043fd421c2
Model: fpadovani/nld-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.6333484779645615,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.164702081680298,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.747,
|
|
"mean_token_accuracy": 0.10336653664708137,
|
|
"num_tokens": 10156.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.202920770645141,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.7119,
|
|
"mean_token_accuracy": 0.10359775125980378,
|
|
"num_tokens": 20669.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.195320177078247,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.8681,
|
|
"mean_token_accuracy": 0.10121074020862579,
|
|
"num_tokens": 30952.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.181957340240478,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.6577,
|
|
"mean_token_accuracy": 0.1091083861887455,
|
|
"num_tokens": 41429.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.118825626373291,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.6923,
|
|
"mean_token_accuracy": 0.10817872360348701,
|
|
"num_tokens": 52166.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.141914033889771,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.7689,
|
|
"mean_token_accuracy": 0.10555170774459839,
|
|
"num_tokens": 61670.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.10323133468628,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.7218,
|
|
"mean_token_accuracy": 0.10314013063907623,
|
|
"num_tokens": 71702.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.063247537612915,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.6454,
|
|
"mean_token_accuracy": 0.1140250563621521,
|
|
"num_tokens": 82840.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.078653001785279,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6923,
|
|
"mean_token_accuracy": 0.10924328565597534,
|
|
"num_tokens": 92290.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.114263725280762,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.6761,
|
|
"mean_token_accuracy": 0.10787095576524734,
|
|
"num_tokens": 101842.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.1406755447387695,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.8497,
|
|
"mean_token_accuracy": 0.09816754907369614,
|
|
"num_tokens": 114030.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.09134955406189,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.7197,
|
|
"mean_token_accuracy": 0.10756717920303345,
|
|
"num_tokens": 124349.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.176341104507446,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.6648,
|
|
"mean_token_accuracy": 0.11219555139541626,
|
|
"num_tokens": 133534.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.076372528076172,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.6403,
|
|
"mean_token_accuracy": 0.10651129633188247,
|
|
"num_tokens": 144383.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.107057428359985,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.7266,
|
|
"mean_token_accuracy": 0.10738222151994706,
|
|
"num_tokens": 155467.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.098436498641968,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.10665511935949326,
|
|
"num_tokens": 165903.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.2096319675445555,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.7301,
|
|
"mean_token_accuracy": 0.10637103319168091,
|
|
"num_tokens": 175812.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.150934505462646,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.10753465890884399,
|
|
"num_tokens": 186357.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.032349157333374,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.662,
|
|
"mean_token_accuracy": 0.11050116717815399,
|
|
"num_tokens": 196124.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.046771717071533,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.6174,
|
|
"mean_token_accuracy": 0.11095911487936974,
|
|
"num_tokens": 205655.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.056599807739258,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.7337,
|
|
"mean_token_accuracy": 0.10731556490063668,
|
|
"num_tokens": 215870.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.120891666412353,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.7045,
|
|
"mean_token_accuracy": 0.10967985242605209,
|
|
"num_tokens": 225027.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.040978288650512,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.7271,
|
|
"mean_token_accuracy": 0.10503458082675934,
|
|
"num_tokens": 236352.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.152721691131592,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.6936,
|
|
"mean_token_accuracy": 0.10544832721352577,
|
|
"num_tokens": 246713.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.054176568984985,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.5967,
|
|
"mean_token_accuracy": 0.11465176194906235,
|
|
"num_tokens": 256950.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.072789001464844,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.7339,
|
|
"mean_token_accuracy": 0.10439727082848549,
|
|
"num_tokens": 267345.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.1310389041900635,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.5981,
|
|
"mean_token_accuracy": 0.11395471319556236,
|
|
"num_tokens": 276782.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.0256682395935055,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.5486,
|
|
"mean_token_accuracy": 0.11245746687054634,
|
|
"num_tokens": 287024.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.054018259048462,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.7399,
|
|
"mean_token_accuracy": 0.10906066820025444,
|
|
"num_tokens": 296597.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.0501950740814205,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.6133,
|
|
"mean_token_accuracy": 0.10526569485664368,
|
|
"num_tokens": 307709.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.057605266571045,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.6945,
|
|
"mean_token_accuracy": 0.11277502775192261,
|
|
"num_tokens": 317976.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.079123687744141,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.5933,
|
|
"mean_token_accuracy": 0.11109811812639236,
|
|
"num_tokens": 327319.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.128782653808594,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.7011,
|
|
"mean_token_accuracy": 0.10989727601408958,
|
|
"num_tokens": 337337.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.004915237426758,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.7058,
|
|
"mean_token_accuracy": 0.10382038801908493,
|
|
"num_tokens": 347779.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.241215610504151,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.7591,
|
|
"mean_token_accuracy": 0.10459929779171943,
|
|
"num_tokens": 358088.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.122637462615967,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.6817,
|
|
"mean_token_accuracy": 0.11470959857106208,
|
|
"num_tokens": 368314.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.984178066253662,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.6326,
|
|
"mean_token_accuracy": 0.11613894402980804,
|
|
"num_tokens": 378406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.127804279327393,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.6571,
|
|
"mean_token_accuracy": 0.10897257775068284,
|
|
"num_tokens": 388912.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.114015007019043,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.6681,
|
|
"mean_token_accuracy": 0.1109640821814537,
|
|
"num_tokens": 399290.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.044808387756348,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.6228,
|
|
"mean_token_accuracy": 0.1105850175023079,
|
|
"num_tokens": 409757.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.091672801971436,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.10671057477593422,
|
|
"num_tokens": 420182.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.1211034774780275,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.6939,
|
|
"mean_token_accuracy": 0.10947111174464226,
|
|
"num_tokens": 429897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.0094818592071535,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.603,
|
|
"mean_token_accuracy": 0.1111033283174038,
|
|
"num_tokens": 441131.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.01470251083374,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.5721,
|
|
"mean_token_accuracy": 0.1163232870399952,
|
|
"num_tokens": 451894.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.03377857208252,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.5484,
|
|
"mean_token_accuracy": 0.11207679361104965,
|
|
"num_tokens": 462613.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.951374053955078,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.625,
|
|
"mean_token_accuracy": 0.11543528139591216,
|
|
"num_tokens": 471606.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.040213966369629,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.5818,
|
|
"mean_token_accuracy": 0.11347164064645768,
|
|
"num_tokens": 481559.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.928437376022339,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.6276,
|
|
"mean_token_accuracy": 0.10829292684793472,
|
|
"num_tokens": 492766.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.141974353790284,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.6785,
|
|
"mean_token_accuracy": 0.11198576018214226,
|
|
"num_tokens": 503357.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.953282117843628,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.5097,
|
|
"mean_token_accuracy": 0.11739670410752297,
|
|
"num_tokens": 513098.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.012547636032105,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.5831,
|
|
"mean_token_accuracy": 0.10788592845201492,
|
|
"num_tokens": 522807.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.048117160797119,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.6139,
|
|
"mean_token_accuracy": 0.10738366693258286,
|
|
"num_tokens": 532670.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.000737524032592,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.5806,
|
|
"mean_token_accuracy": 0.11299150884151458,
|
|
"num_tokens": 543647.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.126269912719726,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.6435,
|
|
"mean_token_accuracy": 0.1134704515337944,
|
|
"num_tokens": 553047.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.067537927627564,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.701,
|
|
"mean_token_accuracy": 0.11085810139775276,
|
|
"num_tokens": 563413.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.03962550163269,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.6526,
|
|
"mean_token_accuracy": 0.1099765919148922,
|
|
"num_tokens": 573790.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.172714424133301,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.1152808465063572,
|
|
"num_tokens": 582958.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.927952623367309,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.12143387421965599,
|
|
"num_tokens": 592886.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.983848524093628,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.5702,
|
|
"mean_token_accuracy": 0.11255928203463554,
|
|
"num_tokens": 603692.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.033806657791137,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.6197,
|
|
"mean_token_accuracy": 0.11122355610132217,
|
|
"num_tokens": 613534.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.102828121185302,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.5846,
|
|
"mean_token_accuracy": 0.11281364113092422,
|
|
"num_tokens": 623862.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.014190912246704,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.6611,
|
|
"mean_token_accuracy": 0.1106049120426178,
|
|
"num_tokens": 633759.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.044203901290894,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.574,
|
|
"mean_token_accuracy": 0.11183998882770538,
|
|
"num_tokens": 643553.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.948228120803833,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.12053277567029,
|
|
"num_tokens": 653714.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.960380268096924,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.11574947088956833,
|
|
"num_tokens": 662817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.0484599590301515,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.572,
|
|
"mean_token_accuracy": 0.11281462013721466,
|
|
"num_tokens": 672381.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.860067987442017,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.4287,
|
|
"mean_token_accuracy": 0.12099420502781869,
|
|
"num_tokens": 682525.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.027956867218018,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.5667,
|
|
"mean_token_accuracy": 0.11456256210803986,
|
|
"num_tokens": 691826.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.891661071777344,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.5254,
|
|
"mean_token_accuracy": 0.12214417457580566,
|
|
"num_tokens": 701319.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.01544942855835,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.6239,
|
|
"mean_token_accuracy": 0.10927131026983261,
|
|
"num_tokens": 710888.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.045792198181152,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.11403941139578819,
|
|
"num_tokens": 722516.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.957839632034302,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.5389,
|
|
"mean_token_accuracy": 0.11641027331352234,
|
|
"num_tokens": 733058.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.970429754257202,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.6529,
|
|
"mean_token_accuracy": 0.11699234396219253,
|
|
"num_tokens": 744832.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.030797052383423,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.5626,
|
|
"mean_token_accuracy": 0.11702975407242774,
|
|
"num_tokens": 755387.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.831810474395752,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.508,
|
|
"mean_token_accuracy": 0.12129609361290931,
|
|
"num_tokens": 764455.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.996274089813232,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.5434,
|
|
"mean_token_accuracy": 0.11287878677248955,
|
|
"num_tokens": 774513.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.123194169998169,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6528,
|
|
"mean_token_accuracy": 0.11267266720533371,
|
|
"num_tokens": 785540.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.854208993911743,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.5076,
|
|
"mean_token_accuracy": 0.11473293825984002,
|
|
"num_tokens": 795233.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.00517725944519,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.5684,
|
|
"mean_token_accuracy": 0.1153998114168644,
|
|
"num_tokens": 804855.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.046209192276001,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5448,
|
|
"mean_token_accuracy": 0.12049437537789345,
|
|
"num_tokens": 814818.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.930202436447144,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.5723,
|
|
"mean_token_accuracy": 0.1219966672360897,
|
|
"num_tokens": 824622.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.988021183013916,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.461,
|
|
"mean_token_accuracy": 0.12495967298746109,
|
|
"num_tokens": 834813.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.910465860366822,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.581,
|
|
"mean_token_accuracy": 0.11608593091368676,
|
|
"num_tokens": 844887.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.913913106918335,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.5676,
|
|
"mean_token_accuracy": 0.11720148175954818,
|
|
"num_tokens": 855430.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.882201671600342,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.4109,
|
|
"mean_token_accuracy": 0.12106730788946152,
|
|
"num_tokens": 866052.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.8772341251373295,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5837,
|
|
"mean_token_accuracy": 0.1082363985478878,
|
|
"num_tokens": 877961.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.97900128364563,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.5385,
|
|
"mean_token_accuracy": 0.12003246918320656,
|
|
"num_tokens": 888386.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.899233198165893,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5949,
|
|
"mean_token_accuracy": 0.11746077984571457,
|
|
"num_tokens": 898272.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.9763165473937985,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5239,
|
|
"mean_token_accuracy": 0.12075713649392128,
|
|
"num_tokens": 908317.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.859544944763184,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5219,
|
|
"mean_token_accuracy": 0.11483893692493438,
|
|
"num_tokens": 917970.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.960592794418335,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5569,
|
|
"mean_token_accuracy": 0.11286390349268913,
|
|
"num_tokens": 928132.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.958032417297363,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.4882,
|
|
"mean_token_accuracy": 0.12300693169236183,
|
|
"num_tokens": 938092.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.903001070022583,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6277,
|
|
"mean_token_accuracy": 0.11332209631800652,
|
|
"num_tokens": 948650.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.926384496688843,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5358,
|
|
"mean_token_accuracy": 0.12150514349341393,
|
|
"num_tokens": 960195.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0210436344146725,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5531,
|
|
"mean_token_accuracy": 0.1211358867585659,
|
|
"num_tokens": 970608.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.86919755935669,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.11455759704113007,
|
|
"num_tokens": 981053.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.865674066543579,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.4754,
|
|
"mean_token_accuracy": 0.12632407173514365,
|
|
"num_tokens": 991560.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.9210834980010985,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5078,
|
|
"mean_token_accuracy": 0.11235889717936516,
|
|
"num_tokens": 1002008.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.893827056884765,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.514,
|
|
"mean_token_accuracy": 0.12223821952939033,
|
|
"num_tokens": 1012349.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.82147445678711,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.4946,
|
|
"mean_token_accuracy": 0.12133770138025284,
|
|
"num_tokens": 1022886.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 6.860337434813034,
|
|
"eval_loss": 6.596576690673828,
|
|
"eval_mean_token_accuracy": 0.11917702073101387,
|
|
"eval_num_tokens": 1022886.0,
|
|
"eval_runtime": 2.0788,
|
|
"eval_samples_per_second": 1652.391,
|
|
"eval_steps_per_second": 206.85,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.038683748245239,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.553,
|
|
"mean_token_accuracy": 0.11260545328259468,
|
|
"num_tokens": 1033129.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.75604133605957,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.4677,
|
|
"mean_token_accuracy": 0.12008674815297127,
|
|
"num_tokens": 1043069.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.91235318183899,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5074,
