_tokens": 12896176.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.279437160491943,
|
|
"epoch": 5.658791458427987,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002588269035156584,
|
|
"loss": 4.8352,
|
|
"mean_token_accuracy": 0.19730358570814133,
|
|
"num_tokens": 12906381.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.3195068359375,
|
|
"epoch": 5.663334847796456,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00025847440867051796,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.1976109579205513,
|
|
"num_tokens": 12915738.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.330824470520019,
|
|
"epoch": 5.667878237164925,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002581219546006314,
|
|
"loss": 4.8359,
|
|
"mean_token_accuracy": 0.20574986487627028,
|
|
"num_tokens": 12926059.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.325921106338501,
|
|
"epoch": 5.672421626533394,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002577695421756441,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.1971482366323471,
|
|
"num_tokens": 12936730.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.3733015060424805,
|
|
"epoch": 5.676965015901863,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002574171722650986,
|
|
"loss": 4.8828,
|
|
"mean_token_accuracy": 0.19474603682756425,
|
|
"num_tokens": 12945687.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.402496337890625,
|
|
"epoch": 5.681508405270332,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00025706484573843273,
|
|
"loss": 4.8922,
|
|
"mean_token_accuracy": 0.2039039671421051,
|
|
"num_tokens": 12955087.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.301148748397827,
|
|
"epoch": 5.6860517946388,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.000256712563464977,
|
|
"loss": 4.8343,
|
|
"mean_token_accuracy": 0.1945272982120514,
|
|
"num_tokens": 12965497.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.3920924186706545,
|
|
"epoch": 5.690595184007269,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002563603263139532,
|
|
"loss": 5.0171,
|
|
"mean_token_accuracy": 0.19039050489664078,
|
|
"num_tokens": 12976520.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.3539612770080565,
|
|
"epoch": 5.695138573375738,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002560081351544711,
|
|
"loss": 4.8443,
|
|
"mean_token_accuracy": 0.20455364286899566,
|
|
"num_tokens": 12987271.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.348012685775757,
|
|
"epoch": 5.699681962744207,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025565599085552776,
|
|
"loss": 4.8685,
|
|
"mean_token_accuracy": 0.19701075702905654,
|
|
"num_tokens": 12998170.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.312058210372925,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00025530389428600394,
|
|
"loss": 4.8732,
|
|
"mean_token_accuracy": 0.20278197526931763,
|
|
"num_tokens": 13008059.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.301107740402221,
|
|
"epoch": 5.708768741481145,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002549518463146631,
|
|
"loss": 4.847,
|
|
"mean_token_accuracy": 0.20441219657659532,
|
|
"num_tokens": 13017904.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.377012920379639,
|
|
"epoch": 5.713312130849614,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002545998478101485,
|
|
"loss": 4.882,
|
|
"mean_token_accuracy": 0.2004920572042465,
|
|
"num_tokens": 13028399.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.335032749176025,
|
|
"epoch": 5.717855520218083,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00025424789964098154,
|
|
"loss": 4.8711,
|
|
"mean_token_accuracy": 0.19663021117448806,
|
|
"num_tokens": 13038420.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.313693475723267,
|
|
"epoch": 5.722398909586552,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002538960026755593,
|
|
"loss": 4.8708,
|
|
"mean_token_accuracy": 0.2039332240819931,
|
|
"num_tokens": 13048514.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.308422040939331,
|
|
"epoch": 5.72694229895502,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00025354415778215255,
|
|
"loss": 4.954,
|
|
"mean_token_accuracy": 0.20012225359678268,
|
|
"num_tokens": 13059728.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.330108594894409,
|
|
"epoch": 5.731485688323489,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00025319236582890364,
|
|
"loss": 4.9549,
|
|
"mean_token_accuracy": 0.1955789476633072,
|
|
"num_tokens": 13069535.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.297762155532837,
|
|
"epoch": 5.736029077691958,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00025284062768382424,
|
|
"loss": 4.8267,
|
|
"mean_token_accuracy": 0.1994822770357132,
|
|
"num_tokens": 13079012.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.3780420303344725,
|
|
"epoch": 5.740572467060427,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0002524889442147932,
|
|
"loss": 4.8438,
|
|
"mean_token_accuracy": 0.2110071912407875,
|
|
"num_tokens": 13089067.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.341086101531983,
|
|
"epoch": 5.745115856428896,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00025213731628955454,
|
|
"loss": 4.8767,
|
|
"mean_token_accuracy": 0.1958890900015831,
|
|
"num_tokens": 13100576.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.343083810806275,
|
|
"epoch": 5.749659245797365,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025178574477571525,
|
|
"loss": 5.0248,
|
|
"mean_token_accuracy": 0.1899324908852577,
|
|
"num_tokens": 13111926.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.291295194625855,
|
|
"epoch": 5.754202635165834,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00025143423054074305,
|
|
"loss": 4.8174,
|
|
"mean_token_accuracy": 0.19743947982788085,
|
|
"num_tokens": 13122362.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.3033052444458,
|
|
"epoch": 5.758746024534303,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002510827744519644,
|
|
"loss": 4.8368,
|
|
"mean_token_accuracy": 0.2028704598546028,
|
|
"num_tokens": 13133372.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.274821329116821,
|
|
"epoch": 5.763289413902772,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00025073137737656226,
|
|
"loss": 4.8149,
|
|
"mean_token_accuracy": 0.2023646876215935,
|
|
"num_tokens": 13143440.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.348180532455444,
|
|
"epoch": 5.76783280327124,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0002503800401815739,
|
|
"loss": 4.8838,
|
|
"mean_token_accuracy": 0.1988438382744789,
|
|
"num_tokens": 13153564.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.329644107818604,
|
|
"epoch": 5.772376192639709,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00025002876373388913,
|
|
"loss": 4.9272,
|
|
"mean_token_accuracy": 0.19203673154115677,
|
|
"num_tokens": 13163679.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.326746892929077,
|
|
"epoch": 5.776919582008178,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00024967754890024737,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.20095942467451094,
|
|
"num_tokens": 13173930.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.342338752746582,
|
|
"epoch": 5.781462971376647,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002493263965472366,
|
|
"loss": 4.8644,
|
|
"mean_token_accuracy": 0.20586053729057313,
|
|
"num_tokens": 13184805.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.324298667907715,
|
|
"epoch": 5.786006360745116,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.0002489753075412901,
|
|
"loss": 4.9039,
|
|
"mean_token_accuracy": 0.1988490864634514,
|
|
"num_tokens": 13196229.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.319838905334473,
|
|
"epoch": 5.790549750113585,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024862428274868533,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.20962711721658706,
|
|
"num_tokens": 13205906.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.291900634765625,
|
|
"epoch": 5.795093139482054,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0002482733230355408,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.20088620334863663,
|
|
"num_tokens": 13215076.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.359968805313111,
|
|
"epoch": 5.799636528850522,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024792242926781495,
|
|
"loss": 4.9433,
|
|
"mean_token_accuracy": 0.1969107687473297,
|
|
"num_tokens": 13225761.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.361367988586426,
|
|
"epoch": 5.804179918218991,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00024757160231130313,
|
|
"loss": 4.9446,
|
|
"mean_token_accuracy": 0.1988942489027977,
|
|
"num_tokens": 13236342.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.2659022331237795,
|
|
"epoch": 5.80872330758746,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00024722084303163607,
|
|
"loss": 4.8094,
|
|
"mean_token_accuracy": 0.21130239069461823,
|
|
"num_tokens": 13244989.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.324373626708985,
|
|
"epoch": 5.813266696955929,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.0002468701522942774,
|
|
"loss": 4.8472,
|
|
"mean_token_accuracy": 0.20092013627290725,
|
|
"num_tokens": 13255495.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.3706563949584964,
|
|
"epoch": 5.817810086324398,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00024651953096452165,
|
|
"loss": 4.9302,
|
|
"mean_token_accuracy": 0.19615861475467683,
|
|
"num_tokens": 13266189.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.30243182182312,
|
|
"epoch": 5.822353475692867,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00024616897990749196,
|
|
"loss": 4.8725,
|
|
"mean_token_accuracy": 0.19918688982725144,
|
|
"num_tokens": 13276194.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.291888284683227,
|
|
"epoch": 5.826896865061336,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 0.0002458184999881385,
|
|
"loss": 4.8816,
|
|
"mean_token_accuracy": 0.19283343553543092,
|
|
"num_tokens": 13286188.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.352660655975342,
|
|
"epoch": 5.831440254429805,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024546809207123534,
|
|
"loss": 4.8417,
|
|
"mean_token_accuracy": 0.20542439669370652,
|
|
"num_tokens": 13295926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.369372367858887,
|
|
"epoch": 5.835983643798274,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002451177570213794,
|
|
"loss": 4.9287,
|
|
"mean_token_accuracy": 0.1979822739958763,
|
|
"num_tokens": 13307373.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.306203937530517,
|
|
"epoch": 5.840527033166742,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00024476749570298746,
|
|
"loss": 4.8528,
|
|
"mean_token_accuracy": 0.19583884477615357,
|
|
"num_tokens": 13316984.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.28069806098938,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024441730898029456,
|
|
"loss": 4.8172,
|
|
"mean_token_accuracy": 0.2061645582318306,
|
|
"num_tokens": 13327188.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.36755633354187,
|
|
"epoch": 5.84961381190368,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024406719771735165,
|
|
"loss": 4.9152,
|
|
"mean_token_accuracy": 0.19454556703567505,
|
|
"num_tokens": 13338108.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.262416124343872,
|
|
"epoch": 5.854157201272149,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002437171627780234,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.20302072465419768,
|
|
"num_tokens": 13347898.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.3314872741699215,
|
|
"epoch": 5.858700590640618,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00024336720502598624,
|
|
"loss": 4.9081,
|
|
"mean_token_accuracy": 0.19833500683307648,
|
|
"num_tokens": 13358649.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.280745220184326,
|
|
"epoch": 5.863243980009087,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002430173253247262,
|
|
"loss": 4.8757,
|
|
"mean_token_accuracy": 0.20123781263828278,
|
|
"num_tokens": 13368447.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.2785228252410885,
|
|
"epoch": 5.867787369377556,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024266752453753655,
|
|
"loss": 4.8707,
|
|
"mean_token_accuracy": 0.1985710546374321,
|
|
"num_tokens": 13377809.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.349882793426514,
|
|
"epoch": 5.872330758746024,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024231780352751615,
|
|
"loss": 4.9875,
|
|
"mean_token_accuracy": 0.19060766845941543,
|
|
"num_tokens": 13389352.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.356575679779053,
|
|
"epoch": 5.876874148114493,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00024196816315756665,
|
|
"loss": 4.8596,
|
|
"mean_token_accuracy": 0.20651315301656722,
|
|
"num_tokens": 13399929.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.35662875175476,
|
|
"epoch": 5.881417537482962,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002416186042903909,
|
|
"loss": 4.7631,
|
|
"mean_token_accuracy": 0.2081802636384964,
|
|
"num_tokens": 13410096.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.319591951370239,
|
|
"epoch": 5.885960926851431,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00024126912778849074,
|
|
"loss": 4.9575,
|
|
"mean_token_accuracy": 0.1967372328042984,
|
|
"num_tokens": 13420384.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.362309217453003,
|
|
"epoch": 5.8905043162199,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00024091973451416472,
|
|
"loss": 4.863,
|
|
"mean_token_accuracy": 0.19924790561199188,
|
|
"num_tokens": 13430887.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.324084901809693,
|
|
"epoch": 5.895047705588369,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024057042532950597,
|
|
"loss": 4.8716,
|
|
"mean_token_accuracy": 0.19691405594348907,
|
|
"num_tokens": 13440914.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.332412481307983,
|
|
"epoch": 5.899591094956838,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002402212010964001,
|
|
"loss": 4.9283,
|
|
"mean_token_accuracy": 0.1952826589345932,
|
|
"num_tokens": 13452476.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.333815813064575,
|
|
"epoch": 5.904134484325307,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023987206267652334,
|
|
"loss": 4.8451,
|
|
"mean_token_accuracy": 0.19982722252607346,
|
|
"num_tokens": 13463779.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.904134484325307,
|
|
"eval_entropy": 5.2524328753005625,
|
|
"eval_loss": 5.737683296203613,
|
|
"eval_mean_token_accuracy": 0.16736519337566785,
|
|
"eval_num_tokens": 13463779.0,
|
|
"eval_runtime": 2.5869,
|
|
"eval_samples_per_second": 1327.834,
|
|
"eval_steps_per_second": 166.221,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.3680657863616945,
|
|
"epoch": 5.908677873693776,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00023952301093133993,
|
|
"loss": 4.9641,
|
|
"mean_token_accuracy": 0.19897969514131547,
|
|
"num_tokens": 13475467.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.32771577835083,
|
|
"epoch": 5.913221263062244,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023917404672210046,
|
|
"loss": 4.8327,
|
|
"mean_token_accuracy": 0.1962883487343788,
|
|
"num_tokens": 13485484.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.318826055526733,
|
|
"epoch": 5.917764652430713,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002388251709098392,
|
|
"loss": 4.8332,
|
|
"mean_token_accuracy": 0.204604834318161,
|
|
"num_tokens": 13494941.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.355777883529663,
|
|
"epoch": 5.922308041799182,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023847638435537283,
|
|
"loss": 4.8814,
|
|
"mean_token_accuracy": 0.2069704219698906,
|
|
"num_tokens": 13504762.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.342716693878174,
|
|
"epoch": 5.926851431167651,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00023812768791929724,
|
|
"loss": 4.8574,
|
|
"mean_token_accuracy": 0.2001414716243744,
|
|
"num_tokens": 13515436.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.32881727218628,
|
|
"epoch": 5.93139482053612,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00023777908246198643,
|
|
"loss": 4.8596,
|
|
"mean_token_accuracy": 0.2037479266524315,
|
|
"num_tokens": 13525211.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.307476758956909,
|
|
"epoch": 5.935938209904589,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00023743056884358954,
|
|
"loss": 4.8556,
|
|
"mean_token_accuracy": 0.1970292240381241,
|
|
"num_tokens": 13534954.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.31577696800232,
|
|
"epoch": 5.940481599273058,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023708214792402942,
|
|
"loss": 4.8928,
|
|
"mean_token_accuracy": 0.20329148173332215,
|
|
"num_tokens": 13545459.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.28683500289917,
|
|
"epoch": 5.945024988641527,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023673382056299997,
|
|
"loss": 4.8597,
|
|
"mean_token_accuracy": 0.1997950941324234,
|
|
"num_tokens": 13554541.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.351372623443604,
|
|
"epoch": 5.949568378009996,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023638558761996436,
|
|
"loss": 4.8396,
|
|
"mean_token_accuracy": 0.20363293588161469,
|
|
"num_tokens": 13564810.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.315874910354614,
|
|
"epoch": 5.954111767378464,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023603744995415277,
|
|
"loss": 4.8339,
|
|
"mean_token_accuracy": 0.2031494364142418,
|
|
"num_tokens": 13574756.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.369516372680664,
|
|
"epoch": 5.958655156746933,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023568940842456026,
|
|
"loss": 4.9033,
|
|
"mean_token_accuracy": 0.19728021174669266,
|
|
"num_tokens": 13584346.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.384568691253662,
|
|
"epoch": 5.963198546115402,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002353414638899447,
|
|
"loss": 4.8813,
|
|
"mean_token_accuracy": 0.19869646281003953,
|
|
"num_tokens": 13595713.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.2159326553344725,
|
|
"epoch": 5.967741935483871,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023499361720882468,
|
|
"loss": 4.7589,
|
|
"mean_token_accuracy": 0.21230109184980392,
|
|
"num_tokens": 13605065.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.319033575057984,
|
|
