3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.3625624716038165,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692006397247315,
|
|
"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
|
|
}
|
|
],
|
|
"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": 4495969830528000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|