4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.1339539868055323,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74211368560791,
|
|
"epoch": 0.00033488496701383075,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7607,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10113.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742099857330322,
|
|
"epoch": 0.0006697699340276615,
|
|
"grad_norm": 4.40625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7537,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 19598.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742137718200684,
|
|
"epoch": 0.0010046549010414923,
|
|
"grad_norm": 4.1875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7462,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 29705.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742128372192383,
|
|
"epoch": 0.001339539868055323,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.7119,
|
|
"mean_token_accuracy": 0.00036584193585440514,
|
|
"num_tokens": 39871.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742036437988281,
|
|
"epoch": 0.0016744248350691537,
|
|
"grad_norm": 4.09375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6391,
|
|
"mean_token_accuracy": 0.0023467960068956017,
|
|
"num_tokens": 49805.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741703224182128,
|
|
"epoch": 0.0020093098020829846,
|
|
"grad_norm": 3.734375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5416,
|
|
"mean_token_accuracy": 0.027319939527660608,
|
|
"num_tokens": 59845.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740657615661622,
|
|
"epoch": 0.0023441947690968153,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4647,
|
|
"mean_token_accuracy": 0.03426684513688087,
|
|
"num_tokens": 69836.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.737139415740966,
|
|
"epoch": 0.002679079736110646,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.3382,
|
|
"mean_token_accuracy": 0.040677194483578205,
|
|
"num_tokens": 80088.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.727961826324464,
|
|
"epoch": 0.0030139647031244767,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.2362,
|
|
"mean_token_accuracy": 0.04172689206898213,
|
|
"num_tokens": 89458.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.714122295379639,
|
|
"epoch": 0.0033488496701383074,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.1927,
|
|
"mean_token_accuracy": 0.04110763110220432,
|
|
"num_tokens": 99523.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.702536869049073,
|
|
"epoch": 0.003683734637152138,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.1602,
|
|
"mean_token_accuracy": 0.04160156771540642,
|
|
"num_tokens": 110080.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.694610786437988,
|
|
"epoch": 0.004018619604165969,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.0683,
|
|
"mean_token_accuracy": 0.043702947348356246,
|
|
"num_tokens": 119412.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.687895107269288,
|
|
"epoch": 0.0043535045711798,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0255,
|
|
"mean_token_accuracy": 0.04217404723167419,
|
|
"num_tokens": 129432.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.679110527038574,
|
|
"epoch": 0.0046883895381936306,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9755,
|
|
"mean_token_accuracy": 0.048846100270748136,
|
|
"num_tokens": 139756.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.670030307769775,
|
|
"epoch": 0.005023274505207461,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.96,
|
|
"mean_token_accuracy": 0.045024541579186915,
|
|
"num_tokens": 148846.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.662967872619628,
|
|
"epoch": 0.005358159472221292,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8931,
|
|
"mean_token_accuracy": 0.04923531375825405,
|
|
"num_tokens": 158647.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.654012298583984,
|
|
"epoch": 0.005693044439235123,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8077,
|
|
"mean_token_accuracy": 0.04514540694653988,
|
|
"num_tokens": 167701.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.641143226623536,
|
|
"epoch": 0.006027929406248953,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.7848,
|
|
"mean_token_accuracy": 0.04806430526077747,
|
|
"num_tokens": 178428.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.615053653717041,
|
|
"epoch": 0.006362814373262784,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.7243,
|
|
"mean_token_accuracy": 0.04973807036876678,
|
|
"num_tokens": 187737.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.592188262939453,
|
|
"epoch": 0.006697699340276615,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.7193,
|
|
"mean_token_accuracy": 0.05120275542140007,
|
|
"num_tokens": 198374.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.56552038192749,
|
|
"epoch": 0.007032584307290445,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.6436,
|
|
"mean_token_accuracy": 0.05372070819139481,
|
|
"num_tokens": 208483.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.523834037780762,
|
|
"epoch": 0.007367469274304276,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.5595,
|
|
"mean_token_accuracy": 0.050585757568478586,
|
|
"num_tokens": 218715.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.464954853057861,
|
|
"epoch": 0.007702354241318108,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.3861,
|
|
"mean_token_accuracy": 0.0644374880939722,
|
|
"num_tokens": 227777.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.395084571838378,
|
|
"epoch": 0.008037239208331938,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3999,
|
|
"mean_token_accuracy": 0.056889945082366465,
|
|
"num_tokens": 238192.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.354606533050537,
|
|
"epoch": 0.008372124175345769,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2388,
|
|
"mean_token_accuracy": 0.06168701723217964,
|
|
"num_tokens": 247259.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.306151485443115,
|
|
"epoch": 0.0087070091423596,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.2634,
|
|
"mean_token_accuracy": 0.05518925692886114,
|
|
"num_tokens": 258100.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.27144947052002,
|
|
"epoch": 0.00904189410937343,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.1445,
|
|
"mean_token_accuracy": 0.061012673750519754,
|
|
"num_tokens": 267620.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.183198928833008,
|
|
"epoch": 0.009376779076387261,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.0077,
|
|
"mean_token_accuracy": 0.06654397621750832,
|
|
"num_tokens": 277654.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.065981578826904,
|
|
"epoch": 0.009711664043401092,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 9.0075,
|
|
"mean_token_accuracy": 0.05672478042542935,
|
|
"num_tokens": 288195.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.022527408599853,
|
|
"epoch": 0.010046549010414923,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.9382,
|
|
"mean_token_accuracy": 0.05744555629789829,
|
|
"num_tokens": 298860.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.896670055389404,
|
|
"epoch": 0.010381433977428753,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.8435,
|
|
"mean_token_accuracy": 0.060239487513899806,
|
|
"num_tokens": 307799.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.799404430389405,
|
|
"epoch": 0.010716318944442584,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.7928,
|
|
"mean_token_accuracy": 0.05781174004077912,
|
|
"num_tokens": 318257.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.696899318695069,
|
|
"epoch": 0.011051203911456415,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.7536,
|
|
"mean_token_accuracy": 0.05891339629888535,
|
|
"num_tokens": 329065.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.617450332641601,
|
|
"epoch": 0.011386088878470245,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.6708,
|
|
"mean_token_accuracy": 0.05866752900183201,
|
|
"num_tokens": 338809.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.496949386596679,
|
|
"epoch": 0.011720973845484076,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.6339,
|
|
"mean_token_accuracy": 0.05763169266283512,
|
|
"num_tokens": 349318.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.336700916290283,
|
|
"epoch": 0.012055858812497907,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.5925,
|
|
"mean_token_accuracy": 0.060683618485927585,
|
|
"num_tokens": 358760.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.284276008605957,
|
|
"epoch": 0.012390743779511737,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.4743,
|
|
"mean_token_accuracy": 0.060685327649116515,
|
|
"num_tokens": 367589.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.175062084197998,
|
|
"epoch": 0.012725628746525568,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4897,
|
|
"mean_token_accuracy": 0.06429480686783791,
|
|
"num_tokens": 377012.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.043092918395995,
|
|
"epoch": 0.013060513713539399,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.4493,
|
|
"mean_token_accuracy": 0.06497252248227596,
|
|
"num_tokens": 386791.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 9.034938144683839,
|
|
"epoch": 0.01339539868055323,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.5439,
|
|
"mean_token_accuracy": 0.057699060812592504,
|
|
"num_tokens": 396620.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 9.003680610656739,
|
|
"epoch": 0.01373028364756706,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.508,
|
|
"mean_token_accuracy": 0.05839668586850166,
|
|
"num_tokens": 406384.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.98305311203003,
|
|
"epoch": 0.01406516861458089,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.4619,
|
|
"mean_token_accuracy": 0.06313387341797352,
|
|
"num_tokens": 416481.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.884008121490478,
|
|
"epoch": 0.014400053581594721,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4978,
|
|
"mean_token_accuracy": 0.06396310143172741,
|
|
"num_tokens": 425697.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.91770076751709,
|
|
"epoch": 0.014734938548608552,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.4105,
|
|
"mean_token_accuracy": 0.06354090645909309,
|
|
"num_tokens": 435836.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.9373610496521,
|
|
"epoch": 0.015069823515622383,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.4483,
|
|
"mean_token_accuracy": 0.06257182881236076,
|
|
"num_tokens": 445926.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.854573345184326,
|
|
"epoch": 0.015404708482636215,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.4182,
|
|
"mean_token_accuracy": 0.06113107353448868,
|
|
"num_tokens": 455594.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.783684635162354,
|
|
"epoch": 0.015739593449650046,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.4544,
|
|
"mean_token_accuracy": 0.0631777100265026,
|
|
"num_tokens": 465928.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.864254093170166,
|
|
"epoch": 0.016074478416663877,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3855,
|
|
"mean_token_accuracy": 0.0648330245167017,
|
|
"num_tokens": 475946.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.782320022583008,
|
|
"epoch": 0.016409363383677707,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.4001,
|
|
"mean_token_accuracy": 0.06609157137572766,
|
|
"num_tokens": 485285.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.789734363555908,
|
|
"epoch": 0.016744248350691538,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.3957,