"epoch": 5.97228532485234,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00023464586923947718,
|
|
"loss": 4.9144,
|
|
"mean_token_accuracy": 0.20242830663919448,
|
|
"num_tokens": 13615028.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.265029191970825,
|
|
"epoch": 5.976828714220809,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00023429822083993596,
|
|
"loss": 4.8356,
|
|
"mean_token_accuracy": 0.1984158307313919,
|
|
"num_tokens": 13625752.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.318032884597779,
|
|
"epoch": 5.981372103589278,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00023395067286798872,
|
|
"loss": 4.8091,
|
|
"mean_token_accuracy": 0.20653698444366456,
|
|
"num_tokens": 13635094.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.3025062084198,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00023360322618117563,
|
|
"loss": 4.8685,
|
|
"mean_token_accuracy": 0.20581616908311845,
|
|
"num_tokens": 13645007.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.286645317077637,
|
|
"epoch": 5.990458882326215,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023325588163678674,
|
|
"loss": 4.7676,
|
|
"mean_token_accuracy": 0.21303020417690277,
|
|
"num_tokens": 13655101.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.2673787593841555,
|
|
"epoch": 5.995002271694684,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00023290864009186036,
|
|
"loss": 4.8294,
|
|
"mean_token_accuracy": 0.20636545568704606,
|
|
"num_tokens": 13665289.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.290776157379151,
|
|
"epoch": 5.999545661063153,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00023256150240318025,
|
|
"loss": 4.9462,
|
|
"mean_token_accuracy": 0.2008124202489853,
|
|
"num_tokens": 13675384.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.346864170498318,
|
|
"epoch": 6.003634711494775,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023221446942727438,
|
|
"loss": 4.8289,
|
|
"mean_token_accuracy": 0.21045461297035217,
|
|
"num_tokens": 13683918.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.3615334033966064,
|
|
"epoch": 6.008178100863244,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002318675420204119,
|
|
"loss": 4.7623,
|
|
"mean_token_accuracy": 0.20907092094421387,
|
|
"num_tokens": 13693999.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.375178813934326,
|
|
"epoch": 6.012721490231713,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023152072103860188,
|
|
"loss": 4.8091,
|
|
"mean_token_accuracy": 0.2048986852169037,
|
|
"num_tokens": 13705465.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.272132492065429,
|
|
"epoch": 6.017264879600182,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023117400733759053,
|
|
"loss": 4.6866,
|
|
"mean_token_accuracy": 0.21176927238702775,
|
|
"num_tokens": 13715858.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.345735549926758,
|
|
"epoch": 6.021808268968651,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002308274017728595,
|
|
"loss": 4.7316,
|
|
"mean_token_accuracy": 0.21593870520591735,
|
|
"num_tokens": 13725639.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.228486251831055,
|
|
"epoch": 6.026351658337119,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002304809051996235,
|
|
"loss": 4.6391,
|
|
"mean_token_accuracy": 0.2174166262149811,
|
|
"num_tokens": 13736015.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.189400339126587,
|
|
"epoch": 6.030895047705588,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002301345184728284,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.2151237577199936,
|
|
"num_tokens": 13745433.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.336166906356811,
|
|
"epoch": 6.035438437074057,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00022978824244714903,
|
|
"loss": 4.7072,
|
|
"mean_token_accuracy": 0.22068069130182266,
|
|
"num_tokens": 13754970.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.300994873046875,
|
|
"epoch": 6.039981826442526,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00022944207797698707,
|
|
"loss": 4.7993,
|
|
"mean_token_accuracy": 0.20574731826782228,
|
|
"num_tokens": 13766356.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.290626859664917,
|
|
"epoch": 6.044525215810995,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002290960259164689,
|
|
"loss": 4.7635,
|
|
"mean_token_accuracy": 0.208756160736084,
|
|
"num_tokens": 13776503.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.33978304862976,
|
|
"epoch": 6.049068605179464,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022875008711944355,
|
|
"loss": 4.7686,
|
|
"mean_token_accuracy": 0.2040879398584366,
|
|
"num_tokens": 13787918.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.221943283081055,
|
|
"epoch": 6.053611994547933,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022840426243948077,
|
|
"loss": 4.7126,
|
|
"mean_token_accuracy": 0.2050175592303276,
|
|
"num_tokens": 13798305.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.321598243713379,
|
|
"epoch": 6.058155383916402,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00022805855272986842,
|
|
"loss": 4.7281,
|
|
"mean_token_accuracy": 0.2086731180548668,
|
|
"num_tokens": 13808867.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.294652652740479,
|
|
"epoch": 6.062698773284871,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00022771295884361088,
|
|
"loss": 4.6109,
|
|
"mean_token_accuracy": 0.22339267134666443,
|
|
"num_tokens": 13820073.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.3435241222381595,
|
|
"epoch": 6.067242162653339,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002273674816334268,
|
|
"loss": 4.7679,
|
|
"mean_token_accuracy": 0.21312017142772674,
|
|
"num_tokens": 13830318.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.207389497756958,
|
|
"epoch": 6.071785552021808,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00022702212195174664,
|
|
"loss": 4.7,
|
|
"mean_token_accuracy": 0.21062860786914825,
|
|
"num_tokens": 13840847.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.267226934432983,
|
|
"epoch": 6.076328941390277,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00022667688065071146,
|
|
"loss": 4.7101,
|
|
"mean_token_accuracy": 0.20911791622638704,
|
|
"num_tokens": 13850478.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.30379810333252,
|
|
"epoch": 6.080872330758746,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022633175858216936,
|
|
"loss": 4.7739,
|
|
"mean_token_accuracy": 0.20125263780355454,
|
|
"num_tokens": 13860749.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.322433710098267,
|
|
"epoch": 6.085415720127215,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00022598675659767515,
|
|
"loss": 4.7699,
|
|
"mean_token_accuracy": 0.2049841672182083,
|
|
"num_tokens": 13869909.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.313140201568603,
|
|
"epoch": 6.089959109495684,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022564187554848663,
|
|
"loss": 4.695,
|
|
"mean_token_accuracy": 0.21080277115106583,
|
|
"num_tokens": 13880235.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.203154420852661,
|
|
"epoch": 6.094502498864153,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00022529711628556387,
|
|
"loss": 4.6208,
|
|
"mean_token_accuracy": 0.21767460703849792,
|
|
"num_tokens": 13890324.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.304999971389771,
|
|
"epoch": 6.099045888232622,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022495247965956567,
|
|
"loss": 4.7369,
|
|
"mean_token_accuracy": 0.20778287053108216,
|
|
"num_tokens": 13900250.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.281234264373779,
|
|
"epoch": 6.10358927760109,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002246079665208489,
|
|
"loss": 4.6945,
|
|
"mean_token_accuracy": 0.2069418653845787,
|
|
"num_tokens": 13909736.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.274271678924561,
|
|
"epoch": 6.108132666969559,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00022426357771946544,
|
|
"loss": 4.7405,
|
|
"mean_token_accuracy": 0.21005418449640273,
|
|
"num_tokens": 13920187.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.289166021347046,
|
|
"epoch": 6.112676056338028,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 0.00022391931410516036,
|
|
"loss": 4.7049,
|
|
"mean_token_accuracy": 0.20222811698913573,
|
|
"num_tokens": 13930858.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.230607366561889,
|
|
"epoch": 6.117219445706497,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00022357517652736992,
|
|
"loss": 4.6338,
|
|
"mean_token_accuracy": 0.2193095266819,
|
|
"num_tokens": 13940771.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.291524362564087,
|
|
"epoch": 6.121762835074966,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00022323116583521948,
|
|
"loss": 4.8052,
|
|
"mean_token_accuracy": 0.20064616501331328,
|
|
"num_tokens": 13951312.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.317718648910523,
|
|
"epoch": 6.126306224443435,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00022288728287752109,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.2047889530658722,
|
|
"num_tokens": 13960862.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.295494413375854,
|
|
"epoch": 6.130849613811904,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002225435285027719,
|
|
"loss": 4.7132,
|
|
"mean_token_accuracy": 0.2094593569636345,
|
|
"num_tokens": 13972228.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.268077564239502,
|
|
"epoch": 6.135393003180373,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00022219990355915148,
|
|
"loss": 4.7005,
|
|
"mean_token_accuracy": 0.2108682230114937,
|
|
"num_tokens": 13982977.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.347118330001831,
|
|
"epoch": 6.139936392548841,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00022185640889452036,
|
|
"loss": 4.6699,
|
|
"mean_token_accuracy": 0.2230816215276718,
|
|
"num_tokens": 13991645.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.336764669418335,
|
|
"epoch": 6.14447978191731,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00022151304535641747,
|
|
"loss": 4.733,
|
|
"mean_token_accuracy": 0.2041908323764801,
|
|
"num_tokens": 14001313.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.2592137336730955,
|
|
"epoch": 6.149023171285779,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002211698137920582,
|
|
"loss": 4.7455,
|
|
"mean_token_accuracy": 0.2111082211136818,
|
|
"num_tokens": 14011248.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.282375907897949,
|
|
"epoch": 6.153566560654248,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00022082671504833223,
|
|
"loss": 4.726,
|
|
"mean_token_accuracy": 0.20515696555376053,
|
|
"num_tokens": 14022290.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.281827068328857,
|
|
"epoch": 6.158109950022717,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00022048374997180175,
|
|
"loss": 4.6991,
|
|
"mean_token_accuracy": 0.2099715143442154,
|
|
"num_tokens": 14032899.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.2815008640289305,
|
|
"epoch": 6.162653339391186,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00022014091940869885,
|
|
"loss": 4.707,
|
|
"mean_token_accuracy": 0.21697249710559846,
|
|
"num_tokens": 14043065.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.265380001068115,
|
|
"epoch": 6.167196728759655,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00021979822420492402,
|
|
"loss": 4.7737,
|
|
"mean_token_accuracy": 0.20450155436992645,
|
|
"num_tokens": 14054594.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.383376169204712,
|
|
"epoch": 6.171740118128124,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00021945566520604337,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.20670277178287505,
|
|
"num_tokens": 14064211.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.287307167053223,
|
|
"epoch": 6.176283507496592,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002191132432572875,
|
|
"loss": 4.7248,
|
|
"mean_token_accuracy": 0.21948544383049012,
|
|
"num_tokens": 14073946.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.230969095230103,
|
|
"epoch": 6.180826896865061,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021877095920354818,
|
|
"loss": 4.6331,
|
|
"mean_token_accuracy": 0.2169025331735611,
|
|
"num_tokens": 14083303.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.282938718795776,
|
|
"epoch": 6.18537028623353,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00021842881388937764,
|
|
"loss": 4.6652,
|
|
"mean_token_accuracy": 0.21995976120233535,
|
|
"num_tokens": 14093475.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.369122076034546,
|
|
"epoch": 6.189913675601999,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00021808680815898508,
|
|
"loss": 4.7995,
|
|
"mean_token_accuracy": 0.2039106696844101,
|
|
"num_tokens": 14104117.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.216143083572388,
|
|
"epoch": 6.194457064970468,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021774494285623598,
|
|
"loss": 4.6655,
|
|
"mean_token_accuracy": 0.20973755270242692,
|
|
"num_tokens": 14115480.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.217545843124389,
|
|
"epoch": 6.199000454338937,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002174032188246487,
|
|
"loss": 4.6714,
|
|
"mean_token_accuracy": 0.21054930090904236,
|
|
"num_tokens": 14125827.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.22686128616333,
|
|
"epoch": 6.203543843707406,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00021706163690739344,
|
|
"loss": 4.7282,
|
|
"mean_token_accuracy": 0.21306169480085374,
|
|
"num_tokens": 14136424.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.347414541244507,
|
|
"epoch": 6.208087233075875,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00021672019794728963,
|
|
"loss": 4.7927,
|
|
"mean_token_accuracy": 0.2028033912181854,
|
|
"num_tokens": 14147478.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.359594249725342,
|
|
"epoch": 6.212630622444343,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00021637890278680406,
|
|
"loss": 4.7982,
|
|
"mean_token_accuracy": 0.21026035249233246,
|
|
"num_tokens": 14159044.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.297825002670288,
|
|
"epoch": 6.217174011812812,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00021603775226804845,
|
|
"loss": 4.745,
|
|
"mean_token_accuracy": 0.2087731823325157,
|
|
"num_tokens": 14169370.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.271664667129516,
|
|
"epoch": 6.221717401181281,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00021569674723277783,
|
|
"loss": 4.7559,
|
|
"mean_token_accuracy": 0.20879374593496322,
|
|
"num_tokens": 14179774.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.275359582901001,
|
|
"epoch": 6.22626079054975,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00021535588852238823,
|
|
"loss": 4.7422,
|
|
"mean_token_accuracy": 0.21396408081054688,
|
|
"num_tokens": 14190182.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.263077926635742,
|
|
"epoch": 6.230804179918219,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021501517697791479,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.21317874491214753,
|
|
"num_tokens": 14199505.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.306978368759156,
|
|
"epoch": 6.235347569286688,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002146746134400291,
|
|
"loss": 4.7549,
|
|
"mean_token_accuracy": 0.20518312603235245,
|
|
"num_tokens": 14209677.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.251652526855469,
|
|
"epoch": 6.239890958655157,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021433419874903803,
|
|
"loss": 4.7013,
|
|
"mean_token_accuracy": 0.2124251276254654,
|
|
"num_tokens": 14219356.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.217204189300537,
|
|
"epoch": 6.244434348023626,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.00021399393374488097,
|
|
"loss": 4.6329,
|
|
"mean_token_accuracy": 0.22693775296211244,
|
|
"num_tokens": 14227992.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.285088777542114,
|
|
"epoch": 6.248977737392094,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00021365381926712797,
|
|
"loss": 4.6988,
|
|
"mean_token_accuracy": 0.21473327428102493,
|
|
"num_tokens": 14238354.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.3047009944915775,
|
|
"epoch": 6.253521126760563,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021331385615497756,
|
|
"loss": 4.7596,
|
|
"mean_token_accuracy": 0.20504184067249298,
|
|
"num_tokens": 14249560.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.269008541107178,
|
|
"epoch": 6.258064516129032,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002129740452472552,
|
|
"loss": 4.7682,
|
|
"mean_token_accuracy": 0.2058011069893837,
|
|
"num_tokens": 14259372.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.256386709213257,
|
|
"epoch": 6.262607905497501,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021263438738241026,
|
|
"loss": 4.6899,
|
|
"mean_token_accuracy": 0.21625061631202697,
|
|
"num_tokens": 14270837.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.284019470214844,
|
|
"epoch": 6.26715129486597,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00021229488339851498,
|
|
"loss": 4.6805,
|
|
"mean_token_accuracy": 0.2135418564081192,
|
|
"num_tokens": 14280138.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.301987361907959,
|
|
"epoch": 6.271694684234439,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021195553413326146,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20581310689449311,
|
|
"num_tokens": 14290131.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.2550191402435305,
|
|
"epoch": 6.276238073602908,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002116163404239605,
|
|
"loss": 4.7402,
|
|
"mean_token_accuracy": 0.20606436282396318,
|
|
"num_tokens": 14299961.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.288954162597657,
|
|
"epoch": 6.280781462971377,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00021127730310753866,
|
|
"loss": 4.6972,
|
|
"mean_token_accuracy": 0.21024607867002487,
|
|
"num_tokens": 14310529.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.259493684768676,
|
|
"epoch": 6.285324852339846,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00021093842302053706,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.22121623754501343,
|
|
"num_tokens": 14320220.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.340792942047119,
|
|
"epoch": 6.289868241708314,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021059970099910835,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.210525743663311,