|
|
"mean_token_accuracy": 0.11930165812373161,
|
|
"num_tokens": 1054049.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.764765977859497,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4839,
|
|
"mean_token_accuracy": 0.11638501510024071,
|
|
"num_tokens": 1065155.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8717132091522215,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.4661,
|
|
"mean_token_accuracy": 0.12135470137000084,
|
|
"num_tokens": 1075510.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.840530157089233,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.492,
|
|
"mean_token_accuracy": 0.11994823887944221,
|
|
"num_tokens": 1086494.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.9312502384185795,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5277,
|
|
"mean_token_accuracy": 0.11823871508240699,
|
|
"num_tokens": 1097311.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.814186382293701,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4851,
|
|
"mean_token_accuracy": 0.12035745903849601,
|
|
"num_tokens": 1107757.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.742862415313721,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3936,
|
|
"mean_token_accuracy": 0.12132056951522827,
|
|
"num_tokens": 1118889.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.9436932563781735,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6034,
|
|
"mean_token_accuracy": 0.11310271173715591,
|
|
"num_tokens": 1128951.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.813915681838989,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4028,
|
|
"mean_token_accuracy": 0.12124144360423088,
|
|
"num_tokens": 1138583.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.7781116485595705,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.12076117545366287,
|
|
"num_tokens": 1149572.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.856851434707641,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.12212569490075112,
|
|
"num_tokens": 1160532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.8364509582519535,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4427,
|
|
"mean_token_accuracy": 0.11744874492287635,
|
|
"num_tokens": 1170915.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.748907423019409,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.1268433950841427,
|
|
"num_tokens": 1180822.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.846159839630127,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4444,
|
|
"mean_token_accuracy": 0.12061617448925972,
|
|
"num_tokens": 1190447.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.761949729919434,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.12140641659498215,
|
|
"num_tokens": 1200951.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.874054861068726,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3761,
|
|
"mean_token_accuracy": 0.12457154095172882,
|
|
"num_tokens": 1211846.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.5768075466156,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3485,
|
|
"mean_token_accuracy": 0.12247997224330902,
|
|
"num_tokens": 1222074.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.724390697479248,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4373,
|
|
"mean_token_accuracy": 0.11953089758753777,
|
|
"num_tokens": 1231741.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.838490581512451,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3576,
|
|
"mean_token_accuracy": 0.12071707099676132,
|
|
"num_tokens": 1241078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.763300848007202,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5515,
|
|
"mean_token_accuracy": 0.1141580305993557,
|
|
"num_tokens": 1251685.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.756213283538818,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4097,
|
|
"mean_token_accuracy": 0.12805262804031373,
|
|
"num_tokens": 1262044.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.798225259780883,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.11838038489222527,
|
|
"num_tokens": 1273647.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.815305852890015,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4587,
|
|
"mean_token_accuracy": 0.11831681355834008,
|
|
"num_tokens": 1284396.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.739588737487793,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4554,
|
|
"mean_token_accuracy": 0.11994302198290825,
|
|
"num_tokens": 1294339.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.743681716918945,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3618,
|
|
"mean_token_accuracy": 0.12565899565815924,
|
|
"num_tokens": 1304572.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.732248544692993,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4008,
|
|
"mean_token_accuracy": 0.1283697485923767,
|
|
"num_tokens": 1314780.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.715030288696289,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4305,
|
|
"mean_token_accuracy": 0.12055126056075097,
|
|
"num_tokens": 1325215.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.737424755096436,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.12564591094851493,
|
|
"num_tokens": 1334745.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.818739223480224,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4545,
|
|
"mean_token_accuracy": 0.12319393679499627,
|
|
"num_tokens": 1345150.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.748230791091919,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.1250821828842163,
|
|
"num_tokens": 1355619.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.675359725952148,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.12429328337311744,
|
|
"num_tokens": 1365933.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.677164220809937,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.3575,
|
|
"mean_token_accuracy": 0.12597684264183046,
|
|
"num_tokens": 1376867.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.880353355407715,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5086,
|
|
"mean_token_accuracy": 0.12199193090200425,
|
|
"num_tokens": 1386739.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.659671306610107,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3756,
|
|
"mean_token_accuracy": 0.12595050930976867,
|
|
"num_tokens": 1397138.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.794150686264038,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4737,
|
|
"mean_token_accuracy": 0.12353222742676735,
|
|
"num_tokens": 1408171.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.678361701965332,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4126,
|
|
"mean_token_accuracy": 0.11492947861552238,
|
|
"num_tokens": 1418259.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.736178636550903,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4123,
|
|
"mean_token_accuracy": 0.1267726466059685,
|
|
"num_tokens": 1428080.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.815822696685791,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4481,
|
|
"mean_token_accuracy": 0.12076391205191613,
|
|
"num_tokens": 1438502.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.598518657684326,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3924,
|
|
"mean_token_accuracy": 0.12057804539799691,
|
|
"num_tokens": 1448396.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.762733507156372,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.449,
|
|
"mean_token_accuracy": 0.11836672648787498,
|
|
"num_tokens": 1459728.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.694298982620239,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3468,
|
|
"mean_token_accuracy": 0.13057461455464364,
|
|
"num_tokens": 1469733.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.584758949279785,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.12913667634129525,
|
|
"num_tokens": 1479004.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.664322137832642,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3013,
|
|
"mean_token_accuracy": 0.12805228009819986,
|
|
"num_tokens": 1489848.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.619141530990601,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3101,
|
|
"mean_token_accuracy": 0.123420599848032,
|
|
"num_tokens": 1500877.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.760257053375244,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4531,
|
|
"mean_token_accuracy": 0.12081244513392449,
|
|
"num_tokens": 1511840.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.64377269744873,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3466,
|
|
"mean_token_accuracy": 0.12521926909685135,
|
|
"num_tokens": 1521839.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.686690950393677,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2627,
|
|
"mean_token_accuracy": 0.13056650534272193,
|
|
"num_tokens": 1531285.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.4388243675231935,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3123,
|
|
"mean_token_accuracy": 0.12405181005597114,
|
|
"num_tokens": 1542027.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.806895637512207,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4322,
|
|
"mean_token_accuracy": 0.1271710440516472,
|
|
"num_tokens": 1552527.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.741839790344239,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4587,
|
|
"mean_token_accuracy": 0.12079727575182915,
|
|
"num_tokens": 1562605.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.614172458648682,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3475,
|
|
"mean_token_accuracy": 0.1223350152373314,
|
|
"num_tokens": 1572128.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.568692779541015,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2926,
|
|
"mean_token_accuracy": 0.12322064191102981,
|
|
"num_tokens": 1582352.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.704865646362305,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3801,
|
|
"mean_token_accuracy": 0.1275831013917923,
|
|
"num_tokens": 1592887.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.567614984512329,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3225,
|
|
"mean_token_accuracy": 0.1287633515894413,
|
|
"num_tokens": 1603564.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.488887453079224,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.1348121538758278,
|
|
"num_tokens": 1613270.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.569432973861694,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2481,
|
|
"mean_token_accuracy": 0.13376593515276908,
|
|
"num_tokens": 1623009.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.583132648468018,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.12445440068840981,
|
|
"num_tokens": 1633209.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.60934681892395,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.13005920574069024,
|
|
"num_tokens": 1643708.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.673225450515747,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3891,
|
|
"mean_token_accuracy": 0.1240218348801136,
|
|
"num_tokens": 1653973.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.610533094406128,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3611,
|
|
"mean_token_accuracy": 0.12812820374965667,
|
|
"num_tokens": 1664856.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.514911937713623,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2889,
|
|
"mean_token_accuracy": 0.12775621116161345,
|
|
"num_tokens": 1675935.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.640262317657471,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3378,
|
|
"mean_token_accuracy": 0.12817456051707268,
|
|
"num_tokens": 1686597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.521266269683838,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.12973518893122674,
|
|
"num_tokens": 1697385.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.757512378692627,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.334,
|
|
"mean_token_accuracy": 0.12559274584054947,
|
|
"num_tokens": 1706680.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.525813150405884,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.13286243975162507,
|
|
"num_tokens": 1716130.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.544560527801513,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.266,
|
|
"mean_token_accuracy": 0.12106607109308243,
|
|
"num_tokens": 1726835.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.625137424468994,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2921,
|
|
"mean_token_accuracy": 0.13414542749524117,
|
|
"num_tokens": 1737708.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.551477527618408,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2803,
|
|
"mean_token_accuracy": 0.13080543354153634,
|
|
"num_tokens": 1747649.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.5627717018127445,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3104,
|
|
"mean_token_accuracy": 0.1241184689104557,
|
|
"num_tokens": 1758327.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.520473337173462,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2597,
|
|
"mean_token_accuracy": 0.12726373523473739,
|
|
"num_tokens": 1769767.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.632397890090942,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3682,
|
|
"mean_token_accuracy": 0.12802448347210885,
|
|
"num_tokens": 1780139.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.603162670135498,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2562,
|
|
"mean_token_accuracy": 0.1305392786860466,
|
|
"num_tokens": 1790783.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.480232000350952,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2484,
|
|
"mean_token_accuracy": 0.1271712563931942,
|
|
"num_tokens": 1801349.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.47876353263855,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2545,
|
|
"mean_token_accuracy": 0.12814290598034858,
|
|
"num_tokens": 1811990.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.468902921676635,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2642,
|
|
"mean_token_accuracy": 0.13316571116447448,
|
|
"num_tokens": 1821725.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.512246942520141,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2265,
|
|
"mean_token_accuracy": 0.12539276257157325,
|
|
"num_tokens": 1831794.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.56938853263855,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3102,
|
|
"mean_token_accuracy": 0.1250581920146942,
|
|
"num_tokens": 1842002.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.550431156158448,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2831,
|
|
"mean_token_accuracy": 0.13148240596055985,
|
|
"num_tokens": 1852282.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.505387592315674,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2905,
|
|
"mean_token_accuracy": 0.12270240858197212,
|
|
"num_tokens": 1862830.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.502415466308594,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2544,
|
|
"mean_token_accuracy": 0.1273520275950432,
|
|
"num_tokens": 1872746.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.531500625610351,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2788,
|
|
"mean_token_accuracy": 0.12095234766602517,
|
|
"num_tokens": 1884353.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.498415279388428,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3099,
|
|
"mean_token_accuracy": 0.12300471439957619,
|
|
"num_tokens": 1895719.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.56763768196106,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2461,
|
|
"mean_token_accuracy": 0.13558438569307327,
|
|
"num_tokens": 1905905.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.554039859771729,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3486,
|
|
"mean_token_accuracy": 0.1283061608672142,
|
|
"num_tokens": 1916010.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.483093357086181,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2355,
|
|
"mean_token_accuracy": 0.12846953123807908,
|
|
"num_tokens": 1926429.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.394514369964599,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.13184919580817223,
|
|
"num_tokens": 1936322.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.586344194412232,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3401,
|
|
"mean_token_accuracy": 0.12153217419981957,
|
|
"num_tokens": 1947885.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.581201982498169,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3061,
|
|
"mean_token_accuracy": 0.1271632768213749,
|
|
"num_tokens": 1957876.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.458595561981201,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.2884,
|
|
"mean_token_accuracy": 0.12774469405412675,
|
|
"num_tokens": 1969531.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.469849681854248,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1585,
|
|
"mean_token_accuracy": 0.13405076786875725,
|
|
"num_tokens": 1980787.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.426231050491333,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.13065283447504045,
|
|
"num_tokens": 1991645.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.458543157577514,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2898,
|
|
"mean_token_accuracy": 0.1293037176132202,
|
|
"num_tokens": 2002544.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.615034198760986,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3558,
|
|
"mean_token_accuracy": 0.12412808239459991,
|
|
"num_tokens": 2012813.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.439260530471802,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2133,
|
|
"mean_token_accuracy": 0.1342377170920372,
|
|
"num_tokens": 2023756.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.462012529373169,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2987,
|
|
"mean_token_accuracy": 0.12924248054623605,
|
|
"num_tokens": 2034245.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.596914148330688,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3289,
|
|
"mean_token_accuracy": 0.12585399001836778,
|
|
"num_tokens": 2045635.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.438915681838989,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2967,
|
|
"mean_token_accuracy": 0.12629913538694382,
|
|
"num_tokens": 2056501.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.50923523902893,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2941,
|
|
"mean_token_accuracy": 0.12796490490436555,
|
|
"num_tokens": 2066540.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.417881159449733,
|
|
"eval_loss": 6.327642440795898,
|
|
"eval_mean_token_accuracy": 0.12967643163405185,
|
|
"eval_num_tokens": 2066540.0,
|
|
"eval_runtime": 2.0068,
|
|
"eval_samples_per_second": 1711.686,
|
|
"eval_steps_per_second": 214.272,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.502445220947266,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.1874,
|
|
"mean_token_accuracy": 0.13226909786462784,
|
|
"num_tokens": 2077008.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.363942956924438,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.112,
|
|
"mean_token_accuracy": 0.13485410138964654,
|
|
"num_tokens": 2087314.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.450909662246704,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.2685,
|
|
"mean_token_accuracy": 0.13152633085846901,
|
|
"num_tokens": 2097928.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.427286386489868,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.12667798325419427,
|
|
"num_tokens": 2108662.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.409921503067016,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.1287694863975048,
|
|
"num_tokens": 2118546.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.523617219924927,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.2745,
|
|
"mean_token_accuracy": 0.12986955642700196,
|
|
"num_tokens": 2128012.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.3750663757324215,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.0978,
|
|
"mean_token_accuracy": 0.132510943710804,
|
|