|
|
"mean_token_accuracy": 0.06265997141599655,
|
|
"num_tokens": 496516.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.808907222747802,
|
|
"epoch": 0.01707913331770537,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3818,
|
|
"mean_token_accuracy": 0.06361959911882878,
|
|
"num_tokens": 506330.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.74974603652954,
|
|
"epoch": 0.0174140182847192,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.4313,
|
|
"mean_token_accuracy": 0.0675040852278471,
|
|
"num_tokens": 515825.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.786038494110107,
|
|
"epoch": 0.01774890325173303,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3628,
|
|
"mean_token_accuracy": 0.06703696362674236,
|
|
"num_tokens": 526303.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.757538318634033,
|
|
"epoch": 0.01808378821874686,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.4598,
|
|
"mean_token_accuracy": 0.06852960139513016,
|
|
"num_tokens": 536805.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.789075469970703,
|
|
"epoch": 0.01841867318576069,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.3273,
|
|
"mean_token_accuracy": 0.07035987079143524,
|
|
"num_tokens": 546381.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.71272668838501,
|
|
"epoch": 0.018753558152774522,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.3438,
|
|
"mean_token_accuracy": 0.06675126180052757,
|
|
"num_tokens": 556278.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.705748081207275,
|
|
"epoch": 0.019088443119788353,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.3522,
|
|
"mean_token_accuracy": 0.07065789699554444,
|
|
"num_tokens": 565312.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.733772277832031,
|
|
"epoch": 0.019423328086802184,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.319,
|
|
"mean_token_accuracy": 0.07612328678369522,
|
|
"num_tokens": 575304.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.657926940917969,
|
|
"epoch": 0.019758213053816014,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.2978,
|
|
"mean_token_accuracy": 0.06872594207525254,
|
|
"num_tokens": 585262.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.737933158874512,
|
|
"epoch": 0.020093098020829845,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.3275,
|
|
"mean_token_accuracy": 0.0715562004595995,
|
|
"num_tokens": 595297.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.71183385848999,
|
|
"epoch": 0.020427982987843676,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.3075,
|
|
"mean_token_accuracy": 0.07006572633981704,
|
|
"num_tokens": 604354.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.618387031555176,
|
|
"epoch": 0.020762867954857506,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.4043,
|
|
"mean_token_accuracy": 0.0666431374847889,
|
|
"num_tokens": 616283.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.783961200714112,
|
|
"epoch": 0.021097752921871337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.322,
|
|
"mean_token_accuracy": 0.06938845105469227,
|
|
"num_tokens": 626059.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.627076625823975,
|
|
"epoch": 0.021432637888885168,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.2453,
|
|
"mean_token_accuracy": 0.07488406710326671,
|
|
"num_tokens": 634483.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.68146858215332,
|
|
"epoch": 0.021767522855899,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.2387,
|
|
"mean_token_accuracy": 0.07477111965417862,
|
|
"num_tokens": 643111.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.574335098266602,
|
|
"epoch": 0.02210240782291283,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2554,
|
|
"mean_token_accuracy": 0.07630948387086392,
|
|
"num_tokens": 653650.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.697111225128173,
|
|
"epoch": 0.02243729278992666,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.252,
|
|
"mean_token_accuracy": 0.08444391936063766,
|
|
"num_tokens": 663143.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.565497016906738,
|
|
"epoch": 0.02277217775694049,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.2062,
|
|
"mean_token_accuracy": 0.07809413745999336,
|
|
"num_tokens": 672434.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.655457496643066,
|
|
"epoch": 0.02310706272395432,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2254,
|
|
"mean_token_accuracy": 0.0773940771818161,
|
|
"num_tokens": 682973.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.558607482910157,
|
|
"epoch": 0.023441947690968152,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2771,
|
|
"mean_token_accuracy": 0.07662329860031605,
|
|
"num_tokens": 692136.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.668216800689697,
|
|
"epoch": 0.023776832657981983,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1824,
|
|
"mean_token_accuracy": 0.08042559921741485,
|
|
"num_tokens": 702257.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.555757141113281,
|
|
"epoch": 0.024111717624995813,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1992,
|
|
"mean_token_accuracy": 0.07901003137230873,
|
|
"num_tokens": 712444.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.65919303894043,
|
|
"epoch": 0.024446602592009644,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.3348,
|
|
"mean_token_accuracy": 0.0759161863476038,
|
|
"num_tokens": 721614.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.671118354797363,
|
|
"epoch": 0.024781487559023475,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.3212,
|
|
"mean_token_accuracy": 0.07223852351307869,
|
|
"num_tokens": 731189.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.591705894470214,
|
|
"epoch": 0.025116372526037305,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.2213,
|
|
"mean_token_accuracy": 0.08042538389563561,
|
|
"num_tokens": 741555.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.528362846374511,
|
|
"epoch": 0.025451257493051136,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1569,
|
|
"mean_token_accuracy": 0.08212503418326378,
|
|
"num_tokens": 751560.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.550797271728516,
|
|
"epoch": 0.025786142460064967,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.2041,
|
|
"mean_token_accuracy": 0.07558429762721061,
|
|
"num_tokens": 761471.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.614824962615966,
|
|
"epoch": 0.026121027427078797,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.2015,
|
|
"mean_token_accuracy": 0.07853580340743065,
|
|
"num_tokens": 771565.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.519598484039307,
|
|
"epoch": 0.026455912394092628,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.1537,
|
|
"mean_token_accuracy": 0.07887223958969117,
|
|
"num_tokens": 781248.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.60389404296875,
|
|
"epoch": 0.02679079736110646,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.2014,
|
|
"mean_token_accuracy": 0.075946843624115,
|
|
"num_tokens": 790907.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.511116981506348,
|
|
"epoch": 0.02712568232812029,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.2054,
|
|
"mean_token_accuracy": 0.07778018116950988,
|
|
"num_tokens": 800073.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.635944366455078,
|
|
"epoch": 0.02746056729513412,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.1885,
|
|
"mean_token_accuracy": 0.07892166152596473,
|
|
"num_tokens": 809782.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.39308671951294,
|
|
"epoch": 0.02779545226214795,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.1461,
|
|
"mean_token_accuracy": 0.07830736264586449,
|
|
"num_tokens": 820329.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.558973217010498,
|
|
"epoch": 0.02813033722916178,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.0641,
|
|
"mean_token_accuracy": 0.08698545768857002,
|
|
"num_tokens": 829405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.460458278656006,
|
|
"epoch": 0.028465222196175612,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.12,
|
|
"mean_token_accuracy": 0.077915009111166,
|
|
"num_tokens": 838700.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.557555007934571,
|
|
"epoch": 0.028800107163189443,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.1831,
|
|
"mean_token_accuracy": 0.08165658637881279,
|
|
"num_tokens": 848892.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.484703731536865,
|
|
"epoch": 0.029134992130203274,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.1107,
|
|
"mean_token_accuracy": 0.08097934126853942,
|
|
"num_tokens": 857737.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.450378036499023,
|
|
"epoch": 0.029469877097217104,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.043,
|
|
"mean_token_accuracy": 0.0830142006278038,
|
|
"num_tokens": 867758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.483449363708496,
|
|
"epoch": 0.029804762064230935,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0694,
|
|
"mean_token_accuracy": 0.0823376640677452,
|
|
"num_tokens": 877273.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.511912822723389,
|
|
"epoch": 0.030139647031244766,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.1597,
|
|
"mean_token_accuracy": 0.0831824667751789,
|
|
"num_tokens": 887109.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.448967266082764,
|
|
"epoch": 0.030474531998258596,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0778,
|
|
"mean_token_accuracy": 0.08146599680185318,
|
|
"num_tokens": 896157.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.426976585388184,
|
|
"epoch": 0.03080941696527243,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0909,
|
|
"mean_token_accuracy": 0.07872407361865044,
|
|
"num_tokens": 907231.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.366917896270753,
|
|
"epoch": 0.03114430193228626,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.997,
|
|
"mean_token_accuracy": 0.08565031662583351,
|
|
"num_tokens": 916756.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.496291255950927,
|
|
"epoch": 0.03147918689930009,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.17,
|
|
"mean_token_accuracy": 0.0742259133607149,
|
|
"num_tokens": 927714.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.47745065689087,
|
|
"epoch": 0.03181407186631392,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.005,
|
|
"mean_token_accuracy": 0.0867729477584362,
|
|
"num_tokens": 936620.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.397373008728028,
|
|
"epoch": 0.03214895683332775,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.0433,
|
|
"mean_token_accuracy": 0.08084618747234344,
|
|
"num_tokens": 945942.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.336866855621338,
|
|
"epoch": 0.032483841800341584,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0883,
|
|
"mean_token_accuracy": 0.08643392398953438,
|
|
"num_tokens": 956685.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.401565170288086,
|
|
"epoch": 0.032818726767355415,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0316,
|
|
"mean_token_accuracy": 0.08411954194307328,
|
|
"num_tokens": 967453.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.446605682373047,
|
|
"epoch": 0.033153611734369245,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 8.0546,
|
|
"mean_token_accuracy": 0.083030104637146,
|
|
"num_tokens": 978220.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.329526329040528,
|
|
"epoch": 0.033488496701383076,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 8.0053,
|
|
"mean_token_accuracy": 0.08742882534861565,