|
|
"num_tokens": 14331700.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.3008955955505375,
|
|
"epoch": 6.294411631076783,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00021026113787901568,
|
|
"loss": 4.7089,
|
|
"mean_token_accuracy": 0.21194118559360503,
|
|
"num_tokens": 14341163.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.264566230773926,
|
|
"epoch": 6.298955020445252,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020992273449562966,
|
|
"loss": 4.6997,
|
|
"mean_token_accuracy": 0.21229505985975267,
|
|
"num_tokens": 14351398.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.319871616363526,
|
|
"epoch": 6.303498409813721,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00020958449168392723,
|
|
"loss": 4.6973,
|
|
"mean_token_accuracy": 0.21329400092363357,
|
|
"num_tokens": 14361920.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.240673160552978,
|
|
"epoch": 6.30804179918219,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002092464102784888,
|
|
"loss": 4.7101,
|
|
"mean_token_accuracy": 0.21550750881433486,
|
|
"num_tokens": 14371980.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.239515781402588,
|
|
"epoch": 6.312585188550659,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020890849111349663,
|
|
"loss": 4.7514,
|
|
"mean_token_accuracy": 0.2096133127808571,
|
|
"num_tokens": 14381397.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.275849771499634,
|
|
"epoch": 6.317128577919128,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00020857073502273266,
|
|
"loss": 4.7675,
|
|
"mean_token_accuracy": 0.20995737612247467,
|
|
"num_tokens": 14393476.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.300877904891967,
|
|
"epoch": 6.321671967287597,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002082331428395765,
|
|
"loss": 4.7491,
|
|
"mean_token_accuracy": 0.21112029999494553,
|
|
"num_tokens": 14404801.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.310664176940918,
|
|
"epoch": 6.326215356656065,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00020789571539700326,
|
|
"loss": 4.7561,
|
|
"mean_token_accuracy": 0.21185617595911027,
|
|
"num_tokens": 14415921.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.25648045539856,
|
|
"epoch": 6.330758746024534,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020755845352758175,
|
|
"loss": 4.7689,
|
|
"mean_token_accuracy": 0.20812695771455764,
|
|
"num_tokens": 14426455.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.296489572525024,
|
|
"epoch": 6.335302135393003,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0002072213580634718,
|
|
"loss": 4.7469,
|
|
"mean_token_accuracy": 0.2141423836350441,
|
|
"num_tokens": 14438720.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.3253504753112795,
|
|
"epoch": 6.339845524761472,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002068844298364233,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.21010375022888184,
|
|
"num_tokens": 14449644.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.305712413787842,
|
|
"epoch": 6.344388914129941,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 0.00020654766967777288,
|
|
"loss": 4.7949,
|
|
"mean_token_accuracy": 0.2050855875015259,
|
|
"num_tokens": 14459712.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.2965716361999515,
|
|
"epoch": 6.34893230349841,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002062110784184431,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.205116169154644,
|
|
"num_tokens": 14469440.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.301122617721558,
|
|
"epoch": 6.353475692866879,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00020587465688893904,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.2024425059556961,
|
|
"num_tokens": 14480587.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.385530042648315,
|
|
"epoch": 6.358019082235348,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020553840591934765,
|
|
"loss": 4.7918,
|
|
"mean_token_accuracy": 0.21337463706731796,
|
|
"num_tokens": 14490360.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.358019082235348,
|
|
"eval_entropy": 5.258032752192298,
|
|
"eval_loss": 5.730602741241455,
|
|
"eval_mean_token_accuracy": 0.1684639310074407,
|
|
"eval_num_tokens": 14490360.0,
|
|
"eval_runtime": 2.7794,
|
|
"eval_samples_per_second": 1235.882,
|
|
"eval_steps_per_second": 154.71,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.2852222442626955,
|
|
"epoch": 6.362562471603816,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002052023263393346,
|
|
"loss": 4.7141,
|
|
"mean_token_accuracy": 0.21100748479366302,
|
|
"num_tokens": 14501135.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.25512466430664,
|
|
"epoch": 6.367105860972285,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00020486641897814295,
|
|
"loss": 4.7276,
|
|
"mean_token_accuracy": 0.20900886356830597,
|
|
"num_tokens": 14511828.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.256915950775147,
|
|
"epoch": 6.371649250340754,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020453068466459067,
|
|
"loss": 4.7451,
|
|
"mean_token_accuracy": 0.20772741734981537,
|
|
"num_tokens": 14521973.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.336220598220825,
|
|
"epoch": 6.376192639709223,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020419512422706893,
|
|
"loss": 4.7916,
|
|
"mean_token_accuracy": 0.21213590502738952,
|
|
"num_tokens": 14532515.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.333255815505981,
|
|
"epoch": 6.380736029077692,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020385973849353957,
|
|
"loss": 4.7131,
|
|
"mean_token_accuracy": 0.20591868311166764,
|
|
"num_tokens": 14542577.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.254451751708984,
|
|
"epoch": 6.385279418446161,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020352452829153369,
|
|
"loss": 4.6596,
|
|
"mean_token_accuracy": 0.21228098571300508,
|
|
"num_tokens": 14552472.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.214546394348145,
|
|
"epoch": 6.38982280781463,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000203189494448149,
|
|
"loss": 4.698,
|
|
"mean_token_accuracy": 0.21906617879867554,
|
|
"num_tokens": 14562615.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.257629919052124,
|
|
"epoch": 6.394366197183099,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00020285463779004837,
|
|
"loss": 4.716,
|
|
"mean_token_accuracy": 0.21322744637727736,
|
|
"num_tokens": 14573200.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.345214462280273,
|
|
"epoch": 6.398909586551568,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020251995914345722,
|
|
"loss": 4.7414,
|
|
"mean_token_accuracy": 0.21387434005737305,
|
|
"num_tokens": 14583042.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.366867351531982,
|
|
"epoch": 6.403452975920036,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020218545933416181,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.2112393006682396,
|
|
"num_tokens": 14594409.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.3571319580078125,
|
|
"epoch": 6.407996365288505,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00020185113918750724,
|
|
"loss": 4.8128,
|
|
"mean_token_accuracy": 0.21132265031337738,
|
|
"num_tokens": 14605370.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.309704399108886,
|
|
"epoch": 6.412539754656974,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002015169995283952,
|
|
"loss": 4.7515,
|
|
"mean_token_accuracy": 0.20657599121332168,
|
|
"num_tokens": 14615056.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.384037017822266,
|
|
"epoch": 6.417083144025443,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020118304118128211,
|
|
"loss": 4.8176,
|
|
"mean_token_accuracy": 0.1990896135568619,
|
|
"num_tokens": 14624959.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.292597007751465,
|
|
"epoch": 6.421626533393912,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00020084926497017703,
|
|
"loss": 4.6955,
|
|
"mean_token_accuracy": 0.21632812172174454,
|
|
"num_tokens": 14635348.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.257171583175659,
|
|
"epoch": 6.426169922762381,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00020051567171863944,
|
|
"loss": 4.7274,
|
|
"mean_token_accuracy": 0.2084020420908928,
|
|
"num_tokens": 14645232.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.313946866989136,
|
|
"epoch": 6.43071331213085,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002001822622497777,
|
|
"loss": 4.7189,
|
|
"mean_token_accuracy": 0.21646155118942262,
|
|
"num_tokens": 14656194.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.321885108947754,
|
|
"epoch": 6.435256701499318,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00019984903738624632,
|
|
"loss": 4.749,
|
|
"mean_token_accuracy": 0.20974619090557098,
|
|
"num_tokens": 14667019.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.319108486175537,
|
|
"epoch": 6.439800090867787,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.00019951599795024478,
|
|
"loss": 4.7091,
|
|
"mean_token_accuracy": 0.21701431423425674,
|
|
"num_tokens": 14676339.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.312474155426026,
|
|
"epoch": 6.444343480236256,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019918314476351462,
|
|
"loss": 4.7647,
|
|
"mean_token_accuracy": 0.20827334821224214,
|
|
"num_tokens": 14688206.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.285890007019043,
|
|
"epoch": 6.448886869604725,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019885047864733803,
|
|
"loss": 4.7045,
|
|
"mean_token_accuracy": 0.21691632717847825,
|
|
"num_tokens": 14698397.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.291223955154419,
|
|
"epoch": 6.453430258973194,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00019851800042253553,
|
|
"loss": 4.6779,
|
|
"mean_token_accuracy": 0.21240607500076295,
|
|
"num_tokens": 14708140.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.280487298965454,
|
|
"epoch": 6.457973648341663,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001981857109094642,
|
|
"loss": 4.7218,
|
|
"mean_token_accuracy": 0.21150214970111847,
|
|
"num_tokens": 14718710.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.261139822006226,
|
|
"epoch": 6.462517037710132,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019785361092801524,
|
|
"loss": 4.763,
|
|
"mean_token_accuracy": 0.20787553042173385,
|
|
"num_tokens": 14729143.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.306623744964599,
|
|
"epoch": 6.467060427078601,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019752170129761244,
|
|
"loss": 4.7502,
|
|
"mean_token_accuracy": 0.21490464806556703,
|
|
"num_tokens": 14739386.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.285168075561524,
|
|
"epoch": 6.47160381644707,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0001971899828372098,
|
|
"loss": 4.7606,
|
|
"mean_token_accuracy": 0.20895636528730394,
|
|
"num_tokens": 14749209.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.316984796524048,
|
|
"epoch": 6.476147205815538,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019685845636528972,
|
|
"loss": 4.7501,
|
|
"mean_token_accuracy": 0.21787956953048707,
|
|
"num_tokens": 14759351.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.318292474746704,
|
|
"epoch": 6.480690595184007,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019652712269986074,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.2124994993209839,
|
|
"num_tokens": 14769069.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.339407444000244,
|
|
"epoch": 6.485233984552476,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019619598265845556,
|
|
"loss": 4.7499,
|
|
"mean_token_accuracy": 0.21520694047212602,
|
|
"num_tokens": 14778563.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.261417102813721,
|
|
"epoch": 6.489777373920945,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00019586503705812969,
|
|
"loss": 4.7178,
|
|
"mean_token_accuracy": 0.21582437455654144,
|
|
"num_tokens": 14788305.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.295353937149048,
|
|
"epoch": 6.494320763289414,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001955342867154582,
|
|
"loss": 4.7875,
|
|
"mean_token_accuracy": 0.20365089178085327,
|
|
"num_tokens": 14799463.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.205623722076416,
|
|
"epoch": 6.498864152657883,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00019520373244653484,
|
|
"loss": 4.6793,
|
|
"mean_token_accuracy": 0.22212345600128175,
|
|
"num_tokens": 14809516.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.307696151733398,
|
|
"epoch": 6.503407542026352,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0001948733750669693,
|
|
"loss": 4.7877,
|
|
"mean_token_accuracy": 0.20058375746011733,
|
|
"num_tokens": 14819605.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.319727420806885,
|
|
"epoch": 6.50795093139482,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019454321539188568,
|
|
"loss": 4.7312,
|
|
"mean_token_accuracy": 0.20737345665693283,
|
|
"num_tokens": 14830241.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.326238775253296,
|
|
"epoch": 6.512494320763289,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00019421325423592007,
|
|
"loss": 4.754,
|
|
"mean_token_accuracy": 0.21210706233978271,
|
|
"num_tokens": 14840294.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.373597097396851,
|
|
"epoch": 6.517037710131758,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001938834924132189,
|
|
"loss": 4.7839,
|
|
"mean_token_accuracy": 0.2110060691833496,
|
|
"num_tokens": 14850156.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.260190820693969,
|
|
"epoch": 6.521581099500227,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001935539307374366,
|
|
"loss": 4.684,
|
|
"mean_token_accuracy": 0.2206605702638626,
|
|
"num_tokens": 14859889.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.325161647796631,
|
|
"epoch": 6.526124488868696,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019322457002173395,
|
|
"loss": 4.7603,
|
|
"mean_token_accuracy": 0.21116570383310318,
|
|
"num_tokens": 14871659.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.309049510955811,
|
|
"epoch": 6.530667878237165,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019289541107877562,
|
|
"loss": 4.74,
|
|
"mean_token_accuracy": 0.21211667805910112,
|
|
"num_tokens": 14882041.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.301411581039429,
|
|
"epoch": 6.535211267605634,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019256645472072876,
|
|
"loss": 4.6901,
|
|
"mean_token_accuracy": 0.21909360140562056,
|
|
"num_tokens": 14891709.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.318630838394165,
|
|
"epoch": 6.539754656974103,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019223770175926032,
|
|
"loss": 4.7497,
|
|
"mean_token_accuracy": 0.21233847141265869,
|
|
"num_tokens": 14901898.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.262526988983154,
|
|
"epoch": 6.544298046342572,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00019190915300553574,
|
|
"loss": 4.6759,
|
|
"mean_token_accuracy": 0.2192450448870659,
|
|
"num_tokens": 14911910.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.281016540527344,
|
|
"epoch": 6.54884143571104,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.0001915808092702162,
|
|
"loss": 4.706,
|
|
"mean_token_accuracy": 0.20835800915956498,
|
|
"num_tokens": 14922055.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.247645854949951,
|
|
"epoch": 6.553384825079509,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0001912526713634574,
|
|
"loss": 4.6861,
|
|
"mean_token_accuracy": 0.21840503811836243,
|
|
"num_tokens": 14931841.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.274683284759521,
|
|
"epoch": 6.557928214447978,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001909247400949069,
|
|
"loss": 4.7289,
|
|
"mean_token_accuracy": 0.21285527050495148,
|
|
"num_tokens": 14942356.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.266970252990722,
|
|
"epoch": 6.562471603816447,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00019059701627370263,
|
|
"loss": 4.6977,
|
|
"mean_token_accuracy": 0.21089812815189363,
|
|
"num_tokens": 14952552.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.242506122589111,
|
|
"epoch": 6.567014993184916,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019026950070847025,
|
|
"loss": 4.705,
|
|
"mean_token_accuracy": 0.21021936535835267,
|
|
"num_tokens": 14962728.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.250985813140869,
|
|
"epoch": 6.571558382553385,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00018994219420732228,
|
|
"loss": 4.7034,
|
|
"mean_token_accuracy": 0.21966067999601363,
|
|
"num_tokens": 14974071.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.311672115325928,
|
|
"epoch": 6.576101771921854,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018961509757785462,
|
|
"loss": 4.7996,
|
|
"mean_token_accuracy": 0.2122173547744751,
|
|
"num_tokens": 14983567.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.2224791049957275,
|
|
"epoch": 6.580645161290323,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 0.00018928821162714587,
|
|
"loss": 4.7171,
|
|
"mean_token_accuracy": 0.21757151186466217,
|
|
"num_tokens": 14994172.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.297982025146484,
|
|
"epoch": 6.585188550658792,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00018896153716175458,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.21487865298986436,
|
|
"num_tokens": 15004094.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.291853046417236,
|
|
"epoch": 6.58973194002726,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001886350749877175,
|
|
"loss": 4.7131,
|
|
"mean_token_accuracy": 0.21426960080862045,
|
|
"num_tokens": 15013986.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.289523553848267,
|
|
"epoch": 6.594275329395729,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 0.0001883088259105476,
|
|
"loss": 4.7314,
|
|
"mean_token_accuracy": 0.21221622377634047,
|
|
"num_tokens": 15024764.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.318004179000854,
|
|
"epoch": 6.598818718764198,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00018798279073523204,
|
|
"loss": 4.7526,
|
|
"mean_token_accuracy": 0.20937420129776002,
|
|
"num_tokens": 15034971.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.310562181472778,
|
|
"epoch": 6.603362108132667,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00018765697026623017,
|
|
"loss": 4.8214,
|
|
