2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.06697699340276615,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 467632251371520.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|