|
|
"num_tokens": 988120.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.36773910522461,
|
|
"epoch": 0.03382338166839691,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 8.0128,
|
|
"mean_token_accuracy": 0.07844792157411576,
|
|
"num_tokens": 997733.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.38361415863037,
|
|
"epoch": 0.03415826663541074,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9798,
|
|
"mean_token_accuracy": 0.08303482234477996,
|
|
"num_tokens": 1007362.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.368096256256104,
|
|
"epoch": 0.03449315160242457,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.928,
|
|
"mean_token_accuracy": 0.08512061238288879,
|
|
"num_tokens": 1016352.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.259499549865723,
|
|
"epoch": 0.0348280365694384,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9417,
|
|
"mean_token_accuracy": 0.08643589839339257,
|
|
"num_tokens": 1025944.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.357065868377685,
|
|
"epoch": 0.03516292153645223,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 8.0056,
|
|
"mean_token_accuracy": 0.0855144739151001,
|
|
"num_tokens": 1035675.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.352817058563232,
|
|
"epoch": 0.03549780650346606,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.9985,
|
|
"mean_token_accuracy": 0.08436472564935685,
|
|
"num_tokens": 1045944.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.339145183563232,
|
|
"epoch": 0.03583269147047989,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9224,
|
|
"mean_token_accuracy": 0.08934126198291778,
|
|
"num_tokens": 1055721.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.342275238037109,
|
|
"epoch": 0.03616757643749372,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 8.0454,
|
|
"mean_token_accuracy": 0.0812804415822029,
|
|
"num_tokens": 1065841.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.365180015563965,
|
|
"epoch": 0.03650246140450755,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 8.0419,
|
|
"mean_token_accuracy": 0.08463779203593731,
|
|
"num_tokens": 1076574.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.389959239959717,
|
|
"epoch": 0.03683734637152138,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.9894,
|
|
"mean_token_accuracy": 0.08296842500567436,
|
|
"num_tokens": 1085761.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.370242595672607,
|
|
"epoch": 0.037172231338535214,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9807,
|
|
"mean_token_accuracy": 0.08592502772808075,
|
|
"num_tokens": 1096036.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.207129859924317,
|
|
"epoch": 0.037507116305549044,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8885,
|
|
"mean_token_accuracy": 0.08698522299528122,
|
|
"num_tokens": 1105103.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.332459354400635,
|
|
"epoch": 0.037842001272562875,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.9138,
|
|
"mean_token_accuracy": 0.09431307911872863,
|
|
"num_tokens": 1115355.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.242012119293213,
|
|
"epoch": 0.038176886239576706,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.9151,
|
|
"mean_token_accuracy": 0.08907190337777138,
|
|
"num_tokens": 1125977.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.372161674499512,
|
|
"epoch": 0.03851177120659054,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9373,
|
|
"mean_token_accuracy": 0.08887948393821717,
|
|
"num_tokens": 1135536.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.19194574356079,
|
|
"epoch": 0.03884665617360437,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8811,
|
|
"mean_token_accuracy": 0.08871764689683914,
|
|
"num_tokens": 1144857.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.234977722167969,
|
|
"epoch": 0.0391815411406182,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.9004,
|
|
"mean_token_accuracy": 0.09209405556321144,
|
|
"num_tokens": 1154188.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.2575364112854,
|
|
"epoch": 0.03951642610763203,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8999,
|
|
"mean_token_accuracy": 0.0915960919111967,
|
|
"num_tokens": 1164583.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.188764095306396,
|
|
"epoch": 0.03985131107464586,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.88,
|
|
"mean_token_accuracy": 0.09105745926499367,
|
|
"num_tokens": 1174163.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.263216972351074,
|
|
"epoch": 0.04018619604165969,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.8911,
|
|
"mean_token_accuracy": 0.09155049026012421,
|
|
"num_tokens": 1184065.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.097884321212769,
|
|
"epoch": 0.04052108100867352,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.7539,
|
|
"mean_token_accuracy": 0.10293057784438134,
|
|
"num_tokens": 1193263.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.254788303375244,
|
|
"epoch": 0.04085596597568735,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.8354,
|
|
"mean_token_accuracy": 0.09084267690777778,
|
|
"num_tokens": 1202940.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.08286714553833,
|
|
"epoch": 0.04119085094270118,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.7684,
|
|
"mean_token_accuracy": 0.09483118429780006,
|
|
"num_tokens": 1212978.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.159868144989014,
|
|
"epoch": 0.04152573590971501,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8011,
|
|
"mean_token_accuracy": 0.09588268324732781,
|
|
"num_tokens": 1222945.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.120132064819336,
|
|
"epoch": 0.04186062087672884,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7818,
|
|
"mean_token_accuracy": 0.09418918639421463,
|
|
"num_tokens": 1232041.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.217391872406006,
|
|
"epoch": 0.042195505843742674,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8217,
|
|
"mean_token_accuracy": 0.0957215555012226,
|
|
"num_tokens": 1241765.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.065081977844239,
|
|
"epoch": 0.042530390810756505,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7955,
|
|
"mean_token_accuracy": 0.09950229525566101,
|
|
"num_tokens": 1250642.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.17001485824585,
|
|
"epoch": 0.042865275777770336,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.797,
|
|
"mean_token_accuracy": 0.09573329538106919,
|
|
"num_tokens": 1260250.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.198292636871338,
|
|
"epoch": 0.043200160744784166,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8765,
|
|
"mean_token_accuracy": 0.08549437895417214,
|
|
"num_tokens": 1269786.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.162594366073609,
|
|
"epoch": 0.043535045711798,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.8204,
|
|
"mean_token_accuracy": 0.08664652034640312,
|
|
"num_tokens": 1280029.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.232313537597657,
|
|
"epoch": 0.04386993067881183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7852,
|
|
"mean_token_accuracy": 0.08978526666760445,
|
|
"num_tokens": 1288983.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.10995364189148,
|
|
"epoch": 0.04420481564582566,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.8219,
|
|
"mean_token_accuracy": 0.09514714106917381,
|
|
"num_tokens": 1298629.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.20604944229126,
|
|
"epoch": 0.04453970061283949,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.8112,
|
|
"mean_token_accuracy": 0.09197623878717423,
|
|
"num_tokens": 1308611.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.12814121246338,
|
|
"epoch": 0.04487458557985332,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.8542,
|
|
"mean_token_accuracy": 0.08948481157422065,
|
|
"num_tokens": 1318150.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.188689994812012,
|
|
"epoch": 0.04520947054686715,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7957,
|
|
"mean_token_accuracy": 0.09157142415642738,
|
|
"num_tokens": 1328933.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.072546482086182,
|
|
"epoch": 0.04554435551388098,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7763,
|
|
"mean_token_accuracy": 0.08876086995005608,
|
|
"num_tokens": 1338614.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.14375991821289,
|
|
"epoch": 0.04587924048089481,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.7277,
|
|
"mean_token_accuracy": 0.08602216690778733,
|
|
"num_tokens": 1348292.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.126710891723633,
|
|
"epoch": 0.04621412544790864,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7691,
|
|
"mean_token_accuracy": 0.09218779653310775,
|
|
"num_tokens": 1358479.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.893112850189209,
|
|
"epoch": 0.04654901041492247,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.6351,
|
|
"mean_token_accuracy": 0.09745336174964905,
|
|
"num_tokens": 1367778.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.056758928298951,
|
|
"epoch": 0.046883895381936304,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7244,
|
|
"mean_token_accuracy": 0.09624183624982834,
|
|
"num_tokens": 1376826.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.083500623703003,
|
|
"epoch": 0.047218780348950135,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7727,
|
|
"mean_token_accuracy": 0.0945653960108757,
|
|
"num_tokens": 1386280.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 8.05593729019165,
|
|
"epoch": 0.047553665315963965,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8068,
|
|
"mean_token_accuracy": 0.08938873261213302,
|
|
"num_tokens": 1396062.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 8.16140775680542,
|
|
"epoch": 0.047888550282977796,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.7817,
|
|
"mean_token_accuracy": 0.09469641521573066,
|
|
"num_tokens": 1406980.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.099840116500854,
|
|
"epoch": 0.04822343524999163,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.8061,
|
|
"mean_token_accuracy": 0.09589127041399478,
|
|
"num_tokens": 1417341.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 8.020673513412476,
|
|
"epoch": 0.04855832021700546,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7569,
|
|
"mean_token_accuracy": 0.0935226060450077,
|
|
"num_tokens": 1427066.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 8.142390918731689,
|
|
"epoch": 0.04889320518401929,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.7412,
|
|
"mean_token_accuracy": 0.09235885515809059,
|
|
"num_tokens": 1437429.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 8.085182189941406,
|
|
"epoch": 0.04922809015103312,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.8405,
|
|
"mean_token_accuracy": 0.080220577865839,
|
|
"num_tokens": 1449128.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 8.032454013824463,
|
|
"epoch": 0.04956297511804695,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6932,
|
|
"mean_token_accuracy": 0.09634066671133042,
|
|
"num_tokens": 1459397.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.99392352104187,
|
|
"epoch": 0.04989786008506078,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.7163,
|
|
"mean_token_accuracy": 0.09166975170373917,
|
|
"num_tokens": 1469424.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.977426147460937,
|
|
"epoch": 0.05023274505207461,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6673,
|
|
"mean_token_accuracy": 0.09820285588502883,
|
|