"mean_token_accuracy": 0.20244472920894624,
|
|
"num_tokens": 15045750.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.315129804611206,
|
|
"epoch": 6.607905497501136,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0001873313653074718,
|
|
"loss": 4.8525,
|
|
"mean_token_accuracy": 0.1981658011674881,
|
|
"num_tokens": 15056840.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.240451145172119,
|
|
"epoch": 6.612448886869605,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00018700597666235452,
|
|
"loss": 4.6997,
|
|
"mean_token_accuracy": 0.21605789959430693,
|
|
"num_tokens": 15067451.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.305985069274902,
|
|
"epoch": 6.616992276238074,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018668080513374268,
|
|
"loss": 4.7388,
|
|
"mean_token_accuracy": 0.21475727260112762,
|
|
"num_tokens": 15077780.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.3194413661956785,
|
|
"epoch": 6.621535665606542,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018635585152396457,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.19901832491159438,
|
|
"num_tokens": 15089100.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.239009428024292,
|
|
"epoch": 6.626079054975011,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000186031116634811,
|
|
"loss": 4.73,
|
|
"mean_token_accuracy": 0.212596794962883,
|
|
"num_tokens": 15099420.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.307452154159546,
|
|
"epoch": 6.63062244434348,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018570660126753291,
|
|
"loss": 4.7435,
|
|
"mean_token_accuracy": 0.21014348268508912,
|
|
"num_tokens": 15109884.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.272648334503174,
|
|
"epoch": 6.635165833711949,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001853823062228399,
|
|
"loss": 4.6908,
|
|
"mean_token_accuracy": 0.22021933943033217,
|
|
"num_tokens": 15119761.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.252042055130005,
|
|
"epoch": 6.639709223080418,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018505823230089758,
|
|
"loss": 4.6681,
|
|
"mean_token_accuracy": 0.22589696943759918,
|
|
"num_tokens": 15129159.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.267010021209717,
|
|
"epoch": 6.644252612448887,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018473438030132633,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.21447397023439407,
|
|
"num_tokens": 15138855.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.281128358840943,
|
|
"epoch": 6.648796001817356,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001844107510231985,
|
|
"loss": 4.7182,
|
|
"mean_token_accuracy": 0.21049480140209198,
|
|
"num_tokens": 15149957.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.281608819961548,
|
|
"epoch": 6.653339391185825,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00018408734526503744,
|
|
"loss": 4.728,
|
|
"mean_token_accuracy": 0.21335206180810928,
|
|
"num_tokens": 15159110.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.300084638595581,
|
|
"epoch": 6.657882780554294,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018376416382481442,
|
|
"loss": 4.7943,
|
|
"mean_token_accuracy": 0.2041315346956253,
|
|
"num_tokens": 15169444.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.403600645065308,
|
|
"epoch": 6.662426169922762,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018344120749994787,
|
|
"loss": 4.9205,
|
|
"mean_token_accuracy": 0.18900640159845353,
|
|
"num_tokens": 15180246.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.245440435409546,
|
|
"epoch": 6.666969559291231,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00018311847708730005,
|
|
"loss": 4.7024,
|
|
"mean_token_accuracy": 0.21034331023693084,
|
|
"num_tokens": 15191578.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.226965618133545,
|
|
"epoch": 6.6715129486597,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0001827959733831765,
|
|
"loss": 4.7467,
|
|
"mean_token_accuracy": 0.20536126643419267,
|
|
"num_tokens": 15202380.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.213947200775147,
|
|
"epoch": 6.676056338028169,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00018247369718332274,
|
|
"loss": 4.7303,
|
|
"mean_token_accuracy": 0.21215869784355162,
|
|
"num_tokens": 15214073.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.261500549316406,
|
|
"epoch": 6.680599727396638,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018215164928292354,
|
|
"loss": 4.6588,
|
|
"mean_token_accuracy": 0.21820634007453918,
|
|
"num_tokens": 15223655.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.346786165237427,
|
|
"epoch": 6.685143116765107,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00018182983047659984,
|
|
"loss": 4.8423,
|
|
"mean_token_accuracy": 0.20172565132379533,
|
|
"num_tokens": 15235567.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.29037652015686,
|
|
"epoch": 6.689686506133576,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00018150824155840778,
|
|
"loss": 4.7286,
|
|
"mean_token_accuracy": 0.2081035077571869,
|
|
"num_tokens": 15246349.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.3224951267242435,
|
|
"epoch": 6.694229895502044,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001811868833218358,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.21410706341266633,
|
|
"num_tokens": 15256443.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.2827043533325195,
|
|
"epoch": 6.698773284870513,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00018086575655980376,
|
|
"loss": 4.7405,
|
|
"mean_token_accuracy": 0.21278878003358842,
|
|
"num_tokens": 15267392.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.2779968738555905,
|
|
"epoch": 6.703316674238982,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018054486206465952,
|
|
"loss": 4.6845,
|
|
"mean_token_accuracy": 0.2189563751220703,
|
|
"num_tokens": 15277371.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.318066549301148,
|
|
"epoch": 6.707860063607451,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001802242006281788,
|
|
"loss": 4.6968,
|
|
"mean_token_accuracy": 0.2128233179450035,
|
|
"num_tokens": 15287490.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.242515802383423,
|
|
"epoch": 6.71240345297592,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00017990377304156165,
|
|
"loss": 4.6876,
|
|
"mean_token_accuracy": 0.21272583305835724,
|
|
"num_tokens": 15298061.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.233055210113525,
|
|
"epoch": 6.716946842344389,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0001795835800954312,
|
|
"loss": 4.6874,
|
|
"mean_token_accuracy": 0.21527092456817626,
|
|
"num_tokens": 15307862.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.206230878829956,
|
|
"epoch": 6.721490231712858,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017926362257983174,
|
|
"loss": 4.7464,
|
|
"mean_token_accuracy": 0.21387675255537034,
|
|
"num_tokens": 15317693.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.300540733337402,
|
|
"epoch": 6.726033621081327,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001789439012842269,
|
|
"loss": 4.7905,
|
|
"mean_token_accuracy": 0.21013894826173782,
|
|
"num_tokens": 15329000.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.330814170837402,
|
|
"epoch": 6.730577010449796,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017862441699749687,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.2140057072043419,
|
|
"num_tokens": 15339967.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.3231593608856205,
|
|
"epoch": 6.735120399818264,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001783051705079375,
|
|
"loss": 4.7837,
|
|
"mean_token_accuracy": 0.2090545892715454,
|
|
"num_tokens": 15350478.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.313417196273804,
|
|
"epoch": 6.739663789186733,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017798616260325767,
|
|
"loss": 4.7155,
|
|
"mean_token_accuracy": 0.2115663021802902,
|
|
"num_tokens": 15360728.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.306729650497436,
|
|
"epoch": 6.744207178555202,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001776673940705778,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.21269436329603195,
|
|
"num_tokens": 15370689.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.234181451797485,
|
|
"epoch": 6.748750567923671,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001773488656964276,
|
|
"loss": 4.6486,
|
|
"mean_token_accuracy": 0.21799594163894653,
|
|
"num_tokens": 15380987.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.321646404266358,
|
|
"epoch": 6.75329395729214,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000177030578266744,
|
|
"loss": 4.7481,
|
|
"mean_token_accuracy": 0.21698492467403413,
|
|
"num_tokens": 15391989.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.339909076690674,
|
|
"epoch": 6.757837346660609,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017671253256686963,
|
|
"loss": 4.768,
|
|
"mean_token_accuracy": 0.20847346484661103,
|
|
"num_tokens": 15402376.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.293379402160644,
|
|
"epoch": 6.762380736029078,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00017639472938155075,
|
|
"loss": 4.7147,
|
|
"mean_token_accuracy": 0.2080901011824608,
|
|
"num_tokens": 15413034.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.222148752212524,
|
|
"epoch": 6.766924125397547,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017607716949493514,
|
|
"loss": 4.6775,
|
|
"mean_token_accuracy": 0.21014216542243958,
|
|
"num_tokens": 15423378.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.2099062442779545,
|
|
"epoch": 6.771467514766016,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017575985369057008,
|
|
"loss": 4.6684,
|
|
"mean_token_accuracy": 0.21039254814386368,
|
|
"num_tokens": 15434092.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.232883596420288,
|
|
"epoch": 6.776010904134484,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001754427827514008,
|
|
"loss": 4.6938,
|
|
"mean_token_accuracy": 0.20632888227701188,
|
|
"num_tokens": 15445178.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.311169290542603,
|
|
"epoch": 6.780554293502953,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017512595745976855,
|
|
"loss": 4.7492,
|
|
"mean_token_accuracy": 0.20916635990142823,
|
|
"num_tokens": 15455456.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.260423374176026,
|
|
"epoch": 6.785097682871422,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00017480937859740793,
|
|
"loss": 4.7146,
|
|
"mean_token_accuracy": 0.22146572023630143,
|
|
"num_tokens": 15465805.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.31777057647705,
|
|
"epoch": 6.789641072239891,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017449304694544608,
|
|
"loss": 4.7642,
|
|
"mean_token_accuracy": 0.20830376893281938,
|
|
"num_tokens": 15476590.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.2772773742675785,
|
|
"epoch": 6.79418446160836,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017417696328439974,
|
|
"loss": 4.6975,
|
|
"mean_token_accuracy": 0.2164659842848778,
|
|
"num_tokens": 15486791.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.288469028472901,
|
|
"epoch": 6.798727850976829,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017386112839417402,
|
|
"loss": 4.704,
|
|
"mean_token_accuracy": 0.21963360756635666,
|
|
"num_tokens": 15497116.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.293762302398681,
|
|
"epoch": 6.803271240345298,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00017354554305405994,
|
|
"loss": 4.7101,
|
|
"mean_token_accuracy": 0.21464851051568984,
|
|
"num_tokens": 15506731.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.2453293800354,
|
|
"epoch": 6.807814629713766,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0001732302080427332,
|
|
"loss": 4.7741,
|
|
"mean_token_accuracy": 0.20957263857126235,
|
|
"num_tokens": 15517333.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.235774850845337,
|
|
"epoch": 6.812358019082235,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017291512413825148,
|
|
"loss": 4.7177,
|
|
"mean_token_accuracy": 0.21368687450885773,
|
|
"num_tokens": 15526995.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.812358019082235,
|
|
"eval_entropy": 5.185723092944123,
|
|
"eval_loss": 5.714527606964111,
|
|
"eval_mean_token_accuracy": 0.17060264421756877,
|
|
"eval_num_tokens": 15526995.0,
|
|
"eval_runtime": 2.6069,
|
|
"eval_samples_per_second": 1317.637,
|
|
"eval_steps_per_second": 164.944,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.258687543869018,
|
|
"epoch": 6.816901408450704,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00017260029211805297,
|
|
"loss": 4.6905,
|
|
"mean_token_accuracy": 0.21180357187986373,
|
|
"num_tokens": 15536686.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.310452127456665,
|
|
"epoch": 6.821444797819173,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017228571275895434,
|
|
"loss": 4.833,
|
|
"mean_token_accuracy": 0.20401362627744674,
|
|
"num_tokens": 15547565.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.23850679397583,
|
|
"epoch": 6.825988187187642,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017197138683714902,
|
|
"loss": 4.6908,
|
|
"mean_token_accuracy": 0.21001136898994446,
|
|
"num_tokens": 15557962.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.272964239120483,
|
|
"epoch": 6.830531576556111,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001716573151282049,
|
|
"loss": 4.7117,
|
|
"mean_token_accuracy": 0.21743513345718385,
|
|
"num_tokens": 15567164.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.224982643127442,
|
|
"epoch": 6.83507496592458,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001713434984070627,
|
|
"loss": 4.6887,
|
|
"mean_token_accuracy": 0.21262873262166976,
|
|
"num_tokens": 15577917.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.293718767166138,
|
|
"epoch": 6.839618355293049,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.00017102993744803392,
|
|
"loss": 4.8002,
|
|
"mean_token_accuracy": 0.2025238648056984,
|
|
"num_tokens": 15589106.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.2875770092010494,
|
|
"epoch": 6.844161744661518,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00017071663302479923,
|
|
"loss": 4.7869,
|
|
"mean_token_accuracy": 0.20705512762069703,
|
|
"num_tokens": 15600368.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.280901336669922,
|
|
"epoch": 6.848705134029986,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017040358591040606,
|
|
"loss": 4.7012,
|
|
"mean_token_accuracy": 0.21470424383878708,
|
|
"num_tokens": 15610671.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.181329774856567,
|
|
"epoch": 6.853248523398455,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000170090796877267,
|
|
"loss": 4.6588,
|
|
"mean_token_accuracy": 0.2182561442255974,
|
|
"num_tokens": 15620403.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.304003429412842,
|
|
"epoch": 6.857791912766924,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00016977826669715783,
|
|
"loss": 4.7406,
|
|
"mean_token_accuracy": 0.21433935612440108,
|
|
"num_tokens": 15631260.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.303240203857422,
|
|
"epoch": 6.862335302135393,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000169465996141216,
|
|
"loss": 4.7173,
|
|
"mean_token_accuracy": 0.2133731573820114,
|
|
"num_tokens": 15641604.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.321141004562378,
|
|
"epoch": 6.866878691503862,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001691539859799377,
|
|
"loss": 4.7284,
|
|
"mean_token_accuracy": 0.20934505760669708,
|
|
"num_tokens": 15652469.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.265742206573487,
|
|
"epoch": 6.871422080872331,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00016884223698317722,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.20861562788486482,
|
|
"num_tokens": 15662615.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.253164577484131,
|
|
"epoch": 6.8759654702408,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001685307499201441,
|
|
"loss": 4.6904,
|
|
"mean_token_accuracy": 0.20715460777282715,
|
|
"num_tokens": 15674868.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.270483160018921,
|
|
"epoch": 6.880508859609268,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001682195255594019,
|
|
"loss": 4.7735,
|
|
"mean_token_accuracy": 0.215412537753582,
|
|
"num_tokens": 15686246.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.280123567581176,
|
|
"epoch": 6.885052248977737,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00016790856466886544,
|
|
"loss": 4.7123,
|
|
"mean_token_accuracy": 0.2133984610438347,
|
|
"num_tokens": 15697025.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.267392921447754,
|
|
"epoch": 6.889595638346206,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00016759786801580018,
|
|
"loss": 4.7658,
|
|
"mean_token_accuracy": 0.2093523845076561,
|
|
"num_tokens": 15706636.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.244884300231933,
|
|
"epoch": 6.894139027714675,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00016728743636681897,
|
|
"loss": 4.7148,
|
|
"mean_token_accuracy": 0.21684873700141907,
|
|
"num_tokens": 15717319.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.256495952606201,
|
|
"epoch": 6.898682417083144,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00016697727048788118,
|
|
"loss": 4.6514,
|
|
"mean_token_accuracy": 0.22218598276376725,
|
|
"num_tokens": 15726546.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.299419641494751,
|
|
"epoch": 6.903225806451613,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00016666737114429017,
|
|
"loss": 4.7527,
|
|
"mean_token_accuracy": 0.20740818977355957,
|
|
"num_tokens": 15737176.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.279237222671509,
|
|
"epoch": 6.907769195820082,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0001663577391006919,
|
|
"loss": 4.7317,
|
|
"mean_token_accuracy": 0.21953113675117492,
|
|
"num_tokens": 15747422.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.289186573028564,
|
|
"epoch": 6.912312585188551,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016604837512107263,
|
|
"loss": 4.7455,
|
|
"mean_token_accuracy": 0.2077797070145607,
|
|
"num_tokens": 15757591.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.25492844581604,
|
|
"epoch": 6.91685597455702,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00016573927996875714,
|
|
"loss": 4.8062,
|
|
"mean_token_accuracy": 0.20606701970100402,
