043fd421c2
Model: fpadovani/nld-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
22277 lines
609 KiB
JSON
22277 lines
609 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.99136756019991,
|
|
"eval_steps": 500,
|
|
"global_step": 11000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.45850100517273,
|
|
"epoch": 3.6378918673330305,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004070228332381266,
|
|
"loss": 5.1016,
|
|
"mean_token_accuracy": 0.19867050349712373,
|
|
"num_tokens": 8297154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.477522230148315,
|
|
"epoch": 3.6424352567014995,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004067364794429908,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.19093022644519805,
|
|
"num_tokens": 8307954.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.41411714553833,
|
|
"epoch": 3.646978646069968,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004064498006011874,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.19603158235549928,
|
|
"num_tokens": 8317714.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.454757261276245,
|
|
"epoch": 3.651522035438437,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040616279742006813,
|
|
"loss": 5.101,
|
|
"mean_token_accuracy": 0.19974977821111678,
|
|
"num_tokens": 8327963.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.468853998184204,
|
|
"epoch": 3.656065424806906,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040587547060778466,
|
|
"loss": 5.0243,
|
|
"mean_token_accuracy": 0.20314159542322158,
|
|
"num_tokens": 8338190.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.4881080150604244,
|
|
"epoch": 3.660608814175375,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004055878208732874,
|
|
"loss": 5.1632,
|
|
"mean_token_accuracy": 0.18801491856575012,
|
|
"num_tokens": 8348564.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.459885692596435,
|
|
"epoch": 3.665152203543844,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040529984892632363,
|
|
"loss": 5.0662,
|
|
"mean_token_accuracy": 0.19465430080890656,
|
|
"num_tokens": 8359463.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.501561164855957,
|
|
"epoch": 3.6696955929123125,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004050115554774353,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.18758884817361832,
|
|
"num_tokens": 8369676.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.469761371612549,
|
|
"epoch": 3.6742389822807815,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004047229412379578,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.20533668994903564,
|
|
"num_tokens": 8379579.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.4417637348175045,
|
|
"epoch": 3.6787823716492505,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004044340069200183,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.19642857760190963,
|
|
"num_tokens": 8390164.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.454135513305664,
|
|
"epoch": 3.683325761017719,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040414475323653326,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.19758211374282836,
|
|
"num_tokens": 8399220.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.481324338912964,
|
|
"epoch": 3.687869150386188,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004038551809012076,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.1947534829378128,
|
|
"num_tokens": 8409488.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.496913242340088,
|
|
"epoch": 3.692412539754657,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004035652906285322,
|
|
"loss": 5.0334,
|
|
"mean_token_accuracy": 0.20396993756294252,
|
|
"num_tokens": 8418817.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.473042583465576,
|
|
"epoch": 3.696955929123126,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040327508313378244,
|
|
"loss": 5.1516,
|
|
"mean_token_accuracy": 0.1938195303082466,
|
|
"num_tokens": 8428742.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.447334337234497,
|
|
"epoch": 3.701499318491595,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004029845591330164,
|
|
"loss": 5.0153,
|
|
"mean_token_accuracy": 0.20452942103147506,
|
|
"num_tokens": 8438016.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.495593547821045,
|
|
"epoch": 3.7060427078600635,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040269371934307336,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.19237001985311508,
|
|
"num_tokens": 8448428.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.398948526382446,
|
|
"epoch": 3.7105860972285325,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040240256448157146,
|
|
"loss": 4.9805,
|
|
"mean_token_accuracy": 0.20304651111364364,
|
|
"num_tokens": 8458444.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.462928819656372,
|
|
"epoch": 3.7151294865970015,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004021110952669065,
|
|
"loss": 5.125,
|
|
"mean_token_accuracy": 0.19258323758840562,
|
|
"num_tokens": 8467721.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.414800214767456,
|
|
"epoch": 3.71967287596547,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004018193124182497,
|
|
"loss": 5.034,
|
|
"mean_token_accuracy": 0.20143663436174392,
|
|
"num_tokens": 8477484.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.4366367816925045,
|
|
"epoch": 3.724216265333939,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004015272166555462,
|
|
"loss": 5.0479,
|
|
"mean_token_accuracy": 0.19863946437835694,
|
|
"num_tokens": 8486902.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.476304149627685,
|
|
"epoch": 3.728759654702408,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004012348086995132,
|
|
"loss": 5.0161,
|
|
"mean_token_accuracy": 0.19606781154870986,
|
|
"num_tokens": 8497626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.514776182174683,
|
|
"epoch": 3.733303044070877,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004009420892716385,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.1879243478178978,
|
|
"num_tokens": 8507518.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.447705411911011,
|
|
"epoch": 3.737846433439346,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004006490590941781,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.19643318504095078,
|
|
"num_tokens": 8517835.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.504614305496216,
|
|
"epoch": 3.7423898228078145,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004003557188901547,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.19166215211153032,
|
|
"num_tokens": 8527552.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.497322273254395,
|
|
"epoch": 3.7469332121762835,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004000620693833562,
|
|
"loss": 5.1139,
|
|
"mean_token_accuracy": 0.19493110626935958,
|
|
"num_tokens": 8538082.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.442343473434448,
|
|
"epoch": 3.7514766015447525,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003997681112983335,
|
|
"loss": 5.0498,
|
|
"mean_token_accuracy": 0.19139655232429503,
|
|
"num_tokens": 8548558.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.468421125411988,
|
|
"epoch": 3.756019990913221,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039947384536039905,
|
|
"loss": 5.0658,
|
|
"mean_token_accuracy": 0.1993381440639496,
|
|
"num_tokens": 8559406.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.50951361656189,
|
|
"epoch": 3.76056338028169,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039917927229562474,
|
|
"loss": 5.0889,
|
|
"mean_token_accuracy": 0.18839718848466874,
|
|
"num_tokens": 8569519.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.42941198348999,
|
|
"epoch": 3.765106769650159,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039888439283084035,
|
|
"loss": 5.1218,
|
|
"mean_token_accuracy": 0.19479948431253433,
|
|
"num_tokens": 8579658.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.436860990524292,
|
|
"epoch": 3.769650159018628,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039858920769363164,
|
|
"loss": 5.0235,
|
|
"mean_token_accuracy": 0.2037848249077797,
|
|
"num_tokens": 8589681.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.517423391342163,
|
|
"epoch": 3.774193548387097,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003982937176123386,
|
|
"loss": 5.1026,
|
|
"mean_token_accuracy": 0.19400613158941268,
|
|
"num_tokens": 8600310.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.452726125717163,
|
|
"epoch": 3.7787369377555655,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039799792331605364,
|
|
"loss": 5.1123,
|
|
"mean_token_accuracy": 0.19437065571546555,
|
|
"num_tokens": 8610920.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.3803472995758055,
|
|
"epoch": 3.7832803271240345,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039770182553461965,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.19948560893535613,
|
|
"num_tokens": 8620536.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.469891834259033,
|
|
"epoch": 3.7878237164925035,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039740542499862866,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.19431438595056533,
|
|
"num_tokens": 8631647.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.491085386276245,
|
|
"epoch": 3.792367105860972,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003971087224394194,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.19009023308753967,
|
|
"num_tokens": 8642659.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.483698225021362,
|
|
"epoch": 3.796910495229441,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003968117185890761,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.2021097645163536,
|
|
"num_tokens": 8651775.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.445494556427002,
|
|
"epoch": 3.80145388459791,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039651441418042606,
|
|
"loss": 5.0628,
|
|
"mean_token_accuracy": 0.19215909242630005,
|
|
"num_tokens": 8662520.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.480993938446045,
|
|
"epoch": 3.805997273966379,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003962168099470384,
|
|
"loss": 5.0943,
|
|
"mean_token_accuracy": 0.1982967585325241,
|
|
"num_tokens": 8672630.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.4850335121154785,
|
|
"epoch": 3.810540663334848,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000395918906623222,
|
|
"loss": 5.1676,
|
|
"mean_token_accuracy": 0.18463584035634995,
|
|
"num_tokens": 8683348.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.442568588256836,
|
|
"epoch": 3.8150840527033165,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003956207049440238,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.19739602953195573,
|
|
"num_tokens": 8693423.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.466849613189697,
|
|
"epoch": 3.8196274420717855,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003953222056452265,
|
|
"loss": 5.0924,
|
|
"mean_token_accuracy": 0.19940629750490188,
|
|
"num_tokens": 8703766.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.571230745315551,
|
|
"epoch": 3.8241708314402545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003950234094633479,
|
|
"loss": 5.1998,
|
|
"mean_token_accuracy": 0.18991368412971496,
|
|
"num_tokens": 8715330.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.47427978515625,
|
|
"epoch": 3.828714220808723,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039472431713563746,
|
|
"loss": 5.0724,
|
|
"mean_token_accuracy": 0.1970528930425644,
|
|
"num_tokens": 8725732.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.535758543014526,
|
|
"epoch": 3.833257610177192,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000394424929400076,
|
|
"loss": 5.0707,
|
|
"mean_token_accuracy": 0.19908757656812667,
|
|
"num_tokens": 8736377.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.539145231246948,
|
|
"epoch": 3.837800999545661,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003941252469953729,
|
|
"loss": 5.084,
|
|
"mean_token_accuracy": 0.19426683634519576,
|
|
"num_tokens": 8746991.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.421463918685913,
|
|
"epoch": 3.84234438891413,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003938252706609647,
|
|
"loss": 5.0763,
|
|
"mean_token_accuracy": 0.1978220373392105,
|
|
"num_tokens": 8756920.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.484556102752686,
|
|
"epoch": 3.846887778282599,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003935250011370132,
|
|
"loss": 5.1512,
|
|
"mean_token_accuracy": 0.18240558803081514,
|
|
"num_tokens": 8768617.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.542027139663697,
|
|
"epoch": 3.851431167651068,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003932244391644037,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.2032419726252556,
|
|
"num_tokens": 8778427.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.447765922546386,
|
|
"epoch": 3.8559745570195365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003929235854847428,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.19676993191242217,
|
|
"num_tokens": 8789093.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.459629058837891,
|
|
"epoch": 3.8605179463880055,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039262244084035724,
|
|
"loss": 5.0678,
|
|
"mean_token_accuracy": 0.1931177034974098,
|
|
"num_tokens": 8799832.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.460925197601318,
|
|
"epoch": 3.8650613357564745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039232100597429143,
|
|
"loss": 5.101,
|
|
"mean_token_accuracy": 0.1948159322142601,
|
|
"num_tokens": 8810149.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.508511114120483,
|
|
"epoch": 3.869604725124943,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039201928163030584,
|
|
"loss": 5.1054,
|
|
"mean_token_accuracy": 0.20215496718883513,
|
|
"num_tokens": 8820087.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.457157230377197,
|
|
"epoch": 3.874148114493412,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039171726855287544,
|
|
"loss": 5.1556,
|
|
"mean_token_accuracy": 0.18760941475629805,
|
|
"num_tokens": 8830959.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.499671936035156,
|
|
"epoch": 3.878691503861881,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039141496748718735,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.1926316186785698,
|
|
"num_tokens": 8842522.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.417900323867798,
|
|
"epoch": 3.88323489323035,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003911123791791395,
|
|
"loss": 4.9993,
|
|
"mean_token_accuracy": 0.20305683463811874,
|
|
"num_tokens": 8852521.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.548113441467285,
|
|
"epoch": 3.887778282598819,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003908095043753384,
|
|
"loss": 5.1517,
|
|
"mean_token_accuracy": 0.18835781067609786,
|
|
"num_tokens": 8862156.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.511874866485596,
|
|
"epoch": 3.8923216719672875,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039050634382309753,
|
|
"loss": 5.0117,
|
|
"mean_token_accuracy": 0.19835689067840576,
|
|
"num_tokens": 8872772.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.464529705047608,
|
|
"epoch": 3.8968650613357565,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003902028982704355,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.19539771527051925,
|
|
"num_tokens": 8883229.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.386270999908447,
|
|
"epoch": 3.9014084507042255,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003898991684660738,
|
|
"loss": 5.0743,
|
|
"mean_token_accuracy": 0.1984972044825554,
|
|
"num_tokens": 8892403.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.458788776397705,
|
|
"epoch": 3.905951840072694,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003895951551594358,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.19079862236976625,
|
|
"num_tokens": 8902617.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.55191650390625,
|
|
"epoch": 3.910495229441163,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000389290859100644,
|
|
"loss": 5.1355,
|
|
"mean_token_accuracy": 0.19410814195871354,
|
|
"num_tokens": 8913099.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.403039884567261,
|
|
"epoch": 3.915038618809632,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003889862810405187,
|
|
"loss": 5.0296,
|
|
"mean_token_accuracy": 0.19456752240657807,
|
|
"num_tokens": 8922469.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.460741949081421,
|
|
"epoch": 3.919582008178101,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000388681421730576,
|
|
"loss": 5.1188,
|
|
"mean_token_accuracy": 0.20009989738464357,
|
|
"num_tokens": 8933261.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.52027382850647,
|
|
"epoch": 3.92412539754657,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003883762819230259,
|
|
"loss": 5.1112,
|
|
"mean_token_accuracy": 0.20046692490577697,
|
|
"num_tokens": 8943873.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.458073282241822,
|
|
"epoch": 3.9286687869150385,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038807086237077057,
|
|
"loss": 5.0107,
|
|
"mean_token_accuracy": 0.21071356683969497,
|
|
"num_tokens": 8953802.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.52363772392273,
|
|
"epoch": 3.9332121762835075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038776516382740237,
|
|
"loss": 5.2019,
|
|
"mean_token_accuracy": 0.18938166052103042,
|
|
"num_tokens": 8964309.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.483035707473755,
|
|
"epoch": 3.9377555656519765,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00038745918704720205,
|
|
"loss": 5.0675,
|
|
"mean_token_accuracy": 0.19790703505277635,
|
|
"num_tokens": 8973611.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.431017875671387,
|
|
"epoch": 3.942298955020445,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038715293278513693,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.19031185507774354,
|
|
"num_tokens": 8985783.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.4657555103302,
|
|
"epoch": 3.946842344388914,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038684640179685896,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.18728139549493789,
|
|
"num_tokens": 8996736.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.536164379119873,
|
|
"epoch": 3.951385733757383,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003865395948387029,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.18692236840724946,
|
|
"num_tokens": 9007635.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.519601440429687,
|
|
"epoch": 3.955929123125852,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038623251266768437,
|
|
"loss": 5.0658,
|
|
"mean_token_accuracy": 0.19407179057598115,
|
|
"num_tokens": 9017892.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.429573059082031,
|
|
"epoch": 3.960472512494321,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038592515604149815,
|
|
"loss": 5.0984,
|
|
"mean_token_accuracy": 0.1974261224269867,
|
|
"num_tokens": 9027594.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.461850786209107,
|
|
"epoch": 3.9650159018627895,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003856175257185162,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.19105612188577653,
|
|
"num_tokens": 9037083.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.505531167984008,
|
|
"epoch": 3.9695592912312585,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038530962245778564,
|
|
"loss": 5.0811,
|
|
"mean_token_accuracy": 0.20364490896463394,
|
|
"num_tokens": 9047176.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.497819805145264,
|
|
"epoch": 3.9741026805997275,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003850014470190272,
|
|
"loss": 5.0464,
|
|
"mean_token_accuracy": 0.2016264632344246,
|
|
"num_tokens": 9057810.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.5166015625,
|
|
"epoch": 3.978646069968196,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003846930001626332,
|
|
"loss": 5.138,
|
|
"mean_token_accuracy": 0.19457311928272247,
|
|
"num_tokens": 9067765.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.522107172012329,
|
|
"epoch": 3.983189459336665,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003843842826496656,
|
|
"loss": 5.0992,
|
|
"mean_token_accuracy": 0.19395556002855302,
|
|
"num_tokens": 9078796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.493750095367432,
|
|
"epoch": 3.987732848705134,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000384075295241854,
|
|
"loss": 5.0456,
|
|
"mean_token_accuracy": 0.20145037919282913,
|
|
"num_tokens": 9088368.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.405688905715943,
|
|
"epoch": 3.992276238073603,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003837660387015944,
|
|
"loss": 4.9982,
|
|
"mean_token_accuracy": 0.20076853781938553,
|
|
"num_tokens": 9099082.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.469166851043701,
|
|
"epoch": 3.996819627442072,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003834565137919463,
|
|
"loss": 5.1443,
|
|
"mean_token_accuracy": 0.19387515038251876,
|
|
"num_tokens": 9110543.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.493903636932373,
|
|
"epoch": 4.000908677873694,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003831467212766318,
|
|
"loss": 5.0678,
|
|
"mean_token_accuracy": 0.19428812464078268,
|
|
"num_tokens": 9119347.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.425588989257813,
|
|
"epoch": 4.005452067242163,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003828366619200331,
|
|
"loss": 4.7491,
|
|
"mean_token_accuracy": 0.2172959938645363,
|
|
"num_tokens": 9129154.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.3712609767913815,
|
|
"epoch": 4.009995456610632,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003825263364871908,
|
|
"loss": 4.7978,
|
|
"mean_token_accuracy": 0.21357453167438506,
|
|
"num_tokens": 9139877.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.454461240768433,
|
|
"epoch": 4.0145388459791,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003822157457438022,
|
|
"loss": 4.8602,
|
|
"mean_token_accuracy": 0.21274617910385132,
|
|
"num_tokens": 9150508.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.433470439910889,
|
|
"epoch": 4.019082235347569,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038190489045621893,
|
|
"loss": 4.8037,
|
|
"mean_token_accuracy": 0.21804286688566207,
|
|
"num_tokens": 9160403.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.495927715301514,
|
|
"epoch": 4.023625624716038,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003815937713914455,
|
|
"loss": 4.8557,
|
|
"mean_token_accuracy": 0.21547941714525223,
|
|
"num_tokens": 9171026.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.4306642532348635,
|
|
"epoch": 4.028169014084507,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038128238931713743,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.21478251516819,
|
|
"num_tokens": 9181252.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.513566017150879,
|
|
"epoch": 4.032712403452976,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038097074500159883,
|
|
"loss": 4.8798,
|
|
"mean_token_accuracy": 0.20811144709587098,
|
|
"num_tokens": 9190803.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.4730432510375975,
|
|
"epoch": 4.037255792821445,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038065883921378123,
|
|
"loss": 4.875,
|
|
"mean_token_accuracy": 0.21120634973049163,
|
|
"num_tokens": 9201077.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.4387514114379885,
|
|
"epoch": 4.041799182189914,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003803466727232811,
|
|
"loss": 4.8921,
|
|
"mean_token_accuracy": 0.21355759650468825,
|
|
"num_tokens": 9211239.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.442941761016845,
|
|
"epoch": 4.046342571558383,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003800342463003383,
|
|
"loss": 4.8641,
|
|
"mean_token_accuracy": 0.21715304553508757,
|
|
"num_tokens": 9222683.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.445368814468384,
|
|
"epoch": 4.050885960926851,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00037972156071583386,
|
|
"loss": 4.817,
|
|
"mean_token_accuracy": 0.21639355570077895,
|
|
"num_tokens": 9232099.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.433478546142578,
|
|
"epoch": 4.05542935029532,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003794086167412884,
|
|
"loss": 4.8795,
|
|
"mean_token_accuracy": 0.21039076447486876,
|
|
"num_tokens": 9242660.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.401778745651245,
|
|
"epoch": 4.059972739663789,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003790954151488601,
|
|
"loss": 4.8987,
|
|
"mean_token_accuracy": 0.201119564473629,
|
|
"num_tokens": 9253981.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.448375129699707,
|
|
"epoch": 4.064516129032258,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003787819567113427,
|
|
"loss": 4.8185,
|
|
"mean_token_accuracy": 0.21393919587135315,
|
|
"num_tokens": 9264262.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.404143857955932,
|
|
"epoch": 4.069059518400727,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003784682422021638,
|
|
"loss": 4.8338,
|
|
"mean_token_accuracy": 0.2134675547480583,
|
|
"num_tokens": 9273562.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.435106182098389,
|
|
"epoch": 4.073602907769196,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003781542723953828,
|
|
"loss": 4.8604,
|
|
"mean_token_accuracy": 0.20632467567920684,
|
|
"num_tokens": 9283171.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.417014169692993,
|
|
"epoch": 4.078146297137665,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00037784004806568885,
|
|
"loss": 4.8388,
|
|
"mean_token_accuracy": 0.21170027554035187,
|
|
"num_tokens": 9292624.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.468687534332275,
|
|
"epoch": 4.082689686506134,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037752556998839934,
|
|
"loss": 4.9181,
|
|
"mean_token_accuracy": 0.20699366182088852,
|
|
"num_tokens": 9303159.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.435090255737305,
|
|
"epoch": 4.087233075874602,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003772108389394576,
|
|
"loss": 4.8535,
|
|
"mean_token_accuracy": 0.21560318768024445,
|
|
"num_tokens": 9314340.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.087233075874602,
|
|
"eval_entropy": 5.251365765859914,
|
|
"eval_loss": 5.6934661865234375,
|
|
"eval_mean_token_accuracy": 0.1705728180186693,
|
|
"eval_num_tokens": 9314340.0,
|
|
"eval_runtime": 1.9925,
|
|
"eval_samples_per_second": 1723.978,
|
|
"eval_steps_per_second": 215.811,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.490742158889771,
|
|
"epoch": 4.091776465243071,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037689585569543137,
|
|
"loss": 4.9377,
|
|
"mean_token_accuracy": 0.21222987323999404,
|
|
"num_tokens": 9325381.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.4283205509185795,
|
|
"epoch": 4.09631985461154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003765806210335103,
|
|
"loss": 4.9066,
|
|
"mean_token_accuracy": 0.20457319170236588,
|
|
"num_tokens": 9336639.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.455983257293701,
|
|
"epoch": 4.100863243980009,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037626513573150473,
|
|
"loss": 4.8121,
|
|
"mean_token_accuracy": 0.21526494473218918,
|
|
"num_tokens": 9346794.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.348176765441894,
|
|
"epoch": 4.105406633348478,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037594940056784315,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.21539445370435714,
|
|
"num_tokens": 9356326.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.4439308643341064,
|
|
"epoch": 4.109950022716947,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003756334163215708,
|
|
"loss": 4.8573,
|
|
"mean_token_accuracy": 0.21661428064107896,
|
|
"num_tokens": 9366402.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.428639459609985,
|
|
"epoch": 4.114493412085416,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00037531718377234723,
|
|
"loss": 4.8061,
|
|
"mean_token_accuracy": 0.21679544746875762,
|
|
"num_tokens": 9377067.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.435898876190185,
|
|
"epoch": 4.119036801453885,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003750007037004451,
|
|
"loss": 4.8693,
|
|
"mean_token_accuracy": 0.2102092131972313,
|
|
"num_tokens": 9387206.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.428882169723511,
|
|
"epoch": 4.123580190822353,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003746839768867473,
|
|
"loss": 4.8619,
|
|
"mean_token_accuracy": 0.21384917497634887,
|
|
"num_tokens": 9397195.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.419287872314453,
|
|
"epoch": 4.128123580190822,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003743670041127458,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.21329186707735062,
|
|
"num_tokens": 9407611.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.413600301742553,
|
|
"epoch": 4.132666969559291,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003740497861605395,
|
|
"loss": 4.9305,
|
|
"mean_token_accuracy": 0.20604915916919708,
|
|
"num_tokens": 9417897.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.425560474395752,
|
|
"epoch": 4.13721035892776,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000373732323812832,
|
|
"loss": 4.8551,
|
|
"mean_token_accuracy": 0.20512489527463912,
|
|
"num_tokens": 9428896.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.412872409820556,
|
|
"epoch": 4.141753748296229,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037341461785293025,
|
|
"loss": 4.7815,
|
|
"mean_token_accuracy": 0.21886242032051087,
|
|
"num_tokens": 9438854.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.331627750396729,
|
|
"epoch": 4.146297137664698,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003730966690647421,
|
|
"loss": 4.8602,
|
|
"mean_token_accuracy": 0.2081120043992996,
|
|
"num_tokens": 9447798.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.411174154281616,
|
|
"epoch": 4.150840527033167,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037277847823277444,
|
|
"loss": 4.8135,
|
|
"mean_token_accuracy": 0.22405674308538437,
|
|
"num_tokens": 9458930.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.34289436340332,
|
|
"epoch": 4.155383916401636,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037246004614213164,
|
|
"loss": 4.8015,
|
|
"mean_token_accuracy": 0.2197309195995331,
|
|
"num_tokens": 9469740.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.3668474674224855,
|
|
"epoch": 4.159927305770104,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003721413735785132,
|
|
"loss": 4.8007,
|
|
"mean_token_accuracy": 0.21807595640420913,
|
|
"num_tokens": 9479823.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.43460865020752,
|
|
"epoch": 4.164470695138573,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000371822461328212,
|
|
"loss": 4.7905,
|
|
"mean_token_accuracy": 0.21840759366750717,
|
|
"num_tokens": 9489995.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.37713098526001,
|
|
"epoch": 4.169014084507042,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037150331017811233,
|
|
"loss": 4.8507,
|
|
"mean_token_accuracy": 0.21205107867717743,
|
|
"num_tokens": 9501311.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.421172094345093,
|
|
"epoch": 4.173557473875511,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003711839209156878,
|
|
"loss": 4.9182,
|
|
"mean_token_accuracy": 0.20520758032798767,
|
|
"num_tokens": 9511901.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.462074947357178,
|
|
"epoch": 4.17810086324398,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003708642943289999,
|
|
"loss": 4.8653,
|
|
"mean_token_accuracy": 0.20613262355327605,
|
|
"num_tokens": 9522908.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.464492607116699,
|
|
"epoch": 4.182644252612449,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003705444312066953,
|
|
"loss": 4.9442,
|
|
"mean_token_accuracy": 0.20734427720308304,
|
|
"num_tokens": 9533363.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.392503833770752,
|
|
"epoch": 4.187187641980918,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003702243323380045,
|
|
"loss": 4.8272,
|
|
"mean_token_accuracy": 0.2122980237007141,
|
|
"num_tokens": 9543302.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.408284902572632,
|
|
"epoch": 4.191731031349387,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00036990399851273965,
|
|
"loss": 4.8489,
|
|
"mean_token_accuracy": 0.20545038282871247,
|
|
"num_tokens": 9553143.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.465175771713257,
|
|
"epoch": 4.196274420717856,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003695834305212925,
|
|
"loss": 4.872,
|
|
"mean_token_accuracy": 0.2122694119811058,
|
|
"num_tokens": 9564236.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.430713224411011,
|
|
"epoch": 4.200817810086324,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036926262915463294,
|
|
"loss": 4.8457,
|
|
"mean_token_accuracy": 0.21467580795288085,
|
|
"num_tokens": 9574363.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.374246454238891,
|
|
"epoch": 4.205361199454793,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036894159520430616,
|
|
"loss": 4.8879,
|
|
"mean_token_accuracy": 0.2082630217075348,
|
|
"num_tokens": 9585631.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.39557728767395,
|
|
"epoch": 4.209904588823262,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036862032946243173,
|
|
"loss": 4.8611,
|
|
"mean_token_accuracy": 0.20572549402713775,
|
|
"num_tokens": 9596826.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.313783645629883,
|
|
"epoch": 4.214447978191731,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00036829883272170087,
|
|
"loss": 4.8134,
|
|
"mean_token_accuracy": 0.21390560418367385,
|
|
"num_tokens": 9608158.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.416778945922852,
|
|
"epoch": 4.2189913675602,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036797710577537486,
|
|
"loss": 4.9206,
|
|
"mean_token_accuracy": 0.2072787404060364,
|
|
"num_tokens": 9619437.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.480019474029541,
|
|
"epoch": 4.223534756928669,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036765514941728295,
|
|
"loss": 4.9038,
|
|
"mean_token_accuracy": 0.21320116966962815,
|
|
"num_tokens": 9629913.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.393963098526001,
|
|
"epoch": 4.228078146297138,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003673329644418204,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.21280837655067444,
|
|
"num_tokens": 9640033.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.4217571258544925,
|
|
"epoch": 4.232621535665607,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036701055164394664,
|
|
"loss": 4.9016,
|
|
"mean_token_accuracy": 0.20516521036624907,
|
|
"num_tokens": 9651010.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.368144464492798,
|
|
"epoch": 4.237164925034075,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036668791181918313,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.2147320553660393,
|
|
"num_tokens": 9661549.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.374210357666016,
|
|
"epoch": 4.241708314402544,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003663650457636117,
|
|
"loss": 4.8724,
|
|
"mean_token_accuracy": 0.20458778887987136,
|
|
"num_tokens": 9671619.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.489054012298584,
|
|
"epoch": 4.246251703771013,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003660419542738722,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.21375705897808076,
|
|
"num_tokens": 9682582.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.3594221591949465,
|
|
"epoch": 4.250795093139482,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003657186381471608,
|
|
"loss": 4.8328,
|
|
"mean_token_accuracy": 0.2152670830488205,
|
|
"num_tokens": 9691997.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.414789438247681,
|
|
"epoch": 4.255338482507951,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036539509818122765,
|
|
"loss": 4.8941,
|
|
"mean_token_accuracy": 0.2105756714940071,
|
|
"num_tokens": 9702105.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.453825330734253,
|
|
"epoch": 4.25988187187642,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003650713351743758,
|
|
"loss": 4.8811,
|
|
"mean_token_accuracy": 0.21139561831951142,
|
|
"num_tokens": 9712347.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.377915716171264,
|
|
"epoch": 4.264425261244889,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003647473499254581,
|
|
"loss": 4.8747,
|
|
"mean_token_accuracy": 0.2110730677843094,
|
|
"num_tokens": 9723116.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.359844493865967,
|
|
"epoch": 4.268968650613358,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000364423143233876,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.21078843325376512,
|
|
"num_tokens": 9732656.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.415643739700317,
|
|
"epoch": 4.273512039981826,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036409871589957723,
|
|
"loss": 4.9445,
|
|
"mean_token_accuracy": 0.20847513973712922,
|
|
"num_tokens": 9743312.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.367781496047973,
|
|
"epoch": 4.278055429350295,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000363774068723054,
|
|
"loss": 4.8728,
|
|
"mean_token_accuracy": 0.20900074541568756,
|
|
"num_tokens": 9753358.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.3835663318634035,
|
|
"epoch": 4.282598818718764,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036344920250534106,
|
|
"loss": 4.7745,
|
|
"mean_token_accuracy": 0.2167047530412674,
|
|
"num_tokens": 9763472.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.387544298171997,
|
|
"epoch": 4.287142208087233,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003631241180480133,
|
|
"loss": 4.7899,
|
|
"mean_token_accuracy": 0.21481245905160903,
|
|
"num_tokens": 9773700.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.360150003433228,
|
|
"epoch": 4.291685597455702,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003627988161531845,
|
|
"loss": 4.8247,
|
|
"mean_token_accuracy": 0.21528870910406112,
|
|
"num_tokens": 9784042.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.441643857955933,
|
|
"epoch": 4.296228986824171,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003624732976235046,
|
|
"loss": 4.9586,
|
|
"mean_token_accuracy": 0.2104850798845291,
|
|
"num_tokens": 9794464.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.411430931091308,
|
|
"epoch": 4.30077237619264,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003621475632621584,
|
|
"loss": 4.8806,
|
|
"mean_token_accuracy": 0.2108780786395073,
|
|
"num_tokens": 9804459.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.400139236450196,
|
|
"epoch": 4.305315765561109,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003618216138728627,
|
|
"loss": 4.8666,
|
|
"mean_token_accuracy": 0.2127412810921669,
|
|
"num_tokens": 9814962.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.400909948348999,
|
|
"epoch": 4.309859154929577,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036149545025986567,
|
|
"loss": 4.9174,
|
|
"mean_token_accuracy": 0.2054225444793701,
|
|
"num_tokens": 9825184.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.473492097854614,
|
|
"epoch": 4.314402544298046,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036116907322794335,
|
|
"loss": 4.9529,
|
|
"mean_token_accuracy": 0.20260304510593413,
|
|
"num_tokens": 9835184.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.428992366790771,
|
|
"epoch": 4.318945933666515,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00036084248358239857,
|
|
"loss": 4.8643,
|
|
"mean_token_accuracy": 0.2087535575032234,
|
|
"num_tokens": 9845358.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.35491681098938,
|
|
"epoch": 4.323489323034984,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00036051568212905897,
|
|
"loss": 4.8124,
|
|
"mean_token_accuracy": 0.21376266032457353,
|
|
"num_tokens": 9856519.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.407486915588379,
|
|
"epoch": 4.328032712403453,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003601886696742745,
|
|
"loss": 4.8722,
|
|
"mean_token_accuracy": 0.21080042719841002,
|
|
"num_tokens": 9867478.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.34118013381958,
|
|
"epoch": 4.332576101771922,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000359861447024916,
|
|
"loss": 4.8585,
|
|
"mean_token_accuracy": 0.21267126202583314,
|
|
"num_tokens": 9877709.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.35651969909668,
|
|
"epoch": 4.337119491140391,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035953401498837285,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.21771082878112794,
|
|
"num_tokens": 9888134.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.374936771392822,
|
|
"epoch": 4.34166288050886,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035920637437255096,
|
|
"loss": 4.8767,
|
|
"mean_token_accuracy": 0.2123656839132309,
|
|
"num_tokens": 9899150.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.41515998840332,
|
|
"epoch": 4.346206269877328,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000358878525985871,
|
|
"loss": 4.8745,
|
|
"mean_token_accuracy": 0.20994329452514648,
|
|
"num_tokens": 9909599.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.362968111038208,
|
|
"epoch": 4.350749659245797,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003585504706372663,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.22168793082237243,
|
|
"num_tokens": 9919607.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.404587078094482,
|
|
"epoch": 4.355293048614266,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003582222091361808,
|
|
"loss": 4.8921,
|
|
"mean_token_accuracy": 0.20846357494592666,
|
|
"num_tokens": 9930431.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.375918006896972,
|
|
"epoch": 4.359836437982735,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003578937422925671,
|
|
"loss": 4.8574,
|
|
"mean_token_accuracy": 0.20886399745941162,
|
|
"num_tokens": 9940647.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.369810628890991,
|
|
"epoch": 4.364379827351204,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003575650709168846,
|
|
"loss": 4.8848,
|
|
"mean_token_accuracy": 0.20917197912931443,
|
|
"num_tokens": 9951110.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.420174026489258,
|
|
"epoch": 4.368923216719673,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035723619582009715,
|
|
"loss": 4.8552,
|
|
"mean_token_accuracy": 0.208492012321949,
|
|
"num_tokens": 9961158.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.389772987365722,
|
|
"epoch": 4.373466606088142,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00035690711781367137,
|
|
"loss": 4.8816,
|
|
"mean_token_accuracy": 0.20689976960420609,
|
|
"num_tokens": 9971051.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.313707447052002,
|
|
"epoch": 4.378009995456611,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003565778377095744,
|
|
"loss": 4.8634,
|
|
"mean_token_accuracy": 0.20521025210618973,
|
|
"num_tokens": 9981990.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.349678659439087,
|
|
"epoch": 4.38255338482508,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003562483563202724,
|
|
"loss": 4.8011,
|
|
"mean_token_accuracy": 0.2155463695526123,
|
|
"num_tokens": 9991668.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.364760875701904,
|
|
"epoch": 4.387096774193548,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035591867445872784,
|
|
"loss": 4.7507,
|
|
"mean_token_accuracy": 0.21755503863096237,
|
|
"num_tokens": 10001519.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.297342014312744,
|
|
"epoch": 4.391640163562017,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000355588792938398,
|
|
"loss": 4.8252,
|
|
"mean_token_accuracy": 0.2145224317908287,
|
|
"num_tokens": 10010856.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.314807891845703,
|
|
"epoch": 4.396183552930486,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003552587125732326,
|
|
"loss": 4.8354,
|
|
"mean_token_accuracy": 0.21747947931289674,
|
|
"num_tokens": 10020647.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.368133497238159,
|
|
"epoch": 4.400726942298955,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035492843417767224,
|
|
"loss": 4.7735,
|
|
"mean_token_accuracy": 0.22234252691268921,
|
|
"num_tokens": 10030114.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.4257495403289795,
|
|
"epoch": 4.405270331667424,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000354597958566646,
|
|
"loss": 4.8514,
|
|
"mean_token_accuracy": 0.2166033834218979,
|
|
"num_tokens": 10039863.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.309572219848633,
|
|
"epoch": 4.409813721035893,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035426728655556954,
|
|
"loss": 4.8143,
|
|
"mean_token_accuracy": 0.2070244088768959,
|
|
"num_tokens": 10049111.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.3559996604919435,
|
|
"epoch": 4.414357110404362,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003539364189603433,
|
|
"loss": 4.8969,
|
|
"mean_token_accuracy": 0.21049066185951232,
|
|
"num_tokens": 10058762.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.407307481765747,
|
|
"epoch": 4.418900499772831,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035360535659735006,
|
|
"loss": 4.8932,
|
|
"mean_token_accuracy": 0.2056547686457634,
|
|
"num_tokens": 10068587.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.431714582443237,
|
|
"epoch": 4.423443889141299,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003532741002834534,
|
|
"loss": 4.8636,
|
|
"mean_token_accuracy": 0.20861863493919372,
|
|
"num_tokens": 10079386.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.370867919921875,
|
|
"epoch": 4.427987278509768,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035294265083599526,
|
|
"loss": 4.7958,
|
|
"mean_token_accuracy": 0.21611275374889374,
|
|
"num_tokens": 10089490.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.376736974716186,
|
|
"epoch": 4.432530667878237,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003526110090727942,
|
|
"loss": 4.8866,
|
|