"num_tokens": 1478980.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.043038415908814,
|
|
"epoch": 0.05056763001908844,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.7344,
|
|
"mean_token_accuracy": 0.09137861505150795,
|
|
"num_tokens": 1488655.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.968122100830078,
|
|
"epoch": 0.05090251498610227,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6783,
|
|
"mean_token_accuracy": 0.09548804014921189,
|
|
"num_tokens": 1498248.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 8.059245347976685,
|
|
"epoch": 0.0512373999531161,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6696,
|
|
"mean_token_accuracy": 0.09389448910951614,
|
|
"num_tokens": 1509496.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.984854793548584,
|
|
"epoch": 0.051572284920129934,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6967,
|
|
"mean_token_accuracy": 0.09822245910763741,
|
|
"num_tokens": 1518741.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 8.041652011871339,
|
|
"epoch": 0.051907169887143764,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.7208,
|
|
"mean_token_accuracy": 0.09426980167627334,
|
|
"num_tokens": 1528928.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.959621095657349,
|
|
"epoch": 0.052242054854157595,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6271,
|
|
"mean_token_accuracy": 0.09875941798090934,
|
|
"num_tokens": 1538255.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.942681169509887,
|
|
"epoch": 0.052576939821171426,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.7488,
|
|
"mean_token_accuracy": 0.09209850579500198,
|
|
"num_tokens": 1549014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 8.029151916503906,
|
|
"epoch": 0.052911824788185256,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.6301,
|
|
"mean_token_accuracy": 0.09733811616897584,
|
|
"num_tokens": 1558905.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.876384210586548,
|
|
"epoch": 0.05324670975519909,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.6259,
|
|
"mean_token_accuracy": 0.0984985277056694,
|
|
"num_tokens": 1568605.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.974690532684326,
|
|
"epoch": 0.05358159472221292,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.6282,
|
|
"mean_token_accuracy": 0.09871783554553985,
|
|
"num_tokens": 1579330.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.894849967956543,
|
|
"epoch": 0.05391647968922675,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.6062,
|
|
"mean_token_accuracy": 0.1039244145154953,
|
|
"num_tokens": 1588702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.9055908679962155,
|
|
"epoch": 0.05425136465624058,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5934,
|
|
"mean_token_accuracy": 0.09964857622981071,
|
|
"num_tokens": 1598789.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.929506158828735,
|
|
"epoch": 0.05458624962325441,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.6202,
|
|
"mean_token_accuracy": 0.1006622552871704,
|
|
"num_tokens": 1608580.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.906660175323486,
|
|
"epoch": 0.05492113459026824,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.6326,
|
|
"mean_token_accuracy": 0.09536490365862846,
|
|
"num_tokens": 1618344.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.959885692596435,
|
|
"epoch": 0.05525601955728207,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5767,
|
|
"mean_token_accuracy": 0.09843075722455978,
|
|
"num_tokens": 1627654.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.949274730682373,
|
|
"epoch": 0.0555909045242959,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.7769,
|
|
"mean_token_accuracy": 0.09027430713176728,
|
|
"num_tokens": 1638591.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 8.063038873672486,
|
|
"epoch": 0.05592578949130973,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.775,
|
|
"mean_token_accuracy": 0.08799890652298928,
|
|
"num_tokens": 1649384.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.971291875839233,
|
|
"epoch": 0.05626067445832356,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.7609,
|
|
"mean_token_accuracy": 0.089073646068573,
|
|
"num_tokens": 1659079.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.96489143371582,
|
|
"epoch": 0.056595559425337394,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.6837,
|
|
"mean_token_accuracy": 0.0922718420624733,
|
|
"num_tokens": 1669245.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.955953884124756,
|
|
"epoch": 0.056930444392351225,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.6108,
|
|
"mean_token_accuracy": 0.09481990039348602,
|
|
"num_tokens": 1678781.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.881563901901245,
|
|
"epoch": 0.057265329359365055,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5798,
|
|
"mean_token_accuracy": 0.09848998859524727,
|
|
"num_tokens": 1688192.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.900932312011719,
|
|
"epoch": 0.057600214326378886,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5616,
|
|
"mean_token_accuracy": 0.10043669566512108,
|
|
"num_tokens": 1698058.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.815959119796753,
|
|
"epoch": 0.05793509929339272,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5988,
|
|
"mean_token_accuracy": 0.0973371610045433,
|
|
"num_tokens": 1708568.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.785998249053955,
|
|
"epoch": 0.05826998426040655,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.4876,
|
|
"mean_token_accuracy": 0.10474598705768585,
|
|
"num_tokens": 1718854.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.980491304397583,
|
|
"epoch": 0.05860486922742038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.6177,
|
|
"mean_token_accuracy": 0.09631310030817986,
|
|
"num_tokens": 1728549.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.822048091888428,
|
|
"epoch": 0.05893975419443421,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5631,
|
|
"mean_token_accuracy": 0.09522234201431275,
|
|
"num_tokens": 1737664.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.948170185089111,
|
|
"epoch": 0.05927463916144804,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.6675,
|
|
"mean_token_accuracy": 0.09205883294343949,
|
|
"num_tokens": 1748791.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.904511594772339,
|
|
"epoch": 0.05960952412846187,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5771,
|
|
"mean_token_accuracy": 0.0962847962975502,
|
|
"num_tokens": 1758323.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.879415035247803,
|
|
"epoch": 0.0599444090954757,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5792,
|
|
"mean_token_accuracy": 0.0964177243411541,
|
|
"num_tokens": 1768992.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.8174121379852295,
|
|
"epoch": 0.06027929406248953,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.6383,
|
|
"mean_token_accuracy": 0.09297752380371094,
|
|
"num_tokens": 1779810.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.833243942260742,
|
|
"epoch": 0.06061417902950336,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.5279,
|
|
"mean_token_accuracy": 0.10349605828523636,
|
|
"num_tokens": 1790071.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.777681875228882,
|
|
"epoch": 0.06094906399651719,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5756,
|
|
"mean_token_accuracy": 0.09668522253632546,
|
|
"num_tokens": 1799818.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.803669404983521,
|
|
"epoch": 0.06128394896353103,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5338,
|
|
"mean_token_accuracy": 0.09885151758790016,
|
|
"num_tokens": 1810559.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.81594672203064,
|
|
"epoch": 0.06161883393054486,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4863,
|
|
"mean_token_accuracy": 0.09520618766546249,
|
|
"num_tokens": 1819810.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.8459617614746096,
|
|
"epoch": 0.06195371889755869,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.6062,
|
|
"mean_token_accuracy": 0.10065344572067261,
|
|
"num_tokens": 1830737.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.940606737136841,
|
|
"epoch": 0.06228860386457252,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5841,
|
|
"mean_token_accuracy": 0.09744365289807319,
|
|
"num_tokens": 1841134.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.8107319355010985,
|
|
"epoch": 0.06262348883158635,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5273,
|
|
"mean_token_accuracy": 0.10333630442619324,
|
|
"num_tokens": 1851962.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.789013481140136,
|
|
"epoch": 0.06295837379860018,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.5789,
|
|
"mean_token_accuracy": 0.09511714950203895,
|
|
"num_tokens": 1862748.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.771980476379395,
|
|
"epoch": 0.06329325876561401,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5325,
|
|
"mean_token_accuracy": 0.09738864228129387,
|
|
"num_tokens": 1872398.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.733763551712036,
|
|
"epoch": 0.06362814373262785,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.4044,
|
|
"mean_token_accuracy": 0.10731609687209129,
|
|
"num_tokens": 1882263.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.868111705780029,
|
|
"epoch": 0.06396302869964167,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5711,
|
|
"mean_token_accuracy": 0.09007921293377877,
|
|
"num_tokens": 1893007.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.693736457824707,
|
|
"epoch": 0.0642979136666555,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.5678,
|
|
"mean_token_accuracy": 0.10205003917217255,
|
|
"num_tokens": 1903357.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.930800437927246,
|
|
"epoch": 0.06463279863366933,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5282,
|
|
"mean_token_accuracy": 0.09658677652478218,
|
|
"num_tokens": 1913793.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.683518171310425,
|
|
"epoch": 0.06496768360068317,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4621,
|
|
"mean_token_accuracy": 0.09894432052969933,
|
|
"num_tokens": 1922540.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.766005039215088,
|
|
"epoch": 0.06530256856769699,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5426,
|
|
"mean_token_accuracy": 0.09709056913852691,
|
|
"num_tokens": 1933706.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.771455478668213,
|
|
"epoch": 0.06563745353471083,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4389,
|
|
"mean_token_accuracy": 0.10492749810218811,
|
|
"num_tokens": 1944169.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.653266859054566,
|
|
"epoch": 0.06597233850172465,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3938,
|
|
"mean_token_accuracy": 0.10923026725649834,
|
|
"num_tokens": 1954017.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.701126337051392,
|
|
"epoch": 0.06630722346873849,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5397,
|
|
"mean_token_accuracy": 0.09850775972008705,
|
|
"num_tokens": 1965075.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.752574443817139,
|
|
"epoch": 0.06664210843575231,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.431,
|
|
"mean_token_accuracy": 0.10416013151407241,
|
|
"num_tokens": 1974808.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.718328952789307,
|
|
"epoch": 0.06697699340276615,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.5221,
|
|
"mean_token_accuracy": 0.09244150072336196,
|
|
"num_tokens": 1985172.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.742373752593994,