|
|
"num_tokens": 15768375.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.248401689529419,
|
|
"epoch": 6.921399363925488,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016543045440640697,
|
|
"loss": 4.6676,
|
|
"mean_token_accuracy": 0.21786013692617417,
|
|
"num_tokens": 15778935.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.281639814376831,
|
|
"epoch": 6.925942753293957,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00016512189919601865,
|
|
"loss": 4.6833,
|
|
"mean_token_accuracy": 0.220657841861248,
|
|
"num_tokens": 15788955.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.293014192581177,
|
|
"epoch": 6.930486142662426,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001648136150989214,
|
|
"loss": 4.6651,
|
|
"mean_token_accuracy": 0.22223425209522246,
|
|
"num_tokens": 15798795.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.25290699005127,
|
|
"epoch": 6.935029532030895,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0001645056028757756,
|
|
"loss": 4.7441,
|
|
"mean_token_accuracy": 0.20813925117254256,
|
|
"num_tokens": 15809251.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.235751819610596,
|
|
"epoch": 6.939572921399364,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00016419786328657066,
|
|
"loss": 4.6592,
|
|
"mean_token_accuracy": 0.21716555953025818,
|
|
"num_tokens": 15819418.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.312016391754151,
|
|
"epoch": 6.944116310767833,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00016389039709062376,
|
|
"loss": 4.7216,
|
|
"mean_token_accuracy": 0.2051782101392746,
|
|
"num_tokens": 15829933.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.2460634231567385,
|
|
"epoch": 6.948659700136302,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016358320504657677,
|
|
"loss": 4.7435,
|
|
"mean_token_accuracy": 0.20967686921358109,
|
|
"num_tokens": 15841032.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.284748125076294,
|
|
"epoch": 6.95320308950477,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016327628791239576,
|
|
"loss": 4.7638,
|
|
"mean_token_accuracy": 0.20051346570253373,
|
|
"num_tokens": 15851126.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.300664663314819,
|
|
"epoch": 6.957746478873239,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016296964644536816,
|
|
"loss": 4.7328,
|
|
"mean_token_accuracy": 0.21026982218027115,
|
|
"num_tokens": 15861832.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.3288445472717285,
|
|
"epoch": 6.962289868241708,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001626632814021015,
|
|
"loss": 4.725,
|
|
"mean_token_accuracy": 0.21172333210706712,
|
|
"num_tokens": 15872295.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.306539869308471,
|
|
"epoch": 6.966833257610177,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00016235719353852079,
|
|
"loss": 4.7328,
|
|
"mean_token_accuracy": 0.2157417505979538,
|
|
"num_tokens": 15882453.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.258709907531738,
|
|
"epoch": 6.971376646978646,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00016205138360986753,
|
|
"loss": 4.7529,
|
|
"mean_token_accuracy": 0.20777132958173752,
|
|
"num_tokens": 15891962.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.262285232543945,
|
|
"epoch": 6.975920036347115,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00016174585237069738,
|
|
"loss": 4.648,
|
|
"mean_token_accuracy": 0.2163541555404663,
|
|
"num_tokens": 15902309.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.220811367034912,
|
|
"epoch": 6.980463425715584,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016144060057487824,
|
|
"loss": 4.6781,
|
|
"mean_token_accuracy": 0.21360816061496735,
|
|
"num_tokens": 15912789.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.300001430511474,
|
|
"epoch": 6.985006815084053,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016113562897558849,
|
|
"loss": 4.7716,
|
|
"mean_token_accuracy": 0.20962308794260026,
|
|
"num_tokens": 15922722.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.227821779251099,
|
|
"epoch": 6.989550204452522,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0001608309383253154,
|
|
"loss": 4.6477,
|
|
"mean_token_accuracy": 0.21954378485679626,
|
|
"num_tokens": 15933069.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.33433985710144,
|
|
"epoch": 6.99409359382099,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001605265293758528,
|
|
"loss": 4.7399,
|
|
"mean_token_accuracy": 0.20759457349777222,
|
|
"num_tokens": 15942427.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.2818702220916744,
|
|
"epoch": 6.998636983189459,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016022240287829947,
|
|
"loss": 4.7263,
|
|
"mean_token_accuracy": 0.20697252154350282,
|
|
"num_tokens": 15952765.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.235715919070774,
|
|
"epoch": 7.002726033621081,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015991855958305724,
|
|
"loss": 4.7137,
|
|
"mean_token_accuracy": 0.21992683079507616,
|
|
"num_tokens": 15963025.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.249240732192993,
|
|
"epoch": 7.00726942298955,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015961500023982945,
|
|
"loss": 4.6287,
|
|
"mean_token_accuracy": 0.22305368036031722,
|
|
"num_tokens": 15973425.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.297963953018188,
|
|
"epoch": 7.011812812358019,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015931172559761848,
|
|
"loss": 4.6478,
|
|
"mean_token_accuracy": 0.2256992757320404,
|
|
"num_tokens": 15984064.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.287467575073242,
|
|
"epoch": 7.016356201726488,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001590087364047243,
|
|
"loss": 4.6347,
|
|
"mean_token_accuracy": 0.2202056735754013,
|
|
"num_tokens": 15994512.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.278939723968506,
|
|
"epoch": 7.020899591094957,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015870603340874267,
|
|
"loss": 4.5643,
|
|
"mean_token_accuracy": 0.22739136815071107,
|
|
"num_tokens": 16004140.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.230801630020141,
|
|
"epoch": 7.025442980463426,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015840361735656333,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.22686371058225632,
|
|
"num_tokens": 16013710.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.320096111297607,
|
|
"epoch": 7.029986369831895,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00015810148899436747,
|
|
"loss": 4.6236,
|
|
"mean_token_accuracy": 0.2198963388800621,
|
|
"num_tokens": 16023846.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.2307733535766605,
|
|
"epoch": 7.034529759200363,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015779964906762706,
|
|
"loss": 4.5813,
|
|
"mean_token_accuracy": 0.22003311663866043,
|
|
"num_tokens": 16034327.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.25165491104126,
|
|
"epoch": 7.039073148568832,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015749809832110197,
|
|
"loss": 4.6293,
|
|
"mean_token_accuracy": 0.22432581186294556,
|
|
"num_tokens": 16045055.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.221637201309204,
|
|
"epoch": 7.043616537937301,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001571968374988389,
|
|
"loss": 4.6255,
|
|
"mean_token_accuracy": 0.21609024703502655,
|
|
"num_tokens": 16055198.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.2456306457519535,
|
|
"epoch": 7.04815992730577,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00015689586734416876,
|
|
"loss": 4.5765,
|
|
"mean_token_accuracy": 0.22491781860589982,
|
|
"num_tokens": 16065030.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.273325347900391,
|
|
"epoch": 7.052703316674239,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015659518859970545,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.22524583339691162,
|
|
"num_tokens": 16075974.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.27882227897644,
|
|
"epoch": 7.057246706042708,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001562948020073441,
|
|
"loss": 4.6808,
|
|
"mean_token_accuracy": 0.2168511226773262,
|
|
"num_tokens": 16086835.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.285020732879639,
|
|
"epoch": 7.061790095411177,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00015599470830825868,
|
|
"loss": 4.6528,
|
|
"mean_token_accuracy": 0.21559265106916428,
|
|
"num_tokens": 16098116.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.335413408279419,
|
|
"epoch": 7.066333484779646,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015569490824290054,
|
|
"loss": 4.6762,
|
|
"mean_token_accuracy": 0.20528496205806732,
|
|
"num_tokens": 16109185.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.248605155944825,
|
|
"epoch": 7.070876874148114,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00015539540255099643,
|
|
"loss": 4.6152,
|
|
"mean_token_accuracy": 0.22007495164871216,
|
|
"num_tokens": 16120025.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.2484081268310545,
|
|
"epoch": 7.075420263516583,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015509619197154705,
|
|
"loss": 4.5918,
|
|
"mean_token_accuracy": 0.2287656381726265,
|
|
"num_tokens": 16131145.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.303710222244263,
|
|
"epoch": 7.079963652885052,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015479727724282476,
|
|
"loss": 4.6623,
|
|
"mean_token_accuracy": 0.2188180074095726,
|
|
"num_tokens": 16141063.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.2515552043914795,
|
|
"epoch": 7.084507042253521,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00015449865910237184,
|
|
"loss": 4.6065,
|
|
"mean_token_accuracy": 0.2225065365433693,
|
|
"num_tokens": 16151059.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.250939607620239,
|
|
"epoch": 7.08905043162199,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015420033828699882,
|
|
"loss": 4.5883,
|
|
"mean_token_accuracy": 0.23163470029830932,
|
|
"num_tokens": 16160679.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.2595781803131105,
|
|
"epoch": 7.093593820990459,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001539023155327829,
|
|
"loss": 4.6203,
|
|
"mean_token_accuracy": 0.2160714477300644,
|
|
"num_tokens": 16170269.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.238500118255615,
|
|
"epoch": 7.098137210358928,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00015360459157506525,
|
|
"loss": 4.6017,
|
|
"mean_token_accuracy": 0.21592795699834824,
|
|
"num_tokens": 16180496.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.261360025405883,
|
|
"epoch": 7.102680599727397,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015330716714845048,
|
|
"loss": 4.6408,
|
|
"mean_token_accuracy": 0.2143355280160904,
|
|
"num_tokens": 16191043.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.325916528701782,
|
|
"epoch": 7.107223989095866,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001530100429868035,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.20944784581661224,
|
|
"num_tokens": 16201551.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.262737655639649,
|
|
"epoch": 7.111767378464334,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015271321982324892,
|
|
"loss": 4.6565,
|
|
"mean_token_accuracy": 0.20836976319551467,
|
|
"num_tokens": 16212928.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.259903144836426,
|
|
"epoch": 7.116310767832803,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015241669839016818,
|
|
"loss": 4.6231,
|
|
"mean_token_accuracy": 0.22414642125368117,
|
|
"num_tokens": 16223233.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.287717247009278,
|
|
"epoch": 7.120854157201272,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00015212047941919854,
|
|
"loss": 4.713,
|
|
"mean_token_accuracy": 0.21017347127199174,
|
|
"num_tokens": 16234902.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.271394968032837,
|
|
"epoch": 7.125397546569741,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015182456364123086,
|
|
"loss": 4.6866,
|
|
"mean_token_accuracy": 0.2182348147034645,
|
|
"num_tokens": 16244846.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.230439329147339,
|
|
"epoch": 7.12994093593821,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00015152895178640788,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.22170813530683517,
|
|
"num_tokens": 16255067.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.264108324050904,
|
|
"epoch": 7.134484325306679,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001512336445841224,
|
|
"loss": 4.6601,
|
|
"mean_token_accuracy": 0.213789926469326,
|
|
"num_tokens": 16265482.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.341061305999756,
|
|
"epoch": 7.139027714675148,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00015093864276301573,
|
|
"loss": 4.644,
|
|
"mean_token_accuracy": 0.2191776990890503,
|
|
"num_tokens": 16275319.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.255118465423584,
|
|
"epoch": 7.143571104043617,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00015064394705097546,
|
|
"loss": 4.6359,
|
|
"mean_token_accuracy": 0.2165455013513565,
|
|
"num_tokens": 16284889.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.311305379867553,
|
|
"epoch": 7.148114493412085,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015034955817513397,
|
|
"loss": 4.6664,
|
|
"mean_token_accuracy": 0.2179744929075241,
|
|
"num_tokens": 16295259.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.206708240509033,
|
|
"epoch": 7.152657882780554,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001500554768618665,
|
|
"loss": 4.6326,
|
|
"mean_token_accuracy": 0.2181015580892563,
|
|
"num_tokens": 16305489.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.213150835037231,
|
|
"epoch": 7.157201272149023,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00014976170383678954,
|
|
"loss": 4.5761,
|
|
"mean_token_accuracy": 0.23275419026613237,
|
|
"num_tokens": 16315167.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.252017831802368,
|
|
"epoch": 7.161744661517492,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00014946823982475883,
|
|
"loss": 4.5801,
|
|
"mean_token_accuracy": 0.23170557320117952,
|
|
"num_tokens": 16325059.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.330133104324341,
|
|
"epoch": 7.166288050885961,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014917508554986754,
|
|
"loss": 4.6931,
|
|
"mean_token_accuracy": 0.21529623717069626,
|
|
"num_tokens": 16336647.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.267104291915894,
|
|
"epoch": 7.17083144025443,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014888224173544476,
|
|
"loss": 4.6091,
|
|
"mean_token_accuracy": 0.21587494015693665,
|
|
"num_tokens": 16346146.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.320767259597778,
|
|
"epoch": 7.175374829622899,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00014858970910405368,
|
|
"loss": 4.7161,
|
|
"mean_token_accuracy": 0.21916465759277343,
|
|
"num_tokens": 16356124.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.2959692001342775,
|
|
"epoch": 7.179918218991368,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014829748837748906,
|
|
"loss": 4.6771,
|
|
"mean_token_accuracy": 0.21392595767974854,
|
|
"num_tokens": 16367197.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.257272338867187,
|
|
"epoch": 7.184461608359836,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001480055802767769,
|
|
"loss": 4.6201,
|
|
"mean_token_accuracy": 0.21948072761297227,
|
|
"num_tokens": 16376757.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.233496904373169,
|
|
"epoch": 7.189004997728305,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00014771398552217114,
|
|
"loss": 4.584,
|
|
"mean_token_accuracy": 0.22719783186912537,
|
|
"num_tokens": 16387454.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.26613483428955,
|
|
"epoch": 7.193548387096774,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00014742270483315319,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.21930038183927536,
|
|
"num_tokens": 16397852.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.2348397731781,
|
|
"epoch": 7.198091776465243,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00014713173892842878,
|
|
"loss": 4.6081,
|
|
"mean_token_accuracy": 0.2179872900247574,
|
|
"num_tokens": 16407752.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.2292236328125,
|
|
"epoch": 7.202635165833712,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 0.00014684108852592765,
|
|
"loss": 4.6383,
|
|
"mean_token_accuracy": 0.21559285819530488,
|
|
"num_tokens": 16418500.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.269316291809082,
|
|
"epoch": 7.207178555202181,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00014655075434280064,
|
|
"loss": 4.6564,
|
|
"mean_token_accuracy": 0.21822898387908934,
|
|
"num_tokens": 16428706.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.286206150054932,
|
|
"epoch": 7.21172194457065,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014626073709541853,
|
|
"loss": 4.6271,
|
|
"mean_token_accuracy": 0.2224705383181572,
|
|
"num_tokens": 16438184.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.234716129302979,
|
|
"epoch": 7.216265333939119,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014597103749937,
|
|
"loss": 4.6103,
|
|
"mean_token_accuracy": 0.22160200625658036,
|
|
"num_tokens": 16449340.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.280458593368531,
|
|
"epoch": 7.220808723307587,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00014568165626946012,
|
|
"loss": 4.6011,
|
|
"mean_token_accuracy": 0.21460297256708144,
|
|
"num_tokens": 16459604.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.34124231338501,
|
|
"epoch": 7.225352112676056,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014539259411970823,
|
|
"loss": 4.7176,
|
|
"mean_token_accuracy": 0.2135590672492981,
|
|
"num_tokens": 16470150.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.295072841644287,
|
|
"epoch": 7.229895502044525,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001451038517633465,
|
|
"loss": 4.6627,
|
|
"mean_token_accuracy": 0.22516481280326844,
|
|
"num_tokens": 16480797.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.3173669338226315,
|
|
"epoch": 7.234438891412994,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014481542991281782,
|
|
"loss": 4.7262,
|
|
"mean_token_accuracy": 0.21400680691003798,
|
|
"num_tokens": 16492629.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.258593797683716,
|