"mean_token_accuracy": 0.2097605675458908,
|
|
"num_tokens": 10100465.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.396946430206299,
|
|
"epoch": 4.437074057246706,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003522791758121434,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.21910446286201476,
|
|
"num_tokens": 10110919.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.3741209506988525,
|
|
"epoch": 4.441617446615175,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035194715187280835,
|
|
"loss": 4.8495,
|
|
"mean_token_accuracy": 0.21505178958177568,
|
|
"num_tokens": 10121116.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.457965087890625,
|
|
"epoch": 4.446160835983644,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003516149380740252,
|
|
"loss": 4.9606,
|
|
"mean_token_accuracy": 0.20762290954589843,
|
|
"num_tokens": 10132269.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.366045713424683,
|
|
"epoch": 4.450704225352113,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003512825352354983,
|
|
"loss": 4.8501,
|
|
"mean_token_accuracy": 0.21736513525247575,
|
|
"num_tokens": 10142856.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.324926900863647,
|
|
"epoch": 4.455247614720582,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035094994417739884,
|
|
"loss": 4.8187,
|
|
"mean_token_accuracy": 0.2131066158413887,
|
|
"num_tokens": 10152377.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.428145217895508,
|
|
"epoch": 4.45979100408905,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000350617165720362,
|
|
"loss": 4.9053,
|
|
"mean_token_accuracy": 0.20840137600898742,
|
|
"num_tokens": 10161961.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.445039558410644,
|
|
"epoch": 4.464334393457519,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035028420068548567,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.2132912188768387,
|
|
"num_tokens": 10171624.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.42944369316101,
|
|
"epoch": 4.468877782825988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003499510498943279,
|
|
"loss": 4.9438,
|
|
"mean_token_accuracy": 0.2046991303563118,
|
|
"num_tokens": 10182642.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.40150728225708,
|
|
"epoch": 4.473421172194457,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034961771416890526,
|
|
"loss": 4.9123,
|
|
"mean_token_accuracy": 0.2135506883263588,
|
|
"num_tokens": 10193307.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.380010175704956,
|
|
"epoch": 4.477964561562926,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003492841943316903,
|
|
"loss": 4.9009,
|
|
"mean_token_accuracy": 0.20946591347455978,
|
|
"num_tokens": 10203506.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.360426807403565,
|
|
"epoch": 4.482507950931395,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003489504912056103,
|
|
"loss": 4.787,
|
|
"mean_token_accuracy": 0.21468471735715866,
|
|
"num_tokens": 10214266.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.273831415176391,
|
|
"epoch": 4.487051340299864,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003486166056140445,
|
|
"loss": 4.784,
|
|
"mean_token_accuracy": 0.21740013360977173,
|
|
"num_tokens": 10224953.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.3163713932037355,
|
|
"epoch": 4.491594729668333,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034828253838082223,
|
|
"loss": 4.8291,
|
|
"mean_token_accuracy": 0.21027301251888275,
|
|
"num_tokens": 10235435.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.350422716140747,
|
|
"epoch": 4.496138119036801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034794829033022134,
|
|
"loss": 4.8323,
|
|
"mean_token_accuracy": 0.21812544912099838,
|
|
"num_tokens": 10245677.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.2526415348052975,
|
|
"epoch": 4.50068150840527,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003476138622869657,
|
|
"loss": 4.8091,
|
|
"mean_token_accuracy": 0.2112659454345703,
|
|
"num_tokens": 10256135.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.348266553878784,
|
|
"epoch": 4.505224897773739,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003472792550762232,
|
|
"loss": 4.8998,
|
|
"mean_token_accuracy": 0.20848560482263565,
|
|
"num_tokens": 10266465.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.478057861328125,
|
|
"epoch": 4.509768287142208,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003469444695236038,
|
|
"loss": 4.8985,
|
|
"mean_token_accuracy": 0.208941650390625,
|
|
"num_tokens": 10278123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.346586418151856,
|
|
"epoch": 4.514311676510677,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034660950645515765,
|
|
"loss": 4.8551,
|
|
"mean_token_accuracy": 0.209741073846817,
|
|
"num_tokens": 10287486.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.356811475753784,
|
|
"epoch": 4.518855065879146,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003462743666973728,
|
|
"loss": 4.8512,
|
|
"mean_token_accuracy": 0.2172999158501625,
|
|
"num_tokens": 10298100.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.4538891315460205,
|
|
"epoch": 4.523398455247615,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003459390510771734,
|
|
"loss": 4.9135,
|
|
"mean_token_accuracy": 0.20884933620691298,
|
|
"num_tokens": 10308549.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.394857454299927,
|
|
"epoch": 4.527941844616084,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003456035604219172,
|
|
"loss": 4.8768,
|
|
"mean_token_accuracy": 0.2091725766658783,
|
|
"num_tokens": 10319030.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.323186922073364,
|
|
"epoch": 4.532485233984552,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003452678955593942,
|
|
"loss": 4.9049,
|
|
"mean_token_accuracy": 0.21294318586587907,
|
|
"num_tokens": 10329613.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.404057168960572,
|
|
"epoch": 4.537028623353021,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000344932057317824,
|
|
"loss": 4.9215,
|
|
"mean_token_accuracy": 0.20849014818668365,
|
|
"num_tokens": 10339844.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.428472185134888,
|
|
"epoch": 4.54157201272149,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003445960465258542,
|
|
"loss": 4.9269,
|
|
"mean_token_accuracy": 0.20683681070804597,
|
|
"num_tokens": 10350906.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.54157201272149,
|
|
"eval_entropy": 5.227729323852894,
|
|
"eval_loss": 5.66549015045166,
|
|
"eval_mean_token_accuracy": 0.17281190358795398,
|
|
"eval_num_tokens": 10350906.0,
|
|
"eval_runtime": 1.9875,
|
|
"eval_samples_per_second": 1728.307,
|
|
"eval_steps_per_second": 216.353,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.326890897750855,
|
|
"epoch": 4.546115402089959,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003442598640125578,
|
|
"loss": 4.8757,
|
|
"mean_token_accuracy": 0.2059277504682541,
|
|
"num_tokens": 10361574.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.4257440090179445,
|
|
"epoch": 4.550658791458428,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000343923510607432,
|
|
"loss": 4.8934,
|
|
"mean_token_accuracy": 0.2083863154053688,
|
|
"num_tokens": 10372714.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.3036415576934814,
|
|
"epoch": 4.555202180826897,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003435869871403952,
|
|
"loss": 4.8008,
|
|
"mean_token_accuracy": 0.2128177896142006,
|
|
"num_tokens": 10382143.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.364067220687867,
|
|
"epoch": 4.559745570195366,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034325029444178565,
|
|
"loss": 4.8418,
|
|
"mean_token_accuracy": 0.20842435210943222,
|
|
"num_tokens": 10392040.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.41144871711731,
|
|
"epoch": 4.564288959563835,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034291343334235915,
|
|
"loss": 4.9153,
|
|
"mean_token_accuracy": 0.2011289969086647,
|
|
"num_tokens": 10403230.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.300043344497681,
|
|
"epoch": 4.568832348932304,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003425764046732869,
|
|
"loss": 4.8133,
|
|
"mean_token_accuracy": 0.2139666795730591,
|
|
"num_tokens": 10413466.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.341774272918701,
|
|
"epoch": 4.573375738300772,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034223920926615376,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.21751508712768555,
|
|
"num_tokens": 10422559.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.363039970397949,
|
|
"epoch": 4.577919127669241,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034190184795295577,
|
|
"loss": 4.812,
|
|
"mean_token_accuracy": 0.21568000465631484,
|
|
"num_tokens": 10432544.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.381924676895141,
|
|
"epoch": 4.58246251703771,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034156432156609844,
|
|
"loss": 4.9062,
|
|
"mean_token_accuracy": 0.20612457841634751,
|
|
"num_tokens": 10442014.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.312402963638306,
|
|
"epoch": 4.587005906406179,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034122663093839466,
|
|
"loss": 4.749,
|
|
"mean_token_accuracy": 0.22080722749233245,
|
|
"num_tokens": 10451093.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.33369779586792,
|
|
"epoch": 4.591549295774648,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003408887769030624,
|
|
"loss": 4.7695,
|
|
"mean_token_accuracy": 0.21237638741731643,
|
|
"num_tokens": 10461678.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.303464603424072,
|
|
"epoch": 4.596092685143117,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000340550760293723,
|
|
"loss": 4.816,
|
|
"mean_token_accuracy": 0.21297024935483932,
|
|
"num_tokens": 10471607.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.391111373901367,
|
|
"epoch": 4.600636074511586,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034021258194439876,
|
|
"loss": 4.8841,
|
|
"mean_token_accuracy": 0.21189649701118468,
|
|
"num_tokens": 10482453.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.291701173782348,
|
|
"epoch": 4.605179463880054,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033987424268951117,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.21905431151390076,
|
|
"num_tokens": 10492424.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.321639823913574,
|
|
"epoch": 4.609722853248523,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00033953574336387867,
|
|
"loss": 4.8132,
|
|
"mean_token_accuracy": 0.21207965761423112,
|
|
"num_tokens": 10500958.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.344120693206787,
|
|
"epoch": 4.614266242616992,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003391970848027148,
|
|
"loss": 4.8917,
|
|
"mean_token_accuracy": 0.21222373545169831,
|
|
"num_tokens": 10510906.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.356979608535767,
|
|
"epoch": 4.618809631985461,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003388582678416257,
|
|
"loss": 4.8071,
|
|
"mean_token_accuracy": 0.2138013705611229,
|
|
"num_tokens": 10521832.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.338470840454102,
|
|
"epoch": 4.62335302135393,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033851929331660876,
|
|
"loss": 4.7956,
|
|
"mean_token_accuracy": 0.21778690069913864,
|
|
"num_tokens": 10532310.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.3425768375396725,
|
|
"epoch": 4.627896410722399,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033818016206404983,
|
|
"loss": 4.8397,
|
|
"mean_token_accuracy": 0.21032117456197738,
|
|
"num_tokens": 10543979.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.480816745758057,
|
|
"epoch": 4.632439800090868,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003378408749207215,
|
|
"loss": 4.9221,
|
|
"mean_token_accuracy": 0.20593876242637635,
|
|
"num_tokens": 10553757.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.355796813964844,
|
|
"epoch": 4.636983189459337,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003375014327237812,
|
|
"loss": 4.8761,
|
|
"mean_token_accuracy": 0.2059740364551544,
|
|
"num_tokens": 10563575.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.378340244293213,
|
|
"epoch": 4.641526578827806,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003371618363107686,
|
|
"loss": 4.9099,
|
|
"mean_token_accuracy": 0.20230767428874968,
|
|
"num_tokens": 10572828.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.335450315475464,
|
|
"epoch": 4.646069968196274,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033682208651960444,
|
|
"loss": 4.8574,
|
|
"mean_token_accuracy": 0.2099425196647644,
|
|
"num_tokens": 10582567.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.337552309036255,
|
|
"epoch": 4.650613357564743,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033648218418858716,
|
|
"loss": 4.849,
|
|
"mean_token_accuracy": 0.2172555670142174,
|
|
"num_tokens": 10594046.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.386779069900513,
|
|
"epoch": 4.655156746933212,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003361421301563922,
|
|
"loss": 4.8351,
|
|
"mean_token_accuracy": 0.21576008945703506,
|
|
"num_tokens": 10604343.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.310541105270386,
|
|
"epoch": 4.659700136301681,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003358019252620691,
|
|
"loss": 4.9066,
|
|
"mean_token_accuracy": 0.20525662302970887,
|
|
"num_tokens": 10615140.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.39592809677124,
|
|
"epoch": 4.66424352567015,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033546157034503954,
|
|
"loss": 4.8647,
|
|
"mean_token_accuracy": 0.21180385798215867,
|
|
"num_tokens": 10624768.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.382642984390259,
|
|
"epoch": 4.668786915038619,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003351210662450955,
|
|
"loss": 4.7867,
|
|
"mean_token_accuracy": 0.2164941281080246,
|
|
"num_tokens": 10635246.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.344247245788575,
|
|
"epoch": 4.673330304407088,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003347804138023968,
|
|
"loss": 4.7835,
|
|
"mean_token_accuracy": 0.22070907056331635,
|
|
"num_tokens": 10645118.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.336080074310303,
|
|
"epoch": 4.677873693775557,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003344396138574698,
|
|
"loss": 4.8017,
|
|
"mean_token_accuracy": 0.21817314624786377,
|
|
"num_tokens": 10655846.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.375018310546875,
|
|
"epoch": 4.682417083144026,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003340986672512042,
|
|
"loss": 4.9008,
|
|
"mean_token_accuracy": 0.20738235116004944,
|
|
"num_tokens": 10665741.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.374571132659912,
|
|
"epoch": 4.686960472512494,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003337575748248521,
|
|
"loss": 4.8576,
|
|
"mean_token_accuracy": 0.21453993022441864,
|
|
"num_tokens": 10676120.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.355083179473877,
|
|
"epoch": 4.691503861880963,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033341633742002503,
|
|
"loss": 4.8483,
|
|
"mean_token_accuracy": 0.20889182835817338,
|
|
"num_tokens": 10686195.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.341623735427857,
|
|
"epoch": 4.696047251249432,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003330749558786924,
|
|
"loss": 4.8775,
|
|
"mean_token_accuracy": 0.2124635860323906,
|
|
"num_tokens": 10697278.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.394570970535279,
|
|
"epoch": 4.700590640617901,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003327334310431791,
|
|
"loss": 4.903,
|
|
"mean_token_accuracy": 0.21280580163002014,
|
|
"num_tokens": 10707937.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.370965766906738,
|
|
"epoch": 4.70513402998637,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003323917637561638,
|
|
"loss": 4.7674,
|
|
"mean_token_accuracy": 0.21251425743103028,
|
|
"num_tokens": 10717963.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.343652009963989,
|
|
"epoch": 4.709677419354839,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033204995486067664,
|
|
"loss": 4.8684,
|
|
"mean_token_accuracy": 0.20727491825819017,
|
|
"num_tokens": 10728294.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.469464635848999,
|
|
"epoch": 4.714220808723308,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033170800520009703,
|
|
"loss": 4.9222,
|
|
"mean_token_accuracy": 0.20771244764328003,
|
|
"num_tokens": 10739617.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.390109920501709,
|
|
"epoch": 4.718764198091776,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003313659156181518,
|
|
"loss": 4.9271,
|
|
"mean_token_accuracy": 0.21015261709690095,
|
|
"num_tokens": 10750704.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.352267122268676,
|
|
"epoch": 4.723307587460245,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033102368695891283,
|
|
"loss": 4.821,
|
|
"mean_token_accuracy": 0.221248596906662,
|
|
"num_tokens": 10760106.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.466974210739136,
|
|
"epoch": 4.727850976828714,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033068132006679554,
|
|
"loss": 4.9894,
|
|
"mean_token_accuracy": 0.2038916304707527,
|
|
"num_tokens": 10770963.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.42772216796875,
|
|
"epoch": 4.732394366197183,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033033881578655605,
|
|
"loss": 4.9328,
|
|
"mean_token_accuracy": 0.20843742787837982,
|
|
"num_tokens": 10781441.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.428295755386353,
|
|
"epoch": 4.736937755565652,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003299961749632898,
|
|
"loss": 4.8974,
|
|
"mean_token_accuracy": 0.21269915699958802,
|
|
"num_tokens": 10791548.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.383069801330566,
|
|
"epoch": 4.741481144934121,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032965339844242864,
|
|
"loss": 4.8561,
|
|
"mean_token_accuracy": 0.21173325479030608,
|
|
"num_tokens": 10801525.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.334641742706299,
|
|
"epoch": 4.74602453430259,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003293104870697399,
|
|
"loss": 4.8225,
|
|
"mean_token_accuracy": 0.2111858010292053,
|
|
"num_tokens": 10811907.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.358064365386963,
|
|
"epoch": 4.750567923671059,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032896744169132305,
|
|
"loss": 4.8317,
|
|
"mean_token_accuracy": 0.21784939765930175,
|
|
"num_tokens": 10822996.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.329522180557251,
|
|
"epoch": 4.755111313039528,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003286242631536087,
|
|
"loss": 4.7795,
|
|
"mean_token_accuracy": 0.20881662219762803,
|
|
"num_tokens": 10832561.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.383299207687378,
|
|
"epoch": 4.759654702407996,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032828095230335555,
|
|
"loss": 4.8763,
|
|
"mean_token_accuracy": 0.20729224979877472,
|
|
"num_tokens": 10842591.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.3380115032196045,
|
|
"epoch": 4.764198091776465,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003279375099876491,
|
|
"loss": 4.891,
|
|
"mean_token_accuracy": 0.207807657122612,
|
|
"num_tokens": 10852888.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.419365406036377,
|
|
"epoch": 4.768741481144934,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003275939370538992,
|
|
"loss": 4.9176,
|
|
"mean_token_accuracy": 0.20689819306135177,
|
|
"num_tokens": 10863792.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.326735591888427,
|
|
"epoch": 4.773284870513403,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032725023434983767,
|
|
"loss": 4.7821,
|
|
"mean_token_accuracy": 0.2197741374373436,
|
|
"num_tokens": 10873206.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.332780599594116,
|
|
"epoch": 4.777828259881872,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032690640272351715,
|
|
"loss": 4.9129,
|
|
"mean_token_accuracy": 0.20579216927289962,
|
|
"num_tokens": 10883686.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.299846887588501,
|
|
"epoch": 4.782371649250341,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003265624430233076,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.2157586082816124,
|
|
"num_tokens": 10894416.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.463975667953491,
|
|
"epoch": 4.78691503861881,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003262183560978956,
|
|
"loss": 4.9578,
|
|
"mean_token_accuracy": 0.21369886994361878,
|
|
"num_tokens": 10904405.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.456334590911865,
|
|
"epoch": 4.791458427987278,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003258741427962814,
|
|
"loss": 4.8626,
|
|
"mean_token_accuracy": 0.21287580877542495,
|
|
"num_tokens": 10915339.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.348273706436157,
|
|
"epoch": 4.796001817355747,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032552980396777714,
|
|
"loss": 4.8437,
|
|
"mean_token_accuracy": 0.2087205708026886,
|
|
"num_tokens": 10926373.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.342918109893799,
|
|
"epoch": 4.800545206724216,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032518534046200464,
|
|
"loss": 4.9234,
|
|
"mean_token_accuracy": 0.2112068086862564,
|
|
"num_tokens": 10936695.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.4380073070526125,
|
|
"epoch": 4.805088596092685,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003248407531288932,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.19943662881851196,
|
|
"num_tokens": 10949741.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.44051308631897,
|
|
"epoch": 4.809631985461154,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000324496042818678,
|
|
"loss": 4.9137,
|
|
"mean_token_accuracy": 0.21087245792150497,
|
|
"num_tokens": 10960786.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.382336902618408,
|
|
"epoch": 4.814175374829623,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032415121038189746,
|
|
"loss": 4.8304,
|
|
"mean_token_accuracy": 0.21864665746688844,
|
|
"num_tokens": 10970423.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.3489118099212645,
|
|
"epoch": 4.818718764198092,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032380625666939125,
|
|
"loss": 4.8563,
|
|
"mean_token_accuracy": 0.21119421124458312,
|
|
"num_tokens": 10980196.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.349702787399292,
|
|
"epoch": 4.823262153566561,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032346118253229834,
|
|
"loss": 4.8648,
|
|
"mean_token_accuracy": 0.21201564818620683,
|
|
"num_tokens": 10989857.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.381877756118774,
|
|
"epoch": 4.82780554293503,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032311598882205496,
|
|
"loss": 4.8642,
|
|
"mean_token_accuracy": 0.2016470640897751,
|
|
"num_tokens": 11000276.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.415317726135254,
|
|
"epoch": 4.832348932303498,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032277067639039216,
|
|
"loss": 4.9412,
|
|
"mean_token_accuracy": 0.19680726081132888,
|
|
"num_tokens": 11012502.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.362420988082886,
|
|
"epoch": 4.836892321671967,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032242524608933423,
|
|
"loss": 4.8063,
|
|
"mean_token_accuracy": 0.20612317323684692,
|
|
"num_tokens": 11023042.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.316581106185913,
|
|
"epoch": 4.841435711040436,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032207969877119587,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.22258250117301942,
|
|
"num_tokens": 11032259.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.419769763946533,
|
|
"epoch": 4.845979100408905,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032173403528858093,
|
|
"loss": 4.9839,
|
|
"mean_token_accuracy": 0.20984968692064285,
|
|
"num_tokens": 11042324.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.342652702331543,
|
|
"epoch": 4.850522489777374,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003213882564943796,
|
|
"loss": 4.877,
|
|
"mean_token_accuracy": 0.2114045262336731,
|
|
"num_tokens": 11053519.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.409711122512817,
|
|
"epoch": 4.855065879145843,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003210423632417666,
|
|
"loss": 4.8511,
|
|
"mean_token_accuracy": 0.21706441938877105,
|
|
"num_tokens": 11063413.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.2955560207366945,
|
|
"epoch": 4.859609268514312,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003206963563841993,
|
|
"loss": 4.8237,
|
|
"mean_token_accuracy": 0.20812415480613708,
|
|
"num_tokens": 11073232.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.305441188812256,
|
|
"epoch": 4.86415265788278,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032035023677541515,
|
|
"loss": 4.8001,
|
|
"mean_token_accuracy": 0.20180709660053253,
|
|
"num_tokens": 11083408.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.412430095672607,
|
|
"epoch": 4.868696047251249,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003200040052694299,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.20900274813175201,
|
|
"num_tokens": 11094125.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.384602069854736,
|
|
"epoch": 4.873239436619718,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031965766272053534,
|
|
"loss": 4.8852,
|
|
"mean_token_accuracy": 0.20741999000310898,
|
|
"num_tokens": 11106293.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.342703342437744,
|
|
"epoch": 4.877782825988187,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031931120998329737,
|
|
"loss": 4.7827,
|
|
"mean_token_accuracy": 0.22554007172584534,
|
|
"num_tokens": 11116485.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.404598903656006,
|
|
"epoch": 4.882326215356656,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003189646479125536,
|
|
"loss": 4.9298,
|
|
"mean_token_accuracy": 0.2034258395433426,
|
|
"num_tokens": 11127876.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.406229400634766,
|
|
"epoch": 4.886869604725125,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003186179773634115,
|
|
"loss": 4.9934,
|
|
"mean_token_accuracy": 0.2083194449543953,
|
|
"num_tokens": 11138871.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.40096607208252,
|
|
"epoch": 4.891412994093594,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031827119919124614,
|
|
"loss": 4.8887,
|
|
"mean_token_accuracy": 0.2108651801943779,
|
|
"num_tokens": 11148872.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.294561862945557,
|
|
"epoch": 4.895956383462063,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031792431425169826,
|
|
"loss": 4.7395,
|
|
"mean_token_accuracy": 0.22432301789522172,
|
|
"num_tokens": 11159716.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.390296745300293,
|
|
"epoch": 4.900499772830532,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00031757732340067195,
|
|
"loss": 4.8398,
|
|
"mean_token_accuracy": 0.21187656819820405,
|
|
"num_tokens": 11169187.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.420208597183228,
|
|
"epoch": 4.905043162199,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00031723022749433267,
|
|
"loss": 4.8024,
|
|
"mean_token_accuracy": 0.21941469609737396,
|
|
"num_tokens": 11179889.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.4078442573547365,
|
|
"epoch": 4.909586551567469,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000316883027389105,
|
|
"loss": 4.9853,
|
|
"mean_token_accuracy": 0.20336132794618605,
|
|
"num_tokens": 11189439.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.362090444564819,
|
|
"epoch": 4.914129940935938,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003165357239416706,
|
|
"loss": 4.8942,
|
|
"mean_token_accuracy": 0.21089393347501756,
|
|
"num_tokens": 11200834.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.363616323471069,
|
|
"epoch": 4.918673330304407,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003161883180089665,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.21451950669288636,
|
|
"num_tokens": 11210475.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.3733124256134035,
|
|
"epoch": 4.923216719672876,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003158408104481821,
|
|
"loss": 4.9019,
|
|
"mean_token_accuracy": 0.2019113630056381,
|
|
"num_tokens": 11220744.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.427550745010376,
|
|
"epoch": 4.927760109041345,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031549320211675776,
|
|
"loss": 4.96,
|
|
"mean_token_accuracy": 0.20023086965084075,
|
|
"num_tokens": 11231108.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.402320623397827,
|
|
"epoch": 4.932303498409814,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031514549387238246,
|
|
"loss": 4.8969,
|
|
"mean_token_accuracy": 0.21418522596359252,
|
|
"num_tokens": 11241160.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.358226251602173,
|
|
"epoch": 4.936846887778283,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003147976865729918,
|
|
"loss": 4.9284,
|
|
"mean_token_accuracy": 0.19985178858041763,
|
|
"num_tokens": 11251711.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.377100801467895,
|
|
"epoch": 4.941390277146752,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031444978107676567,
|
|
"loss": 4.874,
|
|
"mean_token_accuracy": 0.213515043258667,
|
|
"num_tokens": 11263025.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.369738674163818,
|
|
"epoch": 4.94593366651522,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031410177824212635,
|
|
"loss": 4.8655,
|
|
"mean_token_accuracy": 0.21094402223825454,
|
|
"num_tokens": 11273927.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.355129861831665,
|
|
"epoch": 4.950477055883689,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003137536789277362,
|
|
"loss": 4.8607,
|
|
"mean_token_accuracy": 0.20994258671998978,
|
|
"num_tokens": 11284108.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.367716264724732,
|
|
"epoch": 4.955020445252158,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003134054839924957,
|
|
"loss": 4.7895,
|
|
"mean_token_accuracy": 0.21770530343055725,
|
|
"num_tokens": 11294530.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.332775402069092,
|
|
"epoch": 4.959563834620627,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031305719429554124,
|
|
"loss": 4.8469,
|
|
"mean_token_accuracy": 0.20363475680351256,
|
|
"num_tokens": 11304803.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.364111471176147,
|
|
"epoch": 4.964107223989096,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003127088106962429,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.2099867880344391,
|
|
"num_tokens": 11314567.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.375747442245483,
|
|
"epoch": 4.968650613357565,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031236033405420284,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.20828319489955902,
|
|
"num_tokens": 11324452.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.395658922195435,
|
|
"epoch": 4.973194002726034,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031201176522925236,
|
|
"loss": 4.8714,
|
|
"mean_token_accuracy": 0.21161191910505295,
|
|
"num_tokens": 11334930.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.30916748046875,
|
|
"epoch": 4.977737392094502,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003116631050814503,
|
|
"loss": 4.7948,
|
|
"mean_token_accuracy": 0.21638847440481185,
|
|
"num_tokens": 11345487.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.390592908859253,
|
|
"epoch": 4.982280781462971,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031131435447108116,
|
|
"loss": 4.8706,
|
|
"mean_token_accuracy": 0.21087276339530944,
|
|
"num_tokens": 11355150.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.419683265686035,
|
|
"epoch": 4.98682417083144,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003109655142586521,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.21777509897947311,
|
|
"num_tokens": 11366642.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.346822118759155,
|
|
"epoch": 4.991367560199909,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031061658530489195,
|
|
"loss": 4.8814,
|
|
"mean_token_accuracy": 0.21102126240730285,
|
|
"num_tokens": 11377229.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.307001256942749,
|
|
"epoch": 4.995910949568378,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00031026756847074804,
|
|
"loss": 4.9178,
|
|
"mean_token_accuracy": 0.20795635879039764,
|
|
"num_tokens": 11387444.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.995910949568378,
|
|
"eval_entropy": 5.197316982579785,
|
|
"eval_loss": 5.621264457702637,
|
|
"eval_mean_token_accuracy": 0.17553743489606435,
|
|
"eval_num_tokens": 11387444.0,
|
|
"eval_runtime": 2.1912,
|
|
"eval_samples_per_second": 1567.602,
|
|
"eval_steps_per_second": 196.236,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.393439133961995,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.0003099184646173849,
|
|
"loss": 4.8648,
|
|
"mean_token_accuracy": 0.212064567539427,
|
|
"num_tokens": 11396985.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.400833082199097,
|
|
"epoch": 5.004543389368469,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030956927460618126,
|
|
"loss": 4.6722,
|
|
"mean_token_accuracy": 0.22733596861362457,
|
|
"num_tokens": 11408747.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.307932186126709,
|
|
"epoch": 5.009086778736938,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003092199992987291,
|
|
"loss": 4.5785,
|
|
"mean_token_accuracy": 0.23458762168884278,
|
|
"num_tokens": 11418695.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.2772789478302,
|
|
"epoch": 5.013630168105407,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030887063955683044,
|
|
"loss": 4.6619,
|
|
"mean_token_accuracy": 0.22885785847902299,
|
|
"num_tokens": 11428143.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.371218538284301,
|
|
"epoch": 5.018173557473876,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00030852119624249553,
|
|
"loss": 4.7071,
|
|
"mean_token_accuracy": 0.22211153507232667,
|
|
"num_tokens": 11439610.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.375484943389893,
|
|
"epoch": 5.022716946842344,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003081716702179412,
|
|
"loss": 4.6599,
|
|
"mean_token_accuracy": 0.2291506350040436,
|
|
"num_tokens": 11450035.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.301357269287109,
|
|
"epoch": 5.027260336210813,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000307822062345588,
|
|
"loss": 4.6219,
|
|
"mean_token_accuracy": 0.23560964316129684,
|
|
"num_tokens": 11460736.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.33759126663208,
|
|
"epoch": 5.031803725579282,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003074723734880588,
|
|
"loss": 4.6301,
|
|
"mean_token_accuracy": 0.22991645485162734,
|
|
"num_tokens": 11471319.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.296652221679688,
|
|
"epoch": 5.036347114947751,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000307122604508176,
|
|
"loss": 4.6203,
|
|
"mean_token_accuracy": 0.2356238842010498,
|
|
"num_tokens": 11481052.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.260279655456543,
|
|
"epoch": 5.04089050431622,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003067727562689598,
|
|
"loss": 4.6196,
|
|
"mean_token_accuracy": 0.22638531178236007,
|
|
"num_tokens": 11491166.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.341081666946411,
|
|
"epoch": 5.045433893684689,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003064228296336259,
|
|
"loss": 4.6113,
|
|
"mean_token_accuracy": 0.23285963535308837,
|
|
"num_tokens": 11501007.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.345428133010865,
|
|
"epoch": 5.049977283053158,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003060728254655836,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.22604602426290513,
|
|
"num_tokens": 11512197.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.392754697799683,
|
|
"epoch": 5.054520672421627,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030572274462843327,
|
|
"loss": 4.6683,
|
|
"mean_token_accuracy": 0.22846254855394363,
|
|
"num_tokens": 11522396.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.358619737625122,
|
|
"epoch": 5.059064061790095,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030537258798596465,
|
|
"loss": 4.6845,
|
|
"mean_token_accuracy": 0.22879939377307892,
|
|
"num_tokens": 11533034.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.3942962169647215,
|
|
"epoch": 5.063607451158564,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003050223564021543,
|
|
"loss": 4.6943,
|
|
"mean_token_accuracy": 0.22981866151094438,
|
|
"num_tokens": 11543281.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.329396104812622,
|
|
"epoch": 5.068150840527033,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003046720507411641,
|
|
"loss": 4.6505,
|
|
"mean_token_accuracy": 0.2324106961488724,
|
|
"num_tokens": 11552905.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.304842567443847,
|
|
"epoch": 5.072694229895502,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003043216718673383,
|
|
"loss": 4.659,
|
|
"mean_token_accuracy": 0.2374695673584938,
|
|
"num_tokens": 11562397.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.369515609741211,
|
|
"epoch": 5.077237619263971,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030397122064520186,
|
|
"loss": 4.72,
|
|
"mean_token_accuracy": 0.22334132045507432,
|
|
"num_tokens": 11573375.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.357832908630371,
|
|
"epoch": 5.08178100863244,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030362069793945856,
|
|
"loss": 4.6503,
|
|
"mean_token_accuracy": 0.22757819294929504,
|
|
"num_tokens": 11584155.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.276780462265014,
|
|
"epoch": 5.086324398000909,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003032701046149881,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.22619065195322036,
|
|
"num_tokens": 11593929.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.356800603866577,
|
|
"epoch": 5.090867787369378,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003029194415368448,
|
|
"loss": 4.6702,
|
|
"mean_token_accuracy": 0.23150020837783813,
|
|
"num_tokens": 11604891.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.365757989883423,
|
|
"epoch": 5.095411176737846,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003025687095702548,
|
|
"loss": 4.7149,
|
|
"mean_token_accuracy": 0.21905051469802855,
|
|
"num_tokens": 11616377.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.273924255371094,
|
|
"epoch": 5.099954566106315,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003022179095806146,
|
|
"loss": 4.6744,
|
|
"mean_token_accuracy": 0.22456111758947372,
|
|
"num_tokens": 11627079.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.307574605941772,
|
|
"epoch": 5.104497955474784,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000301867042433488,
|
|
"loss": 4.6499,
|
|
"mean_token_accuracy": 0.22698871046304703,
|
|
"num_tokens": 11638014.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.444634485244751,
|
|
"epoch": 5.109041344843253,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030151610899460505,
|
|
"loss": 4.7634,
|
|
"mean_token_accuracy": 0.23238055855035783,
|
|
"num_tokens": 11648759.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.357784223556519,
|
|
"epoch": 5.113584734211722,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030116511012985916,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.22860395461320876,
|
|
"num_tokens": 11659050.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.376965141296386,
|
|
"epoch": 5.118128123580191,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003008140467053048,
|
|
"loss": 4.7151,
|
|
"mean_token_accuracy": 0.22374295443296432,
|
|
"num_tokens": 11670131.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.27871036529541,
|
|
"epoch": 5.12267151294866,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003004629195871565,
|
|
"loss": 4.6064,
|
|
"mean_token_accuracy": 0.23196703791618348,
|
|
"num_tokens": 11680564.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.367365789413452,
|
|
"epoch": 5.127214902317129,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003001117296417852,
|
|
"loss": 4.7367,
|
|
"mean_token_accuracy": 0.2240822896361351,
|
|
"num_tokens": 11691910.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.299062728881836,
|
|
"epoch": 5.131758291685597,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002997604777357173,
|
|
"loss": 4.6601,
|
|
"mean_token_accuracy": 0.2282523438334465,
|
|
"num_tokens": 11702174.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.397566795349121,
|
|
"epoch": 5.136301681054066,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000299409164735632,
|
|
"loss": 4.788,
|
|
"mean_token_accuracy": 0.22190406620502473,
|
|
"num_tokens": 11712841.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.361143970489502,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002990577915083591,
|
|
"loss": 4.698,
|
|
"mean_token_accuracy": 0.22586688846349717,
|
|
"num_tokens": 11722781.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.29167423248291,
|
|
"epoch": 5.145388459791004,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00029870635892087716,
|
|
"loss": 4.5861,
|
|
"mean_token_accuracy": 0.2365378260612488,
|
|
"num_tokens": 11732774.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.287317371368408,
|
|
"epoch": 5.149931849159473,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002983548678403112,
|
|
"loss": 4.6645,
|
|
"mean_token_accuracy": 0.22749639600515364,
|
|
"num_tokens": 11743308.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.340081214904785,
|
|
"epoch": 5.154475238527942,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002980033191339305,
|
|
"loss": 4.7522,
|
|
"mean_token_accuracy": 0.2181773990392685,
|
|
"num_tokens": 11754572.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.3136749267578125,
|
|
"epoch": 5.159018627896411,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002976517136691465,
|
|
"loss": 4.6725,
|
|
"mean_token_accuracy": 0.22588830888271333,
|
|
"num_tokens": 11764406.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.292250394821167,
|
|
"epoch": 5.16356201726488,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029730005231351074,
|
|
"loss": 4.6492,
|
|
"mean_token_accuracy": 0.22577765136957167,
|
|
"num_tokens": 11774121.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.3038726329803465,
|
|
"epoch": 5.168105406633348,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002969483359347127,
|
|
"loss": 4.6929,
|
|
"mean_token_accuracy": 0.22391113936901091,
|
|
"num_tokens": 11784820.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.256404066085816,
|
|
"epoch": 5.172648796001817,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002965965654005776,
|
|
"loss": 4.6678,
|
|
"mean_token_accuracy": 0.23155077397823334,
|
|
"num_tokens": 11794891.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.335179042816162,
|
|
"epoch": 5.177192185370286,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002962447415790642,
|
|
"loss": 4.6316,
|
|
"mean_token_accuracy": 0.23173824250698088,
|
|
"num_tokens": 11804842.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.309024477005005,
|
|
"epoch": 5.181735574738755,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00029589286533826276,
|
|
"loss": 4.6955,
|
|
"mean_token_accuracy": 0.22791049927473067,
|
|
"num_tokens": 11815893.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.276846075057984,
|
|
"epoch": 5.186278964107224,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002955409375463931,
|
|
"loss": 4.6036,
|
|
"mean_token_accuracy": 0.23651532232761383,
|
|
"num_tokens": 11826332.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.361529588699341,
|
|
"epoch": 5.190822353475693,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002951889590718019,
|
|
"loss": 4.7275,
|
|
"mean_token_accuracy": 0.22440107464790343,
|
|
"num_tokens": 11836977.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.320426082611084,
|
|
"epoch": 5.195365742844162,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002948369307829612,
|
|
"loss": 4.6371,
|
|
"mean_token_accuracy": 0.23505959808826446,
|
|
"num_tokens": 11846483.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.319567537307739,
|
|
"epoch": 5.199909132212631,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002944848535484656,
|
|
"loss": 4.6865,
|
|
"mean_token_accuracy": 0.2253303810954094,
|
|
"num_tokens": 11857055.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.242885684967041,
|
|
"epoch": 5.2044525215811,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029413272823703085,
|
|
"loss": 4.5853,
|
|
"mean_token_accuracy": 0.2346850425004959,
|
|
"num_tokens": 11866929.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.317738580703735,
|
|
"epoch": 5.208995910949568,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002937805557174911,
|
|
"loss": 4.7083,
|
|
"mean_token_accuracy": 0.22270682603120803,
|
|
"num_tokens": 11875973.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.359472942352295,
|
|
"epoch": 5.213539300318037,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029342833685879713,
|
|
"loss": 4.7104,
|
|
"mean_token_accuracy": 0.23246724009513856,
|
|
"num_tokens": 11886273.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.4387678623199465,
|
|
"epoch": 5.218082689686506,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002930760725300137,
|
|
"loss": 4.7375,
|
|
"mean_token_accuracy": 0.22568463385105134,
|
|
"num_tokens": 11897462.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.338192272186279,
|
|
"epoch": 5.222626079054975,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00029272376360031833,
|
|
"loss": 4.6812,
|
|
"mean_token_accuracy": 0.22241345196962356,
|
|
"num_tokens": 11908116.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.352029466629029,
|
|
"epoch": 5.227169468423444,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029237141093899804,
|
|
"loss": 4.7469,
|
|
"mean_token_accuracy": 0.22050075829029084,
|
|
"num_tokens": 11918990.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.3737053871154785,
|
|