"num_tokens": 2138241.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.501310014724732,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.2676,
|
|
"mean_token_accuracy": 0.12779484912753106,
|
|
"num_tokens": 2150780.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.38626651763916,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.1633,
|
|
"mean_token_accuracy": 0.12995605394244195,
|
|
"num_tokens": 2161918.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.3813934326171875,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.1939,
|
|
"mean_token_accuracy": 0.13298184126615525,
|
|
"num_tokens": 2172742.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.343666076660156,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.1768,
|
|
"mean_token_accuracy": 0.1389474593102932,
|
|
"num_tokens": 2182122.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.438647651672364,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.2143,
|
|
"mean_token_accuracy": 0.13138292282819747,
|
|
"num_tokens": 2193703.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.378941679000855,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.1416,
|
|
"mean_token_accuracy": 0.1320495679974556,
|
|
"num_tokens": 2204041.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.3784877300262455,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.1832,
|
|
"mean_token_accuracy": 0.13410646393895148,
|
|
"num_tokens": 2214276.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.416499233245849,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.1602,
|
|
"mean_token_accuracy": 0.1342863529920578,
|
|
"num_tokens": 2225390.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.348226547241211,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.1037,
|
|
"mean_token_accuracy": 0.13796225488185881,
|
|
"num_tokens": 2235708.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.427854537963867,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.1452,
|
|
"mean_token_accuracy": 0.13566545397043228,
|
|
"num_tokens": 2246284.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.466590881347656,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.161,
|
|
"mean_token_accuracy": 0.1317994177341461,
|
|
"num_tokens": 2257219.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.246234703063965,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.1683,
|
|
"mean_token_accuracy": 0.13266120254993438,
|
|
"num_tokens": 2268019.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.379768943786621,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.179,
|
|
"mean_token_accuracy": 0.13327732756733895,
|
|
"num_tokens": 2278382.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.361470010545519,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.13881485329733956,
|
|
"num_tokens": 2288515.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.365941667556763,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 5.8731,
|
|
"mean_token_accuracy": 0.14504974335432053,
|
|
"num_tokens": 2298396.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.24194712638855,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 5.9908,
|
|
"mean_token_accuracy": 0.1329639032483101,
|
|
"num_tokens": 2308707.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.378946971893311,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 5.914,
|
|
"mean_token_accuracy": 0.14582742899656295,
|
|
"num_tokens": 2318981.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.239332389831543,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.14213165417313575,
|
|
"num_tokens": 2329647.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.303018856048584,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 5.9275,
|
|
"mean_token_accuracy": 0.14211056828498841,
|
|
"num_tokens": 2339449.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.345061635971069,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 6.0366,
|
|
"mean_token_accuracy": 0.14273955971002578,
|
|
"num_tokens": 2349904.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.281744337081909,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 5.939,
|
|
"mean_token_accuracy": 0.14582237750291824,
|
|
"num_tokens": 2359833.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.344684028625489,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 5.9695,
|
|
"mean_token_accuracy": 0.1404454529285431,
|
|
"num_tokens": 2369857.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.369085693359375,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.14403127133846283,
|
|
"num_tokens": 2380372.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.24423360824585,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.13899227380752563,
|
|
"num_tokens": 2391239.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.212188577651977,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 5.9435,
|
|
"mean_token_accuracy": 0.1507773369550705,
|
|
"num_tokens": 2401178.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.409400558471679,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.025,
|
|
"mean_token_accuracy": 0.1460119053721428,
|
|
"num_tokens": 2412496.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.218855905532837,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 6.0186,
|
|
"mean_token_accuracy": 0.14455554634332657,
|
|
"num_tokens": 2423281.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.283839988708496,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 5.9554,
|
|
"mean_token_accuracy": 0.14712593257427214,
|
|
"num_tokens": 2434078.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.243132781982422,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 5.8953,
|
|
"mean_token_accuracy": 0.14542663395404815,
|
|
"num_tokens": 2443808.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.092622900009156,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 5.9289,
|
|
"mean_token_accuracy": 0.15093712210655214,
|
|
"num_tokens": 2453397.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.379008483886719,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.14048268944025039,
|
|
"num_tokens": 2464295.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.328823566436768,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 6.0681,
|
|
"mean_token_accuracy": 0.1382478453218937,
|
|
"num_tokens": 2475311.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.300699663162232,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 5.9802,
|
|
"mean_token_accuracy": 0.13811480179429053,
|
|
"num_tokens": 2485968.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.234206342697144,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 6.0021,
|
|
"mean_token_accuracy": 0.14477231800556184,
|
|
"num_tokens": 2496183.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.2193512439727785,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 5.9721,
|
|
"mean_token_accuracy": 0.1429495245218277,
|
|
"num_tokens": 2506242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.3535772323608395,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.1466970331966877,
|
|
"num_tokens": 2516443.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.175383520126343,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.1484525978565216,
|
|
"num_tokens": 2525795.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.23672137260437,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.1421683080494404,
|
|
"num_tokens": 2535046.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.270539140701294,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.1446902185678482,
|
|
"num_tokens": 2546489.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.0577202320098875,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 5.9018,
|
|
"mean_token_accuracy": 0.14584060609340668,
|
|
"num_tokens": 2556335.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.330905961990356,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.1401267394423485,
|
|
"num_tokens": 2567647.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.320643138885498,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 6.1235,
|
|
"mean_token_accuracy": 0.13722320273518562,
|
|
"num_tokens": 2578618.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.169188404083252,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 5.9388,
|
|
"mean_token_accuracy": 0.14894965440034866,
|
|
"num_tokens": 2587797.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.263793468475342,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 5.9789,
|
|
"mean_token_accuracy": 0.13785690888762475,
|
|
"num_tokens": 2598403.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.228110361099243,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 5.9828,
|
|
"mean_token_accuracy": 0.1393185056746006,
|
|
"num_tokens": 2608296.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.268101692199707,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 5.9961,
|
|
"mean_token_accuracy": 0.14061041846871375,
|
|
"num_tokens": 2617906.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.230483722686768,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.14462323039770125,
|
|
"num_tokens": 2627699.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.2316888809204105,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.1464667573571205,
|
|
"num_tokens": 2638384.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.133489370346069,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.14992269575595857,
|
|
"num_tokens": 2648279.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.188581752777099,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.13601700812578202,
|
|
"num_tokens": 2657000.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.2308509349823,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.1467374861240387,
|
|
"num_tokens": 2667760.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.164466762542725,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 5.8859,
|
|
"mean_token_accuracy": 0.14483800679445266,
|
|
"num_tokens": 2678450.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.205280447006226,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.13332329392433168,
|
|
"num_tokens": 2689695.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.318863582611084,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 6.0088,
|
|
"mean_token_accuracy": 0.14588286280632018,
|
|
"num_tokens": 2700414.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.0982919216156,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.15696054548025132,
|
|
"num_tokens": 2709985.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.167340230941773,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 5.9796,
|
|
"mean_token_accuracy": 0.14428043439984323,
|
|
"num_tokens": 2720990.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.1923376560211185,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.14820230454206468,
|
|
"num_tokens": 2731285.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.164142608642578,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.1433941274881363,
|
|
"num_tokens": 2741024.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.144341564178466,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.15126630812883377,
|
|
"num_tokens": 2751447.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.191232109069825,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 5.9203,
|
|
"mean_token_accuracy": 0.14985225796699525,
|
|
"num_tokens": 2761283.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.0593242168426515,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.8459,
|
|
"mean_token_accuracy": 0.15530473440885545,
|
|
"num_tokens": 2771354.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.294860458374023,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.0239,
|
|
"mean_token_accuracy": 0.14431079253554344,
|
|
"num_tokens": 2781063.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.217547655105591,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 5.9304,
|
|
"mean_token_accuracy": 0.1437186896800995,
|
|
"num_tokens": 2791103.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.219854211807251,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.144706778973341,
|
|
"num_tokens": 2802382.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.14369158744812,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.14514135867357253,
|
|
"num_tokens": 2813740.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.21032567024231,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.1452763110399246,
|
|
"num_tokens": 2824121.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.214183759689331,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.1472145602107048,
|
|
"num_tokens": 2833356.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.167957735061646,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.9203,
|
|
"mean_token_accuracy": 0.1406264141201973,
|
|
"num_tokens": 2843839.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.170833778381348,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.14574506133794785,
|
|
"num_tokens": 2853523.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.1578349590301515,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 5.9262,
|
|
"mean_token_accuracy": 0.1466248780488968,
|
|
"num_tokens": 2863226.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.242066144943237,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 5.9784,
|
|
"mean_token_accuracy": 0.1478915549814701,
|
|
"num_tokens": 2873822.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.164571857452392,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.859,
|
|
"mean_token_accuracy": 0.15025245249271393,
|
|
"num_tokens": 2883645.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.158694410324097,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 5.968,
|
|
"mean_token_accuracy": 0.14305587261915206,
|
|
"num_tokens": 2894397.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.159107875823975,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.9425,
|
|
"mean_token_accuracy": 0.14893015027046203,
|
|
"num_tokens": 2904947.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.179767179489136,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.14795035123825073,
|
|
"num_tokens": 2916183.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.076482915878296,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.83,
|
|
"mean_token_accuracy": 0.14453127086162568,
|
|
"num_tokens": 2925429.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.1859314918518065,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.1504214696586132,
|
|
"num_tokens": 2935142.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.141068983078003,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 5.9399,
|
|
"mean_token_accuracy": 0.15101734101772307,
|
|
"num_tokens": 2945151.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.25739860534668,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.0315,
|
|
"mean_token_accuracy": 0.14452726691961287,
|
|
"num_tokens": 2956015.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.128569793701172,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.8704,
|
|
"mean_token_accuracy": 0.14910778105258943,
|
|
"num_tokens": 2965879.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.2099706649780275,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 5.9521,
|
|
"mean_token_accuracy": 0.14465083330869674,
|
|
"num_tokens": 2977021.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.051621627807617,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.15206357687711716,
|
|
"num_tokens": 2987169.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.252909708023071,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.14722876995801926,
|
|
"num_tokens": 2997666.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.1304113388061525,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.834,
|
|
"mean_token_accuracy": 0.14632118940353395,
|
|
"num_tokens": 3007021.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.10150933265686,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.8954,
|
|
"mean_token_accuracy": 0.15438694655895233,
|
|
"num_tokens": 3016557.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.080778646469116,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.8474,
|
|
"mean_token_accuracy": 0.15140835493803023,
|
|
"num_tokens": 3026146.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.249139928817749,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 5.914,
|
|
"mean_token_accuracy": 0.14972909688949584,
|
|
"num_tokens": 3037234.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.060841608047485,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.8469,
|
|
"mean_token_accuracy": 0.15097369849681855,
|
|
"num_tokens": 3047282.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.0163311004638675,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.8775,
|
|
"mean_token_accuracy": 0.14596813321113586,
|
|
"num_tokens": 3058648.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.3922360897064205,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 6.0521,
|
|
"mean_token_accuracy": 0.14746224284172058,
|
|
"num_tokens": 3068996.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.0920412063598635,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.8706,
|
|
"mean_token_accuracy": 0.15449940860271455,
|
|
"num_tokens": 3078991.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.061806488037109,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.8313,
|
|
"mean_token_accuracy": 0.14856264963746071,
|
|
"num_tokens": 3090200.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.064583158493042,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 5.8824,
|
|
"mean_token_accuracy": 0.1481930769979954,
|
|
"num_tokens": 3101971.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 6.010310361551684,
|
|
"eval_loss": 6.112026691436768,
|
|
"eval_mean_token_accuracy": 0.14368421690928382,
|
|
"eval_num_tokens": 3101971.0,
|
|
"eval_runtime": 1.9918,
|
|
"eval_samples_per_second": 1724.537,
|
|
"eval_steps_per_second": 215.881,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.155653715133667,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 5.9744,
|
|
"mean_token_accuracy": 0.14244725480675696,
|
|
"num_tokens": 3113381.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.212191724777222,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.14757327735424042,
|
|
"num_tokens": 3125099.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.035371208190918,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.15051676481962203,
|
|
"num_tokens": 3135007.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.139113426208496,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.8755,
|
|
"mean_token_accuracy": 0.14846791476011276,
|
|
"num_tokens": 3145031.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.1009729385375975,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 5.9193,
|
|
"mean_token_accuracy": 0.14607925936579705,
|
|
"num_tokens": 3156044.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.117117691040039,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.8252,
|
|
"mean_token_accuracy": 0.1452489376068115,
|
|
"num_tokens": 3165124.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.991549634933472,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.7562,
|
|
"mean_token_accuracy": 0.15743231400847435,
|
|
"num_tokens": 3175246.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.174799633026123,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.15806265771389008,
|
|
"num_tokens": 3184817.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.039784955978393,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 5.8958,
|
|
"mean_token_accuracy": 0.13960993736982347,
|
|
"num_tokens": 3195142.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.100436353683472,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.8342,
|
|
"mean_token_accuracy": 0.15092215985059737,
|
|
"num_tokens": 3205724.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.136019706726074,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 5.9813,
|
|
"mean_token_accuracy": 0.143601393699646,
|
|
"num_tokens": 3216919.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.22254900932312,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 5.9848,
|
|
"mean_token_accuracy": 0.1447094537317753,
|
|
"num_tokens": 3228929.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.096653318405151,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.150317420065403,
|
|
"num_tokens": 3239541.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.10661678314209,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.1529161736369133,
