043fd421c2
Model: fpadovani/nld-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.3625624716038165,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 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
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 666580809185280.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|