"epoch": 5.231712857791913,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029201901541544824,
|
|
"loss": 4.7315,
|
|
"mean_token_accuracy": 0.22929480224847792,
|
|
"num_tokens": 11929201.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.312944030761718,
|
|
"epoch": 5.236256247160382,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002916665778991696,
|
|
"loss": 4.7088,
|
|
"mean_token_accuracy": 0.21981508135795594,
|
|
"num_tokens": 11939002.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.318306541442871,
|
|
"epoch": 5.240799636528851,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002913140992597668,
|
|
"loss": 4.589,
|
|
"mean_token_accuracy": 0.23162836730480194,
|
|
"num_tokens": 11949255.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.311048889160157,
|
|
"epoch": 5.245343025897319,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002909615803669458,
|
|
"loss": 4.6415,
|
|
"mean_token_accuracy": 0.23259371072053908,
|
|
"num_tokens": 11959108.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.1976605415344235,
|
|
"epoch": 5.249886415265788,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002906090220905119,
|
|
"loss": 4.5772,
|
|
"mean_token_accuracy": 0.23512054830789567,
|
|
"num_tokens": 11969196.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.262940073013306,
|
|
"epoch": 5.254429804634257,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002902564253003677,
|
|
"loss": 4.6134,
|
|
"mean_token_accuracy": 0.23176871836185456,
|
|
"num_tokens": 11979101.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.315931797027588,
|
|
"epoch": 5.258973194002726,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002899037908665106,
|
|
"loss": 4.6701,
|
|
"mean_token_accuracy": 0.22663527578115464,
|
|
"num_tokens": 11988396.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.322834062576294,
|
|
"epoch": 5.263516583371195,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002895511196590312,
|
|
"loss": 4.664,
|
|
"mean_token_accuracy": 0.22784754782915115,
|
|
"num_tokens": 11998355.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.279678583145142,
|
|
"epoch": 5.268059972739664,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002891984125481104,
|
|
"loss": 4.6492,
|
|
"mean_token_accuracy": 0.23198784589767457,
|
|
"num_tokens": 12008938.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.3108151912689205,
|
|
"epoch": 5.272603362108133,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028884567040401814,
|
|
"loss": 4.7271,
|
|
"mean_token_accuracy": 0.22326084226369858,
|
|
"num_tokens": 12019875.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.365749454498291,
|
|
"epoch": 5.277146751476602,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002884928940971105,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.23109504878520964,
|
|
"num_tokens": 12029368.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.321197128295898,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000288140084497828,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.22481284886598588,
|
|
"num_tokens": 12040191.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.258969974517822,
|
|
"epoch": 5.286233530213539,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002877872424766932,
|
|
"loss": 4.6422,
|
|
"mean_token_accuracy": 0.2363392621278763,
|
|
"num_tokens": 12050087.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.247163772583008,
|
|
"epoch": 5.290776919582008,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002874343689043086,
|
|
"loss": 4.6537,
|
|
"mean_token_accuracy": 0.2363737091422081,
|
|
"num_tokens": 12061014.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.3400650978088375,
|
|
"epoch": 5.295320308950477,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000287081464651355,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.22778695225715637,
|
|
"num_tokens": 12071055.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.307133769989013,
|
|
"epoch": 5.299863698318946,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002867285305885882,
|
|
"loss": 4.6802,
|
|
"mean_token_accuracy": 0.22843130975961684,
|
|
"num_tokens": 12081323.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.251366901397705,
|
|
"epoch": 5.304407087687415,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028637556758683816,
|
|
"loss": 4.6364,
|
|
"mean_token_accuracy": 0.2301281988620758,
|
|
"num_tokens": 12091482.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.207917976379394,
|
|
"epoch": 5.308950477055884,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002860225765170058,
|
|
"loss": 4.5679,
|
|
"mean_token_accuracy": 0.23898424208164215,
|
|
"num_tokens": 12101960.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.301973295211792,
|
|
"epoch": 5.313493866424353,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00028566955825006174,
|
|
"loss": 4.6805,
|
|
"mean_token_accuracy": 0.22320136725902556,
|
|
"num_tokens": 12111654.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.393432521820069,
|
|
"epoch": 5.318037255792821,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028531651365704325,
|
|
"loss": 4.7615,
|
|
"mean_token_accuracy": 0.22631869614124298,
|
|
"num_tokens": 12123096.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.297732877731323,
|
|
"epoch": 5.32258064516129,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028496344360905287,
|
|
"loss": 4.6576,
|
|
"mean_token_accuracy": 0.226182821393013,
|
|
"num_tokens": 12133064.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.283245801925659,
|
|
"epoch": 5.327124034529759,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028461034897725587,
|
|
"loss": 4.6885,
|
|
"mean_token_accuracy": 0.2285231575369835,
|
|
"num_tokens": 12144386.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.312462568283081,
|
|
"epoch": 5.331667423898228,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028425723063287803,
|
|
"loss": 4.7184,
|
|
"mean_token_accuracy": 0.22606474608182908,
|
|
"num_tokens": 12156111.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.375386571884155,
|
|
"epoch": 5.336210813266697,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028390408944720385,
|
|
"loss": 4.7632,
|
|
"mean_token_accuracy": 0.2136458560824394,
|
|
"num_tokens": 12166462.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.302098655700684,
|
|
"epoch": 5.340754202635166,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002835509262915741,
|
|
"loss": 4.7816,
|
|
"mean_token_accuracy": 0.2143580973148346,
|
|
"num_tokens": 12178635.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.252813529968262,
|
|
"epoch": 5.345297592003635,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028319774203738375,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.2303081586956978,
|
|
"num_tokens": 12188880.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.2756890773773195,
|
|
"epoch": 5.349840981372104,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002828445375560798,
|
|
"loss": 4.6478,
|
|
"mean_token_accuracy": 0.22301267981529235,
|
|
"num_tokens": 12199733.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.292732906341553,
|
|
"epoch": 5.354384370740572,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00028249131371915927,
|
|
"loss": 4.6897,
|
|
"mean_token_accuracy": 0.22554381787776948,
|
|
"num_tokens": 12209925.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.430313396453857,
|
|
"epoch": 5.358927760109041,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002821380713981668,
|
|
"loss": 4.7481,
|
|
"mean_token_accuracy": 0.22702892869710922,
|
|
"num_tokens": 12219870.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.2865235805511475,
|
|
"epoch": 5.36347114947751,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028178481146469273,
|
|
"loss": 4.6742,
|
|
"mean_token_accuracy": 0.23230669647455215,
|
|
"num_tokens": 12230295.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.3186266899108885,
|
|
"epoch": 5.368014538845979,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002814315347903709,
|
|
"loss": 4.6735,
|
|
"mean_token_accuracy": 0.23260893076658248,
|
|
"num_tokens": 12239700.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.321635913848877,
|
|
"epoch": 5.372557928214448,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002810782422468763,
|
|
"loss": 4.7369,
|
|
"mean_token_accuracy": 0.22189791947603227,
|
|
"num_tokens": 12249529.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.312885522842407,
|
|
"epoch": 5.377101317582917,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00028072493470592317,
|
|
"loss": 4.6328,
|
|
"mean_token_accuracy": 0.23139261901378633,
|
|
"num_tokens": 12259774.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.353052854537964,
|
|
"epoch": 5.381644706951386,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028037161303926265,
|
|
"loss": 4.6747,
|
|
"mean_token_accuracy": 0.22675450444221495,
|
|
"num_tokens": 12270484.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.276238012313843,
|
|
"epoch": 5.386188096319855,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000280018278118681,
|
|
"loss": 4.6597,
|
|
"mean_token_accuracy": 0.2269777923822403,
|
|
"num_tokens": 12281195.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.263835000991821,
|
|
"epoch": 5.390731485688324,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000279664930815997,
|
|
"loss": 4.6535,
|
|
"mean_token_accuracy": 0.22591907232999803,
|
|
"num_tokens": 12292052.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.284007978439331,
|
|
"epoch": 5.395274875056792,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002793115720030599,
|
|
"loss": 4.725,
|
|
"mean_token_accuracy": 0.22450850009918213,
|
|
"num_tokens": 12302562.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.330573606491089,
|
|
"epoch": 5.399818264425261,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002789582025517474,
|
|
"loss": 4.7106,
|
|
"mean_token_accuracy": 0.22729553282260895,
|
|
"num_tokens": 12313875.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.37083854675293,
|
|
"epoch": 5.40436165379373,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002786048233339637,
|
|
"loss": 4.7089,
|
|
"mean_token_accuracy": 0.22028498202562333,
|
|
"num_tokens": 12324500.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.307926607131958,
|
|
"epoch": 5.408905043162199,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027825143522163675,
|
|
"loss": 4.6937,
|
|
"mean_token_accuracy": 0.22768828123807908,
|
|
"num_tokens": 12335474.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.301209402084351,
|
|
"epoch": 5.413448432530668,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002778980390867165,
|
|
"loss": 4.6657,
|
|
"mean_token_accuracy": 0.2288498654961586,
|
|
"num_tokens": 12346174.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.320987176895142,
|
|
"epoch": 5.417991821899137,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027754463580117306,
|
|
"loss": 4.6721,
|
|
"mean_token_accuracy": 0.23781586140394212,
|
|
"num_tokens": 12356108.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.321499061584473,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027719122623699353,
|
|
"loss": 4.7181,
|
|
"mean_token_accuracy": 0.23379721343517304,
|
|
"num_tokens": 12366699.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.407852840423584,
|
|
"epoch": 5.427078600636074,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002768378112661813,
|
|
"loss": 4.7222,
|
|
"mean_token_accuracy": 0.22026822865009307,
|
|
"num_tokens": 12377225.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.375286626815796,
|
|
"epoch": 5.431621990004543,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002764843917607523,
|
|
"loss": 4.7334,
|
|
"mean_token_accuracy": 0.22914440035820008,
|
|
"num_tokens": 12388936.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.269746780395508,
|
|
"epoch": 5.436165379373012,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002761309685927343,
|
|
"loss": 4.6685,
|
|
"mean_token_accuracy": 0.22550414353609086,
|
|
"num_tokens": 12399130.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.316030263900757,
|
|
"epoch": 5.440708768741481,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002757775426341636,
|
|
"loss": 4.6609,
|
|
"mean_token_accuracy": 0.23977339565753936,
|
|
"num_tokens": 12410181.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.367448377609253,
|
|
"epoch": 5.44525215810995,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027542411475708385,
|
|
"loss": 4.7704,
|
|
"mean_token_accuracy": 0.2157247930765152,
|
|
"num_tokens": 12421173.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.39612193107605,
|
|
"epoch": 5.449795547478419,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00027507068583354304,
|
|
"loss": 4.7659,
|
|
"mean_token_accuracy": 0.21552858054637908,
|
|
"num_tokens": 12433195.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.449795547478419,
|
|
"eval_entropy": 5.199006334570951,
|
|
"eval_loss": 5.649643898010254,
|
|
"eval_mean_token_accuracy": 0.17373872605520627,
|
|
"eval_num_tokens": 12433195.0,
|
|
"eval_runtime": 2.0001,
|
|
"eval_samples_per_second": 1717.453,
|
|
"eval_steps_per_second": 214.994,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.309214639663696,
|
|
"epoch": 5.454338936846888,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027471725673559205,
|
|
"loss": 4.6725,
|
|
"mean_token_accuracy": 0.22684060484170915,
|
|
"num_tokens": 12442944.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.260644578933716,
|
|
"epoch": 5.458882326215357,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027436382833528184,
|
|
"loss": 4.6499,
|
|
"mean_token_accuracy": 0.23391480147838592,
|
|
"num_tokens": 12453133.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.278741264343262,
|
|
"epoch": 5.463425715583826,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002740104015046621,
|
|
"loss": 4.7639,
|
|
"mean_token_accuracy": 0.2148856669664383,
|
|
"num_tokens": 12464606.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.277816963195801,
|
|
"epoch": 5.467969104952294,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00027365697711577815,
|
|
"loss": 4.6529,
|
|
"mean_token_accuracy": 0.2247397243976593,
|
|
"num_tokens": 12474490.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.3290900707244875,
|
|
"epoch": 5.472512494320763,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002733035560406697,
|
|
"loss": 4.6597,
|
|
"mean_token_accuracy": 0.23245542496442795,
|
|
"num_tokens": 12485158.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.25628342628479,
|
|
"epoch": 5.477055883689232,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027295013915136805,
|
|
"loss": 4.7111,
|
|
"mean_token_accuracy": 0.22251079678535463,
|
|
"num_tokens": 12496265.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.253548336029053,
|
|
"epoch": 5.481599273057701,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027259672731989436,
|
|
"loss": 4.576,
|
|
"mean_token_accuracy": 0.23487671017646788,
|
|
"num_tokens": 12506719.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.325517177581787,
|
|
"epoch": 5.48614266242617,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000272243321418257,
|
|
"loss": 4.6822,
|
|
"mean_token_accuracy": 0.22805582880973815,
|
|
"num_tokens": 12515990.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.287591934204102,
|
|
"epoch": 5.490686051794639,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002718899223184501,
|
|
"loss": 4.6974,
|
|
"mean_token_accuracy": 0.2257079154253006,
|
|
"num_tokens": 12526557.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.259378671646118,
|
|
"epoch": 5.495229441163108,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027153653089245067,
|
|
"loss": 4.6323,
|
|
"mean_token_accuracy": 0.23873628973960875,
|
|
"num_tokens": 12536657.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.28349814414978,
|
|
"epoch": 5.499772830531577,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000271183148012217,
|
|
"loss": 4.6482,
|
|
"mean_token_accuracy": 0.22884975373744965,
|
|
"num_tokens": 12546958.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.260309028625488,
|
|
"epoch": 5.504316219900045,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002708297745496861,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.2259410724043846,
|
|
"num_tokens": 12556978.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.281688261032104,
|
|
"epoch": 5.508859609268514,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027047641137677213,
|
|
"loss": 4.6318,
|
|
"mean_token_accuracy": 0.23104216903448105,
|
|
"num_tokens": 12566954.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.331643009185791,
|
|
"epoch": 5.513402998636983,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002701230593653632,
|
|
"loss": 4.6781,
|
|
"mean_token_accuracy": 0.22869530767202378,
|
|
"num_tokens": 12577129.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.297736644744873,
|
|
"epoch": 5.517946388005452,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002697697193873207,
|
|
"loss": 4.6662,
|
|
"mean_token_accuracy": 0.2173148661851883,
|
|
"num_tokens": 12586802.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.303196096420288,
|
|
"epoch": 5.522489777373921,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00026941639231447547,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.22031511962413788,
|
|
"num_tokens": 12597732.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.305408048629761,
|
|
"epoch": 5.52703316674239,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002690630790186273,
|
|
"loss": 4.6138,
|
|
"mean_token_accuracy": 0.23197653591632844,
|
|
"num_tokens": 12607185.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.320194101333618,
|
|
"epoch": 5.531576556110859,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00026870978037154137,
|
|
"loss": 4.711,
|
|
"mean_token_accuracy": 0.22427654415369033,
|
|
"num_tokens": 12616821.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.306847858428955,
|
|
"epoch": 5.536119945479328,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002683564972449472,
|
|
"loss": 4.7425,
|
|
"mean_token_accuracy": 0.22166457772254944,
|
|
"num_tokens": 12626698.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.292235708236694,
|
|
"epoch": 5.540663334847796,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002680032305105355,
|
|
"loss": 4.6547,
|
|
"mean_token_accuracy": 0.22907288521528243,
|
|
"num_tokens": 12636588.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.363357591629028,
|
|
"epoch": 5.545206724216265,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002676499810399572,
|
|
"loss": 4.7217,
|
|
"mean_token_accuracy": 0.22510518580675126,
|
|
"num_tokens": 12647196.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.300512742996216,
|
|
"epoch": 5.549750113584734,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026729674970481985,
|
|
"loss": 4.7082,
|
|
"mean_token_accuracy": 0.22529883235692977,
|
|
"num_tokens": 12658631.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.298285913467407,
|
|
"epoch": 5.554293502953203,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026694353737668705,
|
|
"loss": 4.7264,
|
|
"mean_token_accuracy": 0.22505012154579163,
|
|
"num_tokens": 12668905.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.239463996887207,
|
|
"epoch": 5.558836892321672,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026659034492707473,
|
|
"loss": 4.6518,
|
|
"mean_token_accuracy": 0.23246145993471146,
|
|
"num_tokens": 12678270.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.296948671340942,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002662371732274504,
|
|
"loss": 4.6649,
|
|
"mean_token_accuracy": 0.22968941330909728,
|
|
"num_tokens": 12688248.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.332198572158814,
|
|
"epoch": 5.56792367105861,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026588402314923,
|
|
"loss": 4.7644,
|
|
"mean_token_accuracy": 0.22141656130552292,
|
|
"num_tokens": 12698418.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.217432737350464,
|
|
"epoch": 5.572467060427079,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002655308955637763,
|
|
"loss": 4.5965,
|
|
"mean_token_accuracy": 0.23892516344785691,
|
|
"num_tokens": 12708578.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.283588457107544,
|
|
"epoch": 5.577010449795548,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002651777913423965,
|
|
"loss": 4.7139,
|
|
"mean_token_accuracy": 0.21924354434013366,
|
|
"num_tokens": 12718899.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.281859445571899,
|
|
"epoch": 5.581553839164016,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002648247113563402,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.2236953318119049,
|
|
"num_tokens": 12728342.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.393999671936035,
|
|
"epoch": 5.586097228532485,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026447165647679703,
|
|
"loss": 4.7089,
|
|
"mean_token_accuracy": 0.23541267365217208,
|
|
"num_tokens": 12739337.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.281257247924804,
|
|
"epoch": 5.590640617900954,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000264118627574895,
|
|
"loss": 4.6948,
|
|
"mean_token_accuracy": 0.22588736861944197,
|
|
"num_tokens": 12749650.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.2791587829589846,
|
|
"epoch": 5.595184007269423,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026376562552169776,
|
|
"loss": 4.7473,
|
|
"mean_token_accuracy": 0.21881672292947768,
|
|
"num_tokens": 12760573.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.426256704330444,
|
|
"epoch": 5.599727396637892,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026341265118820284,
|
|
"loss": 4.7687,
|
|
"mean_token_accuracy": 0.21999825537204742,
|
|
"num_tokens": 12771070.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.373771476745605,
|
|
"epoch": 5.604270786006361,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026305970544533903,
|
|
"loss": 4.6704,
|
|
"mean_token_accuracy": 0.22763433456420898,
|
|
"num_tokens": 12781896.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.277415418624878,
|
|
"epoch": 5.60881417537483,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026270678916396527,
|
|
"loss": 4.7198,
|
|
"mean_token_accuracy": 0.21930332630872726,
|
|
"num_tokens": 12792675.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.266959619522095,
|
|
"epoch": 5.613357564743298,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026235390321486705,
|
|
"loss": 4.6389,
|
|
"mean_token_accuracy": 0.23069387227296828,
|
|
"num_tokens": 12802371.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.328389358520508,
|
|
"epoch": 5.617900954111767,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026200104846875574,
|
|
"loss": 4.6942,
|
|
"mean_token_accuracy": 0.2304431304335594,
|
|
"num_tokens": 12813281.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.283749914169311,
|
|
"epoch": 5.622444343480236,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000261648225796265,
|
|
"loss": 4.7334,
|
|
"mean_token_accuracy": 0.21898353695869446,
|
|
"num_tokens": 12823859.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.294131851196289,
|
|
"epoch": 5.626987732848705,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002612954360679499,
|
|
"loss": 4.6135,
|
|
"mean_token_accuracy": 0.2349908620119095,
|
|
"num_tokens": 12833288.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.302528953552246,
|
|
"epoch": 5.631531122217174,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000260942680154284,
|
|
"loss": 4.6679,
|
|
"mean_token_accuracy": 0.22561876624822616,
|
|
"num_tokens": 12843698.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.267947387695313,
|
|
"epoch": 5.636074511585643,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002605899589256574,
|
|
"loss": 4.5876,
|
|
"mean_token_accuracy": 0.24070773422718048,
|
|
"num_tokens": 12853956.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.323961162567139,
|
|
"epoch": 5.640617900954112,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026023727325237463,
|
|
"loss": 4.7106,
|
|
"mean_token_accuracy": 0.22130987644195557,
|
|
"num_tokens": 12864680.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.3381139755249025,
|
|
"epoch": 5.645161290322581,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002598846240046527,
|
|
"loss": 4.6289,
|
|
"mean_token_accuracy": 0.23060933351516724,
|
|
"num_tokens": 12873790.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.352335214614868,
|
|
"epoch": 5.64970467969105,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002595320120526183,
|
|
"loss": 4.6927,
|
|
"mean_token_accuracy": 0.22879611402750016,
|
|
"num_tokens": 12884559.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.277868175506592,
|
|
"epoch": 5.654248069059518,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025917943826630666,
|
|
"loss": 4.7238,
|
|
"mean_token_accuracy": 0.2270288199186325,
|
|
"num_tokens": 12896176.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.308832788467408,
|
|
"epoch": 5.658791458427987,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002588269035156584,
|
|
"loss": 4.6684,
|
|
"mean_token_accuracy": 0.21959326565265655,
|
|
"num_tokens": 12906381.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.322448253631592,
|
|
"epoch": 5.663334847796456,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025847440867051796,
|
|
"loss": 4.6437,
|
|
"mean_token_accuracy": 0.22649855315685272,
|
|
"num_tokens": 12915738.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.3148785591125485,
|
|
"epoch": 5.667878237164925,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002581219546006314,
|
|
"loss": 4.6672,
|
|
"mean_token_accuracy": 0.2314092919230461,
|
|
"num_tokens": 12926059.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.2999344825744625,
|
|
"epoch": 5.672421626533394,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002577695421756441,
|
|
"loss": 4.7027,
|
|
"mean_token_accuracy": 0.22299634963274,
|
|
"num_tokens": 12936730.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.284259939193726,
|
|
"epoch": 5.676965015901863,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002574171722650986,
|
|
"loss": 4.7024,
|
|
"mean_token_accuracy": 0.22333675920963286,
|
|
"num_tokens": 12945687.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.359689855575562,
|
|
"epoch": 5.681508405270332,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00025706484573843273,
|
|
"loss": 4.6958,
|
|
"mean_token_accuracy": 0.23166366368532182,
|
|
"num_tokens": 12955087.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.298504829406738,
|
|
"epoch": 5.6860517946388,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000256712563464977,
|
|
"loss": 4.674,
|
|
"mean_token_accuracy": 0.22083057165145875,
|
|
"num_tokens": 12965497.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.361152458190918,
|
|
"epoch": 5.690595184007269,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002563603263139532,
|
|
"loss": 4.8398,
|
|
"mean_token_accuracy": 0.21261918246746064,
|
|
"num_tokens": 12976520.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.290573215484619,
|
|
"epoch": 5.695138573375738,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002560081351544711,
|
|
"loss": 4.6854,
|
|
"mean_token_accuracy": 0.22768215984106063,
|
|
"num_tokens": 12987271.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.317127418518067,
|
|
"epoch": 5.699681962744207,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025565599085552776,
|
|
"loss": 4.7066,
|
|
"mean_token_accuracy": 0.22129882574081422,
|
|
"num_tokens": 12998170.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.291876792907715,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00025530389428600394,
|
|
"loss": 4.6886,
|
|
"mean_token_accuracy": 0.2244232788681984,
|
|
"num_tokens": 13008059.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.293149089813232,
|
|
"epoch": 5.708768741481145,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002549518463146631,
|
|
"loss": 4.6495,
|
|
"mean_token_accuracy": 0.22723347395658494,
|
|
"num_tokens": 13017904.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.373154830932617,
|
|
"epoch": 5.713312130849614,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002545998478101485,
|
|
"loss": 4.71,
|
|
"mean_token_accuracy": 0.22887400388717652,
|
|
"num_tokens": 13028399.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.337195920944214,
|
|
"epoch": 5.717855520218083,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00025424789964098154,
|
|
"loss": 4.7007,
|
|
"mean_token_accuracy": 0.22320223599672318,
|
|
"num_tokens": 13038420.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.3334105014801025,
|
|
"epoch": 5.722398909586552,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002538960026755593,
|
|
"loss": 4.7069,
|
|
"mean_token_accuracy": 0.2268875151872635,
|
|
"num_tokens": 13048514.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.322353315353394,
|
|
"epoch": 5.72694229895502,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025354415778215255,
|
|
"loss": 4.7622,
|
|
"mean_token_accuracy": 0.22440093755722046,
|
|
"num_tokens": 13059728.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.363490295410156,
|
|
"epoch": 5.731485688323489,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025319236582890364,
|
|
"loss": 4.7663,
|
|
"mean_token_accuracy": 0.21983303427696227,
|
|
"num_tokens": 13069535.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.28021388053894,
|
|
"epoch": 5.736029077691958,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00025284062768382424,
|
|
"loss": 4.638,
|
|
"mean_token_accuracy": 0.22660373747348786,
|
|
"num_tokens": 13079012.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.338456010818481,
|
|
"epoch": 5.740572467060427,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002524889442147932,
|
|
"loss": 4.6756,
|
|
"mean_token_accuracy": 0.23634835779666902,
|
|
"num_tokens": 13089067.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.316509103775024,
|
|
"epoch": 5.745115856428896,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025213731628955454,
|
|
"loss": 4.6905,
|
|
"mean_token_accuracy": 0.22679051607847214,
|
|
"num_tokens": 13100576.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.346896076202393,
|
|
"epoch": 5.749659245797365,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025178574477571525,
|
|
"loss": 4.8433,
|
|
"mean_token_accuracy": 0.21215644776821135,
|
|
"num_tokens": 13111926.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.265066289901734,
|
|
"epoch": 5.754202635165834,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025143423054074305,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.22555685341358184,
|
|
"num_tokens": 13122362.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.28929967880249,
|
|
"epoch": 5.758746024534303,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002510827744519644,
|
|
"loss": 4.6593,
|
|
"mean_token_accuracy": 0.22941243797540664,
|
|
"num_tokens": 13133372.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.290349340438842,
|
|
"epoch": 5.763289413902772,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025073137737656226,
|
|
"loss": 4.656,
|
|
"mean_token_accuracy": 0.22604783475399018,
|
|
"num_tokens": 13143440.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.337073183059692,
|
|
"epoch": 5.76783280327124,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002503800401815739,
|
|
"loss": 4.7095,
|
|
"mean_token_accuracy": 0.22297210693359376,
|
|
"num_tokens": 13153564.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.328155279159546,
|
|
"epoch": 5.772376192639709,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025002876373388913,
|
|
"loss": 4.7762,
|
|
"mean_token_accuracy": 0.21255046129226685,
|
|
"num_tokens": 13163679.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.3327796936035154,
|
|
"epoch": 5.776919582008178,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024967754890024737,
|
|
"loss": 4.7018,
|
|
"mean_token_accuracy": 0.22889443337917328,
|
|
"num_tokens": 13173930.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.353774309158325,
|
|
"epoch": 5.781462971376647,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002493263965472366,
|
|
"loss": 4.6826,
|
|
"mean_token_accuracy": 0.23404109925031663,
|
|
"num_tokens": 13184805.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.287617158889771,
|
|
"epoch": 5.786006360745116,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002489753075412901,
|
|
"loss": 4.7241,
|
|
"mean_token_accuracy": 0.22358319610357286,
|
|
"num_tokens": 13196229.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.287916994094848,
|
|
"epoch": 5.790549750113585,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024862428274868533,
|
|
"loss": 4.6122,
|
|
"mean_token_accuracy": 0.23477774262428283,
|
|
"num_tokens": 13205906.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.28186411857605,
|
|
"epoch": 5.795093139482054,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002482733230355408,
|
|
"loss": 4.6384,
|
|
"mean_token_accuracy": 0.22464174777269363,
|
|
"num_tokens": 13215076.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.375334167480469,
|
|
"epoch": 5.799636528850522,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024792242926781495,
|
|
"loss": 4.7723,
|
|
"mean_token_accuracy": 0.21902428269386293,
|
|
"num_tokens": 13225761.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.370274877548217,
|
|
"epoch": 5.804179918218991,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00024757160231130313,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.22328272461891174,
|
|
"num_tokens": 13236342.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.261192560195923,
|
|
"epoch": 5.80872330758746,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024722084303163607,
|
|
"loss": 4.6599,
|
|
"mean_token_accuracy": 0.22834971696138381,
|
|
"num_tokens": 13244989.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.304846954345703,
|
|
"epoch": 5.813266696955929,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002468701522942774,
|
|
"loss": 4.6854,
|
|
"mean_token_accuracy": 0.229074227809906,
|
|
"num_tokens": 13255495.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.387843084335327,
|
|
"epoch": 5.817810086324398,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024651953096452165,
|
|
"loss": 4.7356,
|
|
"mean_token_accuracy": 0.219369338452816,
|
|
"num_tokens": 13266189.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.314527082443237,
|
|
"epoch": 5.822353475692867,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024616897990749196,
|
|
"loss": 4.6946,
|
|
"mean_token_accuracy": 0.22669169008731843,
|
|
"num_tokens": 13276194.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.275722312927246,
|
|
"epoch": 5.826896865061336,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002458184999881385,
|
|
"loss": 4.7228,
|
|
"mean_token_accuracy": 0.216484834253788,
|
|
"num_tokens": 13286188.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.322114706039429,
|
|
"epoch": 5.831440254429805,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024546809207123534,
|
|
"loss": 4.6728,
|
|
"mean_token_accuracy": 0.22921556532382964,
|
|
"num_tokens": 13295926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.3440453052520756,
|
|
"epoch": 5.835983643798274,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002451177570213794,
|
|
"loss": 4.7552,
|
|
"mean_token_accuracy": 0.22437551766633987,
|
|
"num_tokens": 13307373.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.288799905776978,
|
|
"epoch": 5.840527033166742,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024476749570298746,
|
|
"loss": 4.6894,
|
|
"mean_token_accuracy": 0.22196318358182907,
|
|
"num_tokens": 13316984.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.280353450775147,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024441730898029456,
|
|
"loss": 4.6517,
|
|
"mean_token_accuracy": 0.2344815507531166,
|
|
"num_tokens": 13327188.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.4337043285369875,
|
|
"epoch": 5.84961381190368,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024406719771735165,
|
|
"loss": 4.7638,
|
|
"mean_token_accuracy": 0.22509843856096268,
|
|
"num_tokens": 13338108.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.2941734313964846,
|
|
"epoch": 5.854157201272149,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002437171627780234,
|
|
"loss": 4.654,
|
|
"mean_token_accuracy": 0.2326254740357399,
|
|
"num_tokens": 13347898.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.324732398986816,
|
|
"epoch": 5.858700590640618,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024336720502598624,
|
|
"loss": 4.725,
|
|
"mean_token_accuracy": 0.22677038013935089,
|
|
"num_tokens": 13358649.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.279881334304809,
|
|
"epoch": 5.863243980009087,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002430173253247262,
|
|
"loss": 4.7095,
|
|
"mean_token_accuracy": 0.22484695613384248,
|
|
"num_tokens": 13368447.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.283170461654663,
|
|
"epoch": 5.867787369377556,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024266752453753655,
|
|
"loss": 4.6851,
|
|
"mean_token_accuracy": 0.22113156914711,
|
|
"num_tokens": 13377809.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.35424656867981,
|
|
"epoch": 5.872330758746024,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024231780352751615,
|
|
"loss": 4.801,
|
|
"mean_token_accuracy": 0.21904565542936325,
|
|
"num_tokens": 13389352.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.3725049018859865,
|
|
"epoch": 5.876874148114493,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024196816315756665,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.2288314074277878,
|
|
"num_tokens": 13399929.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.336102628707886,
|
|
"epoch": 5.881417537482962,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002416186042903909,
|
|
"loss": 4.6018,
|
|
"mean_token_accuracy": 0.23243868947029114,
|
|
"num_tokens": 13410096.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.31442232131958,
|
|
"epoch": 5.885960926851431,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024126912778849074,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.22108559161424637,
|
|
"num_tokens": 13420384.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.335084629058838,
|
|
"epoch": 5.8905043162199,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024091973451416472,
|
|
"loss": 4.6906,
|
|
"mean_token_accuracy": 0.22831213921308519,
|
|
"num_tokens": 13430887.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.3248645782470705,
|
|
"epoch": 5.895047705588369,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024057042532950597,
|
|
"loss": 4.7008,
|
|
"mean_token_accuracy": 0.22710191309452057,
|
|
"num_tokens": 13440914.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.289365148544311,
|
|
"epoch": 5.899591094956838,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002402212010964001,
|
|
"loss": 4.755,
|
|
"mean_token_accuracy": 0.21540879756212233,
|
|
"num_tokens": 13452476.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.299159669876099,
|
|
"epoch": 5.904134484325307,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023987206267652334,
|
|
"loss": 4.6662,
|
|
"mean_token_accuracy": 0.22873913794755935,
|
|
"num_tokens": 13463779.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.904134484325307,
|
|
"eval_entropy": 5.17384759991668,
|
|
"eval_loss": 5.627315521240234,
|
|
"eval_mean_token_accuracy": 0.17661627174116845,
|
|
"eval_num_tokens": 13463779.0,
|
|
"eval_runtime": 2.2102,
|
|
"eval_samples_per_second": 1554.159,
|
|
"eval_steps_per_second": 194.553,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.342879104614258,
|
|
"epoch": 5.908677873693776,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023952301093133993,
|
|
"loss": 4.7751,
|
|
"mean_token_accuracy": 0.21699213981628418,
|
|
"num_tokens": 13475467.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.289231300354004,
|
|
"epoch": 5.913221263062244,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023917404672210046,
|
|
"loss": 4.6583,
|
|
"mean_token_accuracy": 0.22375408113002776,
|
|
"num_tokens": 13485484.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.318223094940185,
|
|
"epoch": 5.917764652430713,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002388251709098392,
|
|
"loss": 4.6633,
|
|
"mean_token_accuracy": 0.22801149785518646,
|
|
"num_tokens": 13494941.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.369752073287964,
|
|
"epoch": 5.922308041799182,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023847638435537283,
|
|
"loss": 4.7234,
|
|
"mean_token_accuracy": 0.22587551474571227,
|
|
"num_tokens": 13504762.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.364923143386841,
|
|
"epoch": 5.926851431167651,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023812768791929724,
|
|
"loss": 4.6867,
|
|
"mean_token_accuracy": 0.22754170894622802,
|
|
"num_tokens": 13515436.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.299966478347779,
|
|
"epoch": 5.93139482053612,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023777908246198643,
|
|
"loss": 4.6991,
|
|
"mean_token_accuracy": 0.23024591505527497,
|
|
"num_tokens": 13525211.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.298692035675049,
|
|
"epoch": 5.935938209904589,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023743056884358954,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.22784270942211152,
|
|
"num_tokens": 13534954.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.293479776382446,
|
|
"epoch": 5.940481599273058,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023708214792402942,
|
|
"loss": 4.7171,
|
|
"mean_token_accuracy": 0.23137304782867432,
|
|
"num_tokens": 13545459.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.256713008880615,
|
|
"epoch": 5.945024988641527,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023673382056299997,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.2237893372774124,
|
|
"num_tokens": 13554541.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.3175492763519285,
|
|
"epoch": 5.949568378009996,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023638558761996436,
|
|
"loss": 4.6912,
|
|
"mean_token_accuracy": 0.22339745461940766,
|
|
"num_tokens": 13564810.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.322920846939087,
|
|
"epoch": 5.954111767378464,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023603744995415277,
|
|
"loss": 4.6669,
|
|
"mean_token_accuracy": 0.22725937515497208,
|
|
"num_tokens": 13574756.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.398230075836182,
|
|
"epoch": 5.958655156746933,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023568940842456026,
|
|
"loss": 4.7394,
|
|
"mean_token_accuracy": 0.22406996935606002,
|
|
"num_tokens": 13584346.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.369215726852417,
|
|
"epoch": 5.963198546115402,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002353414638899447,
|
|
"loss": 4.7226,
|
|
"mean_token_accuracy": 0.22580974102020263,
|
|
"num_tokens": 13595713.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.168907928466797,
|
|
"epoch": 5.967741935483871,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023499361720882468,
|
|
"loss": 4.5903,
|
|
"mean_token_accuracy": 0.23416789025068283,
|
|
"num_tokens": 13605065.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.30269455909729,
|
|
"epoch": 5.97228532485234,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023464586923947718,
|
|
"loss": 4.755,
|
|
"mean_token_accuracy": 0.22251630276441575,
|
|
"num_tokens": 13615028.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.292668437957763,
|
|
"epoch": 5.976828714220809,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023429822083993596,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.22357941418886185,
|
|
"num_tokens": 13625752.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.367790985107422,
|
|
"epoch": 5.981372103589278,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023395067286798872,
|
|
"loss": 4.6507,
|
|
"mean_token_accuracy": 0.23149229884147643,
|
|
"num_tokens": 13635094.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.296887350082398,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023360322618117563,
|
|
"loss": 4.7023,
|
|
"mean_token_accuracy": 0.23041756451129913,
|
|
"num_tokens": 13645007.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.246177625656128,
|
|
"epoch": 5.990458882326215,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023325588163678674,
|
|
"loss": 4.5978,
|
|
"mean_token_accuracy": 0.23363692164421082,
|
|
"num_tokens": 13655101.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.262781715393066,
|
|
"epoch": 5.995002271694684,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023290864009186036,
|
|
"loss": 4.6599,
|
|
"mean_token_accuracy": 0.2285596415400505,
|
|
"num_tokens": 13665289.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.35294771194458,
|
|
"epoch": 5.999545661063153,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023256150240318025,
|
|
"loss": 4.737,
|
|
"mean_token_accuracy": 0.22992412149906158,
|
|
"num_tokens": 13675384.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.390623993343777,
|
|
"epoch": 6.003634711494775,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023221446942727438,
|
|
"loss": 4.6653,
|
|
"mean_token_accuracy": 0.23654643363422817,
|
|
"num_tokens": 13683918.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.310308837890625,
|
|
"epoch": 6.008178100863244,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002318675420204119,
|
|
"loss": 4.5749,
|
|
"mean_token_accuracy": 0.23771664649248123,
|
|
"num_tokens": 13693999.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.338145065307617,
|
|
"epoch": 6.012721490231713,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023152072103860188,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.23043870627880098,
|
|
"num_tokens": 13705465.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.321788358688354,
|
|
"epoch": 6.017264879600182,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023117400733759053,
|
|
"loss": 4.5091,
|
|
"mean_token_accuracy": 0.24304586797952651,
|
|
"num_tokens": 13715858.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.354307174682617,
|
|
"epoch": 6.021808268968651,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002308274017728595,
|
|
"loss": 4.5456,
|
|
"mean_token_accuracy": 0.24542825073003768,
|
|
"num_tokens": 13725639.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.208072996139526,
|
|
"epoch": 6.026351658337119,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002304809051996235,
|
|
"loss": 4.4568,
|
|
"mean_token_accuracy": 0.24555897265672683,
|
|
"num_tokens": 13736015.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.1614409446716305,