|
|
"num_tokens": 3249527.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.007682275772095,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.7383,
|
|
"mean_token_accuracy": 0.14990039318799972,
|
|
"num_tokens": 3260497.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.1867749214172365,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 5.918,
|
|
"mean_token_accuracy": 0.14759416580200196,
|
|
"num_tokens": 3270865.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.039413547515869,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.14621682912111283,
|
|
"num_tokens": 3281531.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.135113286972046,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.8254,
|
|
"mean_token_accuracy": 0.14780823439359664,
|
|
"num_tokens": 3291970.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.045197772979736,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.8636,
|
|
"mean_token_accuracy": 0.15563471540808677,
|
|
"num_tokens": 3301638.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.2000038623809814,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.1489829644560814,
|
|
"num_tokens": 3311780.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.024671220779419,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.7979,
|
|
"mean_token_accuracy": 0.147994464635849,
|
|
"num_tokens": 3322096.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.203094100952148,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 5.971,
|
|
"mean_token_accuracy": 0.14461974278092385,
|
|
"num_tokens": 3332324.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.002854490280152,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.763,
|
|
"mean_token_accuracy": 0.16030256152153016,
|
|
"num_tokens": 3342037.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.113955974578857,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.8657,
|
|
"mean_token_accuracy": 0.1496028944849968,
|
|
"num_tokens": 3353047.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.031614971160889,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.7719,
|
|
"mean_token_accuracy": 0.14885980635881424,
|
|
"num_tokens": 3363841.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.0577257633209225,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.1518976554274559,
|
|
"num_tokens": 3374086.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.076325702667236,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.15219635143876076,
|
|
"num_tokens": 3384303.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.076239967346192,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.8189,
|
|
"mean_token_accuracy": 0.15623874068260193,
|
|
"num_tokens": 3394858.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.027254343032837,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.8267,
|
|
"mean_token_accuracy": 0.15381198078393937,
|
|
"num_tokens": 3405252.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.210509300231934,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.1505520910024643,
|
|
"num_tokens": 3416007.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.135774326324463,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.9187,
|
|
"mean_token_accuracy": 0.1479334443807602,
|
|
"num_tokens": 3426932.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.015619897842408,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.8104,
|
|
"mean_token_accuracy": 0.15175799578428267,
|
|
"num_tokens": 3437262.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.0340088367462155,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.7805,
|
|
"mean_token_accuracy": 0.1545313701033592,
|
|
"num_tokens": 3447525.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.12761287689209,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.15465754568576812,
|
|
"num_tokens": 3458198.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.002855825424194,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.769,
|
|
"mean_token_accuracy": 0.159763203561306,
|
|
"num_tokens": 3468149.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.033917808532715,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.7608,
|
|
"mean_token_accuracy": 0.15041886121034623,
|
|
"num_tokens": 3478355.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.037674856185913,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.7881,
|
|
"mean_token_accuracy": 0.1532315917313099,
|
|
"num_tokens": 3488726.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.994457054138183,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.7469,
|
|
"mean_token_accuracy": 0.15575082004070281,
|
|
"num_tokens": 3498346.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.982219219207764,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.15100184381008147,
|
|
"num_tokens": 3508213.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.1639477729797365,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.1517024278640747,
|
|
"num_tokens": 3518978.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.086548089981079,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.8769,
|
|
"mean_token_accuracy": 0.1542578913271427,
|
|
"num_tokens": 3529328.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.998582649230957,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.8118,
|
|
"mean_token_accuracy": 0.1479377903044224,
|
|
"num_tokens": 3539347.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.077225923538208,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.8254,
|
|
"mean_token_accuracy": 0.14645756483078004,
|
|
"num_tokens": 3549798.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.097193574905395,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.9228,
|
|
"mean_token_accuracy": 0.14081277772784234,
|
|
"num_tokens": 3560363.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.067219400405884,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.8596,
|
|
"mean_token_accuracy": 0.15149142742156982,
|
|
"num_tokens": 3570080.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.030013561248779,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.15077890902757646,
|
|
"num_tokens": 3580053.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.06266942024231,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.14908838123083115,
|
|
"num_tokens": 3590119.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.13274450302124,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.14301933497190475,
|
|
"num_tokens": 3601809.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.125389242172242,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.1557802826166153,
|
|
"num_tokens": 3611752.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.999754285812378,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.8378,
|
|
"mean_token_accuracy": 0.14936369210481643,
|
|
"num_tokens": 3621137.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.069888114929199,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8157,
|
|
"mean_token_accuracy": 0.15330246984958648,
|
|
"num_tokens": 3631567.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.983861780166626,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.7763,
|
|
"mean_token_accuracy": 0.15117551535367965,
|
|
"num_tokens": 3641145.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.04713807106018,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.8209,
|
|
"mean_token_accuracy": 0.14794272035360337,
|
|
"num_tokens": 3651716.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.039504051208496,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.7659,
|
|
"mean_token_accuracy": 0.15761386901140212,
|
|
"num_tokens": 3662577.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.967999887466431,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.7565,
|
|
"mean_token_accuracy": 0.15783901810646056,
|
|
"num_tokens": 3672984.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.058130693435669,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.8326,
|
|
"mean_token_accuracy": 0.15765131264925003,
|
|
"num_tokens": 3682634.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.012119722366333,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.803,
|
|
"mean_token_accuracy": 0.1533336967229843,
|
|
"num_tokens": 3693022.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.897359848022461,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.6836,
|
|
"mean_token_accuracy": 0.15874901562929153,
|
|
"num_tokens": 3702234.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.99068398475647,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.8666,
|
|
"mean_token_accuracy": 0.14502066224813462,
|
|
"num_tokens": 3712235.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.098167991638183,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.7923,
|
|
"mean_token_accuracy": 0.15776385068893434,
|
|
"num_tokens": 3722745.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.945577621459961,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.8415,
|
|
"mean_token_accuracy": 0.15472140908241272,
|
|
"num_tokens": 3732942.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.090262937545776,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.14960406571626664,
|
|
"num_tokens": 3743806.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.95607361793518,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.8201,
|
|
"mean_token_accuracy": 0.15639638900756836,
|
|
"num_tokens": 3755115.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.01966667175293,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.7753,
|
|
"mean_token_accuracy": 0.15850363969802855,
|
|
"num_tokens": 3765867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.953131103515625,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.7583,
|
|
"mean_token_accuracy": 0.15291280448436737,
|
|
"num_tokens": 3777071.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.994158983230591,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.7438,
|
|
"mean_token_accuracy": 0.15453884303569793,
|
|
"num_tokens": 3786554.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.0475998878479,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.15379742309451103,
|
|
"num_tokens": 3797627.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.9351438045501705,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.7941,
|
|
"mean_token_accuracy": 0.15232791006565094,
|
|
"num_tokens": 3808841.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.059771919250489,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.1511898323893547,
|
|
"num_tokens": 3818995.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.031847953796387,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.8849,
|
|
"mean_token_accuracy": 0.14733135402202607,
|
|
"num_tokens": 3828762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.046364068984985,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.7131,
|
|
"mean_token_accuracy": 0.1525560736656189,
|
|
"num_tokens": 3838752.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.9523547172546385,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.7588,
|
|
"mean_token_accuracy": 0.15982652306556702,
|
|
"num_tokens": 3849664.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.9417627334594725,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.7471,
|
|
"mean_token_accuracy": 0.162150639295578,
|
|
"num_tokens": 3859053.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.992360973358155,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.1509104624390602,
|
|
"num_tokens": 3869654.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.990663051605225,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.8171,
|
|
"mean_token_accuracy": 0.15092908442020417,
|
|
"num_tokens": 3879678.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.9989344596862795,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.7108,
|
|
"mean_token_accuracy": 0.15671169608831406,
|
|
"num_tokens": 3889950.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.903370571136475,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.7897,
|
|
"mean_token_accuracy": 0.1557111695408821,
|
|
"num_tokens": 3900095.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.113225984573364,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.148323355615139,
|
|
"num_tokens": 3911034.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.947522974014282,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.8189,
|
|
"mean_token_accuracy": 0.15346475541591645,
|
|
"num_tokens": 3922219.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.01598973274231,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.15475602000951766,
|
|
"num_tokens": 3932235.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.950323963165284,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.7209,
|
|
"mean_token_accuracy": 0.15549042522907258,
|
|
"num_tokens": 3941845.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.98919267654419,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.8179,
|
|
"mean_token_accuracy": 0.1464027062058449,
|
|
"num_tokens": 3953392.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.958113050460815,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.6751,
|
|
"mean_token_accuracy": 0.15961584746837615,
|
|
"num_tokens": 3963294.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.902626276016235,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.7461,
|
|
"mean_token_accuracy": 0.14919328689575195,
|
|
"num_tokens": 3973660.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.004216289520263,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.7435,
|
|
"mean_token_accuracy": 0.15620660185813903,
|
|
"num_tokens": 3983587.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.990758466720581,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.14908624291419983,
|
|
"num_tokens": 3994346.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.954130840301514,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.7775,
|
|
"mean_token_accuracy": 0.15563612207770347,
|
|
"num_tokens": 4005234.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.907217693328858,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.6992,
|
|
"mean_token_accuracy": 0.1574781894683838,
|
|
"num_tokens": 4015542.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.887025976181031,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.7166,
|
|
"mean_token_accuracy": 0.15762990117073059,
|
|
"num_tokens": 4024874.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.042590951919555,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.8557,
|
|
"mean_token_accuracy": 0.15325795263051986,
|
|
"num_tokens": 4035880.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.995141458511353,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.7431,
|
|
"mean_token_accuracy": 0.1556558609008789,
|
|
"num_tokens": 4046484.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.962204217910767,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.1540881484746933,
|
|
"num_tokens": 4057037.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.083400821685791,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.8815,
|
|
"mean_token_accuracy": 0.15016181543469428,
|
|
"num_tokens": 4066169.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.898530769348144,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.793,
|
|
"mean_token_accuracy": 0.14934501200914382,
|
|
"num_tokens": 4076697.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.959667587280274,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.6796,
|
|
"mean_token_accuracy": 0.15848930925130844,
|
|
"num_tokens": 4087372.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.010518646240234,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.7704,
|
|
"mean_token_accuracy": 0.1524614840745926,
|
|
"num_tokens": 4097653.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.97412519454956,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.7124,
|
|
"mean_token_accuracy": 0.15543478727340698,
|
|
"num_tokens": 4107381.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.963876867294312,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.8254,
|
|
"mean_token_accuracy": 0.15286498814821242,
|
|
"num_tokens": 4118762.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.960270643234253,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7181,
|
|
"mean_token_accuracy": 0.1565595179796219,
|
|
"num_tokens": 4129011.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.032435035705566,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.15029720216989517,
|
|
"num_tokens": 4138975.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.733916078611862,
|
|
"eval_loss": 5.95263147354126,
|
|
"eval_mean_token_accuracy": 0.15108996617586112,
|
|
"eval_num_tokens": 4138975.0,
|
|
"eval_runtime": 2.0045,
|
|
"eval_samples_per_second": 1713.663,
|
|
"eval_steps_per_second": 214.52,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.957431030273438,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.82,
|
|
"mean_token_accuracy": 0.14694613665342332,
|
|
"num_tokens": 4150164.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.8941168785095215,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.7174,
|
|
"mean_token_accuracy": 0.1589890643954277,
|
|
"num_tokens": 4159872.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.01587028503418,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.7925,
|
|
"mean_token_accuracy": 0.1506548911333084,
|
|
"num_tokens": 4170665.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.919941091537476,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.6741,
|
|
"mean_token_accuracy": 0.16530426144599913,
|
|
"num_tokens": 4180476.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.971033048629761,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.82,
|
|
"mean_token_accuracy": 0.1528342068195343,
|
|
"num_tokens": 4190740.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.93886661529541,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.6945,
|
|
"mean_token_accuracy": 0.160904423892498,
|
|
"num_tokens": 4201504.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.844774293899536,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.6354,
|
|
"mean_token_accuracy": 0.16651723235845567,
|
|
"num_tokens": 4211614.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.921957778930664,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.7582,
|
|
"mean_token_accuracy": 0.15619631707668305,
|
|
"num_tokens": 4221141.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.946128082275391,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.7979,
|
|
"mean_token_accuracy": 0.15570344030857086,
|
|
"num_tokens": 4231877.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.904908275604248,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.6137,
|
|
"mean_token_accuracy": 0.16376969814300538,
|
|
"num_tokens": 4242331.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.868312120437622,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.6306,
|
|
"mean_token_accuracy": 0.15482296124100686,
|
|
"num_tokens": 4252406.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.923862314224243,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.7188,
|
|
"mean_token_accuracy": 0.15500643700361252,
|
|
"num_tokens": 4263036.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.9105486392974855,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.6869,
|
|
"mean_token_accuracy": 0.16318628638982774,
|
|
"num_tokens": 4273032.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.891187477111816,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.7554,
|
|
"mean_token_accuracy": 0.15550417006015776,
|
|
"num_tokens": 4283360.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.914634132385254,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.7043,
|
|
"mean_token_accuracy": 0.16062143445014954,
|
|
"num_tokens": 4292623.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.949255800247192,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.7261,
|
|
"mean_token_accuracy": 0.16449115723371505,
|
|
"num_tokens": 4302817.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.9335943222045895,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.7298,
|
|
"mean_token_accuracy": 0.1570841759443283,
|
|
"num_tokens": 4313049.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.869911003112793,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.6044,
|
|
"mean_token_accuracy": 0.16380330324172973,
|
|
"num_tokens": 4322873.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.847801208496094,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.7607,