|
|
"epoch": 0.06731187836977998,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.5187,
|
|
"mean_token_accuracy": 0.10381457358598709,
|
|
"num_tokens": 1996338.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.714392518997192,
|
|
"epoch": 0.06764676333679381,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.5125,
|
|
"mean_token_accuracy": 0.0949136845767498,
|
|
"num_tokens": 2005429.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.7328485488891605,
|
|
"epoch": 0.06798164830380764,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.4857,
|
|
"mean_token_accuracy": 0.0952946312725544,
|
|
"num_tokens": 2015278.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.731113004684448,
|
|
"epoch": 0.06831653327082148,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.4631,
|
|
"mean_token_accuracy": 0.09800807684659958,
|
|
"num_tokens": 2025758.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.606695652008057,
|
|
"epoch": 0.0686514182378353,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.3923,
|
|
"mean_token_accuracy": 0.1038240447640419,
|
|
"num_tokens": 2035029.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.745934104919433,
|
|
"epoch": 0.06898630320484914,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.3838,
|
|
"mean_token_accuracy": 0.10451537966728211,
|
|
"num_tokens": 2044182.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.649461269378662,
|
|
"epoch": 0.06932118817186296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.4697,
|
|
"mean_token_accuracy": 0.09899691492319107,
|
|
"num_tokens": 2054676.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.750053215026855,
|
|
"epoch": 0.0696560731388768,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.4087,
|
|
"mean_token_accuracy": 0.10149102360010147,
|
|
"num_tokens": 2062933.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.57016887664795,
|
|
"epoch": 0.06999095810589062,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.3841,
|
|
"mean_token_accuracy": 0.10415252074599265,
|
|
"num_tokens": 2073146.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.69551362991333,
|
|
"epoch": 0.07032584307290446,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.5109,
|
|
"mean_token_accuracy": 0.09606329277157784,
|
|
"num_tokens": 2083847.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.668817853927612,
|
|
"epoch": 0.07066072803991828,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.4333,
|
|
"mean_token_accuracy": 0.09682594761252403,
|
|
"num_tokens": 2094969.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.698002481460572,
|
|
"epoch": 0.07099561300693212,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.4248,
|
|
"mean_token_accuracy": 0.09735451713204384,
|
|
"num_tokens": 2105275.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.6742143630981445,
|
|
"epoch": 0.07133049797394594,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.4618,
|
|
"mean_token_accuracy": 0.09932760745286942,
|
|
"num_tokens": 2115833.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.704391050338745,
|
|
"epoch": 0.07166538294095978,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.3943,
|
|
"mean_token_accuracy": 0.10237354785203934,
|
|
"num_tokens": 2124747.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.622838020324707,
|
|
"epoch": 0.0720002679079736,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.4275,
|
|
"mean_token_accuracy": 0.10028687044978142,
|
|
"num_tokens": 2135114.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.677819490432739,
|
|
"epoch": 0.07233515287498744,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.4261,
|
|
"mean_token_accuracy": 0.10091969445347786,
|
|
"num_tokens": 2144324.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.634318065643311,
|
|
"epoch": 0.07267003784200127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.09950910061597824,
|
|
"num_tokens": 2153971.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.596848630905152,
|
|
"epoch": 0.0730049228090151,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 7.3452,
|
|
"mean_token_accuracy": 0.10219383016228675,
|
|
"num_tokens": 2164389.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.591188764572143,
|
|
"epoch": 0.07333980777602893,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.3367,
|
|
"mean_token_accuracy": 0.10394962280988693,
|
|
"num_tokens": 2173802.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.628563594818115,
|
|
"epoch": 0.07367469274304277,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.10088860839605332,
|
|
"num_tokens": 2182590.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.564662313461303,
|
|
"epoch": 0.07400957771005659,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 7.341,
|
|
"mean_token_accuracy": 0.10867247581481934,
|
|
"num_tokens": 2192025.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.734026050567627,
|
|
"epoch": 0.07434446267707043,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 7.4933,
|
|
"mean_token_accuracy": 0.10653405413031578,
|
|
"num_tokens": 2201999.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.643482971191406,
|
|
"epoch": 0.07467934764408425,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.4059,
|
|
"mean_token_accuracy": 0.0990131951868534,
|
|
"num_tokens": 2211804.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.540462303161621,
|
|
"epoch": 0.07501423261109809,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 7.2801,
|
|
"mean_token_accuracy": 0.10423837974667549,
|
|
"num_tokens": 2222079.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.603369808197021,
|
|
"epoch": 0.07534911757811191,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.3945,
|
|
"mean_token_accuracy": 0.1089488685131073,
|
|
"num_tokens": 2232213.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.578351306915283,
|
|
"epoch": 0.07568400254512575,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 7.3087,
|
|
"mean_token_accuracy": 0.10233338475227356,
|
|
"num_tokens": 2242017.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.563331890106201,
|
|
"epoch": 0.07601888751213957,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.3008,
|
|
"mean_token_accuracy": 0.10574111714959145,
|
|
"num_tokens": 2251654.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.555946350097656,
|
|
"epoch": 0.07635377247915341,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.3714,
|
|
"mean_token_accuracy": 0.1005162090063095,
|
|
"num_tokens": 2262972.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.546423006057739,
|
|
"epoch": 0.07668865744616724,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 7.3674,
|
|
"mean_token_accuracy": 0.09863484501838685,
|
|
"num_tokens": 2273067.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.524917459487915,
|
|
"epoch": 0.07702354241318107,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 7.2467,
|
|
"mean_token_accuracy": 0.11115464866161347,
|
|
"num_tokens": 2282104.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.571585083007813,
|
|
"epoch": 0.0773584273801949,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 7.3574,
|
|
"mean_token_accuracy": 0.10432401821017265,
|
|
"num_tokens": 2291366.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.562365770339966,
|
|
"epoch": 0.07769331234720873,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.10673468708992004,
|
|
"num_tokens": 2300107.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.523523759841919,
|
|
"epoch": 0.07802819731422256,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 7.2516,
|
|
"mean_token_accuracy": 0.1079891286790371,
|
|
"num_tokens": 2309755.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.5658886432647705,
|
|
"epoch": 0.0783630822812364,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 7.3835,
|
|
"mean_token_accuracy": 0.09904042482376099,
|
|
"num_tokens": 2320042.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.521557331085205,
|
|
"epoch": 0.07869796724825022,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 7.2499,
|
|
"mean_token_accuracy": 0.1096756674349308,
|
|
"num_tokens": 2329592.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.564157485961914,
|
|
"epoch": 0.07903285221526406,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 7.3076,
|
|
"mean_token_accuracy": 0.10216106772422791,
|
|
"num_tokens": 2339219.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.5245466232299805,
|
|
"epoch": 0.07936773718227788,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 7.2776,
|
|
"mean_token_accuracy": 0.10564737990498543,
|
|
"num_tokens": 2349037.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.489606809616089,
|
|
"epoch": 0.07970262214929172,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 7.2917,
|
|
"mean_token_accuracy": 0.1098222590982914,
|
|
"num_tokens": 2359321.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.640200424194336,
|
|
"epoch": 0.08003750711630554,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 7.3224,
|
|
"mean_token_accuracy": 0.10492225512862205,
|
|
"num_tokens": 2368978.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.402497339248657,
|
|
"epoch": 0.08037239208331938,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 7.2502,
|
|
"mean_token_accuracy": 0.11117032766342164,
|
|
"num_tokens": 2378429.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.535825872421265,
|
|
"epoch": 0.0807072770503332,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 7.3044,
|
|
"mean_token_accuracy": 0.0999194212257862,
|
|
"num_tokens": 2389711.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.455561590194702,
|
|
"epoch": 0.08104216201734704,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 7.3199,
|
|
"mean_token_accuracy": 0.10528789013624192,
|
|
"num_tokens": 2399207.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.512408971786499,
|
|
"epoch": 0.08137704698436087,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 7.246,
|
|
"mean_token_accuracy": 0.11051639840006829,
|
|
"num_tokens": 2409236.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.483659648895264,
|
|
"epoch": 0.0817119319513747,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 7.3111,
|
|
"mean_token_accuracy": 0.10474314615130424,
|
|
"num_tokens": 2419588.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.441983795166015,
|
|
"epoch": 0.08204681691838854,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 7.189,
|
|
"mean_token_accuracy": 0.10859855711460113,
|
|
"num_tokens": 2429651.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.5176849365234375,
|
|
"epoch": 0.08238170188540236,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 7.3836,
|
|
"mean_token_accuracy": 0.10394779369235038,
|
|
"num_tokens": 2439942.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.536934804916382,
|
|
"epoch": 0.0827165868524162,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 7.2697,
|
|
"mean_token_accuracy": 0.1037393756210804,
|
|
"num_tokens": 2448843.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.440825605392456,
|
|
"epoch": 0.08305147181943003,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 7.2961,
|
|
"mean_token_accuracy": 0.10528851747512817,
|
|
"num_tokens": 2458878.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.514740467071533,
|
|
"epoch": 0.08338635678644386,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 7.1722,
|
|
"mean_token_accuracy": 0.1126915343105793,
|
|
"num_tokens": 2468527.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.459214782714843,
|
|
"epoch": 0.08372124175345769,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 7.3297,
|
|
"mean_token_accuracy": 0.10487211868166924,
|
|
"num_tokens": 2478996.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.532850980758667,
|
|
"epoch": 0.08405612672047152,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 7.2155,