|
|
"epoch": 6.030895047705588,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002301345184728284,
|
|
"loss": 4.4142,
|
|
"mean_token_accuracy": 0.24534475207328796,
|
|
"num_tokens": 13745433.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.311540222167968,
|
|
"epoch": 6.035438437074057,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00022978824244714903,
|
|
"loss": 4.5187,
|
|
"mean_token_accuracy": 0.24828406125307084,
|
|
"num_tokens": 13754970.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.32213830947876,
|
|
"epoch": 6.039981826442526,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022944207797698707,
|
|
"loss": 4.6201,
|
|
"mean_token_accuracy": 0.23647294640541078,
|
|
"num_tokens": 13766356.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.301358222961426,
|
|
"epoch": 6.044525215810995,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002290960259164689,
|
|
"loss": 4.5666,
|
|
"mean_token_accuracy": 0.2411123439669609,
|
|
"num_tokens": 13776503.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.339816331863403,
|
|
"epoch": 6.049068605179464,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022875008711944355,
|
|
"loss": 4.602,
|
|
"mean_token_accuracy": 0.23380455672740935,
|
|
"num_tokens": 13787918.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.204669713973999,
|
|
"epoch": 6.053611994547933,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022840426243948077,
|
|
"loss": 4.5412,
|
|
"mean_token_accuracy": 0.24024882912635803,
|
|
"num_tokens": 13798305.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.296932172775269,
|
|
"epoch": 6.058155383916402,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022805855272986842,
|
|
"loss": 4.5503,
|
|
"mean_token_accuracy": 0.23801912367343903,
|
|
"num_tokens": 13808867.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.264242839813233,
|
|
"epoch": 6.062698773284871,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022771295884361088,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.25285209268331527,
|
|
"num_tokens": 13820073.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.33353853225708,
|
|
"epoch": 6.067242162653339,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002273674816334268,
|
|
"loss": 4.5973,
|
|
"mean_token_accuracy": 0.23765354752540588,
|
|
"num_tokens": 13830318.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.2007026195526125,
|
|
"epoch": 6.071785552021808,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022702212195174664,
|
|
"loss": 4.5221,
|
|
"mean_token_accuracy": 0.2405605658888817,
|
|
"num_tokens": 13840847.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.29194450378418,
|
|
"epoch": 6.076328941390277,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022667688065071146,
|
|
"loss": 4.5272,
|
|
"mean_token_accuracy": 0.242795892059803,
|
|
"num_tokens": 13850478.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.3235969066619875,
|
|
"epoch": 6.080872330758746,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022633175858216936,
|
|
"loss": 4.6012,
|
|
"mean_token_accuracy": 0.23387123495340348,
|
|
"num_tokens": 13860749.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.324441194534302,
|
|
"epoch": 6.085415720127215,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00022598675659767515,
|
|
"loss": 4.5709,
|
|
"mean_token_accuracy": 0.23779373615980148,
|
|
"num_tokens": 13869909.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.259878635406494,
|
|
"epoch": 6.089959109495684,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022564187554848663,
|
|
"loss": 4.5306,
|
|
"mean_token_accuracy": 0.24562186151742935,
|
|
"num_tokens": 13880235.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.166162919998169,
|
|
"epoch": 6.094502498864153,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022529711628556387,
|
|
"loss": 4.4378,
|
|
"mean_token_accuracy": 0.24997256100177764,
|
|
"num_tokens": 13890324.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.331091022491455,
|
|
"epoch": 6.099045888232622,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022495247965956567,
|
|
"loss": 4.5739,
|
|
"mean_token_accuracy": 0.24052052199840546,
|
|
"num_tokens": 13900250.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.318366479873657,
|
|
"epoch": 6.10358927760109,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002246079665208489,
|
|
"loss": 4.5211,
|
|
"mean_token_accuracy": 0.2371494799852371,
|
|
"num_tokens": 13909736.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.260770559310913,
|
|
"epoch": 6.108132666969559,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022426357771946544,
|
|
"loss": 4.559,
|
|
"mean_token_accuracy": 0.24113679826259612,
|
|
"num_tokens": 13920187.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.267756366729737,
|
|
"epoch": 6.112676056338028,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00022391931410516036,
|
|
"loss": 4.5053,
|
|
"mean_token_accuracy": 0.2288672909140587,
|
|
"num_tokens": 13930858.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.223704051971436,
|
|
"epoch": 6.117219445706497,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00022357517652736992,
|
|
"loss": 4.4705,
|
|
"mean_token_accuracy": 0.25046726912260053,
|
|
"num_tokens": 13940771.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.270145082473755,
|
|
"epoch": 6.121762835074966,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00022323116583521948,
|
|
"loss": 4.6219,
|
|
"mean_token_accuracy": 0.22698426991701126,
|
|
"num_tokens": 13951312.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.288094854354858,
|
|
"epoch": 6.126306224443435,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00022288728287752109,
|
|
"loss": 4.5747,
|
|
"mean_token_accuracy": 0.23753902912139893,
|
|
"num_tokens": 13960862.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.29812650680542,
|
|
"epoch": 6.130849613811904,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002225435285027719,
|
|
"loss": 4.5312,
|
|
"mean_token_accuracy": 0.2384922832250595,
|
|
"num_tokens": 13972228.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.241841983795166,
|
|
"epoch": 6.135393003180373,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022219990355915148,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.24281405210494994,
|
|
"num_tokens": 13982977.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.321514368057251,
|
|
"epoch": 6.139936392548841,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00022185640889452036,
|
|
"loss": 4.5011,
|
|
"mean_token_accuracy": 0.2454376831650734,
|
|
"num_tokens": 13991645.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.315341377258301,
|
|
"epoch": 6.14447978191731,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00022151304535641747,
|
|
"loss": 4.5446,
|
|
"mean_token_accuracy": 0.2368765652179718,
|
|
"num_tokens": 14001313.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.2701410293579105,
|
|
"epoch": 6.149023171285779,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002211698137920582,
|
|
"loss": 4.5811,
|
|
"mean_token_accuracy": 0.23887269496917723,
|
|
"num_tokens": 14011248.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.299800062179566,
|
|
"epoch": 6.153566560654248,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022082671504833223,
|
|
"loss": 4.5327,
|
|
"mean_token_accuracy": 0.23949709236621858,
|
|
"num_tokens": 14022290.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.244191837310791,
|
|
"epoch": 6.158109950022717,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022048374997180175,
|
|
"loss": 4.5185,
|
|
"mean_token_accuracy": 0.24164315462112426,
|
|
"num_tokens": 14032899.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.297331380844116,
|
|
"epoch": 6.162653339391186,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00022014091940869885,
|
|
"loss": 4.5588,
|
|
"mean_token_accuracy": 0.24354132860898972,
|
|
"num_tokens": 14043065.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.290510129928589,
|
|
"epoch": 6.167196728759655,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021979822420492402,
|
|
"loss": 4.5879,
|
|
"mean_token_accuracy": 0.23244184553623198,
|
|
"num_tokens": 14054594.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.409772777557373,
|
|
"epoch": 6.171740118128124,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00021945566520604337,
|
|
"loss": 4.6493,
|
|
"mean_token_accuracy": 0.23465163707733155,
|
|
"num_tokens": 14064211.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.251884174346924,
|
|
"epoch": 6.176283507496592,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002191132432572875,
|
|
"loss": 4.5377,
|
|
"mean_token_accuracy": 0.24627399444580078,
|
|
"num_tokens": 14073946.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.273758363723755,
|
|
"epoch": 6.180826896865061,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021877095920354818,
|
|
"loss": 4.4821,
|
|
"mean_token_accuracy": 0.24490741938352584,
|
|
"num_tokens": 14083303.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.2886029243469235,
|
|
"epoch": 6.18537028623353,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021842881388937764,
|
|
"loss": 4.469,
|
|
"mean_token_accuracy": 0.2500068351626396,
|
|
"num_tokens": 14093475.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.33375039100647,
|
|
"epoch": 6.189913675601999,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021808680815898508,
|
|
"loss": 4.6249,
|
|
"mean_token_accuracy": 0.23255902975797654,
|
|
"num_tokens": 14104117.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.208165788650513,
|
|
"epoch": 6.194457064970468,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021774494285623598,
|
|
"loss": 4.4856,
|
|
"mean_token_accuracy": 0.24499616175889968,
|
|
"num_tokens": 14115480.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.2122283458709715,
|
|
"epoch": 6.199000454338937,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002174032188246487,
|
|
"loss": 4.5025,
|
|
"mean_token_accuracy": 0.24060845673084258,
|
|
"num_tokens": 14125827.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.280530261993408,
|
|
"epoch": 6.203543843707406,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021706163690739344,
|
|
"loss": 4.5329,
|
|
"mean_token_accuracy": 0.24781845957040788,
|
|
"num_tokens": 14136424.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.345702362060547,
|
|
"epoch": 6.208087233075875,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021672019794728963,
|
|
"loss": 4.6362,
|
|
"mean_token_accuracy": 0.22747345864772797,
|
|
"num_tokens": 14147478.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.32397084236145,
|
|
"epoch": 6.212630622444343,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021637890278680406,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.2396610900759697,
|
|
"num_tokens": 14159044.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.273679542541504,
|
|
"epoch": 6.217174011812812,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021603775226804845,
|
|
"loss": 4.5448,
|
|
"mean_token_accuracy": 0.23882315009832383,
|
|
"num_tokens": 14169370.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.285364198684692,
|
|
"epoch": 6.221717401181281,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021569674723277783,
|
|
"loss": 4.5796,
|
|
"mean_token_accuracy": 0.23821156322956086,
|
|
"num_tokens": 14179774.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.293082332611084,
|
|
"epoch": 6.22626079054975,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021535588852238823,
|
|
"loss": 4.5218,
|
|
"mean_token_accuracy": 0.24651459455490113,
|
|
"num_tokens": 14190182.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.233860635757447,
|
|
"epoch": 6.230804179918219,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00021501517697791479,
|
|
"loss": 4.4593,
|
|
"mean_token_accuracy": 0.24869432300329208,
|
|
"num_tokens": 14199505.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.287333297729492,
|
|
"epoch": 6.235347569286688,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002146746134400291,
|
|
"loss": 4.595,
|
|
"mean_token_accuracy": 0.23050543218851088,
|
|
"num_tokens": 14209677.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.270040464401245,
|
|
"epoch": 6.239890958655157,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021433419874903803,
|
|
"loss": 4.5152,
|
|
"mean_token_accuracy": 0.24049790501594542,
|
|
"num_tokens": 14219356.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.198821783065796,
|
|
"epoch": 6.244434348023626,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00021399393374488097,
|
|
"loss": 4.4563,
|
|
"mean_token_accuracy": 0.2569621682167053,
|
|
"num_tokens": 14227992.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.300835752487183,
|
|
"epoch": 6.248977737392094,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021365381926712797,
|
|
"loss": 4.5372,
|
|
"mean_token_accuracy": 0.24133182764053346,
|
|
"num_tokens": 14238354.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.293159055709839,
|
|
"epoch": 6.253521126760563,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021331385615497756,
|
|
"loss": 4.5661,
|
|
"mean_token_accuracy": 0.23737625032663345,
|
|
"num_tokens": 14249560.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.248990535736084,
|
|
"epoch": 6.258064516129032,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002129740452472552,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.237970532476902,
|
|
"num_tokens": 14259372.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.261585760116577,
|
|
"epoch": 6.262607905497501,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021263438738241026,
|
|
"loss": 4.51,
|
|
"mean_token_accuracy": 0.24273274540901185,
|
|
"num_tokens": 14270837.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.280829334259034,
|
|
"epoch": 6.26715129486597,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021229488339851498,
|
|
"loss": 4.5147,
|
|
"mean_token_accuracy": 0.23871345371007918,
|
|
"num_tokens": 14280138.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.339419794082642,
|
|
"epoch": 6.271694684234439,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021195553413326146,
|
|
"loss": 4.633,
|
|
"mean_token_accuracy": 0.2377998948097229,
|
|
"num_tokens": 14290131.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.249248790740967,
|
|
"epoch": 6.276238073602908,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002116163404239605,
|
|
"loss": 4.558,
|
|
"mean_token_accuracy": 0.23310813158750535,
|
|
"num_tokens": 14299961.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.212590932846069,
|
|
"epoch": 6.280781462971377,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021127730310753866,
|
|
"loss": 4.517,
|
|
"mean_token_accuracy": 0.2458632245659828,
|
|
"num_tokens": 14310529.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.2288063049316404,
|
|
"epoch": 6.285324852339846,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021093842302053706,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.25566790997982025,
|
|
"num_tokens": 14320220.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.374371528625488,
|
|
"epoch": 6.289868241708314,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021059970099910835,
|
|
"loss": 4.5983,
|
|
"mean_token_accuracy": 0.24157823771238326,
|
|
"num_tokens": 14331700.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.2958564281463625,
|
|
"epoch": 6.294411631076783,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021026113787901568,
|
|
"loss": 4.5232,
|
|
"mean_token_accuracy": 0.24252172261476518,
|
|
"num_tokens": 14341163.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.2156156539917,
|
|
"epoch": 6.298955020445252,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020992273449562966,
|
|
"loss": 4.5328,
|
|
"mean_token_accuracy": 0.240778349339962,
|
|
"num_tokens": 14351398.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.291507434844971,
|
|
"epoch": 6.303498409813721,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020958449168392723,
|
|
"loss": 4.5439,
|
|
"mean_token_accuracy": 0.24257104396820067,
|
|
"num_tokens": 14361920.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.227987909317017,
|
|
"epoch": 6.30804179918219,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002092464102784888,
|
|
"loss": 4.5326,
|
|
"mean_token_accuracy": 0.241629558801651,
|
|
"num_tokens": 14371980.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.268531703948975,
|
|
"epoch": 6.312585188550659,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020890849111349663,
|
|
"loss": 4.554,
|
|
"mean_token_accuracy": 0.236126609146595,
|
|
"num_tokens": 14381397.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.253949594497681,
|
|
"epoch": 6.317128577919128,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020857073502273266,
|
|
"loss": 4.5727,
|
|
"mean_token_accuracy": 0.23706353455781937,
|
|
"num_tokens": 14393476.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.284581184387207,
|
|
"epoch": 6.321671967287597,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002082331428395765,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.23917243778705596,
|
|
"num_tokens": 14404801.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.311735677719116,
|
|
"epoch": 6.326215356656065,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020789571539700326,
|
|
"loss": 4.5654,
|
|
"mean_token_accuracy": 0.23811901360750198,
|
|
"num_tokens": 14415921.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.235989952087403,
|
|
"epoch": 6.330758746024534,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020755845352758175,
|
|
"loss": 4.5919,
|
|
"mean_token_accuracy": 0.2385019913315773,
|
|
"num_tokens": 14426455.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.302737426757813,
|
|
"epoch": 6.335302135393003,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002072213580634718,
|
|
"loss": 4.5594,
|
|
"mean_token_accuracy": 0.24487907588481903,
|
|
"num_tokens": 14438720.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.280713272094727,
|
|
"epoch": 6.339845524761472,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002068844298364233,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.2363898500800133,
|
|
"num_tokens": 14449644.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.285313749313355,
|
|
"epoch": 6.344388914129941,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020654766967777288,
|
|
"loss": 4.6107,
|
|
"mean_token_accuracy": 0.22875442653894423,
|
|
"num_tokens": 14459712.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.275094747543335,
|
|
"epoch": 6.34893230349841,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002062110784184431,
|
|
"loss": 4.5698,
|
|
"mean_token_accuracy": 0.23609775304794312,
|
|
"num_tokens": 14469440.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.2865218162536625,
|
|
"epoch": 6.353475692866879,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020587465688893904,
|
|
"loss": 4.6116,
|
|
"mean_token_accuracy": 0.23184297382831573,
|
|
"num_tokens": 14480587.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.353555393218994,
|
|
"epoch": 6.358019082235348,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00020553840591934765,
|
|
"loss": 4.629,
|
|
"mean_token_accuracy": 0.2329724282026291,
|
|
"num_tokens": 14490360.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.358019082235348,
|
|
"eval_entropy": 5.1773331065510595,
|
|
"eval_loss": 5.643287181854248,
|
|
"eval_mean_token_accuracy": 0.17553106078574823,
|
|
"eval_num_tokens": 14490360.0,
|
|
"eval_runtime": 1.9963,
|
|
"eval_samples_per_second": 1720.688,
|
|
"eval_steps_per_second": 215.399,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.282586479187012,
|
|
"epoch": 6.362562471603816,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002052023263393346,
|
|
"loss": 4.5435,
|
|
"mean_token_accuracy": 0.24205131232738494,
|
|
"num_tokens": 14501135.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.289597463607788,
|
|
"epoch": 6.367105860972285,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00020486641897814295,
|
|
"loss": 4.5691,
|
|
"mean_token_accuracy": 0.2378766044974327,
|
|
"num_tokens": 14511828.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.263994598388672,
|
|
"epoch": 6.371649250340754,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00020453068466459067,
|
|
"loss": 4.541,
|
|
"mean_token_accuracy": 0.2399464190006256,
|
|
"num_tokens": 14521973.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.3048004627227785,
|
|
"epoch": 6.376192639709223,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020419512422706893,
|
|
"loss": 4.6151,
|
|
"mean_token_accuracy": 0.2408768579363823,
|
|
"num_tokens": 14532515.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.277401924133301,
|
|
"epoch": 6.380736029077692,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020385973849353957,
|
|
"loss": 4.5279,
|
|
"mean_token_accuracy": 0.238243705034256,
|
|
"num_tokens": 14542577.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.2327464580535885,
|
|
"epoch": 6.385279418446161,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020352452829153369,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.24342811107635498,
|
|
"num_tokens": 14552472.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.252198505401611,
|
|
"epoch": 6.38982280781463,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000203189494448149,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.24527997076511382,
|
|
"num_tokens": 14562615.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.231750774383545,
|
|
"epoch": 6.394366197183099,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020285463779004837,
|
|
"loss": 4.5348,
|
|
"mean_token_accuracy": 0.245841121673584,
|
|
"num_tokens": 14573200.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.32502589225769,
|
|
"epoch": 6.398909586551568,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020251995914345722,
|
|
"loss": 4.5661,
|
|
"mean_token_accuracy": 0.24313625544309617,
|
|
"num_tokens": 14583042.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.3683497428894045,
|
|
"epoch": 6.403452975920036,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020218545933416181,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.23867366462945938,
|
|
"num_tokens": 14594409.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.326343679428101,
|
|
"epoch": 6.407996365288505,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020185113918750724,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.23477310836315154,
|
|
"num_tokens": 14605370.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.273496532440186,
|
|
"epoch": 6.412539754656974,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002015169995283952,
|
|
"loss": 4.5728,
|
|
"mean_token_accuracy": 0.23810693770647048,
|
|
"num_tokens": 14615056.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.384608745574951,
|
|
"epoch": 6.417083144025443,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020118304118128211,
|
|
"loss": 4.6642,
|
|
"mean_token_accuracy": 0.22506267577409744,
|
|
"num_tokens": 14624959.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.258981847763062,
|
|
"epoch": 6.421626533393912,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020084926497017703,
|
|
"loss": 4.5038,
|
|
"mean_token_accuracy": 0.24744684398174285,
|
|
"num_tokens": 14635348.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.241959857940674,
|
|
"epoch": 6.426169922762381,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020051567171863944,
|
|
"loss": 4.5608,
|
|
"mean_token_accuracy": 0.23673139214515687,
|
|
"num_tokens": 14645232.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.301164197921753,
|
|
"epoch": 6.43071331213085,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002001822622497777,
|
|
"loss": 4.544,
|
|
"mean_token_accuracy": 0.24436991512775422,
|
|
"num_tokens": 14656194.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.33160400390625,
|
|
"epoch": 6.435256701499318,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019984903738624632,
|
|
"loss": 4.5793,
|
|
"mean_token_accuracy": 0.23906287997961045,
|
|
"num_tokens": 14667019.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.323686218261718,
|
|
"epoch": 6.439800090867787,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00019951599795024478,
|
|
"loss": 4.5596,
|
|
"mean_token_accuracy": 0.24220439344644545,
|
|
"num_tokens": 14676339.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.309477472305298,
|
|
"epoch": 6.444343480236256,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019918314476351462,
|
|
"loss": 4.5697,
|
|
"mean_token_accuracy": 0.2347677990794182,
|
|
"num_tokens": 14688206.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.264942026138305,
|
|
"epoch": 6.448886869604725,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019885047864733803,
|
|
"loss": 4.5364,
|
|
"mean_token_accuracy": 0.24475477933883666,
|
|
"num_tokens": 14698397.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.2720386505126955,
|
|
"epoch": 6.453430258973194,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019851800042253553,
|
|
"loss": 4.5035,
|
|
"mean_token_accuracy": 0.24619390219449996,
|
|
"num_tokens": 14708140.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.283140325546265,
|
|
"epoch": 6.457973648341663,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001981857109094642,
|
|
"loss": 4.5493,
|
|
"mean_token_accuracy": 0.24587465226650237,
|
|
"num_tokens": 14718710.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.295567989349365,
|
|
"epoch": 6.462517037710132,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019785361092801524,
|
|
"loss": 4.5658,
|
|
"mean_token_accuracy": 0.23595952093601227,
|
|
"num_tokens": 14729143.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.2722485065460205,
|
|
"epoch": 6.467060427078601,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019752170129761244,
|
|
"loss": 4.5449,
|
|
"mean_token_accuracy": 0.24405165314674376,
|
|
"num_tokens": 14739386.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.250110769271851,
|
|
"epoch": 6.47160381644707,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001971899828372098,
|
|
"loss": 4.5903,
|
|
"mean_token_accuracy": 0.23640431612730026,
|
|
"num_tokens": 14749209.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.330989265441895,
|
|
"epoch": 6.476147205815538,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019685845636528972,
|
|
"loss": 4.5833,
|
|
"mean_token_accuracy": 0.24869367182254792,
|
|
"num_tokens": 14759351.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.296887922286987,
|
|
"epoch": 6.480690595184007,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00019652712269986074,
|
|
"loss": 4.5593,
|
|
"mean_token_accuracy": 0.23658141046762465,
|
|
"num_tokens": 14769069.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.307418394088745,
|
|
"epoch": 6.485233984552476,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019619598265845556,
|
|
"loss": 4.5747,
|
|
"mean_token_accuracy": 0.23889029622077942,
|
|
"num_tokens": 14778563.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.2172082424163815,
|
|
"epoch": 6.489777373920945,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019586503705812969,
|
|
"loss": 4.5469,
|
|
"mean_token_accuracy": 0.24494533687829972,
|
|
"num_tokens": 14788305.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.283086061477661,
|
|
"epoch": 6.494320763289414,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001955342867154582,
|
|
"loss": 4.6109,
|
|
"mean_token_accuracy": 0.22815056443214415,
|
|
"num_tokens": 14799463.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.22870774269104,
|
|
"epoch": 6.498864152657883,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019520373244653484,
|
|
"loss": 4.5047,
|
|
"mean_token_accuracy": 0.2523532062768936,
|
|
"num_tokens": 14809516.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.283847045898438,
|
|
"epoch": 6.503407542026352,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001948733750669693,
|
|
"loss": 4.6036,
|
|
"mean_token_accuracy": 0.2309289053082466,
|
|
"num_tokens": 14819605.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.298979234695435,
|
|
"epoch": 6.50795093139482,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019454321539188568,
|
|
"loss": 4.5591,
|
|
"mean_token_accuracy": 0.24131494164466857,
|
|
"num_tokens": 14830241.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.301935720443725,
|
|
"epoch": 6.512494320763289,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019421325423592007,
|
|
"loss": 4.5758,
|
|
"mean_token_accuracy": 0.23951137214899063,
|
|
"num_tokens": 14840294.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.309721660614014,
|
|
"epoch": 6.517037710131758,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001938834924132189,
|
|
"loss": 4.6116,
|
|
"mean_token_accuracy": 0.2380874738097191,
|
|
"num_tokens": 14850156.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.2290526866912845,
|
|
"epoch": 6.521581099500227,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001935539307374366,
|
|
"loss": 4.5124,
|
|
"mean_token_accuracy": 0.24687163829803466,
|
|
"num_tokens": 14859889.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.327756786346436,
|
|
"epoch": 6.526124488868696,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019322457002173395,
|
|
"loss": 4.5727,
|
|
"mean_token_accuracy": 0.24156240671873092,
|
|
"num_tokens": 14871659.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.3093681812286375,
|
|
"epoch": 6.530667878237165,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019289541107877562,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.2442413166165352,
|
|
"num_tokens": 14882041.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.287089824676514,
|
|
"epoch": 6.535211267605634,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019256645472072876,
|
|
"loss": 4.5311,
|
|
"mean_token_accuracy": 0.24632636606693267,
|
|
"num_tokens": 14891709.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.288113641738891,
|
|
"epoch": 6.539754656974103,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019223770175926032,
|
|
"loss": 4.5798,
|
|
"mean_token_accuracy": 0.2367118000984192,
|
|
"num_tokens": 14901898.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.233560705184937,
|
|
"epoch": 6.544298046342572,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00019190915300553574,
|
|
"loss": 4.4848,
|
|
"mean_token_accuracy": 0.2470690667629242,
|
|
"num_tokens": 14911910.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.278588056564331,
|
|
"epoch": 6.54884143571104,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001915808092702162,
|
|
"loss": 4.5237,
|
|
"mean_token_accuracy": 0.23909978717565536,
|
|
"num_tokens": 14922055.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.227743196487427,
|
|
"epoch": 6.553384825079509,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001912526713634574,
|
|
"loss": 4.5157,
|
|
"mean_token_accuracy": 0.24920731484889985,
|
|
"num_tokens": 14931841.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.258828639984131,
|
|
"epoch": 6.557928214447978,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001909247400949069,
|
|
"loss": 4.5364,
|
|
"mean_token_accuracy": 0.24476296603679656,
|
|
"num_tokens": 14942356.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.2583231925964355,
|
|
"epoch": 6.562471603816447,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019059701627370263,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.24037134200334548,
|
|
"num_tokens": 14952552.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.212012481689453,
|
|
"epoch": 6.567014993184916,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019026950070847025,
|
|
"loss": 4.5416,
|
|
"mean_token_accuracy": 0.2406063497066498,
|
|
"num_tokens": 14962728.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.222393465042114,
|
|
"epoch": 6.571558382553385,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018994219420732228,
|
|
"loss": 4.5243,
|
|
"mean_token_accuracy": 0.2484747737646103,
|
|
"num_tokens": 14974071.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.307152652740479,
|
|
"epoch": 6.576101771921854,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018961509757785462,
|
|
"loss": 4.6171,
|
|
"mean_token_accuracy": 0.24070972949266434,
|
|
"num_tokens": 14983567.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.2365074634552,
|
|
"epoch": 6.580645161290323,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00018928821162714587,
|
|
"loss": 4.5426,
|
|
"mean_token_accuracy": 0.2441147193312645,
|
|
"num_tokens": 14994172.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.263239526748658,
|
|
"epoch": 6.585188550658792,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018896153716175458,
|
|
"loss": 4.5197,
|
|
"mean_token_accuracy": 0.24267211258411409,
|
|
"num_tokens": 15004094.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.2458192825317385,
|
|
"epoch": 6.58973194002726,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001886350749877175,
|
|
"loss": 4.5413,
|
|
"mean_token_accuracy": 0.2396916076540947,
|
|
"num_tokens": 15013986.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.266790676116943,
|
|
"epoch": 6.594275329395729,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001883088259105476,
|
|
"loss": 4.5652,
|
|
"mean_token_accuracy": 0.240879125893116,
|
|
"num_tokens": 15024764.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.324742364883423,
|
|
"epoch": 6.598818718764198,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018798279073523204,
|
|
"loss": 4.58,
|
|
"mean_token_accuracy": 0.2355024591088295,
|
|
"num_tokens": 15034971.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.318066596984863,
|
|
"epoch": 6.603362108132667,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00018765697026623017,
|
|
"loss": 4.6268,
|
|
"mean_token_accuracy": 0.23213005661964417,
|
|
"num_tokens": 15045750.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.340573453903199,
|
|
"epoch": 6.607905497501136,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001873313653074718,
|
|
"loss": 4.6682,
|
|
"mean_token_accuracy": 0.2236604630947113,
|
|
"num_tokens": 15056840.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.2208765029907225,
|
|
"epoch": 6.612448886869605,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00018700597666235452,
|
|
"loss": 4.5227,
|
|
"mean_token_accuracy": 0.24196203649044037,
|
|
"num_tokens": 15067451.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.300867176055908,
|
|
"epoch": 6.616992276238074,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018668080513374268,
|
|
"loss": 4.573,
|
|
"mean_token_accuracy": 0.2417721465229988,
|
|
"num_tokens": 15077780.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.295457935333252,
|
|
"epoch": 6.621535665606542,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018635585152396457,
|
|
"loss": 4.6417,
|
|
"mean_token_accuracy": 0.22853360176086426,
|
|
"num_tokens": 15089100.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.228935813903808,
|
|
"epoch": 6.626079054975011,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000186031116634811,
|
|
"loss": 4.5628,
|
|
"mean_token_accuracy": 0.2388447940349579,
|
|
"num_tokens": 15099420.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.334658908843994,
|
|
"epoch": 6.63062244434348,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018570660126753291,
|
|
"loss": 4.5792,
|
|
"mean_token_accuracy": 0.23664749711751937,
|
|
"num_tokens": 15109884.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.253931999206543,
|
|
"epoch": 6.635165833711949,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001853823062228399,
|
|
"loss": 4.5187,
|
|
"mean_token_accuracy": 0.249138243496418,
|
|
"num_tokens": 15119761.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.214855003356933,
|
|
"epoch": 6.639709223080418,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00018505823230089758,
|
|
"loss": 4.4966,
|
|
"mean_token_accuracy": 0.25730680525302885,
|
|
"num_tokens": 15129159.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.2968175411224365,
|
|
"epoch": 6.644252612448887,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00018473438030132633,
|
|
"loss": 4.5144,
|
|
"mean_token_accuracy": 0.24423515498638154,
|
|
"num_tokens": 15138855.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.28621826171875,
|
|
"epoch": 6.648796001817356,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001844107510231985,
|
|
"loss": 4.5449,
|
|
"mean_token_accuracy": 0.24247275292873383,
|
|
"num_tokens": 15149957.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.2638030529022215,
|
|
"epoch": 6.653339391185825,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00018408734526503744,
|
|
"loss": 4.5678,
|
|
"mean_token_accuracy": 0.2442340597510338,
|
|
"num_tokens": 15159110.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.282388973236084,
|
|
"epoch": 6.657882780554294,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018376416382481442,
|
|
"loss": 4.6319,
|
|
"mean_token_accuracy": 0.2340039476752281,
|
|
"num_tokens": 15169444.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.404678153991699,
|
|
"epoch": 6.662426169922762,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00018344120749994787,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.22114427983760834,
|
|
"num_tokens": 15180246.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.240191411972046,
|
|
"epoch": 6.666969559291231,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018311847708730005,
|
|
"loss": 4.5357,
|
|
"mean_token_accuracy": 0.23754653334617615,
|
|
"num_tokens": 15191578.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.25492377281189,
|
|
"epoch": 6.6715129486597,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001827959733831765,
|
|
"loss": 4.5641,
|
|
"mean_token_accuracy": 0.23596367985010147,
|
|
"num_tokens": 15202380.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.2239374160766605,
|
|
"epoch": 6.676056338028169,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018247369718332274,
|
|
"loss": 4.5676,
|
|
"mean_token_accuracy": 0.24195376187562942,
|
|
"num_tokens": 15214073.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.244336986541748,
|
|
"epoch": 6.680599727396638,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018215164928292354,
|
|
"loss": 4.5078,
|
|
"mean_token_accuracy": 0.2501635953783989,
|
|
"num_tokens": 15223655.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.369761991500854,
|
|
"epoch": 6.685143116765107,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018182983047659984,
|
|
"loss": 4.6667,
|
|
"mean_token_accuracy": 0.22578510791063308,
|
|
"num_tokens": 15235567.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.279752397537232,
|
|
"epoch": 6.689686506133576,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018150824155840778,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.235878649353981,
|
|
"num_tokens": 15246349.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.322544145584106,
|
|
"epoch": 6.694229895502044,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001811868833218358,
|
|
"loss": 4.6105,
|
|
"mean_token_accuracy": 0.2407063364982605,
|
|
"num_tokens": 15256443.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.274238348007202,
|
|
"epoch": 6.698773284870513,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018086575655980376,
|
|
"loss": 4.5464,
|
|
"mean_token_accuracy": 0.24427748322486878,
|
|
"num_tokens": 15267392.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.260951375961303,
|
|
"epoch": 6.703316674238982,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00018054486206465952,
|
|
"loss": 4.5154,
|
|
"mean_token_accuracy": 0.2466824471950531,
|
|
"num_tokens": 15277371.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.296751213073731,
|
|
"epoch": 6.707860063607451,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001802242006281788,
|
|
"loss": 4.5213,
|
|
"mean_token_accuracy": 0.24309378266334533,
|
|
"num_tokens": 15287490.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.247218751907349,
|
|
"epoch": 6.71240345297592,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017990377304156165,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.24091288596391677,
|
|
"num_tokens": 15298061.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.2216980934143065,
|
|
"epoch": 6.716946842344389,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001795835800954312,
|
|
"loss": 4.5207,
|
|
"mean_token_accuracy": 0.2428148403763771,
|
|
"num_tokens": 15307862.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.170067882537841,
|
|
"epoch": 6.721490231712858,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017926362257983174,
|
|
"loss": 4.5904,
|
|
"mean_token_accuracy": 0.24157393872737884,
|
|
"num_tokens": 15317693.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.2622888565063475,
|
|
"epoch": 6.726033621081327,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001789439012842269,
|
|
"loss": 4.6031,
|
|
"mean_token_accuracy": 0.2358802378177643,
|
|
"num_tokens": 15329000.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.3240193843841555,
|
|
"epoch": 6.730577010449796,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017862441699749687,
|
|
"loss": 4.5931,
|
|
"mean_token_accuracy": 0.24630043655633926,
|
|
"num_tokens": 15339967.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.322586870193481,
|
|
"epoch": 6.735120399818264,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001783051705079375,
|
|
"loss": 4.604,
|
|
"mean_token_accuracy": 0.23926246017217637,
|
|
"num_tokens": 15350478.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.336464166641235,
|
|
"epoch": 6.739663789186733,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017798616260325767,
|
|
"loss": 4.5603,
|
|
"mean_token_accuracy": 0.24136037826538087,
|
|
"num_tokens": 15360728.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.306900835037231,
|
|
"epoch": 6.744207178555202,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001776673940705778,
|
|
"loss": 4.5858,
|
|
"mean_token_accuracy": 0.24291121065616608,
|
|
"num_tokens": 15370689.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.241793870925903,
|
|
"epoch": 6.748750567923671,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001773488656964276,
|
|
"loss": 4.495,
|
|
"mean_token_accuracy": 0.2457278400659561,
|
|
"num_tokens": 15380987.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.31299147605896,
|
|
"epoch": 6.75329395729214,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000177030578266744,
|
|
"loss": 4.5831,
|
|
"mean_token_accuracy": 0.24300929456949233,
|
|
"num_tokens": 15391989.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.297779750823975,
|
|
"epoch": 6.757837346660609,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017671253256686963,
|
|
"loss": 4.6077,
|
|
"mean_token_accuracy": 0.23518924564123153,
|
|
"num_tokens": 15402376.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.251939916610718,
|
|
"epoch": 6.762380736029078,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00017639472938155075,
|
|
"loss": 4.5493,
|
|
"mean_token_accuracy": 0.238115893304348,
|
|
"num_tokens": 15413034.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.222417831420898,
|
|
"epoch": 6.766924125397547,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017607716949493514,
|
|
"loss": 4.5123,
|
|
"mean_token_accuracy": 0.23979702591896057,
|
|
"num_tokens": 15423378.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.199656772613525,
|
|
"epoch": 6.771467514766016,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017575985369057008,
|
|
"loss": 4.509,
|
|
"mean_token_accuracy": 0.24026822298765182,
|
|
"num_tokens": 15434092.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.219292736053466,
|
|
"epoch": 6.776010904134484,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001754427827514008,
|
|
"loss": 4.5207,
|
|
"mean_token_accuracy": 0.24072256982326506,
|
|
"num_tokens": 15445178.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.300052165985107,
|
|
"epoch": 6.780554293502953,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017512595745976855,
|
|
"loss": 4.5693,
|
|
"mean_token_accuracy": 0.24386228919029235,
|
|
"num_tokens": 15455456.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.243346214294434,
|
|
"epoch": 6.785097682871422,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017480937859740793,
|
|
"loss": 4.545,
|
|
"mean_token_accuracy": 0.24423087388277054,
|
|
"num_tokens": 15465805.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.320639419555664,
|
|
"epoch": 6.789641072239891,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017449304694544608,
|
|
"loss": 4.5895,
|
|
"mean_token_accuracy": 0.23743849247694016,
|
|
"num_tokens": 15476590.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.224790859222412,
|
|
"epoch": 6.79418446160836,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017417696328439974,
|
|
"loss": 4.5357,
|
|
"mean_token_accuracy": 0.24247083216905593,
|
|
"num_tokens": 15486791.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.255793476104737,
|
|
"epoch": 6.798727850976829,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017386112839417402,
|
|
"loss": 4.5209,
|
|
"mean_token_accuracy": 0.24733019769191741,
|
|
"num_tokens": 15497116.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.280299043655395,
|
|
"epoch": 6.803271240345298,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017354554305405994,
|
|
"loss": 4.5516,
|
|
"mean_token_accuracy": 0.2452671706676483,
|
|
"num_tokens": 15506731.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.278467416763306,
|
|
"epoch": 6.807814629713766,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001732302080427332,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.23676748126745223,
|
|
"num_tokens": 15517333.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.251539993286133,
|
|
"epoch": 6.812358019082235,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017291512413825148,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.2410186216235161,
|
|
"num_tokens": 15526995.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.812358019082235,
|
|
"eval_entropy": 5.130557036954303,
|
|
"eval_loss": 5.636660575866699,
|
|
"eval_mean_token_accuracy": 0.1763098762825478,
|
|
"eval_num_tokens": 15526995.0,
|
|
"eval_runtime": 2.1865,
|
|
"eval_samples_per_second": 1571.027,
|
|
"eval_steps_per_second": 196.664,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.233345890045166,
|
|
"epoch": 6.816901408450704,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00017260029211805297,
|
|
"loss": 4.5282,
|
|
"mean_token_accuracy": 0.2380431681871414,
|
|
"num_tokens": 15536686.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.311452150344849,
|
|
"epoch": 6.821444797819173,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00017228571275895434,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.23320150822401048,
|
|
"num_tokens": 15547565.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.2007386684417725,
