043fd421c2
Model: fpadovani/nld-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.45433893684688775,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 221422467317760.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|