|
|
"mean_token_accuracy": 0.10624711290001869,
|
|
"num_tokens": 2488697.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.4011256217956545,
|
|
"epoch": 0.08439101168748535,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 7.2525,
|
|
"mean_token_accuracy": 0.10476204007863998,
|
|
"num_tokens": 2498590.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.524996042251587,
|
|
"epoch": 0.08472589665449919,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 7.2281,
|
|
"mean_token_accuracy": 0.11188126131892204,
|
|
"num_tokens": 2507824.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.413699436187744,
|
|
"epoch": 0.08506078162151301,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 7.2556,
|
|
"mean_token_accuracy": 0.10911233723163605,
|
|
"num_tokens": 2516738.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.523989057540893,
|
|
"epoch": 0.08539566658852685,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 7.2787,
|
|
"mean_token_accuracy": 0.10768822953104973,
|
|
"num_tokens": 2525843.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.462717294692993,
|
|
"epoch": 0.08573055155554067,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 7.2549,
|
|
"mean_token_accuracy": 0.10760494396090507,
|
|
"num_tokens": 2534877.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.475369882583618,
|
|
"epoch": 0.08606543652255451,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 7.2682,
|
|
"mean_token_accuracy": 0.1045633964240551,
|
|
"num_tokens": 2545197.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.513711977005005,
|
|
"epoch": 0.08640032148956833,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 7.2268,
|
|
"mean_token_accuracy": 0.11084463000297547,
|
|
"num_tokens": 2555732.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.354825687408447,
|
|
"epoch": 0.08673520645658217,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 7.2796,
|
|
"mean_token_accuracy": 0.1026076965034008,
|
|
"num_tokens": 2565779.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.518953037261963,
|
|
"epoch": 0.087070091423596,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 7.1898,
|
|
"mean_token_accuracy": 0.11239831075072289,
|
|
"num_tokens": 2574822.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.432813692092895,
|
|
"epoch": 0.08740497639060983,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 7.2885,
|
|
"mean_token_accuracy": 0.10602438822388649,
|
|
"num_tokens": 2584780.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.416839838027954,
|
|
"epoch": 0.08773986135762366,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 7.302,
|
|
"mean_token_accuracy": 0.1066212609410286,
|
|
"num_tokens": 2595080.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.546160364151001,
|
|
"epoch": 0.08807474632463749,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 7.3271,
|
|
"mean_token_accuracy": 0.1047338679432869,
|
|
"num_tokens": 2605221.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.424958944320679,
|
|
"epoch": 0.08840963129165132,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 7.2466,
|
|
"mean_token_accuracy": 0.1068871334195137,
|
|
"num_tokens": 2615915.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.492191123962402,
|
|
"epoch": 0.08874451625866515,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 7.2208,
|
|
"mean_token_accuracy": 0.10681156888604164,
|
|
"num_tokens": 2626791.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.396345329284668,
|
|
"epoch": 0.08907940122567898,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 7.2821,
|
|
"mean_token_accuracy": 0.10661014467477799,
|
|
"num_tokens": 2636055.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.509826612472534,
|
|
"epoch": 0.08941428619269282,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 7.223,
|
|
"mean_token_accuracy": 0.10235508903861046,
|
|
"num_tokens": 2645177.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.397401523590088,
|
|
"epoch": 0.08974917115970664,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 7.1743,
|
|
"mean_token_accuracy": 0.11122062802314758,
|
|
"num_tokens": 2655961.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.423821973800659,
|
|
"epoch": 0.09008405612672048,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 7.1747,
|
|
"mean_token_accuracy": 0.1182539276778698,
|
|
"num_tokens": 2666394.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.3939965724945065,
|
|
"epoch": 0.0904189410937343,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 7.2132,
|
|
"mean_token_accuracy": 0.11129515990614891,
|
|
"num_tokens": 2675309.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.412256145477295,
|
|
"epoch": 0.09075382606074814,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 7.2046,
|
|
"mean_token_accuracy": 0.10680647864937783,
|
|
"num_tokens": 2685500.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.455058908462524,
|
|
"epoch": 0.09108871102776196,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 7.1317,
|
|
"mean_token_accuracy": 0.10418605953454971,
|
|
"num_tokens": 2694648.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.392323350906372,
|
|
"epoch": 0.0914235959947758,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 7.2243,
|
|
"mean_token_accuracy": 0.11022280678153037,
|
|
"num_tokens": 2703878.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.443844509124756,
|
|
"epoch": 0.09175848096178962,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 7.2676,
|
|
"mean_token_accuracy": 0.10342767983675002,
|
|
"num_tokens": 2714142.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.408930969238281,
|
|
"epoch": 0.09209336592880346,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 7.1191,
|
|
"mean_token_accuracy": 0.11517907232046128,
|
|
"num_tokens": 2723083.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.352519130706787,
|
|
"epoch": 0.09242825089581728,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 7.2835,
|
|
"mean_token_accuracy": 0.09940153509378433,
|
|
"num_tokens": 2733577.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.593472623825074,
|
|
"epoch": 0.09276313586283112,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 7.2583,
|
|
"mean_token_accuracy": 0.094093906134367,
|
|
"num_tokens": 2744134.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.268954944610596,
|
|
"epoch": 0.09309802082984495,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 7.1409,
|
|
"mean_token_accuracy": 0.1097214013338089,
|
|
"num_tokens": 2754460.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.436053609848022,
|
|
"epoch": 0.09343290579685878,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 7.2179,
|
|
"mean_token_accuracy": 0.10809629186987876,
|
|
"num_tokens": 2763935.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.448184013366699,
|
|
"epoch": 0.09376779076387261,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 7.2089,
|
|
"mean_token_accuracy": 0.10305045768618584,
|
|
"num_tokens": 2774945.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.374975299835205,
|
|
"epoch": 0.09410267573088645,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 7.238,
|
|
"mean_token_accuracy": 0.10469259843230247,
|
|
"num_tokens": 2784799.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.287476062774658,
|
|
"epoch": 0.09443756069790027,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 7.0278,
|
|
"mean_token_accuracy": 0.11678826436400414,
|
|
"num_tokens": 2794506.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.398726749420166,
|
|
"epoch": 0.0947724456649141,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 7.2753,
|
|
"mean_token_accuracy": 0.10628883987665176,
|
|
"num_tokens": 2804547.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.309502363204956,
|
|
"epoch": 0.09510733063192793,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 7.1353,
|
|
"mean_token_accuracy": 0.11116150170564651,
|
|
"num_tokens": 2815279.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.472178888320923,
|
|
"epoch": 0.09544221559894177,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 7.1829,
|
|
"mean_token_accuracy": 0.1019278421998024,
|
|
"num_tokens": 2825754.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.401741695404053,
|
|
"epoch": 0.09577710056595559,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 7.3618,
|
|
"mean_token_accuracy": 0.10463809296488762,
|
|
"num_tokens": 2835775.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.404381275177002,
|
|
"epoch": 0.09611198553296943,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 7.1009,
|
|
"mean_token_accuracy": 0.10979082211852073,
|
|
"num_tokens": 2845065.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.421131610870361,
|
|
"epoch": 0.09644687049998325,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 7.2664,
|
|
"mean_token_accuracy": 0.10228715687990189,
|
|
"num_tokens": 2855493.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.447347784042359,
|
|
"epoch": 0.09678175546699709,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 7.0681,
|
|
"mean_token_accuracy": 0.11278982535004616,
|
|
"num_tokens": 2864449.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.192193031311035,
|
|
"epoch": 0.09711664043401091,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 7.1074,
|
|
"mean_token_accuracy": 0.11356074139475822,
|
|
"num_tokens": 2874056.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.394455814361573,
|
|
"epoch": 0.09745152540102475,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 7.1445,
|
|
"mean_token_accuracy": 0.10959847420454025,
|
|
"num_tokens": 2884283.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.407599925994873,
|
|
"epoch": 0.09778641036803858,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 7.1341,
|
|
"mean_token_accuracy": 0.11027526259422302,
|
|
"num_tokens": 2892720.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.210386323928833,
|
|
"epoch": 0.09812129533505241,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 7.1332,
|
|
"mean_token_accuracy": 0.11168786510825157,
|
|
"num_tokens": 2902671.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.467904567718506,
|
|
"epoch": 0.09845618030206624,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 7.1744,
|
|
"mean_token_accuracy": 0.10181471109390258,
|
|
"num_tokens": 2912683.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.347403192520142,
|
|
"epoch": 0.09879106526908008,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 7.1376,
|
|
"mean_token_accuracy": 0.1077820472419262,
|
|
"num_tokens": 2922888.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.378122711181641,
|
|
"epoch": 0.0991259502360939,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 7.1404,
|
|
"mean_token_accuracy": 0.1046745590865612,
|
|
"num_tokens": 2932284.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.382054424285888,
|
|
"epoch": 0.09946083520310774,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 7.1617,
|
|
"mean_token_accuracy": 0.11370518803596497,
|
|
"num_tokens": 2942061.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.303506755828858,
|
|
"epoch": 0.09979572017012156,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 7.0405,
|
|
"mean_token_accuracy": 0.10833846926689147,
|
|
"num_tokens": 2950216.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.301569938659668,
|
|
"epoch": 0.1001306051371354,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 7.1075,
|
|
"mean_token_accuracy": 0.10717656314373017,
|
|
"num_tokens": 2959578.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.366466760635376,
|
|
"epoch": 0.10046549010414922,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 7.0381,
|
|
"mean_token_accuracy": 0.11490087807178498,
|
|
"num_tokens": 2969043.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.267051839828492,