|
|
"epoch": 6.825988187187642,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017197138683714902,
|
|
"loss": 4.5304,
|
|
"mean_token_accuracy": 0.23952358365058898,
|
|
"num_tokens": 15557962.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.240574502944947,
|
|
"epoch": 6.830531576556111,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001716573151282049,
|
|
"loss": 4.5383,
|
|
"mean_token_accuracy": 0.2436266839504242,
|
|
"num_tokens": 15567164.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.230358409881592,
|
|
"epoch": 6.83507496592458,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001713434984070627,
|
|
"loss": 4.5187,
|
|
"mean_token_accuracy": 0.240547975897789,
|
|
"num_tokens": 15577917.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.314674615859985,
|
|
"epoch": 6.839618355293049,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017102993744803392,
|
|
"loss": 4.6095,
|
|
"mean_token_accuracy": 0.2344943627715111,
|
|
"num_tokens": 15589106.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.280606937408447,
|
|
"epoch": 6.844161744661518,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017071663302479923,
|
|
"loss": 4.6153,
|
|
"mean_token_accuracy": 0.2297571286559105,
|
|
"num_tokens": 15600368.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.302418804168701,
|
|
"epoch": 6.848705134029986,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017040358591040606,
|
|
"loss": 4.5437,
|
|
"mean_token_accuracy": 0.24012238681316375,
|
|
"num_tokens": 15610671.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.177415561676026,
|
|
"epoch": 6.853248523398455,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000170090796877267,
|
|
"loss": 4.4817,
|
|
"mean_token_accuracy": 0.2476690888404846,
|
|
"num_tokens": 15620403.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.311693620681763,
|
|
"epoch": 6.857791912766924,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016977826669715783,
|
|
"loss": 4.5672,
|
|
"mean_token_accuracy": 0.24390484392642975,
|
|
"num_tokens": 15631260.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.262029027938842,
|
|
"epoch": 6.862335302135393,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000169465996141216,
|
|
"loss": 4.5555,
|
|
"mean_token_accuracy": 0.23971976190805436,
|
|
"num_tokens": 15641604.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.294697237014771,
|
|
"epoch": 6.866878691503862,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001691539859799377,
|
|
"loss": 4.5706,
|
|
"mean_token_accuracy": 0.2355852633714676,
|
|
"num_tokens": 15652469.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.240709972381592,
|
|
"epoch": 6.871422080872331,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016884223698317722,
|
|
"loss": 4.5741,
|
|
"mean_token_accuracy": 0.23263973146677017,
|
|
"num_tokens": 15662615.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.2527830600738525,
|
|
"epoch": 6.8759654702408,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001685307499201441,
|
|
"loss": 4.5149,
|
|
"mean_token_accuracy": 0.23653629422187805,
|
|
"num_tokens": 15674868.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.263212776184082,
|
|
"epoch": 6.880508859609268,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001682195255594019,
|
|
"loss": 4.6156,
|
|
"mean_token_accuracy": 0.24403761327266693,
|
|
"num_tokens": 15686246.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.274565267562866,
|
|
"epoch": 6.885052248977737,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016790856466886544,
|
|
"loss": 4.5357,
|
|
"mean_token_accuracy": 0.23703768998384475,
|
|
"num_tokens": 15697025.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.283926010131836,
|
|
"epoch": 6.889595638346206,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016759786801580018,
|
|
"loss": 4.6114,
|
|
"mean_token_accuracy": 0.2332889497280121,
|
|
"num_tokens": 15706636.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.2440262794494625,
|
|
"epoch": 6.894139027714675,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016728743636681897,
|
|
"loss": 4.5355,
|
|
"mean_token_accuracy": 0.24584799259901047,
|
|
"num_tokens": 15717319.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.225335741043091,
|
|
"epoch": 6.898682417083144,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00016697727048788118,
|
|
"loss": 4.464,
|
|
"mean_token_accuracy": 0.24964000284671783,
|
|
"num_tokens": 15726546.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.230144309997558,
|
|
"epoch": 6.903225806451613,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016666737114429017,
|
|
"loss": 4.5891,
|
|
"mean_token_accuracy": 0.22845591008663177,
|
|
"num_tokens": 15737176.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.2555849075317385,
|
|
"epoch": 6.907769195820082,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001663577391006919,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.24325163066387176,
|
|
"num_tokens": 15747422.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.2976850986480715,
|
|
"epoch": 6.912312585188551,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016604837512107263,
|
|
"loss": 4.5899,
|
|
"mean_token_accuracy": 0.22807395607233047,
|
|
"num_tokens": 15757591.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.278740167617798,
|
|
"epoch": 6.91685597455702,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016573927996875714,
|
|
"loss": 4.6127,
|
|
"mean_token_accuracy": 0.23937632888555527,
|
|
"num_tokens": 15768375.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.260160827636719,
|
|
"epoch": 6.921399363925488,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00016543045440640697,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.24449501633644105,
|
|
"num_tokens": 15778935.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.2692841529846195,
|
|
"epoch": 6.925942753293957,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016512189919601865,
|
|
"loss": 4.513,
|
|
"mean_token_accuracy": 0.24943280220031738,
|
|
"num_tokens": 15788955.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.282735538482666,
|
|
"epoch": 6.930486142662426,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001648136150989214,
|
|
"loss": 4.5052,
|
|
"mean_token_accuracy": 0.2515762746334076,
|
|
"num_tokens": 15798795.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.246687698364258,
|
|
"epoch": 6.935029532030895,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001645056028757756,
|
|
"loss": 4.5708,
|
|
"mean_token_accuracy": 0.23648328483104705,
|
|
"num_tokens": 15809251.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.221744060516357,
|
|
"epoch": 6.939572921399364,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016419786328657066,
|
|
"loss": 4.4925,
|
|
"mean_token_accuracy": 0.24388792961835862,
|
|
"num_tokens": 15819418.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.272593641281128,
|
|
"epoch": 6.944116310767833,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016389039709062376,
|
|
"loss": 4.5437,
|
|
"mean_token_accuracy": 0.23660973459482193,
|
|
"num_tokens": 15829933.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.221502256393433,
|
|
"epoch": 6.948659700136302,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016358320504657677,
|
|
"loss": 4.5579,
|
|
"mean_token_accuracy": 0.243408639729023,
|
|
"num_tokens": 15841032.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.296251583099365,
|
|
"epoch": 6.95320308950477,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016327628791239576,
|
|
"loss": 4.6079,
|
|
"mean_token_accuracy": 0.23286572843790054,
|
|
"num_tokens": 15851126.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.3174872398376465,
|
|
"epoch": 6.957746478873239,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016296964644536816,
|
|
"loss": 4.5649,
|
|
"mean_token_accuracy": 0.23511548936367035,
|
|
"num_tokens": 15861832.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.307540988922119,
|
|
"epoch": 6.962289868241708,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001626632814021015,
|
|
"loss": 4.5556,
|
|
"mean_token_accuracy": 0.23890337347984314,
|
|
"num_tokens": 15872295.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.309858798980713,
|
|
"epoch": 6.966833257610177,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016235719353852079,
|
|
"loss": 4.5712,
|
|
"mean_token_accuracy": 0.24022976607084273,
|
|
"num_tokens": 15882453.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.265645360946655,
|
|
"epoch": 6.971376646978646,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00016205138360986753,
|
|
"loss": 4.5883,
|
|
"mean_token_accuracy": 0.23292641788721086,
|
|
"num_tokens": 15891962.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.258993816375733,
|
|
"epoch": 6.975920036347115,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016174585237069738,
|
|
"loss": 4.5062,
|
|
"mean_token_accuracy": 0.2398236498236656,
|
|
"num_tokens": 15902309.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.233528757095337,
|
|
"epoch": 6.980463425715584,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016144060057487824,
|
|
"loss": 4.5142,
|
|
"mean_token_accuracy": 0.23637445718050004,
|
|
"num_tokens": 15912789.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.29595046043396,
|
|
"epoch": 6.985006815084053,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016113562897558849,
|
|
"loss": 4.5907,
|
|
"mean_token_accuracy": 0.2373379498720169,
|
|
"num_tokens": 15922722.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.224090814590454,
|
|
"epoch": 6.989550204452522,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001608309383253154,
|
|
"loss": 4.4691,
|
|
"mean_token_accuracy": 0.24732425659894944,
|
|
"num_tokens": 15933069.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.305351305007934,
|
|
"epoch": 6.99409359382099,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001605265293758528,
|
|
"loss": 4.5626,
|
|
"mean_token_accuracy": 0.23926981240510942,
|
|
"num_tokens": 15942427.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.269901609420776,
|
|
"epoch": 6.998636983189459,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016022240287829947,
|
|
"loss": 4.5635,
|
|
"mean_token_accuracy": 0.2347724288702011,
|
|
"num_tokens": 15952765.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.228502962324354,
|
|
"epoch": 7.002726033621081,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015991855958305724,
|
|
"loss": 4.5241,
|
|
"mean_token_accuracy": 0.24929977622297075,
|
|
"num_tokens": 15963025.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.221128749847412,
|
|
"epoch": 7.00726942298955,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015961500023982945,
|
|
"loss": 4.4624,
|
|
"mean_token_accuracy": 0.2509502485394478,
|
|
"num_tokens": 15973425.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.295962905883789,
|
|
"epoch": 7.011812812358019,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015931172559761848,
|
|
"loss": 4.4845,
|
|
"mean_token_accuracy": 0.25615841150283813,
|
|
"num_tokens": 15984064.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.280064535140991,
|
|
"epoch": 7.016356201726488,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001590087364047243,
|
|
"loss": 4.4619,
|
|
"mean_token_accuracy": 0.255840303003788,
|
|
"num_tokens": 15994512.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.247888469696045,
|
|
"epoch": 7.020899591094957,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00015870603340874267,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.2542534530162811,
|
|
"num_tokens": 16004140.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.2234711170196535,
|
|
"epoch": 7.025442980463426,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015840361735656333,
|
|
"loss": 4.4003,
|
|
"mean_token_accuracy": 0.25526335686445234,
|
|
"num_tokens": 16013710.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.3228026866912845,
|
|
"epoch": 7.029986369831895,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00015810148899436747,
|
|
"loss": 4.4769,
|
|
"mean_token_accuracy": 0.25151526927948,
|
|
"num_tokens": 16023846.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.25654559135437,
|
|
"epoch": 7.034529759200363,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015779964906762706,
|
|
"loss": 4.4143,
|
|
"mean_token_accuracy": 0.25026033371686934,
|
|
"num_tokens": 16034327.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.249836540222168,
|
|
"epoch": 7.039073148568832,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015749809832110197,
|
|
"loss": 4.4668,
|
|
"mean_token_accuracy": 0.25035436302423475,
|
|
"num_tokens": 16045055.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.21235933303833,
|
|
"epoch": 7.043616537937301,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001571968374988389,
|
|
"loss": 4.4471,
|
|
"mean_token_accuracy": 0.2450609877705574,
|
|
"num_tokens": 16055198.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.220421171188354,
|
|
"epoch": 7.04815992730577,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015689586734416876,
|
|
"loss": 4.3979,
|
|
"mean_token_accuracy": 0.2632571801543236,
|
|
"num_tokens": 16065030.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.287376451492309,
|
|
"epoch": 7.052703316674239,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015659518859970545,
|
|
"loss": 4.5061,
|
|
"mean_token_accuracy": 0.25308669805526735,
|
|
"num_tokens": 16075974.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.267031335830689,
|
|
"epoch": 7.057246706042708,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001562948020073441,
|
|
"loss": 4.4888,
|
|
"mean_token_accuracy": 0.2510784834623337,
|
|
"num_tokens": 16086835.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.248017263412476,
|
|
"epoch": 7.061790095411177,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015599470830825868,
|
|
"loss": 4.4767,
|
|
"mean_token_accuracy": 0.25257859379053116,
|
|
"num_tokens": 16098116.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.281676435470581,
|
|
"epoch": 7.066333484779646,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015569490824290054,
|
|
"loss": 4.5009,
|
|
"mean_token_accuracy": 0.23947454541921614,
|
|
"num_tokens": 16109185.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.2365460872650145,
|
|
"epoch": 7.070876874148114,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015539540255099643,
|
|
"loss": 4.4378,
|
|
"mean_token_accuracy": 0.25073806196451187,
|
|
"num_tokens": 16120025.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.271995496749878,
|
|
"epoch": 7.075420263516583,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015509619197154705,
|
|
"loss": 4.4152,
|
|
"mean_token_accuracy": 0.2564996525645256,
|
|
"num_tokens": 16131145.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.313193702697754,
|
|
"epoch": 7.079963652885052,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015479727724282476,
|
|
"loss": 4.4951,
|
|
"mean_token_accuracy": 0.24577263593673707,
|
|
"num_tokens": 16141063.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.244660902023315,
|
|
"epoch": 7.084507042253521,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015449865910237184,
|
|
"loss": 4.434,
|
|
"mean_token_accuracy": 0.25136834681034087,
|
|
"num_tokens": 16151059.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.235743999481201,
|
|
"epoch": 7.08905043162199,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00015420033828699882,
|
|
"loss": 4.435,
|
|
"mean_token_accuracy": 0.25774066895246506,
|
|
"num_tokens": 16160679.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.2251049995422365,
|
|
"epoch": 7.093593820990459,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001539023155327829,
|
|
"loss": 4.4401,
|
|
"mean_token_accuracy": 0.24547605365514755,
|
|
"num_tokens": 16170269.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.244662570953369,
|
|
"epoch": 7.098137210358928,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015360459157506525,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.25179740190505984,
|
|
"num_tokens": 16180496.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.264948606491089,
|
|
"epoch": 7.102680599727397,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015330716714845048,
|
|
"loss": 4.4619,
|
|
"mean_token_accuracy": 0.24535128623247146,
|
|
"num_tokens": 16191043.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.3361451625823975,
|
|
"epoch": 7.107223989095866,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001530100429868035,
|
|
"loss": 4.6062,
|
|
"mean_token_accuracy": 0.238248011469841,
|
|
"num_tokens": 16201551.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.254754590988159,
|
|
"epoch": 7.111767378464334,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015271321982324892,
|
|
"loss": 4.4986,
|
|
"mean_token_accuracy": 0.23770373016595842,
|
|
"num_tokens": 16212928.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.220784378051758,
|
|
"epoch": 7.116310767832803,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015241669839016818,
|
|
"loss": 4.442,
|
|
"mean_token_accuracy": 0.25250548124313354,
|
|
"num_tokens": 16223233.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.275541925430298,
|
|
"epoch": 7.120854157201272,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015212047941919854,
|
|
"loss": 4.5328,
|
|
"mean_token_accuracy": 0.24228892624378204,
|
|
"num_tokens": 16234902.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.2547601699829105,
|
|
"epoch": 7.125397546569741,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015182456364123086,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.2474678263068199,
|
|
"num_tokens": 16244846.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.25298228263855,
|
|
"epoch": 7.12994093593821,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00015152895178640788,
|
|
"loss": 4.4369,
|
|
"mean_token_accuracy": 0.25438217967748644,
|
|
"num_tokens": 16255067.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.251582431793213,
|
|
"epoch": 7.134484325306679,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001512336445841224,
|
|
"loss": 4.4761,
|
|
"mean_token_accuracy": 0.24326016157865524,
|
|
"num_tokens": 16265482.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.290931129455567,
|
|
"epoch": 7.139027714675148,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00015093864276301573,
|
|
"loss": 4.4814,
|
|
"mean_token_accuracy": 0.25080303102731705,
|
|
"num_tokens": 16275319.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.211306715011597,
|
|
"epoch": 7.143571104043617,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00015064394705097546,
|
|
"loss": 4.4694,
|
|
"mean_token_accuracy": 0.24363381415605545,
|
|
"num_tokens": 16284889.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.281263589859009,
|
|
"epoch": 7.148114493412085,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015034955817513397,
|
|
"loss": 4.4943,
|
|
"mean_token_accuracy": 0.24685753434896468,
|
|
"num_tokens": 16295259.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.217783737182617,
|
|
"epoch": 7.152657882780554,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001500554768618665,
|
|
"loss": 4.438,
|
|
"mean_token_accuracy": 0.2509037673473358,
|
|
"num_tokens": 16305489.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.194856452941894,
|
|
"epoch": 7.157201272149023,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00014976170383678954,
|
|
"loss": 4.4213,
|
|
"mean_token_accuracy": 0.260078963637352,
|
|
"num_tokens": 16315167.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.2194420337677006,
|
|
"epoch": 7.161744661517492,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014946823982475883,
|
|
"loss": 4.4089,
|
|
"mean_token_accuracy": 0.2591053992509842,
|
|
"num_tokens": 16325059.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.301082754135132,
|
|
"epoch": 7.166288050885961,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014917508554986754,
|
|
"loss": 4.5397,
|
|
"mean_token_accuracy": 0.2422347903251648,
|
|
"num_tokens": 16336647.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.245672655105591,
|
|
"epoch": 7.17083144025443,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00014888224173544476,
|
|
"loss": 4.4521,
|
|
"mean_token_accuracy": 0.2433578774333,
|
|
"num_tokens": 16346146.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.292208194732666,
|
|
"epoch": 7.175374829622899,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00014858970910405368,
|
|
"loss": 4.536,
|
|
"mean_token_accuracy": 0.24728219956159592,
|
|
"num_tokens": 16356124.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.265067911148071,
|
|
"epoch": 7.179918218991368,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014829748837748906,
|
|
"loss": 4.4941,
|
|
"mean_token_accuracy": 0.2442579373717308,
|
|
"num_tokens": 16367197.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.216362953186035,
|
|
"epoch": 7.184461608359836,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001480055802767769,
|
|
"loss": 4.4541,
|
|
"mean_token_accuracy": 0.24374762773513795,
|
|
"num_tokens": 16376757.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.22951922416687,
|
|
"epoch": 7.189004997728305,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00014771398552217114,
|
|
"loss": 4.4106,
|
|
"mean_token_accuracy": 0.2596891805529594,
|
|
"num_tokens": 16387454.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.26640510559082,
|
|
"epoch": 7.193548387096774,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014742270483315319,
|
|
"loss": 4.4766,
|
|
"mean_token_accuracy": 0.246715047955513,
|
|
"num_tokens": 16397852.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.236388254165649,
|
|
"epoch": 7.198091776465243,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014713173892842878,
|
|
"loss": 4.4401,
|
|
"mean_token_accuracy": 0.2535346120595932,
|
|
"num_tokens": 16407752.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.206421852111816,
|
|
"epoch": 7.202635165833712,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014684108852592765,
|
|
"loss": 4.4667,
|
|
"mean_token_accuracy": 0.24387474209070206,
|
|
"num_tokens": 16418500.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.254835081100464,
|
|
"epoch": 7.207178555202181,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014655075434280064,
|
|
"loss": 4.4968,
|
|
"mean_token_accuracy": 0.24457497894763947,
|
|
"num_tokens": 16428706.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.307448768615723,
|
|
"epoch": 7.21172194457065,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014626073709541853,
|
|
"loss": 4.4766,
|
|
"mean_token_accuracy": 0.2500042006373405,
|
|
"num_tokens": 16438184.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.2290356159210205,
|
|
"epoch": 7.216265333939119,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00014597103749937,
|
|
"loss": 4.4466,
|
|
"mean_token_accuracy": 0.253907798230648,
|
|
"num_tokens": 16449340.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.26077561378479,
|
|
"epoch": 7.220808723307587,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014568165626946012,
|
|
"loss": 4.437,
|
|
"mean_token_accuracy": 0.24509200900793077,
|
|
"num_tokens": 16459604.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.323547792434693,
|
|
"epoch": 7.225352112676056,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014539259411970823,
|
|
"loss": 4.5555,
|
|
"mean_token_accuracy": 0.24566399306058884,
|
|
"num_tokens": 16470150.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.295128059387207,
|
|
"epoch": 7.229895502044525,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001451038517633465,
|
|
"loss": 4.4861,
|
|
"mean_token_accuracy": 0.25775930285453796,
|
|
"num_tokens": 16480797.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.301133441925049,
|
|
"epoch": 7.234438891412994,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014481542991281782,
|
|
"loss": 4.5483,
|
|
"mean_token_accuracy": 0.24379820823669435,
|
|
"num_tokens": 16492629.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.263572645187378,
|
|
"epoch": 7.238982280781463,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014452732927977465,
|
|
"loss": 4.4012,
|
|
"mean_token_accuracy": 0.2515401586890221,
|
|
"num_tokens": 16503225.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.280962133407593,
|
|
"epoch": 7.243525670149932,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001442395505750766,
|
|
"loss": 4.4124,
|
|
"mean_token_accuracy": 0.2605512231588364,
|
|
"num_tokens": 16512446.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.224434089660645,
|
|
"epoch": 7.248069059518401,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014395209450878899,
|
|
"loss": 4.4531,
|
|
"mean_token_accuracy": 0.24781673401594162,
|
|
"num_tokens": 16522944.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.24012942314148,
|
|
"epoch": 7.25261244888687,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000143664961790181,
|
|
"loss": 4.4508,
|
|
"mean_token_accuracy": 0.2510864302515984,
|
|
"num_tokens": 16533100.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.260760688781739,
|
|
"epoch": 7.257155838255338,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001433781531277243,
|
|
"loss": 4.5317,
|
|
"mean_token_accuracy": 0.2428896263241768,
|
|
"num_tokens": 16542893.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.251643323898316,
|
|
"epoch": 7.261699227623807,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001430916692290905,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2514788404107094,
|
|
"num_tokens": 16553209.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.3143439292907715,
|
|
"epoch": 7.266242616992276,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014280551080115032,
|
|
"loss": 4.4899,
|
|
"mean_token_accuracy": 0.24916178733110428,
|
|
"num_tokens": 16564574.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 7.266242616992276,
|
|
"eval_entropy": 5.133170656825221,
|
|
"eval_loss": 5.6493449211120605,
|
|
"eval_mean_token_accuracy": 0.1755293685509715,
|
|
"eval_num_tokens": 16564574.0,
|
|
"eval_runtime": 1.9933,
|
|
"eval_samples_per_second": 1723.266,
|
|
"eval_steps_per_second": 215.722,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.280438423156738,
|
|
"epoch": 7.270786006360745,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014251967854997106,
|
|
"loss": 4.4913,
|
|
"mean_token_accuracy": 0.2498839482665062,
|
|
"num_tokens": 16574517.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.292503118515015,
|
|
"epoch": 7.275329395729214,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014223417318081572,
|
|
"loss": 4.4999,
|
|
"mean_token_accuracy": 0.24552914649248123,
|
|
"num_tokens": 16585311.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.1971556663513185,
|
|
"epoch": 7.279872785097683,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00014194899539814,
|
|
"loss": 4.4716,
|
|
"mean_token_accuracy": 0.2469364196062088,
|
|
"num_tokens": 16596055.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.174793863296509,
|
|
"epoch": 7.284416174466152,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014166414590559197,
|
|
"loss": 4.4334,
|
|
"mean_token_accuracy": 0.24765260517597198,
|
|
"num_tokens": 16606076.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.288254117965698,
|
|
"epoch": 7.288959563834621,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00014137962540600944,
|
|
"loss": 4.4855,
|
|
"mean_token_accuracy": 0.24352458268404006,
|
|
"num_tokens": 16616740.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.239024257659912,
|
|
"epoch": 7.29350295320309,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014109543460141837,
|
|
"loss": 4.4305,
|
|
"mean_token_accuracy": 0.25808209776878355,
|
|
"num_tokens": 16627412.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.212424564361572,
|
|
"epoch": 7.298046342571558,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001408115741930313,
|
|
"loss": 4.4438,
|
|
"mean_token_accuracy": 0.255937822163105,
|
|
"num_tokens": 16637032.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.1984686851501465,
|
|
"epoch": 7.302589731940027,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00014052804488124575,
|
|
"loss": 4.4359,
|
|
"mean_token_accuracy": 0.2461617887020111,
|
|
"num_tokens": 16646596.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.24813871383667,
|
|
"epoch": 7.307133121308496,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014024484736564196,
|
|
"loss": 4.3957,
|
|
"mean_token_accuracy": 0.25568548738956454,
|
|
"num_tokens": 16656404.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.276012945175171,
|
|
"epoch": 7.311676510676965,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001399619823449817,
|
|
"loss": 4.4693,
|
|
"mean_token_accuracy": 0.25020897537469866,
|
|
"num_tokens": 16666372.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.215802907943726,
|
|
"epoch": 7.316219900045434,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013967945051720614,
|
|
"loss": 4.4511,
|
|
"mean_token_accuracy": 0.2522910743951797,
|
|
"num_tokens": 16677150.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.227663373947143,
|
|
"epoch": 7.320763289413903,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013939725257943475,
|
|
"loss": 4.5262,
|
|
"mean_token_accuracy": 0.24668799638748168,
|
|
"num_tokens": 16688712.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.240382671356201,
|
|
"epoch": 7.325306678782372,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013911538922796275,
|
|
"loss": 4.4727,
|
|
"mean_token_accuracy": 0.24570324271917343,
|
|
"num_tokens": 16699214.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.215937805175781,
|
|
"epoch": 7.32985006815084,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013883386115825984,
|
|
"loss": 4.4579,
|
|
"mean_token_accuracy": 0.2494443878531456,
|
|
"num_tokens": 16709294.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.22654242515564,
|
|
"epoch": 7.334393457519309,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00013855266906496867,
|
|
"loss": 4.4243,
|
|
"mean_token_accuracy": 0.2541270345449448,
|
|
"num_tokens": 16718761.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.301029825210572,
|
|
"epoch": 7.338936846887778,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013827181364190288,
|
|
"loss": 4.527,
|
|
"mean_token_accuracy": 0.23994395583868028,
|
|
"num_tokens": 16729075.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.240935039520264,
|
|
"epoch": 7.343480236256247,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013799129558204505,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.2536480128765106,
|
|
"num_tokens": 16739293.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.25622935295105,
|
|
"epoch": 7.348023625624716,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013771111557754577,
|
|
"loss": 4.4793,
|
|
"mean_token_accuracy": 0.24594202935695647,
|
|
"num_tokens": 16749214.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.20169677734375,
|
|
"epoch": 7.352567014993185,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013743127431972124,
|
|
"loss": 4.4297,
|
|
"mean_token_accuracy": 0.2537811368703842,
|
|
"num_tokens": 16758715.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.241025495529175,
|
|
"epoch": 7.357110404361654,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001371517724990522,
|
|
"loss": 4.5105,
|
|
"mean_token_accuracy": 0.2460070475935936,
|
|
"num_tokens": 16768965.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.224939584732056,
|
|
"epoch": 7.361653793730123,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001368726108051813,
|
|
"loss": 4.3781,
|
|
"mean_token_accuracy": 0.2579348772764206,
|
|
"num_tokens": 16778368.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.281821584701538,
|
|
"epoch": 7.366197183098592,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013659378992691244,
|
|
"loss": 4.488,
|
|
"mean_token_accuracy": 0.2504989758133888,
|
|
"num_tokens": 16789396.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.261858940124512,
|
|
"epoch": 7.37074057246706,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013631531055220835,
|
|
"loss": 4.4894,
|
|
"mean_token_accuracy": 0.24530725479125975,
|
|
"num_tokens": 16800063.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.244523525238037,
|
|
"epoch": 7.375283961835529,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013603717336818934,
|
|
"loss": 4.4517,
|
|
"mean_token_accuracy": 0.24993257969617844,
|
|
"num_tokens": 16810344.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.21565990447998,
|
|
"epoch": 7.379827351203998,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013575937906113095,
|
|
"loss": 4.4155,
|
|
"mean_token_accuracy": 0.2538814663887024,
|
|
"num_tokens": 16820233.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.271827983856201,
|
|
"epoch": 7.384370740572467,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0001354819283164634,
|
|
"loss": 4.5119,
|
|
"mean_token_accuracy": 0.24819999933242798,
|
|
"num_tokens": 16829761.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.30153522491455,
|
|
"epoch": 7.388914129940936,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001352048218187686,
|
|
"loss": 4.5624,
|
|
"mean_token_accuracy": 0.24395005106925965,
|
|
"num_tokens": 16841068.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.240445137023926,
|
|
"epoch": 7.393457519309405,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00013492806025177935,
|
|
"loss": 4.4296,
|
|
"mean_token_accuracy": 0.25236994922161105,
|
|
"num_tokens": 16850733.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.241497993469238,
|
|
"epoch": 7.398000908677874,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00013465164429837716,
|
|
"loss": 4.4695,
|
|
"mean_token_accuracy": 0.2459257498383522,
|
|
"num_tokens": 16860416.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.291488790512085,
|
|
"epoch": 7.402544298046343,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001343755746405912,
|
|
"loss": 4.4908,
|
|
"mean_token_accuracy": 0.25597107112407685,
|
|
"num_tokens": 16870662.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.238854122161865,
|
|
"epoch": 7.407087687414811,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013409985195959574,
|
|
"loss": 4.5321,
|
|
"mean_token_accuracy": 0.24090685993432998,
|
|
"num_tokens": 16881758.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.25507698059082,
|
|
"epoch": 7.41163107678328,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013382447693570908,
|
|
"loss": 4.4823,
|
|
"mean_token_accuracy": 0.2433476358652115,
|
|
"num_tokens": 16892399.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.261378574371338,
|
|
"epoch": 7.416174466151749,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013354945024839166,
|
|
"loss": 4.5155,
|
|
"mean_token_accuracy": 0.24344522655010223,
|
|
"num_tokens": 16903696.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.265499830245972,
|
|
"epoch": 7.420717855520218,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013327477257624466,
|
|
"loss": 4.4745,
|
|
"mean_token_accuracy": 0.25274763256311417,
|
|
"num_tokens": 16914326.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.222095012664795,
|
|
"epoch": 7.425261244888687,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001330004445970078,
|
|
"loss": 4.3952,
|
|
"mean_token_accuracy": 0.26050612032413484,
|
|
"num_tokens": 16924533.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.24014983177185,
|
|
"epoch": 7.429804634257156,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013272646698755816,
|
|
"loss": 4.4205,
|
|
"mean_token_accuracy": 0.24926999509334563,
|
|
"num_tokens": 16934157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.175441694259644,
|
|
"epoch": 7.434348023625625,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013245284042390814,
|
|
"loss": 4.3766,
|
|
"mean_token_accuracy": 0.26181563138961794,
|
|
"num_tokens": 16944148.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.259666681289673,
|
|
"epoch": 7.438891412994094,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013217956558120436,
|
|
"loss": 4.5346,
|
|
"mean_token_accuracy": 0.24487215429544448,
|
|
"num_tokens": 16956388.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.245444059371948,
|
|
"epoch": 7.443434802362562,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000131906643133725,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.2540844291448593,
|
|
"num_tokens": 16965216.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.2513939380645756,
|
|
"epoch": 7.447978191731031,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013163407375487926,
|
|
"loss": 4.451,
|
|
"mean_token_accuracy": 0.2518219992518425,
|
|
"num_tokens": 16976250.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.219180488586426,
|
|
"epoch": 7.4525215810995,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013136185811720496,
|
|
"loss": 4.5039,
|
|
"mean_token_accuracy": 0.2443515971302986,
|
|
"num_tokens": 16987375.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.201214218139649,
|
|
"epoch": 7.457064970467969,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001310899968923671,
|
|
"loss": 4.4862,
|
|
"mean_token_accuracy": 0.25370626151561737,
|
|
"num_tokens": 16997526.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.254055452346802,
|
|
"epoch": 7.461608359836438,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001308184907511562,
|
|
"loss": 4.5155,
|
|
"mean_token_accuracy": 0.242329178750515,
|
|
"num_tokens": 17008342.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.262386608123779,
|
|
"epoch": 7.466151749204907,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013054734036348675,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.2520428538322449,
|
|
"num_tokens": 17019555.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.266987419128418,
|
|
"epoch": 7.470695138573376,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013027654639839536,
|
|
"loss": 4.507,
|
|
"mean_token_accuracy": 0.24945030957460404,
|
|
"num_tokens": 17030686.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.274075651168824,
|
|
"epoch": 7.475238527941845,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013000610952403917,
|
|
"loss": 4.4641,
|
|
"mean_token_accuracy": 0.25167514830827714,
|
|
"num_tokens": 17041089.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.184656763076783,
|
|
"epoch": 7.479781917310314,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00012973603040769427,
|
|
"loss": 4.425,
|
|
"mean_token_accuracy": 0.24760058224201204,
|
|
"num_tokens": 17050194.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.240808343887329,
|
|
"epoch": 7.484325306678782,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00012946630971575414,
|
|
"loss": 4.4632,
|
|
"mean_token_accuracy": 0.24651907682418822,
|
|
"num_tokens": 17060596.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.254859542846679,
|
|
"epoch": 7.488868696047251,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012919694811372768,
|
|
"loss": 4.502,
|
|
"mean_token_accuracy": 0.2406336098909378,
|
|
"num_tokens": 17069763.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.177891540527344,
|
|
"epoch": 7.49341208541572,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012892794626623793,
|
|
"loss": 4.4308,
|
|
"mean_token_accuracy": 0.25019834339618685,
|
|
"num_tokens": 17080031.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.231161260604859,
|
|
"epoch": 7.497955474784189,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001286593048370201,
|
|
"loss": 4.4733,
|
|
"mean_token_accuracy": 0.24586824476718902,
|
|
"num_tokens": 17091339.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.2030477046966555,
|
|
"epoch": 7.502498864152658,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012839102448892033,
|
|
"loss": 4.4315,
|
|
"mean_token_accuracy": 0.2525147020816803,
|
|
"num_tokens": 17102264.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.223801708221435,
|
|
"epoch": 7.507042253521127,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012812310588389367,
|
|
"loss": 4.4763,
|
|
"mean_token_accuracy": 0.24396389722824097,
|
|
"num_tokens": 17112493.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.245207023620606,
|
|
"epoch": 7.511585642889596,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012785554968300266,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.2553750976920128,
|
|
"num_tokens": 17122037.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.235300588607788,
|
|
"epoch": 7.516129032258064,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012758835654641548,
|
|
"loss": 4.4747,
|
|
"mean_token_accuracy": 0.25061350762844087,
|
|
"num_tokens": 17133403.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.184303045272827,
|
|
"epoch": 7.520672421626533,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000127321527133405,
|
|
"loss": 4.4182,
|
|
"mean_token_accuracy": 0.25094863772392273,
|
|
"num_tokens": 17143619.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.275459718704224,
|
|
"epoch": 7.525215810995002,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012705506210234583,
|
|
"loss": 4.5192,
|
|
"mean_token_accuracy": 0.2478477567434311,
|
|
"num_tokens": 17154811.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.213428258895874,
|
|
"epoch": 7.529759200363471,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012678896211071422,
|
|
"loss": 4.4432,
|
|
"mean_token_accuracy": 0.24109389930963515,
|
|
"num_tokens": 17166039.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.251965761184692,
|
|
"epoch": 7.53430258973194,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001265232278150854,
|
|
"loss": 4.4541,
|
|
"mean_token_accuracy": 0.2533461660146713,
|
|
"num_tokens": 17176555.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.220733642578125,
|
|
"epoch": 7.538845979100409,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001262578598711323,
|
|
"loss": 4.4928,
|
|
"mean_token_accuracy": 0.2435593456029892,
|
|
"num_tokens": 17188220.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.257509326934814,
|
|
"epoch": 7.543389368468878,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001259928589336239,
|
|
"loss": 4.4687,
|
|
"mean_token_accuracy": 0.2505897268652916,
|
|
"num_tokens": 17199528.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.240171575546265,
|
|
"epoch": 7.547932757837347,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012572822565642378,
|
|
"loss": 4.421,
|
|
"mean_token_accuracy": 0.2479221910238266,
|
|
"num_tokens": 17210030.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.254587316513062,
|
|
"epoch": 7.552476147205816,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012546396069248813,
|
|
"loss": 4.5228,
|
|
"mean_token_accuracy": 0.2425874024629593,
|
|
"num_tokens": 17220231.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.21451096534729,
|
|
"epoch": 7.557019536574284,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012520006469386458,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.2521732211112976,
|
|
"num_tokens": 17229966.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.255106353759766,
|
|
"epoch": 7.561562925942753,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00012493653831169014,
|
|
"loss": 4.4962,
|
|
"mean_token_accuracy": 0.24773452281951905,
|
|
"num_tokens": 17240654.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.184343147277832,
|
|
"epoch": 7.566106315311222,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00012467338219619015,
|
|
"loss": 4.4008,
|
|
"mean_token_accuracy": 0.2623116970062256,
|
|
"num_tokens": 17250653.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.291183710098267,
|
|
"epoch": 7.570649704679691,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000124410596996676,
|
|
"loss": 4.5045,
|
|
"mean_token_accuracy": 0.24988579005002975,
|
|
"num_tokens": 17260223.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.218622779846191,
|
|
"epoch": 7.57519309404816,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001241481833615442,
|
|
"loss": 4.4626,
|
|
"mean_token_accuracy": 0.24890027791261674,
|
|
"num_tokens": 17269998.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.237799739837646,
|
|
"epoch": 7.579736483416629,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012388614193827407,
|
|
"loss": 4.424,
|
|
"mean_token_accuracy": 0.2597310200333595,
|
|
"num_tokens": 17279824.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.202769708633423,
|
|
"epoch": 7.584279872785098,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000123624473373427,
|
|
"loss": 4.4429,
|
|
"mean_token_accuracy": 0.25143535882234574,
|
|
"num_tokens": 17290242.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.181853199005127,
|
|
"epoch": 7.588823262153566,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00012336317831264411,
|
|
"loss": 4.4203,
|
|
"mean_token_accuracy": 0.25428290516138075,
|
|
"num_tokens": 17299515.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.2100183963775635,
|
|
"epoch": 7.593366651522036,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00012310225740064498,
|
|
"loss": 4.4034,
|
|
"mean_token_accuracy": 0.2491419270634651,
|
|
"num_tokens": 17308627.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.166991758346557,
|
|
"epoch": 7.597910040890504,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001228417112812259,
|
|
"loss": 4.3941,
|
|
"mean_token_accuracy": 0.24664377570152282,
|
|
"num_tokens": 17319465.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.174731159210205,
|
|
"epoch": 7.602453430258973,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012258154059725877,
|
|
"loss": 4.423,
|
|
"mean_token_accuracy": 0.24944878220558167,
|
|
"num_tokens": 17329202.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.202054786682129,
|
|
"epoch": 7.606996819627442,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0001223217459906886,
|
|
"loss": 4.3753,
|
|
"mean_token_accuracy": 0.26430008113384246,
|
|
"num_tokens": 17338956.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.302431917190551,
|
|
"epoch": 7.611540208995911,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012206232810253297,
|
|
"loss": 4.5636,
|
|
"mean_token_accuracy": 0.24412962198257446,
|
|
"num_tokens": 17349380.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.2608537673950195,
|
|
"epoch": 7.61608359836438,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012180328757287967,
|
|
"loss": 4.5884,
|
|
"mean_token_accuracy": 0.2421300083398819,
|
|
"num_tokens": 17359770.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.209791088104248,
|
|
"epoch": 7.620626987732849,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012154462504088545,
|
|
"loss": 4.4475,
|
|
"mean_token_accuracy": 0.2503615379333496,
|
|
"num_tokens": 17369209.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.1992157936096195,
|
|
"epoch": 7.625170377101318,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001212863411447744,
|
|
"loss": 4.4522,
|