|
|
"mean_token_accuracy": 0.14720506817102433,
|
|
"num_tokens": 4333339.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.944580936431885,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.7406,
|
|
"mean_token_accuracy": 0.1549614280462265,
|
|
"num_tokens": 4344917.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.832040643692016,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.16366173475980758,
|
|
"num_tokens": 4354140.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.891859245300293,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.157329061627388,
|
|
"num_tokens": 4364976.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.9322466373443605,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.1643754079937935,
|
|
"num_tokens": 4376423.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.966183853149414,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.7472,
|
|
"mean_token_accuracy": 0.15724728852510453,
|
|
"num_tokens": 4387865.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.823707723617554,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.15898389965295792,
|
|
"num_tokens": 4398275.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.970320463180542,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.6557,
|
|
"mean_token_accuracy": 0.1616579294204712,
|
|
"num_tokens": 4407396.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.037105226516724,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.9137,
|
|
"mean_token_accuracy": 0.14634885564446448,
|
|
"num_tokens": 4419682.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.931360530853271,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.7751,
|
|
"mean_token_accuracy": 0.1550420716404915,
|
|
"num_tokens": 4430590.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.982472848892212,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.693,
|
|
"mean_token_accuracy": 0.16307468116283416,
|
|
"num_tokens": 4441979.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.868456172943115,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.7046,
|
|
"mean_token_accuracy": 0.15840312838554382,
|
|
"num_tokens": 4451975.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.8398841381072994,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.6429,
|
|
"mean_token_accuracy": 0.16072934120893478,
|
|
"num_tokens": 4462778.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.95463080406189,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.7018,
|
|
"mean_token_accuracy": 0.16682704985141755,
|
|
"num_tokens": 4472461.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.858252239227295,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.654,
|
|
"mean_token_accuracy": 0.16354322135448457,
|
|
"num_tokens": 4482507.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.905764532089234,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.6723,
|
|
"mean_token_accuracy": 0.16208890974521636,
|
|
"num_tokens": 4492601.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.906562376022339,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.7843,
|
|
"mean_token_accuracy": 0.1586180418729782,
|
|
"num_tokens": 4503696.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.909030771255493,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.6686,
|
|
"mean_token_accuracy": 0.15772291123867035,
|
|
"num_tokens": 4513649.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.879385900497437,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.7026,
|
|
"mean_token_accuracy": 0.1572379268705845,
|
|
"num_tokens": 4524054.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.965790462493897,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.15851562321186066,
|
|
"num_tokens": 4535549.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.887549543380738,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.7269,
|
|
"mean_token_accuracy": 0.1520995393395424,
|
|
"num_tokens": 4545654.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.891722345352173,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.7214,
|
|
"mean_token_accuracy": 0.16194723248481752,
|
|
"num_tokens": 4556230.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.862508985731337,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.16909081571631962,
|
|
"num_tokens": 4564486.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.866947984695434,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.3756,
|
|
"mean_token_accuracy": 0.17403068095445634,
|
|
"num_tokens": 4575026.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.7502704620361325,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.16631340384483337,
|
|
"num_tokens": 4585902.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.013295221328735,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.5023,
|
|
"mean_token_accuracy": 0.17075931578874587,
|
|
"num_tokens": 4597609.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.831322860717774,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.3796,
|
|
"mean_token_accuracy": 0.17327053993940353,
|
|
"num_tokens": 4607531.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.847760105133057,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.4172,
|
|
"mean_token_accuracy": 0.17092780470848085,
|
|
"num_tokens": 4618826.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.864540863037109,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4269,
|
|
"mean_token_accuracy": 0.17099900394678116,
|
|
"num_tokens": 4629526.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.883844804763794,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.4609,
|
|
"mean_token_accuracy": 0.17029182761907577,
|
|
"num_tokens": 4639930.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.8077638149261475,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.3466,
|
|
"mean_token_accuracy": 0.1779349371790886,
|
|
"num_tokens": 4649661.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.753513050079346,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.357,
|
|
"mean_token_accuracy": 0.1814907506108284,
|
|
"num_tokens": 4660178.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.828481006622314,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.17419430166482924,
|
|
"num_tokens": 4670417.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.771015071868897,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.3595,
|
|
"mean_token_accuracy": 0.18370799124240875,
|
|
"num_tokens": 4680221.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.85000958442688,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.17064266949892043,
|
|
"num_tokens": 4691136.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.672495126724243,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.2621,
|
|
"mean_token_accuracy": 0.18813103139400483,
|
|
"num_tokens": 4700879.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.774118232727051,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.16868412047624587,
|
|
"num_tokens": 4711477.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.8024450778961185,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.1725125789642334,
|
|
"num_tokens": 4721973.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.740260934829712,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.409,
|
|
"mean_token_accuracy": 0.16820381283760072,
|
|
"num_tokens": 4732425.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.8100550174713135,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.3775,
|
|
"mean_token_accuracy": 0.17511506080627443,
|
|
"num_tokens": 4743377.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.785009717941284,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.4379,
|
|
"mean_token_accuracy": 0.17519502490758895,
|
|
"num_tokens": 4753534.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.75093822479248,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.3868,
|
|
"mean_token_accuracy": 0.17328918129205703,
|
|
"num_tokens": 4763159.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.745303726196289,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.3603,
|
|
"mean_token_accuracy": 0.17843883931636811,
|
|
"num_tokens": 4773497.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.789405584335327,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.4564,
|
|
"mean_token_accuracy": 0.17826216220855712,
|
|
"num_tokens": 4784291.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.762081146240234,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4045,
|
|
"mean_token_accuracy": 0.17297512739896775,
|
|
"num_tokens": 4794762.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.904514455795288,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.5211,
|
|
"mean_token_accuracy": 0.17019670754671096,
|
|
"num_tokens": 4804449.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.740280532836914,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.1749880075454712,
|
|
"num_tokens": 4814172.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.781768560409546,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.4765,
|
|
"mean_token_accuracy": 0.17494874447584152,
|
|
"num_tokens": 4824909.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.897498083114624,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5098,
|
|
"mean_token_accuracy": 0.17679391503334047,
|
|
"num_tokens": 4834546.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.77701964378357,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.4181,
|
|
"mean_token_accuracy": 0.17149554044008256,
|
|
"num_tokens": 4844809.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.741995573043823,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.4035,
|
|
"mean_token_accuracy": 0.17329975068569184,
|
|
"num_tokens": 4855883.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.856533098220825,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.4256,
|
|
"mean_token_accuracy": 0.18001650273799896,
|
|
"num_tokens": 4865936.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.661838531494141,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.3683,
|
|
"mean_token_accuracy": 0.1719723716378212,
|
|
"num_tokens": 4876621.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.629975748062134,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.16937130242586135,
|
|
"num_tokens": 4886315.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.793085336685181,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.17354542315006255,
|
|
"num_tokens": 4897214.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.7971173286437985,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.17415029108524321,
|
|
"num_tokens": 4907360.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.738719272613525,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.3742,
|
|
"mean_token_accuracy": 0.17936114966869354,
|
|
"num_tokens": 4917591.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.751534557342529,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.4667,
|
|
"mean_token_accuracy": 0.1628737047314644,
|
|
"num_tokens": 4928075.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.775207567214966,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.364,
|
|
"mean_token_accuracy": 0.17666680067777635,
|
|
"num_tokens": 4937898.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.718432950973511,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.4201,
|
|
"mean_token_accuracy": 0.16961177289485932,
|
|
"num_tokens": 4948152.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.774542379379272,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4039,
|
|
"mean_token_accuracy": 0.1735743001103401,
|
|
"num_tokens": 4957973.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.6395844459533695,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.37,
|
|
"mean_token_accuracy": 0.1769695222377777,
|
|
"num_tokens": 4967375.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.748891401290893,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.375,
|
|
"mean_token_accuracy": 0.17679121643304824,
|
|
"num_tokens": 4977286.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.70573058128357,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.16882054805755614,
|
|
"num_tokens": 4987930.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.874832057952881,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.17146590054035188,
|
|
"num_tokens": 4998707.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.6762388229370115,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.3429,
|
|
"mean_token_accuracy": 0.18256889432668685,
|
|
"num_tokens": 5008540.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.65377779006958,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4037,
|
|
"mean_token_accuracy": 0.1740575209259987,
|
|
"num_tokens": 5018199.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.783184003829956,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.4313,
|
|
"mean_token_accuracy": 0.17550634741783142,
|
|
"num_tokens": 5028251.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.690026521682739,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.4305,
|
|
"mean_token_accuracy": 0.1661398559808731,
|
|
"num_tokens": 5038290.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.754314851760864,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.3656,
|
|
"mean_token_accuracy": 0.17466755211353302,
|
|
"num_tokens": 5048337.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.794387769699097,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.4149,
|
|
"mean_token_accuracy": 0.16921241730451583,
|
|
"num_tokens": 5058533.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.736860704421997,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.425,
|
|
"mean_token_accuracy": 0.17878113687038422,
|
|
"num_tokens": 5068556.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.705081224441528,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.17241994738578797,
|
|
"num_tokens": 5079317.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.857335472106934,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.17653426229953767,
|
|
"num_tokens": 5089544.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.69089994430542,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.16911610066890717,
|
|
"num_tokens": 5099491.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.829253530502319,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.17039169669151305,
|
|
"num_tokens": 5109523.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.759532070159912,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.3963,
|
|
"mean_token_accuracy": 0.17784518748521805,
|
|
"num_tokens": 5119940.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.748552751541138,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.1798911511898041,
|
|
"num_tokens": 5129789.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.706392288208008,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.17321959882974625,
|
|
"num_tokens": 5139570.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.681578493118286,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.3826,
|
|
"mean_token_accuracy": 0.16976896524429322,
|
|
"num_tokens": 5150526.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.695160961151123,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.3188,
|
|
"mean_token_accuracy": 0.17343409210443497,
|
|
"num_tokens": 5160895.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.688534450531006,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.322,
|
|
"mean_token_accuracy": 0.17689472585916519,
|
|
"num_tokens": 5171054.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.541668576972429,
|
|
"eval_loss": 5.8642683029174805,
|
|
"eval_mean_token_accuracy": 0.15643741834129013,
|
|
"eval_num_tokens": 5171054.0,
|
|
"eval_runtime": 1.9972,
|
|
"eval_samples_per_second": 1719.94,
|
|
"eval_steps_per_second": 215.305,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.750526094436646,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.4866,
|
|
"mean_token_accuracy": 0.161210834980011,
|
|
"num_tokens": 5180870.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.781940460205078,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.4249,
|
|
"mean_token_accuracy": 0.17138221561908723,
|
|
"num_tokens": 5190388.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.754351997375489,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.16879990994930266,
|
|
"num_tokens": 5200582.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.702089405059814,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.3734,
|
|
"mean_token_accuracy": 0.17459774017333984,
|
|
"num_tokens": 5211079.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.780581617355347,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.5082,
|
|
"mean_token_accuracy": 0.1652182012796402,
|
|
"num_tokens": 5221481.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.765790128707886,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.3764,
|
|
"mean_token_accuracy": 0.18041289746761321,
|
|
"num_tokens": 5231387.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.630071687698364,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.3328,
|
|
"mean_token_accuracy": 0.17884562611579896,
|
|
"num_tokens": 5240758.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.711175584793091,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.3573,
|
|
"mean_token_accuracy": 0.17342581897974013,
|
|
"num_tokens": 5250839.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.805914258956909,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.17285922318696975,
|
|
"num_tokens": 5261554.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.713011932373047,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.4246,
|
|
"mean_token_accuracy": 0.17515779584646224,
|
|
"num_tokens": 5272422.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.695078659057617,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.358,
|
|
"mean_token_accuracy": 0.16814109086990356,
|
|
"num_tokens": 5282522.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.746765947341919,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.17494575083255767,
|
|
"num_tokens": 5293100.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.798492383956909,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.4457,
|
|
"mean_token_accuracy": 0.16868271827697753,
|
|
"num_tokens": 5303103.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.739025545120239,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.1802881196141243,
|
|
"num_tokens": 5313765.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.7982371807098385,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.5092,
|
|
"mean_token_accuracy": 0.1671261727809906,
|
|
"num_tokens": 5324816.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.689096212387085,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.3867,
|
|
"mean_token_accuracy": 0.17445023208856583,
|
|
"num_tokens": 5334882.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.703138446807861,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.3817,
|
|
"mean_token_accuracy": 0.17914554625749587,
|
|
"num_tokens": 5344623.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.751450252532959,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.17148876786231995,
|
|
"num_tokens": 5355786.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.627918767929077,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.4021,
|
|
"mean_token_accuracy": 0.17087053656578063,
|
|
"num_tokens": 5366139.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.862655878067017,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.1666705295443535,
|
|
"num_tokens": 5376890.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.688345241546631,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4076,
|
|
"mean_token_accuracy": 0.17868503630161287,
|
|
"num_tokens": 5386602.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.764010620117188,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.4772,
|
|
"mean_token_accuracy": 0.17160866558551788,
|
|
"num_tokens": 5397749.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.749241018295288,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.1673075184226036,
|
|
"num_tokens": 5407666.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.735607671737671,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5142,
|
|
"mean_token_accuracy": 0.16675720661878585,
|
|
"num_tokens": 5418668.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.680788135528564,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.3883,