|
"epoch": 7.238982280781463,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00014452732927977465,
|
|
"loss": 4.5853,
|
|
"mean_token_accuracy": 0.21934845745563508,
|
|
"num_tokens": 16503225.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.297450828552246,
|
|
"epoch": 7.243525670149932,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001442395505750766,
|
|
"loss": 4.5756,
|
|
"mean_token_accuracy": 0.2287120297551155,
|
|
"num_tokens": 16512446.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.288908243179321,
|
|
"epoch": 7.248069059518401,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00014395209450878899,
|
|
"loss": 4.6254,
|
|
"mean_token_accuracy": 0.2230156406760216,
|
|
"num_tokens": 16522944.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.290480375289917,
|
|
"epoch": 7.25261244888687,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000143664961790181,
|
|
"loss": 4.6108,
|
|
"mean_token_accuracy": 0.22578912675380708,
|
|
"num_tokens": 16533100.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.321746301651001,
|
|
"epoch": 7.257155838255338,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001433781531277243,
|
|
"loss": 4.7134,
|
|
"mean_token_accuracy": 0.2120378151535988,
|
|
"num_tokens": 16542893.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.273731279373169,
|
|
"epoch": 7.261699227623807,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001430916692290905,
|
|
"loss": 4.642,
|
|
"mean_token_accuracy": 0.21866109669208528,
|
|
"num_tokens": 16553209.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.280065202713013,
|
|
"epoch": 7.266242616992276,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00014280551080115032,
|
|
"loss": 4.6328,
|
|
"mean_token_accuracy": 0.22135452479124068,
|
|
"num_tokens": 16564574.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 7.266242616992276,
|
|
"eval_entropy": 5.179921276624813,
|
|
"eval_loss": 5.71056604385376,
|
|
"eval_mean_token_accuracy": 0.17058009593639262,
|
|
"eval_num_tokens": 16564574.0,
|
|
"eval_runtime": 2.5956,
|
|
"eval_samples_per_second": 1323.383,
|
|
"eval_steps_per_second": 165.664,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.280980634689331,
|
|
"epoch": 7.270786006360745,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00014251967854997106,
|
|
"loss": 4.646,
|
|
"mean_token_accuracy": 0.219280706346035,
|
|
"num_tokens": 16574517.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.283842849731445,
|
|
"epoch": 7.275329395729214,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00014223417318081572,
|
|
"loss": 4.6571,
|
|
"mean_token_accuracy": 0.21817802786827087,
|
|
"num_tokens": 16585311.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.232319736480713,
|
|
"epoch": 7.279872785097683,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00014194899539814,
|
|
"loss": 4.6344,
|
|
"mean_token_accuracy": 0.21650594025850295,
|
|
"num_tokens": 16596055.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.239237117767334,
|
|
"epoch": 7.284416174466152,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014166414590559197,
|
|
"loss": 4.5916,
|
|
"mean_token_accuracy": 0.22044324725866318,
|
|
"num_tokens": 16606076.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.304018783569336,
|
|
"epoch": 7.288959563834621,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014137962540600944,
|
|
"loss": 4.6503,
|
|
"mean_token_accuracy": 0.21583789736032485,
|
|
"num_tokens": 16616740.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.274038314819336,
|
|
"epoch": 7.29350295320309,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00014109543460141837,
|
|
"loss": 4.6138,
|
|
"mean_token_accuracy": 0.22221709191799163,
|
|
"num_tokens": 16627412.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.270502710342408,
|
|
"epoch": 7.298046342571558,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001408115741930313,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.22584737539291383,
|
|
"num_tokens": 16637032.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.217905426025391,
|
|
"epoch": 7.302589731940027,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00014052804488124575,
|
|
"loss": 4.6068,
|
|
"mean_token_accuracy": 0.222050341963768,
|
|
"num_tokens": 16646596.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.262633037567139,
|
|
"epoch": 7.307133121308496,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014024484736564196,
|
|
"loss": 4.5619,
|
|
"mean_token_accuracy": 0.22236412465572358,
|
|
"num_tokens": 16656404.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.2764232635498045,
|
|
"epoch": 7.311676510676965,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001399619823449817,
|
|
"loss": 4.6364,
|
|
"mean_token_accuracy": 0.21735679358243942,
|
|
"num_tokens": 16666372.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.235891532897949,
|
|
"epoch": 7.316219900045434,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013967945051720614,
|
|
"loss": 4.6226,
|
|
"mean_token_accuracy": 0.22224539071321486,
|
|
"num_tokens": 16677150.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.238916826248169,
|
|
"epoch": 7.320763289413903,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00013939725257943475,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.21453520208597182,
|
|
"num_tokens": 16688712.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.231528949737549,
|
|
"epoch": 7.325306678782372,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013911538922796275,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.21472105234861374,
|
|
"num_tokens": 16699214.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.262210702896118,
|
|
"epoch": 7.32985006815084,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013883386115825984,
|
|
"loss": 4.6409,
|
|
"mean_token_accuracy": 0.21981458514928817,
|
|
"num_tokens": 16709294.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.275801515579223,
|
|
"epoch": 7.334393457519309,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00013855266906496867,
|
|
"loss": 4.5725,
|
|
"mean_token_accuracy": 0.22105900049209595,
|
|
"num_tokens": 16718761.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.317006254196167,
|
|
"epoch": 7.338936846887778,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013827181364190288,
|
|
"loss": 4.7011,
|
|
"mean_token_accuracy": 0.2107105538249016,
|
|
"num_tokens": 16729075.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.2554709911346436,
|
|
"epoch": 7.343480236256247,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00013799129558204505,
|
|
"loss": 4.571,
|
|
"mean_token_accuracy": 0.2261507347226143,
|
|
"num_tokens": 16739293.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.258328533172607,
|
|
"epoch": 7.348023625624716,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00013771111557754577,
|
|
"loss": 4.6392,
|
|
"mean_token_accuracy": 0.21888426840305328,
|
|
"num_tokens": 16749214.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.202942848205566,
|
|
"epoch": 7.352567014993185,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 0.00013743127431972124,
|
|
"loss": 4.5808,
|
|
"mean_token_accuracy": 0.2263280376791954,
|
|
"num_tokens": 16758715.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.239596176147461,
|
|
"epoch": 7.357110404361654,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0001371517724990522,
|
|
"loss": 4.6784,
|
|
"mean_token_accuracy": 0.2191799521446228,
|
|
"num_tokens": 16768965.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.227122974395752,
|
|
"epoch": 7.361653793730123,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0001368726108051813,
|
|
"loss": 4.5431,
|
|
"mean_token_accuracy": 0.22957678884267807,
|
|
"num_tokens": 16778368.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.308400201797485,
|
|
"epoch": 7.366197183098592,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013659378992691244,
|
|
"loss": 4.6769,
|
|
"mean_token_accuracy": 0.21707866638898848,
|
|
"num_tokens": 16789396.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.286496925354004,
|
|
"epoch": 7.37074057246706,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013631531055220835,
|
|
"loss": 4.6554,
|
|
"mean_token_accuracy": 0.21679040938615798,
|
|
"num_tokens": 16800063.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.264914655685425,
|
|
"epoch": 7.375283961835529,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013603717336818934,
|
|
"loss": 4.6335,
|
|
"mean_token_accuracy": 0.21827109456062316,
|
|
"num_tokens": 16810344.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.232664680480957,
|
|
"epoch": 7.379827351203998,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013575937906113095,
|
|
"loss": 4.5755,
|
|
"mean_token_accuracy": 0.2227008894085884,
|
|
"num_tokens": 16820233.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.292092370986938,
|
|
"epoch": 7.384370740572467,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001354819283164634,
|
|
"loss": 4.6892,
|
|
"mean_token_accuracy": 0.21405888795852662,
|
|
"num_tokens": 16829761.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.329210042953491,
|
|
"epoch": 7.388914129940936,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001352048218187686,
|
|
"loss": 4.7221,
|
|
"mean_token_accuracy": 0.21614449620246887,
|
|
"num_tokens": 16841068.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.266998529434204,
|
|
"epoch": 7.393457519309405,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013492806025177935,
|
|
"loss": 4.6069,
|
|
"mean_token_accuracy": 0.22034084647893906,
|
|
"num_tokens": 16850733.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.247817802429199,
|
|
"epoch": 7.398000908677874,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013465164429837716,
|
|
"loss": 4.6088,
|
|
"mean_token_accuracy": 0.21898606568574905,
|
|
"num_tokens": 16860416.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.2858398914337155,
|
|
"epoch": 7.402544298046343,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001343755746405912,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.22238477170467377,
|
|
"num_tokens": 16870662.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.231678915023804,
|
|
"epoch": 7.407087687414811,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00013409985195959574,
|
|
"loss": 4.7113,
|
|
"mean_token_accuracy": 0.207826629281044,
|
|
"num_tokens": 16881758.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.268845176696777,
|
|
"epoch": 7.41163107678328,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013382447693570908,
|
|
"loss": 4.6704,
|
|
"mean_token_accuracy": 0.2109234169125557,
|
|
"num_tokens": 16892399.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.285771083831787,
|
|
"epoch": 7.416174466151749,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013354945024839166,
|
|
"loss": 4.6986,
|
|
"mean_token_accuracy": 0.2117641180753708,
|
|
"num_tokens": 16903696.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.297069501876831,
|
|
"epoch": 7.420717855520218,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013327477257624466,
|
|
"loss": 4.6397,
|
|
"mean_token_accuracy": 0.21829614341259002,
|
|
"num_tokens": 16914326.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.246400547027588,
|
|
"epoch": 7.425261244888687,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001330004445970078,
|
|
"loss": 4.5562,
|
|
"mean_token_accuracy": 0.23316499590873718,
|
|
"num_tokens": 16924533.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.27253098487854,
|
|
"epoch": 7.429804634257156,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013272646698755816,
|
|
"loss": 4.5861,
|
|
"mean_token_accuracy": 0.22779925763607026,
|
|
"num_tokens": 16934157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.200542688369751,
|
|
"epoch": 7.434348023625625,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013245284042390814,
|
|
"loss": 4.5638,
|
|
"mean_token_accuracy": 0.2263687700033188,
|
|
"num_tokens": 16944148.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.266453695297241,
|
|
"epoch": 7.438891412994094,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013217956558120436,
|
|
"loss": 4.7182,
|
|
"mean_token_accuracy": 0.21434989869594573,
|
|
"num_tokens": 16956388.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.254949283599854,
|
|
"epoch": 7.443434802362562,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.000131906643133725,
|
|
"loss": 4.6575,
|
|
"mean_token_accuracy": 0.2236996576189995,
|
|
"num_tokens": 16965216.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.267396450042725,
|
|
"epoch": 7.447978191731031,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013163407375487926,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.21881519705057145,
|
|
"num_tokens": 16976250.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.254660177230835,
|
|
"epoch": 7.4525215810995,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00013136185811720496,
|
|
"loss": 4.7051,
|
|
"mean_token_accuracy": 0.21344553381204606,
|
|
"num_tokens": 16987375.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.224292182922364,
|
|
"epoch": 7.457064970467969,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001310899968923671,
|
|
"loss": 4.6672,
|
|
"mean_token_accuracy": 0.2236185297369957,
|
|
"num_tokens": 16997526.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.282782506942749,
|
|
"epoch": 7.461608359836438,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001308184907511562,
|
|
"loss": 4.6882,
|
|
"mean_token_accuracy": 0.2164272338151932,
|
|
"num_tokens": 17008342.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.259632110595703,
|
|
"epoch": 7.466151749204907,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013054734036348675,
|
|
"loss": 4.6588,
|
|
"mean_token_accuracy": 0.21742897778749465,
|
|
"num_tokens": 17019555.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.285071229934692,
|
|
"epoch": 7.470695138573376,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013027654639839536,
|
|
"loss": 4.6949,
|
|
"mean_token_accuracy": 0.2155071809887886,
|
|
"num_tokens": 17030686.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.282338714599609,
|
|
"epoch": 7.475238527941845,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013000610952403917,
|
|
"loss": 4.6157,
|
|
"mean_token_accuracy": 0.22847080677747728,
|
|
"num_tokens": 17041089.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.197641611099243,
|
|
"epoch": 7.479781917310314,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012973603040769427,
|
|
"loss": 4.5967,
|
|
"mean_token_accuracy": 0.21643335968255997,
|
|
"num_tokens": 17050194.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.285591459274292,
|
|
"epoch": 7.484325306678782,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00012946630971575414,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.20902557224035262,
|
|
"num_tokens": 17060596.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.30034556388855,
|
|
"epoch": 7.488868696047251,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00012919694811372768,
|
|
"loss": 4.6641,
|
|
"mean_token_accuracy": 0.21397403478622437,
|
|
"num_tokens": 17069763.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.227181529998779,
|
|
"epoch": 7.49341208541572,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00012892794626623793,
|
|
"loss": 4.5945,
|
|
"mean_token_accuracy": 0.22203188240528107,
|
|
"num_tokens": 17080031.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.262538957595825,
|
|
"epoch": 7.497955474784189,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001286593048370201,
|
|
"loss": 4.6399,
|
|
"mean_token_accuracy": 0.21926014721393586,
|
|
"num_tokens": 17091339.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.226081514358521,
|
|
"epoch": 7.502498864152658,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00012839102448892033,
|
|
"loss": 4.5862,
|
|
"mean_token_accuracy": 0.22540078461170196,
|
|
"num_tokens": 17102264.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.244753026962281,
|
|
"epoch": 7.507042253521127,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012812310588389367,
|
|
"loss": 4.6375,
|
|
"mean_token_accuracy": 0.21473031342029572,
|
|
"num_tokens": 17112493.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.246786069869995,
|
|
"epoch": 7.511585642889596,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012785554968300266,
|
|
"loss": 4.6041,
|
|
"mean_token_accuracy": 0.22348957508802414,
|
|
"num_tokens": 17122037.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.263935327529907,
|
|
"epoch": 7.516129032258064,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012758835654641548,
|
|
"loss": 4.6393,
|
|
"mean_token_accuracy": 0.22197726964950562,
|
|
"num_tokens": 17133403.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.208199834823608,
|
|
"epoch": 7.520672421626533,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 0.000127321527133405,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.2214723125100136,
|
|
"num_tokens": 17143619.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.279139518737793,
|
|
"epoch": 7.525215810995002,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012705506210234583,
|
|
"loss": 4.6954,
|
|
"mean_token_accuracy": 0.20976175516843795,
|
|
"num_tokens": 17154811.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.225062656402588,
|
|
"epoch": 7.529759200363471,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00012678896211071422,
|
|
"loss": 4.6115,
|
|
"mean_token_accuracy": 0.21951898336410522,
|
|
"num_tokens": 17166039.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.287937259674072,
|
|
"epoch": 7.53430258973194,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001265232278150854,
|
|
"loss": 4.6338,
|
|
"mean_token_accuracy": 0.2178898945450783,
|
|
"num_tokens": 17176555.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.250502872467041,
|
|
"epoch": 7.538845979100409,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.0001262578598711323,
|
|
"loss": 4.6786,
|
|
"mean_token_accuracy": 0.2107609987258911,
|
|
"num_tokens": 17188220.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.258058786392212,
|
|
"epoch": 7.543389368468878,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001259928589336239,
|
|
"loss": 4.6306,
|
|
"mean_token_accuracy": 0.22054579257965087,
|
|
"num_tokens": 17199528.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.263714408874511,
|
|
"epoch": 7.547932757837347,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012572822565642378,
|
|
"loss": 4.5995,
|
|
"mean_token_accuracy": 0.21510473489761353,
|
|
"num_tokens": 17210030.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.2773202419281,
|
|