|
"mean_token_accuracy": 0.2534232929348946,
|
|
"num_tokens": 17379221.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.265651226043701,
|
|
"epoch": 7.629713766469786,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012102843652183628,
|
|
"loss": 4.3949,
|
|
"mean_token_accuracy": 0.25392640829086305,
|
|
"num_tokens": 17389164.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.305235862731934,
|
|
"epoch": 7.634257155838255,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00012077091180842523,
|
|
"loss": 4.4979,
|
|
"mean_token_accuracy": 0.25081674158573153,
|
|
"num_tokens": 17399392.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.259583425521851,
|
|
"epoch": 7.638800545206724,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012051376763995784,
|
|
"loss": 4.4757,
|
|
"mean_token_accuracy": 0.2494491845369339,
|
|
"num_tokens": 17408933.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.249682235717773,
|
|
"epoch": 7.643343934575193,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012025700465091177,
|
|
"loss": 4.4454,
|
|
"mean_token_accuracy": 0.24946374595165252,
|
|
"num_tokens": 17419391.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.230360412597657,
|
|
"epoch": 7.647887323943662,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00012000062347482413,
|
|
"loss": 4.4998,
|
|
"mean_token_accuracy": 0.24975171089172363,
|
|
"num_tokens": 17430562.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.224960422515869,
|
|
"epoch": 7.652430713312131,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011974462474429015,
|
|
"loss": 4.4499,
|
|
"mean_token_accuracy": 0.2458922818303108,
|
|
"num_tokens": 17441345.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.213955211639404,
|
|
"epoch": 7.6569741026806,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00011948900909096101,
|
|
"loss": 4.4449,
|
|
"mean_token_accuracy": 0.24568468034267427,
|
|
"num_tokens": 17450972.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.217272615432739,
|
|
"epoch": 7.661517492049069,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001192337771455431,
|
|
"loss": 4.4667,
|
|
"mean_token_accuracy": 0.24688876867294313,
|
|
"num_tokens": 17460930.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.281532430648804,
|
|
"epoch": 7.666060881417538,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011897892953779576,
|
|
"loss": 4.4518,
|
|
"mean_token_accuracy": 0.25351732820272443,
|
|
"num_tokens": 17470865.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.22095890045166,
|
|
"epoch": 7.670604270786006,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011872446689653032,
|
|
"loss": 4.3666,
|
|
"mean_token_accuracy": 0.259026363492012,
|
|
"num_tokens": 17481379.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.349306392669678,
|
|
"epoch": 7.675147660154475,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0001184703898496078,
|
|
"loss": 4.5244,
|
|
"mean_token_accuracy": 0.24534447938203813,
|
|
"num_tokens": 17492741.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.250436925888062,
|
|
"epoch": 7.679691049522944,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011821669902393823,
|
|
"loss": 4.5174,
|
|
"mean_token_accuracy": 0.2349986180663109,
|
|
"num_tokens": 17504179.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.192126750946045,
|
|
"epoch": 7.684234438891413,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011796339504547846,
|
|
"loss": 4.466,
|
|
"mean_token_accuracy": 0.25430354923009874,
|
|
"num_tokens": 17514299.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.278289413452148,
|
|
"epoch": 7.688777828259882,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00011771047853923086,
|
|
"loss": 4.4612,
|
|
"mean_token_accuracy": 0.251273638010025,
|
|
"num_tokens": 17524454.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.242253017425537,
|
|
"epoch": 7.693321217628351,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011745795012924169,
|
|
"loss": 4.4441,
|
|
"mean_token_accuracy": 0.2602087497711182,
|
|
"num_tokens": 17534887.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.2132587909698485,
|
|
"epoch": 7.69786460699682,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00011720581043859988,
|
|
"loss": 4.5129,
|
|
"mean_token_accuracy": 0.2462400823831558,
|
|
"num_tokens": 17544944.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.236640453338623,
|
|
"epoch": 7.702407996365288,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011695406008943495,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.25605556219816206,
|
|
"num_tokens": 17556481.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.309423780441284,
|
|
"epoch": 7.706951385733757,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00011670269970291584,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.24638880044221878,
|
|
"num_tokens": 17566776.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.209789514541626,
|
|
"epoch": 7.711494775102226,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001164517298992493,
|
|
"loss": 4.4072,
|
|
"mean_token_accuracy": 0.25440947711467743,
|
|
"num_tokens": 17575441.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.27567343711853,
|
|
"epoch": 7.716038164470695,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011620115129767848,
|
|
"loss": 4.5544,
|
|
"mean_token_accuracy": 0.2442579224705696,
|
|
"num_tokens": 17585792.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.239242935180664,
|
|
"epoch": 7.720581553839164,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011595096451648111,
|
|
"loss": 4.4487,
|
|
"mean_token_accuracy": 0.24456769824028016,
|
|
"num_tokens": 17595820.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.720581553839164,
|
|
"eval_entropy": 5.117334855988968,
|
|
"eval_loss": 5.646170616149902,
|
|
"eval_mean_token_accuracy": 0.17648643755982088,
|
|
"eval_num_tokens": 17595820.0,
|
|
"eval_runtime": 1.9934,
|
|
"eval_samples_per_second": 1723.147,
|
|
"eval_steps_per_second": 215.707,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.218241453170776,
|
|
"epoch": 7.725124943207633,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00011570117017296819,
|
|
"loss": 4.4245,
|
|
"mean_token_accuracy": 0.25033349096775054,
|
|
"num_tokens": 17605791.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.2263350009918215,
|
|
"epoch": 7.729668332576102,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011545176888348237,
|
|
"loss": 4.4676,
|
|
"mean_token_accuracy": 0.24659403115510942,
|
|
"num_tokens": 17616159.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.212412023544312,
|
|
"epoch": 7.734211721944571,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00011520276126339667,
|
|
"loss": 4.4278,
|
|
"mean_token_accuracy": 0.25336240977048874,
|
|
"num_tokens": 17626612.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.232559394836426,
|
|
"epoch": 7.73875511131304,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00011495414792711259,
|
|
"loss": 4.4549,
|
|
"mean_token_accuracy": 0.24381217807531358,
|
|
"num_tokens": 17636254.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.323142766952515,
|
|
"epoch": 7.743298500681508,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011470592948805882,
|
|
"loss": 4.5495,
|
|
"mean_token_accuracy": 0.238371804356575,
|
|
"num_tokens": 17648097.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.211718511581421,
|
|
"epoch": 7.747841890049977,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011445810655868957,
|
|
"loss": 4.4329,
|
|
"mean_token_accuracy": 0.25164134353399276,
|
|
"num_tokens": 17658361.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.178594970703125,
|
|
"epoch": 7.752385279418446,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011421067975048355,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.2596516579389572,
|
|
"num_tokens": 17668640.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.171883916854858,
|
|
"epoch": 7.756928668786915,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011396364967394152,
|
|
"loss": 4.3467,
|
|
"mean_token_accuracy": 0.25842963606119157,
|
|
"num_tokens": 17678257.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.188561153411865,
|
|
"epoch": 7.761472058155384,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011371701693858582,
|
|
"loss": 4.4617,
|
|
"mean_token_accuracy": 0.25200554430484773,
|
|
"num_tokens": 17688635.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.2542259216308596,
|
|
"epoch": 7.766015447523853,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011347078215295817,
|
|
"loss": 4.5068,
|
|
"mean_token_accuracy": 0.23866744339466095,
|
|
"num_tokens": 17699917.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.280057048797607,
|
|
"epoch": 7.770558836892322,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00011322494592461839,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.24951765388250352,
|
|
"num_tokens": 17710278.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.238742923736572,
|
|
"epoch": 7.77510222626079,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011297950886014292,
|
|
"loss": 4.4499,
|
|
"mean_token_accuracy": 0.2503645449876785,
|
|
"num_tokens": 17720375.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.224109506607055,
|
|
"epoch": 7.779645615629259,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001127344715651234,
|
|
"loss": 4.4737,
|
|
"mean_token_accuracy": 0.2371811792254448,
|
|
"num_tokens": 17730299.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.281082439422607,
|
|
"epoch": 7.784189004997728,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011248983464416497,
|
|
"loss": 4.487,
|
|
"mean_token_accuracy": 0.25235176682472227,
|
|
"num_tokens": 17741096.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.212363052368164,
|
|
"epoch": 7.788732394366197,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00011224559870088488,
|
|
"loss": 4.4556,
|
|
"mean_token_accuracy": 0.24940920174121856,
|
|
"num_tokens": 17751502.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.224365949630737,
|
|
"epoch": 7.793275783734666,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00011200176433791101,
|
|
"loss": 4.4843,
|
|
"mean_token_accuracy": 0.2504249319434166,
|
|
"num_tokens": 17761955.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.186493635177612,
|
|
"epoch": 7.797819173103135,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001117583321568805,
|
|
"loss": 4.4215,
|
|
"mean_token_accuracy": 0.24875668734312056,
|
|
"num_tokens": 17772330.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.234844398498535,
|
|
"epoch": 7.802362562471604,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011151530275843805,
|
|
"loss": 4.3928,
|
|
"mean_token_accuracy": 0.2582192331552505,
|
|
"num_tokens": 17781688.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.166111087799072,
|
|
"epoch": 7.806905951840073,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011127267674223449,
|
|
"loss": 4.4201,
|
|
"mean_token_accuracy": 0.25508803129196167,
|
|
"num_tokens": 17792037.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.232910442352295,
|
|
"epoch": 7.811449341208542,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011103045470692532,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.25035879760980606,
|
|
"num_tokens": 17802705.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.298873519897461,
|
|
"epoch": 7.81599273057701,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011078863725016955,
|
|
"loss": 4.5518,
|
|
"mean_token_accuracy": 0.2403860107064247,
|
|
"num_tokens": 17813758.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.204554700851441,
|
|
"epoch": 7.820536119945479,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011054722496862759,
|
|
"loss": 4.4287,
|
|
"mean_token_accuracy": 0.25158497095108034,
|
|
"num_tokens": 17823465.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.239207601547241,
|
|
"epoch": 7.825079509313948,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011030621845796021,
|
|
"loss": 4.4398,
|
|
"mean_token_accuracy": 0.25536522418260577,
|
|
"num_tokens": 17833719.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.208041381835938,
|
|
"epoch": 7.829622898682417,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011006561831282703,
|
|
"loss": 4.416,
|
|
"mean_token_accuracy": 0.2577873095870018,
|
|
"num_tokens": 17844519.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.195179224014282,
|
|
"epoch": 7.834166288050886,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010982542512688515,
|
|
"loss": 4.425,
|
|
"mean_token_accuracy": 0.2520665869116783,
|
|
"num_tokens": 17856463.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.160019111633301,
|
|
"epoch": 7.838709677419355,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010958563949278711,
|
|
"loss": 4.4179,
|
|
"mean_token_accuracy": 0.25291280299425123,
|
|
"num_tokens": 17867449.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.1565569877624515,
|
|
"epoch": 7.843253066787824,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010934626200218032,
|
|
"loss": 4.4255,
|
|
"mean_token_accuracy": 0.2475459799170494,
|
|
"num_tokens": 17878522.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.254718446731568,
|
|
"epoch": 7.847796456156293,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00010910729324570488,
|
|
"loss": 4.4813,
|
|
"mean_token_accuracy": 0.2505793243646622,
|
|
"num_tokens": 17888544.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.266549348831177,
|
|
"epoch": 7.852339845524762,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00010886873381299248,
|
|
"loss": 4.5344,
|
|
"mean_token_accuracy": 0.24304690212011337,
|
|
"num_tokens": 17897432.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.264706230163574,
|
|
"epoch": 7.85688323489323,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010863058429266469,
|
|
"loss": 4.4617,
|
|
"mean_token_accuracy": 0.25387533754110336,
|
|
"num_tokens": 17907868.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.266523790359497,
|
|
"epoch": 7.861426624261699,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00010839284527233192,
|
|
"loss": 4.4314,
|
|
"mean_token_accuracy": 0.24720678478479385,
|
|
"num_tokens": 17917887.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.208467054367065,
|
|
"epoch": 7.865970013630168,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010815551733859153,
|
|
"loss": 4.4167,
|
|
"mean_token_accuracy": 0.2504715070128441,
|
|
"num_tokens": 17928030.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.225989103317261,
|
|
"epoch": 7.870513402998637,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00010791860107702658,
|
|
"loss": 4.5406,
|
|
"mean_token_accuracy": 0.23917509317398072,
|
|
"num_tokens": 17937990.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.1877237319946286,
|
|
"epoch": 7.875056792367106,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010768209707220435,
|
|
"loss": 4.5187,
|
|
"mean_token_accuracy": 0.2439128950238228,
|
|
"num_tokens": 17948262.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.241894674301148,
|
|
"epoch": 7.879600181735575,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00010744600590767504,
|
|
"loss": 4.4863,
|
|
"mean_token_accuracy": 0.2449999198317528,
|
|
"num_tokens": 17958942.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.1922307968139645,
|
|
"epoch": 7.884143571104044,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010721032816597012,
|
|
"loss": 4.4705,
|
|
"mean_token_accuracy": 0.2487598404288292,
|
|
"num_tokens": 17969362.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.250648832321167,
|
|
"epoch": 7.888686960472512,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000106975064428601,
|
|
"loss": 4.4443,
|
|
"mean_token_accuracy": 0.2524044573307037,
|
|
"num_tokens": 17979009.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.275392532348633,
|
|
"epoch": 7.893230349840981,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010674021527605742,
|
|
"loss": 4.5032,
|
|
"mean_token_accuracy": 0.24657540321350097,
|
|
"num_tokens": 17989064.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.19702558517456,
|
|
"epoch": 7.89777373920945,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010650578128780656,
|
|
"loss": 4.4311,
|
|
"mean_token_accuracy": 0.2546486765146255,
|
|
"num_tokens": 17998574.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.325407791137695,
|
|
"epoch": 7.902317128577919,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001062717630422908,
|
|
"loss": 4.6386,
|
|
"mean_token_accuracy": 0.2380664438009262,
|
|
"num_tokens": 18010735.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.239076232910156,
|
|
"epoch": 7.906860517946388,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00010603816111692699,
|
|
"loss": 4.4603,
|
|
"mean_token_accuracy": 0.2524801388382912,
|
|
"num_tokens": 18020316.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.274438858032227,
|
|
"epoch": 7.911403907314857,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00010580497608810455,
|
|
"loss": 4.4634,
|
|
"mean_token_accuracy": 0.2451927751302719,
|
|
"num_tokens": 18031058.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.218672132492065,
|
|
"epoch": 7.915947296683326,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00010557220853118456,
|
|
"loss": 4.4533,
|
|
"mean_token_accuracy": 0.25305635929107667,
|
|
"num_tokens": 18040603.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.290171718597412,
|
|
"epoch": 7.920490686051795,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010533985902049755,
|
|
"loss": 4.5312,
|
|
"mean_token_accuracy": 0.24244319051504135,
|
|
"num_tokens": 18051369.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.200766563415527,
|
|
"epoch": 7.925034075420264,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010510792812934307,
|
|
"loss": 4.4097,
|
|
"mean_token_accuracy": 0.25834730118513105,
|
|
"num_tokens": 18062283.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.2463642120361325,
|
|
"epoch": 7.929577464788732,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010487641642998746,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.24884833246469498,
|
|
"num_tokens": 18072125.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.241290616989136,
|
|
"epoch": 7.934120854157201,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010464532449366281,
|
|
"loss": 4.5253,
|
|
"mean_token_accuracy": 0.24301611334085466,
|
|
"num_tokens": 18083812.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.270674371719361,
|
|
"epoch": 7.93866424352567,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010441465289056548,
|
|
"loss": 4.4861,
|
|
"mean_token_accuracy": 0.2527700439095497,
|
|
"num_tokens": 18095399.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.253411293029785,
|
|
"epoch": 7.943207632894139,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010418440218985482,
|
|
"loss": 4.4872,
|
|
"mean_token_accuracy": 0.25460486859083176,
|
|
"num_tokens": 18106234.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.207377195358276,
|
|
"epoch": 7.947751022262608,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010395457295965143,
|
|
"loss": 4.3873,
|
|
"mean_token_accuracy": 0.26062434166669846,
|
|
"num_tokens": 18116129.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.310530281066894,
|
|
"epoch": 7.952294411631077,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010372516576703617,
|
|
"loss": 4.513,
|
|
"mean_token_accuracy": 0.24756571352481843,
|
|
"num_tokens": 18126757.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.229206466674805,
|
|
"epoch": 7.956837800999546,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001034961811780484,
|
|
"loss": 4.4794,
|
|
"mean_token_accuracy": 0.2512534290552139,
|
|
"num_tokens": 18136645.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.272875356674194,
|
|
"epoch": 7.961381190368014,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00010326761975768492,
|
|
"loss": 4.5234,
|
|
"mean_token_accuracy": 0.24583324044942856,
|
|
"num_tokens": 18146458.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.260607624053955,
|
|
"epoch": 7.965924579736483,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010303948206989827,
|
|
"loss": 4.4651,
|
|
"mean_token_accuracy": 0.25206867456436155,
|
|
"num_tokens": 18156151.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.243887805938721,
|
|
"epoch": 7.970467969104952,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010281176867759556,
|
|
"loss": 4.4901,
|
|
"mean_token_accuracy": 0.24807964861392975,
|
|
"num_tokens": 18166695.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.21069655418396,
|
|
"epoch": 7.975011358473421,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010258448014263682,
|
|
"loss": 4.4311,
|
|
"mean_token_accuracy": 0.25389720797538756,
|
|
"num_tokens": 18177011.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.260270833969116,
|
|
"epoch": 7.97955474784189,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010235761702583405,
|
|
"loss": 4.5274,
|
|
"mean_token_accuracy": 0.2463166519999504,
|
|
"num_tokens": 18187179.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.218239593505859,
|
|
"epoch": 7.984098137210359,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010213117988694944,
|
|
"loss": 4.478,
|
|
"mean_token_accuracy": 0.2412942185997963,
|
|
"num_tokens": 18197106.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.157071113586426,
|
|
"epoch": 7.988641526578828,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010190516928469407,
|
|
"loss": 4.4492,
|
|
"mean_token_accuracy": 0.254221522808075,
|
|
"num_tokens": 18207729.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.215344715118408,
|
|
"epoch": 7.993184915947297,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010167958577672654,
|
|
"loss": 4.4061,
|
|
"mean_token_accuracy": 0.253074212372303,
|
|
"num_tokens": 18217885.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.248755359649659,
|
|
"epoch": 7.997728305315766,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010145442991965198,
|
|
"loss": 4.5076,
|
|
"mean_token_accuracy": 0.24511915296316147,
|
|
"num_tokens": 18229633.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.197750621371799,
|
|
"epoch": 8.001817355747388,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010122970226901981,
|
|
"loss": 4.4731,
|
|
"mean_token_accuracy": 0.2480786227517658,
|
|
"num_tokens": 18239194.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.273721313476562,
|
|
"epoch": 8.006360745115856,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010100540337932339,
|
|
"loss": 4.4338,
|
|
"mean_token_accuracy": 0.25612025558948515,
|
|
"num_tokens": 18250061.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.2211884498596195,
|
|
"epoch": 8.010904134484326,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00010078153380399773,
|
|
"loss": 4.3643,
|
|
"mean_token_accuracy": 0.26501971781253814,
|
|
"num_tokens": 18260977.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.238532638549804,
|
|
"epoch": 8.015447523852794,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010055809409541902,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.263276244699955,
|
|
"num_tokens": 18271866.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.232587480545044,
|
|
"epoch": 8.019990913221264,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010033508480490222,
|
|
"loss": 4.4637,
|
|
"mean_token_accuracy": 0.2523852840065956,
|
|
"num_tokens": 18283353.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.1490062236785885,
|
|
"epoch": 8.024534302589732,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00010011250648270081,
|
|
"loss": 4.4114,
|
|
"mean_token_accuracy": 0.2560569867491722,
|
|
"num_tokens": 18293619.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.239298963546753,
|
|
"epoch": 8.0290776919582,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.98903596780046e-05,
|
|
"loss": 4.3875,
|
|
"mean_token_accuracy": 0.264839743077755,
|
|
"num_tokens": 18303411.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.226376295089722,
|
|
"epoch": 8.03362108132667,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.966864493893876e-05,
|
|
"loss": 4.3469,
|
|
"mean_token_accuracy": 0.27113326638936996,
|
|
"num_tokens": 18312798.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.205983448028564,
|
|
"epoch": 8.038164470695138,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.94473628125623e-05,
|
|
"loss": 4.413,
|
|
"mean_token_accuracy": 0.2535378307104111,
|
|
"num_tokens": 18324111.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.221177864074707,
|
|
"epoch": 8.042707860063608,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.922651384486695e-05,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.2544989064335823,
|
|
"num_tokens": 18335197.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.2729676246643065,
|
|
"epoch": 8.047251249432076,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.900609858077562e-05,
|
|
"loss": 4.4231,
|
|
"mean_token_accuracy": 0.2547672167420387,
|
|
"num_tokens": 18345896.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.302001953125,
|
|
"epoch": 8.051794638800546,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.878611756414094e-05,
|
|
"loss": 4.4615,
|
|
"mean_token_accuracy": 0.25303480923175814,
|
|
"num_tokens": 18356941.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.209436702728271,
|
|
"epoch": 8.056338028169014,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.856657133774424e-05,
|
|
"loss": 4.3465,
|
|
"mean_token_accuracy": 0.26235208958387374,
|
|
"num_tokens": 18366767.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.2463689804077145,
|
|
"epoch": 8.060881417537482,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.834746044329411e-05,
|
|
"loss": 4.4753,
|
|
"mean_token_accuracy": 0.25343468189239504,
|
|
"num_tokens": 18376698.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.227330112457276,
|
|
"epoch": 8.065424806905952,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.81287854214248e-05,
|
|
"loss": 4.3655,
|
|
"mean_token_accuracy": 0.255557207763195,
|
|
"num_tokens": 18386678.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.244167232513428,
|
|
"epoch": 8.06996819627442,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.791054681169522e-05,
|
|
"loss": 4.4059,
|
|
"mean_token_accuracy": 0.25678226798772813,
|
|
"num_tokens": 18397020.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.180817604064941,
|
|
"epoch": 8.07451158564289,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.769274515258739e-05,
|
|
"loss": 4.3801,
|
|
"mean_token_accuracy": 0.25308036208152773,
|
|
"num_tokens": 18407330.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.194338941574097,
|
|
"epoch": 8.079054975011358,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.747538098150539e-05,
|
|
"loss": 4.3886,
|
|
"mean_token_accuracy": 0.25813722610473633,
|
|
"num_tokens": 18417046.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.2376251220703125,
|
|
"epoch": 8.083598364379828,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.725845483477352e-05,
|
|
"loss": 4.4509,
|
|
"mean_token_accuracy": 0.2572389915585518,
|
|
"num_tokens": 18427907.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.20089750289917,
|
|
"epoch": 8.088141753748296,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 9.704196724763565e-05,
|
|
"loss": 4.4042,
|
|
"mean_token_accuracy": 0.25502693355083467,
|
|
"num_tokens": 18437972.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.1834228515625,
|
|
"epoch": 8.092685143116766,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.682591875425323e-05,
|
|
"loss": 4.4176,
|
|
"mean_token_accuracy": 0.25532554984092715,
|
|
"num_tokens": 18448188.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.222577667236328,
|
|
"epoch": 8.097228532485234,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.661030988770463e-05,
|
|
"loss": 4.4413,
|
|
"mean_token_accuracy": 0.2507751315832138,
|
|
"num_tokens": 18458130.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.237573766708374,
|
|
"epoch": 8.101771921853702,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.63951411799831e-05,
|
|
"loss": 4.3965,
|
|
"mean_token_accuracy": 0.2596951812505722,
|
|
"num_tokens": 18468257.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.195806837081909,
|
|
"epoch": 8.106315311222172,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.618041316199617e-05,
|
|
"loss": 4.4225,
|
|
"mean_token_accuracy": 0.25173446387052534,
|
|
"num_tokens": 18478720.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.263276863098144,
|
|
"epoch": 8.11085870059064,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.596612636356383e-05,
|
|
"loss": 4.3923,
|
|
"mean_token_accuracy": 0.2563210591673851,
|
|
"num_tokens": 18489543.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.216325092315674,
|
|
"epoch": 8.11540208995911,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.575228131341745e-05,
|
|
"loss": 4.4168,
|
|
"mean_token_accuracy": 0.2557149201631546,
|
|
"num_tokens": 18499505.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.293788003921509,
|
|
"epoch": 8.119945479327578,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.55388785391984e-05,
|
|
"loss": 4.4864,
|
|
"mean_token_accuracy": 0.2508104547858238,
|
|
"num_tokens": 18510044.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.234007930755615,
|
|
"epoch": 8.124488868696048,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.532591856745689e-05,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.25259285271167753,
|
|
"num_tokens": 18520537.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.232175874710083,
|
|
"epoch": 8.129032258064516,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.511340192365044e-05,
|
|
"loss": 4.4706,
|
|
"mean_token_accuracy": 0.24527376890182495,
|
|
"num_tokens": 18531840.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.232611417770386,
|
|
"epoch": 8.133575647432984,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 9.490132913214277e-05,
|
|
"loss": 4.3758,
|
|
"mean_token_accuracy": 0.25673303604125974,
|
|
"num_tokens": 18542097.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.235378885269165,
|
|
"epoch": 8.138119036801454,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.468970071620231e-05,
|
|
"loss": 4.4293,
|
|
"mean_token_accuracy": 0.25237356424331664,
|
|
"num_tokens": 18551629.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.216175174713134,
|
|
"epoch": 8.142662426169922,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.447851719800126e-05,
|
|
"loss": 4.3804,
|
|
"mean_token_accuracy": 0.2556073233485222,
|
|
"num_tokens": 18562248.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.2167715549469,
|
|
"epoch": 8.147205815538392,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.426777909861395e-05,
|
|
"loss": 4.356,
|
|
"mean_token_accuracy": 0.25932457596063613,
|
|
"num_tokens": 18572107.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.217586660385132,
|
|
"epoch": 8.15174920490686,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.405748693801562e-05,
|
|
"loss": 4.4452,
|
|
"mean_token_accuracy": 0.2526706650853157,
|
|
"num_tokens": 18582665.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.312609434127808,
|
|
"epoch": 8.15629259427533,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.384764123508124e-05,
|
|
"loss": 4.4534,
|
|
"mean_token_accuracy": 0.2551994949579239,
|
|
"num_tokens": 18593737.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.247857761383057,
|
|
"epoch": 8.160835983643798,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.363824250758443e-05,
|
|
"loss": 4.418,
|
|
"mean_token_accuracy": 0.26122285425662994,
|
|
"num_tokens": 18604180.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.164255952835083,
|
|
"epoch": 8.165379373012268,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.34292912721955e-05,
|
|
"loss": 4.3384,
|
|
"mean_token_accuracy": 0.25732200443744657,
|
|
"num_tokens": 18614328.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.225952863693237,
|
|
"epoch": 8.169922762380736,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.322078804448095e-05,
|
|
"loss": 4.3861,
|
|
"mean_token_accuracy": 0.26659134328365325,
|
|
"num_tokens": 18625357.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.170964574813842,
|
|
"epoch": 8.174466151749204,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.301273333890171e-05,
|
|
"loss": 4.3756,
|
|
"mean_token_accuracy": 0.2602734431624413,
|
|
"num_tokens": 18636306.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 8.174466151749204,
|
|
"eval_entropy": 5.109929851044056,
|
|
"eval_loss": 5.6533660888671875,
|
|
"eval_mean_token_accuracy": 0.17622696754717548,
|
|
"eval_num_tokens": 18636306.0,
|
|
"eval_runtime": 2.0014,
|
|
"eval_samples_per_second": 1716.318,
|
|
"eval_steps_per_second": 214.852,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.2696582794189455,
|
|
"epoch": 8.179009541117674,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.280512766881227e-05,
|
|
"loss": 4.4247,
|
|
"mean_token_accuracy": 0.25623800456523893,
|
|
"num_tokens": 18647134.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.235560703277588,
|
|
"epoch": 8.183552930486142,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 9.25979715464588e-05,
|
|
"loss": 4.3653,
|
|
"mean_token_accuracy": 0.26241896152496336,
|
|
"num_tokens": 18656948.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.254617977142334,
|
|
"epoch": 8.188096319854612,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.23912654829784e-05,
|
|
"loss": 4.3919,
|
|
"mean_token_accuracy": 0.24983571320772172,
|
|
"num_tokens": 18667682.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.256754207611084,
|
|
"epoch": 8.19263970922308,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.218500998839794e-05,
|
|
"loss": 4.4683,
|
|
"mean_token_accuracy": 0.2545133665204048,
|
|
"num_tokens": 18678124.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.198268938064575,
|
|
"epoch": 8.19718309859155,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.197920557163215e-05,
|
|
"loss": 4.4178,
|
|
"mean_token_accuracy": 0.2516356140375137,
|
|
"num_tokens": 18689063.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.255533456802368,
|
|
"epoch": 8.201726487960018,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.177385274048312e-05,
|
|
"loss": 4.4685,
|
|
"mean_token_accuracy": 0.24894030392169952,
|
|
"num_tokens": 18700118.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.2536211013793945,
|
|
"epoch": 8.206269877328488,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.156895200163837e-05,
|
|
"loss": 4.4318,
|
|
"mean_token_accuracy": 0.2506991297006607,
|
|
"num_tokens": 18711683.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.242490196228028,
|
|
"epoch": 8.210813266696956,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.13645038606703e-05,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.25329880565404894,
|
|
"num_tokens": 18722033.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.2758465766906735,
|
|
"epoch": 8.215356656065424,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.116050882203427e-05,
|
|
"loss": 4.4426,
|
|
"mean_token_accuracy": 0.2527270942926407,
|
|
"num_tokens": 18732387.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.254736852645874,
|
|
"epoch": 8.219900045433894,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.09569673890678e-05,
|
|
"loss": 4.4476,
|
|
"mean_token_accuracy": 0.2497573524713516,
|
|
"num_tokens": 18743573.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.157283115386963,
|
|
"epoch": 8.224443434802362,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.075388006398907e-05,
|
|
"loss": 4.33,
|
|
"mean_token_accuracy": 0.2536814913153648,
|
|
"num_tokens": 18753897.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.247425222396851,
|
|
"epoch": 8.228986824170832,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.0551247347896e-05,
|
|
"loss": 4.4537,
|
|
"mean_token_accuracy": 0.24927113354206085,
|
|
"num_tokens": 18764849.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.1850786209106445,
|
|
"epoch": 8.2335302135393,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.034906974076449e-05,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.25925577282905576,
|
|
"num_tokens": 18776404.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.238427543640137,
|
|
"epoch": 8.23807360290777,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.014734774144784e-05,
|
|
"loss": 4.4347,
|
|
"mean_token_accuracy": 0.24885546267032624,
|
|
"num_tokens": 18786803.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.256671619415283,
|
|
"epoch": 8.242616992276238,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.994608184767491e-05,
|
|
"loss": 4.4603,
|
|
"mean_token_accuracy": 0.24975856840610505,
|
|
"num_tokens": 18797135.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.2317280769348145,
|
|
"epoch": 8.247160381644706,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 8.97452725560495e-05,
|
|
"loss": 4.3543,
|
|
"mean_token_accuracy": 0.26288229674100877,
|
|
"num_tokens": 18806495.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.235189294815063,
|
|
"epoch": 8.251703771013176,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.954492036204829e-05,
|
|
"loss": 4.4063,
|
|
"mean_token_accuracy": 0.25647003799676893,
|
|
"num_tokens": 18816925.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.236835575103759,
|
|
"epoch": 8.256247160381644,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.934502576002059e-05,
|
|
"loss": 4.3538,
|
|
"mean_token_accuracy": 0.26107853204011916,
|
|
"num_tokens": 18826604.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.2334247589111325,
|
|
"epoch": 8.260790549750114,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.914558924318643e-05,
|
|
"loss": 4.4397,
|
|
"mean_token_accuracy": 0.255289252102375,
|
|
"num_tokens": 18837278.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.2445292472839355,
|
|
"epoch": 8.265333939118582,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.894661130363557e-05,
|
|
"loss": 4.4486,
|
|
"mean_token_accuracy": 0.2527014195919037,
|
|
"num_tokens": 18846588.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.185458850860596,
|
|
"epoch": 8.269877328487052,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.874809243232624e-05,
|
|
"loss": 4.3908,
|
|
"mean_token_accuracy": 0.2678797051310539,
|
|
"num_tokens": 18856345.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.2380904197692875,
|
|
"epoch": 8.27442071785552,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.85500331190841e-05,
|
|
"loss": 4.4763,
|
|
"mean_token_accuracy": 0.2429969996213913,
|
|
"num_tokens": 18867000.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.182608556747437,
|
|
"epoch": 8.27896410722399,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 8.835243385260083e-05,
|
|
"loss": 4.4654,
|
|
"mean_token_accuracy": 0.24813367873430253,
|
|
"num_tokens": 18876925.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.250661754608155,
|
|
"epoch": 8.283507496592458,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 8.815529512043293e-05,
|
|
"loss": 4.4057,
|
|
"mean_token_accuracy": 0.25946542620658875,
|
|
"num_tokens": 18886376.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.219332981109619,
|
|
"epoch": 8.288050885960926,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.795861740900061e-05,
|
|
"loss": 4.3985,
|
|
"mean_token_accuracy": 0.25563151985406873,
|
|
"num_tokens": 18897258.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.263466310501099,
|
|
"epoch": 8.292594275329396,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.77624012035866e-05,
|
|
"loss": 4.4039,
|
|
"mean_token_accuracy": 0.26377025693655015,
|
|
"num_tokens": 18906959.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.240538549423218,
|
|
"epoch": 8.297137664697864,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.756664698833492e-05,
|
|
"loss": 4.4897,
|
|
"mean_token_accuracy": 0.24155495464801788,
|
|
"num_tokens": 18917930.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.288996076583862,
|
|
"epoch": 8.301681054066334,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.737135524624959e-05,
|
|
"loss": 4.4972,
|
|
"mean_token_accuracy": 0.2507754281163216,
|
|
"num_tokens": 18928506.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.236560726165772,
|
|
"epoch": 8.306224443434802,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.717652645919352e-05,
|
|
"loss": 4.3948,
|
|
"mean_token_accuracy": 0.25807899981737137,
|
|
"num_tokens": 18938811.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.189000272750855,
|
|
"epoch": 8.310767832803272,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.698216110788753e-05,
|
|
"loss": 4.3996,
|
|
"mean_token_accuracy": 0.2591201767325401,
|
|
"num_tokens": 18948487.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.215292453765869,
|
|
"epoch": 8.31531122217174,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.67882596719086e-05,
|
|
"loss": 4.4059,
|
|
"mean_token_accuracy": 0.2608529910445213,
|
|
"num_tokens": 18958021.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.224228286743164,
|
|
"epoch": 8.319854611540208,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.659482262968943e-05,
|
|
"loss": 4.3969,
|
|
"mean_token_accuracy": 0.2592949092388153,
|
|
"num_tokens": 18967937.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.236398553848266,
|
|
"epoch": 8.324398000908678,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.640185045851653e-05,
|
|
"loss": 4.4238,
|
|
"mean_token_accuracy": 0.25412029325962066,
|
|
"num_tokens": 18978119.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.2046715259552006,
|
|
"epoch": 8.328941390277146,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.620934363452973e-05,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.2542385131120682,
|
|
"num_tokens": 18989036.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.256027460098267,
|
|
"epoch": 8.333484779645616,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.601730263272034e-05,
|
|
"loss": 4.4658,
|
|
"mean_token_accuracy": 0.25843254327774046,
|
|
"num_tokens": 18999561.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.1840790748596195,
|
|
"epoch": 8.338028169014084,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.58257279269305e-05,
|
|
"loss": 4.3601,
|
|
"mean_token_accuracy": 0.2570444107055664,
|
|
"num_tokens": 19010086.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.159156131744385,
|
|
"epoch": 8.342571558382554,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.563461998985172e-05,
|
|
"loss": 4.3169,
|
|
"mean_token_accuracy": 0.2626626893877983,
|
|
"num_tokens": 19019697.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.27059416770935,
|
|
"epoch": 8.347114947751022,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 8.544397929302389e-05,
|
|
"loss": 4.4622,
|
|
"mean_token_accuracy": 0.2535486787557602,
|
|
"num_tokens": 19030429.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.275853872299194,
|
|
"epoch": 8.351658337119492,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.525380630683389e-05,
|
|
"loss": 4.4232,
|
|
"mean_token_accuracy": 0.25707777589559555,
|
|
"num_tokens": 19041052.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.182751321792603,
|
|
"epoch": 8.35620172648796,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 8.506410150051478e-05,
|
|
"loss": 4.366,
|
|
"mean_token_accuracy": 0.2649271324276924,
|
|
"num_tokens": 19050936.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.187569999694825,
|
|
"epoch": 8.360745115856428,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.487486534214423e-05,
|
|
"loss": 4.358,
|
|
"mean_token_accuracy": 0.25637385845184324,
|
|
"num_tokens": 19060884.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.216471862792969,
|
|
"epoch": 8.365288505224898,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.468609829864366e-05,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.2543789714574814,
|
|
"num_tokens": 19071405.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.235090208053589,
|
|
"epoch": 8.369831894593366,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.449780083577697e-05,
|
|
"loss": 4.3909,
|
|
"mean_token_accuracy": 0.2528301551938057,
|
|
"num_tokens": 19081186.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.2720459461212155,
|
|
"epoch": 8.374375283961836,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.430997341814947e-05,
|
|
"loss": 4.4466,
|
|
"mean_token_accuracy": 0.25407021939754487,
|
|
"num_tokens": 19091042.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.197043943405151,
|
|
"epoch": 8.378918673330304,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 8.41226165092066e-05,
|
|
"loss": 4.4016,
|
|
"mean_token_accuracy": 0.25541582256555556,
|
|
"num_tokens": 19101408.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.19293704032898,
|
|
"epoch": 8.383462062698774,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.393573057123299e-05,
|
|
"loss": 4.3834,
|
|
"mean_token_accuracy": 0.2604291334748268,
|
|
"num_tokens": 19111773.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.268459463119507,
|
|
"epoch": 8.388005452067242,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 8.374931606535099e-05,
|
|
"loss": 4.4372,
|
|
"mean_token_accuracy": 0.26066552847623825,
|
|
"num_tokens": 19121532.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.210270595550537,
|
|
"epoch": 8.392548841435712,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.356337345152002e-05,
|
|
"loss": 4.4414,
|
|
"mean_token_accuracy": 0.24708416908979416,
|
|
"num_tokens": 19131389.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.166284275054932,
|
|
"epoch": 8.39709223080418,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 8.33779031885348e-05,
|
|
"loss": 4.3856,
|
|
"mean_token_accuracy": 0.2531314089894295,
|
|
"num_tokens": 19141304.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.1725029945373535,
|
|
"epoch": 8.401635620172648,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 8.319290573402494e-05,
|
|
"loss": 4.3558,
|
|
"mean_token_accuracy": 0.2540024146437645,
|
|
"num_tokens": 19151039.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.169120407104492,
|
|
"epoch": 8.406179009541118,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.300838154445314e-05,
|
|
"loss": 4.3741,
|
|
"mean_token_accuracy": 0.2564253702759743,
|
|
"num_tokens": 19160909.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.234465169906616,
|
|
"epoch": 8.410722398909586,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.282433107511466e-05,
|
|
"loss": 4.4047,
|
|
"mean_token_accuracy": 0.2557554543018341,
|
|
"num_tokens": 19171906.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.239300155639649,
|
|
"epoch": 8.415265788278056,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.264075478013552e-05,
|
|
"loss": 4.4422,
|
|
"mean_token_accuracy": 0.25529737770557404,
|
|
"num_tokens": 19182688.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.217407369613648,
|
|