|
|
"mean_token_accuracy": 0.17995792180299758,
|
|
"num_tokens": 5429098.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.669166088104248,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.3324,
|
|
"mean_token_accuracy": 0.17956721186637878,
|
|
"num_tokens": 5438997.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.698221731185913,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.3997,
|
|
"mean_token_accuracy": 0.17252164483070373,
|
|
"num_tokens": 5449315.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.705436658859253,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4358,
|
|
"mean_token_accuracy": 0.1685012996196747,
|
|
"num_tokens": 5460523.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.762444067001343,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.4614,
|
|
"mean_token_accuracy": 0.1750774085521698,
|
|
"num_tokens": 5470012.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.695530652999878,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4067,
|
|
"mean_token_accuracy": 0.17659069299697877,
|
|
"num_tokens": 5480364.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.823437452316284,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.5352,
|
|
"mean_token_accuracy": 0.1670384019613266,
|
|
"num_tokens": 5492138.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.681525564193725,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.3091,
|
|
"mean_token_accuracy": 0.18115043491125107,
|
|
"num_tokens": 5501935.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.701013469696045,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.16933519542217254,
|
|
"num_tokens": 5513993.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.753058958053589,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.1767013266682625,
|
|
"num_tokens": 5524822.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.75729718208313,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.4127,
|
|
"mean_token_accuracy": 0.16921148002147673,
|
|
"num_tokens": 5535583.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.675585603713989,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.3964,
|
|
"mean_token_accuracy": 0.1775403916835785,
|
|
"num_tokens": 5546199.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.69486632347107,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.3642,
|
|
"mean_token_accuracy": 0.17965541183948516,
|
|
"num_tokens": 5556409.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.725578355789184,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.17723667919635772,
|
|
"num_tokens": 5565854.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.748017168045044,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.4639,
|
|
"mean_token_accuracy": 0.16633473336696625,
|
|
"num_tokens": 5576194.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.706389665603638,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.3706,
|
|
"mean_token_accuracy": 0.17574018239974976,
|
|
"num_tokens": 5586679.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.6699425220489506,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.4121,
|
|
"mean_token_accuracy": 0.16650531440973282,
|
|
"num_tokens": 5597268.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.722821760177612,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.3871,
|
|
"mean_token_accuracy": 0.17458881437778473,
|
|
"num_tokens": 5608046.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.689449501037598,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.31,
|
|
"mean_token_accuracy": 0.17710677087306975,
|
|
"num_tokens": 5617541.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.7190234661102295,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.3635,
|
|
"mean_token_accuracy": 0.16786014437675476,
|
|
"num_tokens": 5628428.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.757678413391114,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.5451,
|
|
"mean_token_accuracy": 0.15948943719267844,
|
|
"num_tokens": 5639557.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.786512804031372,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.5296,
|
|
"mean_token_accuracy": 0.17010488212108613,
|
|
"num_tokens": 5650586.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.750928783416748,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.428,
|
|
"mean_token_accuracy": 0.17482876777648926,
|
|
"num_tokens": 5661525.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.733806705474853,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4546,
|
|
"mean_token_accuracy": 0.16926867961883546,
|
|
"num_tokens": 5671272.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.708602237701416,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.3802,
|
|
"mean_token_accuracy": 0.17592664510011674,
|
|
"num_tokens": 5680384.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.721498632431031,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.401,
|
|
"mean_token_accuracy": 0.1755047246813774,
|
|
"num_tokens": 5690983.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.691464233398437,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.4169,
|
|
"mean_token_accuracy": 0.16735788732767104,
|
|
"num_tokens": 5702144.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.688137054443359,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.16610317528247834,
|
|
"num_tokens": 5713283.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.680425405502319,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4358,
|
|
"mean_token_accuracy": 0.16601063907146454,
|
|
"num_tokens": 5724533.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.711917686462402,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.3873,
|
|
"mean_token_accuracy": 0.17389980256557463,
|
|
"num_tokens": 5735186.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.721314525604248,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.17304984778165816,
|
|
"num_tokens": 5746537.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.734245824813843,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.4203,
|
|
"mean_token_accuracy": 0.17387858033180237,
|
|
"num_tokens": 5756666.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.7586259841918945,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.4455,
|
|
"mean_token_accuracy": 0.17220781594514847,
|
|
"num_tokens": 5766870.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.695646190643311,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4351,
|
|
"mean_token_accuracy": 0.17539490461349488,
|
|
"num_tokens": 5776775.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.776819038391113,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.4271,
|
|
"mean_token_accuracy": 0.17059330493211747,
|
|
"num_tokens": 5786993.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.665747785568238,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4193,
|
|
"mean_token_accuracy": 0.17776434123516083,
|
|
"num_tokens": 5796230.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.786862230300903,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.4103,
|
|
"mean_token_accuracy": 0.17467432469129562,
|
|
"num_tokens": 5807086.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.641618251800537,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.4449,
|
|
"mean_token_accuracy": 0.17003494650125503,
|
|
"num_tokens": 5817897.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.739160108566284,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.3422,
|
|
"mean_token_accuracy": 0.18196709752082824,
|
|
"num_tokens": 5827794.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.618197011947632,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.4253,
|
|
"mean_token_accuracy": 0.1728539928793907,
|
|
"num_tokens": 5837603.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.6844886302947994,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.3399,
|
|
"mean_token_accuracy": 0.1776605576276779,
|
|
"num_tokens": 5847013.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.677564477920532,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.3449,
|
|
"mean_token_accuracy": 0.1727038323879242,
|
|
"num_tokens": 5856783.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.68329758644104,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.3387,
|
|
"mean_token_accuracy": 0.18341223597526551,
|
|
"num_tokens": 5866241.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.6050855159759525,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.3906,
|
|
"mean_token_accuracy": 0.1770755559206009,
|
|
"num_tokens": 5876135.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.68485279083252,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.1759236678481102,
|
|
"num_tokens": 5886841.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.7248931407928465,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.3912,
|
|
"mean_token_accuracy": 0.17576023787260056,
|
|
"num_tokens": 5896884.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.649676322937012,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.4261,
|
|
"mean_token_accuracy": 0.17579701840877532,
|
|
"num_tokens": 5906194.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.673415470123291,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.3524,
|
|
"mean_token_accuracy": 0.1809778705239296,
|
|
"num_tokens": 5917217.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.642754220962525,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.17782741785049438,
|
|
"num_tokens": 5926307.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.656463956832885,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.3875,
|
|
"mean_token_accuracy": 0.1743921473622322,
|
|
"num_tokens": 5936718.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.604428958892822,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.35,
|
|
"mean_token_accuracy": 0.17654486894607543,
|
|
"num_tokens": 5947870.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.674286794662476,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.5238,
|
|
"mean_token_accuracy": 0.17045848071575165,
|
|
"num_tokens": 5959090.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.770153999328613,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.16692567467689515,
|
|
"num_tokens": 5970435.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.686746835708618,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.3168,
|
|
"mean_token_accuracy": 0.18975419849157332,
|
|
"num_tokens": 5979732.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.674428415298462,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.17490168213844298,
|
|
"num_tokens": 5989965.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.7551796436309814,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.4457,
|
|
"mean_token_accuracy": 0.17994607239961624,
|
|
"num_tokens": 6000752.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.6631536960601805,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.4078,
|
|
"mean_token_accuracy": 0.17297106683254243,
|
|
"num_tokens": 6011318.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.683416843414307,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4123,
|
|
"mean_token_accuracy": 0.17364877313375474,
|
|
"num_tokens": 6022024.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.628732824325562,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.373,
|
|
"mean_token_accuracy": 0.17655983418226243,
|
|
"num_tokens": 6032118.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.689882612228393,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.3866,
|
|
"mean_token_accuracy": 0.18142862915992736,
|
|
"num_tokens": 6042607.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.61193356513977,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.3967,
|
|
"mean_token_accuracy": 0.1752897545695305,
|
|
"num_tokens": 6053625.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.695006561279297,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.3518,
|
|
"mean_token_accuracy": 0.18216099143028258,
|
|
"num_tokens": 6063823.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.661330366134644,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.3518,
|
|
"mean_token_accuracy": 0.17229808866977692,
|
|
"num_tokens": 6073914.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.658997678756714,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.3349,
|
|
"mean_token_accuracy": 0.17754623740911485,
|
|
"num_tokens": 6085121.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.612066221237183,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.261,
|
|
"mean_token_accuracy": 0.18452934622764589,
|
|
"num_tokens": 6095324.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.62063684463501,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.3896,
|
|
"mean_token_accuracy": 0.1811824172735214,
|
|
"num_tokens": 6104704.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.682147741317749,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.16916490495204925,
|
|
"num_tokens": 6116569.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.7397918701171875,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.1748810663819313,
|
|
"num_tokens": 6126166.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.68087682723999,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.18225707560777665,
|
|
"num_tokens": 6136169.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.696323394775391,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.17932215034961702,
|
|
"num_tokens": 6146249.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.750275993347168,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.17407476156949997,
|
|
"num_tokens": 6156892.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.737551021575928,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.3638,
|
|
"mean_token_accuracy": 0.18510779440402986,
|
|
"num_tokens": 6166915.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.62639627456665,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.2982,
|
|
"mean_token_accuracy": 0.18099016845226287,
|
|
"num_tokens": 6176837.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.587598085403442,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.3918,
|
|
"mean_token_accuracy": 0.17472818940877916,
|
|
"num_tokens": 6188038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.667161703109741,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.1745101973414421,
|
|
"num_tokens": 6199835.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.750896167755127,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.3499,
|
|
"mean_token_accuracy": 0.17623832821846008,
|
|
"num_tokens": 6209122.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.453245504512343,
|
|
"eval_loss": 5.784785747528076,
|
|
"eval_mean_token_accuracy": 0.16304942005595496,
|
|
"eval_num_tokens": 6209122.0,
|
|
"eval_runtime": 2.0071,
|
|
"eval_samples_per_second": 1711.42,
|
|
"eval_steps_per_second": 214.239,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.6080889225006105,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4026,
|
|
"mean_token_accuracy": 0.17455950230360032,
|
|
"num_tokens": 6219317.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.703277206420898,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.4271,
|
|
"mean_token_accuracy": 0.176008839905262,
|
|
"num_tokens": 6229561.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.705724906921387,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.4027,
|
|
"mean_token_accuracy": 0.17291686087846755,
|
|
"num_tokens": 6240150.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.610035562515259,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.18002589493989946,
|
|
"num_tokens": 6250641.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.635480499267578,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.3878,
|
|
"mean_token_accuracy": 0.16999149918556214,
|
|
"num_tokens": 6261239.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.697523498535157,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.389,
|
|
"mean_token_accuracy": 0.1768403857946396,
|
|
"num_tokens": 6270919.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.7210699081420895,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.4686,
|
|
"mean_token_accuracy": 0.16550424695014954,
|
|
"num_tokens": 6283007.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.5354899883270265,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.18222982287406922,
|
|
"num_tokens": 6293596.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.652623891830444,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.3718,
|
|
"mean_token_accuracy": 0.1778726100921631,
|
|
"num_tokens": 6304568.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.651189088821411,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.3416,
|
|
"mean_token_accuracy": 0.17796921730041504,
|
|
"num_tokens": 6315275.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.692053127288818,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.4535,
|
|
"mean_token_accuracy": 0.17656845301389695,
|
|
"num_tokens": 6324986.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.660140800476074,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.4493,
|
|
"mean_token_accuracy": 0.16845283061265945,
|
|
"num_tokens": 6336318.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.602566337585449,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.3237,
|
|
"mean_token_accuracy": 0.18113580495119094,
|
|
"num_tokens": 6345986.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.6237877368927,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.4042,
|
|
"mean_token_accuracy": 0.1757053703069687,
|
|
"num_tokens": 6357025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.704830837249756,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.4035,
|
|
"mean_token_accuracy": 0.1736297681927681,
|
|
"num_tokens": 6367885.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.54007773399353,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.3366,
|
|
"mean_token_accuracy": 0.1801164522767067,
|
|
"num_tokens": 6377729.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.600256299972534,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.3303,
|
|
"mean_token_accuracy": 0.17931140810251237,
|
|
"num_tokens": 6387368.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.675865364074707,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.17015431225299835,
|
|
"num_tokens": 6397771.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.682026052474976,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.3057,
|
|
"mean_token_accuracy": 0.1849225789308548,
|
|
"num_tokens": 6408238.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.673222589492798,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.4458,
|
|
"mean_token_accuracy": 0.16830612868070602,
|
|
"num_tokens": 6419196.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.629798126220703,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.4008,
|
|
"mean_token_accuracy": 0.17280980944633484,
|
|
"num_tokens": 6430342.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.650047159194946,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.2996,
|
|
"mean_token_accuracy": 0.18436145782470703,
|
|
"num_tokens": 6440218.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.607442235946655,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.302,
|
|
"mean_token_accuracy": 0.18424032330513002,
|
|
"num_tokens": 6450128.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.5780956745147705,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.3012,
|
|
"mean_token_accuracy": 0.18950312286615373,
|
|
"num_tokens": 6459916.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.6977263450622555,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.3548,
|
|
"mean_token_accuracy": 0.17936145961284639,
|
|
"num_tokens": 6470250.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.6951452732086185,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.17187969088554383,
|
|
"num_tokens": 6481623.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.640903949737549,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.3481,
|
|
"mean_token_accuracy": 0.17956022173166275,
|
|
"num_tokens": 6492143.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.665499019622803,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.3607,
|
|
"mean_token_accuracy": 0.1813562586903572,
|
|
"num_tokens": 6501643.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.587842512130737,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.3037,
|
|
"mean_token_accuracy": 0.1881158947944641,