|
|
"epoch": 0.10080037507116306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 7.1022,
|
|
"mean_token_accuracy": 0.11307050734758377,
|
|
"num_tokens": 2978765.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.34058256149292,
|
|
"epoch": 0.10113526003817688,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 7.0988,
|
|
"mean_token_accuracy": 0.11385324522852898,
|
|
"num_tokens": 2988878.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.379073524475098,
|
|
"epoch": 0.10147014500519072,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 7.1492,
|
|
"mean_token_accuracy": 0.11306478753685952,
|
|
"num_tokens": 2999048.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.370050287246704,
|
|
"epoch": 0.10180502997220454,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 7.0417,
|
|
"mean_token_accuracy": 0.11641403511166573,
|
|
"num_tokens": 3008847.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.267196941375732,
|
|
"epoch": 0.10213991493921838,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 7.1068,
|
|
"mean_token_accuracy": 0.11009391099214554,
|
|
"num_tokens": 3018514.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.325578784942627,
|
|
"epoch": 0.1024747999062322,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 7.016,
|
|
"mean_token_accuracy": 0.10993625149130822,
|
|
"num_tokens": 3029096.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.281036233901977,
|
|
"epoch": 0.10280968487324604,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 7.0964,
|
|
"mean_token_accuracy": 0.11679442673921585,
|
|
"num_tokens": 3037163.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.350086259841919,
|
|
"epoch": 0.10314456984025987,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 7.1711,
|
|
"mean_token_accuracy": 0.10986532866954804,
|
|
"num_tokens": 3046763.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.394155740737915,
|
|
"epoch": 0.1034794548072737,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 7.0185,
|
|
"mean_token_accuracy": 0.11672415882349015,
|
|
"num_tokens": 3056250.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.247371864318848,
|
|
"epoch": 0.10381433977428753,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 7.0457,
|
|
"mean_token_accuracy": 0.11317939907312394,
|
|
"num_tokens": 3067238.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.295481586456299,
|
|
"epoch": 0.10414922474130137,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 7.2072,
|
|
"mean_token_accuracy": 0.10626539513468743,
|
|
"num_tokens": 3077479.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.352273797988891,
|
|
"epoch": 0.10448410970831519,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 7.1346,
|
|
"mean_token_accuracy": 0.1156134270131588,
|
|
"num_tokens": 3086639.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.486025619506836,
|
|
"epoch": 0.10481899467532903,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 7.1736,
|
|
"mean_token_accuracy": 0.10581467673182487,
|
|
"num_tokens": 3096501.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.295076704025268,
|
|
"epoch": 0.10515387964234285,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 7.0005,
|
|
"mean_token_accuracy": 0.12267342209815979,
|
|
"num_tokens": 3106026.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.231325626373291,
|
|
"epoch": 0.10548876460935669,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 7.0646,
|
|
"mean_token_accuracy": 0.11261114478111267,
|
|
"num_tokens": 3114744.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.308481884002686,
|
|
"epoch": 0.10582364957637051,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 7.0507,
|
|
"mean_token_accuracy": 0.11041012555360794,
|
|
"num_tokens": 3123996.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.295450687408447,
|
|
"epoch": 0.10615853454338435,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 7.0963,
|
|
"mean_token_accuracy": 0.11200174242258072,
|
|
"num_tokens": 3133301.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.312792110443115,
|
|
"epoch": 0.10649341951039817,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 7.0417,
|
|
"mean_token_accuracy": 0.11235228106379509,
|
|
"num_tokens": 3142505.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.286295175552368,
|
|
"epoch": 0.10682830447741201,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 7.0886,
|
|
"mean_token_accuracy": 0.11075412482023239,
|
|
"num_tokens": 3152858.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.293213748931885,
|
|
"epoch": 0.10716318944442584,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 7.0954,
|
|
"mean_token_accuracy": 0.1056660145521164,
|
|
"num_tokens": 3163241.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.245172357559204,
|
|
"epoch": 0.10749807441143967,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 7.0358,
|
|
"mean_token_accuracy": 0.12011707574129105,
|
|
"num_tokens": 3172461.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.342641592025757,
|
|
"epoch": 0.1078329593784535,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 7.0444,
|
|
"mean_token_accuracy": 0.11463872864842414,
|
|
"num_tokens": 3181686.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.122599506378174,
|
|
"epoch": 0.10816784434546733,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.9989,
|
|
"mean_token_accuracy": 0.12019835412502289,
|
|
"num_tokens": 3191123.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.259741258621216,
|
|
"epoch": 0.10850272931248116,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.94,
|
|
"mean_token_accuracy": 0.11922215074300765,
|
|
"num_tokens": 3200346.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.32574348449707,
|
|
"epoch": 0.108837614279495,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 7.1218,
|
|
"mean_token_accuracy": 0.11046352982521057,
|
|
"num_tokens": 3211345.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.289894533157349,
|
|
"epoch": 0.10917249924650882,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 7.1476,
|
|
"mean_token_accuracy": 0.10400293096899986,
|
|
"num_tokens": 3221621.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.386224317550659,
|
|
"epoch": 0.10950738421352266,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 7.1045,
|
|
"mean_token_accuracy": 0.10810255780816078,
|
|
"num_tokens": 3231535.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.236555194854736,
|
|
"epoch": 0.10984226918053648,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 7.0087,
|
|
"mean_token_accuracy": 0.11163441389799118,
|
|
"num_tokens": 3241649.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.286080074310303,
|
|
"epoch": 0.11017715414755032,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 7.033,
|
|
"mean_token_accuracy": 0.11515506654977799,
|
|
"num_tokens": 3251690.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.257711172103882,
|
|
"epoch": 0.11051203911456414,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.9924,
|
|
"mean_token_accuracy": 0.11696538552641869,
|
|
"num_tokens": 3260852.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.225003290176391,
|
|
"epoch": 0.11084692408157798,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 7.0525,
|
|
"mean_token_accuracy": 0.11066322326660157,
|
|
"num_tokens": 3270695.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.253508138656616,
|
|
"epoch": 0.1111818090485918,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.9633,
|
|
"mean_token_accuracy": 0.11173780634999275,
|
|
"num_tokens": 3279417.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.200061988830567,
|
|
"epoch": 0.11151669401560564,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 7.0637,
|
|
"mean_token_accuracy": 0.11319149881601334,
|
|
"num_tokens": 3289787.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.308526706695557,
|
|
"epoch": 0.11185157898261947,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.9691,
|
|
"mean_token_accuracy": 0.11160579174757004,
|
|
"num_tokens": 3298454.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.146737957000733,
|
|
"epoch": 0.1121864639496333,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.996,
|
|
"mean_token_accuracy": 0.11499083116650581,
|
|
"num_tokens": 3308794.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.292838478088379,
|
|
"epoch": 0.11252134891664713,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 7.0596,
|
|
"mean_token_accuracy": 0.10847899839282035,
|
|
"num_tokens": 3318371.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.228676080703735,
|
|
"epoch": 0.11285623388366096,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 7.0088,
|
|
"mean_token_accuracy": 0.10878502950072289,
|
|
"num_tokens": 3327717.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.209138488769531,
|
|
"epoch": 0.11319111885067479,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.9442,
|
|
"mean_token_accuracy": 0.12112146839499474,
|
|
"num_tokens": 3337188.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.208089971542359,
|
|
"epoch": 0.11352600381768863,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.9729,
|
|
"mean_token_accuracy": 0.11790513694286346,
|
|
"num_tokens": 3345665.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.215198612213134,
|
|
"epoch": 0.11386088878470245,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 7.0227,
|
|
"mean_token_accuracy": 0.11154457107186318,
|
|
"num_tokens": 3355178.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.239945411682129,
|
|
"epoch": 0.11419577375171629,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 7.0399,
|
|
"mean_token_accuracy": 0.11386125013232232,
|
|
"num_tokens": 3364379.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.294025516510009,
|
|
"epoch": 0.11453065871873011,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 7.0327,
|
|
"mean_token_accuracy": 0.11665912345051765,
|
|
"num_tokens": 3374187.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.173009729385376,
|
|
"epoch": 0.11486554368574395,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.9908,
|
|
"mean_token_accuracy": 0.11849205568432808,
|
|
"num_tokens": 3383602.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.283406448364258,
|
|
"epoch": 0.11520042865275777,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 7.0612,
|
|
"mean_token_accuracy": 0.11695406585931778,
|
|
"num_tokens": 3395542.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.281910800933838,
|
|
"epoch": 0.11553531361977161,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 7.0224,
|
|
"mean_token_accuracy": 0.1149914063513279,
|
|
"num_tokens": 3404528.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.259571361541748,
|
|
"epoch": 0.11587019858678543,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 7.0302,
|
|
"mean_token_accuracy": 0.10809605866670609,
|
|
"num_tokens": 3414525.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.183487796783448,
|
|
"epoch": 0.11620508355379927,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.997,
|
|
"mean_token_accuracy": 0.11267608925700187,
|
|
"num_tokens": 3424743.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.199751615524292,
|
|
"epoch": 0.1165399685208131,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 7.0015,
|
|
"mean_token_accuracy": 0.1154498003423214,
|
|
"num_tokens": 3435516.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.218643569946289,
|
|
"epoch": 0.11687485348782693,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 7.0353,
|
|
"mean_token_accuracy": 0.11391362622380256,
|
|
"num_tokens": 3445017.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.2358087539672855,
|
|
"epoch": 0.11720973845484076,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.9598,
|
|
"mean_token_accuracy": 0.1185583457350731,
|
|