"epoch": 8.419809177646524,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.245765311247216e-05,
|
|
"loss": 4.3622,
|
|
"mean_token_accuracy": 0.2584868907928467,
|
|
"num_tokens": 19192429.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.269301128387451,
|
|
"epoch": 8.424352567014994,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.227502652390968e-05,
|
|
"loss": 4.502,
|
|
"mean_token_accuracy": 0.24442875534296035,
|
|
"num_tokens": 19202454.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.3100803852081295,
|
|
"epoch": 8.428895956383462,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.2092875465061e-05,
|
|
"loss": 4.4944,
|
|
"mean_token_accuracy": 0.254664708673954,
|
|
"num_tokens": 19213057.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.170656394958496,
|
|
"epoch": 8.43343934575193,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.191120038536575e-05,
|
|
"loss": 4.3252,
|
|
"mean_token_accuracy": 0.2607087418437004,
|
|
"num_tokens": 19222417.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.2512328147888185,
|
|
"epoch": 8.4379827351204,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.173000173308921e-05,
|
|
"loss": 4.438,
|
|
"mean_token_accuracy": 0.24674825370311737,
|
|
"num_tokens": 19233743.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.300572347640991,
|
|
"epoch": 8.442526124488868,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.154927995532098e-05,
|
|
"loss": 4.5697,
|
|
"mean_token_accuracy": 0.24793995469808577,
|
|
"num_tokens": 19245000.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.296702432632446,
|
|
"epoch": 8.447069513857338,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.136903549797408e-05,
|
|
"loss": 4.4294,
|
|
"mean_token_accuracy": 0.2511420503258705,
|
|
"num_tokens": 19255738.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.259768152236939,
|
|
"epoch": 8.451612903225806,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.118926880578378e-05,
|
|
"loss": 4.4642,
|
|
"mean_token_accuracy": 0.25454760789871217,
|
|
"num_tokens": 19266268.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.372559022903443,
|
|
"epoch": 8.456156292594276,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.100998032230656e-05,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.2511442109942436,
|
|
"num_tokens": 19276721.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.231683969497681,
|
|
"epoch": 8.460699681962744,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.083117048991895e-05,
|
|
"loss": 4.384,
|
|
"mean_token_accuracy": 0.25560591369867325,
|
|
"num_tokens": 19287241.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.223943853378296,
|
|
"epoch": 8.465243071331214,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.065283974981638e-05,
|
|
"loss": 4.4966,
|
|
"mean_token_accuracy": 0.24226788133382798,
|
|
"num_tokens": 19298065.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.284431266784668,
|
|
"epoch": 8.469786460699682,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.047498854201222e-05,
|
|
"loss": 4.4733,
|
|
"mean_token_accuracy": 0.24765658527612686,
|
|
"num_tokens": 19308108.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.218800115585327,
|
|
"epoch": 8.47432985006815,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 8.029761730533678e-05,
|
|
"loss": 4.4462,
|
|
"mean_token_accuracy": 0.25874957740306853,
|
|
"num_tokens": 19317909.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.262027454376221,
|
|
"epoch": 8.47887323943662,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 8.012072647743569e-05,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.2571576550602913,
|
|
"num_tokens": 19328278.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.154941082000732,
|
|
"epoch": 8.483416628805088,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.994431649476974e-05,
|
|
"loss": 4.3947,
|
|
"mean_token_accuracy": 0.25647547245025637,
|
|
"num_tokens": 19338995.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.2232883930206295,
|
|
"epoch": 8.487960018173558,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.976838779261285e-05,
|
|
"loss": 4.4354,
|
|
"mean_token_accuracy": 0.2548922702670097,
|
|
"num_tokens": 19349177.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.344258213043213,
|
|
"epoch": 8.492503407542026,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.959294080505179e-05,
|
|
"loss": 4.4617,
|
|
"mean_token_accuracy": 0.25421054363250734,
|
|
"num_tokens": 19358871.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.256483936309815,
|
|
"epoch": 8.497046796910496,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.941797596498435e-05,
|
|
"loss": 4.4813,
|
|
"mean_token_accuracy": 0.2526631414890289,
|
|
"num_tokens": 19368594.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.2026783466339115,
|
|
"epoch": 8.501590186278964,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.924349370411904e-05,
|
|
"loss": 4.3691,
|
|
"mean_token_accuracy": 0.25401529520750044,
|
|
"num_tokens": 19379973.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.248098087310791,
|
|
"epoch": 8.506133575647432,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.906949445297342e-05,
|
|
"loss": 4.3977,
|
|
"mean_token_accuracy": 0.25969956517219545,
|
|
"num_tokens": 19390762.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.2358564853668215,
|
|
"epoch": 8.510676965015902,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.889597864087336e-05,
|
|
"loss": 4.4578,
|
|
"mean_token_accuracy": 0.260677570104599,
|
|
"num_tokens": 19401918.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.187251186370849,
|
|
"epoch": 8.51522035438437,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.872294669595185e-05,
|
|
"loss": 4.4028,
|
|
"mean_token_accuracy": 0.2571165829896927,
|
|
"num_tokens": 19411973.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.221611499786377,
|
|
"epoch": 8.51976374375284,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.855039904514808e-05,
|
|
"loss": 4.3954,
|
|
"mean_token_accuracy": 0.24927058070898056,
|
|
"num_tokens": 19422888.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.299443054199219,
|
|
"epoch": 8.524307133121308,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.837833611420618e-05,
|
|
"loss": 4.4601,
|
|
"mean_token_accuracy": 0.25554895848035813,
|
|
"num_tokens": 19433611.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.152651643753051,
|
|
"epoch": 8.528850522489778,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.820675832767435e-05,
|
|
"loss": 4.3867,
|
|
"mean_token_accuracy": 0.25867066234350206,
|
|
"num_tokens": 19443712.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.207858276367188,
|
|
"epoch": 8.533393911858246,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.803566610890365e-05,
|
|
"loss": 4.321,
|
|
"mean_token_accuracy": 0.2611874669790268,
|
|
"num_tokens": 19453410.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.190402221679688,
|
|
"epoch": 8.537937301226716,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.786505988004722e-05,
|
|
"loss": 4.3756,
|
|
"mean_token_accuracy": 0.26464569866657256,
|
|
"num_tokens": 19463306.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.260924386978149,
|
|
"epoch": 8.542480690595184,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.769494006205894e-05,
|
|
"loss": 4.4739,
|
|
"mean_token_accuracy": 0.25533499419689176,
|
|
"num_tokens": 19474127.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.303929710388184,
|
|
"epoch": 8.547024079963652,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.752530707469252e-05,
|
|
"loss": 4.4992,
|
|
"mean_token_accuracy": 0.25830257385969163,
|
|
"num_tokens": 19486472.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.21991081237793,
|
|
"epoch": 8.551567469332122,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.73561613365005e-05,
|
|
"loss": 4.4393,
|
|
"mean_token_accuracy": 0.25962396115064623,
|
|
"num_tokens": 19497130.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.204071044921875,
|
|
"epoch": 8.55611085870059,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 7.718750326483321e-05,
|
|
"loss": 4.3507,
|
|
"mean_token_accuracy": 0.2619276776909828,
|
|
"num_tokens": 19507321.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.2814795017242435,
|
|
"epoch": 8.56065424806906,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 7.701933327583766e-05,
|
|
"loss": 4.4851,
|
|
"mean_token_accuracy": 0.252509181201458,
|
|
"num_tokens": 19516244.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.209652709960937,
|
|
"epoch": 8.565197637437528,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.685165178445658e-05,
|
|
"loss": 4.386,
|
|
"mean_token_accuracy": 0.2546782001852989,
|
|
"num_tokens": 19526594.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.186282205581665,
|
|
"epoch": 8.569741026805998,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.668445920442732e-05,
|
|
"loss": 4.3869,
|
|
"mean_token_accuracy": 0.24856673628091813,
|
|
"num_tokens": 19536835.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.188549661636353,
|
|
"epoch": 8.574284416174466,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.651775594828112e-05,
|
|
"loss": 4.4691,
|
|
"mean_token_accuracy": 0.25034874081611636,
|
|
"num_tokens": 19546474.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.297825908660888,
|
|
"epoch": 8.578827805542936,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.635154242734146e-05,
|
|
"loss": 4.4895,
|
|
"mean_token_accuracy": 0.24505223631858825,
|
|
"num_tokens": 19556604.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.166006946563721,
|
|
"epoch": 8.583371194911404,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.618581905172385e-05,
|
|
"loss": 4.3724,
|
|
"mean_token_accuracy": 0.2559832289814949,
|
|
"num_tokens": 19566824.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.214518690109253,
|
|
"epoch": 8.587914584279872,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.60205862303342e-05,
|
|
"loss": 4.3603,
|
|
"mean_token_accuracy": 0.25817672461271285,
|
|
"num_tokens": 19576917.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.24187388420105,
|
|
"epoch": 8.592457973648342,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.585584437086806e-05,
|
|
"loss": 4.4169,
|
|
"mean_token_accuracy": 0.25363878905773163,
|
|
"num_tokens": 19587651.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.231191968917846,
|
|
"epoch": 8.59700136301681,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.569159387980958e-05,
|
|
"loss": 4.4121,
|
|
"mean_token_accuracy": 0.25634774267673494,
|
|
"num_tokens": 19598421.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.15672345161438,
|
|
"epoch": 8.60154475238528,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.552783516243056e-05,
|
|
"loss": 4.353,
|
|
"mean_token_accuracy": 0.2655596613883972,
|
|
"num_tokens": 19608715.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.234029722213745,
|
|
"epoch": 8.606088141753748,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.536456862278936e-05,
|
|
"loss": 4.4352,
|
|
"mean_token_accuracy": 0.25343277454376223,
|
|
"num_tokens": 19619676.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.262057161331176,
|
|
"epoch": 8.610631531122218,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.520179466372988e-05,
|
|
"loss": 4.4414,
|
|
"mean_token_accuracy": 0.2569874331355095,
|
|
"num_tokens": 19630837.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.167800807952881,
|
|
"epoch": 8.615174920490686,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.503951368688068e-05,
|
|
"loss": 4.2984,
|
|
"mean_token_accuracy": 0.26397351175546646,
|
|
"num_tokens": 19640607.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.233308506011963,
|
|
"epoch": 8.619718309859154,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.487772609265401e-05,
|
|
"loss": 4.4819,
|
|
"mean_token_accuracy": 0.25315805822610854,
|
|
"num_tokens": 19651354.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.200176000595093,
|
|
"epoch": 8.624261699227624,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 7.471643228024466e-05,
|
|
"loss": 4.3714,
|
|
"mean_token_accuracy": 0.2602162599563599,
|
|
"num_tokens": 19661266.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.225366926193237,
|
|
"epoch": 8.628805088596092,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.455563264762903e-05,
|
|
"loss": 4.4265,
|
|
"mean_token_accuracy": 0.25524410605430603,
|
|
"num_tokens": 19671275.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 8.628805088596092,
|
|
"eval_entropy": 5.11975522706675,
|
|
"eval_loss": 5.653097629547119,
|
|
"eval_mean_token_accuracy": 0.17652752248700274,
|
|
"eval_num_tokens": 19671275.0,
|
|
"eval_runtime": 2.0105,
|
|
"eval_samples_per_second": 1708.49,
|
|
"eval_steps_per_second": 213.872,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.20487699508667,
|
|
"epoch": 8.633348477964562,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.439532759156422e-05,
|
|
"loss": 4.3035,
|
|
"mean_token_accuracy": 0.2675573259592056,
|
|
"num_tokens": 19682360.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.25970139503479,
|
|
"epoch": 8.63789186733303,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.423551750758711e-05,
|
|
"loss": 4.4478,
|
|
"mean_token_accuracy": 0.2524769514799118,
|
|
"num_tokens": 19692393.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.2202863693237305,
|
|
"epoch": 8.6424352567015,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.407620279001315e-05,
|
|
"loss": 4.3612,
|
|
"mean_token_accuracy": 0.2596442461013794,
|
|
"num_tokens": 19702045.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.28242974281311,
|
|
"epoch": 8.646978646069968,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.391738383193553e-05,
|
|
"loss": 4.4928,
|
|
"mean_token_accuracy": 0.24726963937282562,
|
|
"num_tokens": 19713038.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.181081104278564,
|
|
"epoch": 8.651522035438438,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.375906102522423e-05,
|
|
"loss": 4.4283,
|
|
"mean_token_accuracy": 0.2537392690777779,
|
|
"num_tokens": 19723708.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.250782108306884,
|
|
"epoch": 8.656065424806906,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.360123476052519e-05,
|
|
"loss": 4.4985,
|
|
"mean_token_accuracy": 0.24797665029764177,
|
|
"num_tokens": 19735271.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.287685441970825,
|
|
"epoch": 8.660608814175374,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.344390542725882e-05,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.24982582330703734,
|
|
"num_tokens": 19744593.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.18670129776001,
|
|
"epoch": 8.665152203543844,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.328707341361975e-05,
|
|
"loss": 4.3599,
|
|
"mean_token_accuracy": 0.2609484940767288,
|
|
"num_tokens": 19754600.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.229602670669555,
|
|
"epoch": 8.669695592912312,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.313073910657535e-05,
|
|
"loss": 4.4653,
|
|
"mean_token_accuracy": 0.2549873560667038,
|
|
"num_tokens": 19766376.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.207068729400635,
|
|
"epoch": 8.674238982280782,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.297490289186495e-05,
|
|
"loss": 4.3555,
|
|
"mean_token_accuracy": 0.25822059214115145,
|
|
"num_tokens": 19776015.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.208258104324341,
|
|
"epoch": 8.67878237164925,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.281956515399891e-05,
|
|
"loss": 4.3642,
|
|
"mean_token_accuracy": 0.2564428150653839,
|
|
"num_tokens": 19785794.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.259734344482422,
|
|
"epoch": 8.68332576101772,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 7.266472627625776e-05,
|
|
"loss": 4.396,
|
|
"mean_token_accuracy": 0.2570520922541618,
|
|
"num_tokens": 19795202.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.242871713638306,
|
|
"epoch": 8.687869150386188,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 7.251038664069094e-05,
|
|
"loss": 4.4175,
|
|
"mean_token_accuracy": 0.24852036088705062,
|
|
"num_tokens": 19805623.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.243520975112915,
|
|
"epoch": 8.692412539754656,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.235654662811618e-05,
|
|
"loss": 4.4111,
|
|
"mean_token_accuracy": 0.2537427082657814,
|
|
"num_tokens": 19815792.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.2155543804168705,
|
|
"epoch": 8.696955929123126,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 7.22032066181184e-05,
|
|
"loss": 4.3962,
|
|
"mean_token_accuracy": 0.253386464715004,
|
|
"num_tokens": 19825460.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.202692461013794,
|
|
"epoch": 8.701499318491594,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.205036698904888e-05,
|
|
"loss": 4.4484,
|
|
"mean_token_accuracy": 0.25165548771619795,
|
|
"num_tokens": 19836630.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.240066719055176,
|
|
"epoch": 8.706042707860064,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 7.18980281180242e-05,
|
|
"loss": 4.4042,
|
|
"mean_token_accuracy": 0.2588438853621483,
|
|
"num_tokens": 19846445.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.187673330307007,
|
|
"epoch": 8.710586097228532,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.174619038092537e-05,
|
|
"loss": 4.3945,
|
|
"mean_token_accuracy": 0.26222456693649293,
|
|
"num_tokens": 19856738.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.193586635589599,
|
|
"epoch": 8.715129486597002,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.159485415239685e-05,
|
|
"loss": 4.4311,
|
|
"mean_token_accuracy": 0.24731139689683915,
|
|
"num_tokens": 19866134.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.1923255443573,
|
|
"epoch": 8.71967287596547,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.144401980584584e-05,
|
|
"loss": 4.3383,
|
|
"mean_token_accuracy": 0.26650340259075167,
|
|
"num_tokens": 19877384.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.25066351890564,
|
|
"epoch": 8.72421626533394,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.129368771344108e-05,
|
|
"loss": 4.3771,
|
|
"mean_token_accuracy": 0.25965592116117475,
|
|
"num_tokens": 19887836.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.209628772735596,
|
|
"epoch": 8.728759654702408,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 7.114385824611203e-05,
|
|
"loss": 4.4151,
|
|
"mean_token_accuracy": 0.2571896627545357,
|
|
"num_tokens": 19897323.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.219085597991944,
|
|
"epoch": 8.733303044070876,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.099453177354797e-05,
|
|
"loss": 4.4432,
|
|
"mean_token_accuracy": 0.24737028330564498,
|
|
"num_tokens": 19908959.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.182817506790161,
|
|
"epoch": 8.737846433439346,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 7.084570866419727e-05,
|
|
"loss": 4.3518,
|
|
"mean_token_accuracy": 0.2701904371380806,
|
|
"num_tokens": 19918918.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.230889225006104,
|
|
"epoch": 8.742389822807814,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 7.069738928526597e-05,
|
|
"loss": 4.4752,
|
|
"mean_token_accuracy": 0.2594154953956604,
|
|
"num_tokens": 19928804.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.232018661499024,
|
|
"epoch": 8.746933212176284,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.054957400271751e-05,
|
|
"loss": 4.4313,
|
|
"mean_token_accuracy": 0.25021475553512573,
|
|
"num_tokens": 19938706.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.173541307449341,
|
|
"epoch": 8.751476601544752,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 7.040226318127142e-05,
|
|
"loss": 4.3288,
|
|
"mean_token_accuracy": 0.26788181364536284,
|
|
"num_tokens": 19948548.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.144832801818848,
|
|
"epoch": 8.756019990913222,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.025545718440245e-05,
|
|
"loss": 4.3731,
|
|
"mean_token_accuracy": 0.25677865743637085,
|
|
"num_tokens": 19958568.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.222461748123169,
|
|
"epoch": 8.76056338028169,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.01091563743398e-05,
|
|
"loss": 4.4216,
|
|
"mean_token_accuracy": 0.2535837396979332,
|
|
"num_tokens": 19968889.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.226753616333008,
|
|
"epoch": 8.76510676965016,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.996336111206619e-05,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.25026918649673463,
|
|
"num_tokens": 19978813.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.200356388092041,
|
|
"epoch": 8.769650159018628,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.981807175731698e-05,
|
|
"loss": 4.3645,
|
|
"mean_token_accuracy": 0.2586676299571991,
|
|
"num_tokens": 19988689.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.23523359298706,
|
|
"epoch": 8.774193548387096,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.967328866857915e-05,
|
|
"loss": 4.4047,
|
|
"mean_token_accuracy": 0.25654312819242475,
|
|
"num_tokens": 19999006.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.243108701705933,
|
|
"epoch": 8.778736937755566,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 6.952901220309063e-05,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.25419641435146334,
|
|
"num_tokens": 20009038.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.274879455566406,
|
|
"epoch": 8.783280327124034,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.938524271683918e-05,
|
|
"loss": 4.5269,
|
|
"mean_token_accuracy": 0.24386754930019378,
|
|
"num_tokens": 20020908.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.298810291290283,
|
|
"epoch": 8.787823716492504,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.924198056456185e-05,
|
|
"loss": 4.4472,
|
|
"mean_token_accuracy": 0.25290532857179643,
|
|
"num_tokens": 20032441.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.283793115615845,
|
|
"epoch": 8.792367105860972,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.90992260997436e-05,
|
|
"loss": 4.5106,
|
|
"mean_token_accuracy": 0.24532811790704728,
|
|
"num_tokens": 20043772.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.197098731994629,
|
|
"epoch": 8.796910495229442,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 6.895697967461699e-05,
|
|
"loss": 4.3945,
|
|
"mean_token_accuracy": 0.2602959305047989,
|
|
"num_tokens": 20053058.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.1930260181427,
|
|
"epoch": 8.80145388459791,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.881524164016087e-05,
|
|
"loss": 4.478,
|
|
"mean_token_accuracy": 0.24866838306188582,
|
|
"num_tokens": 20064289.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.201801681518555,
|
|
"epoch": 8.805997273966378,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.867401234609985e-05,
|
|
"loss": 4.3992,
|
|
"mean_token_accuracy": 0.26191624701023103,
|
|
"num_tokens": 20074813.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.238270092010498,
|
|
"epoch": 8.810540663334848,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.853329214090304e-05,
|
|
"loss": 4.4229,
|
|
"mean_token_accuracy": 0.25312917232513427,
|
|
"num_tokens": 20085019.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.279025030136109,
|
|
"epoch": 8.815084052703316,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.839308137178363e-05,
|
|
"loss": 4.4494,
|
|
"mean_token_accuracy": 0.2584308236837387,
|
|
"num_tokens": 20094773.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.222107696533203,
|
|
"epoch": 8.819627442071786,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.825338038469778e-05,
|
|
"loss": 4.3836,
|
|
"mean_token_accuracy": 0.2547050669789314,
|
|
"num_tokens": 20105612.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.204859113693237,
|
|
"epoch": 8.824170831440254,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.811418952434375e-05,
|
|
"loss": 4.4204,
|
|
"mean_token_accuracy": 0.25361269563436506,
|
|
"num_tokens": 20117032.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.262170839309692,
|
|
"epoch": 8.828714220808724,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.797550913416111e-05,
|
|
"loss": 4.3992,
|
|
"mean_token_accuracy": 0.2600389778614044,
|
|
"num_tokens": 20127539.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.227006483078003,
|
|
"epoch": 8.833257610177192,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.783733955633004e-05,
|
|
"loss": 4.4265,
|
|
"mean_token_accuracy": 0.2557223752140999,
|
|
"num_tokens": 20138430.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.185586261749267,
|
|
"epoch": 8.837800999545662,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.769968113177016e-05,
|
|
"loss": 4.3995,
|
|
"mean_token_accuracy": 0.24679649174213408,
|
|
"num_tokens": 20149261.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.232632637023926,
|
|
"epoch": 8.84234438891413,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.756253420013996e-05,
|
|
"loss": 4.3755,
|
|
"mean_token_accuracy": 0.25613555908203123,
|
|
"num_tokens": 20158617.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.205956935882568,
|
|
"epoch": 8.846887778282598,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.742589909983586e-05,
|
|
"loss": 4.4169,
|
|
"mean_token_accuracy": 0.2557495623826981,
|
|
"num_tokens": 20169028.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.207171392440796,
|
|
"epoch": 8.851431167651068,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.728977616799144e-05,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.2633624792098999,
|
|
"num_tokens": 20178926.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.214498376846313,
|
|
"epoch": 8.855974557019536,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.715416574047645e-05,
|
|
"loss": 4.3508,
|
|
"mean_token_accuracy": 0.2672609344124794,
|
|
"num_tokens": 20188602.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.181071853637695,
|
|
"epoch": 8.860517946388006,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.701906815189613e-05,
|
|
"loss": 4.3789,
|
|
"mean_token_accuracy": 0.25686743557453157,
|
|
"num_tokens": 20198623.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.256879663467407,
|
|
"epoch": 8.865061335756474,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.688448373559037e-05,
|
|
"loss": 4.4267,
|
|
"mean_token_accuracy": 0.25361049473285674,
|
|
"num_tokens": 20208627.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.268274211883545,
|
|
"epoch": 8.869604725124944,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.675041282363286e-05,
|
|
"loss": 4.4482,
|
|
"mean_token_accuracy": 0.2538605034351349,
|
|
"num_tokens": 20218920.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.223486232757568,
|
|
"epoch": 8.874148114493412,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.661685574683021e-05,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.2548517256975174,
|
|
"num_tokens": 20229938.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.2245128631591795,
|
|
"epoch": 8.87869150386188,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.648381283472128e-05,
|
|
"loss": 4.43,
|
|
"mean_token_accuracy": 0.25186797231435776,
|
|
"num_tokens": 20240135.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.187893581390381,
|
|
"epoch": 8.88323489323035,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.635128441557615e-05,
|
|
"loss": 4.4218,
|
|
"mean_token_accuracy": 0.2564263105392456,
|
|
"num_tokens": 20250239.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.231339788436889,
|
|
"epoch": 8.887778282598818,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.621927081639563e-05,
|
|
"loss": 4.4753,
|
|
"mean_token_accuracy": 0.25219457149505614,
|
|
"num_tokens": 20260498.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.237685251235962,
|
|
"epoch": 8.892321671967288,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.608777236291003e-05,
|
|
"loss": 4.4851,
|
|
"mean_token_accuracy": 0.2495596468448639,
|
|
"num_tokens": 20271830.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.228261232376099,
|
|
"epoch": 8.896865061335756,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.59567893795788e-05,
|
|
"loss": 4.451,
|
|
"mean_token_accuracy": 0.2535568356513977,
|
|
"num_tokens": 20282519.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.1761415004730225,
|
|
"epoch": 8.901408450704226,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.58263221895894e-05,
|
|
"loss": 4.3703,
|
|
"mean_token_accuracy": 0.2555522441864014,
|
|
"num_tokens": 20292285.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.270341873168945,
|
|
"epoch": 8.905951840072694,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.569637111485666e-05,
|
|
"loss": 4.4327,
|
|
"mean_token_accuracy": 0.2536139041185379,
|
|
"num_tokens": 20302191.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.232021999359131,
|
|
"epoch": 8.910495229441164,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.556693647602191e-05,
|
|
"loss": 4.3993,
|
|
"mean_token_accuracy": 0.26209924519062044,
|
|
"num_tokens": 20313839.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.18741545677185,
|
|
"epoch": 8.915038618809632,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.543801859245229e-05,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.24978577792644502,
|
|
"num_tokens": 20324016.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.236614084243774,
|
|
"epoch": 8.9195820081781,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.530961778223988e-05,
|
|
"loss": 4.4007,
|
|
"mean_token_accuracy": 0.2559172555804253,
|
|
"num_tokens": 20334085.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.267110443115234,
|
|
"epoch": 8.92412539754657,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.518173436220089e-05,
|
|
"loss": 4.4847,
|
|
"mean_token_accuracy": 0.2506674647331238,
|
|
"num_tokens": 20344647.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.282026481628418,
|
|
"epoch": 8.928668786915038,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.505436864787489e-05,
|
|
"loss": 4.4589,
|
|
"mean_token_accuracy": 0.2543804332613945,
|
|
"num_tokens": 20354697.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.297132396697998,
|
|
"epoch": 8.933212176283508,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.49275209535242e-05,
|
|
"loss": 4.4874,
|
|
"mean_token_accuracy": 0.24608783423900604,
|
|
"num_tokens": 20365652.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.251209402084351,
|
|
"epoch": 8.937755565651976,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.480119159213288e-05,
|
|
"loss": 4.4255,
|
|
"mean_token_accuracy": 0.2523648291826248,
|
|
"num_tokens": 20375629.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.281072759628296,
|
|
"epoch": 8.942298955020446,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 6.467538087540606e-05,
|
|
"loss": 4.4785,
|
|
"mean_token_accuracy": 0.24840882420539856,
|
|
"num_tokens": 20385361.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.1643671035766605,
|
|
"epoch": 8.946842344388914,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.455008911376914e-05,
|
|
"loss": 4.3488,
|
|
"mean_token_accuracy": 0.26244045197963717,
|
|
"num_tokens": 20394900.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.169113254547119,
|
|
"epoch": 8.951385733757384,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.442531661636715e-05,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.25792319923639295,
|
|
"num_tokens": 20405476.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.136023902893067,
|
|
"epoch": 8.955929123125852,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.430106369106381e-05,
|
|
"loss": 4.3495,
|
|
"mean_token_accuracy": 0.2627858996391296,
|
|
"num_tokens": 20414536.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.311545467376709,
|
|
"epoch": 8.96047251249432,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.417733064444083e-05,
|
|
"loss": 4.5346,
|
|
"mean_token_accuracy": 0.24159520566463472,
|
|
"num_tokens": 20426046.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.175200462341309,
|
|
"epoch": 8.96501590186279,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.405411778179722e-05,
|
|
"loss": 4.3493,
|
|
"mean_token_accuracy": 0.2689951419830322,
|
|
"num_tokens": 20435788.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.202020740509033,
|
|
"epoch": 8.969559291231258,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.39314254071485e-05,
|
|
"loss": 4.3783,
|
|
"mean_token_accuracy": 0.26226664036512376,
|
|
"num_tokens": 20445166.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.18081374168396,
|
|
"epoch": 8.974102680599728,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.380925382322584e-05,
|
|
"loss": 4.4124,
|
|
"mean_token_accuracy": 0.25862699598073957,
|
|
"num_tokens": 20455034.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.230588912963867,
|
|
"epoch": 8.978646069968196,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.368760333147555e-05,
|
|
"loss": 4.4151,
|
|
"mean_token_accuracy": 0.2576820343732834,
|
|
"num_tokens": 20465092.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.24817476272583,
|
|
"epoch": 8.983189459336666,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.356647423205809e-05,
|
|
"loss": 4.4265,
|
|
"mean_token_accuracy": 0.253504478931427,
|
|
"num_tokens": 20474793.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.197743892669678,
|
|
"epoch": 8.987732848705134,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.344586682384748e-05,
|
|
"loss": 4.4396,
|
|
"mean_token_accuracy": 0.2490181177854538,
|
|
"num_tokens": 20486201.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.228555297851562,
|
|
"epoch": 8.992276238073602,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.332578140443045e-05,
|
|
"loss": 4.4914,
|
|
"mean_token_accuracy": 0.24439710527658462,
|
|
"num_tokens": 20497264.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.194231462478638,
|
|
"epoch": 8.996819627442072,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 6.320621827010595e-05,
|
|
"loss": 4.3826,
|
|
"mean_token_accuracy": 0.2616214632987976,
|
|
"num_tokens": 20507279.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.212401602003309,
|
|
"epoch": 9.000908677873694,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.308717771588407e-05,
|
|
"loss": 4.3644,
|
|
"mean_token_accuracy": 0.2581610447830624,
|
|
"num_tokens": 20516472.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.2000823497772215,
|
|
"epoch": 9.005452067242162,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.296866003548555e-05,
|
|
"loss": 4.4164,
|
|
"mean_token_accuracy": 0.2542212620377541,
|
|
"num_tokens": 20527251.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.221954250335694,
|
|
"epoch": 9.009995456610632,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.285066552134091e-05,
|
|
"loss": 4.3507,
|
|
"mean_token_accuracy": 0.262487056851387,
|
|
"num_tokens": 20537810.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.193523359298706,
|
|
"epoch": 9.0145388459791,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.273319446459002e-05,
|
|
"loss": 4.3679,
|
|
"mean_token_accuracy": 0.2576700493693352,
|
|
"num_tokens": 20548000.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.221700859069824,
|
|
"epoch": 9.01908223534757,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.261624715508095e-05,
|
|
"loss": 4.3744,
|
|
"mean_token_accuracy": 0.2606788232922554,
|
|
"num_tokens": 20557798.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.236097288131714,
|
|
"epoch": 9.023625624716038,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.249982388136958e-05,
|
|
"loss": 4.4126,
|
|
"mean_token_accuracy": 0.26334539502859117,
|
|
"num_tokens": 20568172.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.227021741867065,
|
|
"epoch": 9.028169014084508,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.238392493071873e-05,
|
|
"loss": 4.3842,
|
|
"mean_token_accuracy": 0.2636853262782097,
|
|
"num_tokens": 20578306.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.25737566947937,
|
|
"epoch": 9.032712403452976,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.226855058909762e-05,
|
|
"loss": 4.3793,
|
|
"mean_token_accuracy": 0.2644745856523514,
|
|
"num_tokens": 20588820.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.198901176452637,
|
|
"epoch": 9.037255792821444,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.215370114118092e-05,
|
|
"loss": 4.3262,
|
|
"mean_token_accuracy": 0.26528919488191605,
|
|
"num_tokens": 20598907.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.204660081863404,
|
|
"epoch": 9.041799182189914,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.203937687034822e-05,
|
|
"loss": 4.4184,
|
|
"mean_token_accuracy": 0.2534408584237099,
|
|
"num_tokens": 20609023.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.255126142501831,
|
|
"epoch": 9.046342571558382,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.192557805868328e-05,
|
|
"loss": 4.4001,
|
|
"mean_token_accuracy": 0.25857075601816176,
|
|
"num_tokens": 20619868.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.27453203201294,
|
|
"epoch": 9.050885960926852,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.181230498697343e-05,
|
|
"loss": 4.4602,
|
|
"mean_token_accuracy": 0.25735954791307447,
|
|
"num_tokens": 20631674.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.267403507232666,
|
|
"epoch": 9.05542935029532,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.169955793470859e-05,
|
|
"loss": 4.4022,
|
|
"mean_token_accuracy": 0.2630584120750427,
|
|
"num_tokens": 20641878.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.264423131942749,
|
|
"epoch": 9.05997273966379,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.158733718008102e-05,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.25045574754476546,
|
|
"num_tokens": 20652663.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.190480804443359,
|
|
"epoch": 9.064516129032258,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.147564299998422e-05,
|
|
"loss": 4.3873,
|
|
"mean_token_accuracy": 0.262087582051754,
|
|
"num_tokens": 20662651.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.184243679046631,
|
|
"epoch": 9.069059518400726,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.136447567001251e-05,
|
|
"loss": 4.3562,
|
|
"mean_token_accuracy": 0.263031879067421,
|
|
"num_tokens": 20672531.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.213598108291626,
|
|
"epoch": 9.073602907769196,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 6.125383546446015e-05,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.256279955804348,
|
|
"num_tokens": 20681944.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.214720153808594,
|
|
"epoch": 9.078146297137664,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.114372265632095e-05,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.2585111245512962,
|
|
"num_tokens": 20692491.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.182323312759399,
|
|
"epoch": 9.082689686506134,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.1034137517287256e-05,
|
|
"loss": 4.3616,
|
|
"mean_token_accuracy": 0.26114430725574495,
|
|
"num_tokens": 20702649.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 9.082689686506134,
|
|
"eval_entropy": 5.107128910685694,
|
|
"eval_loss": 5.655239105224609,
|
|
"eval_mean_token_accuracy": 0.17679712422365365,
|
|
"eval_num_tokens": 20702649.0,
|
|
"eval_runtime": 1.991,
|
|
"eval_samples_per_second": 1725.243,
|
|
"eval_steps_per_second": 215.969,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.2076622486114506,
|
|
"epoch": 9.087233075874602,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.092508031774954e-05,
|
|
"loss": 4.3497,
|
|
"mean_token_accuracy": 0.26091153770685194,
|
|
"num_tokens": 20712543.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.203850650787354,
|
|
"epoch": 9.091776465243072,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.081655132679551e-05,
|
|
"loss": 4.3424,
|
|
"mean_token_accuracy": 0.261168384552002,
|
|
"num_tokens": 20723500.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.198302316665649,
|
|
"epoch": 9.09631985461154,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.07085508122098e-05,
|
|
"loss": 4.3534,
|
|
"mean_token_accuracy": 0.26363862603902816,
|
|
"num_tokens": 20734141.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.224570322036743,
|
|
"epoch": 9.10086324398001,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.060107904047288e-05,
|
|
"loss": 4.3328,
|
|
"mean_token_accuracy": 0.25850091129541397,
|
|
"num_tokens": 20743486.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.257355690002441,
|
|
"epoch": 9.105406633348478,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.0494136276760635e-05,
|
|
"loss": 4.39,
|
|
"mean_token_accuracy": 0.2677877306938171,
|
|
"num_tokens": 20753577.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.243933868408203,
|
|
"epoch": 9.109950022716946,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.03877227849437e-05,
|
|
"loss": 4.3625,
|
|
"mean_token_accuracy": 0.25641720443964006,
|
|
"num_tokens": 20763829.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.184901857376099,
|
|
"epoch": 9.114493412085416,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.028183882758685e-05,
|
|
"loss": 4.3073,
|
|
"mean_token_accuracy": 0.26530774831771853,
|
|
"num_tokens": 20773450.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.1816511154174805,
|
|
"epoch": 9.119036801453884,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.01764846659482e-05,
|
|
"loss": 4.3675,
|
|
"mean_token_accuracy": 0.2567674249410629,
|
|
"num_tokens": 20783574.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.197167682647705,
|
|
"epoch": 9.123580190822354,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.0071660559978645e-05,
|
|
"loss": 4.3455,
|
|
"mean_token_accuracy": 0.2591617897152901,
|
|
"num_tokens": 20793130.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.22406325340271,
|
|
"epoch": 9.128123580190822,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.996736676832124e-05,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.2474212184548378,
|
|
"num_tokens": 20802766.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.191367721557617,
|
|
"epoch": 9.132666969559292,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.9863603548310646e-05,
|
|
"loss": 4.3901,
|
|
"mean_token_accuracy": 0.25761930346488954,
|
|
"num_tokens": 20812620.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.231948804855347,
|
|
"epoch": 9.13721035892776,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.976037115597215e-05,
|
|
"loss": 4.382,
|
|
"mean_token_accuracy": 0.2633367836475372,
|
|
"num_tokens": 20822401.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.278076314926148,
|
|
"epoch": 9.141753748296228,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.965766984602149e-05,
|
|
"loss": 4.4585,
|
|
"mean_token_accuracy": 0.24974486827850342,
|
|
"num_tokens": 20832122.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.273944091796875,
|
|
"epoch": 9.146297137664698,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.9555499871863955e-05,
|
|
"loss": 4.4202,
|
|
"mean_token_accuracy": 0.2575241357088089,
|
|
"num_tokens": 20843223.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.2227349281311035,
|
|
"epoch": 9.150840527033166,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.9453861485593794e-05,
|
|
"loss": 4.3939,
|
|
"mean_token_accuracy": 0.2575162410736084,
|
|
"num_tokens": 20854452.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.25346097946167,
|
|
"epoch": 9.155383916401636,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.9352754937993586e-05,
|
|
"loss": 4.4,
|
|
"mean_token_accuracy": 0.25986931920051576,
|
|
"num_tokens": 20864668.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.202000665664673,
|
|
"epoch": 9.159927305770104,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.925218047853371e-05,
|
|
"loss": 4.3681,
|
|
"mean_token_accuracy": 0.2664692625403404,
|
|
"num_tokens": 20874667.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.2086732387542725,
|
|
"epoch": 9.164470695138574,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.9152138355371646e-05,
|
|
"loss": 4.4451,
|
|
"mean_token_accuracy": 0.2573858544230461,
|
|
"num_tokens": 20884578.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.226881074905395,
|
|
"epoch": 9.169014084507042,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.9052628815351416e-05,
|
|
"loss": 4.384,
|
|
"mean_token_accuracy": 0.26110417395830154,
|
|
"num_tokens": 20894239.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.277196216583252,
|
|
"epoch": 9.173557473875512,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.8953652104002825e-05,
|
|
"loss": 4.3491,
|
|
"mean_token_accuracy": 0.26434845179319383,
|
|
"num_tokens": 20904960.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.228878927230835,
|
|
"epoch": 9.17810086324398,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.885520846554119e-05,
|
|
"loss": 4.3962,
|
|
"mean_token_accuracy": 0.25515228062868117,
|
|
"num_tokens": 20915215.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.233779621124268,
|
|
"epoch": 9.182644252612448,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.875729814286634e-05,
|
|
"loss": 4.3691,
|
|
"mean_token_accuracy": 0.2567964047193527,
|
|
"num_tokens": 20925112.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.204053163528442,
|
|
"epoch": 9.187187641980918,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.865992137756225e-05,
|
|
"loss": 4.3697,
|
|
"mean_token_accuracy": 0.2613224759697914,
|
|
"num_tokens": 20935393.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.2038086414337155,
|
|
"epoch": 9.191731031349386,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.856307840989642e-05,
|
|
"loss": 4.4194,
|
|
"mean_token_accuracy": 0.25171790421009066,
|
|
"num_tokens": 20944876.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.205584287643433,
|
|
"epoch": 9.196274420717856,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.846676947881927e-05,
|
|
"loss": 4.3848,
|
|
"mean_token_accuracy": 0.26110183000564574,
|
|
"num_tokens": 20955115.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.2102972030639645,
|
|
"epoch": 9.200817810086324,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.8370994821963514e-05,
|
|
"loss": 4.4187,
|
|
"mean_token_accuracy": 0.25984649956226347,
|
|
"num_tokens": 20964914.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.218361043930054,
|
|
"epoch": 9.205361199454794,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.8275754675643565e-05,
|
|
"loss": 4.3491,
|
|
"mean_token_accuracy": 0.2610879808664322,
|
|
"num_tokens": 20975768.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.179225921630859,
|
|