|
|
"num_tokens": 6511442.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.587015914916992,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.2357,
|
|
"mean_token_accuracy": 0.18846675604581833,
|
|
"num_tokens": 6521303.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.584110736846924,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.3678,
|
|
"mean_token_accuracy": 0.17428743839263916,
|
|
"num_tokens": 6531143.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.695782089233399,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.1739306181669235,
|
|
"num_tokens": 6540899.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.757632493972778,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.4668,
|
|
"mean_token_accuracy": 0.17104335576295854,
|
|
"num_tokens": 6552131.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.586829948425293,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.3952,
|
|
"mean_token_accuracy": 0.17037197798490525,
|
|
"num_tokens": 6563275.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.6519371509552006,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.2532,
|
|
"mean_token_accuracy": 0.18573469817638397,
|
|
"num_tokens": 6573109.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.556792402267456,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.3776,
|
|
"mean_token_accuracy": 0.1706434965133667,
|
|
"num_tokens": 6584978.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.572997808456421,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.3734,
|
|
"mean_token_accuracy": 0.17502374947071075,
|
|
"num_tokens": 6595608.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.5755220413208,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.17897515147924423,
|
|
"num_tokens": 6605983.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.654182195663452,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.3711,
|
|
"mean_token_accuracy": 0.17443785071372986,
|
|
"num_tokens": 6616682.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.737604475021362,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.4166,
|
|
"mean_token_accuracy": 0.17461859434843063,
|
|
"num_tokens": 6628715.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.655596923828125,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.3999,
|
|
"mean_token_accuracy": 0.17531391829252244,
|
|
"num_tokens": 6639219.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.554110765457153,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.2806,
|
|
"mean_token_accuracy": 0.1882808655500412,
|
|
"num_tokens": 6649516.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.685878038406372,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.17875213623046876,
|
|
"num_tokens": 6659470.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.578742456436157,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.1780405804514885,
|
|
"num_tokens": 6669637.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.572076034545899,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.3223,
|
|
"mean_token_accuracy": 0.17649637013673783,
|
|
"num_tokens": 6680450.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.675052356719971,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.17175417840480806,
|
|
"num_tokens": 6691526.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.653611278533935,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.362,
|
|
"mean_token_accuracy": 0.18099160641431808,
|
|
"num_tokens": 6702543.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.644366931915283,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.432,
|
|
"mean_token_accuracy": 0.173517544567585,
|
|
"num_tokens": 6712374.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.681358289718628,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.3281,
|
|
"mean_token_accuracy": 0.177579765021801,
|
|
"num_tokens": 6722091.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.562648773193359,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.3032,
|
|
"mean_token_accuracy": 0.1880867600440979,
|
|
"num_tokens": 6732024.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.6726624965667725,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.4343,
|
|
"mean_token_accuracy": 0.1747943416237831,
|
|
"num_tokens": 6741480.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.685817384719849,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.3025,
|
|
"mean_token_accuracy": 0.18109937906265258,
|
|
"num_tokens": 6752504.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.645481538772583,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.358,
|
|
"mean_token_accuracy": 0.17322249561548234,
|
|
"num_tokens": 6762945.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.603443384170532,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.1814771443605423,
|
|
"num_tokens": 6772862.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.667225980758667,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.3583,
|
|
"mean_token_accuracy": 0.17745987176895142,
|
|
"num_tokens": 6782912.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.701634693145752,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.3862,
|
|
"mean_token_accuracy": 0.1821895956993103,
|
|
"num_tokens": 6793249.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.635831356048584,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.3755,
|
|
"mean_token_accuracy": 0.17844953536987304,
|
|
"num_tokens": 6803721.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.752778673171997,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.3653,
|
|
"mean_token_accuracy": 0.18217482417821884,
|
|
"num_tokens": 6813637.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.610965538024902,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.3568,
|
|
"mean_token_accuracy": 0.1780506655573845,
|
|
"num_tokens": 6824171.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.5364542484283445,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.2167,
|
|
"mean_token_accuracy": 0.18581394702196122,
|
|
"num_tokens": 6833668.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.594743039872911,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.1484,
|
|
"mean_token_accuracy": 0.18705875509315067,
|
|
"num_tokens": 6842297.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.664475727081299,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.18605576306581498,
|
|
"num_tokens": 6853124.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.591793251037598,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.0094,
|
|
"mean_token_accuracy": 0.20247548967599868,
|
|
"num_tokens": 6861990.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.624763631820679,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.0144,
|
|
"mean_token_accuracy": 0.19900811612606048,
|
|
"num_tokens": 6872024.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.57113471031189,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.0806,
|
|
"mean_token_accuracy": 0.1964230164885521,
|
|
"num_tokens": 6881205.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.496262645721435,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.0401,
|
|
"mean_token_accuracy": 0.2007389023900032,
|
|
"num_tokens": 6891741.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.511004686355591,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.1915447786450386,
|
|
"num_tokens": 6902122.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.632760047912598,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.0626,
|
|
"mean_token_accuracy": 0.19105150550603867,
|
|
"num_tokens": 6912129.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.541226196289062,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 4.9527,
|
|
"mean_token_accuracy": 0.20671245455741882,
|
|
"num_tokens": 6921409.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.493974208831787,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.0771,
|
|
"mean_token_accuracy": 0.19682952910661697,
|
|
"num_tokens": 6932346.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.608472728729248,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.1942780375480652,
|
|
"num_tokens": 6942139.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.59161605834961,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.0934,
|
|
"mean_token_accuracy": 0.19804326295852662,
|
|
"num_tokens": 6952344.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.5765156745910645,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.0691,
|
|
"mean_token_accuracy": 0.19665647447109222,
|
|
"num_tokens": 6962430.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.636666584014892,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.138,
|
|
"mean_token_accuracy": 0.18742588460445403,
|
|
"num_tokens": 6973213.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.541379404067993,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.0742,
|
|
"mean_token_accuracy": 0.19189217835664749,
|
|
"num_tokens": 6983896.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.524691247940064,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.0403,
|
|
"mean_token_accuracy": 0.20257313102483748,
|
|
"num_tokens": 6993549.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.529317045211792,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.0797,
|
|
"mean_token_accuracy": 0.19670161455869675,
|
|
"num_tokens": 7003893.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.579141569137573,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.1785,
|
|
"mean_token_accuracy": 0.1873283088207245,
|
|
"num_tokens": 7014826.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.658420038223267,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.1128,
|
|
"mean_token_accuracy": 0.19410496652126313,
|
|
"num_tokens": 7024758.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.520545101165771,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.0797,
|
|
"mean_token_accuracy": 0.1996367961168289,
|
|
"num_tokens": 7034063.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.516371583938598,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.0894,
|
|
"mean_token_accuracy": 0.18975686132907868,
|
|
"num_tokens": 7045137.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.492123985290528,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.0139,
|
|
"mean_token_accuracy": 0.19682970494031907,
|
|
"num_tokens": 7056381.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.642992448806763,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.18474567979574202,
|
|
"num_tokens": 7066462.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.620988368988037,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.1779,
|
|
"mean_token_accuracy": 0.18971589356660842,
|
|
"num_tokens": 7078438.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.528593635559082,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.018,
|
|
"mean_token_accuracy": 0.1973418489098549,
|
|
"num_tokens": 7088981.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.5474175930023195,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.091,
|
|
"mean_token_accuracy": 0.19311785995960234,
|
|
"num_tokens": 7099407.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.5198742866516115,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 4.9994,
|
|
"mean_token_accuracy": 0.20123654156923293,
|
|
"num_tokens": 7109454.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.478142929077149,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.0888,
|
|
"mean_token_accuracy": 0.19594966173171996,
|
|
"num_tokens": 7119639.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.541839599609375,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.19495830088853836,
|
|
"num_tokens": 7129872.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.512017393112183,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 4.9933,
|
|
"mean_token_accuracy": 0.2008160024881363,
|
|
"num_tokens": 7140456.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.468423748016358,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.025,
|
|
"mean_token_accuracy": 0.1984676942229271,
|
|
"num_tokens": 7149787.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.509213495254516,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.0056,
|
|
"mean_token_accuracy": 0.20098166018724442,
|
|
"num_tokens": 7160482.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.542237043380737,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.0447,
|
|
"mean_token_accuracy": 0.19270921051502227,
|
|
"num_tokens": 7170818.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.532478761672974,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.0817,
|
|
"mean_token_accuracy": 0.19675188064575194,
|
|
"num_tokens": 7180770.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.453178882598877,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 4.9938,
|
|
"mean_token_accuracy": 0.20483460128307343,
|
|
"num_tokens": 7190751.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.509433937072754,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.20054661780595778,
|
|
"num_tokens": 7200093.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.63069429397583,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.1234,
|
|
"mean_token_accuracy": 0.19962280094623566,
|
|
"num_tokens": 7210526.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.497222518920898,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.19520049244165422,
|
|
"num_tokens": 7222127.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.426847505569458,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.1989658460021019,
|
|
"num_tokens": 7233418.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.531636142730713,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.19284862875938416,
|
|
"num_tokens": 7243743.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.294256185930829,
|
|
"eval_loss": 5.751738548278809,
|
|
"eval_mean_token_accuracy": 0.16606204542656278,
|
|
"eval_num_tokens": 7243743.0,
|
|
"eval_runtime": 1.9999,
|
|
"eval_samples_per_second": 1717.591,
|
|
"eval_steps_per_second": 215.011,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.5030608654022215,
|
|
"epoch": 3.1835529304861425,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004338989704678364,
|
|
"loss": 5.1264,
|
|
"mean_token_accuracy": 0.18420005291700364,
|
|
"num_tokens": 7254906.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.528319406509399,
|
|
"epoch": 3.1880963198546115,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043364854852027566,
|
|
"loss": 5.0618,
|
|
"mean_token_accuracy": 0.19399550259113313,
|
|
"num_tokens": 7264193.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.596078014373779,
|
|
"epoch": 3.1926397092230805,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004333977351231924,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.1943611443042755,
|
|
"num_tokens": 7273723.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.515632915496826,
|
|
"epoch": 3.1971830985915495,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043314653089544336,
|
|
"loss": 5.1419,
|
|
"mean_token_accuracy": 0.1859172075986862,
|
|
"num_tokens": 7284142.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.510475730895996,
|
|
"epoch": 3.201726487960018,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004328949364568504,
|
|
"loss": 5.0403,
|
|
"mean_token_accuracy": 0.20122268050909042,
|
|
"num_tokens": 7294228.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.51758337020874,
|
|
"epoch": 3.206269877328487,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004326429524281975,
|
|
"loss": 5.0656,
|
|
"mean_token_accuracy": 0.1953857108950615,
|
|
"num_tokens": 7304555.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.540026998519897,
|
|
"epoch": 3.210813266696956,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004323905794312303,
|
|
"loss": 5.1641,
|
|
"mean_token_accuracy": 0.18480420261621475,
|
|
"num_tokens": 7315955.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.542173051834107,
|
|
"epoch": 3.215356656065425,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043213781808865414,
|
|
"loss": 5.0956,
|
|
"mean_token_accuracy": 0.18931326568126677,
|
|
"num_tokens": 7326757.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.4850963115692135,
|
|
"epoch": 3.2199000454338935,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004318846690241324,
|
|
"loss": 5.0687,
|
|
"mean_token_accuracy": 0.19748374819755554,
|
|
"num_tokens": 7335969.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.50565733909607,
|
|
"epoch": 3.2244434348023625,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043163113286228533,
|
|
"loss": 5.0681,
|
|
"mean_token_accuracy": 0.19274979531764985,
|
|
"num_tokens": 7345824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.556841421127319,
|
|
"epoch": 3.2289868241708315,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004313772102286882,
|
|
"loss": 5.1442,
|
|
"mean_token_accuracy": 0.18979265540838242,
|
|
"num_tokens": 7356292.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.53232216835022,
|
|
"epoch": 3.2335302135393005,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004311229017498698,
|
|
"loss": 5.1441,
|
|
"mean_token_accuracy": 0.18825449645519257,
|
|
"num_tokens": 7366809.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.594621801376343,
|
|
"epoch": 3.238073602907769,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000430868208053311,
|
|
"loss": 5.053,
|
|
"mean_token_accuracy": 0.20324303656816484,
|
|
"num_tokens": 7376076.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.48758020401001,
|
|
"epoch": 3.242616992276238,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043061312976744326,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.19461447298526763,
|
|
"num_tokens": 7386338.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.502114915847779,
|
|
"epoch": 3.247160381644707,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004303576675216469,
|
|
"loss": 5.0528,
|
|
"mean_token_accuracy": 0.19673527777194977,
|
|
"num_tokens": 7395792.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.491319179534912,
|
|
"epoch": 3.251703771013176,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043010182194624957,
|
|
"loss": 5.0757,
|
|
"mean_token_accuracy": 0.19470140486955642,
|
|
"num_tokens": 7405765.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.488127040863037,
|
|
"epoch": 3.2562471603816445,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004298455936725248,
|
|
"loss": 5.1261,
|
|
"mean_token_accuracy": 0.19317075312137605,
|
|
"num_tokens": 7415213.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.584626817703247,
|
|
"epoch": 3.2607905497501135,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004295889833326903,
|
|
"loss": 5.0902,
|
|
"mean_token_accuracy": 0.1905794084072113,
|
|
"num_tokens": 7425722.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.599624967575073,
|
|
"epoch": 3.2653339391185825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042933199155990675,
|
|
"loss": 5.1524,
|
|
"mean_token_accuracy": 0.19233058393001556,
|
|
"num_tokens": 7436240.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.509469032287598,
|
|
"epoch": 3.2698773284870515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004290746189882757,
|
|
"loss": 5.1361,
|
|
"mean_token_accuracy": 0.18685731738805772,
|
|
"num_tokens": 7447878.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.56361141204834,
|
|
"epoch": 3.27442071785552,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042881686625283844,
|
|
"loss": 5.0981,
|
|
"mean_token_accuracy": 0.19554533213376998,
|
|
"num_tokens": 7458635.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.541248416900634,
|
|
"epoch": 3.278964107223989,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004285587339895742,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.18703659921884536,
|
|
"num_tokens": 7468740.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.629860496520996,
|
|
"epoch": 3.283507496592458,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042830022283539863,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.18779015243053437,
|
|
"num_tokens": 7480977.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.533052587509156,
|
|
"epoch": 3.288050885960927,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004280413334281623,
|
|
"loss": 5.1774,
|
|
"mean_token_accuracy": 0.19077278971672057,
|
|
"num_tokens": 7491758.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.600966930389404,
|
|