"epoch": 7.552476147205816,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00012546396069248813,
|
|
"loss": 4.6786,
|
|
"mean_token_accuracy": 0.2183949589729309,
|
|
"num_tokens": 17220231.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.243138027191162,
|
|
"epoch": 7.557019536574284,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012520006469386458,
|
|
"loss": 4.6344,
|
|
"mean_token_accuracy": 0.2199437752366066,
|
|
"num_tokens": 17229966.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.26244044303894,
|
|
"epoch": 7.561562925942753,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012493653831169014,
|
|
"loss": 4.6694,
|
|
"mean_token_accuracy": 0.2191854879260063,
|
|
"num_tokens": 17240654.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.202248620986938,
|
|
"epoch": 7.566106315311222,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00012467338219619015,
|
|
"loss": 4.5454,
|
|
"mean_token_accuracy": 0.23462310284376145,
|
|
"num_tokens": 17250653.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.280370092391967,
|
|
"epoch": 7.570649704679691,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000124410596996676,
|
|
"loss": 4.6583,
|
|
"mean_token_accuracy": 0.21644940376281738,
|
|
"num_tokens": 17260223.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.216563081741333,
|
|
"epoch": 7.57519309404816,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001241481833615442,
|
|
"loss": 4.6151,
|
|
"mean_token_accuracy": 0.21821609884500504,
|
|
"num_tokens": 17269998.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.264539909362793,
|
|
"epoch": 7.579736483416629,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012388614193827407,
|
|
"loss": 4.6097,
|
|
"mean_token_accuracy": 0.22745223492383956,
|
|
"num_tokens": 17279824.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.234407520294189,
|
|
"epoch": 7.584279872785098,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000123624473373427,
|
|
"loss": 4.5958,
|
|
"mean_token_accuracy": 0.22410191595554352,
|
|
"num_tokens": 17290242.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.20608549118042,
|
|
"epoch": 7.588823262153566,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00012336317831264411,
|
|
"loss": 4.5884,
|
|
"mean_token_accuracy": 0.2290993645787239,
|
|
"num_tokens": 17299515.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.240670871734619,
|
|
"epoch": 7.593366651522036,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012310225740064498,
|
|
"loss": 4.5597,
|
|
"mean_token_accuracy": 0.22444207966327667,
|
|
"num_tokens": 17308627.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.1907960891723635,
|
|
"epoch": 7.597910040890504,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001228417112812259,
|
|
"loss": 4.5499,
|
|
"mean_token_accuracy": 0.215601746737957,
|
|
"num_tokens": 17319465.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.201975536346436,
|
|
"epoch": 7.602453430258973,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012258154059725877,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.21769667863845826,
|
|
"num_tokens": 17329202.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.238059139251709,
|
|
"epoch": 7.606996819627442,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001223217459906886,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.2367899537086487,
|
|
"num_tokens": 17338956.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.314474964141846,
|
|
"epoch": 7.611540208995911,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012206232810253297,
|
|
"loss": 4.6936,
|
|
"mean_token_accuracy": 0.21824212223291398,
|
|
"num_tokens": 17349380.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.276782274246216,
|
|
"epoch": 7.61608359836438,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012180328757287967,
|
|
"loss": 4.7614,
|
|
"mean_token_accuracy": 0.2101784199476242,
|
|
"num_tokens": 17359770.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.227610778808594,
|
|
"epoch": 7.620626987732849,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00012154462504088545,
|
|
"loss": 4.6164,
|
|
"mean_token_accuracy": 0.22567343562841416,
|
|
"num_tokens": 17369209.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.229271173477173,
|
|
"epoch": 7.625170377101318,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001212863411447744,
|
|
"loss": 4.6323,
|
|
"mean_token_accuracy": 0.21572960764169694,
|
|
"num_tokens": 17379221.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.267580938339234,
|
|
"epoch": 7.629713766469786,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012102843652183628,
|
|
"loss": 4.5497,
|
|
"mean_token_accuracy": 0.2248469188809395,
|
|
"num_tokens": 17389164.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.326529169082642,
|
|
"epoch": 7.634257155838255,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00012077091180842523,
|
|
"loss": 4.6676,
|
|
"mean_token_accuracy": 0.21617236286401748,
|
|
"num_tokens": 17399392.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.286958265304565,
|
|
"epoch": 7.638800545206724,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012051376763995784,
|
|
"loss": 4.6283,
|
|
"mean_token_accuracy": 0.2201067715883255,
|
|
"num_tokens": 17408933.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.255212831497192,
|
|
"epoch": 7.643343934575193,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012025700465091177,
|
|
"loss": 4.6173,
|
|
"mean_token_accuracy": 0.21879734843969345,
|
|
"num_tokens": 17419391.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.263277053833008,
|
|
"epoch": 7.647887323943662,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012000062347482413,
|
|
"loss": 4.6492,
|
|
"mean_token_accuracy": 0.21866900622844695,
|
|
"num_tokens": 17430562.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.244241428375244,
|
|
"epoch": 7.652430713312131,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011974462474429015,
|
|
"loss": 4.6304,
|
|
"mean_token_accuracy": 0.21740393042564393,
|
|
"num_tokens": 17441345.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.241474342346192,
|
|
"epoch": 7.6569741026806,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011948900909096101,
|
|
"loss": 4.6268,
|
|
"mean_token_accuracy": 0.21534114629030227,
|
|
"num_tokens": 17450972.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.236260557174683,
|
|
"epoch": 7.661517492049069,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001192337771455431,
|
|
"loss": 4.6164,
|
|
"mean_token_accuracy": 0.2184592068195343,
|
|
"num_tokens": 17460930.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.282356119155883,
|
|
"epoch": 7.666060881417538,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011897892953779576,
|
|
"loss": 4.6266,
|
|
"mean_token_accuracy": 0.22112844735383988,
|
|
"num_tokens": 17470865.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.245069885253907,
|
|
"epoch": 7.670604270786006,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011872446689653032,
|
|
"loss": 4.5521,
|
|
"mean_token_accuracy": 0.22996515184640884,
|
|
"num_tokens": 17481379.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.358064317703247,
|
|
"epoch": 7.675147660154475,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001184703898496078,
|
|
"loss": 4.6852,
|
|
"mean_token_accuracy": 0.2209307000041008,
|
|
"num_tokens": 17492741.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.286445760726929,
|
|
"epoch": 7.679691049522944,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011821669902393823,
|
|
"loss": 4.6778,
|
|
"mean_token_accuracy": 0.20831394493579863,
|
|
"num_tokens": 17504179.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.21372799873352,
|
|
"epoch": 7.684234438891413,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011796339504547846,
|
|
"loss": 4.6245,
|
|
"mean_token_accuracy": 0.22467038184404373,
|
|
"num_tokens": 17514299.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.302642154693603,
|
|
"epoch": 7.688777828259882,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011771047853923086,
|
|
"loss": 4.6237,
|
|
"mean_token_accuracy": 0.22399120330810546,
|
|
"num_tokens": 17524454.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.264330339431763,
|
|
"epoch": 7.693321217628351,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011745795012924169,
|
|
"loss": 4.5928,
|
|
"mean_token_accuracy": 0.22503731697797774,
|
|
"num_tokens": 17534887.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.261464405059814,
|
|
"epoch": 7.69786460699682,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00011720581043859988,
|
|
"loss": 4.6735,
|
|
"mean_token_accuracy": 0.21746076196432113,
|
|
"num_tokens": 17544944.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.246092462539673,
|
|
"epoch": 7.702407996365288,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011695406008943495,
|
|
"loss": 4.6281,
|
|
"mean_token_accuracy": 0.2283418819308281,
|
|
"num_tokens": 17556481.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.32862491607666,
|
|
"epoch": 7.706951385733757,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011670269970291584,
|
|
"loss": 4.6796,
|
|
"mean_token_accuracy": 0.21634739488363267,
|
|
"num_tokens": 17566776.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.188323879241944,
|
|
"epoch": 7.711494775102226,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001164517298992493,
|
|
"loss": 4.5798,
|
|
"mean_token_accuracy": 0.2168860375881195,
|
|
"num_tokens": 17575441.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.2943048000335695,
|
|
"epoch": 7.716038164470695,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011620115129767848,
|
|
"loss": 4.7061,
|
|
"mean_token_accuracy": 0.21458061933517455,
|
|
"num_tokens": 17585792.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.226095676422119,
|
|
"epoch": 7.720581553839164,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011595096451648111,
|
|
"loss": 4.6305,
|
|
"mean_token_accuracy": 0.21615330576896669,
|
|
"num_tokens": 17595820.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.720581553839164,
|
|
"eval_entropy": 5.183981141378713,
|
|
"eval_loss": 5.699517726898193,
|
|
"eval_mean_token_accuracy": 0.17211287583029547,
|
|
"eval_num_tokens": 17595820.0,
|
|
"eval_runtime": 2.587,
|
|
"eval_samples_per_second": 1327.777,
|
|
"eval_steps_per_second": 166.214,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.234664249420166,
|
|
"epoch": 7.725124943207633,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011570117017296819,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.22798995971679686,
|
|
"num_tokens": 17605791.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.263302946090699,
|
|
"epoch": 7.729668332576102,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011545176888348237,
|
|
"loss": 4.6661,
|
|
"mean_token_accuracy": 0.2146323412656784,
|
|
"num_tokens": 17616159.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.222070169448853,
|
|
"epoch": 7.734211721944571,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011520276126339667,
|
|
"loss": 4.6041,
|
|
"mean_token_accuracy": 0.22045547068119048,
|
|
"num_tokens": 17626612.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.27053747177124,
|
|
"epoch": 7.73875511131304,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011495414792711259,
|
|
"loss": 4.6305,
|
|
"mean_token_accuracy": 0.2170681521296501,
|
|
"num_tokens": 17636254.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.3440176486969,
|
|
"epoch": 7.743298500681508,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00011470592948805882,
|
|
"loss": 4.7165,
|
|
"mean_token_accuracy": 0.2079721748828888,
|
|
"num_tokens": 17648097.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.260576438903809,
|
|
"epoch": 7.747841890049977,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011445810655868957,
|
|
"loss": 4.5919,
|
|
"mean_token_accuracy": 0.22387945055961608,
|
|
"num_tokens": 17658361.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.19376654624939,
|
|
"epoch": 7.752385279418446,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011421067975048355,
|
|
"loss": 4.5635,
|
|
"mean_token_accuracy": 0.22718183547258378,
|
|
"num_tokens": 17668640.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.173579359054566,
|
|
"epoch": 7.756928668786915,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011396364967394152,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.23119159191846847,
|
|
"num_tokens": 17678257.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.203030824661255,
|
|
"epoch": 7.761472058155384,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011371701693858582,
|
|
"loss": 4.6275,
|
|
"mean_token_accuracy": 0.2199384644627571,
|
|
"num_tokens": 17688635.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.251443576812744,
|
|
"epoch": 7.766015447523853,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011347078215295817,
|
|
"loss": 4.6805,
|
|
"mean_token_accuracy": 0.21199290305376053,
|
|
"num_tokens": 17699917.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.286216592788696,
|
|
"epoch": 7.770558836892322,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00011322494592461839,
|
|
"loss": 4.6348,
|
|
"mean_token_accuracy": 0.22105531245470048,
|
|
"num_tokens": 17710278.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.228350019454956,
|
|
"epoch": 7.77510222626079,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011297950886014292,
|
|
"loss": 4.5967,
|
|
"mean_token_accuracy": 0.2304760232567787,
|
|
"num_tokens": 17720375.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.253705453872681,
|
|
"epoch": 7.779645615629259,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001127344715651234,
|
|
"loss": 4.6304,
|
|
"mean_token_accuracy": 0.21359452605247498,
|
|
"num_tokens": 17730299.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.291649198532104,
|
|
"epoch": 7.784189004997728,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011248983464416497,
|
|
"loss": 4.6522,
|
|
"mean_token_accuracy": 0.21793753057718276,
|
|
"num_tokens": 17741096.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.229942178726196,
|
|
"epoch": 7.788732394366197,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011224559870088488,
|
|
"loss": 4.6225,
|
|
"mean_token_accuracy": 0.22049803286790848,
|
|
"num_tokens": 17751502.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.252112913131714,
|
|
"epoch": 7.793275783734666,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011200176433791101,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.2199626937508583,
|
|
"num_tokens": 17761955.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.188704872131348,
|
|
"epoch": 7.797819173103135,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001117583321568805,
|
|
"loss": 4.5726,
|
|
"mean_token_accuracy": 0.2232927680015564,
|
|
"num_tokens": 17772330.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.24195671081543,
|
|
"epoch": 7.802362562471604,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011151530275843805,
|
|
"loss": 4.536,
|
|
"mean_token_accuracy": 0.23166336715221406,
|
|
"num_tokens": 17781688.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.197198486328125,
|
|
"epoch": 7.806905951840073,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00011127267674223449,
|
|
"loss": 4.6085,
|
|
"mean_token_accuracy": 0.22127853631973265,
|
|
"num_tokens": 17792037.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.275870132446289,
|
|
"epoch": 7.811449341208542,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011103045470692532,
|
|
"loss": 4.673,
|
|
"mean_token_accuracy": 0.22075311094522476,
|
|
"num_tokens": 17802705.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.30716986656189,
|
|
"epoch": 7.81599273057701,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011078863725016955,
|
|
"loss": 4.7255,
|
|
"mean_token_accuracy": 0.21222663819789886,
|
|
"num_tokens": 17813758.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.2064660549163815,
|
|
"epoch": 7.820536119945479,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011054722496862759,
|
|
"loss": 4.5904,
|
|
"mean_token_accuracy": 0.2239968553185463,
|
|
"num_tokens": 17823465.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.265296554565429,
|
|
"epoch": 7.825079509313948,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011030621845796021,
|
|
"loss": 4.6044,
|
|
"mean_token_accuracy": 0.22265194058418275,
|
|
"num_tokens": 17833719.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.226050472259521,
|
|
"epoch": 7.829622898682417,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011006561831282703,
|
|
"loss": 4.5837,
|
|
"mean_token_accuracy": 0.22592074275016785,
|
|
"num_tokens": 17844519.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.246757078170776,
|
|
"epoch": 7.834166288050886,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010982542512688515,
|
|
"loss": 4.5877,
|
|
"mean_token_accuracy": 0.2262299418449402,
|
|
"num_tokens": 17856463.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.196876335144043,
|
|
"epoch": 7.838709677419355,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 0.00010958563949278711,
|
|
"loss": 4.5795,
|
|
"mean_token_accuracy": 0.2216697260737419,
|
|
"num_tokens": 17867449.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.184791517257691,
|
|
"epoch": 7.843253066787824,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010934626200218032,
|
|
"loss": 4.6027,
|
|
"mean_token_accuracy": 0.21649682521820068,
|
|
"num_tokens": 17878522.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.260023784637451,
|
|
"epoch": 7.847796456156293,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010910729324570488,
|
|
"loss": 4.6505,
|
|
"mean_token_accuracy": 0.2220764324069023,
|
|
"num_tokens": 17888544.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.274901533126831,
|
|
"epoch": 7.852339845524762,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010886873381299248,
|
|
"loss": 4.6926,
|
|
"mean_token_accuracy": 0.21742896884679794,
|
|
"num_tokens": 17897432.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.261731386184692,
|
|
"epoch": 7.85688323489323,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010863058429266469,
|
|
"loss": 4.6316,
|
|
"mean_token_accuracy": 0.22556807547807695,
|
|
"num_tokens": 17907868.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.258230495452881,
|
|
"epoch": 7.861426624261699,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010839284527233192,
|
|
"loss": 4.5704,
|
|
"mean_token_accuracy": 0.22311976850032805,
|
|
"num_tokens": 17917887.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.195612382888794,
|
|
"epoch": 7.865970013630168,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00010815551733859153,
|
|