"epoch": 9.209904588823262,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.8181049274855013e-05,
|
|
"loss": 4.3445,
|
|
"mean_token_accuracy": 0.2621371492743492,
|
|
"num_tokens": 20985496.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.268178081512451,
|
|
"epoch": 9.214447978191732,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.80868788532741e-05,
|
|
"loss": 4.4368,
|
|
"mean_token_accuracy": 0.2563245490193367,
|
|
"num_tokens": 20995243.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.187670421600342,
|
|
"epoch": 9.2189913675602,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.7993243643256854e-05,
|
|
"loss": 4.4234,
|
|
"mean_token_accuracy": 0.25567215383052827,
|
|
"num_tokens": 21005768.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.247197580337525,
|
|
"epoch": 9.223534756928668,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.790014387583895e-05,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.24975521713495255,
|
|
"num_tokens": 21015615.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.278665208816529,
|
|
"epoch": 9.228078146297138,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.780757978073474e-05,
|
|
"loss": 4.456,
|
|
"mean_token_accuracy": 0.2546222761273384,
|
|
"num_tokens": 21026482.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.244924926757813,
|
|
"epoch": 9.232621535665606,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.771555158633699e-05,
|
|
"loss": 4.4647,
|
|
"mean_token_accuracy": 0.25331505089998246,
|
|
"num_tokens": 21037709.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.175563049316406,
|
|
"epoch": 9.237164925034076,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.762405951971607e-05,
|
|
"loss": 4.2947,
|
|
"mean_token_accuracy": 0.27216711193323134,
|
|
"num_tokens": 21047306.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.193836355209351,
|
|
"epoch": 9.241708314402544,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.75331038066196e-05,
|
|
"loss": 4.356,
|
|
"mean_token_accuracy": 0.27288040369749067,
|
|
"num_tokens": 21057271.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.263077688217163,
|
|
"epoch": 9.246251703771014,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.744268467147174e-05,
|
|
"loss": 4.4038,
|
|
"mean_token_accuracy": 0.25779581516981126,
|
|
"num_tokens": 21066404.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.219763088226318,
|
|
"epoch": 9.250795093139482,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.7352802337372724e-05,
|
|
"loss": 4.377,
|
|
"mean_token_accuracy": 0.25186321288347246,
|
|
"num_tokens": 21077101.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.347354412078857,
|
|
"epoch": 9.25533848250795,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.726345702609824e-05,
|
|
"loss": 4.5166,
|
|
"mean_token_accuracy": 0.24541837126016616,
|
|
"num_tokens": 21087640.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.156579971313477,
|
|
"epoch": 9.25988187187642,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.7174648958099e-05,
|
|
"loss": 4.2895,
|
|
"mean_token_accuracy": 0.26307797580957415,
|
|
"num_tokens": 21098394.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.24271330833435,
|
|
"epoch": 9.264425261244888,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.7086378352500096e-05,
|
|
"loss": 4.4179,
|
|
"mean_token_accuracy": 0.2526425182819366,
|
|
"num_tokens": 21108949.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.230097007751465,
|
|
"epoch": 9.268968650613358,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.6998645427100456e-05,
|
|
"loss": 4.4277,
|
|
"mean_token_accuracy": 0.25788921415805816,
|
|
"num_tokens": 21119204.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.211864614486695,
|
|
"epoch": 9.273512039981826,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.691145039837231e-05,
|
|
"loss": 4.3361,
|
|
"mean_token_accuracy": 0.2648081734776497,
|
|
"num_tokens": 21128959.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.217974853515625,
|
|
"epoch": 9.278055429350296,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.682479348146081e-05,
|
|
"loss": 4.3589,
|
|
"mean_token_accuracy": 0.2637791097164154,
|
|
"num_tokens": 21139465.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.2602463722229,
|
|
"epoch": 9.282598818718764,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.673867489018322e-05,
|
|
"loss": 4.3994,
|
|
"mean_token_accuracy": 0.25979944318532944,
|
|
"num_tokens": 21150871.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.224378252029419,
|
|
"epoch": 9.287142208087234,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.6653094837028626e-05,
|
|
"loss": 4.3667,
|
|
"mean_token_accuracy": 0.2631355568766594,
|
|
"num_tokens": 21161905.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.143555736541748,
|
|
"epoch": 9.291685597455702,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.6568053533157275e-05,
|
|
"loss": 4.34,
|
|
"mean_token_accuracy": 0.2646746292710304,
|
|
"num_tokens": 21172367.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.2840334415435795,
|
|
"epoch": 9.29622898682417,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.648355118840018e-05,
|
|
"loss": 4.4728,
|
|
"mean_token_accuracy": 0.2514013424515724,
|
|
"num_tokens": 21182471.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.1992615222930905,
|
|
"epoch": 9.30077237619264,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.639958801125842e-05,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.25406992733478545,
|
|
"num_tokens": 21193292.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.208943128585815,
|
|
"epoch": 9.305315765561108,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.6316164208902844e-05,
|
|
"loss": 4.3277,
|
|
"mean_token_accuracy": 0.26488006561994554,
|
|
"num_tokens": 21203711.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.251263809204102,
|
|
"epoch": 9.309859154929578,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.623327998717334e-05,
|
|
"loss": 4.3918,
|
|
"mean_token_accuracy": 0.25829970091581345,
|
|
"num_tokens": 21214037.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.271260118484497,
|
|
"epoch": 9.314402544298046,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.615093555057853e-05,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.2567081332206726,
|
|
"num_tokens": 21225514.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.225193977355957,
|
|
"epoch": 9.318945933666516,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.606913110229511e-05,
|
|
"loss": 4.3971,
|
|
"mean_token_accuracy": 0.25886778980493547,
|
|
"num_tokens": 21236002.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.27762656211853,
|
|
"epoch": 9.323489323034984,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.598786684416738e-05,
|
|
"loss": 4.4544,
|
|
"mean_token_accuracy": 0.2548588290810585,
|
|
"num_tokens": 21245785.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.241774845123291,
|
|
"epoch": 9.328032712403452,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.59071429767069e-05,
|
|
"loss": 4.4087,
|
|
"mean_token_accuracy": 0.2586813747882843,
|
|
"num_tokens": 21255396.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.266464471817017,
|
|
"epoch": 9.332576101771922,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.5826959699091725e-05,
|
|
"loss": 4.428,
|
|
"mean_token_accuracy": 0.2657110184431076,
|
|
"num_tokens": 21265341.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.204911231994629,
|
|
"epoch": 9.33711949114039,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.574731720916617e-05,
|
|
"loss": 4.3978,
|
|
"mean_token_accuracy": 0.2508070558309555,
|
|
"num_tokens": 21276344.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.276037788391113,
|
|
"epoch": 9.34166288050886,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.56682157034401e-05,
|
|
"loss": 4.3899,
|
|
"mean_token_accuracy": 0.25527921617031096,
|
|
"num_tokens": 21286836.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.244865560531617,
|
|
"epoch": 9.346206269877328,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.558965537708867e-05,
|
|
"loss": 4.3695,
|
|
"mean_token_accuracy": 0.2564277946949005,
|
|
"num_tokens": 21297064.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.248440837860107,
|
|
"epoch": 9.350749659245798,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.551163642395165e-05,
|
|
"loss": 4.3959,
|
|
"mean_token_accuracy": 0.2626134932041168,
|
|
"num_tokens": 21307687.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.1443897724151615,
|
|
"epoch": 9.355293048614266,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.543415903653307e-05,
|
|
"loss": 4.3786,
|
|
"mean_token_accuracy": 0.2537171706557274,
|
|
"num_tokens": 21317596.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.217080545425415,
|
|
"epoch": 9.359836437982736,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.535722340600065e-05,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.2537093013525009,
|
|
"num_tokens": 21327515.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.272422409057617,
|
|
"epoch": 9.364379827351204,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.528082972218548e-05,
|
|
"loss": 4.4951,
|
|
"mean_token_accuracy": 0.2536563128232956,
|
|
"num_tokens": 21338583.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.273396301269531,
|
|
"epoch": 9.368923216719672,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.520497817358129e-05,
|
|
"loss": 4.4576,
|
|
"mean_token_accuracy": 0.2548402577638626,
|
|
"num_tokens": 21349119.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.287933731079102,
|
|
"epoch": 9.373466606088142,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.5129668947344275e-05,
|
|
"loss": 4.4435,
|
|
"mean_token_accuracy": 0.2619324862957001,
|
|
"num_tokens": 21359997.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.225732278823853,
|
|
"epoch": 9.37800999545661,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.505490222929246e-05,
|
|
"loss": 4.3772,
|
|
"mean_token_accuracy": 0.25897299498319626,
|
|
"num_tokens": 21369457.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.276122045516968,
|
|
"epoch": 9.38255338482508,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.498067820390537e-05,
|
|
"loss": 4.4536,
|
|
"mean_token_accuracy": 0.2542008697986603,
|
|
"num_tokens": 21380230.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.171463966369629,
|
|
"epoch": 9.387096774193548,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.4906997054323245e-05,
|
|
"loss": 4.3652,
|
|
"mean_token_accuracy": 0.25925483256578447,
|
|
"num_tokens": 21390235.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.209560203552246,
|
|
"epoch": 9.391640163562018,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.483385896234714e-05,
|
|
"loss": 4.3912,
|
|
"mean_token_accuracy": 0.26418634355068205,
|
|
"num_tokens": 21400034.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.210446405410766,
|
|
"epoch": 9.396183552930486,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.4761264108437966e-05,
|
|
"loss": 4.3536,
|
|
"mean_token_accuracy": 0.25638953894376754,
|
|
"num_tokens": 21410136.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.2516130924224855,
|
|
"epoch": 9.400726942298956,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.4689212671716335e-05,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.25658652782440183,
|
|
"num_tokens": 21420333.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.224611806869507,
|
|
"epoch": 9.405270331667424,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.461770482996194e-05,
|
|
"loss": 4.3884,
|
|
"mean_token_accuracy": 0.26114570051431657,
|
|
"num_tokens": 21430804.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.198682260513306,
|
|
"epoch": 9.409813721035892,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.4546740759613354e-05,
|
|
"loss": 4.3743,
|
|
"mean_token_accuracy": 0.2556040197610855,
|
|
"num_tokens": 21440769.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.267255544662476,
|
|
"epoch": 9.414357110404362,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.447632063576734e-05,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.2514883756637573,
|
|
"num_tokens": 21450648.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.226421117782593,
|
|
"epoch": 9.41890049977283,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.4406444632178515e-05,
|
|
"loss": 4.4217,
|
|
"mean_token_accuracy": 0.24768737703561783,
|
|
"num_tokens": 21461038.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.25992259979248,
|
|
"epoch": 9.4234438891413,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.4337112921259005e-05,
|
|
"loss": 4.4182,
|
|
"mean_token_accuracy": 0.2541766494512558,
|
|
"num_tokens": 21471702.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.248213386535644,
|
|
"epoch": 9.427987278509768,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.426832567407794e-05,
|
|
"loss": 4.3503,
|
|
"mean_token_accuracy": 0.2626885578036308,
|
|
"num_tokens": 21482434.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.202988719940185,
|
|
"epoch": 9.432530667878238,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.420008306036098e-05,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.267041078209877,
|
|
"num_tokens": 21492738.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.200833797454834,
|
|
"epoch": 9.437074057246706,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.413238524848998e-05,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.25755934566259386,
|
|
"num_tokens": 21503154.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.217644691467285,
|
|
"epoch": 9.441617446615174,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.4065232405502576e-05,
|
|
"loss": 4.4189,
|
|
"mean_token_accuracy": 0.254086522758007,
|
|
"num_tokens": 21513097.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.264949178695678,
|
|
"epoch": 9.446160835983644,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.399862469709175e-05,
|
|
"loss": 4.4127,
|
|
"mean_token_accuracy": 0.25145620852708817,
|
|
"num_tokens": 21522888.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.2430463314056395,
|
|
"epoch": 9.450704225352112,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.3932562287605364e-05,
|
|
"loss": 4.3656,
|
|
"mean_token_accuracy": 0.26143275648355485,
|
|
"num_tokens": 21533998.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.2002850532531735,
|
|
"epoch": 9.455247614720582,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.38670453400459e-05,
|
|
"loss": 4.453,
|
|
"mean_token_accuracy": 0.2559526652097702,
|
|
"num_tokens": 21544042.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.1758781433105465,
|
|
"epoch": 9.45979100408905,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.3802074016069835e-05,
|
|
"loss": 4.3559,
|
|
"mean_token_accuracy": 0.26852357387542725,
|
|
"num_tokens": 21553449.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.252255487442016,
|
|
"epoch": 9.46433439345752,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.373764847598756e-05,
|
|
"loss": 4.4391,
|
|
"mean_token_accuracy": 0.2540691688656807,
|
|
"num_tokens": 21564485.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.279834032058716,
|
|
"epoch": 9.468877782825988,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.3673768878762564e-05,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.24613537341356279,
|
|
"num_tokens": 21574889.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.198553419113159,
|
|
"epoch": 9.473421172194458,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.361043538201152e-05,
|
|
"loss": 4.3728,
|
|
"mean_token_accuracy": 0.2664152875542641,
|
|
"num_tokens": 21584913.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.205464315414429,
|
|
"epoch": 9.477964561562926,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 5.3547648142003454e-05,
|
|
"loss": 4.3154,
|
|
"mean_token_accuracy": 0.26633008420467374,
|
|
"num_tokens": 21594711.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.188658952713013,
|
|
"epoch": 9.482507950931394,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.348540731365968e-05,
|
|
"loss": 4.3506,
|
|
"mean_token_accuracy": 0.2613437160849571,
|
|
"num_tokens": 21605100.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.2172278881073,
|
|
"epoch": 9.487051340299864,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.342371305055323e-05,
|
|
"loss": 4.4132,
|
|
"mean_token_accuracy": 0.25739458203315735,
|
|
"num_tokens": 21616153.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.245951366424561,
|
|
"epoch": 9.491594729668332,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.3362565504908595e-05,
|
|
"loss": 4.4303,
|
|
"mean_token_accuracy": 0.26548206359148024,
|
|
"num_tokens": 21626102.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.168992805480957,
|
|
"epoch": 9.496138119036802,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.330196482760126e-05,
|
|
"loss": 4.3045,
|
|
"mean_token_accuracy": 0.2718330562114716,
|
|
"num_tokens": 21635338.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.260129976272583,
|
|
"epoch": 9.50068150840527,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.324191116815735e-05,
|
|
"loss": 4.4522,
|
|
"mean_token_accuracy": 0.25513927787542345,
|
|
"num_tokens": 21645783.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.152850246429443,
|
|
"epoch": 9.50522489777374,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.31824046747533e-05,
|
|
"loss": 4.332,
|
|
"mean_token_accuracy": 0.2635300442576408,
|
|
"num_tokens": 21656112.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.27111930847168,
|
|
"epoch": 9.509768287142208,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.312344549421547e-05,
|
|
"loss": 4.3843,
|
|
"mean_token_accuracy": 0.2625214233994484,
|
|
"num_tokens": 21667886.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.210653162002563,
|
|
"epoch": 9.514311676510676,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.306503377201981e-05,
|
|
"loss": 4.3455,
|
|
"mean_token_accuracy": 0.26023375540971755,
|
|
"num_tokens": 21677726.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.2178879737854,
|
|
"epoch": 9.518855065879146,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.300716965229141e-05,
|
|
"loss": 4.4449,
|
|
"mean_token_accuracy": 0.2562070205807686,
|
|
"num_tokens": 21688552.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.2279016971588135,
|
|
"epoch": 9.523398455247614,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.294985327780423e-05,
|
|
"loss": 4.4073,
|
|
"mean_token_accuracy": 0.25724405497312547,
|
|
"num_tokens": 21699206.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.224678993225098,
|
|
"epoch": 9.527941844616084,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.289308478998076e-05,
|
|
"loss": 4.3415,
|
|
"mean_token_accuracy": 0.2646951645612717,
|
|
"num_tokens": 21709083.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.178432989120483,
|
|
"epoch": 9.532485233984552,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.2836864328891533e-05,
|
|
"loss": 4.3663,
|
|
"mean_token_accuracy": 0.2578443750739098,
|
|
"num_tokens": 21719092.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.18172869682312,
|
|
"epoch": 9.537028623353022,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.278119203325501e-05,
|
|
"loss": 4.4343,
|
|
"mean_token_accuracy": 0.2536148399114609,
|
|
"num_tokens": 21729335.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 9.537028623353022,
|
|
"eval_entropy": 5.097596596562585,
|
|
"eval_loss": 5.656391143798828,
|
|
"eval_mean_token_accuracy": 0.17632159588988436,
|
|
"eval_num_tokens": 21729335.0,
|
|
"eval_runtime": 2.0015,
|
|
"eval_samples_per_second": 1716.173,
|
|
"eval_steps_per_second": 214.834,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 5.2120733737945555,
|
|
"epoch": 9.54157201272149,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.272606804043701e-05,
|
|
"loss": 4.3875,
|
|
"mean_token_accuracy": 0.2632221668958664,
|
|
"num_tokens": 21740135.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 5.152893304824829,
|
|
"epoch": 9.54611540208996,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.267149248645054e-05,
|
|
"loss": 4.2949,
|
|
"mean_token_accuracy": 0.2657772719860077,
|
|
"num_tokens": 21749481.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 5.272723436355591,
|
|
"epoch": 9.550658791458428,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.2617465505955324e-05,
|
|
"loss": 4.4804,
|
|
"mean_token_accuracy": 0.24432436972856522,
|
|
"num_tokens": 21760175.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 5.209526777267456,
|
|
"epoch": 9.555202180826896,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.256398723225758e-05,
|
|
"loss": 4.3904,
|
|
"mean_token_accuracy": 0.25785136967897415,
|
|
"num_tokens": 21771138.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 5.176654863357544,
|
|
"epoch": 9.559745570195366,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.251105779730963e-05,
|
|
"loss": 4.4173,
|
|
"mean_token_accuracy": 0.25356009006500246,
|
|
"num_tokens": 21781459.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 5.160875940322876,
|
|
"epoch": 9.564288959563834,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.2458677331709585e-05,
|
|
"loss": 4.3661,
|
|
"mean_token_accuracy": 0.26698879301548006,
|
|
"num_tokens": 21792265.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 5.232065153121948,
|
|
"epoch": 9.568832348932304,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.240684596470104e-05,
|
|
"loss": 4.4386,
|
|
"mean_token_accuracy": 0.25140133649110796,
|
|
"num_tokens": 21802379.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 5.224978303909301,
|
|
"epoch": 9.573375738300772,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.2355563824172744e-05,
|
|
"loss": 4.4482,
|
|
"mean_token_accuracy": 0.24882517606019974,
|
|
"num_tokens": 21812910.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 5.234821557998657,
|
|
"epoch": 9.577919127669242,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.23048310366583e-05,
|
|
"loss": 4.4264,
|
|
"mean_token_accuracy": 0.25467455834150315,
|
|
"num_tokens": 21823619.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 5.272902202606201,
|
|
"epoch": 9.58246251703771,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.2254647727335795e-05,
|
|
"loss": 4.5269,
|
|
"mean_token_accuracy": 0.2495165541768074,
|
|
"num_tokens": 21833661.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 5.176737833023071,
|
|
"epoch": 9.58700590640618,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.220501402002755e-05,
|
|
"loss": 4.3222,
|
|
"mean_token_accuracy": 0.26104004085063937,
|
|
"num_tokens": 21843528.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 5.199957227706909,
|
|
"epoch": 9.591549295774648,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.215593003719981e-05,
|
|
"loss": 4.3512,
|
|
"mean_token_accuracy": 0.26327215433120726,
|
|
"num_tokens": 21854282.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 5.225586175918579,
|
|
"epoch": 9.596092685143116,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.210739589996245e-05,
|
|
"loss": 4.4083,
|
|
"mean_token_accuracy": 0.2516780227422714,
|
|
"num_tokens": 21864631.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 5.188154029846191,
|
|
"epoch": 9.600636074511586,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.2059411728068595e-05,
|
|
"loss": 4.3937,
|
|
"mean_token_accuracy": 0.2625514522194862,
|
|
"num_tokens": 21875298.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 5.2554277896881105,
|
|
"epoch": 9.605179463880054,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.201197763991444e-05,
|
|
"loss": 4.4424,
|
|
"mean_token_accuracy": 0.25002156049013136,
|
|
"num_tokens": 21886317.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 5.247936058044433,
|
|
"epoch": 9.609722853248524,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.196509375253887e-05,
|
|
"loss": 4.4113,
|
|
"mean_token_accuracy": 0.25962085872888563,
|
|
"num_tokens": 21897459.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 5.246512937545776,
|
|
"epoch": 9.614266242616992,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.19187601816232e-05,
|
|
"loss": 4.4069,
|
|
"mean_token_accuracy": 0.25629227608442307,
|
|
"num_tokens": 21909122.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 5.238268184661865,
|
|
"epoch": 9.618809631985462,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.187297704149094e-05,
|
|
"loss": 4.3547,
|
|
"mean_token_accuracy": 0.2582015201449394,
|
|
"num_tokens": 21918610.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 5.2464919090271,
|
|
"epoch": 9.62335302135393,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.182774444510744e-05,
|
|
"loss": 4.3594,
|
|
"mean_token_accuracy": 0.26762671321630477,
|
|
"num_tokens": 21927697.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 5.210334825515747,
|
|
"epoch": 9.627896410722398,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.178306250407963e-05,
|
|
"loss": 4.3416,
|
|
"mean_token_accuracy": 0.26951252967119216,
|
|
"num_tokens": 21936750.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 5.2311598777771,
|
|
"epoch": 9.632439800090868,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.1738931328655714e-05,
|
|
"loss": 4.3742,
|
|
"mean_token_accuracy": 0.2652585610747337,
|
|
"num_tokens": 21946291.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 5.192223787307739,
|
|
"epoch": 9.636983189459336,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 5.1695351027725054e-05,
|
|
"loss": 4.3415,
|
|
"mean_token_accuracy": 0.2649568811058998,
|
|
"num_tokens": 21956102.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 5.1826841831207275,
|
|
"epoch": 9.641526578827806,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.1652321708817646e-05,
|
|
"loss": 4.3636,
|
|
"mean_token_accuracy": 0.26272705346345904,
|
|
"num_tokens": 21966953.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 5.273648881912232,
|
|
"epoch": 9.646069968196274,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.160984347810409e-05,
|
|
"loss": 4.4512,
|
|
"mean_token_accuracy": 0.25576180070638654,
|
|
"num_tokens": 21977461.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 5.260091924667359,
|
|
"epoch": 9.650613357564744,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.15679164403952e-05,
|
|
"loss": 4.448,
|
|
"mean_token_accuracy": 0.24706888496875762,
|
|
"num_tokens": 21988571.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 5.280228853225708,
|
|
"epoch": 9.655156746933212,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.152654069914177e-05,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.2564290761947632,
|
|
"num_tokens": 22000503.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 5.174888563156128,
|
|
"epoch": 9.65970013630168,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.1485716356434315e-05,
|
|
"loss": 4.2762,
|
|
"mean_token_accuracy": 0.26997085809707644,
|
|
"num_tokens": 22010927.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 5.207941913604737,
|
|
"epoch": 9.66424352567015,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.144544351300285e-05,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.2583809942007065,
|
|
"num_tokens": 22020033.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 5.251324939727783,
|
|
"epoch": 9.668786915038618,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.14057222682166e-05,
|
|
"loss": 4.4597,
|
|
"mean_token_accuracy": 0.24586455821990966,
|
|
"num_tokens": 22031124.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 5.2188805103302,
|
|
"epoch": 9.673330304407088,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.136655272008381e-05,
|
|
"loss": 4.3842,
|
|
"mean_token_accuracy": 0.260808727145195,
|
|
"num_tokens": 22041624.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 5.209733486175537,
|
|
"epoch": 9.677873693775556,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.1327934965251445e-05,
|
|
"loss": 4.4434,
|
|
"mean_token_accuracy": 0.25377359241247177,
|
|
"num_tokens": 22053344.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 5.270387029647827,
|
|
"epoch": 9.682417083144026,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.128986909900499e-05,
|
|
"loss": 4.3787,
|
|
"mean_token_accuracy": 0.26355412751436236,
|
|
"num_tokens": 22063519.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 5.217002964019775,
|
|
"epoch": 9.686960472512494,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.125235521526814e-05,
|
|
"loss": 4.4699,
|
|
"mean_token_accuracy": 0.2497694179415703,
|
|
"num_tokens": 22074443.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 5.252590894699097,
|
|
"epoch": 9.691503861880964,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.121539340660269e-05,
|
|
"loss": 4.4298,
|
|
"mean_token_accuracy": 0.2544380068778992,
|
|
"num_tokens": 22084664.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 5.232533407211304,
|
|
"epoch": 9.696047251249432,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.117898376420827e-05,
|
|
"loss": 4.3528,
|
|
"mean_token_accuracy": 0.26296240389347075,
|
|
"num_tokens": 22093539.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 5.221664333343506,
|
|
"epoch": 9.7005906406179,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.1143126377922e-05,
|
|
"loss": 4.3794,
|
|
"mean_token_accuracy": 0.2560381844639778,
|
|
"num_tokens": 22103441.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 5.226282548904419,
|
|
"epoch": 9.70513402998637,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.110782133621843e-05,
|
|
"loss": 4.3847,
|
|
"mean_token_accuracy": 0.26153538227081297,
|
|
"num_tokens": 22113770.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 5.2043976306915285,
|
|
"epoch": 9.709677419354838,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.107306872620926e-05,
|
|
"loss": 4.412,
|
|
"mean_token_accuracy": 0.2507258027791977,
|
|
"num_tokens": 22124452.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 5.211108922958374,
|
|
"epoch": 9.714220808723308,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.10388686336431e-05,
|
|
"loss": 4.3387,
|
|
"mean_token_accuracy": 0.25481604784727097,
|
|
"num_tokens": 22133719.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 5.192142963409424,
|
|
"epoch": 9.718764198091776,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.1005221142905254e-05,
|
|
"loss": 4.366,
|
|
"mean_token_accuracy": 0.2656842455267906,
|
|
"num_tokens": 22144271.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 5.200067567825317,
|
|
"epoch": 9.723307587460246,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.0972126337017575e-05,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.254022428393364,
|
|
"num_tokens": 22154581.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 5.208414745330811,
|
|
"epoch": 9.727850976828714,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.093958429763821e-05,
|
|
"loss": 4.469,
|
|
"mean_token_accuracy": 0.25241209119558333,
|
|
"num_tokens": 22165831.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 5.202087926864624,
|
|
"epoch": 9.732394366197184,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.090759510506138e-05,
|
|
"loss": 4.3905,
|
|
"mean_token_accuracy": 0.26246648132801054,
|
|
"num_tokens": 22176407.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 5.207535028457642,
|
|
"epoch": 9.736937755565652,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.087615883821728e-05,
|
|
"loss": 4.4106,
|
|
"mean_token_accuracy": 0.2610371232032776,
|
|
"num_tokens": 22187451.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 5.250697755813599,
|
|
"epoch": 9.74148114493412,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 5.084527557467175e-05,
|
|
"loss": 4.437,
|
|
"mean_token_accuracy": 0.2567033916711807,
|
|
"num_tokens": 22196893.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 5.166067266464234,
|
|
"epoch": 9.74602453430259,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.0814945390626174e-05,
|
|
"loss": 4.2541,
|
|
"mean_token_accuracy": 0.2716713011264801,
|
|
"num_tokens": 22207072.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 5.257362937927246,
|
|
"epoch": 9.750567923671058,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.0785168360917285e-05,
|
|
"loss": 4.4048,
|
|
"mean_token_accuracy": 0.2555985078215599,
|
|
"num_tokens": 22217708.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 5.189595031738281,
|
|
"epoch": 9.755111313039528,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.075594455901691e-05,
|
|
"loss": 4.3719,
|
|
"mean_token_accuracy": 0.261403451859951,
|
|
"num_tokens": 22229213.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 5.1922760009765625,
|
|
"epoch": 9.759654702407996,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.072727405703189e-05,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.2574868455529213,
|
|
"num_tokens": 22240399.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 5.1992175579071045,
|
|
"epoch": 9.764198091776466,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.0699156925703846e-05,
|
|
"loss": 4.391,
|
|
"mean_token_accuracy": 0.2557691618800163,
|
|
"num_tokens": 22250842.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 5.2123888492584225,
|
|
"epoch": 9.768741481144934,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.0671593234409e-05,
|
|
"loss": 4.3508,
|
|
"mean_token_accuracy": 0.26600344479084015,
|
|
"num_tokens": 22261829.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 5.248690748214722,
|
|
"epoch": 9.773284870513404,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.0644583051158005e-05,
|
|
"loss": 4.4042,
|
|
"mean_token_accuracy": 0.2556778475642204,
|
|
"num_tokens": 22273515.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 5.191469335556031,
|
|
"epoch": 9.777828259881872,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.061812644259582e-05,
|
|
"loss": 4.4068,
|
|
"mean_token_accuracy": 0.25911415070295335,
|
|
"num_tokens": 22284167.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 5.188240814208984,
|
|
"epoch": 9.78237164925034,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.059222347400151e-05,
|
|
"loss": 4.4028,
|
|
"mean_token_accuracy": 0.2598406359553337,
|
|
"num_tokens": 22295442.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 5.27122917175293,
|
|
"epoch": 9.78691503861881,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.056687420928805e-05,
|
|
"loss": 4.4101,
|
|
"mean_token_accuracy": 0.26193796545267106,
|
|
"num_tokens": 22305717.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 5.208701467514038,
|
|
"epoch": 9.791458427987278,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.0542078711002246e-05,
|
|
"loss": 4.3152,
|
|
"mean_token_accuracy": 0.2711173802614212,
|
|
"num_tokens": 22315802.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 5.216084384918213,
|
|
"epoch": 9.796001817355748,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.0517837040324515e-05,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.23840076327323914,
|
|
"num_tokens": 22326161.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 5.221318483352661,
|
|
"epoch": 9.800545206724216,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.049414925706877e-05,
|
|
"loss": 4.4332,
|
|
"mean_token_accuracy": 0.25220952481031417,
|
|
"num_tokens": 22335760.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 5.169746923446655,
|
|
"epoch": 9.805088596092686,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.0471015419682273e-05,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.255529397726059,
|
|
"num_tokens": 22345954.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 5.1574851989746096,
|
|
"epoch": 9.809631985461154,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.044843558524547e-05,
|
|
"loss": 4.243,
|
|
"mean_token_accuracy": 0.271256747841835,
|
|
"num_tokens": 22354638.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 5.257366132736206,
|
|
"epoch": 9.814175374829622,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.042640980947183e-05,
|
|
"loss": 4.4659,
|
|
"mean_token_accuracy": 0.24865181148052215,
|
|
"num_tokens": 22365831.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 5.244766187667847,
|
|
"epoch": 9.818718764198092,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.0404938146707766e-05,
|
|
"loss": 4.4366,
|
|
"mean_token_accuracy": 0.24995417296886444,
|
|
"num_tokens": 22376285.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 5.158493232727051,
|
|
"epoch": 9.82326215356656,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.038402064993251e-05,
|
|
"loss": 4.3802,
|
|
"mean_token_accuracy": 0.2524024173617363,
|
|
"num_tokens": 22385587.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 5.225384616851807,
|
|
"epoch": 9.82780554293503,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.036365737075789e-05,
|
|
"loss": 4.4277,
|
|
"mean_token_accuracy": 0.25224491506814956,
|
|
"num_tokens": 22396273.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 5.204897451400757,
|
|
"epoch": 9.832348932303498,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.034384835942826e-05,
|
|
"loss": 4.4195,
|
|
"mean_token_accuracy": 0.24936858117580413,
|
|
"num_tokens": 22408002.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 5.207967138290405,
|
|
"epoch": 9.836892321671968,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.032459366482041e-05,
|
|
"loss": 4.4025,
|
|
"mean_token_accuracy": 0.25296441167593003,
|
|
"num_tokens": 22418910.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 5.194256210327149,
|
|
"epoch": 9.841435711040436,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.0305893334443346e-05,
|
|
"loss": 4.3505,
|
|
"mean_token_accuracy": 0.25782200396060945,
|
|
"num_tokens": 22429549.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 5.185819339752197,
|
|
"epoch": 9.845979100408904,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.02877474144383e-05,
|
|
"loss": 4.3976,
|
|
"mean_token_accuracy": 0.25230061560869216,
|
|
"num_tokens": 22440220.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 5.244987106323242,
|
|
"epoch": 9.850522489777374,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.0270155949578526e-05,
|
|
"loss": 4.4636,
|
|
"mean_token_accuracy": 0.26096087545156477,
|
|
"num_tokens": 22450877.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 5.270755052566528,
|
|
"epoch": 9.855065879145842,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.025311898326922e-05,
|
|
"loss": 4.4493,
|
|
"mean_token_accuracy": 0.25141348987817763,
|
|
"num_tokens": 22461047.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 5.237989902496338,
|
|
"epoch": 9.859609268514312,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.02366365575474e-05,
|
|
"loss": 4.4361,
|
|
"mean_token_accuracy": 0.25142005681991575,
|
|
"num_tokens": 22472183.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 5.278855037689209,
|
|
"epoch": 9.86415265788278,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.022070871308179e-05,
|
|
"loss": 4.4242,
|
|
"mean_token_accuracy": 0.2563687592744827,
|
|
"num_tokens": 22482576.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 5.196455287933349,
|
|
"epoch": 9.86869604725125,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.020533548917279e-05,
|
|
"loss": 4.4022,
|
|
"mean_token_accuracy": 0.2539734810590744,
|
|
"num_tokens": 22491871.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 5.228411912918091,
|
|
"epoch": 9.873239436619718,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.01905169237523e-05,
|
|
"loss": 4.3954,
|
|
"mean_token_accuracy": 0.259551477432251,
|
|
"num_tokens": 22501738.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 5.1722899913787845,
|
|
"epoch": 9.877782825988188,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.017625305338363e-05,
|
|
"loss": 4.3462,
|
|
"mean_token_accuracy": 0.26701701879501344,
|
|
"num_tokens": 22511692.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 5.202678155899048,
|
|
"epoch": 9.882326215356656,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.01625439132615e-05,
|
|
"loss": 4.4268,
|
|
"mean_token_accuracy": 0.25392004400491713,
|
|
"num_tokens": 22520975.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 5.275298404693603,
|
|
"epoch": 9.886869604725124,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.0149389537211824e-05,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.24713047742843627,
|
|
"num_tokens": 22531365.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 5.197598171234131,
|
|
"epoch": 9.891412994093594,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.013678995769173e-05,
|
|
"loss": 4.3916,
|
|
"mean_token_accuracy": 0.2561886370182037,
|
|
"num_tokens": 22540949.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 5.215548801422119,
|
|
"epoch": 9.895956383462062,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.012474520578942e-05,
|
|
"loss": 4.4017,
|
|
"mean_token_accuracy": 0.2525309160351753,
|
|
"num_tokens": 22551342.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 5.202882385253906,
|
|
"epoch": 9.900499772830532,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.0113255311224107e-05,
|
|
"loss": 4.4362,
|
|
"mean_token_accuracy": 0.25373011976480486,
|
|
"num_tokens": 22562251.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 5.271402072906494,
|
|
"epoch": 9.905043162199,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.0102320302345996e-05,
|
|
"loss": 4.5025,
|
|
"mean_token_accuracy": 0.24296278804540633,
|
|
"num_tokens": 22573347.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 5.227095222473144,
|
|
"epoch": 9.90958655156747,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.0091940206136123e-05,
|
|
"loss": 4.3656,
|
|
"mean_token_accuracy": 0.2709994077682495,
|
|
"num_tokens": 22583790.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 5.274792337417603,
|
|
"epoch": 9.914129940935938,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.008211504820635e-05,
|
|
"loss": 4.4832,
|
|
"mean_token_accuracy": 0.2469862088561058,
|
|
"num_tokens": 22594912.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 5.238311100006103,
|
|
"epoch": 9.918673330304408,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.007284485279925e-05,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.257322458922863,
|
|
"num_tokens": 22605602.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 5.258810234069824,
|
|
"epoch": 9.923216719672876,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.006412964278816e-05,
|
|
"loss": 4.4184,
|
|
"mean_token_accuracy": 0.24470768570899964,
|
|
"num_tokens": 22616365.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 5.22624626159668,
|
|
"epoch": 9.927760109041344,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.0055969439676925e-05,
|
|
"loss": 4.3723,
|
|
"mean_token_accuracy": 0.26570749282836914,
|
|
"num_tokens": 22626238.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 5.251295137405395,
|
|
"epoch": 9.932303498409814,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.004836426360008e-05,
|
|
"loss": 4.4099,
|
|
"mean_token_accuracy": 0.25762871503829954,
|
|
"num_tokens": 22636420.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 5.208899021148682,
|
|
"epoch": 9.936846887778282,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.004131413332263e-05,
|
|
"loss": 4.3725,
|
|
"mean_token_accuracy": 0.26067101061344145,
|
|
"num_tokens": 22646968.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 5.24845929145813,
|
|
"epoch": 9.941390277146752,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.003481906624009e-05,
|
|
"loss": 4.4424,
|
|
"mean_token_accuracy": 0.24958671033382415,
|
|
"num_tokens": 22657243.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 5.205689144134522,
|
|
"epoch": 9.94593366651522,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.002887907837835e-05,
|
|
"loss": 4.3385,
|
|
"mean_token_accuracy": 0.2735374540090561,
|
|
"num_tokens": 22668143.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 5.262158155441284,
|
|
"epoch": 9.95047705588369,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.002349418439379e-05,
|
|
"loss": 4.489,
|
|
"mean_token_accuracy": 0.2449803665280342,
|
|
"num_tokens": 22680731.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 5.240759229660034,
|
|
"epoch": 9.955020445252158,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.001866439757306e-05,
|
|
"loss": 4.4048,
|
|
"mean_token_accuracy": 0.25890006870031357,
|
|
"num_tokens": 22691309.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 5.213824939727783,
|
|
"epoch": 9.959563834620628,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.00143897298332e-05,
|
|
"loss": 4.3928,
|
|
"mean_token_accuracy": 0.25896636098623277,
|
|
"num_tokens": 22702636.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 5.2173011302948,
|
|
"epoch": 9.964107223989096,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.001067019172151e-05,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.25222279131412506,
|
|
"num_tokens": 22714801.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 5.207951831817627,
|
|
"epoch": 9.968650613357564,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.00075057924156e-05,
|
|
"loss": 4.373,
|
|
"mean_token_accuracy": 0.2573188692331314,
|
|
"num_tokens": 22726149.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 5.21246747970581,
|
|
"epoch": 9.973194002726034,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.000489653972332e-05,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.2578797787427902,
|
|
"num_tokens": 22736084.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 5.179901218414306,
|
|
"epoch": 9.977737392094502,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.0002842440082713e-05,
|
|
"loss": 4.3719,
|
|
"mean_token_accuracy": 0.264464907348156,
|
|
"num_tokens": 22745646.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 5.200182104110718,
|
|
"epoch": 9.982280781462972,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.000134349856208e-05,
|
|
"loss": 4.3384,
|
|
"mean_token_accuracy": 0.2635975077748299,
|
|
"num_tokens": 22755822.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 5.1718817234039305,
|
|
"epoch": 9.98682417083144,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.000039971885989e-05,
|
|
"loss": 4.3621,
|
|
"mean_token_accuracy": 0.2650574520230293,
|
|
"num_tokens": 22765773.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 5.240612888336182,
|
|
"epoch": 9.99136756019991,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.0000011103304856e-05,
|
|
"loss": 4.4255,
|
|
"mean_token_accuracy": 0.2587421178817749,
|
|
"num_tokens": 22776353.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 9.99136756019991,
|
|
"eval_entropy": 5.10820281893708,
|
|
"eval_loss": 5.655570030212402,
|
|
"eval_mean_token_accuracy": 0.17649583632862845,
|
|
"eval_num_tokens": 22776353.0,
|
|
"eval_runtime": 2.0274,
|
|
"eval_samples_per_second": 1694.281,
|
|
"eval_steps_per_second": 212.093,
|
|
"step": 11000
|
|
}
|
|
],
|
|
"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": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4899031485143040.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|