"num_tokens": 3454200.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.164967584609985,
|
|
"epoch": 0.1175446234218546,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.9548,
|
|
"mean_token_accuracy": 0.11466429457068443,
|
|
"num_tokens": 3464631.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.177149629592895,
|
|
"epoch": 0.11787950838886842,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.888,
|
|
"mean_token_accuracy": 0.1215953677892685,
|
|
"num_tokens": 3474564.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.144310569763183,
|
|
"epoch": 0.11821439335588226,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.9856,
|
|
"mean_token_accuracy": 0.11216104701161385,
|
|
"num_tokens": 3485464.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.19792594909668,
|
|
"epoch": 0.11854927832289608,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.9271,
|
|
"mean_token_accuracy": 0.11837260350584984,
|
|
"num_tokens": 3495194.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.142852163314819,
|
|
"epoch": 0.11888416328990992,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.9649,
|
|
"mean_token_accuracy": 0.11836648434400558,
|
|
"num_tokens": 3505737.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.164193153381348,
|
|
"epoch": 0.11921904825692374,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.998,
|
|
"mean_token_accuracy": 0.11925284788012505,
|
|
"num_tokens": 3515520.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.218386220932007,
|
|
"epoch": 0.11955393322393758,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.8363,
|
|
"mean_token_accuracy": 0.11765402480959893,
|
|
"num_tokens": 3524795.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.1260106563568115,
|
|
"epoch": 0.1198888181909514,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.8932,
|
|
"mean_token_accuracy": 0.12132256329059601,
|
|
"num_tokens": 3535022.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.213653326034546,
|
|
"epoch": 0.12022370315796524,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 7.0069,
|
|
"mean_token_accuracy": 0.10930481627583503,
|
|
"num_tokens": 3544560.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.243549633026123,
|
|
"epoch": 0.12055858812497906,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.9755,
|
|
"mean_token_accuracy": 0.1146695539355278,
|
|
"num_tokens": 3554298.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.128929853439331,
|
|
"epoch": 0.1208934730919929,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.9815,
|
|
"mean_token_accuracy": 0.11245792731642723,
|
|
"num_tokens": 3563574.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.113662958145142,
|
|
"epoch": 0.12122835805900672,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.9993,
|
|
"mean_token_accuracy": 0.11119666472077369,
|
|
"num_tokens": 3573529.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.312092447280884,
|
|
"epoch": 0.12156324302602056,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 7.0139,
|
|
"mean_token_accuracy": 0.11274924501776695,
|
|
"num_tokens": 3583604.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.1304497718811035,
|
|
"epoch": 0.12189812799303439,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.8591,
|
|
"mean_token_accuracy": 0.11975032612681388,
|
|
"num_tokens": 3593341.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.103163576126098,
|
|
"epoch": 0.12223301296004822,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 7.0363,
|
|
"mean_token_accuracy": 0.10505542308092117,
|
|
"num_tokens": 3603654.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.410056734085083,
|
|
"epoch": 0.12256789792706206,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 7.1241,
|
|
"mean_token_accuracy": 0.10470781922340393,
|
|
"num_tokens": 3613651.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.134521055221557,
|
|
"epoch": 0.12290278289407588,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.9188,
|
|
"mean_token_accuracy": 0.11272940561175346,
|
|
"num_tokens": 3623860.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.209635782241821,
|
|
"epoch": 0.12323766786108972,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.9595,
|
|
"mean_token_accuracy": 0.11435427442193032,
|
|
"num_tokens": 3633614.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.188301849365234,
|
|
"epoch": 0.12357255282810355,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.9331,
|
|
"mean_token_accuracy": 0.11353752762079239,
|
|
"num_tokens": 3644282.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.215322494506836,
|
|
"epoch": 0.12390743779511738,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 7.0218,
|
|
"mean_token_accuracy": 0.11466898620128632,
|
|
"num_tokens": 3653572.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.160602998733521,
|
|
"epoch": 0.12424232276213121,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.11843540817499161,
|
|
"num_tokens": 3664732.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.1696264266967775,
|
|
"epoch": 0.12457720772914505,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.9607,
|
|
"mean_token_accuracy": 0.1168016955256462,
|
|
"num_tokens": 3674457.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.144387102127075,
|
|
"epoch": 0.12491209269615887,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.8783,
|
|
"mean_token_accuracy": 0.12042904272675514,
|
|
"num_tokens": 3683754.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.109519815444946,
|
|
"epoch": 0.1252469776631727,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.9568,
|
|
"mean_token_accuracy": 0.11945500671863556,
|
|
"num_tokens": 3692529.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.256045913696289,
|
|
"epoch": 0.12558186263018653,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.9898,
|
|
"mean_token_accuracy": 0.11463228091597558,
|
|
"num_tokens": 3702013.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.220793533325195,
|
|
"epoch": 0.12591674759720037,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.908,
|
|
"mean_token_accuracy": 0.12105381786823273,
|
|
"num_tokens": 3712548.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.046376323699951,
|
|
"epoch": 0.1262516325642142,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.8764,
|
|
"mean_token_accuracy": 0.11862443760037422,
|
|
"num_tokens": 3722307.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.132947492599487,
|
|
"epoch": 0.12658651753122802,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.843,
|
|
"mean_token_accuracy": 0.12242488786578179,
|
|
"num_tokens": 3732665.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.15237283706665,
|
|
"epoch": 0.12692140249824185,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.9137,
|
|
"mean_token_accuracy": 0.12014789804816246,
|
|
"num_tokens": 3742667.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.08045802116394,
|
|
"epoch": 0.1272562874652557,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.8803,
|
|
"mean_token_accuracy": 0.11937455907464027,
|
|
"num_tokens": 3753961.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.215423440933227,
|
|
"epoch": 0.12759117243226953,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 7.0593,
|
|
"mean_token_accuracy": 0.11359925791621209,
|
|
"num_tokens": 3763493.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.217863655090332,
|
|
"epoch": 0.12792605739928334,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.8634,
|
|
"mean_token_accuracy": 0.12124418541789055,
|
|
"num_tokens": 3772673.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.060074377059936,
|
|
"epoch": 0.12826094236629718,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.9514,
|
|
"mean_token_accuracy": 0.11748291477560997,
|
|
"num_tokens": 3782601.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.124035215377807,
|
|
"epoch": 0.128595827333311,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.8538,
|
|
"mean_token_accuracy": 0.1251473009586334,
|
|
"num_tokens": 3793277.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.247915887832642,
|
|
"epoch": 0.12893071230032485,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.9187,
|
|
"mean_token_accuracy": 0.1132840670645237,
|
|
"num_tokens": 3802882.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.155371379852295,
|
|
"epoch": 0.12926559726733866,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.8811,
|
|
"mean_token_accuracy": 0.11853846460580826,
|
|
"num_tokens": 3812520.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.082964658737183,
|
|
"epoch": 0.1296004822343525,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.8706,
|
|
"mean_token_accuracy": 0.11984705701470375,
|
|
"num_tokens": 3822943.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 7.128271675109863,
|
|
"epoch": 0.12993536720136634,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.8484,
|
|
"mean_token_accuracy": 0.12053665444254875,
|
|
"num_tokens": 3832762.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.086871862411499,
|
|
"epoch": 0.13027025216838017,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.8535,
|
|
"mean_token_accuracy": 0.12352325096726417,
|
|
"num_tokens": 3842631.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.143572854995727,
|
|
"epoch": 0.13060513713539398,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.9091,
|
|
"mean_token_accuracy": 0.11555887237191201,
|
|
"num_tokens": 3852508.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.1171807765960695,
|
|
"epoch": 0.13094002210240782,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.7942,
|
|
"mean_token_accuracy": 0.12326547503471375,
|
|
"num_tokens": 3861305.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.089008808135986,
|
|
"epoch": 0.13127490706942166,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.10690325349569321,
|
|
"num_tokens": 3870786.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.117994117736816,
|
|
"epoch": 0.1316097920364355,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.8141,
|
|
"mean_token_accuracy": 0.1205119639635086,
|
|
"num_tokens": 3879723.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.148208427429199,
|
|
"epoch": 0.1319446770034493,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.9016,
|
|
"mean_token_accuracy": 0.12056270465254784,
|
|
"num_tokens": 3888755.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.1228758811950685,
|
|
"epoch": 0.13227956197046314,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.8287,
|
|
"mean_token_accuracy": 0.11535504683852196,
|
|
"num_tokens": 3898543.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.07359676361084,
|
|
"epoch": 0.13261444693747698,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.8781,
|
|
"mean_token_accuracy": 0.12056904509663582,
|
|
"num_tokens": 3909130.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.0354334831237795,
|
|
"epoch": 0.13294933190449082,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.782,
|
|
"mean_token_accuracy": 0.12990196645259858,
|
|
"num_tokens": 3918214.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.060833883285523,
|
|
"epoch": 0.13328421687150463,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.8646,
|
|
"mean_token_accuracy": 0.11948921456933022,
|
|
"num_tokens": 3927629.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.094577646255493,
|
|
"epoch": 0.13361910183851847,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.11988667100667953,
|
|
"num_tokens": 3937048.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.189002180099488,
|
|
"epoch": 0.1339539868055323,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.9051,
|
|
"mean_token_accuracy": 0.11899922117590904,
|
|
"num_tokens": 3947615.0,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 934107155988480.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|