"loss": 4.5806,
|
|
"mean_token_accuracy": 0.2201479882001877,
|
|
"num_tokens": 17928030.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.231483793258667,
|
|
"epoch": 7.870513402998637,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010791860107702658,
|
|
"loss": 4.6837,
|
|
"mean_token_accuracy": 0.21030519902706146,
|
|
"num_tokens": 17937990.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.214671039581299,
|
|
"epoch": 7.875056792367106,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010768209707220435,
|
|
"loss": 4.6775,
|
|
"mean_token_accuracy": 0.21529633849859237,
|
|
"num_tokens": 17948262.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.263279438018799,
|
|
"epoch": 7.879600181735575,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010744600590767504,
|
|
"loss": 4.6429,
|
|
"mean_token_accuracy": 0.21998861879110337,
|
|
"num_tokens": 17958942.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.243550252914429,
|
|
"epoch": 7.884143571104044,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010721032816597012,
|
|
"loss": 4.6301,
|
|
"mean_token_accuracy": 0.21741609126329423,
|
|
"num_tokens": 17969362.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.283176517486572,
|
|
"epoch": 7.888686960472512,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000106975064428601,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.22129491716623306,
|
|
"num_tokens": 17979009.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.30822982788086,
|
|
"epoch": 7.893230349840981,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010674021527605742,
|
|
"loss": 4.667,
|
|
"mean_token_accuracy": 0.2217469185590744,
|
|
"num_tokens": 17989064.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.23407826423645,
|
|
"epoch": 7.89777373920945,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010650578128780656,
|
|
"loss": 4.5797,
|
|
"mean_token_accuracy": 0.2259630784392357,
|
|
"num_tokens": 17998574.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.319518136978149,
|
|
"epoch": 7.902317128577919,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001062717630422908,
|
|
"loss": 4.8027,
|
|
"mean_token_accuracy": 0.21051065027713775,
|
|
"num_tokens": 18010735.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.259637641906738,
|
|
"epoch": 7.906860517946388,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010603816111692699,
|
|
"loss": 4.6373,
|
|
"mean_token_accuracy": 0.22269640266895294,
|
|
"num_tokens": 18020316.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.2859502792358395,
|
|
"epoch": 7.911403907314857,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010580497608810455,
|
|
"loss": 4.6343,
|
|
"mean_token_accuracy": 0.21350328475236893,
|
|
"num_tokens": 18031058.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.250996875762939,
|
|
"epoch": 7.915947296683326,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010557220853118456,
|
|
"loss": 4.5952,
|
|
"mean_token_accuracy": 0.2240217611193657,
|
|
"num_tokens": 18040603.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.317535305023194,
|
|
"epoch": 7.920490686051795,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010533985902049755,
|
|
"loss": 4.6959,
|
|
"mean_token_accuracy": 0.2140206515789032,
|
|
"num_tokens": 18051369.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.237325286865234,
|
|
"epoch": 7.925034075420264,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010510792812934307,
|
|
"loss": 4.5804,
|
|
"mean_token_accuracy": 0.22363612949848174,
|
|
"num_tokens": 18062283.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.26497802734375,
|
|
"epoch": 7.929577464788732,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010487641642998746,
|
|
"loss": 4.6576,
|
|
"mean_token_accuracy": 0.21539966017007828,
|
|
"num_tokens": 18072125.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.24119610786438,
|
|
"epoch": 7.934120854157201,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010464532449366281,
|
|
"loss": 4.6913,
|
|
"mean_token_accuracy": 0.21975600123405456,
|
|
"num_tokens": 18083812.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.271639490127564,
|
|
"epoch": 7.93866424352567,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010441465289056548,
|
|
"loss": 4.659,
|
|
"mean_token_accuracy": 0.2207971528172493,
|
|
"num_tokens": 18095399.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.286943006515503,
|
|
"epoch": 7.943207632894139,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010418440218985482,
|
|
"loss": 4.6348,
|
|
"mean_token_accuracy": 0.22928458899259568,
|
|
"num_tokens": 18106234.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.232179784774781,
|
|
"epoch": 7.947751022262608,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010395457295965143,
|
|
"loss": 4.5499,
|
|
"mean_token_accuracy": 0.23280995786190034,
|
|
"num_tokens": 18116129.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.29691104888916,
|
|
"epoch": 7.952294411631077,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010372516576703617,
|
|
"loss": 4.6634,
|
|
"mean_token_accuracy": 0.21703089624643326,
|
|
"num_tokens": 18126757.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.252694034576416,
|
|
"epoch": 7.956837800999546,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0001034961811780484,
|
|
"loss": 4.6252,
|
|
"mean_token_accuracy": 0.22240649312734603,
|
|
"num_tokens": 18136645.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.311610889434815,
|
|
"epoch": 7.961381190368014,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010326761975768492,
|
|
"loss": 4.6805,
|
|
"mean_token_accuracy": 0.2211284413933754,
|
|
"num_tokens": 18146458.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.266968059539795,
|
|
"epoch": 7.965924579736483,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00010303948206989827,
|
|
"loss": 4.6258,
|
|
"mean_token_accuracy": 0.2242304041981697,
|
|
"num_tokens": 18156151.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.247860527038574,
|
|
"epoch": 7.970467969104952,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010281176867759556,
|
|
"loss": 4.6461,
|
|
"mean_token_accuracy": 0.22247770726680755,
|
|
"num_tokens": 18166695.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.2387152194976805,
|
|
"epoch": 7.975011358473421,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010258448014263682,
|
|
"loss": 4.5931,
|
|
"mean_token_accuracy": 0.22284553200006485,
|
|
"num_tokens": 18177011.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.272799825668335,
|
|
"epoch": 7.97955474784189,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010235761702583405,
|
|
"loss": 4.7027,
|
|
"mean_token_accuracy": 0.21780301928520202,
|
|
"num_tokens": 18187179.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.26063723564148,
|
|
"epoch": 7.984098137210359,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010213117988694944,
|
|
"loss": 4.6358,
|
|
"mean_token_accuracy": 0.21801599562168122,
|
|
"num_tokens": 18197106.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.1941478729248045,
|
|
"epoch": 7.988641526578828,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010190516928469407,
|
|
"loss": 4.6157,
|
|
"mean_token_accuracy": 0.22230374664068223,
|
|
"num_tokens": 18207729.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.227265548706055,
|
|
"epoch": 7.993184915947297,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010167958577672654,
|
|
"loss": 4.5575,
|
|
"mean_token_accuracy": 0.22351201623678207,
|
|
"num_tokens": 18217885.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.2653289318084715,
|
|
"epoch": 7.997728305315766,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010145442991965198,
|
|
"loss": 4.6536,
|
|
"mean_token_accuracy": 0.21540001630783082,
|
|
"num_tokens": 18229633.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.233368502722846,
|
|
"epoch": 8.001817355747388,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010122970226901981,
|
|
"loss": 4.6487,
|
|
"mean_token_accuracy": 0.22228972779379952,
|
|
"num_tokens": 18239194.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.29127950668335,
|
|
"epoch": 8.006360745115856,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010100540337932339,
|
|
"loss": 4.5771,
|
|
"mean_token_accuracy": 0.21876836121082305,
|
|
"num_tokens": 18250061.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.258141899108887,
|
|
"epoch": 8.010904134484326,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010078153380399773,
|
|
"loss": 4.5141,
|
|
"mean_token_accuracy": 0.2349844232201576,
|
|
"num_tokens": 18260977.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.235427284240723,
|
|
"epoch": 8.015447523852794,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00010055809409541902,
|
|
"loss": 4.5482,
|
|
"mean_token_accuracy": 0.23030997812747955,
|
|
"num_tokens": 18271866.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.25554723739624,
|
|
"epoch": 8.019990913221264,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.00010033508480490222,
|
|
"loss": 4.6198,
|
|
"mean_token_accuracy": 0.22217823565006256,
|
|
"num_tokens": 18283353.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.179054498672485,
|
|
"epoch": 8.024534302589732,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010011250648270081,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.22902377247810363,
|
|
"num_tokens": 18293619.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.251852035522461,
|
|
"epoch": 8.0290776919582,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.98903596780046e-05,
|
|
"loss": 4.5612,
|
|
"mean_token_accuracy": 0.23248142898082733,
|
|
"num_tokens": 18303411.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.264840316772461,
|
|
"epoch": 8.03362108132667,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.966864493893876e-05,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.23986872881650925,
|
|
"num_tokens": 18312798.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.24741530418396,
|
|
"epoch": 8.038164470695138,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.94473628125623e-05,
|
|
"loss": 4.5825,
|
|
"mean_token_accuracy": 0.22044241577386856,
|
|
"num_tokens": 18324111.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.234966373443603,
|
|
"epoch": 8.042707860063608,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.922651384486695e-05,
|
|
"loss": 4.549,
|
|
"mean_token_accuracy": 0.2262202888727188,
|
|
"num_tokens": 18335197.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.2563910484313965,
|
|
"epoch": 8.047251249432076,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 9.900609858077562e-05,
|
|
"loss": 4.5933,
|
|
"mean_token_accuracy": 0.2232024312019348,
|
|
"num_tokens": 18345896.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.296202278137207,
|
|
"epoch": 8.051794638800546,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.878611756414094e-05,
|
|
"loss": 4.5993,
|
|
"mean_token_accuracy": 0.22564305663108825,
|
|
"num_tokens": 18356941.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.2453583717346195,
|
|
"epoch": 8.056338028169014,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.856657133774424e-05,
|
|
"loss": 4.4922,
|
|
"mean_token_accuracy": 0.2315317213535309,
|
|
"num_tokens": 18366767.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.2860345363616945,
|
|
"epoch": 8.060881417537482,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.834746044329411e-05,
|
|
"loss": 4.6373,
|
|
"mean_token_accuracy": 0.22589844316244126,
|
|
"num_tokens": 18376698.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.254580020904541,
|
|
"epoch": 8.065424806905952,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.81287854214248e-05,
|
|
"loss": 4.518,
|
|
"mean_token_accuracy": 0.22918906211853027,
|
|
"num_tokens": 18386678.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.282481575012207,
|
|
"epoch": 8.06996819627442,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.791054681169522e-05,
|
|
"loss": 4.5768,
|
|
"mean_token_accuracy": 0.22233218401670457,
|
|
"num_tokens": 18397020.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.2205650329589846,
|
|
"epoch": 8.07451158564289,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.769274515258739e-05,
|
|
"loss": 4.5623,
|
|
"mean_token_accuracy": 0.21989136934280396,
|
|
"num_tokens": 18407330.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.22734899520874,
|
|
"epoch": 8.079054975011358,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.747538098150539e-05,
|
|
"loss": 4.5403,
|
|
"mean_token_accuracy": 0.22780284583568572,
|
|
"num_tokens": 18417046.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.239780473709106,
|
|
"epoch": 8.083598364379828,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.725845483477352e-05,
|
|
"loss": 4.6022,
|
|
"mean_token_accuracy": 0.2249958723783493,
|
|
"num_tokens": 18427907.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.2481772899627686,
|
|
"epoch": 8.088141753748296,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 9.704196724763565e-05,
|
|
"loss": 4.5682,
|
|
"mean_token_accuracy": 0.22553861886262894,
|
|
"num_tokens": 18437972.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.248580884933472,
|
|
"epoch": 8.092685143116766,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.682591875425323e-05,
|
|
"loss": 4.5788,
|
|
"mean_token_accuracy": 0.22634052336215973,
|
|
"num_tokens": 18448188.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.252760171890259,
|
|
"epoch": 8.097228532485234,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 9.661030988770463e-05,
|
|
"loss": 4.6036,
|
|
"mean_token_accuracy": 0.22171541899442673,
|
|
"num_tokens": 18458130.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.26653265953064,
|
|
"epoch": 8.101771921853702,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.63951411799831e-05,
|
|
"loss": 4.548,
|
|
"mean_token_accuracy": 0.2292411670088768,
|
|
"num_tokens": 18468257.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.199504327774048,
|
|
"epoch": 8.106315311222172,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 9.618041316199617e-05,
|
|
"loss": 4.5723,
|
|
"mean_token_accuracy": 0.2224802017211914,
|
|
"num_tokens": 18478720.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.259245681762695,
|
|
"epoch": 8.11085870059064,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.596612636356383e-05,
|
|
"loss": 4.5536,
|
|
"mean_token_accuracy": 0.22462934404611587,
|
|
"num_tokens": 18489543.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.24628529548645,
|
|
"epoch": 8.11540208995911,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.575228131341745e-05,
|
|
"loss": 4.5984,
|
|
"mean_token_accuracy": 0.22231434732675553,
|
|
"num_tokens": 18499505.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.30615348815918,
|
|
"epoch": 8.119945479327578,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 9.55388785391984e-05,
|
|
"loss": 4.6759,
|
|
"mean_token_accuracy": 0.21941571682691574,
|
|
"num_tokens": 18510044.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.254226160049439,
|
|
"epoch": 8.124488868696048,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.532591856745689e-05,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.223791241645813,
|
|
"num_tokens": 18520537.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.256456327438355,
|
|
"epoch": 8.129032258064516,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.511340192365044e-05,
|
|
"loss": 4.6249,
|
|
"mean_token_accuracy": 0.2146080896258354,
|
|
"num_tokens": 18531840.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.248765516281128,
|
|
"epoch": 8.133575647432984,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.490132913214277e-05,
|
|
"loss": 4.5556,
|
|
"mean_token_accuracy": 0.22279821932315827,
|
|
"num_tokens": 18542097.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.264424657821655,
|
|
"epoch": 8.138119036801454,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.468970071620231e-05,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.2185577929019928,
|
|
"num_tokens": 18551629.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.247019338607788,
|
|
"epoch": 8.142662426169922,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.447851719800126e-05,
|
|
"loss": 4.5437,
|
|
"mean_token_accuracy": 0.22308824509382247,
|
|
"num_tokens": 18562248.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.219873046875,
|
|
"epoch": 8.147205815538392,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.426777909861395e-05,
|
|
"loss": 4.5129,
|
|
"mean_token_accuracy": 0.22822321057319642,
|
|
"num_tokens": 18572107.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.260891771316528,
|
|
"epoch": 8.15174920490686,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.405748693801562e-05,
|
|
"loss": 4.594,
|
|
"mean_token_accuracy": 0.22418171763420106,
|
|
"num_tokens": 18582665.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.310816764831543,
|
|
"epoch": 8.15629259427533,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.384764123508124e-05,
|
|
"loss": 4.6212,
|
|
"mean_token_accuracy": 0.22273018956184387,
|
|
"num_tokens": 18593737.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.27907075881958,
|
|
"epoch": 8.160835983643798,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.363824250758443e-05,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.23119494318962097,
|
|
"num_tokens": 18604180.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.205421018600464,
|
|
"epoch": 8.165379373012268,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.34292912721955e-05,
|
|
"loss": 4.4991,
|
|
"mean_token_accuracy": 0.22659600228071214,
|
|
"num_tokens": 18614328.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.27348780632019,
|
|
"epoch": 8.169922762380736,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.322078804448095e-05,
|
|
"loss": 4.5743,
|
|
"mean_token_accuracy": 0.2356656402349472,
|
|
"num_tokens": 18625357.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.203095579147339,
|
|
"epoch": 8.174466151749204,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.301273333890171e-05,
|
|
"loss": 4.5267,
|
|
"mean_token_accuracy": 0.22977028638124466,
|
|
"num_tokens": 18636306.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 8.174466151749204,
|
|
"eval_entropy": 5.165834321532139,
|
|
"eval_loss": 5.695127487182617,
|
|
"eval_mean_token_accuracy": 0.17244828222102898,
|
|
"eval_num_tokens": 18636306.0,
|
|
"eval_runtime": 2.5786,
|
|
"eval_samples_per_second": 1332.114,
|
|
"eval_steps_per_second": 166.757,
|
|
"step": 9000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"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": 2.7038237668992e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|