"epoch": 3.292594275329396,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042778206640664915,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.19315629601478576,
|
|
"num_tokens": 7500486.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.44861068725586,
|
|
"epoch": 3.2971376646978645,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004275224224105748,
|
|
"loss": 5.1165,
|
|
"mean_token_accuracy": 0.19958876222372054,
|
|
"num_tokens": 7510371.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.533408546447754,
|
|
"epoch": 3.3016810540663335,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042726240208058484,
|
|
"loss": 5.144,
|
|
"mean_token_accuracy": 0.19472647756338118,
|
|
"num_tokens": 7520697.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.492074060440063,
|
|
"epoch": 3.3062244434348025,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004270020060582538,
|
|
"loss": 5.0553,
|
|
"mean_token_accuracy": 0.19772299081087114,
|
|
"num_tokens": 7530757.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.51636757850647,
|
|
"epoch": 3.310767832803271,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042674123498608285,
|
|
"loss": 5.1056,
|
|
"mean_token_accuracy": 0.19229757487773896,
|
|
"num_tokens": 7541810.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.626980352401733,
|
|
"epoch": 3.31531122217174,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042648008950749877,
|
|
"loss": 5.2965,
|
|
"mean_token_accuracy": 0.17989146560430527,
|
|
"num_tokens": 7552754.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.574767255783081,
|
|
"epoch": 3.319854611540209,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000426218570266852,
|
|
"loss": 5.0767,
|
|
"mean_token_accuracy": 0.19806645810604095,
|
|
"num_tokens": 7562906.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.434383964538574,
|
|
"epoch": 3.324398000908678,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042595667790941536,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.19282221794128418,
|
|
"num_tokens": 7574149.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.442363357543945,
|
|
"epoch": 3.328941390277147,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042569441308138204,
|
|
"loss": 5.0817,
|
|
"mean_token_accuracy": 0.20203310549259185,
|
|
"num_tokens": 7584610.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.542620754241943,
|
|
"epoch": 3.3334847796456155,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004254317764298645,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.1923823893070221,
|
|
"num_tokens": 7594954.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.598101663589477,
|
|
"epoch": 3.3380281690140845,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004251687686028927,
|
|
"loss": 5.1653,
|
|
"mean_token_accuracy": 0.17998976409435272,
|
|
"num_tokens": 7606069.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.43584451675415,
|
|
"epoch": 3.3425715583825535,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042490539024941216,
|
|
"loss": 5.0002,
|
|
"mean_token_accuracy": 0.20193633139133454,
|
|
"num_tokens": 7615671.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.508904314041137,
|
|
"epoch": 3.347114947751022,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042464164201928286,
|
|
"loss": 5.1961,
|
|
"mean_token_accuracy": 0.18532725274562836,
|
|
"num_tokens": 7626740.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.539602279663086,
|
|
"epoch": 3.351658337119491,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042437752456327734,
|
|
"loss": 5.1183,
|
|
"mean_token_accuracy": 0.1940290555357933,
|
|
"num_tokens": 7636459.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.612280559539795,
|
|
"epoch": 3.35620172648796,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042411303853307907,
|
|
"loss": 5.0713,
|
|
"mean_token_accuracy": 0.19265790432691574,
|
|
"num_tokens": 7646974.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.406089162826538,
|
|
"epoch": 3.360745115856429,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004238481845812811,
|
|
"loss": 5.0436,
|
|
"mean_token_accuracy": 0.1934661850333214,
|
|
"num_tokens": 7657372.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.427165126800537,
|
|
"epoch": 3.365288505224898,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042358296336138436,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.19385142922401427,
|
|
"num_tokens": 7668972.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.572782802581787,
|
|
"epoch": 3.3698318945933665,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004233173755277956,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.2041905164718628,
|
|
"num_tokens": 7679742.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.518075370788575,
|
|
"epoch": 3.3743752839618355,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042305142173582657,
|
|
"loss": 5.13,
|
|
"mean_token_accuracy": 0.19715550690889358,
|
|
"num_tokens": 7690759.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.507992267608643,
|
|
"epoch": 3.3789186733303045,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042278510264169165,
|
|
"loss": 5.0818,
|
|
"mean_token_accuracy": 0.19355220198631287,
|
|
"num_tokens": 7700482.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.482635736465454,
|
|
"epoch": 3.3834620626987735,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042251841890250696,
|
|
"loss": 5.0661,
|
|
"mean_token_accuracy": 0.1878388524055481,
|
|
"num_tokens": 7710414.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.54592809677124,
|
|
"epoch": 3.388005452067242,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000422251371176288,
|
|
"loss": 5.1005,
|
|
"mean_token_accuracy": 0.19825109988451003,
|
|
"num_tokens": 7721214.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.497478485107422,
|
|
"epoch": 3.392548841435711,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042198396012194854,
|
|
"loss": 5.0537,
|
|
"mean_token_accuracy": 0.20017760396003723,
|
|
"num_tokens": 7732551.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.509873962402343,
|
|
"epoch": 3.39709223080418,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042171618639929865,
|
|
"loss": 5.1599,
|
|
"mean_token_accuracy": 0.19333608895540239,
|
|
"num_tokens": 7742266.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.488704776763916,
|
|
"epoch": 3.401635620172649,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042144805066904345,
|
|
"loss": 5.009,
|
|
"mean_token_accuracy": 0.1954241394996643,
|
|
"num_tokens": 7753152.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.4690587520599365,
|
|
"epoch": 3.4061790095411175,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042117955359278125,
|
|
"loss": 5.1049,
|
|
"mean_token_accuracy": 0.19189360439777375,
|
|
"num_tokens": 7764420.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.535941791534424,
|
|
"epoch": 3.4107223989095865,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042091069583300177,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.1919311374425888,
|
|
"num_tokens": 7776044.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.4821361064910885,
|
|
"epoch": 3.4152657882780555,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004206414780530849,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.19450199455022812,
|
|
"num_tokens": 7786737.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.452231979370117,
|
|
"epoch": 3.4198091776465245,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004203719009172987,
|
|
"loss": 5.0369,
|
|
"mean_token_accuracy": 0.19422636479139327,
|
|
"num_tokens": 7796974.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.535924434661865,
|
|
"epoch": 3.424352567014993,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000420101965090798,
|
|
"loss": 5.1955,
|
|
"mean_token_accuracy": 0.18865609019994736,
|
|
"num_tokens": 7808221.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.435919141769409,
|
|
"epoch": 3.428895956383462,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004198316712396226,
|
|
"loss": 5.0423,
|
|
"mean_token_accuracy": 0.19782525449991226,
|
|
"num_tokens": 7818422.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.497764015197754,
|
|
"epoch": 3.433439345751931,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041956102003069573,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.19576893448829652,
|
|
"num_tokens": 7828313.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.454093313217163,
|
|
"epoch": 3.4379827351204,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004192900121318223,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.20414010733366011,
|
|
"num_tokens": 7839493.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.5003832340240475,
|
|
"epoch": 3.4425261244888685,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004190186482116874,
|
|
"loss": 5.0992,
|
|
"mean_token_accuracy": 0.1865068718791008,
|
|
"num_tokens": 7849796.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.463807201385498,
|
|
"epoch": 3.4470695138573375,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004187469289398545,
|
|
"loss": 5.1444,
|
|
"mean_token_accuracy": 0.19304777979850768,
|
|
"num_tokens": 7860865.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.473004722595215,
|
|
"epoch": 3.4516129032258065,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041847485498676385,
|
|
"loss": 5.0822,
|
|
"mean_token_accuracy": 0.20007635802030563,
|
|
"num_tokens": 7871144.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.469412326812744,
|
|
"epoch": 3.4561562925942755,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041820242702373094,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.19716734886169435,
|
|
"num_tokens": 7880916.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.4438964366912845,
|
|
"epoch": 3.460699681962744,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004179296457229448,
|
|
"loss": 5.0592,
|
|
"mean_token_accuracy": 0.19640869945287703,
|
|
"num_tokens": 7891813.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.542184829711914,
|
|
"epoch": 3.465243071331213,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004176565117574659,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.19370521306991578,
|
|
"num_tokens": 7902697.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.495546579360962,
|
|
"epoch": 3.469786460699682,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004173830258012255,
|
|
"loss": 5.0892,
|
|
"mean_token_accuracy": 0.19609837383031845,
|
|
"num_tokens": 7912777.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.4906168460845945,
|
|
"epoch": 3.474329850068151,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004171091885290226,
|
|
"loss": 5.1313,
|
|
"mean_token_accuracy": 0.19004505276679992,
|
|
"num_tokens": 7923255.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.467664766311645,
|
|
"epoch": 3.4788732394366195,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004168350006165238,
|
|
"loss": 5.0017,
|
|
"mean_token_accuracy": 0.1963506817817688,
|
|
"num_tokens": 7934268.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.386504459381103,
|
|
"epoch": 3.4834166288050885,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041656046274026034,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.19830360263586044,
|
|
"num_tokens": 7944526.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.490135526657104,
|
|
"epoch": 3.4879600181735575,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004162855755776272,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.19211395829916,
|
|
"num_tokens": 7954512.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.497490501403808,
|
|
"epoch": 3.4925034075420265,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041601033980688106,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.1822715610265732,
|
|
"num_tokens": 7965703.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.550517416000366,
|
|
"epoch": 3.497046796910495,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004157347561071389,
|
|
"loss": 5.1132,
|
|
"mean_token_accuracy": 0.19571154564619064,
|
|
"num_tokens": 7977822.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.465801143646241,
|
|
"epoch": 3.501590186278964,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041545882515837604,
|
|
"loss": 5.0358,
|
|
"mean_token_accuracy": 0.19362603574991227,
|
|
"num_tokens": 7988318.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.491103267669677,
|
|
"epoch": 3.506133575647433,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041518254764142477,
|
|
"loss": 5.0062,
|
|
"mean_token_accuracy": 0.1980455219745636,
|
|
"num_tokens": 7998076.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.463761329650879,
|
|
"epoch": 3.510676965015902,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041490592423797226,
|
|
"loss": 5.1176,
|
|
"mean_token_accuracy": 0.1959502801299095,
|
|
"num_tokens": 8007913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.431157064437866,
|
|
"epoch": 3.515220354384371,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004146289556305594,
|
|
"loss": 5.0744,
|
|
"mean_token_accuracy": 0.1907247170805931,
|
|
"num_tokens": 8018340.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.571393203735352,
|
|
"epoch": 3.5197637437528395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004143516425025786,
|
|
"loss": 5.1197,
|
|
"mean_token_accuracy": 0.1900700196623802,
|
|
"num_tokens": 8028312.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.578557729721069,
|
|
"epoch": 3.5243071331213085,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004140739855382724,
|
|
"loss": 5.1481,
|
|
"mean_token_accuracy": 0.19704046100378036,
|
|
"num_tokens": 8038580.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.52321343421936,
|
|
"epoch": 3.5288505224897775,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041379598542273197,
|
|
"loss": 5.0953,
|
|
"mean_token_accuracy": 0.19009448140859603,
|
|
"num_tokens": 8048365.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.478733491897583,
|
|
"epoch": 3.533393911858246,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041351764284189464,
|
|
"loss": 5.152,
|
|
"mean_token_accuracy": 0.18652414083480834,
|
|
"num_tokens": 8059638.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.55508828163147,
|
|
"epoch": 3.537937301226715,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004132389584825433,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.18808463215827942,
|
|
"num_tokens": 8070993.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.467888402938843,
|
|
"epoch": 3.542480690595184,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041295993303230383,
|
|
"loss": 5.0415,
|
|
"mean_token_accuracy": 0.2027706667780876,
|
|
"num_tokens": 8081203.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.417987108230591,
|
|
"epoch": 3.547024079963653,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004126805671796438,
|
|
"loss": 5.0529,
|
|
"mean_token_accuracy": 0.1934710845351219,
|
|
"num_tokens": 8091564.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.467380571365356,
|
|
"epoch": 3.551567469332122,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004124008616138707,
|
|
"loss": 5.0387,
|
|
"mean_token_accuracy": 0.20112892985343933,
|
|
"num_tokens": 8100675.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.476384687423706,
|
|
"epoch": 3.5561108587005905,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00041212081702513015,
|
|
"loss": 5.0971,
|
|
"mean_token_accuracy": 0.20031499713659287,
|
|
"num_tokens": 8110605.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.512406873703003,
|
|
"epoch": 3.5606542480690595,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004118404341044044,
|
|
"loss": 5.1441,
|
|
"mean_token_accuracy": 0.19020797908306122,
|
|
"num_tokens": 8120910.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.500950288772583,
|
|
"epoch": 3.5651976374375285,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041155971354351034,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.19211931824684142,
|
|
"num_tokens": 8131827.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.483038759231567,
|
|
"epoch": 3.569741026805997,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041127865603509815,
|
|
"loss": 5.0742,
|
|
"mean_token_accuracy": 0.1954970568418503,
|
|
"num_tokens": 8141785.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.507791328430176,
|
|
"epoch": 3.574284416174466,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004109972622726493,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.1933991476893425,
|
|
"num_tokens": 8151732.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.525823974609375,
|
|
"epoch": 3.578827805542935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041071553295047485,
|
|
"loss": 5.1917,
|
|
"mean_token_accuracy": 0.1891781210899353,
|
|
"num_tokens": 8162334.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.454549932479859,
|
|
"epoch": 3.583371194911404,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004104334687637139,
|
|
"loss": 5.0662,
|
|
"mean_token_accuracy": 0.1937805950641632,
|
|
"num_tokens": 8172810.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.443389749526977,
|
|
"epoch": 3.587914584279873,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041015107040833187,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.19253693670034408,
|
|
"num_tokens": 8183240.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.505223894119263,
|
|
"epoch": 3.5924579736483415,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004098683385811185,
|
|
"loss": 5.034,
|
|
"mean_token_accuracy": 0.19396354407072067,
|
|
"num_tokens": 8193824.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.459287023544311,
|
|
"epoch": 3.5970013630168105,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040958527397968666,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.20360795110464097,
|
|
"num_tokens": 8203792.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.523562383651734,
|
|
"epoch": 3.6015447523852795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004093018773024699,
|
|
"loss": 5.2223,
|
|
"mean_token_accuracy": 0.19078201055526733,
|
|
"num_tokens": 8214436.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.517093276977539,
|
|
"epoch": 3.606088141753748,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004090181492487215,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.19075654000043868,
|
|
"num_tokens": 8224518.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.515325736999512,
|
|
"epoch": 3.610631531122217,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040873409051851223,
|
|
"loss": 5.2131,
|
|
"mean_token_accuracy": 0.18362343162298203,
|
|
"num_tokens": 8235234.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.451943302154541,
|
|
"epoch": 3.615174920490686,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004084497018127286,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.19037348181009292,
|
|
"num_tokens": 8246612.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.555245971679687,
|
|
"epoch": 3.619718309859155,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004081649838330717,
|
|
"loss": 5.0575,
|
|
"mean_token_accuracy": 0.20067218393087388,
|
|
"num_tokens": 8256720.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.484178972244263,
|
|
"epoch": 3.624261699227624,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040787993728205464,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.19589620977640151,
|
|
"num_tokens": 8267666.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.520815324783325,
|
|
"epoch": 3.6288050885960925,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040759456286300154,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.19454220235347747,
|
|
"num_tokens": 8277882.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.499479579925537,
|
|
"epoch": 3.6333484779645615,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040730886128004544,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.19699462950229646,
|
|
"num_tokens": 8286968.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.6333484779645615,
|
|
"eval_entropy": 5.315180792919425,
|
|
"eval_loss": 5.70609188079834,
|
|
"eval_mean_token_accuracy": 0.168532062148632,
|
|
"eval_num_tokens": 8286968.0,
|
|
"eval_runtime": 2.0048,
|
|
"eval_samples_per_second": 1713.389,
|
|
"eval_steps_per_second": 214.485,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": 1789765102725120.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|